Monday, August 27, 2007

Part III

Department of Health and Human Services Food and Drug Administration

21 CFR Parts 347 and 352 Sunscreen Drug Products for Over-the- Counter Human Use; Proposed Amendment of Final Monograph; Proposed Rule

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49070 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

DEPARTMENT OF HEALTH AND 305), Food and Drug Administration, Sunscreen Drug Products With HUMAN SERVICES 5630 Fishers Lane, rm. 1061, Rockville, UVA Protection MD 20852. F. Comments on the Labeling of Food and Drug Administration To ensure more timely processing of Sunscreen Drug Products With High comments, FDA is no longer accepting SPF Values 21 CFR Parts 347 and 352 comments submitted to the agency by e- G. Comments on Indications for [Docket No. 1978N–0038] (formerly Docket mail. FDA encourages you to continue Sunscreen Drug Products No. 78N–0038) to submit electronic comments by using H. Comments on Directions for the Federal eRulemaking Portal or the Sunscreen Drug Products RIN 0910–AF43 agency Web site, as described in the I. General Comments on SPF Testing Electronic Submissions portion of this Procedure Sunscreen Drug Products for Over-the- paragraph. J. Comments on the Sunscreen Counter Human Use; Proposed Instructions: All submissions received Standard for SPF Testing Procedure Amendment of Final Monograph must include the agency name, docket K. Comments on Artificial Light AGENCY: Food and Drug Administration, number and regulatory information Sources for SPF Testing Procedure HHS. number (RIN) for this rulemaking. All L. Comments on the Design/Analysis of SPF Testing Procedure ACTION: Proposed rule. comments received may be posted without change to http://www.fda.gov/ M. General Comments on UVA SUMMARY: The Food and Drug ohrms/dockets/default.htm, including Testing Procedure Administration (FDA) is issuing a any personal information provided. For N. Comments on UVA Testing proposed rule that would amend the additional information on submitting Procedure Design and Testing final monograph (FM) for over-the- comments, see the ‘‘Request for Criteria counter (OTC) sunscreen drug products Comments’’ heading of the O. Comments on the Photostability of as part of FDA’s ongoing review of OTC SUPPLEMENTARY INFORMATION section of Sunscreen Drug Products drug products. This amendment this document. IV. FDA’s Tentative Conclusions and addresses formulation, labeling, and Docket: For access to the docket to Proposals testing requirements for both ultraviolet read background documents or V. Analysis of Impacts B (UVB) and ultraviolet A (UVA) comments received, go to http:// A. Background radiation protection. FDA is issuing this www.fda.gov/ohrms/dockets/ B. Number of Products Affected proposed rule after considering public default.htm and insert the docket C. Cost to Relabel D. Cost to Test or Retest Products for comments and new data and number, found in brackets in the UVA Protection information that have come to FDA’s heading of this document, into the E. Total Incremental Costs attention. This rule proposes to lift the ‘‘Search’’ box and follow the prompts F. Small Business Impact stays of 21 CFR 347.20(d) and 21 CFR and/or go to the Division of Dockets G. Analysis of Alternatives Management, 5630 Fishers Lane, rm. Part 352 when FDA publishes a final VI. Paperwork Reduction Act of 1995 1061, Rockville, MD 20852. rule based on this proposed rule. VII. Environmental Impact DATES: Submit written or electronic FOR FURTHER INFORMATION CONTACT: VIII. Federalism comments by November 26, 2007. Matthew R. Holman, Office of IX. Request for Comments Submit written or electronic comments Nonprescription Products, Center for X. Proposed Effective and Compliance on FDA’s economic impact Drug Evaluation and Research, Food Dates determination by November 26, 2007. and Drug Administration, 10903 New XI. References Please see section X of this document Hampshire Ave., Bldg. 22, rm. 5414, for the effective and compliance dates of Silver Spring, MD 20993, 301–796– I. Background any final rule that may publish based on 2090. In the Federal Register of May 12, this proposal. SUPPLEMENTARY INFORMATION: 1993 (58 FR 28194), FDA published a ADDRESSES: You may submit comments, notice of proposed rulemaking in the identified by Docket No. 1978N–0038 Table of Contents form of a tentative final monograph and RIN number 0910–AF43, by any of I. Background (TFM) for OTC sunscreen drug the following methods: II. Summary of Major Changes to the FM products. In the TFM, FDA proposed A. Ingredients the conditions under which OTC Electronic Submissions B. UVB (SPF) Labeling sunscreen drug products would be Submit electronic comments in the C. UVA Labeling considered generally recognized as safe following ways: D. Indications and effective (GRASE), under section • Federal eRulemaking Portal: http:// E. Warnings 201(p) of the Federal Food, Drug, and www.regulations.gov. Follow the F. Directions Cosmetic Act (the act) (21 U.S.C. instructions for submitting comments. G. UVB Testing 321(p)), and not misbranded, under • Agency Web site: http:// H. UVA Testing section 502 of the act (21 U.S.C. 352). www.fda.gov/dockets/ecomments. III. FDA’s Tentative Conclusions on the In the Federal Register of April 5, Follow the instructions for submitting Comments 1994 (59 FR 16042), FDA reopened the comments on the agency Web site. A. General Comments on OTC administrative record until July 31, Sunscreen Drug Products 1994, to allow additional submissions Written Submissions B. Comments on Tanning and on UVA-related issues and announced a Submit written submissions in the Tanning Preparations public meeting for May 12, 1994, to following ways: C. Comments on Specific Sunscreen discuss UVA testing procedures. As • FAX: 301–827–6870. Active Ingredients explained in that Federal Register • Mail/Hand delivery/Courier (for D. General Comments on the Labeling notice, the TFM included proposed paper, disk, or CD–ROM submissions): of Sunscreen Drug Products UVB (i.e., 290–320 nm) testing and Division of Dockets Management (HFA– E. Comments on the Labeling of labeling. The sun protection factor (SPF)

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49071

test and corresponding labeling reflects water resistant test methods and establishing conditions under which the level of protection against sunburn, cosmetic labeling) as items that could be OTC skin protectant products are which is caused primarily by UVB finalized within the timeframe set by generally recognized as safe and radiation. The TFM also explained the FDAMA. Because of outstanding issues effective and not misbranded. This final importance of protection against UVA related to the development of testing rule lifted the stay of 21 CFR part 352 radiation (i.e., 320–400 nm), the other standards and labeling for UVA to amend the final monograph for OTC UV component of sunlight (58 FR 28194 radiation protection, FDA deferred final sunscreen drug products to include at 28232 and 28233). The TFM action on these items. sunscreen-skin protectant combination referenced published UVA test methods Therefore, in the Federal Register of drug products. This final rule concluded but did not propose a method (58 FR May 21, 1999 (64 FR 27666), FDA by placing a stay on both part 352 and published the FM for OTC sunscreen 28194 at 28248 to 28250). Rather, the on § 347.20(d). The proposed rule that is drug products in part 352 (21 CFR part TFM stated that a product could be the subject of this document provides labeled as ‘‘broad spectrum’’ or a similar 352) with an effective date of May 21, UVA testing and labeling that is claim if it protected against UVA 2001, but deferred UVA testing and necessary on sunscreen and sunscreen- radiation. Thus, FDA held the 1994 labeling for future regulatory action. skin protectant combination drug public meeting to gather further FDA stated that more time was required information about an appropriate UVA to review comments from interested products. This proposed rule, therefore, test method and labeling. parties on active ingredients, labeling, proposes that the stays of both part 352 In the Federal Register of June 8, 1994 and test methods for products intended and § 347.20(d) be lifted when this rule (59 FR 29706), FDA proposed to amend to provide UVA protection. This is finalized. These stays will be the TFM (and reopened the comment proposed amendment to the FM for OTC maintained until a final rule based on period until August 22, 1994) to remove sunscreen drug products will complete this proposed rule becomes effective. five proposed sunscreen ingredients the FM by addressing both UVB and In the Federal Register of September from the TFM because of lack of interest UVA testing and labeling. 3, 2004 (69 FR 53801), FDA delayed the in establishing United States In the Federal Register of June 8, 2000 implementation date for OTC sunscreen Pharmacopeia—National Formulary (65 FR 36319), FDA reopened the drug products subject to the final rule (USP–NF) monographs. FDA also administrative record of the rulemaking that established standardized format reiterated that all sunscreen ingredients for OTC sunscreen drug products to and content requirements for the must have a USP–NF monograph before allow for specific comment on high SPF labeling of OTC drug products (i.e., being included in the FM for OTC and UVA radiation testing and labeling. Drug Facts rule). FDA explained that we sunscreen drug products. FDA also extended the effective date for postponed the Drug Facts In the Federal Register of August 15, the FM to December 31, 2002. implementation date because we did not 1996 (61 FR 42398), FDA reopened the In the Federal Register of December expect to complete the final amendment administrative record until December 6, 31, 2001 (66 FR 67485), FDA stayed the of the sunscreen monograph to include 1996, to allow additional submissions December 31, 2002, effective date of the on zinc oxide and titanium dioxide as FM for OTC sunscreen drug products in UVA testing and labeling by the Drug well as sunscreen photostability. FDA part 352 until we provided further Facts implementation date of May 16, also announced a public meeting for notice in a future issue of the Federal 2005 (64 FR 13254 at 13273 and 13274, September 19 and 20, 1996, to discuss Register. FDA took this action because March 17, 1999). Thus, FDA delayed the the safety and efficacy of these two we planned to amend part 352 to implementation date of the Drug Facts ingredients and photostability of address formulation, labeling, and rule with respect to OTC sunscreen drug sunscreens in general. testing requirements for both UVB and products until further notice to avoid In the Federal Registers of September UVA radiation protection. This issuing successive relabeling 16, 1996 (61 FR 48645) and October 22, document proposes such changes. This requirements for sunscreen drug 1998 (63 FR 56584), FDA amended the document also proposes an effective products at two closely related time TFM to add the UVA-absorbing date related to publication of an intervals, as required by the Drug Facts sunscreen ingredients avobenzone and amended FM (see section X of this rule and the final amendment to the zinc oxide to the proposed list of document). The existing stay of the sunscreen monograph. monograph ingredients. FDA also effective date for part 352 remains in II. Summary of Major Changes to the proposed indications for these effect at this time. ingredients. As a result of this In the Federal Register of June 20, FM amendment to the TFM, in the Federal 2002 (67 FR 41821), FDA published a In response to the TFM and FM, FDA Register of April 30, 1997 (62 FR technical amendment to change the received substantial data and 23350), FDA announced an enforcement names of four sunscreen active information regarding UVA and UVB policy allowing interim marketing of ingredients in § 352.10 of the active ingredients, claims, and testing OTC sunscreen drug products monograph to be consistent with name procedures, as well as on other issues containing avobenzone. changes that appeared in USP 24. The addressed in this document. FDA On November 21, 1997, Congress new names, which are simpler and more summarizes these issues and proposed enacted the Food and Drug convenient, are meradimate for menthyl changes to the FM in this section. Administration Modernization Act of anthranilate, octinoxate for octyl 1997 (FDAMA). Section 129 of FDAMA methoxycinnamate, octisalate for octyl A. Ingredients stated that ‘‘Not later than 18 months salicylate, and ensulizole for after the date of enactment of this Act, phenylbenzimidazole sulfonic acid. FDA proposes to add combinations of the Secretary of Health and Human Because the names became official on avobenzone with zinc oxide and Services shall issue regulations for over- March 1, 2001, manufacturers could avobenzone with ensulizole as the-counter sunscreen products for the begin using them at any time after that permitted combinations of active prevention or treatment of sunburn.’’ date. sunscreen ingredients in the FM (see FDA identified the UVB portions of the In the Federal Register of June 4, 2003 section III.C, comment 7 of this monograph (and related provisions on (68 FR 33362), FDA issued a final rule document).

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49072 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

B. UVB (SPF) Labeling OTC sunscreen products tested in § 352.52(f), which are not required to The FM allowed specific labeled SPF accordance with the proposed UVA include this statement under proposed values up to, but not exceeding, 30. OTC testing procedures in §§ 352.71 and § 352.52(f)(1)(v) and (f)(1)(vi) (see sunscreen drug products with SPF 352.72 (see section III.E, comment 14 section III.G, comment 19 of this values greater than 30 could be labeled and section III.N, comment 45 of this document). The statement in proposed with the collective term ‘‘30+.’’ In this document). § 352.52(c)(1) reads as follows: ‘‘UV exposure from the sun increases the risk amendment, FDA proposes to increase D. Indications of skin cancer, premature skin aging, the specific labeled SPF value to 50 and The FM allowed the following two and other skin damage. It is important revise the collective term to ‘‘50+.’’ FDA UVB indications in § 352.52(b)(1): to decrease UV exposure by limiting will consider higher specific labeled • ‘‘helps prevent sunburn’’ time in the sun, wearing protective SPF values upon receipt of adequate, • ‘‘higher SPF gives more sunburn clothing, and using a sunscreen.’’ FDA validated data (see section III.F, protection’’ proposes that the statement appear in comment 15 of this document). In this amendment, FDA proposes to bold type as the first statement in the In addition, FDA proposes to revise revise the first statement to read ‘‘low,’’ ‘‘Warnings’’ section. the following FM labeling: ‘‘medium,’’ ‘‘high,’’ or ‘‘highest’’ ‘‘UVB • The phrase ‘‘sun protection’’ to sunburn protection’’ in proposed F. Directions ‘‘sunburn protection’’ where used in § 352.52(b)(1)(i) through (b)(1)(iv). FDA FDA proposes changes to the §§ 352.3(b)(1), (b)(2), (b)(3), and (d) and is proposing to revise the additional directions to reduce the likelihood that 352.52(e)(1)(i), (e)(1)(ii), and (e)(1)(iii) indications in § 352.52(b)(2) to reflect OTC sunscreen drug products are (see section III.D, comment 10 of this the new PCD ranges in proposed underapplied. Section 352.52(d)(1)(i) document); and § 352.3(c) (e.g., SPF of 2 to under 12 currently provides manufacturers the • Section 352.50(a) to include the becomes SPF of 2 to under 15) and option to select one or more of the term ‘‘UVB’’ before the term ‘‘SPF’’ on create the new PCD range over SPF 50. following terms: ‘‘liberally,’’ the principal display panel (PDP), along These proposed revisions are based ‘‘generously,’’ ‘‘smoothly,’’ or ‘‘evenly.’’ with the product category designation upon the revised PCD categories in FDA is proposing to allow the choice of (PCD) (see section III.E, comment 14 of proposed § 352.3(c) (see section III.G, one of two required terms (i.e., this document). comment 16 of this document). FDA ‘‘liberally’’ or ‘‘generously’’) and to FDA also proposes to revise the PCD proposes that the second statement in include ‘‘evenly’’ as an additional SPF ranges in § 352.3(b)(1), (b)(2), and current § 352.52(b)(1) (‘‘higher SPF gives optional term. FDA is proposing to (b)(3) (proposed § 352.3(c)(1) through more sunburn protection’’) no longer be eliminate the term ‘‘smoothly’’ because (c)(4)) to reflect the following: required and proposes an additional it is vague. • The current standard public health indication regarding UVA protection FDA also proposes to add a new message concerning use of sunscreens, (see proposed § 352.52(b)(2)(v)). direction ‘‘apply and reapply as directed • The proposed increase of the In proposed § 352.52(b)(2)(v), FDA to avoid lowering protection’’ (proposed labeled SPF value to ‘‘50+,’’ and includes a new indication for UVA § 352.52(d)(1)(ii)). Because new • The proposed addition of the term protection that involves selection of the information demonstrates the ‘‘UVB’’ before the word ‘‘sunburn.’’ appropriate descriptor (‘‘low,’’ importance of sunscreen reapplication, Proposed § 352.3(c)(4) contains a new ‘‘medium,’’ ‘‘high,’’ or ‘‘highest’’) to FDA also proposes to make the optional PCD of ‘‘highest UVB sunburn describe the level of protection. In directions in paragraph (d)(2) a protection product’’ for products that proposed § 352.52(b)(2)(vi), FDA requirement. As a result of this change, provide an SPF value over 50. FDA includes a modified version of the FDA is proposing to remove the current further proposes to revise current sunburn ‘‘Uses’’ statement required by language in paragraph (d)(3) because it § 352.3(b)(1) and (b)(2) to replace the proposed § 352.52(b)(1)(i) through is no longer necessary. Instead, FDA is current category descriptors of (b)(1)(iv) when the additional statement proposing, in paragraph (d)(3), required ‘‘minimal’’ and ‘‘moderate’’ with the in proposed § 352.52(b)(2)(v) is used information for products that do not terms ‘‘low’’ and ‘‘medium,’’ and bears the same category descriptor satisfy the water resistant testing respectively. FDA considers the new as the SPF value (e.g., medium UVA/ procedures in § 352.76. FDA is also terms to be simpler and uniform with UVB protection from sunburn) (see proposing a required reapplication the proposed UVB and UVA ‘‘Uses’’ section III.G, comment 17 of this statement in § 352.52(d)(1)(ii). The statements. Proposed changes to PCDs document). reapplication information in current and category descriptors also occur in § 352.52(d)(2) appears in proposed proposed § 352.52(e)(1) (see section E. Warnings § 352.52(d)(2) and (d)(3) of this III.D, comment 13 and section III.G, FDA is proposing to shorten the document (see section III.H, comment comment 16 of this document). In warning in § 352.52(c)(1)(ii) (proposed 22 of this document). addition, FDA proposes optional UVB § 352.52(c)(3)) under the subheading radiation protection statements (see ‘‘Stop use and ask a doctor if’’ from G. UVB Testing proposed § 352.52(e)(2) and (e)(3)). ‘‘[bullet] rash or irritation develops and FDA is proposing to revise the SPF lasts’’ to ‘‘[bullet] skin rash occurs.’’ (UVB) testing procedure (see section III, C. UVA Labeling FDA proposes removing the optional paragraphs I through L of this FDA proposes new labeling to ‘‘sun alert’’ product performance document) and to move the SPF testing designate the level of UVA protection statement (current § 352.52(e)(2)) and procedure currently in §§ 352.70 on the PDP of OTC sunscreen drug requiring a revised ‘‘sun alert’’ through 352.73 to proposed § 352.70. products. FDA proposes the use of statement in the ‘‘Warnings’’ section FDA proposes a padimate O/ symbols (‘‘stars’’) in conjunction with a (proposed § 352.52(c)(1)). FDA proposes oxybenzone sunscreen standard in descriptor (i.e., ‘‘low,’’ ‘‘medium,’’ that this revised statement be required § 352.70 that will be required for testing ‘‘high,’’ or ‘‘highest’’). FDA also on all OTC sunscreen drug products sunscreen products with SPF values proposes to add new § 352.50(b) except lip cosmetic-drug and lip over 15. Manufacturers may use either specifying the required PDP labeling for protectant-sunscreen products subject to this padimate O/oxybenzone standard

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49073

or the homosalate standard to test to 3 years after publication of the FM as monographs now state the same products with SPF values of 2 to 15. appropriate or necessary for conditions for lawfully marketing these FDA proposes a high pressure liquid implementation. Several of these combination products. The existing chromatography (HPLC) method to comments added that the date should be language in §§ 347.20(d) and 352.20(b) replace the spectrophotometric method in the June/July time period because the would include the two new used to assay the homosalate and shipping season is practically over at combinations that FDA is proposing to padimate O/oxybenzone standards. that time and manufacturing for the next add to the sunscreen monograph (see FDA proposes the following season is just beginning. section II.A, comment 7 of this modifications to the SPF testing FDA understands the seasonal nature document). procedure: of the sunscreen industry and the time • required for product testing and B. Comments on Tanning and Tanning Specifications for the solar Preparations simulator in § 352.71 (proposed relabeling. FDA is also aware that more § 352.70(b)), than 1 year may be needed for (Comment 4) One comment requested • Instructions for the application of implementation. FDA is proposing an that the effective date of § 740.19 (21 test materials and response criteria in 18- to 24-month implementation date CFR 740.19) be extended to December § 352.72 (proposed § 352.70(c)), and and will try to have it coincide with the 31, 2002, consistent with the delay of • Doses and determination of June/July time period (see section XI of the effective date for § 310.545(a)(29) minimal erythema dose (MED) in this document). and (d)(31), part 352, and § 700.35 (65 § 352.73 (proposed § 352.70(d)). (Comment 2) One comment requested FR 36319). The comment stated that FDA proposes to continue requiring a that FDA and the Federal Trade singling out § 740.19 to become effective finger cot to be used in the application Commission (FTC) take steps to make earlier might constitute an arbitrary and of sunscreen standard and test product sure that sunscreen manufacturers capricious decision by FDA. as specified in § 352.72(e) (proposed provide information to the American The May 21, 1999, final rule set a 2- § 352.70(c)(5)). However, FDA now public to help them understand and use year effective date (May 21, 2001) for proposes that the finger cot be the Ultraviolet Index (UVI) to determine § 310.545(a)(29) and (d)(31), part 352, pretreated. These two proposed UVB their risk of sunburn. and § 700.35. In the Federal Register of testing changes also apply to UVA in The National Weather Service, the June 8, 2000 (65 FR 36319), FDA vivo testing. Environmental Protection Agency extended the effective date for (EPA), and the Centers for Disease compliance with § 310.545(a)(29) and H. UVA Testing Control and Prevention (CDC) (d)(31), part 352, and § 700.35 until FDA proposes a combination of developed the UVI, which has been in December 31, 2002, to provide time for spectrophotometric (in vitro) and use since 1995. This index is an completion of a more comprehensive clinical (in vivo) UVA test procedures in indication of the amount of UV UVA/UVB FM for OTC sunscreen drug proposed §§ 352.71 and 352.72, radiation reaching the surface of the products. On December 31, 2001, FDA respectively. To assure UVA protection earth as a function of ozone data, then stayed the effective date of part 352 for ‘‘water resistant’’ and ‘‘very water atmospheric pressure, temperature, and (but not § 310.545(a)(29) and (d)(31), resistant’’ sunscreen products, FDA cloudiness and is generated for 58 cities and § 700.35) until further notice (66 FR proposes that the in vivo UVA test be around the United States. 67485). FDA took this action because we Usage information required by the conducted after the appropriate water are amending part 352 to address OTC sunscreen drug product immersion period for OTC sunscreen formulation, labeling, and testing monograph applies regardless of the requirements for both UVA and UVB drug products making a UVA claim. UVI value. Therefore, FDA believes that radiation protection. The May 21, 1999, Therefore, FDA proposes modification UVI information need not be required in final rule also set a 1-year effective date of § 352.76 to state that the water the monograph for the safe and effective (May 22, 2000) for new § 740.19, which resistance claim applies to the SPF and, use of these products and should not be addresses a warning statement for if appropriate, UVA values determined included in the ‘‘Drug Facts’’ labeling. cosmetic suntanning preparations that after the appropriate water immersion However, manufacturers who wish to do do not contain a sunscreen active period as described in proposed so may voluntarily include such ingredient. These products are not § 352.70 and, if appropriate, proposed information in their labeling outside the subject to the monograph for OTC § 352.72. ‘‘Drug Facts’’ box. sunscreen drug products in part 352. III. FDA’s Tentative Conclusions on the (Comment 3) One comment requested FDA considered this warning to be Comments that FDA make clear, through either the sufficiently important for safety reasons FM for skin protectant or sunscreen when we issued the final rule (64 FR A. General Comments on OTC drug products, or both, that combination 27666 at 27669) to require a 12-month Sunscreen Drug Products products containing sunscreen and skin effective date as opposed to the 24- (Comment 1) Several comments asked protectant ingredients may be lawfully month effective date for the other that FDA provide more time to comply marketed. sections of the rule. Further, FDA’s with requirements of the FM in order to Section 347.20(d) of the skin primary reason for extending the avoid an adverse economic impact on protectant FM (21 CFR 347.20(d)), effective date of those other sections to the suncare industry and consumers. which published in the Federal Register December 31, 2002, and then staying The comments described the seasonal of June 4, 2003 (68 FR 33362), provides part 352 to address formulation, dynamics of the suncare industry (i.e., for combinations of sunscreen labeling, and testing requirements for products are sold in two marketing ingredients and specific skin protectant both UVA and UVB protection, was to cycles over a period of 18 months) and ingredients. The final rule for OTC skin allow FDA to develop a comprehensive stated that the industry would need protectant drug products also included UVB/UVA final monograph. This reason more time to develop products that meet an amendment to the sunscreen FM, does not apply to § 740.19. Accordingly, the FM requirements and allow for adding new § 352.20(b), which allows FDA did not extend the effective date shipment of the previous year’s returns. combinations of sunscreen and skin for § 740.19, and § 740.19 is in effect at The comments mentioned times from 2 protectant active ingredients. Thus, both this time. FDA concludes that this

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49074 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

decision is not arbitrary and capricious, • Oral and dermal toxicity studies, too few subjects were enrolled, and the but is based on valid health concerns • A chronic skin painting study subjects were not representative related to the products subject to the carcinogenicity study in mice, of the average sunscreen user. warning requirement in § 740.19. • Comedogenecity tests in rabbits, Well-controlled clinical trials with (Comment 5) One comment requested • Repeated insult patch test in DHA alone are lacking. Although some that FDA and FTC take steps to ensure humans, and investigations described by the sunscreen manufacturers inform • Photoallergy tests. comments suggest that DHA may help consumers that their natural skin Effectiveness studies consisted of protect the normal skin of psoriasis pigmentation provides protection from published articles using either humans patients, concerns remain about the sunlight. The comment stated that these or photosensitized rats. Another usefulness of DHA products in the OTC adaptive individuals might not require a comment discussed investigations with market. For example, one comment daily application of a sunscreen. DHA on psoriasis patients sensitized stated that photoprotection provided by Another comment submitted a copy of with 8-methoxypsoralen (8–MOP). DHA depends upon the way the product a patent for an electronic sensor device FDA is not proposing to include DHA polymerizes in the stratum corneum and to measure solar radiation. The in the monograph as a single active that polymerization depends on the skin comment stated that the personal device ingredient in OTC sunscreen products. of each individual. Therefore, the could alert consumers to their level of Although there were no product photoprotection provided by DHA UV exposure so they could either come submissions to the Advisory Review varies from person to person and has to out of the sun or apply a sunscreen to Panel on Topical Analgesic, be determined for each person by avoid sunburn and skin cancer. Antirheumatic, Otic, Burn, and Sunburn diffuse reflectance spectroscopy. Given FDA has no objection to sunscreen Prevention and Treatment Drug these statements, it is not clear how manufacturers informing consumers Products (the Panel) using DHA as a appropriate OTC drug product labeling that their natural skin pigmentation sunscreen ingredient, the Panel could be written to aid consumers in provides protection from sunlight. discussed available scientific evidence proper selection and use of a DHA However, FDA has no basis to require for DHA as a single sunscreen sunscreen. such information as part of the required ingredient. The Panel concluded that FDA concludes that current labeling for OTC sunscreen drug DHA is not a sunscreen but a cosmetic; information is inadequate to include products. Thus, manufacturers may it is a sunscreen only when used with DHA in the monograph as a single include this information in labeling lawsone (43 FR 38206 at 38215 to sunscreen ingredient. None of the outside of the ‘‘Drug Facts’’ box, but are 38216, August 25, 1978). Although one comments provided information to not required to include this information. comment stated that DHA alters the establish the appropriate number of FDA considers the comment regarding structure of the skin, it did not provide consecutive product applications and the UV measuring device to be outside data to support this claim. Thus, at this the timing of these applications (how far the scope of this rulemaking, which time, FDA agrees with the Panel that apart or how soon before sun exposure) evaluates the safety, effectiveness, and DHA is a cosmetic. that are necessary to achieve the desired labeling of OTC drug products. FDA acknowledges that DHA is the protection using products containing subject of an approved color additive various concentrations of DHA. In two C. Comments on Specific Sunscreen petition and its safety as a color additive Active Ingredients submitted studies, a preparation has been established. However, the containing 3 percent DHA was applied (Comment 6) Several comments submitted chronic (life-span) skin six times prior to sun exposure and a requested that dihydroxyacetone (DHA) painting study in mice does not support preparation containing 15 percent DHA be added to the monograph as a single the safe use of DHA as a sunscreen preparation was applied one time 24 active ingredient for UVA protection. because no group of mice was included hours prior to sun exposure, The comments claimed that DHA alone in the study to determine the possible respectively (Ref. 1). The comments did provides an SPF of 2 to 4. One comment photocarcinogenic effect of DHA. This not include any information on claimed that a 15 percent topical effect needs to be studied because DHA appropriate regimens for various skin solution of DHA provided a is associated with carbonyl compounds types, which is necessary because the photoprotective factor of 10 in the UVA known to react with pyrimidine bases in level of photoprotection provided by region. Other comments contended that the presence of UV radiation, and it DHA is dependent on skin type. the brown color produced by DHA, appears to be a potent inducer of Therefore, based upon this lack of resembling melanin, should potentiate thymine dimers, premutagenic information, it is not clear how to state the action of sunscreens. Another deoxyribonucleic acid (DNA) lesions. appropriate label directions for comment stated that DHA alone is not Therefore, its safety, in terms of the consumer use. FDA needs additional a sunscreen, but forms a sunscreen type, extent, and location of photo- information from clinical studies to when combined with lawsone. The induced DNA damage, is of concern and determine the effective concentration of comment cited unpublished should be determined. Whether DHA DHA in sunscreen product formulations observations by two independent contributes or promotes UV and the frequency and timing of product investigators that the melanoidins of carcinogenesis is not known. application. DHA-induced skin pigment resemble The submitted studies on the (Comment 7) One comment submitted melanin in that they absorb UVB effectiveness of DHA as a single UVA data to support the combination of strongly, with decreasing absorbance sunscreen ingredient add only avobenzone with ensulizole and through the UVA region and into visible qualitative information. Many of the avobenzone with zinc oxide (Ref. 2). light. The comment added that, because studies utilized animal models; few The safety data included the following: DHA alters the structure of the skin included human subjects. One study • A repeat insult patch test, surface, it is, by definition, a drug. involved only five subjects, three with • A phototoxicity study, and One comment provided information erythropoietic protoporphyria and two • A photoallergy study. on the safety and UVA effectiveness of with polymorphic light eruptions. The effectiveness data involved a DHA alone (Ref. 1). Safety studies Another study involved six subjects clinical study using the in vitro ‘‘critical included the following: sensitized with 8–MOP. In both studies, wavelength’’ (CW) method and the in

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49075

vivo ‘‘protection factor A’’ (PFA) appropriate for these products. Based on recognized that some product packages method to support the UVA radiation the labeling in § 352.52, the comment were too small to accommodate all of protection potential of the combination proposed eight additional modifications the required labeling. Therefore, under products. The PFA test data were from for all other OTC sunscreen products § 201.66(d)(10) (21 CFR 201.66(d)(10)), a double blind clinical study using five regardless of package size: FDA allows labeling format sunscreen formulations. 1. Delete ‘‘Drug Facts’’ title because it modifications for all OTC drug products The safety studies demonstrated that is inappropriate and unnecessary for sold in small packages. In the final rule the following combinations of active sunscreens. establishing ‘‘Drug Facts’’ labeling, FDA ingredients have a low potential for 2. Omit ‘‘Purpose’’ because it is also stated that we may allow reduced irritation, allergenic sensitization, and repetitive of the statement of identity on labeling requirements beyond those phototoxicity: the PDP and ‘‘Uses’’ information. specified under § 201.66(d)(10) for OTC • 3 percent or less avobenzone with 3. Revise ‘‘higher SPF gives more drug products that meet the criteria 2 percent ensulizole sunburn protection’’ in ‘‘Uses’’ to read listed in the preceding paragraph (see • 3 percent or less avobenzone with ‘‘higher SPF products give more sun section III.D, comment 9 of this 5 percent zinc oxide protection, but are not intended to document). The data further suggested that the extend the time spent in the sun,’’ and In the final rule for OTC sunscreen photoallergenic potential of avobenzone require this statement only on products drug products (64 FR 27666 at 27681 to is not augmented by its combination with an SPF value over 30. 27682), FDA recognized that some OTC with either ensulizole or zinc oxide. 4. Omit ‘‘For external use only’’ sunscreen drug products meet these The clinical study using the PFA in warning because it is self-evident for criteria for reduced labeling. vivo method demonstrated that the sunscreen products. Specifically, FDA identified OTC following combinations of active 5. Revise ‘‘When using this product sunscreen drug products that qualify for ingredients are significantly more [bullet] keep out of eyes. Rinse with the small package specifications in effective than 1.5 percent ensulizole or water to remove’’ to read ‘‘Keep out of § 201.66(d)(10) and are labeled for use 3 percent zinc oxide alone in protecting eyes.’’ only on specific small areas of the face against UVA radiation: 6. Revise ‘‘Stop use and ask a doctor as meeting the criteria for reduced • 3 percent avobenzone with 1.5 if [bullet] rash or irritation develops and labeling. Therefore, FDA allows content percent ensulizole lasts’’ to read ‘‘Stop use if skin rash and format modifications for these • 3 percent avobenzone with 4 occurs.’’ products under § 352.52(f). FDA allows percent zinc oxide 7. Omit barlines, hairlines, and box further modifications for lip products FDA’s detailed comments on the safety enclosure. containing sunscreen because these and effectiveness studies are on file in 8. Allow the option to list inactive products for small areas of the face are the Division of Dockets Management ingredients in a different location on the sold in even smaller packages than the (Ref. 3). label or in labeling accompanying the other sunscreen products marketed FDA considers the data submitted by product. under § 352.52(f) (68 FR 33362 at 33371; the comment sufficient to support the The comment stated that these 64 FR 13254 at 13270). FDA believes safety and effectiveness of avobenzone modifications would allow reduced that sunscreen products labeled for use with ensulizole and avobenzone with Drug Facts labeling for all OTC only on small areas of the face, zinc oxide when used in the sunscreen drug products. including lip products containing concentrations established for each The comment contended that sunscreen, serve an important public ingredient in § 352.10 of the sunscreen sunscreen products meet all of FDA’s health need and FDA does not want to monograph. Accordingly, FDA is criteria for reduced labeling (64 FR discourage manufacturers from proposing to amend § 352.20(a)(2) by 13254 at 13270): marketing these products (64 FR 13254 adding ensulizole and zinc oxide. • Packaged in small amounts, at 13270). Marketing of products containing • High therapeutic index, FDA does not find it appropriate to avobenzone with ensulizole and • Extremely low risk in actual extend the labeling modifications for avobenzone with zinc oxide will not be consumer use situations, OTC sunscreen drug products marketed permitted unless and until the following • A favorable public health benefit, under § 352.52(f) to all OTC sunscreen three actions occur: • No specified dosage limitation, and drug products. FDA disagrees with the 1. The comment period specific to • Few specific warnings and no comment’s argument that all sunscreen this proposal closes. general warnings (e.g., pregnancy or products meet the criteria for reduced 2. FDA has evaluated all comments on overdose warnings). Drug Facts labeling (64 FR 13254 at these combination products submitted The comment added that OTC 13270), because most sunscreen in response to the proposal. sunscreen products are a unique products are not sold in small packages. 3. FDA publishes a Federal Register category substantially different from Therefore, because sunscreen products notice announcing our determination to most other types of OTC drug products do not generally meet all of the criteria permit the marketing of OTC sunscreen because they are recommended for use for reduced Drug Facts labeling, FDA is drug products containing these on a daily basis to prevent serious not proposing reduced labeling for all combinations. disease. The comment concluded that OTC sunscreen products. FDA’s rationale for standardized FDA does not consider sunscreens as D. General Comments on the Labeling of labeling format and content a unique category substantially different Sunscreen Drug Products requirements does not necessarily from other types of OTC drug products (Comment 8) One comment agreed transfer to OTC sunscreen products and because they are recommended for use that the labeling modifications allowed specifically not to drug-cosmetic on a daily basis to prevent serious by the FM in § 352.52 for OTC products with a sunscreen. disease, as argued by the comment. sunscreen products marketed as a When FDA created the standardized Other OTC drug products are used on a lipstick or labeled for use only on labeling format and content daily basis, some to prevent serious specific small areas of the face (e.g., lips, requirements (i.e., ‘‘Drug Facts’’ disease and some for other reasons. For nose, ears, and/or around eyes) are labeling) for OTC drug products, we example, anticaries drug products are

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49076 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

used daily to prevent dental caries. remove.’’ This is useful information if a containers of 2 ounces (oz) or less (by Antiperspirant drug products can be sunscreen product gets into the eyes. weight or liquid measure). used daily to reduce underarm wetness. FDA agrees with part of the proposed The comment added that most facial FDA has concluded that these various shortened warning for OTC sunscreen makeup products are typically packaged products should generally be labeled drug products to ‘‘Stop use if skin rash in small containers. The comment stated using the standardized content and occurs’’ in place of ‘‘Stop use and ask that to meet any of FDA’s concerns that format in § 201.66. The standardized a doctor [bullet] if rash or irritation lotions and moisturizers sold in larger labeling allows consumers to more develops and lasts.’’ Therefore, FDA is packages may be used over the entire easily recognize that these products are, proposing to amend § 352.52(c)(1)(ii) body despite labeling that restricts use in fact, drug products and to more easily (proposed § 352.52(c)(3)) to state: ‘‘Stop to the face or hands, FDA could limit read and understand the labeling use and ask a doctor if [bullet] skin rash the flexible labeling to containers of 2 oz information. occurs.’’ or less. Furthermore, the comment The same principle applies when the FDA finds no reason to omit barlines, added that containers of 2 oz or less product is a drug cosmetic product (e.g., hairlines, or the box enclosure for all could not feasibly include the full OTC sunscreen moisturizer or antiperspirant OTC sunscreen drug products regardless drug labeling. deodorant). Consumers need to be of package size. These labeling formats The second comment contended that informed that the product has a drug help consumers identify a product as a the modified labeling in § 352.52(f) is effect, and the uniform Drug Facts drug and help make labeling particularly compelling for color labeling for all OTC drug and drug information easier to read and cosmetic products for the face that cosmetic products helps convey this understand. Thus, they should be contain sunscreens (i.e., ‘‘facial message. FDA applied this rationale included when package size allows. The makeups with sunscreen’’). The when it finalized the requirements in FM already allows horizontal barlines comment added that these products and the final rule that established § 201.66. and hairlines and the box enclosure to OTC sunscreen drug products for use FDA agrees that some OTC sunscreen be omitted if a small package meets the only on specific small areas of the face drug products meet the criteria for criteria in §§ 352.52(f) and have the same overall safety profile, reduced information for safe and 201.66(d)(10). and, therefore, FDA should allow these effective use (64 FR 13254 at 13270, 64 products to be labeled similarly. Finally, FDA has no basis to provide FR 27666 at 27681 to 27682). However, A third comment strongly disagreed an option for sunscreen products to list FDA disagrees with most of the with a specific labeling exemption for inactive ingredients in labeling that modifications proposed by the comment makeup with sunscreen and moisturizer accompanies the products. FDA for all package sizes of OTC sunscreen products for use on the face and hands. interprets section 502(e)(1)(A)(iii) of the products. FDA disagrees with deletion The comment contended that an act (21 U.S.C. 352(e)(1)(A)(iii)) as of the ‘‘Drug Facts’’ title and the exemption would not be in the best requiring the inactive ingredients to be ‘‘Purpose’’ information because many interest of consumers. The comment sunscreen products do not meet the listed on the outside container of a retail also argued that consumer confusion parameters for reduced Drug Facts package or on the immediate container and subsequent misuse of sunscreen labeling. if there is no outside container or products, particularly failure to apply FDA disagrees that the ‘‘Purpose’’ wrapper (§ 201.66(c)). Because this adequate amounts of sunscreen or to information is repetitive and, therefore, information, by law, must appear either reapply a product after certain activities, disagrees that it may be omitted where on the outside container or immediate will occur if FDA permits reduced there is sufficient labeling space. The container of the product, FDA does not labeling for these products. The ‘‘Purpose’’ section is a standard part of find a basis for allowing an option to list comment added that many consumers Drug Facts labeling and is intended to the inactive ingredients in a different use face and hand cosmetic products inform consumers which ingredients are location, such as other labeling with sunscreen as their primary and sunscreens in a product. This accompanying the product. In only source of UV radiation protection information is even more important accordance with § 201.66(c)(8), the for those areas of the body. Moreover, when a sunscreen is marketed in a inactive ingredients must be listed on consumers are more likely to use these combination product. For example, in a the product label in the ‘‘Drug Facts’’ products properly if they contain full sunscreen skin protectant drug product, box. sunscreen drug labeling. The comment the ‘‘Purpose’’ section informs (Comment 9) Two comments concluded that makeup foundations, consumers which ingredients are supported extending the labeling in tints, blushes, rouges, and moisturizers sunscreens and which are skin § 352.52(f) for products intended for use that are intended to be used on a daily protectants. only on specific small areas of the face or frequent basis to protect against the FDA has revised the ‘‘Uses’’ section and sold in small packages to all OTC adverse health and skin aging effects of and deleted the statement ‘‘higher SPF sunscreen products. The comments acute and chronic sun exposure must be gives more sunburn protection’’ (see contended that all OTC sunscreen drug labeled as drugs similar to other OTC section III.G, comment 16 of this products meet most of FDA’s criteria for sunscreen products. document). FDA disagrees with omitting products that require minimal FDA is not proposing to extend the the ‘‘For external use only’’ warning for information for safe and effective use labeling modifications in § 352.52(f), all OTC sunscreen drug products. FDA (64 FR 13254 at 13270) (see section which is specific for products used only finds no basis to exclude all OTC III.G, comment 8 of this document). on small areas of the face and sold in sunscreen products from this The first comment added that FDA small packages, to all OTC sunscreen requirement. Likewise, FDA finds no should permit the labeling products. FDA has determined that most reason to omit the two standard modifications in § 352.52(f) for the OTC sunscreen products should have subheadings that accompany the following products: full drug labeling information using the warning statements, as proposed by the • Makeup products (as defined in 21 standardized content and format in comment. Further, FDA disagrees with CFR 720.4(c)(7)) with sunscreen, and § 201.66 to ensure the safe and effective the comment’s suggestion to omit the • Lotions and moisturizers for the use of these products. In establishing statement ‘‘Rinse with water to hands or face with sunscreen in the labeling modifications in § 352.52(f),

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49077

FDA determined how the labeling for use only on specific small areas of and clarifying that the labeling information for sunscreen drug the face and that meet the criteria in modifications in § 352.52(f) apply to products, including drug cosmetic § 201.66(d)(10). However, FDA does not both sunscreen and makeup with products, could best be presented on agree that these labeling modifications sunscreen products. Furthermore, products with limited labeling space should apply to all makeup products because lip products with sunscreen and still provide consumers with identified in § 720.4(c) (21 CFR 720.4(c)) have substantially less labeling space adequate information to use these that contain sunscreen, because most than the nonlip products with products safely and effectively. are not sold in small packages and, sunscreen used only on specific small Although any sunscreen products sold therefore, do not meet all of the criteria areas of the face and sold in small in small packages that meet the criteria for reduced labeling (64 FR 13254 at packages, proposed § 352.52(f)(1)(vi) in § 201.66(d)(10) are allowed the format 13270). Thus, most of these products allows more labeling exemptions for lip exemptions under that section, FDA is can accommodate full Drug Facts products with sunscreen than other also proposing content exemptions for labeling, and FDA finds no reason to products that are within the scope of sunscreen products marketed under extend the labeling modifications in § 352.52(f). § 352.52(f). FDA is proposing these § 352.52(f) to all makeup with (Comment 10) Several comments exemptions under § 352.52(f) because sunscreens products. recommended changing the acronym sunscreen products labeled for use only As explained in the previous ‘‘SPF’’ from ‘‘sun protection factor’’ to on small areas of the face and sold in paragraph, the labeling modifications in ‘‘sunburn protection factor’’ because the small packages are generally sold in § 352.52(f) apply to makeup with latter definition is more descriptive of packages substantially smaller than sunscreen products labeled for use only the use of OTC sunscreen drug products other sunscreen products, even those on specific small areas of the face and and avoids giving consumers the sunscreen products labeled for other sold in small packages. FDA also impression of solar invincibility and a uses that meet the criteria in believes that any sunscreen products false sense of security. § 201.66(d)(10). that are used only on specific small FDA agrees. In § 352.52(b) of the FDA continues to believe that areas of the face and sold in small sunscreen FM, FDA included only requiring full Drug Facts labeling on packages meet FDA’s reduced labeling indications for sunburn protection (e.g., sunscreen products used only on criteria regardless of whether they are ‘‘helps prevent sunburn’’) (64 FR 27666 specific small areas of the face and sold drug or drug-cosmetic products. at 27691). In this document, FDA is in small packages (i.e., § 352.52(f)) Therefore, FDA is proposing to amend proposing to change the word ‘‘sun’’ to would discourage manufacturers from the heading of § 352.52(f) to read as ‘‘sunburn’’ in § 352.3(b)(1), (b)(2), (b)(3), marketing some of these products for follows: ‘‘Products, including cosmetic- and (d) and § 352.52(e)(1)(i), (e)(1)(ii), drug use. Many of these products, such drug products, containing any and (e)(1)(iii). as sunscreen-lip protectant products, are ingredient identified in § 352.10 labeled Manufacturers can continue to use sold in extremely small packages that for use only on specific small areas of existing labeling until the compliance cannot accommodate the required the face (e.g., lips, nose, ears, and/or dates of a final rule based on this labeling even with the format around the eyes) and that meet the proposal. However, FDA encourages exemptions allowed under criteria established in § 201.66(d)(10) of § 201.66(d)(10). As explained in a manufacturers to revise any labeling this chapter.’’ that states ‘‘sun protection’’ attributed to number of rulemakings (64 FR 27666 at In addition, FDA is proposing to sunscreen active ingredient(s) to the 27681 to 27682; 68 FR 33362 at 33371; extend the labeling exemptions, with new term ‘‘sunburn protection’’ as early 64 FR 13254 at 13270), these products some modifications, currently allowed as possible. meet the criteria for additional reduced for lipsticks in § 352.52(f)(1)(vi) to the labeling. Removal of these products following lip products with sunscreen, (Comment 11) Some comments from the OTC market would have a as defined in § 720.4(c): questioned the constitutionality of the negative impact on public health. FDA • Lipsticks, FM’s labeling provisions. Specifically, believes that the benefit of UV radiation • Lip products to prolong wear of the comments contended that the FM’s protection provided by these products lipstick, prohibition on the labeling of SPF outweighs the need for manufacturers to • Lip gloss, and products over 30, its restrictions on skin include all sunscreen labeling • Lip balm. aging claims, and its limitation of the information. In contrast, FDA believes FDA has identified lip products to indications for use for OTC sunscreen manufacturers of sunscreen products prolong wear of lipstick as ‘‘makeup drug products all violate the first that are not within the scope of fixatives’’ under § 720.4(c)(7)(viii). Lip amendment to the U.S. Constitution. § 352.52(f) will continue to market their gloss and lip balm fall under ‘‘other The comments asserted that these bans products even though full Drug Facts makeup preparations’’ in on allegedly truthful labeling in the FM labeling is required. Unlike sunscreen § 720.4(c)(7)(ix). As long as these lip go well beyond constitutionally products that meet § 352.52(f), the products with sunscreen are used only permissible restrictions on commercial package size of products that do not on specific small areas of the face and free speech. meet § 352.52(f) will accommodate full are sold in small packages (i.e., meet the One comment contended that FDA Drug Facts labeling. criteria in § 201.66(d)(10)), they would had failed to meet its burden to Although FDA is not extending the meet FDA’s reduced labeling criteria. As demonstrate that the claims at issue are labeling modifications in § 352.52(f) to discussed earlier in this comment, FDA misleading or that the restrictions on all OTC sunscreen products, as believes not allowing Drug Facts speech directly advance any substantial requested by the first and second labeling exemptions for these products governmental purpose. In addition, the comments, we are allowing these would discourage manufacturers from comment claimed that any interest FDA labeling modifications for certain marketing some of these products for has asserted in restricting the speech at makeup with sunscreen products. drug use. In proposed § 352.52(f)(1)(vi), issue is served equally well, if not Specifically, these labeling FDA is proposing to extend the labeling better, by regulations that do not restrict modifications would apply to makeup modifications for lipsticks to other lip speech to the same extent as FDA’s with sunscreen products that are labeled cosmetic products containing sunscreen regulations.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49078 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

FDA disagrees with the comments for addresses, in part, the comments’ promote directly the proper use of the following reasons. OTC drug request that FDA allow specific labeled sunscreens, which, in turn, would better monographs establish conditions under SPF values over 30. ensure the protection of public health. which ingredients for certain OTC uses Elsewhere in this document, FDA In addition, it would not be ‘‘unduly are generally recognized as safe and explains the reasons for the specific burdensome’’ to sunscreen effective (GRASE) and are not labeling proposals, such as the required manufacturers to require these labeling misbranded. General recognition of SPF labeling, revised ‘‘sun alert’’ in the disclosures. Finally, it is important to safety and effectiveness in an OTC drug ‘‘Warnings’’ section of the Drug Facts note that a sunscreen manufacturer monograph means that experts qualified box, and indications for use (see section could pursue alternative labeling by scientific training and experience III.F, comment 15 and section III.G, conditions for its product by filing an recognize the conditions as safe and comments 16, 17, and 19 of this NDA with the appropriate evidence effective for OTC marketing for the use document). FDA also explains our demonstrating the product’s safety and recommended or suggested in the denial of specific labeling claims effectiveness under the proposed product’s labeling. An OTC drug suggested by the comments, including conditions. monograph establishes, among other the prohibition on specific SPF values In any event, FDA believes that the things, specific indications that are over a certain threshold (SPF 50), skin labeling requirements outlined in this appropriate for the safe and effective use aging claims, and additional indications proposed rule would pass muster when of a drug. An OTC drug product with for use (see section III.F, comments 15 analyzed under the four-part test for labeled indications different than those and 17 of this document). As noted restrictions on commercial speech set set forth in an applicable OTC drug earlier in this comment, any variation fourth by the Supreme Court in Central monograph would not be considered from these labeling conditions in the Hudson Gas & Electric Corporation v. GRASE. monograph, if finalized, would cause an Public Service Commission, 447 U.S. OTC drug monographs allow OTC sunscreen drug product to be a 557 (1980). Under the test, the first manufacturers to market those products new drug requiring an approved NDA question is whether the commercial satisfying the monograph standard before it could be legally marketed in speech at issue is false, misleading, or without requiring the specific approval the United States. concerns unlawful activity, because of the product by means of a new drug The labeling requirements in this such speech is beyond the first application (NDA) under section 505 of proposed rule would not violate the first amendment’s protection and may be the act. FDA has issued numerous OTC amendment. FDA’s requirements for the prohibited. If the speech is truthful, drug monographs for certain categories disclosure of information in the labeling nonmisleading, and concerns lawful of OTC drug products. If an OTC drug of OTC sunscreen drug products are activity, the Government may product subject to a final monograph is constitutionally permissible because nonetheless regulate it if the labeled for indications that differ from they are reasonably related to the government interest asserted to justify those set forth in the monograph, then Government’s interest in promoting the the regulation is substantial, the it would be a ‘‘new drug’’ under section health, safety, and welfare of consumers regulation directly advances the 201(p) of the act. In order to be legally and because they are not an ‘‘unjustified asserted governmental interest, and the marketed and distributed in interstate or unduly burdensome’’ disclosure regulation is no more extensive than commerce, the drug manufacturer requirement that offends the first necessary to serve the government would be required to obtain approval amendment (see Zauderer v. Office of interest (Id. at 566). The Supreme Court from FDA for that product, and those Disciplinary Counsel, 471 U.S. 626, 651 has explained that the last element of conditions varying from the monograph, (1985); see also Ibanez v. Florida Dep’t the test is not a ‘‘least restrictive means’’ in an NDA under section 505 of the act. of Bus. and Prof’l Regulation, 512 U.S. requirement but, rather, requires narrow All OTC drug monographs place 136, 146 (1994)). The reasonable tailoring (i.e., ‘‘a fit that is not limits on the conditions that have been relationship between the required necessarily perfect, but reasonable’’ found acceptable for inclusion in the labeling disclosures proposed herein between means and ends) (Board of monograph by an administrative and the Government’s interest is plain Trustees of the State Univ. of N.Y. v. rulemaking process based on scientific here. Fox, 109 S.Ct. 3028, 3032–35 (1989)). In data. Here, FDA set certain limits on the The proposed labeling disclosures subsequent decisions, the Court has also labeling of sunscreen drug products in addressed by the comments, such as the clarified that ‘‘misleading’’ in the first the final rule, such as the prohibition on SPF value, indications for use, and element of the test refers to speech that specific SPF values over 30, certain skin revised ‘‘sun alert,’’ would contribute is inherently or actually misleading. aging claims, and other indications for directly to the safe and effective use of Thus, if the speech to be regulated use. FDA is maintaining similar labeling OTC sunscreen drug products. The SPF concerns lawful activity and is not restrictions in this proposed rule with value and indications for use are critical inherently or actually misleading, the respect to skin aging claims and other components of labeling that allow remainder of the test applies (see In re indications proposed by the comments. consumers to understand more clearly a R.M.J., 455 U.S. 191, 203 (1982)). Also, as described elsewhere in this sunscreen product’s use in preventing Based on the data currently available, document, the revised ‘‘sun alert’’ in the sunburn and relative level of UVA/UVB FDA believes that the labeling ‘‘Warnings’’ section does not include protection. As explained elsewhere in statements proposed by the comments any skin aging claims (see section III.G, this document, the revised ‘‘sun alert’’ (i.e., specific SPF values above FDA’s comment 19 of this document). we propose to require in the established threshold, skin aging claims, However, FDA is proposing to increase ‘‘Warnings’’ section would help and certain other indications) would not the SPF labeling limit from 30 to 50, consumers understand more clearly the be protected speech and may be based on additional data that was role of sunscreens as part of a prohibited under the first prong of the submitted subsequent to the issuance of comprehensive sun protection program Central Hudson test. FDA has the FM. FDA is also proposing that the (see section III.F, comment 19 of this tentatively determined that these term ‘‘SPF 50+’’ can be used, rather than document). The greater consumer proposed labeling statements would be the term ‘‘SPF 30+’’ allowed in the FM. understanding resulting from all of inherently misleading on OTC This increase in the SPF labeling limit these labeling conditions would sunscreen products sold and, thus,

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49079

misbrand the products under section SPF values above 50, there is no these short-term adverse effects of UV 502(a) and 201(n) of the act. Because assurance that the specific values radiation to a long-term, chronic effect FDA believes these labeling statements themselves are in fact truthful and not such as skin cancer. In addition, like are inherently misleading, they would misleading. Thus, labeling of specific skin aging, the complete action not be subject to protection under the values above SPF 50 without spectrum for skin cancer is not known first prong of the Central Hudson test. appropriate validation (which FDA at this time. With respect to the labeling currently lacks) would be inherently For all these reasons, FDA has limitations for SPF values, based on misleading. As noted elsewhere, FDA tentatively concluded that the available current data, FDA believes that the invited any interested parties to submit evidence fails to show that sunscreen labeling of sunscreens with specific SPF such validation data for consideration use alone helps prevents skin cancer or values greater than 50 would be by FDA and possible inclusion of premature skin aging. Thus, the anti- inherently misleading. As discussed specific values above SPF 50 in the FM. aging, skin cancer, and sun damage elsewhere in this document, FDA is With respect to anti-aging, skin claims proposed by the comments concerned with the accuracy and cancer, and sun damage claims would be false or misleading due to lack reproducibility of test results showing proposed by the comments, as discussed of sufficient data in support of these protection greater than SPF 50 due to elsewhere in this proposed rule, FDA is claims. For example, the statement the lack of adequate validation data (see concerned that these statements would proposed by one comment that section III.F, comment 15 of this be false or misleading due to lack of sunscreen use ‘‘may help prevent sun- document). FDA had the same concern sufficient data in support of these induced skin damage, such as with SPF values above 30 when we claims (see section III.F, comment 17 of premature skin aging’’ would be published the FM in 1999. At that time, this document). FDA has reviewed the inherently misleading to consumers by FDA had only received data submitted articles concerning UV- suggesting that sunscreen use alone may demonstrating that the SPF test induced skin damage (i.e., premature help prevent premature skin aging. As produces accurate results for products aging and cancer) along with the articles explained in this response, the available with SPF values of 30 or less. Since obtained from a search of scientific data fail to show that sunscreen use publication of the FM, FDA has received literature (Refs. 26 through 34). As alone helps prevent premature skin additional SPF testing data for discussed elsewhere, although FDA has aging and skin cancer. sunscreen products with SPF values concluded that the studies support the As described elsewhere, FDA is between 30 and 50 (Ref. 13). However, conclusion that exposure to UV rays proposing a revised ‘‘sun alert’’ so that FDA has not received any data for increase the risk of premature skin the labeling of OTC sunscreen drug sunscreen products with SPF values aging, the study data fails to show that products include the most accurate greater than 50. The data submitted to sunscreen use alone helps prevent information, based on the available FDA indicate that the SPF test is premature skin aging and skin cancer scientific evidence, concerning the accurate and reproducible for sunscreen for several reasons (see section III. F., relationship of sunscreen use to the products with SPF values up to 50 (Ref. comment 17 of this document). prevention of sunburn, skin cancer, and 13). However, these data cannot be First, with respect to premature skin premature skin aging caused by UV extrapolated to SPF values above 50. aging, the studies have not completely exposure (see section III.F, comment 19 Thus, FDA is proposing to allow defined the action spectrum for the of this document). The revised ‘‘sun specific labeled SPF values only up to majority of UV radiation-induced effects alert’’ also includes a statement about 50. on human skin. Second, the inability to limiting sun exposure and wearing Increasing variability in test results is identify the exact UVB and UVA protective clothing because FDA has likely with increasing SPF values. If wavelengths that induce each tentatively determined that it is critical there is large variability in test results, histological change in skin derives from for consumers to understand the role of then the SPF value determined from the the study designs. Without knowing sunscreen use in a comprehensive sun test is not accurate (i.e., an SPF 60 which UVB and UVA wavelengths protection program. As FDA has product may not actually be an SPF 60 induce each histological change in the explained, the available evidence product). The submitted data skin, FDA is unable to determine which strongly suggests that consumers rely demonstrated that variability is not an wavelengths are most important to more heavily on sunscreens alone issue for sunscreen products with SPF causing skin aging and cannot without taking other protective values up to 50. However, FDA is determine the action spectrum for aging. measures against sunlight, particularly concerned that variability will become Third, the studies did not examine the when the labeling of products indicates an issue for sunscreen products with chronic, long-term consequences of UV the potential for greater protection (see SPF values over 50. radiation exposure in human skin. section III.F, comment 19 of this For those sunscreens with SPF values Fourth, although the studies that document). By indicating the potential above 50, FDA is proposing that the examined the ability of sunscreens to for greater protection than is supported labeling can denote such values by a protect against UV radiation-induced by the available evidence, the proposed ‘‘50+’’ designation. As discussed histological changes in the skin provide anti-aging, skin cancer, and other elsewhere in this document, FDA has useful data, it is difficult for FDA to related claims would mislead sufficient assurance that a result over 50 conclude that sunscreen use alone helps consumers into relying more heavily on from the required SPF test is, in fact, prevent skin aging based on these sunscreens alone. Such excessive greater than 50 and can be labeled studies. reliance would undermine consumers’ ‘‘50+’’ (see section III. F, comment 15 of Likewise, FDA is not aware of data protection from the sun and, thus, this document). Thus, FDA believes that demonstrating that sunscreen use alone FDA’s public health mission. the term ‘‘50+’’ is truthful and helps prevent skin cancer. Like skin FDA has also preliminarily nonmisleading on the label of OTC aging, these are studies examining the determined that the proposed labeling sunscreen drug products for which the effects of sunscreen drug products on statements would concern unlawful SPF test in the monograph has indicated short-term factors for skin cancer, such activity which are not protected speech an SPF value greater than 50. However, as sunburn and other cellular damage. under the first prong of the Central without proper validation of specific However, it is difficult to extrapolate Hudson test.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49080 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

FDA is proposing specific conditions activity would obviate the need for FDA Consumers are accustomed to using the in the monograph under which OTC to address the other three prongs of the label as their primary source of sunscreen drug products would be Central Hudson test, we believe that the information about a drug product’s GRASE. Elsewhere, FDA explains how labeling requirements proposed in this contents and use. Neither a public the labeling statements proposed by the document would satisfy each of the education campaign, nor encouraging comments would not be appropriate parts of this test. With respect to the OTC drug product manufacturers to monograph indications for these second prong, FDA’s interest in the provide information, such as that in the sunscreen products (see section III.G, required labeling disclosures and proposed revised ‘‘sun alert,’’ to comment 17 of this document). Thus, prohibitions addressed by the comments consumers by other means, would the proposed labeling statements would contribute directly to the safe ensure that people have the information outside the proposed indications of the and effective use of these OTC they need about sunscreen products at final monograph, as FDA proposes to sunscreen drug products, which is the point of sale or use. Likewise, with revise it, would promote a sunscreen critical for the protection of public respect to the alternative labeling drug product for use as an unapproved health. FDA’s interest in protecting the statements proposed by the comments, new drug, which is illegal. In addition, public health has been previously FDA’s proposed indications and revised any variation in the statements in a upheld as a substantial government ‘‘sun alert’’ present the relevant public ‘‘Warnings’’ section of a final interest under Central Hudson (see health information to consumers in the monograph, such as the revised ‘‘sun Pearson v. Shalala, 164 F.3d 650, 656 clearest and most direct manner. Thus, alert’’ statement in this proposed rule, (D.C. Cir. 1999) (citing Rubin v. Coors FDA’s proposed indications and would be outside the monograph Brewing Co., 514 U.S. 476, 484–485 prohibition of other labeling statements conditions and, thus, would promote (1995)). are not more extensive than necessary. the product as an unapproved new drug. The proposed labeling requirements In this way, the required labeling The marketing and distribution in would directly advance this interest, disclosures and prohibitions proposed interstate commerce of an OTC thereby satisfying the third prong of the in this document would meet the fourth sunscreen drug product with such Central Hudson test. By requiring prong of the test. labeling variations would be prohibited labeling disclosure of the SPF value, the Furthermore, the proposed under sections 301(d) and 505(a) of the proposed revised ‘‘sun alert,’’ and prohibition of claims in a final act. Speech promoting such an illegal indications for use, FDA can better monograph does not prevent such activity may be restricted without assure that consumers understand more claims from being approved in an NDA. violating the first amendment (Central clearly the use of sunscreens in As explained previously, a final Hudson, 447 U.S. at 563–564). preventing sunburn, their relative UVA/ monograph sets forth those conditions, If a manufacturer could circumvent UVB protection, and their role as part of including labeling, under which an OTC the requirements and restrictions a comprehensive sun protection drug product would be considered imposed by a final monograph by program. The greater consumer GRASE and not misbranded. In issuing including nonmonograph labeling understanding resulting from all of monographs, FDA considers whether statements, or excluding required these labeling conditions would the available scientific evidence monograph statements, based on its own promote directly the proper use of demonstrates that OTC drug products assertions of the alleged appropriateness sunscreens, which, in turn, would better within a therapeutic category are and truthfulness of the statements, then ensure the protection of the public GRASE. A final monograph does not such activity would significantly health. constitute an FDA decision regarding an undermine the monograph system and Likewise, this proposed rule’s NDA for an OTC drug proposing FDA’s assurance that OTC drugs are safe exclusion from the monograph of the variations in these conditions. Thus, and effective for their labeled labeling statements proposed by the FDA’s proposals in this document conditions. FDA has assessed the comments also directly advances FDA’s would not prohibit any interested labeling statements proposed by the public health interest. FDA has manufacturer from filing an NDA, with comments and preliminarily determined preliminarily determined from the the appropriate evidence, for any that they are not justified by the available evidence that these statements variations from the monograph labeling available scientific evidence as GRASE would not be appropriate conditions for conditions. Because of this significant conditions for the monograph. Instead, OTC use under the monograph. Thus, available option to manufacturers for in order to legally market a sunscreen the statements would directly proposing alternative labeling drug product with such labeling undermine the protection of public statements, FDA’s proposed labeling statements, an interested manufacturer health. In addition, it is important to requirements and prohibitions are not would have to submit an NDA to FDA note that the Pearson court, in assessing more extensive than necessary. with the appropriate evidence to show whether the specific dietary supplement In conclusion, FDA believes it has the safety and effectiveness of the drug regulations at issue directly advanced complied with its burdens under the under the proposed nonmonograph FDA’s stated public health goals under first amendment to support the labeling labeling conditions. Requiring the third prong of the Central Hudson requirements of this proposed rule. premarket FDA review and test, explained that its findings under (Comment 12) One comment stated authorization of such nonmonograph this prong did not apply to drugs, where that voluntary professional labeling can drug claims ensures that such claims ‘‘the potential harm is presumably much be provided to physicians that will will be evaluated by a public health greater’’ than other products (Pearson, allow them to select or recommend agency that has scientific and medical 164 F. 3d at 656, n 13). sunscreen products for their patients’ expertise so that only products that are Finally, under the fourth prong of the needs, based on more detailed safe and effective will be permitted to be Central Hudson test, there are not information describing the quantity sold for therapeutic purposes. numerous and obvious (Cincinnati v. (protection factor) and the range of UV Although this preliminary- Discovery Network, 507 U.S. 410, 418 n. protection (e.g., UVB, UVA, or UVB/ determination that the labeling 13 (1993)) alternatives to the required UVA protection). Another comment statements at issue would be inherently labeling statements or labeling stated that FDA should not require misleading and would concern unlawful prohibitions proposed herein. professional labeling because complete

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49081

and accurate product labeling should be FDA disagrees with the comment This comment referred to The Japan available to all consumers, not just to contending that manufacturers will only Cosmetic Industry Association (JCIA) their health care providers. produce the minimum SPF value in a Measurement Standards for UVA FDA defines professional labeling in given PCD range because they can use Protection Efficacy (Ref. 6), which OTC drug monographs as labeling that less active ingredients and get the same recommend labeling UVA protection as is provided to health professionals but PCD classification. Section 352.50 of the three grades: (1) PA+, (2) PA++, or (3) not to the general public (i.e., not current FM requires the SPF value to PA+++. directly to consumers) (for example, see appear on a sunscreen product’s PDP. Several comments recommended two § 331.80 (21 CFR 331.80)). In the final This proposed rule would not change categories of UV protection labeling rule, FDA stated that it would consider that requirement. Thus, while the PCD based on the ratio of UVA radiation professional labeling, such as protection provides additional information about protection factor to SPF value: against photosensitization reactions, if the SPF value, consumers seeking • ‘‘with UV protection’’ if ratio equals data were received (64 FR 22666 at higher SPF values can readily identify 0.20 27674). FDA has not received any data such products by the SPF value stated • ‘‘with extra UV protection’’ if ratio to date. Therefore, FDA is not proposing on a sunscreen product’s PDP. equals 0.25 any professional labeling in this The proposed ratio is based on the UVA document. FDA will consider E. Comments on the Labeling of radiation protection factor as professional labeling for OTC sunscreen Sunscreen Drug Products With UVA determined by the persistent pigment drug products in the future if specific Protection darkening (PPD) test method (see supportive data are provided. (Comment 14) Many comments section III.N, comment 46 of this (Comment 13) Some comments discussed ways to categorize, phrase, document). These comments stated that, objected to the ranges of SPF values that and display UVA/UVB radiation because the ratio of damage from solar define the product category designations protection on an OTC sunscreen drug UVB radiation to that of solar UVA (PCDs) in § 352.3(b). Stating that product label. All of the comments radiation is 80:20 over a day, a standard public health messages stated that the SPF value should retain sunscreen must protect against an 80:20 recommend use of a sunscreen with at preeminence on the label’s PDP and be ratio of UVB to UVA radiation. The least an SPF of 15, the comments the consumers’ criteria for choosing an comments also recommended that contended that the ‘‘moderate’’ PCD OTC sunscreen product. Some products labeled ‘‘with UV protection’’ (SPF values of 12 to under 30) may comments recommended that UVA or ‘‘with extra UV protection’’ exhibit cause consumers to believe that SPF radiation protection be stated on the absorbance of 360 nanometers (nm) and values of less than 15 provide adequate PDP in descriptive words or simple longer wavelengths. protection. One comment further stated phrases, rather than numbers or Another comment suggested two that if the PCD range is from SPF 12 to symbols, for the following reasons: categories to state overall UV radiation 29, manufacturers will only produce the • Simplicity, protection: ‘‘regular’’ and ‘‘broad minimum SPF value as they can use less • Clarity, spectrum.’’ The comment proposed that active ingredients and get the same PCD • To avoid confusion with SPF, and the ratio of a sunscreen product’s SPF • classification. To maximize consumer value to its UVA protection factor be the As discussed in the final rule (64 FR comprehension. single criterion for the ‘‘broad 27666 at 27681), the PCD ranges in Some comments referenced consumer spectrum’’ designation, with the § 352.3(b) and § 352.52(e) reflect a research, discussed in subsequent maximum ratio no greater than 4:1. For modified, simpler, combined version of paragraphs, to support this example, an SPF 16 product would need the previously proposed five PCDs and recommendation (Refs. 4 and 5). to provide a UVA protection factor of at the ‘‘Recommended Product Guide.’’ One comment suggested the following least 4 to be designated ‘‘broad However, FDA agrees with the labeling statements: spectrum.’’ • ‘‘Protects against UVA rays’’ comments that the current standard • One comment disagreed with the public health message from public ‘‘screens out UVA rays’’ previous comment, stating that there is • ‘‘shields from UVA rays’’ health organizations generally • ‘‘broad spectrum sunscreen’’ no supportable scientific basis for the recommends use of a sunscreen with an • ‘‘UVA/UVB protection’’ relevance of the 4:1 ratio. The comment SPF value of at least 15 (see section • ‘‘provides protection against both argued that the ratio inappropriately III.G, comment 19 of this document). We UVB and UVA rays’’ combines, in the same equation, SPF also agree that allowing SPF values • other truthful and nonmisleading values obtained with a solar simulator below 15 in any but the lowest PCD statements describing a quantification of and solar irradiance values at low sun range may appear to contradict this the product’s UVA radiation protection angles. message. Therefore, FDA is proposing to The comment stated that quantification Another comment suggested that modify the PCD SPF value range in of the UVA radiation protection should sunscreen products with an SPF value proposed § 352.3(c)(1) from ‘‘2 to under be allowed in labeling, but not required, of 2 or greater must have a UVA 12’’ to ‘‘2 to under 15’’ and in proposed so that consumers can have additional protection factor of at least 2 to be § 352.3(c)(2) from ‘‘12 to under 30’’ to product performance information to labeled ‘‘UVA/UVB’’ or ‘‘broad ‘‘15 to under 30.’’ FDA is also proposing help them select appropriate products. spectrum protection.’’ The comment to replace the PCD terms ‘‘minimal’’ and Another comment stated that UVA stated that products with SPF values of ‘‘moderate’’ with the simpler terms radiation protection should be labeled at least 15 and UVA protection factors ‘‘low’’ and ‘‘medium,’’ respectively, and only as grades of effectiveness (multiple of at least 4 may be labeled ‘‘extra (or to use these simpler terms for the UVA levels) for the following reasons: extended or enhanced) UVA radiation protection categories (see • UVA radiation irritation induces protection.’’ The comment stated that section III.E, comment 14 of this various skin reactions (e.g., erythema, these criteria are independent of test document). These labeling changes will pigment darkening, skin cancer, and method and should apply to any of the provide consumers with familiar and photodermatitis), and proposed UVA radiation test methods. consistent terms describing both UVA • Some action spectra of damages Another comment proposed and UVB radiation protection. have not been determined. establishing PCDs based on the UVA

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49082 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

radiation protection value obtained by and ease of product selection (Ref. 7). • 3-level descriptive (labeled blank if the PPD test method. The comment Other comments provided consumer no UVA radiation protection provided suggested four PCDs that would enable research data, discussed elsewhere in or labeled ‘‘UVA and UVB Protection’’ consumers to choose the desired levels this comment, suggesting this approach or ‘‘UVB Plus Extended UVA of protection: was least preferred by consumers (Refs. Protection,’’ depending on the level of • ‘‘moderate’’ 4 and 8). UVA radiation protection provided). • ‘‘high’’ One comment stated that UVA The studies included focus group • ‘‘very high’’ radiation protection claims should be discussions and indepth interviews. The • ‘‘extra’’ allowed for sunscreen products with comment stated that the data suggested Another comment recommended three SPF values of 4 and higher. The that a numeric designation for UVA PCDs: comment added that, for products radiation protection (in addition to the • ‘‘low UVA protection’’ claiming to protect against UVA and SPF value) created confusion for • ‘‘moderate UVA protection’’ • UVB radiation, a minimum UVA consumers and that symbols (i.e., stars) ‘‘maximum UVA protection’’ protection factor of 2 should be required misled consumers into giving equal or Another comment suggested using the if the SPF value is less than or equal to greater importance to the UVA radiation five PCDs proposed in the TFM (58 FR 12. rating compared to the SPF value. The 28194 at 28295) and added a UVA Several comments stated that comment concluded that a descriptive protection factor number for each PCD sunscreen drug products labeled as ‘‘full approach better conveyed to consumers based on the immediate pigment spectrum’’ or ‘‘broad spectrum’’ should the added benefit of UVA protection darkening (IPD) test method. protect consumers from substantially all without detracting from the SPF value. Two comments recommended a four- of the harmful effects of the sun, Another comment described two star rating system to describe UVA including sunburn associated with UVA consumer research studies conducted in radiation protection. The comments radiation. According to one comment, 1999 (Ref. 7) in which 2,238 consumers stated that this system, based on the sunscreen drug products labeled ‘‘full assessed three sunscreen product ratio of UVA to UVB radiation spectrum’’ or ‘‘broad spectrum’’ that do labeling systems: absorbance, would provide a simple not protect against nearly all UVB and • A pass/fail descriptive (labeled method for consumers to determine the UVA radiation wavelengths seriously blank if no UVA protection provided protective nature of an OTC sunscreen risk misleading consumers into (i.e., fails) or labeled ‘‘Broad Spectrum drug product. The absorbance ratio believing they are fully and completely UVA and UVB Protection’’ if UVA would range from 0 for products protected from the dangers of the sun. radiation protection provided (i.e., exhibiting no protection against UVA One comment recommended using the passes)), radiation to 1 for products exhibiting claim ‘‘full spectrum’’ rather than • A 3-level descriptive (labeled blank equal absorption at all wavelengths ‘‘broad spectrum’’ to describe products if no UVA radiation protection provided throughout the UVA/UVB radiation that attenuate more than 90 percent of or labeled ‘‘UVA and UVB Protection’’ spectrum. Using this ratio, products UVA radiation and are at least SPF 15. or ‘‘UVB Plus Extended UVA would be classified in one of the The comment suggested no UVA Protection,’’ depending on the level of following five categories: radiation protection claims be allowed if UVA radiation protection provided), • 0 to < 0.2 = no UVA radiation the product is below SPF 15. and protection claim • • 0.2 to < 0.4 = Moderate () In support of their proposed UVA A 3-level grapho/numerical (a bar • 0.4 to < 0.6 = Good () labeling, a number of comments graph indicating a level, 4, 8, or 12, with • 0.6 to < 0.8 = Superior () provided results from consumer the corresponding number appearing • 0.8 plus = Maximum () research studies that assessed consumer alongside the graph). Another comment recommended a labeling preferences for stating UVA The comment stated that the data five point rating system using the radiation protection. One comment suggested the pass/fail descriptor, described a 1996 survey (Ref. 4) in ‘‘critical wavelength’’ (CW) (λc) test ‘‘broad spectrum,’’ was significantly method. This system uses a scale which 275 subjects compared two superior to the other labels and analogous to the star rating system to labeling systems: recommended that FDA use this • assign products a ‘‘broad spectrum’’ 3-level descriptive (‘‘light,’’ labeling to designate UVA radiation rating as follows: ‘‘intermediate,’’ or ‘‘extended’’ ‘‘UVA protection. •λ< 325 = ‘‘0’’ protection’’) and Another comment described a c • • 325 < λc < 335 = ‘‘1’’ Grapho/numerical (a bar graph consumer research study conducted in • 335 < λc < 350 = ‘‘2’’ indicating a level, 0, 4, 8, or 12, with the 2000 (Ref. 8) at 20 urban and suburban • 350 < λc < 370 = ‘‘3’’ corresponding number appearing shopping malls in which 1,921 subjects • 370 < λc = ‘‘4’’ alongside the graph). ranked four labeling systems: Several comments supported a single The comment stated that the survey data • 4-level numerical, claim, such as ‘‘provides broad suggested that, while equally able to • 4-level symbolic, spectrum protection against UVB and understand both types of labels, the • 4-level descriptive, and UVA radiation,’’ based on determining a panelists preferred the grapho/ • Pass/fail descriptive (‘‘with/without sunscreen pass/fail CW (λc). Comments numerical system over the descriptive broad spectrum UVA/UVB protection’’). that supported this ‘‘broad spectrum system. The numerical labeling system was protection’’ claim stated that, in Another comment described two shown as Arabic numerals ‘‘1, 2, 3, 4’’ combination with SPF, it provides consumer research studies, conducted with the number ‘‘2’’ highlighted. The simple and accurate labeling that is in 1994 and 1995 (Ref. 9), in which 235 descriptor labeling system was shown as easily understood by consumers. The subjects compared three potential UVA the words ‘‘Minimum, Moderate, High, comments referred to a research study radiation labeling options: Maximum’’ with the word ‘‘Moderate’’ that suggested this approach to UVA • Numerical (2, 3, or 5), highlighted. The symbolic labeling radiation protection labeling was • Symbolic (4 stars with 1, 2, 3, or 4 system was shown as a picture of four superior for consumer comprehension stars filled), and stars with two stars highlighted.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49083

The comment concluded that the symbols) similar to those described by category descriptors for UVA radiation subjects had a significant preference for some of the comments. The category protection must be the same size as the a labeling system based on descriptive descriptors and stars will designate SPF value and its descriptors. All four words or numbers because of clarity, relative levels of UVA radiation stars must appear and be preceded by specificity, and ease of comprehension. protection as measured by the UVA the term ‘‘UVA’’ and followed by the Subjects least preferred the pass/fail radiation test methods. The level of appropriate category descriptor (e.g., system because they found it unclear, UVA radiation protection identified on UVA ✰ High). All star borders and nonspecific, and lacking sufficient the label reflects the following: the color inside a solid star must be the information to compare sunscreen • A numerical ‘‘UVA protection same while the color of ‘‘empty’’ stars products. This study also revealed that factor’’ (from the clinical test), and must be lighter and distinctively the numerical labeling system was one • A numerical ratio of UVA I (340 to different than solid stars. The color of the top two choices because numbers 400 nm) radiation absorption to UVB/ inside a solid star must be distinctively were ‘‘clearer, more specific, and easier UVA (290 to 400 nm) radiation different than the background color. The to understand.’’ Age, gender, and absorption (from the in vitro test). stars must be filled in starting with the educational or ethnic background were The test that indicates the lowest level first star on the left and must appear in reported as not affecting the study of UVA radiation protection determines a straight horizontal line. results. the level identified on the label. For As requested by some comments, an In the TFM for OTC sunscreen drug example, if the clinical test indicates OTC sunscreen drug product that does products (58 FR 28194 at 28233), FDA ‘‘low’’ protection and the in vitro test not provide the minimum UVA proposed to allow claims relating to indicates ‘‘medium’’ protection for a protection, as determined by the ‘‘broad spectrum protection’’ or ‘‘UVA product, the product is labeled as proposed UVA test methods, may only radiation protection’’ for OTC sunscreen providing ‘‘low’’ UVA radiation display an SPF value on the PDP. An products that meet the following two protection. This system comprises four OTC sunscreen drug product is not criteria: categories of UVA radiation protection required to provide UVA protection and 1. Contain sunscreen active as described in table 1 of this document. may bear only a sunburn (UVB/SPF) ingredients with absorption spectra protection claim. However, FDA is extending to 360 nm or above, and TABLE 1.—OVERALL UVA PROTEC- proposing that a sunscreen product that 2. Demonstrate meaningful UVA TION OF A SUNSCREEN DRUG does not provide at least a ‘‘low’’ level radiation protection using appropriate PRODUCT of UVA protection include the following testing procedures to be developed. statement on the PDP: ‘‘no UVA In the FM for OTC sunscreen drug Star category Category descriptor protection.’’ This statement must be the products (64 FR 27666 at 27672), FDA same font size as the SPF value and its stated that UVA radiation labeling of ✰✰✰ Low descriptor. FDA is not proposing four ✰✰ OTC sunscreen drug products could Medium empty stars because we are concerned continue in accordance with the TFM ✰ High that consumers may confuse products and its amendments until addressed in Highest providing no UVA protection (i.e., four a future issue of the Federal Register. empty stars) with those providing the Elsewhere in this document, FDA is Some of the comments argued that the highest UVA protection (i.e., four filled proposing test methods for determining UVB radiation protection labeling is more important than UVA radiation stars). the UVA radiation protection potential In developing this UVA radiation protection and should be emphasized in of an OTC sunscreen drug product (see protection labeling, FDA has the labeling over UVA radiation section III.N, comment 46). particularly considered the label FDA believes that the existing data do protection. FDA disagrees with the comprehension studies (Refs. 4, 7, 8, not clearly define the relationship comments and proposes that the UVA and 9). These studies used multiple between UVA radiation and skin radiation protection designation appear methodologies and report a diverse damage. The principal reason for not on the PDP along with the SPF value in range of preferences for each labeling better understanding this relationship is an equally prominent manner that does system: not conflict with the SPF value. Because that the action spectra for specific types • Category descriptors, of UVA radiation-induced skin damage action spectra for UV-induced skin • Graphics, (i.e., which wavelengths of UVA cause damage have not been clearly defined, • Symbols, which types of skin damage) have not FDA is unable to specify labeling for • Numerics, and been established. However, most OTC sunscreen drug products that • ‘‘Pass/fail’’ descriptors. scientific data demonstrate that UVA indicates what ranges of UV radiation The diverse results and varying radiation is harmful to the skin. Thus, are most harmful to consumers. In other methodology make it difficult to until these action spectra are known, words, FDA cannot conclude whether identify a clear preference for one FDA believes that more protection UVB or UVA radiation is more harmful labeling system. However, the studies against UVA radiation damage is better to humans based on the scientific data indicate an overall preference for for consumers’ health. Therefore, FDA collected to date. Therefore, FDA category descriptors. believes it is important, as with the SPF considers both UVB and UVA radiation In agreement with the studies, FDA is value, to designate UVA radiation protection equally important at this time proposing category descriptors to protection in a straightforward manner because scientific data demonstrates indicate the relative level of UVA that consumers clearly understand. that both have harmful effects on the radiation protection. As discussed in FDA proposes that the UVA radiation skin. preceding paragraphs, FDA believes protection of an OTC sunscreen drug So that consumers consider UVB and consumers should consider UVB and product determined from these UVA UVA radiation protection equally in UVA radiation protection equally when test methods be designated on the PDP selecting an OTC sunscreen drug selecting an OTC sunscreen drug using a combination of category product, FDA is proposing a number of product. For this reason, FDA is descriptors (i.e., ‘‘low,’’ ‘‘medium,’’ labeling requirements. Under this proposing that stars be used with ‘‘high,’’ or ‘‘highest’’) and stars (i.e., proposal, the font size of the stars and category descriptors. FDA believes that

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49084 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

the category descriptor and star labeling sunscreen product labels (alongside To prevent consumer confusion about for UVA radiation protection will give it these more general statements). UV radiation protection, FDA is equal prominence with UVB radiation Although none of the studies proposing changes to UVB radiation protection (i.e., category descriptor and combined labeling systems as proposed protection labeling (i.e., the SPF value). SPF) on the PDP. in this document, FDA believes the SPF values indicate how effective a FDA is not proposing grapho/numeric studies support use of category sunscreen product is in protecting labeling because we are concerned that descriptors and symbols together. One against sunburn. By displaying the consumers may be confused by a second study suggested that symbols may imply relative level of sunburn protection on number on the PDP (i.e., in addition to importance over SPF values (Ref. 9). the sunscreen drug product PDP in the SPF value). FDA is also not However, FDA believes consumers will terms of an SPF value, consumers can proposing any of the simple two- not place greater importance on UVA choose their desired level of UVB category designations suggested by the protection because we are proposing a radiation protection. To further improve comments: required statement to inform consumers consumers’ understanding of the • With/without UVA protection, about the importance of both UVB and sunburn protection level provided by a • With UVA protection/with extra UVA protection. We are proposing to certain sunscreen product, FDA is UVA protection, or require one of the following statements proposing to require descriptive terms • Regular/broad spectrum protection. on the PDP of all OTC sunscreen drug of relative sunburn protection (i.e., FDA agrees with one of the comments, products: ‘‘low,’’ ‘‘medium,’’ ‘‘high,’’ and which argued that these types of • ‘‘UV rays from the sun are made of ‘‘highest’’) to accompany the SPF value statements are misleading. FDA does UVB and UVA. It is important to protect on the PDP. FDA is further proposing not consider this labeling as providing against both UVB & UVA rays.’’ that the SPF value must be preceded by consumers with enough information • ‘‘UV rays from the sun are made of the term ‘‘UVB’’ to further differentiate about the magnitude of UVA protection UVB and UVA. It is important to protect the SPF value from the UVA symbol/ offered by an OTC sunscreen product. against both UVB & UVA rays to prevent descriptor on the PDP. FDA believes However, FDA does not object to the use sunburn and other skin damage.’’ that numerical labeling for UVB of the following four statements for OTC FDA believes that the use of one of these protection, symbolic labeling for UVA sunscreen drug products that satisfy the statements, along with the proposed protection, and the same descriptive requirements of proposed § 352.73 for a UVB and UVA radiation protection labeling for UVB and UVA protection labeled UVA protection value: labeling, including the format will allow consumers to easily • ‘‘broad spectrum sunscreen’’, requirements described in preceding understand and choose from relative • ‘‘provides [select one of the paragraphs, will lead consumers to view levels of UVB and UVA radiation following: ‘UVB and UVA,’ or ‘broad UVB and UVA radiation protection as protection. spectrum’] protection’’, equally important. FDA is aware that consumers have • ‘‘protects from UVB and UVA In addition, this statement will used and become accustomed to [select one of the following: ‘rays’ or educate consumers about UVA choosing OTC sunscreen drug products ‘radiation’]’’, and radiation, which will be a new term and based on the SPF value for many years. • [select one of the following: concept to many consumers. The Likewise, FDA believes that, over a ‘‘absorbs’’ or ‘‘protects’’] ‘‘within the proposed statement should help period of time, consumers will similarly UVA spectrum’’. consumers better understand the new become accustomed to the proposed These statements may appear elsewhere UVB and UVA labeling when it is labeling using symbols and descriptors in product labeling outside the ‘‘Drug initially introduced to the OTC market. to designate relative UVA radiation Facts’’ box or enclosure but not Thus, FDA believes that the consumer protection. Furthermore, FDA believes intermixed with the information label comprehension studies, along with consumer familiarity with similar star required on the PDP under § 352.50. the proposed educational statement rating systems (e.g., movies, hotels, and FDA agrees with some comments that about UVB and UVA radiation, support restaurants) used for many years in the these statements, by themselves, may be the stars and descriptor UVA radiation United States provide a basis for misleading by implying that a sunscreen protection labeling proposed in this consumers’ understanding of this protects against nearly all UVB and document. Moreover, a similar ‘‘star proposed labeling for OTC sunscreen UVA radiation. However, FDA does not rating system’’ for UVA radiation drug products. believe these optional statements will be protection (i.e., the Boots Star System) FDA is providing a number of misleading in the context of the entire has been used to label sunscreen examples of how the UVA/UVB label, because the relative level of UVB products throughout Europe for over 10 protection designations could appear on and UVA protection must be stated on years. the PDP.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49085

FDA believes that, as with SPF values, want the many currently marketed A foreign industry organization identifying the relative level of UVA sunscreens that are labeled as ‘‘SPF 45, suggested an upper limit for labeled SPF radiation protection provides the most SPF 50, etc.’’ Thus, the comments values of 50+ and provided three useful information for consumers. argued that U.S. consumers will be at a reasons: Consumers who desire more protection disadvantage within the international • Unreasonably high SPF values will from the sun will be able to identify community, because products providing lead consumers to expect ‘‘too much products with higher UVB (SPF) and SPF values over 30 are available in other effectiveness’’ from sunscreen products. UVA radiation protection. FDA agrees countries. In addition, the comments • Higher concentrations of sunscreen with the comments that a product must stated that many prominent medical active ingredients are not ‘‘in the provide at least some minimum level of authorities maintain the need for high interest of safety.’’ • UVA radiation protection (as with SPF SPF sunscreens for individuals at ‘‘high Higher SPF values will invite values) to be labeled as providing UVA risk’’ based on medical and/or excessive, meaningless competition in radiation protection. Therefore, FDA is occupational concerns and individuals the industry. proposing minimum criteria for the who desire increased protection from The comment explained that lowest UVA category in its proposed photoaging and lengthy/intensive sun competition would be meaningless test procedures (see section III.N, exposure situations. The comments because the amount of UV protection comment 46 of this document). argued that the need for high SPF provided by products with SPF values sunscreens is supported by findings that above 50 is not significantly greater than F. Comments on the Labeling of products with an SPF of 50. Sunscreen Drug Products With High SPF UV exposures in several cities are considerably higher than previously Another comment from a sunscreen Values manufacturer agreed with FDA’s recognized and because high SPF (Comment 15) Several comments concern about the possibility of products can reduce cumulative UV objected to FDA limiting specific increasing variability when testing high exposure. The comments stated that labeled SPF values ‘‘up to but not above SPF sunscreens. The comment consumer desire for high SPF products 30.’’ The comments stated that data and suggested a modified ‘‘binomial’’ test is demonstrated by sales data showing information supplied to FDA since method and labeling requirements for that products with an SPF value of 45 publication of the sunscreen FM SPF values over 20 that would allow for are one of the fastest growing segments demonstrate that SPF values over 30 can high SPF products. be safely tested with accuracy. The of the total sunscreen market. Another comment submitted a comments also argued that removing the The remaining comments discussed published survey of 208 sunbathers on limit will not lead to consumers the consequences of limiting the Miami’s South Beach during July 2001 spending more time in the sun when specific labeled SPF value. For example, with the goal of measuring UV radiation using high SPF sunscreens in one comment noted that if exposure and probable injury (Ref. 10). comparison to low SPF sunscreens. To manufacturers cannot state the SPF The ‘‘worst case’’ scenario identified by address that point, one comment level above 30, they will no longer have the survey was based on sunbathers proposed labeling to help reduce an incentive to fund research for better with Type I skin (persons most sensitive potential consumer misuse of sunscreens. In addition, manufacturers to sunlight who burn easily and never sunscreens with SPF values over 30: may reformulate products to reduce tan) exposed to UV radiation near the ‘‘higher SPF products give more sun active ingredients and, thus, reduce the longest day and highest sun angle of the protection, but are not intended to level of UV protection. A comment year at the ‘‘southern-most major beach’’ extend the time spent in the sun.’’ argued that another adverse in the United States. The survey was a Another comment noted that the SPF consequence results from most followup to one conducted in 1993 with value, in addition to proper sunscreen consumers failing to achieve the labeled 62 sunbathers and evaluated by FDA in application and reapplication, is only SPF value because they do not apply the FM (64 FR 27666 at 27674). The part of a comprehensive sun protection enough sunscreen and/or reapply it too 2001 survey determined MEDs absorbed program. infrequently. Because high SPF by the following three steps: Other comments explained the need products can help make up for such 1. Measuring incident UV radiation for high SPF sunscreen products. The improper use, limiting the specific (using three dosimeters), comments contended that consumers labeled SPF value to 30 has a negative 2. Multiplying by an adjusting factor and physicians are familiar with and impact on UV protection. for skin type (using a 30 percent

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.003 49086 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

increase in sensitivity between skin active ingredients are not ‘‘in the document, FDA is proposing UVA types), and interest of safety’’ did not supply any radiation testing methods and labeling 3. Dividing by the SPF worn by the new data to support its contention. FDA that will categorize the relative levels of sunbather. will continue to monitor adverse drug protection provided by the SPF and The survey suggests that sunbathers experience reports for sunscreen drug UVA values of the sunscreen product with Type I skin might receive a products reported to its Medwatch (see section III.E, comment 14 and cumulative dose of 49.5 MEDs with 8 program and in the medical literature. section III.N, comment 45 of this hours of exposure. The comment As noted by one comment, some document), allowing consumers to concluded that, while SPF values up to, researchers have raised the concern that compare products and choose the levels and including, 50 are warranted, values sunscreen use may lead to increased sun of UVB and UVA radiation protection over 50 are unwarranted in any exposure. The ‘‘compensation desired. condition for sunburn protection. hypothesis’’ states that consumers who An SPF 30 sunscreen product may Two comments submitted testing data use high SPF sunscreens spend more provide adequate sunburn protection for for sunscreens with SPF values between time in the sun and/or use less many consumers. However, FDA 30 and 50 using the test method in the protective clothing. The only double believes that appropriately tested and FM. The comments concluded that the blind, randomized trial that addressed labeled high SPF value sunscreen test method was valid for these high this issue showed a significant increase products should be available for SPF values. In addition, one comment in sun exposure time when comparing consumers who desire or need high indicated that a very water resistant test use of SPF 30 to SPF 10 (Ref. 14). In levels of UV protection, in particular, for an SPF 45 to 50 sunscreen would addition, two retrospective survey those who burn easily. Such products take nearly 4.5 hours using the skin studies showed that sun exposure time would do the following: types of subjects in the SPF testing is longer when using sunscreen • Help compensate for inadequate procedures in the FM (i.e., skin types I, compared to not using sunscreen (Refs. application and/or reapplication, II, and III) (Ref. 13). The comment 15 and 16). Other studies cited by the • Provide additional sunburn concluded that it is beyond the practical comment to support the premise that protection during intense UV radiation endurance capabilities of many people the ‘‘compensation hypothesis’’ is conditions, in the test to spend more than 5 to 6 incorrect and either did not provide • Help reduce cumulative UV hours in front of a UV radiation lamp data about the length of sun exposure or radiation exposure (when used in and that fatigue can lead to errors in test the study method did not allow for data conjunction with other measures to results. The comment also noted that interpretation (Refs. 17 through 20). reduce overall sun exposure), and • the potential for intra and Based on all of this data, FDA believes Generally provide consumers interlaboratory variability in test results that some consumers may increase total incremental increases in sunburn increases as sunscreen SPF values UV exposure through over-reliance on protection. increase. sunscreens. The apparent divergent FDA agrees that SPF values should be FDA concluded in the FM (64 FR results on the validity of the supported by scientific evidence. In the 27666 at 27675) that test methods ‘‘compensation hypothesis’’ between FM, FDA limited the specific labeled supported specific SPF label values up studies may indicate that sun protection SPF value to 30. At that time, FDA had to 30. FDA invited interested persons to behaviors vary greatly for each person. only received data demonstrating that submit data in support of high SPF test More specifically, there is a spectrum of the SPF test produces accurate results methods and to consider proposed attitudes about the sun, from those for products with SPF values of 30 or methods for communicating the level of individuals who seek dark suntans to less. Since publication of the FM, FDA protection in labeling. Data and those who seek to avoid the sun and has received additional SPF testing data information on high SPF testing and consequent UV skin damage (Ref. 21). for sunscreen products with SPF values labeling were submitted to FDA at, and Such evidence underscores the need for between 30 and 50 (Ref. 13). However, following, public meetings on July 22, adequate labeling so consumers can FDA has not received any data for 1999, and October 26, 1999, and after make informed decisions regarding their sunscreen products with SPF values reopening of the administrative record use of OTC sunscreen drug products. greater than 50. The data submitted to (65 FR 36319) (see section III.I, FDA agrees that the SPF value is one FDA indicate that the SPF test is comment 24 of this document) (Refs. 11 factor in a comprehensive sun accurate and reproducible for sunscreen and 12). protection program. However, the SPF is products with SPF values up to 50 (Ref. FDA continues to be aware that many only a measure of protection from 13). However, these data cannot be OTC sunscreen products with specific erythema (i.e., UVB radiation-induced extrapolated to SPF values above 50. labeled SPF values over 30 are currently sunburn) and does not measure Thus, FDA proposes to allow specific marketed, both nationally and protection from other UV skin damage, labeled SPF values up to 50. internationally, and are increasingly such as that induced by UVA radiation. FDA agrees with the sunscreen used by consumers and recommended While increased short wavelength UVA manufacturer that increasing variability by health professionals (64 FR 27666 at radiation protection generally increases in test results is likely with increasing 27675). FDA agrees that these products with increasing SPF values, studies SPF values. If there is large variability should be available for those sun- using in vivo or in vitro UVA radiation in test results, then the SPF value sensitive consumers who require such testing methods demonstrate that determined from the test is not accurate products based upon personal sunscreen products with the same SPF (i.e., an SPF 50 product may not knowledge, planned sun exposure, values can have markedly different actually be an SPF 50 product). The geographical location, or advice of a levels of UVA protection, especially for submitted data demonstrate that health professional. FDA previously long wavelength UVA radiation (Refs. variability is not an issue for sunscreen noted the lack of any known safety 22 and 23). These studies also indicate products with SPF values up to 50. problems for sunscreen products with that a specific high SPF product can However, FDA is concerned that SPF values greater than 30 (64 FR 27666 provide much less UVA radiation variability will become an issue for at 27675). The comment that argued protection than a product with a much sunscreen products with SPF values higher concentrations of sunscreen lower SPF value. Elsewhere in this over 50.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49087

FDA recognizes that future data may products with specific SPF values over because it is duplicative and no longer demonstrate that variability may not be 50. As explained earlier in this necessary. This phrase would only a problem for sunscreen products with comment, FDA is not proposing that the lengthen the ‘‘Uses’’ statement. SPF values over 50. Therefore, FDA will specific value over 50 be stated in the Furthermore, consumers will now be consider specific SPF values greater labeling because there is no data, at this able to equate a product’s UVB radiation than 50 upon receipt of data time, demonstrating the accuracy and protection rating (i.e., SPF value) demonstrating that accurate and reproducibility of the specific value directly to the relative level of sunburn reproducible results can be obtained over 50. Based upon the proposed protection. from the SPF test for sunscreen products labeling, improvements to SPF testing The second indication ‘‘higher SPF with SPF values over 50. Generally, methods, and specific high SPF test gives more sunburn protection’’ is no such data should include results from data, FDA is proposing to modify the longer needed because the relative level multiple laboratories using the same labeled SPF values in current of sunburn protection is provided in the sunscreen formulations and using the § 352.50(a)(1) and (a)(2) by changing the new ‘‘Uses’’ statements. In addition, SPF test proposed in this document, SPF values from ‘‘30’’ to ‘‘50.’’ without clarification, the statement may along with a statistical analysis of the encourage consumers to spend more overall results. In addition, FDA G. Comments on Indications for time in the sun. Clarification is believes that the modified ‘‘binomial’’ Sunscreen Drug Products necessary because, as discussed in test method submitted by one comment (Comment 16) One comment comment 19 of this document, surveys has merit for high SPF sunscreens and requested that the ‘‘Uses’’ statement, reveal that consumers spend more time is requesting others’ views on this ‘‘higher SPF gives more sunburn in the sun with increasingly higher SPF method during the comment period for protection,’’ be omitted except for sunscreen products (Refs. 14, 15, and this rulemaking (see section III.I, products with an SPF over 30. This and 16). Therefore, FDA is not allowing this comment 24 of this document). other comments suggested that FDA’s statement in the ‘‘Uses’’ section. In the FM (64 FR 27666 at 27675), labeling concerns regarding high SPF However, under proposed § 352.52(e)(2), FDA disagreed with the comment that sunscreens could be alleviated if the FDA is proposing the following optional manufacturers would have no incentive following statement was required on statement under ‘‘Other information’’ or to fund research for better sunscreens sunscreens over SPF 30: ‘‘Higher SPF anywhere outside of the ‘‘Drug Facts’’ and may reformulate to less protective products give more sun protection, but box or enclosure: ‘‘higher SPF products products if there is an upper limit to are not intended to extend the time give more sun protection, but are not specific labeled SPF values. Although spent in the sun.’’ intended to extend the time spent in the FDA would not want to decrease FDA is proposing to revise the sun.’’ The phrase ‘‘but are not intended research incentive, FDA is more sunscreen FM ‘‘Uses’’ statement ‘‘helps to extend the time spent in the sun’’ is concerned about valid scientific data prevent sunburn’’ and delete the ‘‘Uses’’ additional information not included in demonstrating the ability of multiple statement ‘‘higher SPF gives more the FM indication. FDA believes this laboratories to accurately and sunburn protection’’ in current revised indication statement will reproducibly determine SPF values. § 352.52(b). The first indication, ‘‘helps discourage consumers from spending However, FDA does not believe it is prevent sunburn,’’ is being revised to more time in the sun when using a necessary to arbitrarily limit specific one of the following, which would be higher SPF product. labeled SPF values. To the contrary, required on all sunscreens: FDA is proposing additional revisions both in the FM and in this proposal, • ‘‘low UVB sunburn protection’’ in ‘‘Uses’’ in § 352.52(b)(1) to include FDA has specifically stated that high • ‘‘medium UVB sunburn protection’’ UVA claims and other information (see SPF sunscreens should be available for • ‘‘high UVB sunburn protection’’ section III.G, comments 17 and 18 of those individuals desiring such • ‘‘highest UVB sunburn protection’’ this document). The proposed revisions products. The maximum allowable The relative level of sunburn protection will help consumers to more fully specific labeled SPF value, both in the is determined from the SPF value: understand the uses and expected FM and in this proposal, is based upon • low = SPF 2 to under 15 results for individual sunscreen the review of data and information • medium = SPF 15 to under 30 products. These changes are necessary submitted to FDA. FDA purposely did • high = SPF 30 to 50 because the PDP for a sunscreen product not limit labeled SPF values at 30 in the • highest = SPF over 50 will now include two performance FM. Instead, FDA used the value of Thus, relative descriptors (low, ratings (see section III.E, comment 14 of ‘‘30+,’’ pending the receipt of adequate medium, high, and highest) describe this document): data to support any higher specific label SPF values, which are relative and not • The well-accepted SPF value and values. absolute levels of sunburn protection new descriptor rating for UVB radiation Similarly, in this document, FDA is intended to help consumers determine protection, and proposing the collective value ‘‘50+.’’ differences in sunburn protection • A new star/descriptor rating for FDA has sufficient assurance that a offered by different sunscreen products UVA radiation protection. result over 50 from the required SPF test (see section III.I, comment 23 of this Consequently, FDA considers it is, in fact, greater than 50 and can be document). important that the ‘‘Uses’’ statements in labeled ‘‘50+.’’ Thus, FDA believes that FDA considers it important that the ‘‘Drug Facts’’ box accurately reflect the term ‘‘SPF 50+’’ is truthful and consumers be made aware of the relative product claims related to specific nonmisleading on the label of OTC level of sunburn protection provided by indications, UVA and UVB radiation, sunscreen drug products for which the a product in addition to its indication and the level of anticipated protection SPF test in the monograph has indicated for sunburn protection. Individuals may (low, medium, high, or highest) an SPF value greater than 50. FDA select a low, medium, high, or highest determined by the UVA and UVB believes that allowing manufacturers to sunburn protection product to meet product ratings. As with the label sunscreens as ‘‘SPF 50+’’ may their specific needs. The descriptor introduction of SPF labeling years ago, encourage further research in human ‘‘UVB’’ is included to describe the it will take the combined efforts of skin photobiology and the development predominant rays that are screened. The government, manufacturers, consumer of safe and effective sunscreen drug phrase ‘‘helps prevent’’ is being deleted organizations, and the health care

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49088 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

community to educate consumers to conditions. FDA believes that the sunscreen prevented the development of fully understand these labeling articles with clinical data provide more solar elastosis, a condition in which initiatives to enhance their safe and meaningful results, as they can be easily skin loses its elasticity after chronic effective use of sunscreen products. extrapolated to consumer actual use exposure to the sun (Ref. 33). However, (Comment 17) One comment stated conditions. Therefore, FDA is focusing these method and data analyses raise that FDA’s ‘‘sun alert’’ statement in the discussion in this document on the questions about the validity of the FM recognized that sun-induced skin clinical studies. In agreement with reported conclusion: damage can contribute to photoaging Garland (Ref. 25) and Gasparro (Ref. 24), • Discrepancies were noted and increase the risk of skin cancer. FDA does not believe, as a whole, that concerning demographic characteristics This statement reads: ‘‘Sun alert: the studies demonstrate that sunscreens of subjects, sunscreen application, and Limiting sun exposure, wearing alone help prevent skin aging or skin compliance rates. protective clothing, and using cancer. • Skin biopsy data at all three time sunscreens may reduce the risks of skin Some of the clinical studies examined points in the study were available from aging, skin cancer, and other harmful the role of UVB and UVA radiation in only 10 of the 35 subjects. effects of the sun.’’ The comment urged producing histological changes • The only statistically significant FDA to allow other truthful use indicative of skin aging due to the sun. difference between the sunscreen and statements, such as the following: Lowe et al. demonstrated that high placebo treatment groups was achieved • ‘‘helps protect against skin damage doses of UVA radiation (320 to 400 nm) in a computerized evaluation of solar caused by the sun’’ increased melanization of human skin elastosis at baseline and 24 months. • ‘‘helps protect against skin aging more than lower doses of UVA or solar The second study demonstrated caused by the sun’’ simulating UV radiation at 290 to 400 significant contribution of a sunscreen • ‘‘regular use helps protect against nm (Ref. 26). Seite et al. demonstrated in preventing UV radiation-induced certain forms of skin cancer caused by that melanization of human skin skin damage (Ref. 32). The use of the sun’’ increased with exposure to UVB/UVA sunscreens with absorption spectra • ‘‘helps protect against fine lines and radiation at 290 to 400 nm (Ref. 32) and covering the 290 to 400 nm range wrinkles caused by the sun’’ UVA radiation at 330 to 440 nm (Ref. prevented all of the effects of chronic • ‘‘helps protect against pigmentary 27). Seite et al. also showed that human exposure (6 weeks) to UV radiation changes due to sun exposure’’ skin hydration decreased after chronic evaluated in the study. The third study Another comment urged FDA to include exposure to UV radiation at the showed a photoprotective effect of an the first three use statements suggested wavelengths studied. SPF 15 sunscreen product from damage by the first comment, as well as ‘‘helps Five studies revealed stratum induced by short term exposure to UVB protect against the harmful effects of the corneum thickening produced by both radiation (Ref. 30). The fourth study sun’’ and ‘‘helps protect against (select UVB and UVA radiation (Refs. 26 showed that a UVB only sunscreen did one: ‘casual,’ ‘incidental,’ ‘intermittent,’ through 29 and 32). Stratum granulosum not provide protection against chronic or ‘daily’) sun exposure.’’ The comment thickening was transiently induced after exposure to UVA radiation (Ref. 29). contended that, when used effectively 6 weeks of exposure to UV radiation The studies provide evidence that as part of a sun protection program, (UVB/UVA) at 290 to 400 nm (Ref. 32). both UVB and UVA radiation induce sunscreens may prevent very serious The same effects were seen with solar histological changes in the skin disease conditions. simulated radiation and high and low consistent with skin aging. Thus, the Another comment provided citations doses of UVA radiation after 12 weeks studies support the conclusion that from the medical literature to support of exposure (Ref. 26). Viable epidermal exposure to UV rays increases the risk its contention that claims of sunscreens thickening was seen after 6 weeks of of premature skin aging. However, the preventing skin cancer induction may exposure to UV radiation at 290 to 400 study data fails to show that sunscreen be false, deceptive, misleading, and nm in one study (Ref. 32) and after 9 use alone helps prevent premature skin unsubstantiated. The comment days of exposure to UVA radiation at aging for several reasons. First, the mentioned an article by Garland (Ref. 335 to 345 nm in another study (Ref. studies have not completely defined the 25) that states the following: ‘‘No 31). action spectrum for the majority of UV epidemiological studies were identified Inflammation and lysozyme radiation-induced effects on human that showed a protective effect of use of deposition along the dermal elastic skin. While studies demonstrate that a chemical sunscreen on risk of fibers were increased more in human given histological change, such as melanoma or other cutaneous skin exposed to UVA than UVB thickening of the stratum corneum, is malignancies in humans.’’ The comment radiation (Refs. 26, 28, 29, and 31). induced by certain wavelengths within also mentioned an article by Gasparro Sunburn cell appearance, a typical the UVB and UVA region, studies have (Ref. 24) that states the following: response to UVB radiation, was also not examined the ability of the ‘‘Although some have promoted daily found to be present after exposure to remaining UVB and UVA regions use (of sunscreen) for the prevention of different UVA radiation regimens in two outside of these wavelengths to induce premature aging of the skin and the studies (Refs. 28 and 31) but not found the same change. For example, studies prevention of skin cancer, actual data in a third study (Ref. 27). Thus, FDA may have shown that 290 nm to 310 nm are lacking to support these concludes that these studies and 360 nm to 400 nm radiation induce recommendations.’’ demonstrated that both UVB and UVA stratum corneum thickening, but it is FDA has reviewed the submitted radiation induce histological changes not known whether 311 nm to 359 nm articles concerning UV-induced skin associated with skin aging. radiation induces the same histological damage (i.e., premature aging and Four of these studies focused on the change. cancer) along with articles obtained histological changes within the skin Second, the inability to identify the from a search of the scientific literature induced by UVB and UVA radiation and exact UVB and UVA wavelengths that (Refs. 26 through 34). Many of the explored the ability of sunscreens to induce each histological change in the articles involved preclinical data, which protect human skin against these skin derives from the study designs. can be difficult to extrapolate to changes (Refs. 29, 30, 32, and 33). The Each study differed in the following consumer (human) actual use first study suggested that an SPF 29 parameters:

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49089

• UV radiation wavelengths, effects of UV radiation to a long-term, comment, along with the third • UV exposure regimens, chronic effect such as skin cancer. In comment, seemed to agree with FDA’s • Sunscreen doses, addition, like skin aging, the complete conclusions in this proposed rule • Sunscreen application techniques, action spectrum for skin cancer is not concerning the need for consumers to and known at this time. • use sunscreens in conjunction with Endpoints. Unlike skin cancer and premature other sun protection measures. Therefore, FDA cannot combine all of skin aging, FDA has evidence that In addition, the reference in the new the data from these studies to define a sunscreens alone help prevent sunburn. warning statement to sunscreen use complete action spectrum for each The SPF test measures the effectiveness combined with limiting sun exposure histological change in the skin. of sunscreens with sunburn (erythema) and wearing protective clothing is Furthermore, the action spectrum for as the endpoint. Thus, the impact of consistent with recommendations by each histological change would need to sunscreens on sunburn can be measured other public health organizations. For be combined to define a single action directly. In contrast, it is difficult to example, the World Health spectrum for skin aging, which is a measure directly the impact of Organization’s International Agency for cumulation of these histological sunscreens on skin cancer or premature Research on Cancer (IARC) (Ref. 21) changes. Without knowing which UVB skin aging because these are long-term, makes the following assessments and and UVA wavelengths induce each cumulative adverse effects of UV recommendations: histological change in the skin, FDA is exposure. • There is inadequate evidence in unable to determine which wavelengths Thus, for all of the reasons discussed humans for a cancer preventative effect are most important in causing skin aging in this comment, FDA concludes that of sunscreens against basal cell or and cannot determine the action the available evidence fails to show that malignant melanoma cancers. spectrum for aging. sunscreens alone help prevent skin • There is only limited evidence for Third, the studies did not examine the cancer or premature skin aging. Based a preventive effect of sunscreens against chronic, long-term consequences of UV on this conclusion, FDA is not squamous cell cancer. radiation exposure in human skin. proposing the indication statements • Sunscreens should not be the first Thus, it is not possible for FDA to proposed by the first and second choice for skin cancer prevention or extrapolate the data to longer time comments, because these claims are for used as the sole agent for protection points at which the short-term protection from premature skin aging, against UV radiation. histological changes may cumulate to skin cancer, and related factors (e.g., Likewise, the CDC recommends that produce visible signs of skin aging. ‘‘helps protect against skin aging caused sunscreens be used as a complementary Fourth, although the studies that by the sun’’). FDA also is not proposing measure in an overall sun protection examined the ability of sunscreens to claims that sunscreens protect against program (Ref. 35). protect against UV radiation-induced ‘‘casual, incidental, intermittent, or FDA believes that additional histological changes in the skin provide daily’’ sun exposure, as proposed by the information from controlled clinical useful data, it is difficult for FDA to second comment, because the studies do studies is needed to better understand conclude that sunscreens alone help not support these claims. Furthermore, the role of sunscreens in preventing prevent skin aging based on these FDA considers these terms as lacking premature skin aging and skin cancer. studies. The number of participants in sufficient meaning to be useful to Studies examining premature skin aging each study was relatively small, with consumers. (using solar radiation or simulated solar only 10 to 35 subjects per study. As described elsewhere in this radiation) are needed to determine the Different sunscreen formulations, with document (see section III.G, comment following in humans: differing absorption spectra, were used 19), FDA is proposing to require a • Measurable skin properties such as in each study. As explained previously, revised ‘‘sun alert’’ statement in the elasticity, collagen/elastin ratios and these studies do not identify exactly form of a new warning. The new properties, wrinkling, pigmentation which UVB and UVA wavelengths warning statement is based on FDA’s changes and visual grades, leading to contribute the most to skin aging (i.e., review of the available evidence accepted quantitative definitions of the studies do not define the skin aging concerning UV exposure and skin chronological and sun-induced skin action spectrum). For all of these cancer, premature skin aging, and other aging; reasons, the studies do not prove that skin damage. The new warning • The relationship between sunlight sunscreens alone help prevent statement clarifies that UV exposure exposure and skin aging, stratified by premature skin aging. from the sun increases the risk of skin skin type; Likewise, FDA is not aware of data cancer, premature skin aging, and other • An action spectrum for photoaging demonstrating that sunscreens alone skin damage. In addition, the new of skin; help prevent skin cancer. It has been warning statement specifies that • A dose response for UV radiation- known for many years that UV radiation consumers should use complementary induced skin aging; increases the risk of skin cancer. It has sun protection measures along with • Quantitative estimates of realistic also been known for many years that a sunscreen (i.e., limit sun exposure and ‘‘worst case,’’ long-term exposures to higher incidence of sunburn earlier in wear protective clothing). FDA has sunlight in relevant UVA and UVB life corresponds to a higher incidence of concluded from the available evidence radiation spectral ranges (i.e., the level skin cancer later in life. However, FDA that it is important to adopt a complete of UVB and UVA protection needed); is not aware of any studies sun protection program (sunscreen, sun and demonstrating that the use of avoidance, and protective clothing) to • How UV radiation-induced sunscreens alone decreases the risk of decrease UV exposure. In fact, the processes that occur at a given skin cancer. Like skin aging, there are second comment argued for new wavelength affect UV radiation-induced studies examining the effects of indication statements by considering the processes that occur at other sunscreens on short-term factors for skin sunscreen use as part of such a sun wavelengths. cancer, such as sunburn and other protection program (i.e., in conjunction Similar information is needed for skin cellular damage. However, it is difficult with limiting time in sun and wearing cancer, except that studies should to extrapolate these short-term adverse protective clothing). Thus, the second examine the different types of skin

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49090 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

cancer, rather than examining different following UVA indication statements, time, FDA encouraged manufacturers to skin properties. In addition, IARC has depending on the level of UVA voluntarily include this statement in provided recommendations for research protection provided by a product: labeling, make it available at the point on skin cancer prevention and • ‘‘low UVA protection’’ of purchase, and/or make it available sunscreens. These recommendations • ‘‘medium UVA protection’’ through consumer education programs. can also be used as a guide in designing • ‘‘high UVA protection’’ FDA is now proposing a revised ‘‘sun studies to examine the role of • ‘‘highest UVA protection’’ alert’’ statement be required in the sunscreens in preventing premature The level of protection (i.e., low, ‘‘Warnings’’ section of the ‘‘Drug Facts’’ skin aging due to the sun (Ref. 21). FDA medium, high, or highest) is determined box. FDA is proposing the statement to encourages interested parties to submit from the UVA rating obtained from read as follows: ‘‘UV exposure from the study protocols to FDA for review to product testing (see section III.N, sun increases the risk of skin cancer, ensure that studies are as informative as comment 45 of this document). premature skin aging, and other skin possible. FDA also invites comments by Manufacturers who wish to combine the damage. It is important to decrease UV interested parties on the feasibility and ‘‘Uses’’ statements about UVA exposure by limiting time in the sun, validity of surrogate endpoints for protection and UVB sunburn protection wearing protective clothing, and using a studies to determine whether the use of may do so if the descriptors (i.e., levels sunscreen. FDA is proposing that the sunscreens alone help prevent skin of protection) are the same. For statement appear in bold type as the cancer, premature skin aging, or other example, if the levels of UVA and UVB first statement in the ‘‘Warnings’’ skin damage. protection are medium, the ‘‘Use’’ may section. FDA believes the statement is (Comment 18) As discussed in section read: ‘‘medium UVA/UVB sunburn most appropriate in the ‘‘Warnings’’ III.E of this document, FDA received protection’’. section because it warns consumers that several comments discussing ways to FDA is not including the fifth effective protection from the sun does categorize, phrase, and display UVA/ indication because FDA does not not involve only the application of UVB radiation protection on an OTC consider ‘‘skin aging’’ or ‘‘skin damage’’ sunscreens, as many consumers believe. sunscreen drug product label. In the claims adequately supported at this In addition, it warns consumers that UV amendment to include avobenzone in time. As discussed elsewhere in this radiation not only increases the risk of the monograph (61 FR 48645 at 48655), document (see section III.G, comment sunburn but also increases the risk of FDA proposed the following indications 19), FDA is proposing a statement in the skin cancer and premature skin aging, for UVB and UVA radiation protection ‘‘Drug Facts’’ box that informs which many consumers may not know. by sunscreen drug products containing consumers that sunscreens may reduce FDA believes the new warning will avobenzone: the risks of skin aging, skin cancer, and encourage consumers to use sunscreen, 1. ‘‘Broad spectrum sunscreen’’; other harmful effects from the sun when limit time in the sun, and wear 2. ‘‘Provides’’ (select one of the used in a regular program that relies protective clothing to reduce UV following: ‘‘UVB and UVA,’’ or ‘‘broad upon limiting sun exposure and wearing exposure. Because of the importance of spectrum’’) ‘‘protection’’; protective clothing. Therefore, FDA warning statements and the need for 3. ‘‘Protects from UVB and UVA’’ believes the fifth indication statement consumers to receive a uniform message (select one of the following: ‘‘Rays’’ or would mislead consumers by not concerning such warnings, no variations ‘‘radiation’’); discussing sun exposure and protective in wording are allowed under 4. (Select one of the following: clothing. § 330.1(c)(2). ‘‘Absorbs,’’ ‘‘Protects,’’ ‘‘Screens,’’ or (Comment 19) As discussed in section FDA acknowledges that the new ‘‘Shields’’) ‘‘throughout the UVA III.G of this document, FDA received warning statement differs from the spectrum’’; and several comments concerning the ‘‘sun’’ wording of the voluntary ‘‘sun alert’’ in 5. ‘‘Provides protection from the UVA alert statement. In § 352.52(e)(2) of the the FM. These differences are based on rays that may contribute to skin damage FM, FDA included the optional FDA’s assessment of the additional and premature aging of the skin’’. statement: ‘‘Sun alert: Limiting sun evidence available since publication of Likewise, in the amendment to include exposure, wearing protective clothing, the FM in 1999. As explained in zinc oxide in the monograph (63 FR and using sunscreens may reduce the comment 17 of this document, FDA 56584 at 56588), FDA proposed similar risks of skin aging, skin cancer, and does not believe that the available data labeling for UVA and UVB radiation other harmful effects of the sun.’’ This support a claim concerning the use of protection for products containing zinc statement’s emphasis of the need for a sunscreen and a reduction in the risk of oxide (substituting the word ‘‘within’’ comprehensive sun protection program premature skin aging and skin cancer. for the word ‘‘throughout’’ in the fourth (64 FR 27666 at 27679) was based on the The revised wording of the statement statement). FDA did not include these findings of numerous groups, including more accurately reflects the scientific indications in the FM but has allowed the following: conclusions that can be drawn from this their use until the UVA portion of the • The American Academy of evidence. monograph is established. Dermatology (AAD), FDA is proposing the warning FDA has reconsidered these UVA • The CDC, because we continue to be concerned protection indications. FDA is • The Australian Government; and about adequate consumer understanding proposing to allow all of them except • The New Zealand Government. of a sun protection program that the fifth statement. In proposed These groups have recommended that includes sun avoidance and wearing § 352.52(e), the first four statements are sunscreens be considered an adjunct to protective clothes along with sunscreen optional statements allowed for other UV protection strategies, such as use. This proposed rule provides for products that demonstrate UVA avoiding the sun near midday, seeking even higher SPF values and a new rating protection according to the proposed shade, and wearing protective clothing system for UVA protection. Consumers testing (see section III.N, comment 45 of and hats. may believe that sunscreens with higher this document). The statements can only The FM provided that the ‘‘sun alert’’ SPF values (especially with UVA be included in labeling outside of the appear under the heading ‘‘Other protection) provide complete UV ‘‘Drug Facts’’ box. Within the ‘‘Drug information’’ or anywhere outside of the radiation protection. Subsequently, Facts’’ box, FDA is proposing one of the ‘‘Drug Facts’’ box or enclosure. At that consumers may prolong sun exposure

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49091

because they think higher SPF values • An estimated 8,110 persons would purchase sunscreens to provide equate to longer times in the sun die from melanoma and 2,000 persons protection from the sun damage that is without burning. FDA is aware of a would die from other skin cancers. not immediately recognizable. For these double-blind, randomized clinical study Skin cancer affects roughly the same reasons, the comment requested claims that showed a significant increase in number of people as all other cancers such as the following: combined. In view of the continuing sun exposure time of persons using high • ‘‘helps protect against casual or SPF sunscreens compared to persons increase in the incidence of all types of skin cancer and the lack of data incidental or intermittent daily sun using low SPF sunscreens (Ref. 14). In exposure’’ addition, two questionnaire-based demonstrating that sunscreens alone • surveys showed that sun exposure time prevent skin cancer, FDA considers the ‘‘helps protect against the harmful is prolonged for persons using new warning important for the effects of the sun’’ sunscreens compared to persons not protection of the public health. Another comment acknowledged that using sunscreens (Refs. 15 and 16). By FDA is proposing that the new facial makeups with sunscreen provide educating consumers about a sun warning be required on all OTC protection from sunburn, but that is not protection program, we believe sunscreen drug products except lip the primary reason why consumers use requiring this new proposed warning cosmetic-drug and lip protectant- these products. The comment sunscreen products subject to will decrease the likelihood of contended that requiring the ‘‘sunburn’’ § 352.52(f). FDA continues to believe consumers spending more time in the indication would be inappropriate and that all sunscreen products should have sun when using a sunscreen. misleading labeling for most facial labeling to ensure that consumers are The new proposed warning also makeups with sunscreen. The comment, adequately protected against informs consumers that use of instead, requested a claim such as sunscreens alone is not the sole measure overexposure to UV radiation (64 FR 27666 at 27673). Thus, sunscreen ‘‘protects against the harmful rays of the of protection from UV exposure, even sun.’’ with the use of high SPF products that products labeled for use only on specific FDA notes that the second comment provide UVA protection. Although it is small areas of the face and sold in small acknowledged that facial makeups with well established that sunscreens protect packages (i.e., sunscreen products sunscreen provide protection from against UV radiation, the following subject to § 352.52(f)) must include the sunburn. Not every consumer who uses factors affect the level of protection new warning. The only sunscreen provided by a sunscreen for each products not required to include the color cosmetics or facial makeups with new warning are those lip cosmetic- individual: sunscreen meets the following criteria: drug and lip protectant-sunscreen • Variations between individuals, • Has a dark skin type, or products subject to § 352.52(f), as • UV radiation absorption, • Uses these products solely to • proposed in § 352.52(f)(1)(ii). FDA is Ability of sunscreens to adhere to making this proposal because lip provide protection from sun damage and be absorbed by the skin, cosmetic and lip protectant products are that is not immediately recognizable. • Exposure conditions, and often sold in packages that are As noted in section III.D, comment 9 of • Conditions of use (e.g., inadequate substantially smaller than those of other this document, many consumers use application amount or reapplication products that fall under § 352.52(f). FDA facial products with sunscreen as their frequency). believes requiring the new warning on primary and only source of sunscreen Therefore, FDA agrees with the lip cosmetic-sunscreen and lip protection for that area of the body. As numerous groups that promote protectant-sunscreen products may discussed in section III.G, comment 16 sunscreen use as part of a total sun discourage manufacturers from of this document, sunscreen products protection program. marketing these products because it will be required to bear a claim of low, FDA reviewed the relationship requires a significant amount of labeling medium, high, or highest UVB sunburn between sunscreen use and skin cancer space. protection. FDA does not consider it incidence in the scientific literature and FDA has limited labeling inappropriate or misleading for color did not find confirmatory evidence that requirements as much as possible for cosmetic or facial makeup products sunscreens alone protect against the sunscreen products subject to containing sunscreens to have this development of skin cancer. The § 352.52(f). However, FDA believes sunburn protection claim of low, incidence of skin cancer continues to consumers are at great risk for UV- medium, high, or highest. rise in the United States. The incidence induced skin damage, including cancer, of the most serious form of skin cancer, on the face. Therefore, consumers who Sunscreen products that provide UVA malignant melanoma, grew 6.1 percent purchase products specifically for use radiation protection may also bear a per year during the 1970s (Refs. 14 and on the face need to be informed about claim about the level of protection. In 36). The rate is still rising an average 2.8 the information contained in the new addition, all OTC sunscreen products, percent annually, with a rate of 14.3 warning. Although these products are except lip cosmetic-drug and lip percent per 100,000 persons in 1997. marketed in small package sizes, FDA protectant-sunscreen products subject to Melanoma is one of the top 10 cancers, has determined that the products’ § 352.52(f), will be required to bear the by incidence, for persons with white labeling needs to include this important revised ‘‘sun alert’’ statement, which is skin. The American Cancer Society information in order to protect now included in the ‘‘Warnings’’ section (ACS) estimated the following statistics consumers. of the ‘‘Drug Facts’’ box. FDA considers concerning skin cancer in 2007 (Ref. (Comment 20) One comment stated the information in this new ‘‘Warnings’’ 37): that consumers who use color cosmetics statement much more beneficial to • More than 1 million new cases of or facial moisturizers with sunscreens consumers than the statements curable basal cell and squamous cell make the informed decision to purchase proposed by the comments. FDA carcinomas would be detected, them as an additional benefit to their rejected the terms ‘‘casual, incidental, • Approximately 59,940 new cases of cosmetic use. The comment contended and intermittent,’’ as explained in malignant melanoma would be that a significant number of people with section III.G, comment 17 of this diagnosed, and dark skin types, who do not burn easily, document.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49092 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

H. Comments on Directions for incidental sun exposure in addition to at 2 mg/cm2 (Refs. 49, 50, and 51). Sunscreen Drug Products intentional sun exposure, such as Gasparro notes that statements such as (Comment 21) Several comments sunbathing. ‘‘apply liberally and frequently’’ are too requested alternative directions for For these reasons, FDA considers it vague to be informative (Ref. 24). makeup with sunscreen products. One important that consumers using makeup FDA is concerned that, in practice, with sunscreen products not labeled for comment requested ‘‘apply smoothly or consumers may be getting less use only on specific small areas of the protection than the labeled SPF value evenly before sun exposure and/or as face recognize that these products are and believes that further information needed.’’ The comment added that sunscreens and use them appropriately should be included in the labeling for ‘‘before sun exposure’’ may not always to maximize UV protection. Therefore, sunscreen drug products to reduce the be appropriate as these makeup FDA is not proposing modified likelihood of underapplication. FDA products are not exclusively or even directions for these makeup with believes that this information is better primarily used for protection against sunscreen products. communicated as revised product sun exposure. A second comment (Comment 22) One comment directions rather than a warning. FDA requested ‘‘apply smoothly or evenly requested that FDA require sunscreen is, therefore, proposing to revise before sun exposure and reapply as manufacturers to provide accurate and § 352.52(d)(1). The directions will needed.’’ A third comment did not appropriate instructions about how continue to state that OTC sunscreen suggest any specific language, but much sunscreen should be applied to drug products should be applied requested flexibility to recognize the the body. The comment also suggested ‘‘liberally’’ or ‘‘generously’’ because it product’s primary use as a makeup, that a warning about the dangers of would be cumbersome to specify while providing adequate information sunburn from applying suboptimal quantitative amounts for all possible about the sunscreen component. This amounts be included in sunscreen body areas and the various uses on the comment added that the direction to product labeling. A second comment label. However, FDA is proposing to consult a doctor for children under 6 stated that it was not aware of any study make optional the directions in months of age was clearly unnecessary indicating that consumers use adequate § 352.52(d)(1)(i) to apply ‘‘evenly.’’ FDA for facial makeup with sunscreen amounts of sunscreen. The comment believes that this term, if used alone, because these products cannot supplied data and other information may not convey the appropriate message reasonably be expected to be used on concerning the dependency of the SPF to ensure that consumers apply children that age. value on the total quantity of sunscreen sufficient sunscreen. In addition, FDA is FDA agrees that flexibility is applied (Ref. 49). proposing to remove the term appropriate for the directions for Section 352.52(d)(1) currently ‘‘smoothly’’ from § 352.52(d)(1)(i) makeup with sunscreen products. provides manufacturers the option to because FDA considers that term to be Elsewhere in this document, FDA is select one or more of the following vague and it may have different proposing to allow labeling application terms for a sunscreen meanings to different consumers. FDA modifications for makeup with product: ‘‘liberally, generously, also believes this term is more likely to sunscreen products used only on smoothly, or evenly.’’ Manufacturers result in product underapplication. specific small areas of the face and sold may also include optional directions In addition to labeling directing in small packages (see section III.D, that state ‘‘[bullet] reapply as needed or consumers to apply sufficient amounts comment 9 of this document). Those after towel drying, swimming, or (select of sunscreen, FDA is also proposing to modifications include modified one of the following: ‘sweating’ or revise the labeling requirements directions for cosmetic lip products ‘perspiring’).’’ In the final rule, FDA had concerning reapplication of the containing sunscreen that are within the concluded that the directions in sunscreen product. In § 352.52(d) of the scope of proposed § 352.52(f). FDA is § 352.52(d)(1) to apply ‘‘liberally’’ or FM, the general reapplication statement not extending the proposed ‘‘generously’’ convey the appropriate ‘‘and as needed’’ was the only required modifications to all makeup with message to ensure that consumers information. FDA made specific sunscreen products. Makeup with adequately apply the sunscreen (64 FR reapplication directions in sunscreen products not labeled only for 27666 at 27679). § 352.52(d)(2) of the FM optional in an specific small areas of the face may be Several studies suggest that, in effort to equalize requirements between applied to a large area of the face or practice, consumers may apply amounts sunscreens with and without water other areas of the body. As explained of sunscreen below the density of 2 resistant claims (64 FR 27666 at 27681). later in this comment, FDA would have milligrams/square centimeter (mg/cm2), FDA now believes that more detailed concerns with the modifications being which is the amount of product required reapplication directions must be applied to these products. for the SPF determination in § 352.72(e) included on all OTC sunscreen Whether intentional or not, makeup (proposed § 352.71(e)). These data products, because sunscreens may be with sunscreen products may be the suggest that consumers may apply as underapplied as suggested by the primary sunscreen for many consumers. little as 0.5 to 1.0 mg/cm2 (Refs. 50 comments. A recent study examined sunscreen use through 54). One comment reported FDA came to this conclusion after patterns (Ref. 48). Participants were that, to achieve the rated protection over reviewing studies concerning sunscreen instructed to apply sunscreen every day. the whole body, a typical adult with a reapplication as well as Of those who used sunscreen surface area of 1.73 square meters (m2) recommendations of public health infrequently, the majority spent some would need to apply 35 milliliters (mL) organizations. Wright, et al. suggests time outdoors with 11 percent spending of sunscreen, roughly one-third of a 4 oz that inadvertent sunburn may be due to the majority of their time outdoors. bottle per application (Ref. 55). Studies the failure to use and reapply sunscreen These same participants explained that indicate that SPF values determined at appropriately (Ref. 56). Study subjects they did not believe sunscreen was an application rate of 1 mg/cm2 are who reapplied sunscreen every 1 to 2 necessary because of their planned approximately 50 percent of those hours and after swimming did not activities. The authors cited this finding determined at 2 mg/cm2, and when report sunburn. Rigel et al. reported in advocating educating consumers on applied at 0.65 mg/cm2, the SPF values that, even under intense solar the need for sunscreen for frequent are 20 to 30 percent of those determined conditions, those reapplying an SPF 15

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49093

sunscreen every 2 hours or sooner were one of the following: ‘40 minutes of’ or ‘80 protection under consumer use five times less likely to sunburn minutes of’ for products that satisfy either the conditions is affected by a number of compared to those who reapplied every water resistant or very water resistant test factors. Some of the key factors are 2.5 or more hours (Ref. 57). The AAD procedures in § 352.76, respectively] • Application density, (Refs. 38, 58, and 59), the ACS (Ref. 60), swimming or [select one of the following: • Reapplication frequency, ‘sweating’ or ‘perspiring’] and after towel • Skin type (e.g., burns easily versus and the EPA (Ref. 40) recommend drying. Otherwise, reapply at least every 2 reapplying sunscreens every 2 hours or never burns), hours’’. • Time of day during sun exposure, sooner and also recommend application (3) For products that do not satisfy the and to all exposed areas of the body (Refs. water resistant or very water resistant testing • 60, 61, and 62). procedures identified in § 352.76. The Geographical location during sun Because the frequency of application labeling states ‘‘[bullet] reapply at least every exposure. appears to be critical for proper 2 hours and after towel drying, swimming, or Thus, SPF values reflect relative and not protection, FDA is proposing to add the [select one of the following: ‘sweating’ or absolute levels of sunburn protection. statement ‘‘apply and reapply as ‘perspiring’]’’. Although SPF values do not convey directed to avoid lowering protection.’’ As discussed in the FM (64 FR 27666 at actual levels of sunburn protection, In addition, FDA is proposing to further 27679), manufacturers who have data to when comparing multiple sunscreen revise the directions in § 352.52(d) to support different reapplication products, SPF values enable consumers include the following reapplication directions based on specific to determine which products provide statement: ‘‘reapply at least every 2 substantiation information may submit the most sunburn protection. For hours.’’ Likewise, for those products the information for approval of those example, FDA believes most consumers making a water resistant claim, FDA is directions via an NDA deviation as would correctly identify an SPF 20 proposing to include the number of provided in § 330.11 (21 CFR 330.11). product as providing more sunburn minutes (i.e., 40 or 80) that the product protection than an SPF 10 product. I. General Comments on SPF Testing Thus, lowering the sunscreen maintains its water resistance before the Procedure ‘‘swimming/sweating’’ term. FDA application density would not be believes these additional proposed (Comment 23) One comment necessary to more accurately reflect the directions will alert consumers about suggested that the SPF test incorporate degree of relative sunburn protection. the hazards of using insufficient an amount of product that more closely FDA agrees that, in addition to amounts of sunscreen product and reflects the amount applied by bringing SPF values closer to encourage reapplication after the consumers. More specifically, the representing absolute levels of appropriate time. FDA considers these comment requested that FDA replace protection, lowering the sunscreen 2 specific, informative reapplication the 2 mg/cm required in § 352.72(e) application density might also reduce statements, instead of ‘‘and as needed,’’ (proposed § 352.70(c)(5)) to a value some of the inaccuracies and limitations 2 to be necessary on all OTC sunscreen between 0.5 and 1.0 mg/cm . The encountered when testing high SPF products. FDA is also proposing the comment argued that the protection sunscreen products. Thus, FDA invites optional direction ‘‘apply to all skin afforded during actual usage may be interested parties to submit data exposed to the sun.’’ FDA is proposing only one-quarter to one-half the labeled supporting a smaller application density that this direction be optional because SPF value (see section III.H, comment for SPF testing of all sunscreen dosage we believe most consumers know to 22 of this document). The comment also forms in accordance with § 352.77. apply sunscreen to all exposed skin. suggested that SPF could be stated using However, developing a single global However, if a sunscreen product can descriptive terms, such as ‘‘light,’’ method and labeling would require a accommodate this direction, it will ‘‘moderate,’’ or ‘‘heavy’’ protection, coordinated effort between the serve to remind consumers that all instead of a numerical value. regulatory agencies in many countries exposed skin is susceptible to UV FDA is not proposing the suggested around the world. Because FDA does damage. These proposed directions, as a change in test method at this time. This not have data to validate the SPF test whole, should serve to better protect issue was discussed in detail in the using a lowering sunscreen density, consumers, particularly those who tend TFM (58 FR 28194 at 28264 to 28266). FDA is proposing directions that we to underapply sunscreen, from The majority of comments advocated believe will encourage consumers to continuing the use of an application apply greater densities of sunscreen overexposure to the sun. 2 2 Accordingly, FDA is proposing to density of 2 mg/cm . The current (i.e., closer to 2 mg/cm ) (see section change § 352.52(d) to read as follows: comment did not provide data III.H, comment 22 of this document). (d) Directions. *** demonstrating the suitability of a FDA does not find that there are (1) For products containing any ingredient smaller test amount. FDA is concerned sufficient benefits for using descriptors in § 352.10. (i) The labeling states ‘‘[bullet] that a uniform distribution of sunscreen instead of numerical values for SPF on apply [select one of the following: ‘liberally’ over the test area might be difficult the PDP. Consumers are familiar with or ‘generously’] [and, as an option: ‘and using a smaller amount of sunscreen. numerical SPF values from over 20 evenly’] [insert appropriate time interval, if a Further, the standard application years of usage. As described in section waiting period is needed] before sun III.G, comment 16 of this document, exposure’’. density used worldwide in the SPF test 2 (ii) The labeling states ‘‘[bullet] apply and is 2 mg/cm (Ref. 63). FDA believes that the use of descriptors reapply as directed to avoid lowering FDA agrees that SPF values do not in combination with numerical values protection’’. reflect exact levels of sunburn on the PDP may be beneficial to (iii) As an option, the labeling may state protection that consumers receive under consumer understanding of the level of ‘‘[bullet] apply to all skin exposed to the actual use conditions. The required SPF sunburn protection provided by a sun’’. test is a clinical test conducted with product. Thus, as explained in comment (iv) The labeling states ‘‘[bullet] children strict control over factors such as 16, FDA is proposing to include a under 6 months of age: ask a doctor’’. (2) For products that satisfy the water product application density. However, descriptive term of relative sunburn resistant or very water resistant testing under actual use conditions, these protection (i.e., low, medium, high, or procedures identified in § 352.76. The factors are not controlled and vary highest) with the proposed sunburn labeling states ‘‘[bullet] reapply after [select greatly. The actual level of sunburn protection statement in the ‘‘Uses’’

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49094 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

section and on the PDP. The intent of as there would be less than a 95 percent using standard probability computer this dual descriptive and numerical probability the actual SPF value was software, FDA calculates that the values sunburn protection measure is to allow higher than the expected SPF value. If for the maximum number of failures in consumers to more easily differentiate less than six subsites show perceptible table 2 of this document for subjects one the level of sunburn protection provided erythema, the product passes, as there through five should be 0, 1, 2, 4, and 5, by different sunscreen products. In would be greater than a 95 percent respectively, rather than the values addition, this proposed labeling for probability that the actual SPF value provided by the comment. sunburn protection is similar to the was more than the expected SPF value. FDA would also consider three proposed UVA protection labeling (see The comment proposed the following: modifications to the method described section III.G, comment 14 of this by the comment and invites comment. document). TABLE 2.—PROBABILITY TABLE First, each subject may have test FDA is also aware of sunscreen drug successes and failures due to multiple products marketed in dosage forms that No. of subjects Maximum no. Probability subsites on each subject. Statistically, may not be addressed by current SPF of failures these will not be independent testing procedures. The SPF testing observations, which is a condition 1 (n=4) 0 0.06251 procedure described in § 352.72 2 (n=8) 2 0.0352 needed for a binomial probability (proposed § 352.70) references oils, 3 (n=12) 3 0.0200 calculation. Therefore, FDA is lotions, creams, gels, butters, pastes, and 4 (n=16) 5 0.0383 considering that a test panel should ointments. FDA invites interested 5 (n=20) 5 0.0207 consist of 20 to 25 subjects and that only parties to submit SPF testing one site be tested on each subject. A 1 n is not sufficient to make a 95 percent modifications for new dosage forms prediction pass/fail determination would be made (e.g., mousses, foams, and towelettes) in for each individual. accordance with § 352.77. The comment further proposed that if Second, as an alternate, a double (Comment 24) One comment all eight subsites of the first two subjects sampling plan based on Taylor’s Guide recommended a pass/fail (binomial) test pass, then the product passes and the to Acceptance Sampling may replace to determine SPF values (Ref. 49). The remaining three subjects would not be the five-layered plan proposed by the test would demonstrate that subjects evaluated. The probability of this comment (Ref. 64). With the double have no reaction to a quantity of UV happening would be 1/256 unless the sampling plan, two subjects are tested energy equivalent to an expected SPF product is over the expected SPF value. simultaneously with up to a maximum value (for products passing the test). For FDA agrees that, currently, there may of four subjects, each having four example, subjects being tested with a not be enough experience and test data subsites tested. If no more than one of product with an expected SPF value of for products with SPF values of 30 and the first eight subsites has perceptible 30 would be dosed only at the SPF 30 over on which to determine the sample erythema, the product passes. If three to level, and the product would either pass size needed to obtain an acceptable 95 eight subsites have perceptible or fail. A product passing this test percent confidence interval. As erythema, the product fails. If exactly would actually have an SPF value of 30 discussed in section III.L, comment 37 two of the eight subsites have or over, whereas a product failing this of this document, to account for perceptible erythema, then the second test would have an SPF value below 30. increased variability in SPF values for group of two subjects is tested. If two to The comment argued that while the sunscreens with SPF values over 30, four subsites from four subjects have monograph SPF test is probably FDA proposes to increase the sample perceptible erythema, the product adequate for products with low SPF size to at least 25 subjects. Therefore, passes. Otherwise, the product fails. values, it is not adequate for testing high the comment may be correct in arguing According to this scheme, if probability SPF products because differences in that large numbers of subjects may be p = 0.10 that the product tested would solar simulators can provide as much as required for testing products with high produce any recognizable erythema, a 200 percent variation in results SPF values. FDA believes that the pass/ then the probability = 0.95 that the depending on the formulation. The fail test has merit and could provide a product will pass. If probability p = 0.5 comment further argued that an reasonable substitute for the current SPF that the product tested would produce impossibly high number of subjects method for products with expected SPF any recognizable erythema, then the would be required for the current SPF value of 30 or higher. However, before probability = 0.05 that the product will method to obtain a 95 percent the method can be accepted, method pass. confidence level and that the test validation data are required that Third, an alternative to the probability exposes subjects to a potentially demonstrate the method can be calculation is a margin of error dangerous condition, sunburn. performed satisfactorily by multiple approach. With this method, a margin of According to the comment, the laboratories using the same sunscreen error for the expected SPF value is average MED for each skin type can be formulation(s). FDA invites such data. defined before testing. The margin of predicted from existing solar simulator If the pass/fail method is accepted, error is used to determine the calibration data. During the pass/fail FDA may stipulate that the method be tolerability interval around the expected test, each test subject is screened for used only for products with SPF values SPF value. The 90 percent confidence skin type and then given a first day of 30 and higher because of the large interval for the product’s test result (one range of energy that does not exceed the number of subjects that would be result per subject) must fall within the expected MED. The comment proposed required for high SPF products under tolerability interval to be labeled with using a panel of five subjects. Using the the current test method. A pass/fail that SPF value. For example, if a 10 MED information obtained on the first method would require fewer test percent margin of error is claimed for a day, each subject is given four UV subjects. Low SPF products can be product with an expected SPF value of radiation exposures corresponding to adequately tested under the current 40, then the tolerability interval would the expected SPF value. Each subsite is method without large numbers of be 40 ± 4, or 36 to 44. If the related 90 then evaluated for erythema. If six or subjects. In addition, FDA would likely percent confidence interval is from 37 to more of the 20 subsites show require that all 20 subsites be evaluated 43, an SPF value of 40 is assigned to the perceptible erythema, the product fails, even if the first 2 subjects pass. Further, product. If the related 90 percent

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49095

confidence interval is from 35 to 45, an method require further exploration. readings at 16 and 24 hours. Thus, FDA SPF value of 40 could not be assigned Items such as the cut-off to define is proposing to revise the MED to the product and the product may be photostability need further explanation determination time in §§ 352.72(h) and retested at an expected SPF of 30. and validation. It should also be pointed 352.73(c) (proposed §§ 352.70(c)(8) and FDA invites discussion of these out that the current SPF test method 352.70(d)(3), respectively) from ‘‘22 to suggested modifications to the does not directly measure 24 hours’’ to ‘‘16 to 24 hours.’’ comment’s pass/fail method for testing photostability, but it accounts for sunscreen drug products having an SPF photostability. More specifically, the J. Comments on the Sunscreen Standard value of 30 or higher. SPF value is determined after a for SPF Testing Procedure (Comment 25) One comment sunscreen is exposed to UV radiation, so (Comment 27) Several comments described an in vitro method it the SPF represents UVB protection suggested that standard controls with developed for simultaneously predicting provided by whatever fraction of the SPF values of 15 or higher be developed SPF and assessing photostability. The sunscreen has not decomposed. to test high SPF sunscreen products. method utilizes a 150 watt xenon arc FDA agrees that in vitro tests are One comment stated that such standards lamp to irradiate sunscreen applied at a generally rapid and less expensive than would improve test accuracy and 2 level of 1 to 2 mg/cm to a flat collagen in vivo tests and, for SPF measurements, provide a consistent and adequate membrane substrate placed in the would reduce exposure of human benchmark for compliance. One opening of an integrating sphere subjects to UV radiation. FDA is willing comment mentioned use of a control attached to a spectroradiometer. The to consider alternate methods for SPF SPF 15 formula routinely in SPF spectral irradiance of the source and the testing if they are adequately supported evaluation and considered it a more spectral irradiance of the substrate alone with data and are shown to be valuable control than the 8-percent are measured from 290 to 400 nm, at 1 equivalent to established in vivo homosalate SPF 4 standard. Another nm intervals. The spectral irradiance methods by collaborative studies. If the comment supplied ‘‘round-robin,’’ transmitted by the sunscreen/substrate methods are equivalent, then the same collaborative SPF testing data from 7 combination is measured at 1 minute SPF values should be determined for laboratories on a total of 153 subjects intervals until the total erythemal- each sunscreen tested according to the with 2 potential SPF 15 sunscreen effective dose transmitted by the SPF method and the alternate method. standard preparations, ‘‘Formulation A’’ sunscreen exceeds 1 MED, where 1 MED The comments have not provided data on 147 subjects and ‘‘Formulation B’’ on equals 0.02 erythema-effective Joules (J)/ from such studies. Therefore, FDA is not 146 subjects (Refs. 13, 68, and 69). The cm2. Each 1 minute interval represents proposing to include the described in comment concluded that differences two to three MEDs. The time course of vitro method in the monograph at this between the two preparations were not the sunscreen’s SPF is then computed time. significant (p=0.653) but ‘‘Formulation (Ref. 65). This information reveals the (Comment 26) Several comments B’’ was preferred due to its less complex photostability of a sunscreen. If a urged FDA to revise § 352.72(h) and formula and slightly more consistent sunscreen is photostable, it will not reinstate the requirement for results. The comment added that the decompose when exposed to UV determining MED at 16 to 24 hours after data showed that different laboratories radiation, and the SPF will not change exposure, rather than 22 to 24 hours. with increasing UV exposure. If a The comments submitted data showing can obtain valid, reproducible results sunscreen is not photostable, it will that, for an SPF 30 product and for the when testing high SPF sunscreens. decompose when exposed to UV 8 percent homosalate standard, Another comment stated that it radiation, and the SPF will decrease determining the MED at 16 or 24 hours provided test results on 20 subjects with increasing UV exposure. Another does not result in any clinical or using an SPF 25 product as the control comment asked FDA to consider statistical difference in the SPF (Refs. 66 (Ref. 70). Three comments suggested replacing the human SPF test with and 67). Comments argued that that the European Cosmetic, Toiletry, equivalent in vitro technology and immediate pigmentation fades rapidly and Perfumery Association (COLIPA) chemical engineering, but did not and does not interfere with MED ‘‘European low SPF Standard Code suggest a suitable method. readings. One comment further argued Number COL492/1 (formerly the DIN FDA does not agree that an in vitro that the 16 to 24 hour time is standard)’’ be included in the OTC method is adequate to replace the in universally accepted by the European sunscreen drug product monograph as a vivo SPF test. In vitro tests are generally Union, Australia, and Japan and FDA permissible standard sunscreen inadequate as the sole measure of SPF should adopt this time in the interest of preparation, in addition to the 8-percent because substrates cannot mimic international harmonization. homosalate standard, and that either sweating, skin absorption, or certain The Panel recommended that the standard should be allowed in the SPF interactions with skin that influence MED be evaluated 16 to 24 hours after testing procedures. The comments SPF. Some sunscreen ingredients do not exposure (43 FR 38206 at 38262). FDA contended that this approach will serve behave similarly in vitro and in vivo. At proposed a post exposure time of 22 to to permit international marketing and this time, the comment’s method has 24 hours based upon information eliminate duplicative testing. Another not been validated, and the chosen provided by comments to the Panel’s comment asked FDA to adopt the JCIA substrate has not been demonstrated to report that immediate pigmentation may SPF 15 ‘‘P3’’ standard, but did not possess penetration characteristics and persist with higher doses of UV provide supporting data. surface chemistry similar to human radiation up to 24 hours or, in some The comment concerning the SPF 25 skin. cases, for 36 to 48 hours after prolonged control provided data from comparative The described in vitro method does exposure (58 FR 28194 at 28268 to tests on 20 subjects, using the 8-percent have potential utility for measuring 28269). Comments had indicated that homosalate standard, an SPF 15 photostability of a sunscreen product. immediate pigmentation might interfere sunscreen drug product, and an SPF 25 Measuring the erythemal-effective dose with an investigator’s perception of sunscreen drug product (Ref. 70). FDA transmitted through the sunscreen in minimally perceptible erythema. finds that this study is inadequate to vitro over time seems like a reasonable FDA agrees that these new data show support the comment’s request because approach. However, portions of the no significant difference in MED the study did not do the following:

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49096 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

• Include sufficient numbers of SPF ranges. The comment claimed that values over 15 may be erroneous in subjects, the 8-percent homosalate SPF 4 other studies. Better assurance of an • Address suitability of the standard standard is appropriate to test products accurate SPF value is obtained by using across different laboratories, and at any SPF level and that the choice of a standard that is closer in SPF value to • Document some properties required whether to use the SPF 4 or SPF 15 the sunscreen product being tested. in a sunscreen standard to test high SPF control formulation should rest with the The use of an SPF 15 standard would sunscreen products. manufacturer. Several other comments be reasonable to test products with SPF The following properties of a agreed with this comment. values below 15. SPF 15 is in the sunscreen standard were not addressed Another comment provided data middle (geometrically) of the 4 to 50 but need to be addressed: using the 8-percent homosalate standard range. The ratio of SPF 15 to SPF 4 is • Low level of interlaboratory to test product formulations with 3.75, and the ratio of SPF 50 to SPF 15 variation, estimated SPF values of 15, 30, and 45 is 3.33. Thus, there would be equal • Sensitivity to experimental error, on 20 subjects (Ref. 67). The comment coverage of all ranges. Therefore, FDA is and concluded that the data showed testing proposing that Formulation B may be • Ease of preparation with a procedures in the FM can differentiate used to test sunscreen drug products reasonable degree of accuracy. high SPF sunscreens using the with SPF 2 and over, and is required for These data are also needed for the JCIA homosalate SPF 4 standard. The testing sunscreen drug products with standard. comment requested that the homosalate SPF over 15 (proposed Although comments provided data on SPF 4 standard be allowed to be used § 352.70(a)(1)(ii)). The 8-percent 20 subjects in each of 4 laboratories for products with an SPF value over or homosalate standard may be used for using the COLIPA COL492/1 standard, below 15. testing sunscreen drug products with FDA is not proposing to include this FDA does not consider the data SPF of 2 to 15. standard as an alternate to the 8-percent adequate to support the suggestion that (Comment 29) Several comments homosalate standard because we do not the 8-percent homosalate standard suggested that a modern, HPLC method believe that using the COL492/1 currently used to evaluate sunscreen is superior to the older standard will make the monograph drug products with SPF values up to 15 spectrophotometric assay in § 352.70(c) method comparable to the European is equally applicable to products with of the FM. One comment provided method, as other differences exist SPF values over 15 (Ref. 67). The study technical information about the HPLC between the two methods. For example, had the following deficiencies: method and stated that it is now the monograph method requires 20 • Did not include sufficient numbers commonly used by analytical evaluable subjects, while the European of subjects, laboratories to assay sunscreen method requires only 10 evaluable • Did not address suitability of the formulations (Ref. 71). Although this subjects. Therefore, the COL492/1 standard across different laboratories, HPLC assay method was used in the standard is a valid standard under the and study of two SPF 15 sunscreen standard • European method but may not be a valid Did not document certain preparations (see section III.J, comment standard under the monograph method. properties required in a sunscreen 27 of this document), one comment Finally, FDA finds that the 8-percent standard to test high SPF sunscreen noted that there are limited data on this homosalate standard is a suitable products. method with the SPF 15 control control for testing sunscreen drug The following sunscreen standard formulation because FDA has not yet products with SPF 15 or below (see properties were not addressed but need published this formula as an accepted section III.J, comment 28 of this to be addressed: • standard. document). Low level of interlaboratory FDA agrees that an HPLC method is FDA agrees with the comment that the variation, and • superior to the spectrophotometric submitted collaborative data from seven Sensitivity to experimental error. method, which was originally published FDA agrees that the two standards are laboratories support ‘‘Formulation B’’ as by FDA in 1978, in specificity and method controls rather than calibration an appropriate SPF 15 sunscreen precision. Validation data provided by tools. As such, the standard used should standard. The mean SPF for the comment documented the following: ‘‘Formulation B’’ was 16.3 in 146 approximate the expected SPF of the • Specificity, subjects tested, with 1.7 percent product being tested to better verify that • Accuracy, standard error of the mean, and all aspects of the testing method are • Limit of detection, laboratory means ranging from SPF 15.6 performing properly at the expected SPF • Linearity, to 18.5. Therefore, FDA is proposing to level. • Precision, and include the ‘‘Formulation B’’ SPF 15 Using the SPF 4 standard to measure • Reproducibility of the method. standard in the FM to be used for SPF values over 15 is more likely to The validation data included sunscreen drug products with an SPF produce erroneous results than using a chromatograms and demonstrated that value over 15 (optional for SPF values standard with an SPF of 15. In the HPLC method is suitable for both of 2 to 15). measuring SPF values over 15, much the SPF 4 and SPF 15 standards. (Comment 28) One comment noted higher light energies (J/cm2) are used in Further, FDA validated the method in that there are two recognized standard comparison to measuring SPF values its laboratories and concludes that the control formulations: below 15. Problems in the accurate method is acceptable for quality control 1. An 8-percent homosalate quantitation of high light intensities and regulatory purposes (Ref. 72). preparation with an SPF value of 4 may not be detected if the SPF 4 Finally, the spectrophotometric method (§ 352.70(b) of the FM), and standard is used for SPF values over 15. has not been validated for the SPF 15 2. Formulation B (padimate O/ While the SPF 4 standard may give standard, and the HPLC method has oxybenzone) with an SPF value of 15. acceptable results for products with SPF been validated for both the SPF 4 and The comment stated that the function of values over 15 in some studies, the SPF 15 standards. Therefore, FDA is the standard formulation is quality extrapolation of these results to proposing to revise § 352.70 to replace assurance for method control and not as approximately 4 to 13 fold higher light the outdated spectrophotometric a calibration standard to bracket specific energies used to test products with SPF method with the HPLC method and to

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49097

use the HPLC method to assay both the and is not requiring that a specific as the Percentage Relative Cumulative SPF 4 and SPF 15 standards. source maintain and supply the Erythemal Effectiveness (%RCEE). (Comment 30) Two comments sunscreen standard at this time. According to the COLIPA specifications disagreed with the requirement in and consistent with § 352.71, K. Comments on Artificial Light Sources § 352.70(a) for concomitant use of a wavelengths below 290 nm should be for SPF Testing Procedure standard sunscreen for each SPF test. excluded from any source by One comment suggested that a standard (Comment 32) Several comments appropriate filters. Likewise, could be run twice yearly. Another suggested that FDA replace the wavelengths above 400 nm should be comment suggested that data to evaluate specifications in § 352.71 that state ‘‘sun limited as much as possible and are not proper laboratory test procedures could at a zenith angle of 10°’’ and ‘‘less than included in the calculation of %RCEE. be obtained from panels of a standard 1 percent of its total energy output Because RCEE values are calculated as run as part of ‘‘the ongoing laboratory contributed by nonsolar wavelengths relative percentages, measuring the operation.’’ A third comment stated that shorter than 290 nm’’ with the COLIPA spectral irradiance in absolute energy a standard preparation should be run table of ‘‘percent erythemal units is not necessary. Relative units are each time an SPF determination is contribution’’ as the spectral power sufficient. The spectral irradiance of the made. distribution standard for the light source source is multiplied by the Commission FDA discussed this issue in comment used in the SPF test procedures (Ref. International de L’Eclairage (CIE) (1998) 78 of the TFM (58 FR 28194 at 28253 73). The comments suggested that the standard skin erythemal action to 28254). FDA disagreed with one spectra of currently used solar spectrum to obtain the erythemal comment that the standard could be run simulators (especially around 290 nm effectiveness of the source. The spectral once or twice a year and reaffirmed the and above 350 nm) could cause erythemal effectiveness values of the Panel’s recommendation that overestimation of SPF values for high source spectrum are then integrated concomitant testing is necessary in SPF SPF sunscreens. Because shorter from 250 nm to the various successive determinations to ensure uniform wavelengths can make a very large reference wavelength values shown in evaluation of OTC sunscreen drug contribution to erythema, the comments the COLIPA table in order to produce products and to serve as an internal stated that small errors in the 290 nm the cumulative erythemal effectiveness indicator of experimental errors. The region of solar simulator spectra could for each spectral waveband, and the comments requesting a change did not have considerable effects. The total erythemal effectiveness is provide any supporting data. In the comments noted that spectral power calculated up to 400 nm. Finally, the absence of supporting data, FDA is not deficiencies above 350 nm may give %RCEE is calculated at the reference persuaded to change the concomitant artificially high SPF values for waveband as the percentage ratio of the use requirement in § 352.70(a). sunscreen drug products that absorb cumulative erythemal effectiveness in (Comment 31) One comment poorly in the long wavelength UVA each of these wavebands to the total suggested that there is a need for a region. integrated value from 250 nm to 400 specific source to maintain and supply The comments added that there is nm. sunscreen standards. The comment general agreement in the industry that Based on these calculations, the contended that a few testing laboratories § 352.71 should be revised to permit COLIPA table includes limits up to 400 are reporting differences in the tested compliance with the COLIPA standard nm. In contrast, when FDA requested SPF of the 8-percent homosalate for solar simulators. The comments comments on this issue, we included a standard preparation depending on further recommended one modification modified COLIPA table that includes whether the standard is prepared by the to the COLIPA standard: The energy for limits up to 350 nm (65 FR 36319 at laboratory or purchased from one wavelengths below 290 nm should be 36321). However, the modified COLIPA company that manufactured this limited to ‘‘less than 0.1 percent’’ rather table published by FDA was erroneous. standard. The comment stated that than ‘‘less than 1.0 percent,’’ as stated FDA agrees with the comment (and either the testing procedures or the in the COLIPA standard. The comments COLIPA) that it is necessary to include standard itself have changed since the stated that a more restrictive all UV erythemal wavelengths (i.e., up original formula was published (earlier specification of ‘‘0.01 percent,’’ as to 400 nm) when standardizing solar standard SPF values were 3.7/3.8 to 4.2/ mentioned by FDA (65 FR 36319 at simulator output. As argued by the 4.3 with an average of 4.1, while current 36321), would result more in testing the comment, the erythemal contribution values are 4.3 to 4.9/5.0). limits of the measurement from long-wavelength UVA radiation Data supporting the reliability and spectroradiometer rather than the true (i.e., 350 nm to 400 nm) can become wide acceptance of the 8-percent output of the solar simulator. One important when a high SPF product is homosalate standard preparation were comment that supported the COLIPA tested. However, FDA believes that the previously discussed in the TFM (58 FR standard subsequently suggested that limits for the 290 to 350 waveband 28194 at 28250 through 28252). The the spectral limits be further narrowed should be changed from 93.5 to 99.0 comment did not provide any data to to prevent excessive variability of SPF percent to 93.5 to 98.5 percent. This support its contention concerning values for certain sunscreen products modification will address some of the discrepancies in the SPF of 8-percent (Ref. 74). errors in SPF that are attributed to the homosalate standard preparations and One comment discussed the lack of match between the solar FDA is not aware of any new data that calculations to obtain the source simulator and actual solar spectra. FDA support the need for a specific source to spectral specification according to invites comments on these proposed maintain and supply this standard. The COLIPA (Ref. 73). In the COLIPA table, changes. standard is a control to validate the the source spectral specification is FDA does not agree, at this time, with testing procedure, equipment, and described in terms of cumulative the comment’s suggestion to further facilities rather than a calibration tool erythemal effectiveness by successive narrow the COLIPA standard to the for setting SPF values of sunscreen wavebands. The erythemal effectiveness spectral limits that it proposed. The products. FDA considers the parameters of each waveband is expressed as a comment based its suggestion on a established in § 352.70 of the FM percentage of the total erythemal theoretical argument and did not supply adequate to assure a uniform standard effectiveness from 250 nm to 400 nm, or the complete emission spectra of the

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49098 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

four solar simulators used in its two solar simulators operated at or below a requirement that places a quantifiable referenced studies. There may be total irradiance limit of 1,500 W/m2. limit of 20 percent on time related significant differences in the 290 to 350 The comments added that the ‘‘out of fluctuations of the radiation emissions nm range in these studies that can band’’ specification is not possible with of the solar simulator. account for the reported differences in existing solar simulators and new Accordingly, FDA is proposing to SPF test results. Further, FDA has systems would need to be designed, revise portions of § 352.71 (proposed concerns about the ability of currently tested, manufactured, and distributed to § 352.70(b)(2)) to read as follows: used solar simulators to meet the provide equipment capable of meeting (2) Operation. A solar simulator should comment’s suggested spectral standard this specification. The comments have no significant time related fluctuations and invites comments on the changes concluded that replacing the ‘‘out of (within 20 percent) in radiation emissions suggested by the comment. band’’ specification with a limit would after an appropriate warmup time and good FDA agrees with the comments that improve the testing of all products, beam uniformity (within 20 percent) in the the COLIPA approach provides a more exposure plane. The average delivered dose including high SPF products. to the UV exposure site must be within 10 appropriate description for solar FDA believes that it is important to percent of the prescribed dose. simulators. FDA’s original proposal that limit total energy delivered to the skin (Comment 35) Several comments solar simulators have a spectral power during the SPF test so that skin recommended that the last sentence of distribution ‘‘similar to sunlight at a temperature does not reach a point that ° § 352.71 be modified to include zenith angle of 10 ’’ is nonquantitative may compromise dose reciprocity. FDA additional requirements for the periodic and may not be practical, considering concurs with the comments and is testing of solar simulators. The proposing to replace the ‘‘out of band’’ the types of solar simulators that are comments suggested that periodic specification in § 352.71 (proposed generally available. Accordingly, FDA is measurements be made twice a year and § 352.70(b)) with a limit of 1,500 W/m2 proposing to revise the first part of that measurements be done after on total solar simulator irradiance § 352.71 (proposed § 352.70(b)) as changes in the optical filtering follows: between 250 and 1,400 nm. (Comment 34) Two comments components. (b) Light source (solar simulator)—(1) FDA agrees with the comments and is Emission spectrum. A solar simulator used recommended that FDA change the for determining the SPF of a sunscreen drug solar simulator specification in § 352.71 proposing to revise the last part of product should be filtered so that it provides from ‘‘good beam uniformity (within 10 § 352.71 (proposed § 352.70(b)(3)) to a continuous emission spectrum from 290 to percent) in the exposure plane’’ to ‘‘the read as follows: 400 nanometers (nm) with * * * the delivered dose to the UV exposure sites (3) Periodic measurement. To ensure that following percentage of erythema-effective the solar simulator delivers the appropriate radiation in each specified range of be within 10 percent of the prescribed spectrum of UV radiation, the emission wavelengths: dose with good beam uniformity’’ spectrum of the solar simulator must be (without defining ‘‘good beam measured every 6 months with an SOLAR SIMULATOR EMISSION uniformity’’). The comments contended appropriate and accurately calibrated SPECTRUM that although ‘‘reasonable’’ or ‘‘good’’ spectroradiometer system (results should be beam uniformity is desirable, beam traceable to the National Institute for Wavelength range Percent erythemal uniformity within 10 percent is virtually Standards and Technology). In addition, the (nm) contribution impossible to measure or achieve for the solar simulator must be recalibrated if there vast majority of solar simulators. is any change in the lamp bulb or the optical < 290 < 0.1 FDA agrees that ‘‘dose’’ accuracy is a filtering components (i.e., filters, mirrors, 290–310 46.0–67.0 lenses, collimating devices, or focusing critical variable and the delivered dose devices). Daily solar simulator radiation 290–320 80.0–91.0 to the UV exposure sites should be 290–330 86.5–95.0 intensity should be monitored with a 290–340 90.5–97.0 within 10 percent of the prescribed broadband radiometric device that is 290–350 93.5–98.5 dose. Because FDA considers sensitive primarily to UV radiation. The 290–400 93.5–100.0 quantification of ‘‘good beam broadband radiometric device should be uniformity’’ to be an important issue, it calibrated using side by side comparison (Comment 33) Several comments is keeping a specification for this with the spectroradiometer at the time of the suggested the following revisions to the parameter. However, FDA believes that semiannual spectroradiometric measurement light source (solar simulator) a specification of 20 percent is more of the solar simulator. If a lamp must be replaced due to failure or aging during a requirements in § 352.71: achievable than the proposed 10 • phototest, broadband device readings Delete the ‘‘out of band’’ percent. Beam uniformity can be consistent with those obtained for the specification that not more than 5 measured with broadband UV detectors original calibrated lamp will suffice until percent of a solar simulator’s total that have been modified to provide a measurements can be performed with the energy output can be contributed by small input aperture to the detector. For spectroradiometer at the earliest possible wavelengths longer than 400 nm. example, for a single beam simulator opportunity. • In place of this 5 percent ‘‘out of with a subsite exposure area of band’’ limitation, allow a limit such as approximately 1 cm2, an appropriate L. Comments on the Design/Analysis of 1,250 to 1,500 watts/square meter (W/ input aperture would be 0.25 cm2. Beam SPF Testing Procedure m2) on the total solar simulator uniformity can then be checked by (Comment 36) Several comments irradiance delivered to the skin for all making a measurement in the center of contended that the series of seven wavelengths. each of the four quadrants of the exposure doses in § 352.73(c) should be One comment provided data exposure field. These readings should modified to eliminate the two doses comparing solar simulators with and be within 20 percent of the peak placed symmetrically around the without a 50 percent neutral density reading. The same principle can be middle exposure. One comment filter to demonstrate that there is no applied to larger exposure fields. provided data comparing the seven- measurable impact of heat load on the Additionally, the average of these four exposure series against the five- outcome of SPF testing (Ref. 13). The readings should be within 10 percent of exposure series and concluded that the comment stated that thermal overload the prescribed dose for a given exposure seven-exposure series did not increase does not occur for COLIPA-compliant site. In addition, FDA is proposing a the precision of the test (Ref. 66).

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49099

Comments also argued that the seven- required for high SPF products and section. If more than 5 subjects are rejected exposure series would require longer proposed a binomial test method to based on paragraph (c)(9) of this section, the testing times, thus increasing exposure reduce the number of subjects (see panel is disqualified, and a new panel must risk and discomfort to subjects, and that section III.I, comment 24 of this be created. (ii) For products with an expected SPF of the five-exposure series is as accurate as document). Another comment stated 30 or over. A test panel shall consist of 25 the seven-exposure series even at high that the 20 of 25 subject limitation may to 30 subjects with at least 25 subjects who SPF values. be an issue for products with high SPF produce valid data for analysis. Data are FDA discussed its rationale for seven values due to the high variability in the valid unless rejected in accordance with versus five exposure doses in the TFM responses obtained and suggested that paragraph (c)(9) of this section. If more than (58 FR 28194 at 28269 to 28272). FDA the number of subjects be increased 5 subjects are rejected based on paragraph sought an exposure format that would when evaluating sunscreen products (c)(9) of this section, the panel is provide better accuracy and precision to with high SPF values. disqualified, and a new panel must be SPF measurements, particularly at As discussed in section III.I, comment created. higher SPF values. FDA reasoned that 24 of this document, the binomial test In the 1978 advance notice of the seven-exposure series in § 352.73(c), method deserves further investigation proposed rulemaking (ANPRM), the with two additional exposures and may prove to be a reasonable Panel recommended that studies enroll at least 20 subjects, adding that ‘‘the symmetrically placed around the approach as additional data and ± middle exposure of the geometric series, experience become available. In standard error shall not exceed 5 would increase precision and eliminate addition, based on the current SPF test percent of the mean’’ (43 FR 38206 at possible overestimation of the true SPF method, FDA agrees with the comment 38261). Following publication of the value of a product with a high SPF. recommending deletion of the ANPRM, FDA held a public meeting on FDA has evaluated the data and other requirement to fix the number of January 26, 1988 (52 FR 33598 at 33600 information submitted by the comments subjects per panel in advance. This to 33601). During that meeting, and agrees they demonstrate that the requirement is unnecessary because the attendees argued the following four additional two exposure doses do not panel is limited to a range of 20 to 25 points related to the number of subjects: make the test more precise. Therefore, subjects (under current § 352.72(g)). 1. Test panels should consist of at FDA is proposing to modify § 352.73(c) Thus, if 20 subjects produce valid data least 20 subjects. 2. The size of the test panel should be (proposed § 352.70(d)(3)) as follows: in accordance with proposed fixed in advance. * * * Administer a series of five UV § 352.70(c)(9), then it would be 2 3. The limitation that the standard radiation doses expressed as J/m -eff unnecessary to test additional subjects. error should be less than ± 5 percent (adjusted to the erythema action spectrum In addition, some subjects may not calculated according to paragraph (d)(1) of should not apply. this section) to the subsites within each test produce valid data in accordance with 4. The testing procedures should site on a subject using an accurately proposed § 352.70(c)(9) (e.g., no make it clear that the addition of calibrated solar simulator. The five UV doses erythema produced), requiring testing of subjects to the test panel to achieve the will be a geometric series as described in additional subjects (not exceeding 25 desired minimum is acceptable under paragraph (d)(2) of this section, where the subjects). FDA agrees that the number of specific conditions (58 FR 28194 at middle exposure represents the expected subjects should be based on error about 28267). SPF. For products with an expected SPF less the mean SPF, but disagrees that the than 8, use exposures that are the product of In the 1993 TFM, FDA based § 352.72(g) minimum number of subjects can be on these comments and the Panel’s the initial unprotected MED times 0.64X, lowered to 10. As described later in this 0.80X, 1.00X, 1.25X, and 1.56X, where X recommendation. equals the expected SPF of the test product. comment, FDA has reevaluated the The calculations of the sample size For products with an expected SPF between proposed minimum number of subjects and confidence interval in § 352.72(g) 8 and 15, use exposures that are the initial based on error about the mean SPF. are based on the assumption that there unprotected MED times 0.69X, 0.83X, 1.00X, FDA agrees with one comment that is a normal distribution about the mean 1.20X, and 1.44X, where X equals the more subjects are needed when testing (i.e., a bell curve). Based on this expected SPF of the test product. For products with high SPF values. FDA assumption, the t-test is used for products with an expected SPF greater that believes that a minimum sample size of statistical analysis. Based on the t-test, 15, use exposures that are the initial 20 subjects is adequate for products unprotected MED times 0.76X, 0.87X, 1.00X, FDA calculated that a panel of 20 with an expected SPF value of 30 or subjects should result in an acceptable 1.15X, and 1.32X, where X equals the less. However, current data and expected SPF of the test product. * * * error about the mean. However, in some (Comment 37) Several comments experience with products having SPF cases, a panel of 10 subjects would suggested changes to the number of values over 30 are not sufficient to probably result in an error about the subjects per test panel in § 352.72(g). determine an appropriate sample size. mean that is unacceptably large. There One comment suggested deletion of the Therefore, to account for increased is inherently higher variability in testing phrase ‘‘with the number fixed in variability in SPF values for sunscreens and, consequently, larger error about the advance by the investigator.’’ The with SPF values over 30, FDA proposes mean for products with high SPF comment reasoned that if the first 20 to increase the sample size to at least 25 values. Therefore, FDA believes a subjects provided data that can be subjects. FDA invites data greater number of subjects is necessary evaluated, risk to human subjects could demonstrating an appropriate panel size when testing products with high SPF be curtailed by not impaneling another for sunscreens with SPF values over 30. values. FDA believes a panel of 25 to 30 5 subjects. Other comments At this time, FDA is proposing to revise subjects should result in an acceptable recommended using 10 to 20 subjects, § 352.72(g) (proposed § 352.70(c)(7)) as error about the mean for products with arguing that the criterion for accuracy follows: high SPF values. FDA invites additional (7) Number of subjects—(i) For products should not be the number of subjects, with an expected SPF value under 30. A test data demonstrating adequate numbers of but the relative deviation of individual panel shall consist of 20 to 25 subjects with subjects, especially for products with SPF measurements. One comment used at least 20 subjects who produce valid data high SPF values. absorbance instead of the SPF value to for analysis. Data are valid unless rejected in (Comment 38) One comment stated calculate the number of subjects accordance with paragraph (c)(9) of this that one factor affecting the SPF of a

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49100 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

product is the erythemal threshold of there was a large systematic deviation using a smaller test subsite area before the skin, or MED(US). The comment between the FDA and DIN methods. As amending the monograph test. FDA will argued that SPF decreases with this study was not designed specifically consider a reduction in test subsite area increasing erythemal threshold. The to compare irradiation areas, three if adequate supporting data are comment maintained that, because different test subsite areas were used, provided. The studies should do the MED(US) varies only with skin type, the and none was 0.5 cm2. FDA cannot following: MED(US) of each subject in a test group determine the suitability of a 0.5 cm2 • Compare the smaller subsite area to should be within reasonably similar test subsite area compared to a 1 cm2 1 cm2 on the same subjects, limits. The comment suggested that the test subsite area based on this study. • Utilize high SPF products as well as MED(US) of each subject should be 50 The comment advocating the 0.8 cm products with SPF values below 15, and to 150 percent of the median MED(US). test subsite diameter argued that setting • Demonstrate comparable results The comment also suggested that a lower area limit has the following four among several laboratories. subjects with an MED(US) that is twice benefits: (Comment 40) Several comments the median should be excluded • Does not preclude the use of larger either agreed or disagreed with the regardless of skin type. irradiation areas, blinding procedures for the application FDA is not proposing the revisions • Will not affect the accuracy of of test materials described in suggested by the comment. FDA based resulting measurements, § 352.72(e). One comment stated that § 352.73(b), which describes • Permits lower wattage lamps as unblinded SPF testing is bad science, determination of an MED(US), on the well as liquid light guides that have and that exposure sites within test areas Panel recommendation in the ANPRM. apertures of 0.8 cm diameter, and should always be randomized no matter • The procedure for determining Provides more skin area for testing. how many products are being tested. MED(US) requires irradiation of subjects The comment provided statistical Another comment stated that the with a geometric series of UV doses. analysis of a study comparing multi-port blinding procedure is an unnecessary When developing this procedure, the and single-port solar simulators (Ref. complication and does not contribute to Panel explained that the geometric 66). SPF 15 or SPF 4 products were the accuracy of the test. One comment series provides the same relative level of tested along with the homosalate agreed that, in order to approximate true uncertainty independent of the subject’s standard sunscreen. Two subsite areas blinding, the individual who grades sensitivity to UV light (i.e., independent were exposed to the multi-port solar erythemal responses should not be the of skin type) (43 FR 38206 at 38266). simulator, and two were exposed to the same clinician who applied the test Thus, the Panel disagreed that skin type single-port solar simulator. The materials. Another comment contended affects MED(US). The comment did not comment concluded that similar SPF that it is not reasonable to randomly provide any data or other information values are determined using the two irradiate test sites with varying doses of demonstrating that skin type, in fact, types of solar simulators. However, the UV radiation. One comment affects MED(US). FDA is not aware of study report did not include details recommended making the use of finger any data demonstrating this such as subject selection, product cots optional because some product phenomenon. FDA will revise the application, or specifications for the vehicles are incompatible with finger proposed test criteria if we receive data solar simulators. More importantly, the cot material. Another comment or information demonstrating that the study report did not specify the size of suggested that the amount of product criteria are not appropriate or other each subsite. Thus, FDA cannot draw remaining on the finger cot is a source criteria are more suitable. any conclusions regarding appropriate of variability in the SPF test and (Comment 39) Several comments test subsite area from the submitted suggested that the extent of this urged FDA to reduce the minimum 1 study. variability be fully evaluated. cm2 test subsite area in § 352.72(d)(2). The comment supporting the 0.15 cm FDA agrees with the comments that One comment proposed the minimum test subsite diameter referenced two favor blinding and randomization and is test subsite area be decreased to 0.5 cm2. studies (Ref. 76). Significant not proposing to remove the blinding Two comments suggested that the test discrepancies in the information and randomization requirements from subsite area be defined by minimum submitted for the first study prevented § 352.72(e) (proposed § 352.70(c)(5)). diameters of 0.8 cm (circular area of 0.5 evaluation of this study. The comment According to § 352.72, blinding and cm2) and 0.15 cm (circular area of 0.017 did not submit full details of the second randomization is required only when cm2), respectively. study. Therefore, FDA could not reach two or more sunscreen drug products The comment supporting the 0.5 cm2 any conclusions from the submitted are being evaluated at the same time. test subsite area referenced a study studies. Because a test product is always tested published in 1987 (Ref. 75) that was FDA agrees, in principle, with the in conjunction with the standard mentioned in relation to artificial light advantages of a smaller test subsite area. sunscreen, FDA proposes to delete the sources in comment 86 of the TFM (58 The Panel stated that, depending on statement, ‘‘If only one sunscreen drug FR 28258 to 28261). This study was instrumental design, irradiation test product is being tested, testing subsites designed to evaluate the FDA sequential subsite areas less than 1 cm2 can be should be exposed to varying doses of technique of dosing using a single-port utilized and that test subsite diameters UV radiation in a randomized manner.’’ solar simulator (SPSS), a series greater than 0.4 cm present no difficulty Section 352.72(h) (proposed sequential method using a multi-port in determining skin erythema (43 FR § 352.70(c)(8)) specifies that the person xenon arc solar simulator (MPSS), and 38206 at 38260). While FDA does not who evaluates the MED responses must the Deutsches Institut fu¨ r Normung consider the information provided by not be the same person who applied the (DIN) simultaneous technique of dosing the comments adequate to support the sunscreen or administered the dose of using an Osram Ultravitalux lamp. Five suggested test subsite areas, it UV radiation. The comments that sunscreen formulations with SPF values recognizes that considerable advances disagreed did not provide evidence from 4 to 15 were tested. The authors have been made since the Panel met. demonstrating that these requirements suggested that there was little However, FDA requires data are unnecessary. systematic difference in estimates demonstrating that the monograph test With regard to the suggestion that the obtained using the SPSS and MPSS, but produces valid and reproducible results use of finger cots be made optional, the

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49101

Panel’s review of data found that that define the total amount of equivalency test than would be numerous investigators have obtained erythema-effective energy applied to the determined using the FM method. By more reproducible results by spreading testing subsite (i.e., as J/m2). FDA contrast, a higher SPF would never be a product using a finger cot than by discussed replacing ‘‘exposure time determined for a sunscreen using the spreading with a glass or plastic rod (43 interval’’ with ‘‘erythema-effective FM method than would be determined FR 38206 at 38261). FDA agrees with exposure (dose),’’ but inadvertently using the equivalency test. Thus, the FM the comment that some formulations used ‘‘exposure time interval’’ instead of method results in a more conservative may be chemically incompatible with ‘‘dose’’ in § 352.73(b). FDA agrees that SPF value than the equivalency test. latex finger cots, but there are finger cots § 352.73(b) (proposed § 352.70(d)(2)) FDA believes it is in the best interest of composed of other materials that should should be modified and is amending public health to label sunscreens with be compatible with these sunscreens. this section as the comment suggested. the more conservative SPF value. If FDA Therefore, to increase reproducibility in (Comment 43) Several comments adopted the equivalency test after over sunscreen application, FDA is suggested an alternative statistical 30 years of using the FM method, proposing to revise the application procedure for calculating product SPF consumers may, in some cases, requirement in § 352.72(e) (proposed values and PCD in current § 352.73(d). overestimate the protection provided by § 352.70(c)(5)) to read as follows: The comments argued that the a sunscreen based on a higher SPF * * * Use a finger cot compatible with the procedure described in the FM would number resulting from the equivalency sunscreen to spread the product as evenly as result in significant lowering of SPF test. possible. Pretreat the finger cot by saturating values. The comments advocated with the sunscreen and then wiping off clinical equivalency testing (i.e., using a M. General Comments on UVA Testing material before application. Pretreatment is lower one-sided 95 percent confidence Procedure meant to ensure that sunscreen is applied at the correct density of 2 mg/cm2. interval or a one-sided t test, with a (Comment 44) Many comments FDA urges manufacturers of sunscreen delta of 5 percent). The comments noted discussed UVA radiation action spectra drug products to investigate the extent that an upper and lower bound and skin damage (erythema, of variability in the SPF test that may be equivalency procedure with a delta of photocarcinogenesis, DNA damage, caused by various applicators. 20 percent would be an appropriate photosensitivity reactions, photoaging, (Comment 41) One comment procedure. The comments added that mutagenicity, and immunosuppression). addressed illumination at the test site in SPF is not a precise value, but rather a Some comments described various types § 352.72(h) and recommended that a valid estimate of product performance. of solar-induced skin damage and the level of at least 1,000 lux be used. The Another comment suggested using the wavelengths contributing to the specific comment contended that 450 to 550 lux mean of the results to find the actual biological events. Some comments is too low to provide adequate number and then round-off (either up or stated that UVA II radiation (320 to 340 illumination for reading erythema. down) to the nearest whole number. nm) is much more damaging than UVA As discussed in the TFM, the Panel FDA is not proposing to modify the I radiation (340 to 400 nm). recommended an incandescent or warm calculation of product SPF values and Other comments stated that there is fluorescent illumination source but did PCD in § 352.73(d) (proposed presently no convincing evidence that not specify a required illumination level § 352.70(d)(4)) at this time. The distinct the action spectra for damage from UV (58 FR 28194 at 28269). In the TFM, advantage of the t-test is that it provides radiation have been clearly defined. One FDA agreed with the Panel about the a simple computational procedure for a comment stated that until the separate illumination source. FDA also proposed statistical test that makes inferences dangers and risks of each portion of the that the illumination level be 450 to 550 about the population. The SPF is UVB and UVA radiation action spectra lux. The comment did not provide any determined to be the largest whole are precisely and scientifically data to support its contention that 1,000 number that is excluded by a lower one- identified and quantified, FDA should lux is the appropriate illumination sided 95 percent confidence interval. consider the entire UVA radiation range level. Thus, FDA is not revising the lux Simply finding a mean value, as one as having significant biological risk. range in § 352.72(h) (proposed comment suggested, is not adequate Another comment stated that protection § 352.70(c)(8)) at this time. FDA invites because such a value does not provide against all UVA radiation wavelengths data and information on levels of information about the validity of the test would seem to be both desirable and illumination currently used to evaluate (e.g., standard deviation) that should be prudent considering the present state of MED responses in SPF testing taken into consideration. our knowledge. FDA’s evaluation of the equivalency laboratories and will consider FDA agrees that the action spectra for testing approach for calculating SPF adequately supported alternatives. various harmful effects on human skin values indicates the method is less (Comment 42) One comment stated from chronic UVA radiation have not stringent than the FM method. The that the third sentence in § 352.73(b) been clearly defined and that it may be proposed equivalency test is essentially should be modified to read: ‘‘*** misleading to associate damage with testing the following hypothesis: wherein each exposure dose is 25 any specific action spectrum based H : µ ≤ 0.95L versus H : µ > 0.95L percent greater than the previous 0 a upon current knowledge. Information where: H0 = null hypothesis exposure dose to maintain the same Ha = alternative hypothesis provided by comments suggests a relative uncertainty * * *.’’ The µ = population mean relatively greater role for UVA radiation comment explained that defining the L = confidence limit than UVB radiation in long-term sun exposure dose in terms of ‘‘time’’ is FDA acknowledges that the equivalency damage even though there is little incorrect. test may be a valid method for consensus about the amount of UVA FDA discussed the Panel’s definition determining SPF. In many cases, the radiation protection required. Therefore, of dose in terms of time intervals in same SPF would be determined for a FDA is proposing UVA radiation test comment 84 of the TFM (58 FR 28194 sunscreen using either the equivalency methods that assess protection at 28256 to 28257). FDA stated that it is test or the FM method. However, in throughout the UVA spectrum (see more accurate to express dose as the some cases, a higher SPF would be section III.N, comment 45 of this ‘‘erythema-effective exposure,’’ in units determined for a sunscreen using the document).

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49102 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

N. Comments on UVA Testing • Practicability, convenience, and absorption spectrum of the product Procedure Design and Testing Criteria safety. throughout the UV radiation range. Stating that there is currently no (Comment 45) FDA is proposing that Then, one calculates the wavelength convincing evidence that the action λ both an in vitro and an in vivo test be value c (the ‘‘critical wavelength’’), spectrum for UVA radiation damage has conducted to determine UVA radiation where the area under the absorption been clearly defined, another comment λ protection. The proposed in vitro test is spectrum from 290 nm to c is 90 suggested that protection from UV the ratio of long wavelength UVA percent of the integral of the absorption radiation be measured using two factors absorbance (UVA I) to total UV spectrum from 290 to 400 nm, and uses based on the degree of attenuation of UV absorbance (i.e., UVB + UVA). The a five-point scale to classify products as radiation across the full spectrum. One follows: proposed in vivo test is the PPD test, factor, the SPF value, is erythemally which is similar to the SPF test except weighted and gives an indication of the TABLE 3.—BROAD SPECTRUM RAT- the endpoint is pigment darkening power of protection provided by the rather than erythema. FDA is proposing ING BASED ON CRITICAL WAVE- product. The second factor should take LENGTH that UVA labeling consist of a UVA into account the shape of the rating reflecting both the in vitro and in transmittance curve measured by either vivo test results. The rating will be the Critical Wavelength Broad Spectrum in vivo or in vitro means. The comment (nm) Rating lowest ‘‘high’’ protection, then the stated that it is potentially dangerous to sunscreen would be labeled as λ associate skin damage with any single c < 325 0 ≤ λ providing ‘‘medium’’ UVA protection. action spectrum (e.g., IPD, PPD, or PFA). 325 c < 335 1 FDA is proposing these UVA testing 335 ≤ λc < 350 2 The comment argued that all of these ≤ λ requirements based on many comments indicators are wavelength-specific and 350 c < 370 3 370 < λ 4 submitted in response to the TFM that protection from specific wavelengths c contained data and information on does not mean protection from damage. The comment concluded that this test possible test methods (and The comment added that if only the method makes no underlying combinations or modifications of these erythema action spectrum is used, it assumptions about the form of action methods). The comments discussed the virtually ignores the effects of spectra for either acute or chronic following in vivo and in vitro test wavelengths over 320 nm. The comment photobiological damage. Because the procedures: contended that using an SPF value • efficiency of UV radiation to induce a IPD, augmented by the shape of the • given photobiological endpoint tends to PPD, transmission curve would give • PFA, consumers the information necessary to decrease with increasing wavelength, • Photosensitivity methods, make an effective and safe judgment the method utilizes wavelength • UVA radiation protection percent, about the protection provided by a intervals for classifying the ‘‘broad • Diffey/Robson method and sunscreen drug product. For example, spectrum’’ rating, which increases in an modifications of that method, the comment noted that a product with approximately logarithmic manner. • Standards Association of Australia, a high SPF and a uniform high level of One comment submitted a protocol • Diffuse reflectance method, attenuation across the spectrum (i.e., for the ‘‘critical wavelength’’ (CW) • 2 Skin method, and equal attenuation at all UVB and UVA modification of the Diffey/Robson • Psoralen photoadduct method. wavelengths) will provide the most method for classifying the relative On May 12, 1994, FDA held a public protection. The comment added that, at degree of UVA radiation protection of meeting to discuss these UVA radiation a later date, if sufficient evidence sunscreen drug products (Ref. 80). The testing procedures (Ref. 77). becomes available to describe a credible comment addressed product One comment suggested using either UVA radiation damage spectrum, this photostability by pre-irradiation of the or both PPD and erythema skin combined system could be used by sunscreen product with a UV radiation responses to measure the UVA radiation convoluting the attenuation curve with dose corresponding to one-third the protection effectiveness of OTC the action spectrum curve. labeled SPF value. The comment sunscreen drug products. The comment One comment proposed a reported recommendations based on the maintained that these two test methods modification (‘‘critical wavelength’’) of results of a round-robin evaluation of have the following similarities: the Diffey/Robson test method (Refs. 78 the proposed CW method involving six • Same UVA radiation source, and 79). The comment noted that, when laboratories using four test sunscreen • Same dose range, and people are outdoors, they are not formulations with various substrates. • Similar post exposure time lags for exposed to only UVB or UVA radiation The comment concluded that the CW observation. but are exposed to solar UV radiation, method is a convenient, reproducible in The only difference is in the skin types which always contains both. In vitro method for measuring the used, thus giving a variable balance in addition, biological effects against uniformity of sunscreen absorbance PPD and erythema responses. The which people may wish to be protected spectra across the UV radiation comment added that such a are caused by all wavelengths in the spectrum to classify products into broad combination of methods has the solar UV radiation spectrum. The UVA radiation protection categories. following advantages: comment contended that investigators In response to the June 8, 2000, • Reproducibility and stability, should not be exposing subjects to reopening of the administrative record • Relevance, sources of radiation with spectra that for the rulemaking for OTC sunscreen • Persistence of skin response have no practical application and using drug products (65 FR 36319), FDA through 1 to 24 hours, irrelevant biological effects as endpoints received additional comments on UVA • Independence of source flux and (e.g., IPD). radiation testing methods. While all accuracy, The comment proposed to assess the comments supported some type of • Utilization for static as well as for UVA radiation protection potential of an testing to differentiate the UVA water resistance photoprotective OTC sunscreen drug product by first radiation protection potential of predictions, and spectrophotometrically determining the sunscreen products, they disagreed

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49103

about the use of in vivo versus in vitro further stated that the proposed in vivo because two products can have the same testing methods. methods modeled after the SPF test CW with very different UVA radiation Comments from a group of sunscreen generate protection factors that are absorbance curves and, thus, provide product manufacturers contended that protocol dependent and of different protection for consumers. an in vivo test method, such as PPD or indeterminate clinical relevance, as Some comments stated that a PFA, best describes the photoprotective none are surrogates for long term combination of methods may be characteristics of a sunscreen drug concerns like cancer and photoaging. appropriate for assessing the complete product. These comments stated that an Another comment added that the PPD UVA radiation protection potential of a in vivo method measures the actual and PFA tests do not adequately assess sunscreen product. One comment effect of UVA radiation on the skin and the breadth of UVA radiation protection suggested combining either the PPD or estimates the expected product and that the biologic effects of full PFA method with an in vitro method for performance under actual use spectrum UV radiation differ from the a meaningful and rigorous test of both conditions. effects of isolated wavelengths. the magnitude and breadth of the One comment presented test data that Several comments recommended biological protection (i.e., the level of suggested PPD and PFA values are using an in vitro method, and most protection and the UVB and UVA comparable (Ref. 6). The comment considered the CW method as wavelengths that are protected against) stated that an advantage of the PFA appropriate. One comment stated that provided by a sunscreen product. method is that it allows inclusion of CW allows for broad spectrum activity Another comment stated that complete skin type I, whereas the PPD test is regardless of SPF so that, if consumers assessment of a sunscreen product’s conducted on darker skin types (II and use a low SPF product, they will at least UVA radiation protection must include III). However, the comment added that have the option of choosing one that both of the following: the PPD test has been accepted since provides a wide breadth of activity. • An in vitro measurement of the 1996 by the JCIA for the assessment of Another comment stated that CW absorbance above 360 nm (i.e., UVA radiation protection efficacy of provides a simple, reproducible, and demonstrate adequate breadth of sunscreen products. adaptable method that can account for absorbance), and One comment contended that the PPD sunscreen photostability and insure • An in vivo measurement of the test should be used for the following UVA radiation protection that is both quantity of UV radiation protection (i.e., reasons: commensurate with and independent demonstrate adequate magnitude of • It requires a relatively low dose of from the SPF value. Another comment absorbance). UV radiation. added that CW accounts for Other comments stated that a • The reaction is stabilized in 2 to 4 proportionality because, in order for a combination of the in vivo SPF method hours. sunscreen to maintain a given CW, and the in vitro CW method provide a • The test subject is left with no mark protection from both long and short complete description of a product’s of irradiation and receives little or no UVA radiation wavelengths must inherent photoprotective characteristics injury. increase as UVB radiation protection with the SPF value describing the • The test can be conducted with increases. amplitude of protection and CW high precision. Several comments stated that the CW providing a reliable measure of the Another comment stated that PPD threshold should be 370 nm for a ‘‘broad product’s spectral absorption capability. values demonstrate the same correlative spectrum’’ claim on a sunscreen. Other One comment suggested a UVA/UVB benefits that exist for SPF values and, comments recommended a threshold of radiation proportionality scheme. The therefore, do not give false impressions 360 nm. One comment stated that if comment referred to FDA’s previous of magnitude. Another comment stated FDA were to arbitrarily select a standard discussions about UVA/UVB radiation that products with the same SPF can higher than 360 nm, it would cause a proportionality (Refs. 11 and 81) and a have different levels of UVA radiation major reformulation effort within the recommendation from the AAD that ‘‘an protection. Thus, PFA or PPD is not industry, higher prices to consumers, increase in SPF of a sunscreen must be redundant with the SPF value. and a shortage of ‘‘broad spectrum’’ accompanied by a proportional increase Comments from other sunscreen products in the OTC marketplace. The in the UVA protection value’’ (Ref. 82). product manufacturers opposed an in comments did not provide data to The comment added that the vivo method to determine UVA support the use of a specific threshold proportional contribution to sunburn radiation protection. One of these number in relation to the prevention of from solar UVB and UVA radiation is 80 comments stated that in vivo tests specific photobiological effects. to 20 (4 to 1), respectively, and that this expose human subjects to doses of UVA Other comments opposed the CW relationship gives the minimum UVA radiation with unknown human health method as not appropriate. One radiation attenuation needed to provide consequences. The comment added that comment, which favored an in vivo proportional UVA/UVB radiation because exposure to UVA radiation method, stated that the CW method, protection for any SPF value. The alone is never encountered in nature, based on an arbitrary, nonbiological comment concluded that a minimum full spectrum light is most relevant for criterion, fails to provide an accurate UVA protection value of 2 should be product evaluations. This comment measure of the protection efficacy of a required even at low SPF levels with contended that PFA values are sunscreen product. This comment proportionately higher UVA protection redundant with SPF testing because of provided data to demonstrate that a values for higher SPF values. an overemphasis on short wavelength significant failure of the CW method is One comment suggested that the UVA UVA radiation (UVA II), and PFA values its inherent inability to differentiate protection value should be determined give a false impression of the magnitude UVA radiation protection levels of with an in vivo method while CW is of absorption differences. For example, sunscreen products relative to biological appropriate to determine spectral the comment stated that two products endpoints (e.g., premature skin aging) broadness. Another comment stated that with PFA values of 5 and 10 may (Ref. 23). A second comment agreed CW accounts for proportionality attenuate 80 and 90 percent of UVA with this assertion, while a third because both long and short UVA radiation, respectively. Thus, the real comment expressed concern that CW radiation protection must increase as difference is small. The comment measurements may be misleading UVB radiation protection increases in

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49104 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

order for a sunscreen to maintain a suggested test methods. On the other Breadth can best be determined by given CW. Another comment provided hand, the IPD response has not been appropriate in vitro test methods. data (Ref. 23) for two products with the shown to represent a direct or surrogate At this time, FDA believes a same CW value but different SPF values endpoint for biological damage. The IPD combination of existing in vivo and in and concluded that the product with the response is also extremely difficult to vitro UVA radiation testing methods higher SPF value did not provide greater read. addresses the inadequacies of either UVA protection. Other comments stated The PFA test method uses endpoints method when used alone and provides that there is no biological basis for that reflect actual damage that can occur a more complete UVA radiation establishing strict UVB/UVA radiation to normal skin as a result of UVA attenuation profile for use in labeling proportionality and that the radiation exposure (i.e., erythema or OTC sunscreen drug products. establishment of this kind of ratio is tanning). The erythema action spectra Requiring the two test methods will arbitrary. may be similar to the action spectra of ensure that both the magnitude and The AAD (Ref. 83) referenced an known chronic skin damage (e.g., solar breadth of UVA protection is international consensus conference on elastosis) (Ref. 85). However, the PFA determined. As discussed later in this UVA radiation protection of sunscreens test method may not determine response, the proposed UVA labeling and recommended the following: protection against skin melanoma or will reflect the results of both tests and, 1. Both an in vitro and an in vivo other skin damage thought to be caused therefore, will reflect magnitude and testing method must be used to measure by chronic exposure to UVA radiation breadth of UVA protection. FDA UVA radiation protection. (Refs. 29 and 86). believes that the methods and labeling 2. CW is the preferred method of in The CW method can assess how currently being proposed provide the vitro testing for a broad spectrum claim broadly a sunscreen can absorb across best assurance for consumers to receive (with a threshold for this claim at 370 the UV radiation spectrum, but provides adequate protection across the entire nm). 3. CW must be combined with an in no information concerning product UVA radiation spectrum. vivo method such as either PPD or PFA. performance after interaction with FDA is proposing the PPD method as 4. There must be a minimum four-fold human skin. While in vivo methods to the in vivo part of the test to determine increase in PPD or PFA value in the assess UVA radiation protection may UVA radiation protection of a sunscreen presence of a sunscreen (relative to the have possible sources of variability drug product. This test assesses UVA absence of sunscreen). similar to the SPF test (e.g., test product radiation attenuation by measuring UVA In the Federal Registers of May 12, application, differences in light sources, radiation-induced tanning, a direct 1993 (58 FR 28194 at 28248 to 28250), etc.), in vitro methods also possess effect induced by UVA exposure. The September 16, 1996 (61 FR at 48645 at possible sources of inherent variability PPD test is relatively easy to perform 48652), and October 22, 1998 (63 FR (e.g., test product evaporation time, and relies on a stable, biological 56584 at 56587), FDA discussed substrate orientation, instrumentation, endpoint that can describe the photosensitivity and erythemal UVA use with color change sunscreen magnitude of UVA radiation protection radiation testing procedures for OTC formulations, etc.). of sunscreen products. It is similar to sunscreen drug products. Criteria In general, FDA would prefer the the SPF determination as it is a ratio of discussed for UVA radiation claims standard UVA radiation test method to a minimum pigmenting dose (MPD) on included the requirement for an have a clinically significant endpoint. unprotected skin to that on protected absorption spectrum extending to 360 After reviewing the data and skin. The endpoint is the PPD response, nm or above, plus the demonstration of information provided by the comments, which is the stable, lasting residual part meaningful UVA radiation protection FDA agrees that there is no convincing of the immediate pigment darkening or via testing procedures. IPD/PPD, PFA, evidence that the action spectra for all blue gray pigment that develops photosensitivity, and in vitro UVA possible types of UVA-induced damage immediately during exposure to UVA radiation testing methodologies were have been clearly defined and that no radiation and quickly fades at the end also discussed at a public meeting on one method is without disadvantages. of exposure. It provides consumers with May 12, 1994 (Ref. 77). At this time, FDA agrees with the a means to specifically compare the The selection of an appropriate UVA recommendation provided by the AAD amount of UVA radiation protection radiation testing procedure for OTC and other comments that an in vivo between products and select an sunscreen drug products has been method is appropriate in combination appropriate sunscreen product. The PPD difficult for a number of reasons. The with an in vitro testing method to assess test has been shown to produce reliable, scientific community does not agree on the UVA radiation protection. reproducible data and to distinguish which testing procedure is most Because the action spectrum for UVA- between varying levels of UVA radiation appropriate. For example, Cole induced skin damage is not clearly attenuation (Refs. 87 and 88). It has been discusses the virtues and shortcomings known, FDA considers it necessary to shown to detect protection provided by of a variety of in vivo and in vitro test measure both the magnitude and ‘‘broad spectrum’’ sunscreens against methods (Ref. 84). In addition, each test breadth of UVA protection. The both short and long wavelength UVA procedure has its own distinct magnitude of UVA absorbance is a radiation. The endpoint is a stable skin advantages and disadvantages, as measure of how well a product absorbs response that is linearly dependent on discussed in the following paragraphs. UVA radiation. The magnitude of UVA the amount of UVA radiation that enters FDA believes the IPD test method absorbance is best measured by an in the viable epidermis. FDA also agrees provides an appropriate endpoint for vivo method. An in vivo method with one comment that a UVA determining UVA protection, because measures a biological response on the protection value of 2 should define the pigment darkening is caused primarily skin (e.g., pigment darkening) and, lowest end of acceptable PPD test by UVA (and not UVB) radiation. This therefore, correlates to actual use results relative to the consideration of method is advantageous over other conditions. The breadth of the UVA acceptable UVA radiation claims (see suggested test methods in that it uses absorbance is a measure of how broadly proposed § 352.72(d)(3)). FDA considers low doses of radiation and, therefore, a product absorbs UVA radiation across it desirable to incorporate measurable exposes subjects to less risk than other the entire UVA radiation spectrum. UVA radiation protection at all SPF

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00036 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49105

levels for products that claim to protect As the current state of technology over the UV radiation range from 290 to against both UVB and UVA radiation. allows for an instrumental 400 nm by measuring the quantity of UV As one comment noted, the PPD test measurement/quantification of skin radiation transmitted through surgical has been accepted and validated as the color via spectral reflectance, FDA also tape (TransporeTM tape) before and after JCIA method since 1996 (Ref. 23) and is invites comments regarding colorimetry application of a sunscreen drug product. one of two in vivo methods suggested by as a method of evaluating pigment The test product (2 mg/cm2) is applied the AAD (Ref. 83). Although data darkening. By avoiding the subjectivity to the textured surface of the provided to FDA indicate that the PPD of detecting pigment change by the TransporeTM tape. A xenon arc solar and PFA in vivo tests provide human eye, the reproducibility of the simulator is used as the UV radiation comparable results (Ref. 6), the PPD test PPD method should increase. source. Transmitted UV energy is provides the practical benefit of a Colorimetry could likewise be used in collected and measured at 5 nm shorter post exposure reading time. FDA SPF testing if submitted data intervals over the UVB and UVA agrees with the comments that PPD demonstrated increased accuracy and radiation range, which provides a values are not redundant with SPF reproducibility of colorimetry over profile of UV radiation absorbance. values as sunscreen drug products with visual inspection. Mathematical calculations are made the same SPF value can have very As the PPD method is similar, overall, separately of the areas under the UVB different levels of UVA radiation to the SPF method, FDA is also and UVA radiation parts of the curve. protection as measured by the PPD test. proposing that the directions for the The ratio below the curve is determined Accordingly, FDA is including the PPD PPD method be similar to those for the as follows: method in proposed § 352.72 as part of SPF test for determining MPDs on the testing to determine the UVA unprotected skin, individual UVA radiation protection potential of an OTC protection factors, test product UVA protection factors, and PCDs. Further, as sunscreen drug product. As the ratio increases, the degree of FDA agrees with the comments that discussed in section III.L, comment 37 of this document regarding the SPF test, UVA radiation protection increases. suggested modifications to the PPD FDA is proposing that a PPD test panel FDA is concerned that this method, as method (i.e., the JCIA standard). consist of 20 subjects who produce valid described in previous paragraphs, Therefore, FDA is proposing data, similar to the panel size for determines the ratio of the entire UVA modifications to the PPD method. One sunscreens having SPF values less than to UVB radiation spectra. Therefore, a group of sunscreen manufacturers 30. sunscreen drug product that absorbs suggested that the previously validated FDA is concerned, however, that use strongly in the UVA II radiation area, ‘‘high SPF’’ padimate O/oxybenzone of the PPD method alone could result in but does not absorb strongly in the UVA standard sunscreen under consideration some products yielding high UVA I radiation area, might still have an by FDA (see section III.J, comment 27 of radiation protection factors without adequate ratio of UVA to UVB radiation this document) should also be used as having broad absorbance throughout the protection to fulfill the test the control formulation for in vivo UVA UVA radiation spectrum due to strong requirements, but would not provide radiation testing (Ref. 6). Based upon absorbance in the UVA II region. In adequate protection in the UVA data provided by the comment, FDA is other words, a sunscreen could absorb radiation region where absorbance is proposing the referenced ‘‘high SPF’’ high levels of UVA II but very little lacking. FDA believes that this padimate O/oxybenzone standard UVA I and achieve a high UVA rating deficiency can be corrected by revising sunscreen for use as the standard under the PPD method. Therefore, FDA the calculations to take into account the sunscreen in the in vivo UVA radiation is proposing that an in vitro method be ratio of UVA I and/or UVA II test in proposed § 352.72. FDA invites used (to assess the breadth of individually to UV radiation. Some comment on the suitability of this absorbance across the UV radiation comments were concerned that UVA II formulation as a UVA radiation test spectrum) in conjunction with the PPD radiation may be the portion of the UVA standard, on alternative standards, and method to more completely assess a spectrum most represented in the PPD on preparation/assay/validation data for product’s UVA radiation protection. test. FDA agrees that the UVA II any suggested alternatives. FDA disagrees with the comments spectrum is well represented by the PPD FDA also notes that the JCIA light that the CW method should be used as test. Therefore, to provide for a more source specification states that ‘‘UV rays the in vitro testing method and proposes balanced method, FDA is proposing that shorter than 320 nm shall be excluded using a modification of the Boots the in vitro component of the through the use of an appropriate filter.’’ adaptation of the Diffey/Robson method monograph UVA radiation method only FDA considers it important to set an (Ref. 90). Both the CW and the in vitro need provide a measure of the relative exact limit for this specification and is test proposed by FDA measure the UVA I radiation absorbance. proposing that optical radiation from absorbance of a sunscreen product using FDA is proposing to measure UVA I the light source between 250 and 320 in vitro spectrophotometry. However, radiation absorbance relative to UV nm be less than 0.1 percent of the FDA’s proposed method calculates the radiation absorbance rather than relative optical radiation between 320 and 400 ratio of long wavelength UVA to UVB radiation absorbance. If UVA I nm. Also, the observation of pigment absorbance (UVA I) to total UV radiation protection is measured relative darkening in the JCIA standard is at 2 absorbance to provide a measure of the to UVB radiation, then the test does not to 4 hours post irradiation. FDA notes relative UVA I radiation protection account for UVA II radiation protection. that it appears the pigment darkening is provided by a sunscreen drug product. FDA’s proposed modification of the most stable about 3 hours or more after FDA believes that this test, in Boots adaptation of the Diffey/Robson post irradiation (Ref. 89), and is thus combination with the PPD method, method accounts for the entire UV proposing that this observation occur at provides a better assessment of overall radiation spectrum. Further, the ratio of 3 to 24 hours post irradiation. This time UVA radiation protection. UVA I radiation to UV radiation has a range provides increased flexibility in The Boots adaptation of the Diffey/ convenient finite range and allows for the test method without sacrificing Robson test method assesses the the use of defined values to categorize accuracy. absorbance of a sunscreen drug product UVA radiation protection.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.004 49106 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

FDA is proposing a modified Boots quartz plates for its in vitro studies and be used with the following number of adaptation of the Diffey/Robson method that quartz plates are reusable and inert. graphical representations in the form of instead of the CW method. The CW Diffey et al. have also used quartz plates solid ‘‘stars’’: determination only reveals the shortest as the substrate for the CW method (Ref. wavelength at which 90 percent of total 91). Accordingly, at this time, FDA is TABLE 5.—GRAPHICAL UVA RATING UVB and UVA radiation is absorbed by proposing that roughened quartz plates BASED ON CATEGORY a sunscreen. Thus, this method does not be specified as the substrate in the in directly reveal the breadth of UV vitro portion of its UVA test method. Combined Category Rating Star absorption, whereas the modified Boots FDA requests comment regarding the Rating adaptation of the Diffey/Robson method suitability and availability of quartz Low ✰✰✰ does. This point is demonstrated by data plates and other possible substrates. Medium ✰✰ submitted by one comment (Ref. 23). FDA agrees with one comment that High ✰ The comment submitted the UV there is no biological basis for Highest absorption spectra of two sunscreens establishing a strict UVA to UVB ratio having nearly identical SPF and CW and that such a ratio would be arbitrary. FDA invites comment on these values. The absorption spectra FDA is proposing that data from the proposed test methods/criteria and demonstrate that two sunscreens with proposed in vitro and in vivo tests be encourages the continued development similar CWs can have significantly integrated into a single labeled UVA of biologically meaningful test different UVA absorption spectra. The rating. Similar to suggestions from some procedures. ratios of UVA I/UV radiation absorbance comments, FDA is proposing the for these formulations were markedly categories of low, medium, high, and O. Comments on the Photostability of different: 0.85 and 0.52. Thus, FDA highest (corresponding to one, two, Sunscreen Drug Products believes that the ratio method generally three, and four ‘‘stars,’’ respectively). (Comment 46) Various comments allows for better discrimination of Based on test data submitted by one discussed the photostability of OTC products with these types of absorbance comment (Ref. 6), FDA is proposing that sunscreen formulations and active spectra. test results for each in vitro or in vivo ingredients. One comment stated that FDA is also concerned that the test be categorized as follows: photostability is important because activity of the sunscreen ingredients in many sunscreen ingredient the product may be diminished by TABLE 4.—UVA RATING CATEGORIES combinations with avobenzone are not exposure to UV radiation, i.e., that the believed to be photostable. This sunscreen ingredients in the product Category In vitro result In vivo result comment stressed that a sunscreen drug might not be photostable. Therefore, in Low 0.2 to 0.39 2 to under 4 product should maintain most of its order to account for changes in UVA and UVB radiation protection absorbance as a function of UV radiation Medium 0.40 to 0.69 4 to under 8 High 0.70 to 0.95 8 to under 12 throughout the expected consumer time exposure, FDA is proposing to revise the Highest greater than 0.95 12 or more in the sun. Another comment stated that Boots modification of the Diffey/Robson the integrity of a sunscreen drug method by incorporating pre-irradiation FDA is aware of the difficulty for product depends on its degree of dose (PID), which is defined as follows current sunscreen formulations to meet photostability and that a photostable (see section III.O, comment 46 of this the ‘‘highest’’ category and believes that product should maintain its protection document): over a wide range of UV radiation 2 allowing such a category will foster PID (J/m -eff) = SPF * 1 MED * 2/3, additional research and development in spectra. where 1 MED = 200 J/m2-eff FDA is also concerned about this area. Some comments supported a standard specifying the use of TransporeTM tape FDA is proposing that the overall method using pre-irradiation to account (used in the original Diffey/Robson UVA radiation category for use in for photostability of sunscreen method), an artificial substrate that product labeling be the lowest category ingredients. One comment favoring the mimics the surface topography of determined by the in vitro and in vivo CW method for measuring UVA human stratum corneum. When test results. For example, if the test radiation protection submitted a sunscreen emulsions are applied to results for a sunscreen indicate an in formula to establish a pre-irradiation TransporeTM tape (Refs. 7 and 77), the vitro category of ‘‘low’’ and an in vivo dose to assess photostability (Ref. 7). emulsions may experience a micro category of ‘‘high’’ (or the reverse), then This comment stated that pre-irradiation environment that differs from human the overall UVA classification on the provides a reasonable estimate of what skin in several key aspects, including sunscreen product label would be ‘‘low’’ a consumer might expect when using the following: (i.e., the lower of the two categories). the product and stressed that the dose • Lack of electrolyte effect, FDA believes that using the lower of the should be both full spectrum (290 to 400 • Lack of moisturization/humectant two categories takes into account the nm) and sufficient to detect significant plasticization of the substrate, following situations: changes in CW as a function of UV • Differences in pH and wetting • A product that has a high in vivo radiation exposure. This comment effects, and rating because of substantial UVA II considered its pre-irradiation dose of • Different degrees of sunscreen absorbance, but a low in vitro rating solar-simulated UV radiation to be penetration and retention by the because of poor UVA I absorbance, or equivalent to about 1 1/2 hours of substrate. • A product that has a low in vivo noonday sun or 3 hours of sun exposure The fourth aspect, different degrees of rating because of poor UVA II in the early morning or late afternoon. penetration and retention, is especially absorbance, but a high in vitro rating One comment noted that avobenzone- significant for oil soluble sunscreen because of substantial UVA I containing formulations can be ingredients. One comment suggested absorbance. photostabilized by the addition of that either roughened quartz plates or a FDA is further proposing that each suitable ingredients and supported a synthetic collagen should be used as the overall UVA radiation category protocol developed by Sayre and Dowdy substrate, noting that COLIPA has used correspond to and (on product labeling) for measuring UVA radiation protection

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00038 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49107

following a measured exposure of the simulator with the reference action Mandating warnings in an OTC drug test formulation to solar radiation (290 spectrum for erythema as defined by monograph does not require a finding to 400 nm) (Ref. 92). CIE. A typical weighted value (J/m2-eff) that any or all of the OTC drug products Another comment stressed the for an MED in a Skin Type II individual covered by the monograph actually importance of a standard pre-irradiation is 200 J/m2-eff (Ref. 93). Thus, FDA is caused an adverse event, and FDA does dose and included data suggesting that proposing to use the following formula not so find. Nor does FDA’s requirement a ‘‘UVB-only’’ sunscreen product to determine the required pre- of warnings repudiate the prior OTC formulation, at high pre-irradiation irradiation dose: drug monographs and monograph doses, could qualify for UVA ‘‘broad PID (J/m2-eff) = SPF * 1 MED * 2/3 rulemakings under which the affected spectrum’’ labeling by the CW method where 1 MED = 200 J/m2-eff drug products have been lawfully (Ref. 23). This comment concluded that In considering the selection of the marketed. Rather, as a consumer pre-irradiation does not always account appropriate pre-irradiation dose of protection agency, FDA has determined for photostability and appears to be very solar-simulated UV radiation, FDA that warnings are necessary to ensure formulation specific. agrees that the maximum pre-irradiation that these OTC drug products continue Another comment submitted an in exposure would be a dose of UV to be safe and effective for their labeled vitro method for simultaneously radiation that equaled the SPF of the indications under ordinary conditions predicting SPF and assessing product times the MED. However, FDA of use as those terms are defined in the believes that this calculated dose is photostability of sunscreen formulas act. This judgment balances the benefits probably greater than the dose that a (Ref. 65). The comment stated that pre- of these drug products against their sunscreen product would incur during irradiation with measured UV radiation potential risks (see 21 CFR 330.10(a)). doses has permitted more accurate in typical consumer usage. Thus, the dose FDA’s decision to act in this instance vitro estimates of SPF. was reduced by a factor of one-third to FDA agrees that it is important to represent a more reasonable exposure need not meet the standard of proof address the photostability for sunscreen condition. required to prevail in a private tort drug product formulations. Unstable action (Glastetter v. Novartis IV. FDA’s Tentative Conclusions and Pharmaceuticals Corp., 252 F.3d 986, product formulations present the Proposals problem of degradation of product 991 (8th Cir. 2001)). To mandate effectiveness during actual use. The FDA tentatively concludes that the warnings, or take similar regulatory assessment of overall protection FM for OTC sunscreen drug products action, FDA need not show, nor do we provided by such formulations is should be amended to include the allege, actual causation. For an difficult due to product effectiveness combinations of avobenzone with expanded discussion of the case law being heavily dependent on the UV ensulizole and avobenzone with zinc supporting FDA’s authority to require radiation exposure dose. Sayre and oxide when used in the concentrations such warnings without evidence of Dowdy demonstrated, through a series established for each ingredient in actual causation, see Labeling of of in vitro studies, how the UV radiation § 352.10 (see section III.C, comment 7 of Diphenhydramine-Containing Drug transmission of an avobenzone this document). However, before Products for Over-the-Counter Human containing formula changes with UV marketing may begin, the comment Use, final rule (67 FR 72555, December radiation exposure and that most of the period for this proposal must end and 6, 2002). loss of protection occurred in the UVA FDA must publish another Federal V. Analysis of Impacts radiation spectrum (Ref. 92). Register notice setting forth our FDA is proposing to address determination concerning interim FDA has examined the impacts of this photostability by adding a pre- marketing before publication of the final proposed rule under Executive Order irradiation step to the in vitro test rule for OTC sunscreen drug products. 12866, the Regulatory Flexibility Act (5 method for measuring UVA radiation FDA followed this procedure previously U.S.C. 601–612), and the Unfunded protection (see section III.N, comment for avobenzone as a single active Mandates Reform Act of 1995 (2 U.S.C. 45 of this document). As noted in the ingredient and in combination with 1501 et seq.). Executive Order 12866 scientific literature, the choice of a pre some GRASE active ingredients other directs agencies to assess all costs and irradiation dose is ‘‘somewhat arbitrary, than ensulizole or zinc oxide (62 FR benefits of available regulatory yet critical to the outcome of the test’’ 23350). alternatives and, when regulation is (Ref. 84). FDA received one comment FDA considers the UVA-related necessary, to select regulatory with supporting data for a proposed pre- labeling in this proposal to supersede approaches that maximize net benefits irradiation dose (Ref. 7). The comment the labeling proposed in the TFM and (including potential economic, suggested using a dose equivalent to the its amendments of September 16, 1996, environmental, public health and safety, SPF times 2 J/cm2 multiplied by a factor and October 22, 1998. While the prior and other advantages; distributive of 2/3. The comment stated that 2 J/cm2 proposed labeling can continue to be impacts; and equity). Under the from a xenon arc solar simulator with 1 used until a FM is issued, FDA Regulatory Flexibility Act, if a rule has millimeter (mm) WG-320 and 1 mm UG- encourages manufacturers of OTC a significant economic impact on a 5 filters was equivalent to one MED. sunscreen drug products to voluntarily substantial number of small entities, an Because all solar simulators used by the implement the UVA-related labeling agency must analyze regulatory options industry may not use this exact filter changes as soon as possible after that would minimize any significant combination and the spectral publication of this proposal, especially impact of the rule on small entities. transmittance of filters can vary from lot if product relabeling occurs in the Section 202(a) of the Unfunded to lot, FDA is proposing to specify the normal course of business. We note, Mandates Reform Act requires that pre-irradiation dose in terms of though, that any relabeling prior to agencies prepare a written statement of ‘‘erythemal effective dose.’’ The issuance of the FM is subject to the anticipated costs and benefits before erythemal effective dose of a solar possibility that FDA may change some proposing any rule that may result in an simulator can by calculated as described of the labeling requirements as a result expenditure in any one year by State, in proposed § 352.70(d) by weighting of comments filed in response to this local, and tribal governments, in the the output spectrum of the solar proposal. aggregate, or by the private sector, of

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00039 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49108 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

$100 million (adjusted annually for manufacturers will choose not to test for costs that would be incurred by OTC inflation). UVA protection and the labeling for sunscreen manufacturers. RTI estimated FDA believes that this proposed rule those sunscreens will state, ‘‘No UVA that graphic design and prepress and is consistent with the principles set out Protection.’’ engraving costs would range from in the Executive Order 12866 and in $1,970 to $13,800 per SKU depending B. Number of Products Affected these two statutes. The proposed rule is on the type of packaging and printing not a significant regulatory action as Estimating the number of products method used. There would also be defined by the Executive order and, affected is difficult because we lack data administrative costs to account for therefore, is not subject to review under on the number of products currently contracting costs and obtaining final the Executive order. Further, because marketed. Our Drug Listing System approvals for the new labels. RTI this proposed rule is not expected to currently does not have accurate estimated administrative costs to range result in any 1-year expenditure that information on the number of marketed from $360 to $880 depending on the would exceed $100 million adjusted for OTC sunscreen products, especially the size of the firm. For this analysis, we are inflation, FDA need not prepare drug-cosmetic combination products. assuming an average design price of additional analyses under the Unfunded Proprietary databases that track retail $7,000 per SKU and average Mandates Reform Act. Because the rule sales of OTC drugs and other products administrative costs of $600 per SKU.1 may have a significant economic impact do not distinguish cosmetics containing Therefore, the total relabeling cost per on a substantial number of small sunscreens from other cosmetic SKU would be $7,600 (i.e., $600 + entities, this section of the preamble products and their surveys do not $7,000). constitutes FDA’s regulatory flexibility include many of the outlets where While all sunscreen SKUs would need analysis. sunscreen products are sold. Based on to be relabeled to comply with the An analysis of the costs and benefits earlier estimates (64 FR 27666 at 27684) proposed rule, we estimate that the of this regulation, conducted under and our knowledge of the industry, we timing of the scheduled relabeling Executive Order 12866, was discussed assume there are about 3,000 OTC would coincide with the regulatory- in the FM (64 FR 27666 at 27683 to sunscreen drug products (different mandated changes for 50 percent of the 27686), which was later stayed (66 FR formulations, not including products SKUs (i.e., 6,000 SKUs). We estimate the 67485). This analysis reflects the that differ only by color), including total labeling cost of the proposed incremental costs of the revised or new drug-cosmetic combinations, and about labeling changes for the SKUs with the requirements in this proposed 12,000 individual stock keeping units coinciding scheduled redesign would be amendment of the FM. (SKUs) (individual products, packages, 50 percent of the administrative cost and sizes). All 12,000 SKUs will need to A. Background (i.e., $300). Therefore, the total one-time be relabeled, but manufacturers can cost to industry for relabeling would be The purpose of this document is to choose whether to test their sunscreen about $47.5 million (i.e., (6,000 x amend the conditions under which OTC products for UVA protection. We $7,600) + (6,000 x $300)). sunscreen drug products are generally assume that about 75 percent (2,250) of D. Cost to Test or Retest Products for recognized as safe and effective the sunscreen products would be tested UVA Protection (GRASE) and not misbranded. This for UVA protection. We request amendment addresses formulation, comment on the accuracy of this This proposed rule will result in labeling, and testing requirements for assumption. testing costs for products that make both UVB and UVA radiation UVA protection claims. The protection. C. Cost to Relabel approximate costs are $2,200 for in vivo Manufacturers would not need to The cost to relabel varies greatly UVA testing and $200 for in vitro UVA reformulate their sunscreen products to depending on the printing method and testing. Based on the number of comply with the proposed number of colors used. The majority of sunscreen products currently labeled as requirements. Manufacturers also would sunscreen products are packaged in providing UVA protection, we estimate not need to retest their sunscreen plastic bottles or tubes with the label that 75 percent (2,250) of the sunscreen products for UVB protection (i.e., they printed directly on the container or products will be tested according to the would not need to retest for SPF). The applied as a decal or paper label during proposed UVA tests. Therefore, FDA labeled SPF value determined from the the packaging process. The proposed estimates a one-time UVA testing cost of SPF test in the FM would not likely labeling requirements impact both the approximately $5.4 million (i.e., 2,250 x change if a sunscreen product was PDP and the Drug Facts section of the $2,400). retested using the modifications to the package and would be considered a E. Total Incremental Costs SPF test proposed in this document. In major redesign. addition, manufacturers who have Frequent label redesigns are typical The estimated total one-time tested and labeled their sunscreen for OTC sunscreen products, with incremental cost of this proposed rule is products as ‘‘SPF 30+’’ can relabel their redesigns generally implemented every $53 million (i.e., $47.5 million + $5.4 products with the specific SPF value 1 to 2 years for a product. To the extent million). The incremental cost for the above 30 (but no greater than 50) that a scheduled redesign coincides UVA testing could be less should the without retesting. with the regulatory-mandated rule become final because many However, all manufacturers would relabeling, the impact on the manufacturers may voluntarily comply incur some relabeling costs due to manufacturer will be negligible. with the proposed rule when proposed revisions to both the PDP and We used a model developed for FDA reformulating current products or the Drug Facts section of the product by the consulting firm RTI to derive an marketing new products. Although the label. If manufacturers wish to label estimate of the cost to relabel sunscreen FM is not effective, manufacturers of their sunscreen products as providing products (Ref. 94). The model was sunscreen products comply with the UVA protection, then manufacturers of developed to estimate the cost of food those sunscreen products would also labels. However, we believe that the 1 We did not select the midpoint of the ranges because of the large number of private label incur UVA testing costs. Because UVA graphic and design estimates from that products that have lower design and administrative testing is not required, some study are an appropriate proxy for the costs than branded goods.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00040 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49109

UVB (SPF) test in the FM for nearly all their production schedules. By has determined that the proposed rule, sunscreen products. Therefore, it is providing an additional 6 months to if finalized as proposed, would have a likely that manufacturers of sunscreen implement the changes, compliance preemptive effect on State law. Section products will also voluntarily comply costs to manufacturers will be reduced. 4(a) of the Executive order requires with the proposed UVA tests in this In addition, FDA reduced compliance agencies to ‘‘construe * * * a Federal document. costs when we chose to stay the labeling statute to preempt State law only where It should also be noted that sunscreen requirements for the FM (64 FR 27666), the statute contains an express products that are already distributed by sparing industry the cost of an preemption provision or there is some the effective date of the FM will not be additional regulatory-mandated label other clear evidence that the Congress required to be relabeled or retested in change. In the stay, FDA estimated a intended preemption of State law, or conformity with these FM conditions, cost savings of $1.5 million to industry. where the exercise of State authority unless these products are subsequently It should be noted that labeling costs conflicts with the exercise of Federal relabeled or repackaged after the were significantly less in the FM than in authority under the Federal statute.’’ effective date. Therefore, there is no this proposed rule primarily because we Section 751 of the Federal Food, Drug, one-time cost associated with disposing assumed in the FM that the majority of and Cosmetic Act (the act) (21 U.S.C. of sunscreens that are already on the relabeling would coinside with 379r) is an express preemption market at the time of the rule’s effective scheduled voluntary label redesigns at provision. Section 751(a) of the act (21 date. no additional cost. Manufacturers were U.S.C. 379r(a)) provides that ‘‘no State F. Small Business Impact also able to avoid or postpone incurring or political subdivision of a State may an additional industry total of $5 establish or continue in effect any In the FM (64 FR 27666 at 27685), million when FDA chose to stay the requirement—* * * (1) that relates to FDA estimated that 78 percent of the UVB testing requirements of the FM. the regulation of a drug that is not 180 domestic companies that FDA invites public comment subject to the requirements of section manufacture OTC sunscreen products regarding any substantial or significant 503(b)(1) or 503(f)(1)(A); and (2) that is would be considered a small business economic impact that this proposed rule different from or in addition to, or that (defined as fewer than 750 employees). would have on manufacturers of OTC is otherwise not identical with, a FDA cannot estimate with certainty the sunscreen drug products. Comments requirement under this Act, the Poison number of small firms that will need to regarding the impact of this rulemaking Prevention Packaging Act of 1970 (15 test or retest their OTC sunscreen on such manufacturers should be U.S.C. 1471 et seq.), or the Fair products to provide for UVA protection accompanied by appropriate Packaging and Labeling Act (15 U.S.C. claims, but projects that approximately documentation. FDA is providing a 1451 et seq.).’’ Currently, this provision 75 percent of all products may need to period of 90 days from the date of operates to preempt States from be tested for UVA protection. Costs will publication of this proposed rule in the imposing requirements related to the vary by firm, depending on the number Federal Register for comments to be regulation of nonprescription drug of products requiring testing. The firm- developed and submitted. FDA will products. Section 751(b) through (e) of specific impact may vary inversely with evaluate any comments and supporting the act outlines the scope of the express the volume of product sales, because per data that are received and will reassess preemption provision, the exemption unit costs will be lower for products the economic impact of this rulemaking procedures, and the exceptions to the with high volume sales. Thus, the in the final rule. provision. relative economic impact of product This proposed rule, if finalized as retesting may be greater for small firms VI. Paperwork Reduction Act of 1995 proposed, would amend the labeling than for large firms. Because the OTC FDA tentatively concludes that the and include new UVA testing for OTC drug industry is highly regulated, all labeling requirements in this document sunscreen drug products. Any final rule firms are expected to have access to the are not subject to review by the Office would have a preemptive effect in that necessary professional skills on staff or of Management and Budget because it would preclude States from issuing to have contractual arrangements to they do not constitute a ‘‘collection of requirements related to the labeling and comply with the testing requirements of information’’ under the Paperwork testing of OTC sunscreen drug products this rule. Reduction Act of 1995 (44 U.S.C. 3501 that are different from or in addition to, or not otherwise identical with a G. Analysis of Alternatives et seq.). Rather, the proposed labeling statements are a ‘‘public disclosure of requirement in the final rule. This FDA could have proposed only an in information originally supplied by the preemptive effect is consistent with vivo or an in vitro test for UVA. FDA Federal Government to the recipient for what Congress set forth in section 751 recognizes that requiring only the in the purpose of disclosure to the public’’ of the act. Section 751(a) of the act vitro test would mean significantly less (5 CFR 1320.3(c)(2)). displaces both State legislative cost to manufacturers. However, the requirements and State common law proposed in vivo test measures the VII. Environmental Impact duties. We also note that even where the magnitude of UVA protection. The FDA has determined under 21 CFR express preemption provision in section proposed in vitro test measures the 25.31(a) that this action is of a type that 751(a) of the act is not applicable, breadth of UVA protection. FDA does not individually or cumulatively implied preemption may arise (see Geier believes it is important to conduct both have a significant effect on the human v. American Honda Co., 529 US 861 tests to determine the magnitude and environment. Therefore, neither an (2000)). breadth of UVA protection. environmental assessment nor an FDA believes that the preemptive FDA plans to grant an extended environmental impact statement is effect of the proposed rule, if finalized compliance period when this proposed required. as proposed, would be consistent with rule is finalized. Given the seasonal Executive Order 13132. Section 4(e) of nature of these products, FDA is VIII. Federalism the Executive order provides that ‘‘when concerned that some manufacturers may FDA has analyzed this proposed rule an agency proposes to act through not have sufficient time to incorporate in accordance with the principles set adjudication or rulemaking to preempt labeling changes without disrupting forth in Executive Order 13132. FDA State law, the agency shall provide all

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00041 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49110 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

affected State and local officials notice In addition to AHAs, FDA seeks X. Proposed Effective and Compliance and an opportunity for appropriate comment on titanium dioxide and zinc Dates participation in the proceedings.’’ FDA oxide formulated in particle sizes as FDA is proposing that any final rule is providing an opportunity for State small as a few nanometers. FDA that may issue based on this proposal and local officials to comment on this addressed issues concerning micronized become effective 18 months after its rulemaking. sunscreen ingredients in the FM (64 FR date of publication in the Federal IX. Request for Comments 27666 at 27671 to 27672). The FM stated Register. The compliance date for that FDA did not consider micronized In the Federal Register of January 10, products with annual sales less than titanium dioxide to be a new ingredient $25,000 would be 24 months after 2005 (70 FR 1721), FDA announced the but rather a specific grade of the same availability of a final guidance for publication of the final rule in the active ingredient. The FM also stated Federal Register. industry entitled ‘‘Labeling for that FDA was aware of concerns about Topically Applied Cosmetic Products potential risks associated with increased XI. References Containing Alpha Hydroxy Acids as dermal penetration of such small The following references are on Ingredients.’’ The purpose of this particles. However, the FM explained display in the Division of Dockets guidance is twofold: that, based on the safety data submitted Management (see ADDRESSES) under • To educate consumers about the to FDA before publication of the FM, Docket No. 1978N–0038 and may be potential for increased skin sensitivity FDA was not aware of any evidence at seen by interested persons between 9 to the sun from the topical use of that time demonstrating a safety concern a.m. and 4 p.m., Monday through cosmetics containing alpha hydroxy from the use of micronized titanium acids (AHAs) as ingredients. Friday. dioxide in sunscreen products (64 FR 1. Comment No. CP6. • To educate manufacturers to help 27666 at 27671 to 27672). 2. Comment Nos. CP8, C548, SUP22, and ensure that their labeling for cosmetic C555. products containing AHAs as FDA recognizes that more sunscreens containing small particle size titanium 3. Comment Nos. LET166 and LET169. ingredients is not false or misleading. 4. Comment No. C538. As discussed in the guidance, AHAs dioxide and zinc oxide ingredients enter 5. Comment No. C576. may increase skin sensitivity to UV the market each year. FDA is interested 6. Comment No. C565. radiation. Therefore, FDA recommends in receiving comments and data about 7. Comment No. C581. that manufacturers of cosmetic products these sunscreen ingredients and 8. Comment No. C567. containing AHAs include the following products that contain these ingredients, 9. Comment No. C515. warning: their safety and effectiveness, and how 10. Comment No. C597. Sunburn Alert: This product contains an they should be regulated. FDA received 11. Comment No. MM22. alpha hydroxy acid (AHA) that may increase a citizen petition shortly before 12. Comment No. MM21. 13. Comment No. C573. your skin’s sensitivity to the sun and publication of this document that, particularly the possibility of sunburn. Use a 14. Autier, P. et al., ‘‘Sunscreen Use and among other things, raises these issues. Duration of Sun Exposure: a Double-blind, sunscreen and limit sun exposure while FDA is currently evaluating the citizen using this product and for a week afterwards. Randomized Trial,’’ Journal of the National The guidance addresses only cosmetic petition, which is filed as CP17 in Cancer Institute, 91(15):1304–1309, 1999. products containing AHAs and does not Docket No. 1978N–0038. FDA 15. Reynolds, K.D. et al., ‘‘Predictors of encourages other parties to submit Sun Exposure in Adolescents in a address sunscreen drug products Southeastern U.S. Population,’’ Journal of containing AHAs (i.e., drug-cosmetic additional data or information on the safety and effectiveness of sunscreen Adolescent Health, 19(6):409–415, 1996. products). FDA is considering an 16. Robinson, J.K., D.S. Rigel, and R.A. additional warning or direction for ingredients formulated in particle sizes Amonette, ‘‘Summertime Sun Protection sunscreen drug products containing as small as a few nanometers. Used by Adults for Their Children,’’ Journal AHAs similar to the warning for the On April 14, 2006, FDA announced in American Academy of Dermatology, cosmetic products described in the the Federal Register that we were 42(5):746–753, 2000. guidance for industry. However, FDA planning a public meeting on FDA- 17. Gallagher, R.P. et al., ‘‘Broad-spectrum Sunscreen Use and the Development of New invites interested parties to submit regulated products containing nanotechnology materials (71 FR Nevi in White Children: A Randomized comments and data regarding such Controlled Trial,’’ Journal of the American labeling. In particular, FDA would like 19523). As explained in the notice, the Medical Association, 283(22):2955–2960, the following questions addressed: purpose of the meeting was to help FDA 2000. 1. Does the body of existing evidence further its understanding of 18. Stender, I.M., J.L. Andersen, and H.C. on AHAs and skin sensitivity warrant developments in nanotechnology Wulf, ‘‘Sun Exposure and Sunscreen Use voluntary or mandatory labeling on OTC materials that pertain to FDA-regulated among Sunbathers in Denmark,’’ Acta sunscreen drug products containing products. The meeting was held on Dermato-Venereologica, 76:31–33, 1996. AHAs regarding possible risks of October 10, 2006, and FDA has received 19. Zitser, B.S. et al., ‘‘A Survey of comments from interested members of Sunbathing Practices on Three Connecticut increased sun damage (e.g., sunburn)? State Beaches,’’ Connecticut Medicine, 2. If additional labeling is warranted, the public which have been filed in the 60(10):591–594, 1996. what information should be conveyed in docket for this public meeting (Docket 20. Green, A. et al., ‘‘Daily Sunscreen the labeling and why? No. 2006N– 0107). Some of these Application and Betacarotene Comments along with supporting data comments concern sunscreen Supplementation in Prevention of Basal-cell will help enable FDA to determine how ingredients formulated with and Squamous-cell Carcinomas of the Skin: and what information, if any, related to nanotechnology materials. FDA will file a Randomized Controlled Trial,’’ The Lancet, UV hypersensitivity due to AHAs in any comments concerning sunscreen 354:723–729, 1999. sunscreen-cosmetic products should be ingredients formulated in nanometer 21. IARC Handbooks of Cancer Prevention communicated to consumers. FDA will particle sizes received in response to Volume 5 Sunscreens, H. Vainio, ed., International Agency for Research on Cancer, also be evaluating any comments or data this proposed rule in the docket for this Lyon, France, p. 62, 2001. submitted in response to the final rulemaking and the citizen petition 22. Diffey, B.L. et al., ‘‘In Vitro Assessment guidance for cosmetic products (Docket No. 1978N–0038) and the of the Broad-spectrum Ultraviolet Protection containing AHAs. docket for the nanotechnology meeting. of Sunscreen Products,’’ Journal of the

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00042 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49111

American Academy of Dermatology, 41. Sun Protection Facts to Help You 60. Sun Basics-Skin Protection Made 43(6):1024–1035, 2000. Prevent Skin Cancer, American Cancer Simple for Everyone Under the Sun, 23. Comment No. C572. Society, 2000. American Cancer Society, Inc., 1999. 24. Gasparro, F.P., ‘‘Sunscreens, Skin 42. What You Need to Know About Skin 61. Skin Protection Guide for Everyone Photobiology, and Skin Cancer: the Need for Cancer, National Cancer Institute, 1998. Under the Sun. A Parents Guide to Sun UVA Protection and Evaluation of Efficacy,’’ 43. Committee on Environmental Health, Protection, American Cancer Society, Inc., Environmental Health Perspectives, 108 American Academy of Pediatrics, 1999. Suppl 1:71–78, 2000. ‘‘Ultraviolet Light: A Hazard to Children,’’ 62. Can Melanoma Be Prevented?, 25. Garland, C.F., F.C. Garland, and E.D. Pediatrics, 104:328–333, 1999. American Cancer Society, Inc., 2001. Gorham, ‘‘Lack of Efficacy of Common 44. Green, A. et al., ‘‘Site Distribution of 63. Diffey, B., ‘‘Has the Sun Protection Sunscreens in Melanoma Prevention’’ in Cutaneous Melanoma in Queensland,’’ Factor Had Its Day?,’’ British Medical Journal, Epidemiology, Causes and Prevention of Skin International Journal of Cancer, 53:232–236, 320:176–177, 2000. Diseases, Grob, J.J. et al. eds., Blackwell 1992. 64. Taylor, W.A., ‘‘Double Sampling Plan’’ Science Ltd., Malden, MA, pp. 151–159, 45. Franceschi, S. et al., ‘‘Site Distribution in Guide to Acceptance Sampling, R. R. 1997. of Different Types of Skin Cancer: New Donnelly & Sons, U.S.A., pp. 117–142, 1992. 26. Lowe, N.J. et al., ‘‘Low Doses of Aetiological Clues,’’ International Journal of 65. Comment No. C574. Repetitive Ultraviolet A Induce Morphologic Cancer, 67:24–28, 1996. 66. Comment No. C405. Changes in Human Skin,’’ Journal of 46. Raasch, B. et al., ‘‘Body Site Specific 67. Comment No. C404. Investigative Dermatology, 105(6):739–743, Incidence of Basal and Squamous Cell 68. Comment No. C111. 1995. Carcinoma in an Exposed Population, 69. Comment No. RPT7. 27. Seite, S. et al., ‘‘Effects of Repeated Townsville, Australia,’’ Mutation Research, 70. Comment No. C442. Suberythemal Doses of UVA in Human 422:101–106, 1998. 71. Comment No. SUP29. Skin,’’ European Journal of Dermatology, 47. Osterlind, A., K. Hou-Jensen, and O.M. 72. Memorandum from W.H. DeCamp, 7:204–209, 1997. Jensen, ‘‘Incidence of Cutaneous Malignant FDA, to C. Ganley, FDA, July 7, 2000. 28. Lavker, R.M. et al., ‘‘Quantitative Melanoma in Denmark 1978–1982. Anatomic 73. Comment No. CP12. Assessment of Cumulative Damage from Site Distribution, Histologic Types, and 74. Comment No. SUP33. Repetitive Exposures to Suberythemogenic Comparison with Non-melanoma Skin 75. Gabriel, K.L. et al., ‘‘Sun Protection Doses of UVA in Human Skin,’’ Cancer,’’ British Journal of Cancer, 58:385– Factor Testing: Comparison of FDA and DIN Photodermatology, Photoimmunology, and 391, 1988. Methods,’’ Journal of Toxicology-Cutaneous Photomedicine, 62(2):348–352, 1995. 48. Neale, R., G. Williams, and A. Green, and Ocular Toxicology, 6:357–370, 1987. 29. Lavker, R.M. et al., ‘‘Cumulative Effects ‘‘Application Patterns Among Participants 76. Comment No. C491. from Repeated Exposures to Suberythemal 77. Comment No. TR2. Randomized to Daily Sunscreen Use in a Doses of UVB and UVA in Human Skin,’’ 78. Diffey, B.L., ‘‘A Method for Broad Skin Cancer Prevention Trial,’’ Archives of Journal of the American Academy of Spectrum Classification of Sunscreens,’’ Dermatology, 138:1319–1325, 2002. Dermatology, 32(1):53–62, 1995. International Journal of Cosmetic Science, 49. Comment No. C584. 30. Elmets, C.A., A. Vargas, and C. Oresajo, 16:47–52, 1994. 50. Steinberg, C., and O. Larko, ‘‘Sunscreen ‘‘Photoprotective Effects of Sunscreens in 79. Diffey, B.L., and J. Robson, ‘‘A New Cosmetics on Sunburn and Langerhans Cell Application and its Importance for Sun Substrate to Measure Sunscreen Protection Photodamage,’’ Photodermatology, Protection Factor,’’ Archives of Dermatology, Factors Throughout the Ultraviolet Photoimmunology, and Photomedicine, 121:1400–1402, 1985. Spectrum,’’ Journal of the Society of 9(3):113–120, 1992. 51. Stokes, R., and B. Diffey, ‘‘How Well Cosmetic Chemists, 40:127–133, 1989. 31. Lavker, R., and K. Kaidbey, ‘‘The Are Sunscreen Users Protected?,’’ 80. Comment No. RPT9. Spectral Dependence for UVA-induced Photodermatology, Photoimmunology, and 81. Comment No. LET170. Cumulative Damage in Human Skin,’’ Journal Photomedicine, 13:186–188, 1997. 82. Press Release: ‘‘The Future of of Investigative Dermatology, 108(1):17–21, 52. Sayre, R.M. et al., ‘‘Product Application Sunscreen Labeling: Recommendations of the 1997. Technique Alters the Sun Protection Factor,’’ Consensus Conference on UVA Protection of 32. Seite, S. et al., ‘‘A Full-UV Spectrum Photodermatology, Photoimmunology, and Sunscreens Convened by the American Absorbing Daily Use Cream Protects Human Photomedicine, 8:222–224, 1991. Academy of Dermatology,’’ New York, NY, Skin Against Biological Changes Occurring in 53. Azurda, R.M. et al., ‘‘Sunscreen April 26, 2000. Photoaging,’’ Photodermatology, Application Technique in Photosensitive 83. Lim, H.W. et al., ‘‘American Academy Photoimmunology, and Photomedicine, Patients: a Quantitative Assessment and the of Dermatology Consensus Conference on 16(4):147–155, 2000. Effect of Education,’’ Photodermatology, UVA Protection of Sunscreens: Summary and 33. Boyd, A.S. et al., ‘‘The Effects of Photoimmunology and Photomedicine, Recommendations,’’ Journal of the American Chronic Sunscreen Use on the Histologic 16:53–66, 2000. Academy of Dermatology, 44:505–508, 2001. Changes of Dermatoheliosis,’’ Journal of the 54. Bech-Thomsen, N., and H.C. Wulf, 84. Cole, C., ‘‘Sunscreen Protection in the American Academy of Dermatology, ‘‘Sunbather’s Application of Sunscreen is Ultraviolet A Region: How to Measure 33(6):941–946, 1995. Probably Inadequate to Obtain Sun Effectiveness,’’ Photodermatology, 34. Comment No. CP15. Protection Factor Assigned to the Photoimmunology, and Photomedicine, 17:2– 35. Center for Disease Control and Preparation,’’ Photodermatology, 10, 2001. Prevention, ‘‘Guidelines for School Programs Photoimmunolology and Photomedicine, 85. Comment No. C137. To Prevent Skin Cancer’’ in Morbidity and 9:242–244, 1992. 86. Setlow, R.B., et al., ‘‘Wavelengths Mortality Weekly Report, 1–18, 2002. 55. Diffey, B., ‘‘People Do Not Apply Effective in Induction of Malignant 36. Ries, A.G. et al., ‘‘The Annual Report Enough Sunscreen for Protection,’’ British Melanoma,’’ Proceedings of the American to the Nation on the Status of Cancer, 1973– Medical Journal, 313:942, 1996. Academy of Science, U.S.A., 90:6666–6670, 1997, with a Special Section on Colorectal 56. Wright, M.W. et al., ‘‘Mechanisms of 1993. Cancer,’’ Cancer, 88(10):2398–2424, 2000. Sunscreen Failure,’’ Journal of the American 87. Moyal, D. et al., ‘‘UVA Protection 37. Cancer Facts and Figures 2007, Academy of Dermatology, 44:781–784, 2001. Efficacy of Sunscreens Can Be Determined by American Cancer Society, Inc., 2007. 57. Rigel, D. et al., ‘‘American Academy of the Persistent Pigment Darkening (PPD) 38. Sun Safety Tips, The American Dermatology’s Melanoma/Skin Cancer Method (Part 2),’’ Photodermatology, Academy of Dermatology, 1999. Detection and Prevention Month Press Photoimmunology, and Photomedicine, 39. Skin Cancer: Preventing America’s Release,’’ April 25, 2001. 16:250–255, 2000. Most Common Cancer 2001 Choose Your 58. The ABCs for Fun in the Sun 88. Moyal, D. et al., ‘‘Determination of Cover, Centers for Disease Control and Educational Pamphlet, The American UVA Protection Factors Using the Persistent Prevention, 2001. Academy of Dermatology, 1999. Pigment Darkening (PPD) Method as the 40. The Sun, UV, and You: A Guide to 59. Ultraviolet Index—What You Need to Endpoint (Part 1) Calibration of the Method,’’ SunWise Behavior, Environmental Protection Know, The American Academy of Photodermatology, Photoimmunology, and Agency, 1999. Dermatology, 1996. Photomedicine, 16:245–249, 2000.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00043 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49112 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

89. Chardon, A. et al., ‘‘Persistent Pigment- § 352.3 Definitions. ingredients used in the combination Darkening Response as a Method for * * * * * multiplied by 2. Evaluation of Ultraviolet A Protection (b) Minimal pigmenting dose (MPD). * * * * * Assays’’ in Sunscreens: Development, The quantity of erythema-effective Evaluation, and Regulatory Aspects, 2nd 7. Section 352.50 is revised to read as edition, Marcel Dekker, Inc., New York, NY, energy (expressed as Joules per square follows: pp. 559–582, 1997. meter) required to produce the first § 352.50 Principal display panel of all 90. Comment No. TS3. perceptible pigment darkening. sunscreen drug products. 91. Diffey, B.L. et al., ‘‘Suncare Product (c) Product category designation Photostability: a Key Parameter for a More (PCD). A labeling designation for (a) UVB sunburn protection Realistic In Vitro Efficacy Evaluation,’’ sunscreen drug products to aid in designation—(1) For products with an European Journal of Dermatology, 7:226–228, selecting the type of product best suited SPF of 2 to under 15. The labeling states 1997. to an individual’s complexion ‘‘UVB SPF [insert tested SPF value of 92. Sayre, R.M., and J. Dowdy, (pigmentation) and desired response to the product] low’’. ‘‘Photostability Testing of Avobenzone,’’ ultraviolet (UV) radiation. (2) For products with an SPF of 15 to Cosmetics and Toiletries, 114:85–86, 88, 90– (1) Low UVB sunburn protection under 30. The labeling states ‘‘UVB SPF 91, 1999. [insert tested SPF value of the product] 93. Urbach, F., ‘‘Man and Ultraviolet product. A sunscreen product that Radiation’’ in Human Exposure to Ultraviolet provides a sunburn protection factor medium’’. Radiation: Risks and Regulations, Passchier, (SPF) value of 2 to under 15. (3) For products with an SPF of 30 to W.F. and Bosnjakovic, B.F.M. (eds.), Elsevier (2) Medium UVB sunburn protection 50. The labeling states ‘‘UVB SPF [insert Science, New York, NY, pp. 3–6, 1987. product. A sunscreen product that tested SPF value of the product] high’’. 94. RTI International, ‘‘FDA Labeling Cost provides an SPF value of 15 to under 30. (4) For products with an SPF over 50. Model, Final Report,’’ prepared by Mary (3) High UVB sunburn protection The labeling states ‘‘UVB SPF 50 [select Muth, Erica Gledhill, and Shawn Karns, RTI. product. A sunscreen product that one of the following: ‘plus’ or ‘+’] Prepared for Amber Jessup, FDA Center for provides an SPF value of 30 to 50. highest’’. Any statement accompanying Food Safety and Applied Nutrition, Revised (4) Highest UVB sunburn protection the marketed product that states a January 2003. product. A sunscreen product that specific SPF value over 50 or similar List of Subjects provides an SPF value over 50. language indicating a person can stay in the sun more than 50 times longer than 21 CFR Part 347 * * * * * (e) Sunburn protection factor (SPF) without sunscreen will cause the Labeling, Over-the-counter drugs. value. The UV energy required to product to be misbranded under section 21 CFR Part 352 produce an MED on protected skin 502 of the Federal Food, Drug, and divided by the UV energy required to Cosmetic Act (the act) (21 U.S.C. 352). Labeling, Over-the-counter drugs, produce an MED on unprotected skin, (b) UVA protection designation—(1) Incorporation by reference. which may also be defined by the For products not providing UVA Therefore, under the Federal Food, following ratio: SPF value = MED protection according to § 352.73. The Drug, and Cosmetic Act, and under (protected skin (PS))/MED (unprotected labeling states ‘‘no UVA protection’’. authority delegated to the Commissioner skin (US)), where MED(PS) is the (i) The UVA protection designation of Food and Drugs, it is proposed that minimal erythema dose for protected shall appear on the principal display 21 CFR parts 347 and 352 be amended skin after application of 2 milligrams panel along with the UVB protection as follows: per square centimeter of the final designation in an equally prominent manner that does not conflict with the PART 347—SKIN PROTECTANT DRUG formulation of the sunscreen product, and MED(US) is the minimal erythema UVB protection designation. PRODUCTS FOR OVER-THE- (ii) The font size of the UVA COUNTER HUMAN USE dose for unprotected skin (i.e., skin to which no sunscreen product has been protection designation shall be the same 1. The authority citation for 21 CFR applied). In effect, the SPF value is the size as the UVB protection designation. part 347 continues to read as follows: reciprocal of the effective transmission (2) For products providing UVA of the product viewed as a UV radiation protection according to § 352.73. The Authority: 21 U.S.C. 321, 351, 352, 353, labeling states ‘‘UVA [select one of the 355, 360, 371. filter. 6. Section 352.20 is amended by following in accordance with § 352.73: 2. FDA is proposing to lift the stay of ✰✰✰ ✰✰ revising paragraph (a)(2) to read as ‘ Low,’ ‘ Medium,’ § 347.20(d) as published at 68 FR 33362, ✰ June 4, 2003. follows: ‘ High,’ or ‘ Highest’]’’. (i) The UVA protection designation PART 352—SUNSCREEN DRUG § 352.20 Permitted combinations of active shall appear on the principal display PRODUCTS FOR OVER THE COUNTER ingredients. panel along with the UVB protection HUMAN USE * * * * * designation in an equally prominent (a) * * * manner that does not conflict with the 3. The authority citation for 21 CFR (2) Avobenzone in § 352.10(b) may be UVB protection designation. part 352 continues to read as follows: combined with one or more sunscreen (ii) The font size of the UVA Authority: 21 U.S.C. 321, 351, 352, 353, active ingredients identified in protection designation shall be the same 355, 360, 371. § 352.10(c), (e), (f), (i) through (l), (n), size as the UVB protection designation. 4. FDA is proposing to lift the stay of (o), (q), and (r) in a single product when (iii) All star borders and the color 21 CFR part 352 as published at 68 FR used in the concentrations established inside a solid star shall be the same 33362, June 4, 2003. for each ingredient in § 352.10. The while the color of ‘‘empty’’ stars must be 5. Section 352.3 is amended by concentration of each active ingredient lighter and distinctly different than redesignating paragraphs (b) through (d) must be sufficient to contribute a solid stars. The color inside a solid star as (c) through (e), respectively; revising minimum SPF of not less than 2 to the should be distinctly different than the newly redesignated paragraphs (c) and finished product. The finished product background color. (e); and adding new paragraph (b) to must have a minimum SPF of not less (iv) The stars are to be filled in read as follows: than the number of sunscreen active starting with the first star on the left and

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00044 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49113

are to appear in a straight horizontal (v) For products not providing UVA § 352.52(b)(1)(vi) as follows: ‘‘[Bullet] line. protection according to § 352.73. The [select one of the following descriptors (c) Select one of the following: ‘‘UV labeling states ‘‘[bullet] no UVA in accordance with §§ 352.70 and rays from the sun are made of UVB and protection.’’ 352.73: ‘low,’ ‘medium,’ ‘high,’ or UVA. It is important to protect against (vi) For products providing UVA ‘highest’] UVB sunburn/UVA both UVB & UVA rays.’’ or ‘‘UV rays protection according to § 352.73. The protection’’. from the sun are made of UVB and labeling states ‘‘[bullet] [select one of (c) Warnings. The labeling of the UVA. It is important to protect against the following in accordance with product contains the following warnings both UVB & UVA rays to prevent § 352.73: ‘Low,’ ‘medium,’ ‘high,’ or under the heading ‘‘Warnings:’’ sunburn and other skin damage.’’ ‘highest’] UVA protection’’. (1) The labeling states in bold type (d) For products that satisfy the water (vii) For products that satisfy the ‘‘UV exposure from the sun increases resistant sunscreen product testing water resistant testing procedures the risk of skin cancer, premature skin procedures in § 352.76. The labeling identified in § 352.76. The labeling aging, and other skin damage. It is states (select one of the following: states ‘‘[bullet] retains SPF after 40 important to decrease UV exposure by ‘‘water,’’ ‘‘water/sweat,’’ or ‘‘water/ minutes of [select one or more of the limiting time in the sun, wearing perspiration’’) ‘‘resistant.’’ following: ‘activity in the water,’ protective clothing, and using a (e) For products that satisfy the very ‘swimming,’ ‘sweating,’ ‘perspiring,’ sunscreen.’’ water resistant sunscreen product ‘swimming/sweating,’ or ‘swimming/ (2) The labeling states ‘‘When using testing procedures in § 352.76. The perspiring’]’’. this product [bullet] keep out of eyes. labeling states ‘‘very’’ (select one of the (viii) For products that satisfy the very Rinse with water to remove.’’ following: ‘‘water,’’ ‘‘water/sweat,’’ or water resistant testing procedures (3) The labeling states ‘‘Stop use and ‘‘water/perspiration’’) ‘‘resistant.’’ identified in § 352.76. The labeling ask a doctor if [bullet] skin rash occurs’’. 8. Section 352.52 is amended by states ‘‘[bullet] retains SPF after 80 (d) Directions. The labeling of the revising paragraphs (b), (c), (d), (e), the minutes of [select one or more of the product contains the following heading of paragraph (f), paragraphs following: ‘activity in the water,’ statements, as appropriate, under the (f)(1)(ii) through (f)(1)(vi) to read as ‘swimming,’ ‘sweating,’ ‘perspiring,’ heading ‘‘Directions.’’ More detailed follows: ‘swimming/sweating,’ or ‘swimming/ directions applicable to a particular perspiring’]’’. product formulation (e.g., cream, gel, § 352.52 Labeling of sunscreen drug (2) Additional indications. In addition lotion, oil, spray, etc.) may also be products. to the indications provided in paragraph included. * * * * * (b)(1) of this section, the following may (1) For products containing any (b) Indications. The labeling of the be used for products containing any ingredient in § 352.10. (i) The labeling product states, under the heading ingredient in § 352.10: states ‘‘[bullet] apply [select one of the ‘‘Uses,’’ all of the phrases listed in (i) For products with an SPF of 2 to following: ‘liberally’ or ‘generously’] paragraph (b)(1) of this section that are under 15. Select one or both of the [and, as an option: ‘and evenly’] [insert applicable to the product and may following: ‘‘[Bullet] provides low appropriate time interval, if a waiting contain any of the additional phrases protection against [select one of the period is needed] before sun exposure’’. listed in paragraph (b)(2) of this section, following: ‘sunburn’ or ‘sunburn and (ii) The labeling states ‘‘[bullet] apply as appropriate. Other truthful and tanning’]’’ or ‘‘[bullet] for skin that and reapply as directed to avoid nonmisleading statements, describing sunburns minimally’’. lowering protection’’. only the uses that have been established (ii) For products with an SPF of 15 to (iii) As an option, the labeling may and listed in this paragraph (b), may under 30. Select one or both of the state ‘‘[bullet] apply to all skin exposed also be used, as provided in § 330.1(c)(2) following: ‘‘[Bullet] provides medium to the sun’’. of this chapter, subject to the provisions protection against [select one of the (iv) The labeling states ‘‘[bullet] of section 502 of the act (21 U.S.C. 352) following: ‘sunburn’ or ‘sunburn and children under 6 months of age: ask a relating to misbranding and the tanning’]’’ or ‘‘[bullet] for skin that doctor’’. prohibition in section 301(d) of the act sunburns moderately’’. (2) For products that satisfy the water (21 U.S.C. 331(d)) against the (iii) For products with an SPF of 30 resistant or very water resistant testing introduction or delivery for introduction to 50. Select one or both of the procedures identified in § 352.76. The into interstate commerce of unapproved following: ‘‘[Bullet] [select one of the labeling states ‘‘[bullet] reapply after new drugs in violation of section 505(a) following: ‘provides high’ or ‘high’] [select one of the following: ‘40 minutes of the act (21 U.S.C. 355(a)). protection against [select one of the of’ or ‘80 minutes of’ for products that (1) For products containing any following: ‘sunburn’ or ‘sunburn and satisfy either the water resistant or very ingredient in § 352.10. (i) For products tanning’]’’ or ‘‘[bullet] for skin highly water resistant test procedures in with an SPF of 2 to under 15. The sensitive to sunburn’’. § 352.76, respectively] swimming or labeling states ‘‘[bullet]1 low UVB (iv) For products with an SPF over 50. [select one or more of the following: sunburn protection’’. Select one or both of the following: ‘sweating’ or ‘perspiring’] and after (ii) For products with an SPF of 15 to ‘‘[Bullet] [select one of the following: towel drying. Otherwise, reapply at least under 30. The labeling states ‘‘[bullet] ‘provides highest’ or ‘highest’] every 2 hours’’. medium UVB sunburn protection’’. protection against [select one of the (3) For products that do not satisfy the (iii) For products with an SPF of 30 following: ‘sunburn’ or ‘sunburn and water resistant or very water resistant to 50. The labeling states ‘‘[bullet] high tanning’]’’ or ‘‘[bullet] for skin extremely testing procedures identified in § 352.76. UVB sunburn protection’’. sensitive to sunburn’’. The labeling states ‘‘[bullet] reapply at (iv) For products with an SPF over 50. (v) If the UVA descriptor in least every 2 hours and after towel The labeling states ‘‘[bullet] highest § 352.52(b)(1)(vi) is the same as the SPF drying, swimming, or [select one of the UVB sunburn protection’’. descriptor in § 352.52(b)(1)(i) through following: ‘sweating’ or ‘perspiring’]’’. (b)(1)(iv), then the statement in (e) Statement on product 1 See § 201.66(b)(4) of this chapter for definition § 352.52(b)(1)(i) through (b)(1)(iv) may performance—(1) For products of bullet symbol. be combined with the statement in containing any ingredient identified in

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00045 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49114 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

§ 352.10. The following product ‘‘UV exposure from the sun increases paragraph. When the time intervals or category designation (PCD) labeling the risk of skin cancer, premature skin age limitations for administration of the claims may be used under the heading aging, and other skin damage. It is individual ingredients differ, the ‘‘Other information’’ or anywhere important to decrease UV exposure by directions for the combination product outside of the ‘‘Drug Facts’’ box or limiting time in the sun, wearing may not contain any dosage that enclosure and shall not be intermixed protective clothing, and using a exceeds those established for any with the information required under sunscreen. [in bold type]’’ ‘‘[bullet] keep individual ingredient in the applicable § 352.50(a). out of eyes’’ ‘‘[bullet] stop use if skin OTC drug monograph(s), and may not (i) For products with an SPF of 2 to rash occurs.’’ provide for use by any age group lower under 15. The labeling states ‘‘low (iv) The warning in § 201.66(c)(5)(x) than the highest minimum age limit sunburn protection product’’. of this chapter may be limited to the established for any individual (ii) For products with an SPF of 15 to following: ‘‘Keep out of reach of ingredient. For permitted combinations under 30. The labeling states ‘‘medium children.’’ containing a sunscreen and a skin sunburn protection product’’. (v) For lip protectant products protectant identified in § 352.20(b), the (iii) For products with an SPF of 30 containing any ingredient identified in directions for sunscreens in § 352.52(d) to 50. The labeling states ‘‘high sunburn § 352.10. The heading and the must be used. For products marketed as protection product’’. indication required by § 201.66(c)(4) of a lip protectant with sunscreen, (iv) For products with an SPF over 50. this chapter may be limited to ‘‘Use [in § 352.52(f)(1)(vi) applies. bold type] helps prevent sunburn and The labeling states ‘‘highest sunburn 10. Sections 352.70 through 352.73 chapped lips’’. The warnings required protection product’’. are revised as follows: (2) For products containing any in paragraph (f)(1)(iii) of this section ingredient identified in § 352.10. The may be limited to the following: ‘‘Stop Subpart D—Testing Procedures following labeling statement may be use if skin rash occurs.’’ The warning Sec. used under the heading ‘‘Other required in paragraph (f)(1)(iv) of this 352.70 SPF testing procedure. information’’ or anywhere outside of the section may be omitted. The directions 352.71 UVA in vitro testing procedure. ‘‘Drug Facts’’ box or enclosure and shall in paragraphs (d)(2) and (d)(3) of this 352.72 UVA in vivo testing procedure. not be intermixed with the information section may be limited to the following: 352.73 Determination of the labeled UVA required under § 352.50(a). The labeling ‘‘apply liberally and reapply at least protective value. states ‘‘higher SPF products give more every 2 hours for sunburn protection’’. * * * * * (vi) For lipsticks, lip products to sun protection, but are not intended to § 352.70 SPF testing procedure. extend the time spent in the sun’’. prolong wear of lipstick, lip gloss, and (3) For products containing any lip balm containing any ingredient (a) Standard sunscreens—(1) ingredient identified in § 352.10 and identified in § 352.10 and identified in Laboratory validation. A standard that satisfy the requirements in § 352.73 § 720.4(c)(7) of this chapter. The sunscreen shall be used concomitantly for a labeled UVA protection value. The labeling is identical to that in paragraph in the testing procedures for following labeling statements may be (f)(1)(v) of this section except the determining the SPF value of a used anywhere outside of the ‘‘Drug heading and the indication required by sunscreen drug product to ensure the Facts’’ box or enclosure and shall not be § 201.66(c)(4) of this chapter are limited uniform evaluation of sunscreen drug intermixed with the information to ‘‘Use [in bold type] helps prevent products. required under § 352.50(a). sunburn’’. (i) For products with an SPF of 2 to (i) The labeling states ‘‘broad * * * * * 15. The standard sunscreen shall be an spectrum sunscreen’’. 9. Section 352.60 is amended by 8-percent homosalate preparation with a (ii) The labeling states ‘‘provides revising paragraphs (c) and (d) to read mean SPF value of 4.47 (standard [select one of the following: ‘UVA and as follows: deviation = 1.28). In order for the SPF UVB,’ or ‘broad spectrum’] protection’’. § 352.60 Labeling of permitted determination of a test product to be (iii) The labeling states ‘‘protects from combinations of active ingredients. considered valid, the SPF of the standard sunscreen must fall within the UVA and UVB [select one of the * * * * * following: ‘rays’ or ‘radiation’]’’. standard deviation range of the expected (c) Warnings. The labeling of the ± (iv) The labeling states ‘‘[select one of product states, under the heading SPF (i.e., 4.47 1.28). Optionally, the the following: ‘absorbs’ or ‘protects’] ‘‘Warnings,’’ the warning(s) for each standard sunscreen in paragraph within the UVA spectrum’’. ingredient in the combination, as (a)(1)(ii) of this section may be used. (f) Products, including cosmetic-drug established in the warnings section of (ii) For products with an SPF over 15 products, containing any ingredient the applicable OTC drug monographs, (optional for SPF values of 2 to 15). The identified in § 352.10 labeled for use except that the warning for skin standard sunscreen shall be an SPF 15 only on specific small areas of the face protectants in § 347.50(c)(3) of this formulation containing 7 percent (e.g., lips, nose, ears, and/or around the chapter is not required for permitted padimate O and 3 percent oxybenzone eyes) and that meet the criteria combinations containing a sunscreen with a mean SPF value of 16.3 (standard established in § 201.66(d)(10) of this and a skin protectant identified in deviation = 3.43). In order for the SPF chapter. *** § 352.20(b). For products marketed as a determination of a test product to be (1) * * * lip protectant with sunscreen, considered valid, the SPF of the * * * * * § 352.52(f)(1)(vi) applies. standard sunscreen must fall within the (ii) The indication required by (d) Directions. The labeling of the standard deviation range of the expected § 201.66(c)(4) of this chapter may be product states, under the heading SPF (i.e., 16.3 ± 3.43). limited to the following: ‘‘Use [in bold ‘‘Directions,’’ directions that conform to (2) Standard homosalate sunscreen— type] helps prevent sunburn.’’ the directions established for each (i) Preparation of the standard (iii) The warnings required by ingredient in the directions sections of homosalate sunscreen. (A) The standard § 201.66(c)(5)(i) through (c)(5)(ix) of this the applicable OTC drug monographs, homosalate sunscreen is prepared from chapter may be limited to the following: unless otherwise stated in this two different preparations (preparation

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00046 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49115

A and preparation B) with the following preparation by the following method to (2) Add approximately 75 milliliters compositions: ensure proper concentration: of isopropanol and heat with swirling (A) Reagents. (1) Acetic acid, glacial, until the sample is evenly dispersed. COMPOSITION OF PREPARATION A AND ACS grade. (3) Cool to room temperature (15 to 30 PREPARATION B OF THE HOMOSALATE (2) Isopropanol, HPLC grade. °C) and dilute to volume with STANDARD SUNSCREEN (3) Methanol, HPLC grade. isopropanol. Mix well. (4) Homosalate, USP reference (4) Pipet 25.0 milliliters of this sample Percent by standard. Ingredients weight preparation into a 100-milliliter (B) Instrumentation. Equilibrate a volumetric flask and dilute to volume Preparation A suitable liquid chromatograph to the with isopropanol. Mix well. following or equivalent conditions: (E) System suitability. (1) Three Lanolin 5.00 Column ...... Ultrasphere ODS 150 x 4.6 replicate injections of the standard Homosalate 8.00 millimeters (5 microns), or preparation (described in paragraph White petrolatum 2.50 Ultrasphere ODS 250 x 4.6 (a)(2)(ii)(C)(2) of this section) will yield Stearic acid 4.00 millimeters (5 microns) Propylparaben 0.05 Mobile Phase ..... 85:15:0.5 meth- a relative standard deviation of not more anol:water:acetic acid than 2.0 percent calculated on peak Preparation B Flow Rate ...... 1.5 milliliters per minute areas for homosalate. Temperature ...... Ambient (2) In case a system fails to meet this Methylparaben 0.10 Detector ...... UV spectrophotometer at 308 nanometers criterion, adjusting the mobile phase or Edetate disodium 0.05 replacing the column may be necessary Propylene glycol 5.00 Attenuation ...... As needed Injection Amount 10 microliters Triethanolamine 1.00 to obtain suitable chromatography. Purified water USP 74.30 (C) Standard preparation. (1) (F) Analysis. (1) Inject 10 microliters Accurately weigh 0.50 gram of of the standard preparation (described (B) Preparation A and preparation B homosalate USP reference standard into in paragraph (a)(2)(ii)(C) of this section) are heated separately to 77 to 82 °C, a 250-milliliter volumetric flask. in triplicate and collect data for about with constant stirring, until the contents Dissolve and dilute to volume with 15 minutes or until both homosalate of each part are solubilized. Add isopropanol. Mix well. (two isomers) peaks have completely preparation A slowly to preparation B (2) Accurately pipet 20.0 milliliters of eluted. while stirring. Continue stirring until the homosalate solution (described in (2) Similarly inject 10 microliters of the emulsion formed is cooled to room paragraph (a)(2)(ii)(C)(1) of this section) each sample preparation. temperature (15 to 30 °C). Add sufficient into a 100-milliliter volumetric flask. (3) The system suitability purified water to obtain 100 grams of Dilute to volume with isopropanol and requirements must be met. standard sunscreen preparation. mix well. This is the standard (G) Calculation. Sum the peak areas of (ii) High performance liquid preparation. the two homosalate isomers for each chromatography (HPLC) assay of the (D) Sample preparation. (1) injection and calculate the percent standard homosalate sunscreen. Assay Accurately weigh 2.0 grams of sample (weight/weight) homosalate content in the standard homosalate sunscreen into a 100-milliliter volumetric flask. the sample preparation as follows:

(3) Standard padimate O/oxybenzone COMPOSITION OF THE PADIMATE O/ COMPOSITION OF THE PADIMATE O/ sunscreen—(i) Preparation of the OXYBENZONE STANDARD SUN- OXYBENZONE STANDARD SUN- standard padimate O/oxybenzone SCREEN SCREEN—Continued sunscreen. The standard sunscreen is prepared from four different parts (parts Percent Percent A, B, C, and D) with the following Ingredients by Ingredients by weight weight compositions: Part A Oxybenzone ...... 3.00 Propylparaben ...... 0.10 Lanolin ...... 4.50 Cocoa butter ...... 2.00 Part B Glyceryl monostearate ...... 3.00 Stearic acid ...... 2.00 Purified water USP ...... 71.60 Padimate O ...... 7.00 Sorbitol solution ...... 5.00

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00047 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.005 49116 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

COMPOSITION OF THE PADIMATE O/ ensure that the standard sunscreen (D) Sample preparation. (1) Weigh 1.0 OXYBENZONE STANDARD SUN- contains proper amounts of padimate O gram of sample into a 50-milliliter SCREEN—Continued and oxybenzone, analyze it against USP volumetric flask. reference standards for padimate O and (2) Add approximately 30 milliliters Percent oxybenzone in a high performance of isopropanol and heat with swirling Ingredients by liquid chromatography procedure using until the sample is evenly dispersed. weight the following parameters: (3) Cool to room temperature (15 to 30 ° Triethanolamine, 99 percent .... 1.00 (A) Reagents. (1) Acetic acid, glacial, C) and dilute to volume with Methylparaben ...... 0.30 ACS grade. isopropanol. Mix well. (2) Isopropanol, HPLC grade. (4) Pipet 5.0 milliliters of this sample Part C (3) Methanol, HPLC grade. preparation into a 50-milliliter Benzyl alcohol ...... 0.50 (4) Oxybenzone, USP reference volumetric flask and dilute to volume standard. with isopropanol. Mix well. Part D (5) Padimate O, USP reference (E) System suitability. (1) Three standard. replicate injections of the standard Purified water USP ...... QS1 (B) Instrumentation. Equilibrate a preparation (described in paragraph 1 Quantity sufficient to make 100 grams suitable liquid chromatograph to the (a)(3)(ii)(C) of this section) will yield a following or equivalent conditions: relative standard deviation of not more (A) Step 1. Add the ingredients of Part than 2.0 percent calculated on peak A into a suitable stainless steel kettle Column ...... Ultrasphere ODS 250 x 4.6 areas for oxybenzone and padimate O. equipped with a propeller agitator. Mix millimeters (5 microns), or at 77 to 82 °C until uniform. Supelcosil LC-18 DB 250 x (2) A calculated resolution between 4.6 millimeters (5 microns) the oxybenzone and padimate O peaks (B) Step 2. Add the water of Part B Mobile Phase ..... 85:15:0.5 meth- into a suitable stainless steel kettle anol:water:acetic acid will be not less than 3.0. equipped with a propeller agitator and Flow Rate ...... 1.5 milliliters per minute (3) In case a system fails to meet this begin mixing and heating to 77 to 82 °C. Temperature ...... Ambient criterion, adjusting the mobile phase or Detector ...... UV spectrophotometer at 308 Add the remaining ingredients of Part B nanometers replacing the column may be necessary and mix until uniform. Maintain Attenuation ...... As needed to obtain suitable chromatography. temperature at 77 to 82 °C. Injection Amount 10 microliters (F) Analysis. (1) Inject 10 microliters (C) Step 3. Add the batch of Step 1 at (C) Standard preparation. (1) Weigh of the standard preparation (described 77 to 82 °C to the batch of Step 2 at 77 in paragraph (a)(3)(ii)(C) of this section) to 82 °C, and mix until smooth and 0.50 gram of oxybenzone reference standard into a 250-milliliter volumetric in triplicate and collect data for about uniform. Slowly cool the batch to 49 to 15 minutes or until the padimate O peak 54 °C. flask. Dissolve and dilute to volume with isopropanol. Mix well. has completely eluted. Elution order is (D) Step 4. Add the benzyl alcohol of oxybenzone, then padimate O. Part C to the batch of Step 3 at 49 to 54 (2) Weigh 0.50 gram of padimate O (2) Similarly inject 10 microliters of °C. Mix until uniform. Continue to cool reference standard into a 250-milliliter each sample preparation. batch to 35 to 41 °C. volumetric flask. Dissolve and dilute to (E) Step 5. Add sufficient water of volume with isopropanol. Mix well. (3) The system suitability Part D to the batch of Step 4 at 35 to 41 (3) Pipet 3.0 milliliters of the requirements must be met. °C to obtain 100 grams of standard oxybenzone solution and 7.0 milliliters (G) Calculation. Calculate the percent sunscreen preparation. Mix until of the padimate O solution into a 100- (weight/weight) of each sunscreen uniform. Cool batch to 27 to 32 °C. milliliter volumetric flask. Dilute to ingredient in the sample preparation as (ii) HPLC assay of the standard volume with isopropanol and mix well. follows: padimate O/oxybenzone sunscreen. To This is the standard preparation. (1) Oxybenzone (percent weight)

(2) Padimate O (percent weight)

(b) Light source (solar simulator)—(1) emission spectrum from 290 to 400 and the following percentage of Emission spectrum. A solar simulator nanometers (nm) with a limit of 1,500 erythema-effective radiation in each used for determining the SPF of a watts per square meter (W/m2) on total specified range of wavelengths: sunscreen drug product should be solar simulator irradiance for all filtered so that it provides a continuous wavelengths between 250 and 1400 nm

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00048 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.006 EP27AU07.007 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49117

SOLAR SIMULATOR EMISSION IV: Burns minimally; always tans well will be at least 1 square centimeter SPECTRUM (moderate brown) (normal). (cm2) in area and will be separated from V: Rarely burns; tans profusely (dark brown) each other by at least 1 cm. Mark the (insensitive). location of each test subsite with Wavelength range Percent erythemal VI: Never burns; deeply pigmented (nm) contribution (insensitive). indelible ink. (5) Application of test materials. < 290 < 0.1 Determine that the subject is not taking Apply the test sunscreen product and 290–310 46.0–67.0 topical or systemic medication that is the standard sunscreen at 2 milligrams 290–320 80.0–91.0 known to alter responses to ultraviolet per square centimeter (mg/cm2) to their 290–330 86.5–95.0 radiation and that the subject has no 290–340 90.5–97.0 respective test sites to establish standard 290–350 93.5–98.5 history of sensitivities to topical films. Test sites will be randomly 290–400 93.5–100.0 products and/or abnormal responses to located on the back in a blinded sunlight, such as a phototoxic or manner. Use a finger cot compatible (2) Operation. A solar simulator photoallergic response. with the sunscreen to spread the should have no significant time related (2) Physical examination. Conduct a product as evenly as possible. Pretreat fluctuations (within 20 percent) in physical examination to determine the the finger cot by saturating with the radiation emissions after an appropriate presence of sunburn, suntan, scars, sunscreen and then wiping off material warmup time and good beam uniformity active dermal lesions, and uneven skin before application. Pretreatment is (within 20 percent) in the exposure tones on the areas of the back to be meant to ensure that sunscreen is plane. The average delivered dose to the tested. A suitable source of low power applied at the correct density of 2 mg/ UV exposure site must be within 10 UVA, such as a Woods lamp, is helpful cm2. percent of the prescribed dose. in this process. If any of these (6) Waiting period. Before exposing (3) Periodic measurement. To ensure conditions are present, the subject is not the test site areas after applying a that the solar simulator delivers the qualified to participate in the study. The product, wait at least 15 minutes. appropriate spectrum of UV radiation, presence of nevi, blemishes, or moles (7) Number of subjects—(i) For the emission spectrum of the solar will be acceptable if in the physician’s products with an expected SPF under simulator must be measured every 6 judgment they will neither compromise 30. A test panel shall consist of 20 to 25 months with an appropriate and the study, nor jeopardize subject safety. subjects with at least 20 subjects who accurately calibrated spectroradiometer Subjects with dysplastic nevi should not produce valid data for analysis. Data are system (results should be traceable to be enrolled. Excess hair on the back is valid unless rejected in accordance with the National Institute for Standards and acceptable if the hair is clipped. paragraph (c)(9) of this section. If more Technology). In addition, the solar Shaving is unacceptable because it may than 5 subjects are rejected based on simulator must be recalibrated if there is remove a significant portion of the paragraph (c)(9) of this section, the any change in the lamp bulb or the stratum corneum and temporarily panel is disqualified, and a new panel optical filtering components (i.e., filters, increase skin permeability to ultraviolet must be created. mirrors, lenses, collimating devices, or radiation. (ii) For products with an expected focusing devices). Daily solar simulator (3) Informed consent. Obtain legally SPF of 30 or over. A test panel shall radiation intensity should be monitored effective written informed consent from consist of 25 to 30 subjects with at least with a broadband radiometric device all subjects. 25 subjects who produce valid data for that is sensitive primarily to UV (4) Test site delineation—(i) Test site. analysis. Data are valid unless rejected radiation. The broadband radiometric A test site is the location on the back for in accordance with paragraph (c)(9) of device should be calibrated using side determining the subject’s initial and this section. If more than 5 subjects are by side comparison with the final minimal erythema dose (MED) for rejected based on paragraph (c)(9) of this spectroradiometer at the time of the unprotected skin and for determining section, the panel is disqualified, and a semiannual spectroradiometric SPF values after application of the new panel must be created. measurement of the solar simulator. If a sunscreen standard and the test (8) Response criteria. In order that the lamp must be replaced due to failure or sunscreen product(s). There typically person who evaluates the MED aging during a phototest, broadband are 4 to 6 test sites for each subject. Test responses is not biased, he/she must not device readings consistent with those sites should be located on the back be the same person who applied the obtained for the original calibrated lamp between the beltline and the shoulder sunscreen drug product to the test site will suffice until measurements can be blades (scapulae) and lateral to the or administered the doses of UV performed with the spectroradiometer at midline. Each test site shall be a radiation. After UV radiation exposure the earliest possible opportunity. minimum of 50 square centimeters, e.g., from the solar simulator is completed, (c) General testing procedures—(1) 5 x 10 centimeters. Outline the test sites all immediate responses shall be Medical history. Obtain a medical to which the sunscreen standard and the recorded. These may include an history from each subject with emphasis test sunscreen product(s) will be immediate darkening or tanning, on the effects of sunlight on his/her applied with indelible ink. If the subject typically grayish or purplish in color, skin. Determine that each subject is in is to receive the doses of ultraviolet which fades in 30 to 60 minutes; an good general health with skin type I, II, radiation in an upright (seated) position, immediate reddening at the subsite, due or III (as described in this paragraph). draw the lines on the skin with the to heating of the skin, which fades Skin Type and Sunburn and Tanning subject upright (seated). If the subject is rapidly; and an immediate generalized History (Based on first 30 to 45 minutes to receive the doses of ultraviolet heat response, spreading beyond the of sun exposure after a winter season of radiation while prone, draw the lines subsite, which fades in 30 to 60 no sun exposure). with the subject prone. minutes. After the immediate responses (ii) Test subsite. Test subsites are the are noted, each subject shall shield the I: Always burns easily; never tans (sensitive). II: Always burns easily; tans minimally locations to which ultraviolet radiation exposed area from further UV radiation (sensitive). is administered within a test site. At until the MED response is evaluated. III: Burns moderately; tans gradually (light least 5 test subsites will receive UV Determine the MED 16 to 24 hours after brown) (normal). doses within each test site. Test subsites exposure. Evaluate the erythema

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00049 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 49118 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

responses of each test site using either more intense redness. For subsites exposed testing sites from further UV tungsten or warm white fluorescent showing an erythema response, the radiation until the MED is read. lighting that provides 450 to 550 lux of maximal exposure should be no more (d) Determination of SPF—(1) illumination at the test site. For the than twice the total energy of the Determination of erythema action evaluation, the test subject should be in minimal exposure. spectrum. (i) Use the following erythema action spectrum as weighting the same position used when the test (9) Rejection of test data. Reject test factors to calculate the erythema- site was irradiated. For each test site, data if the exposure series fails to elicit determine the smallest UV dose that effective exposure produced by a solar an MED response on either the treated simulator: produced redness reaching the borders or unprotected skin sites; or all subsites Vi (λ) = 1.0 (250 < λ < 298 nm) of the test subsite. The MED is the 0.094 * (298 - ) within a test site show more intense Vi (λ) = 10 l (298 < λ < 328 quantity of erythema-effective energy responses than the threshold erythema nanometers) required to produce the first perceptible, λ 0.015 * (140 - ) λ response; or the responses are Vi ( ) = 10 l (328 < < 400 redness reaction with clearly defined inconsistent with the series of UV doses nanometers) borders at 16 to 24 hours post-exposure. administered; or the subject was (ii) Integrate the erythemally-effective To determine the MED, there must be at noncompliant, e.g., the subject spectral irradiance over wavelength and least one subsite that received a smaller withdraws from the test due to illness time to calculate the erythema-effective UV dose and does not produce redness or work conflicts or does not shield the UV dose delivered by a solar simulator as well as a subsite(s) with somewhat as follows:

(iii) The erythema action spectrum Select the doses that are a geometric product. For products with an expected may be determined using a handheld series represented by (1.25n), wherein SPF between 8 and 15, use exposures radiometer with a response weighted to each exposure dose is 25 percent greater that are the initial unprotected MED match the spectrum in ‘‘CIE S 007/E than the previous exposure dose to times 0.69X, 0.83X, 1.00X, 1.20X, and Erythemal Reference Action Spectrum maintain the same relative uncertainty 1.44X, where X equals the expected SPF and Standard Erythema Dose,’’ dated (expressed as a constant percentage), of the test product. For products with an 1998, which is incorporated by independent of the subject’s sensitivity expected SPF greater that 15, use reference in accordance with 5 U.S.C. to UV radiation. Usually, the UV exposures that are the initial 552(a) and 1 CFR part 51. Copies are radiation for determining the initial unprotected MED times 0.76X, 0.87X, available from CIE Central Bureau, unprotected MED is administered the 1.00X, 1.15X, and 1.32X, where X Kegelgasse 27, A–1030, Vienna, Austria, day prior to applying the sunscreen equals the expected SPF of the test or may be examined at the Center for product and standard sunscreen, and product. The MED is the smallest Drug Evaluation and Research, 10903 the responses then are evaluated erythemally-effective UV dose required New Hampshire Ave., Bldg. 22, Silver immediately prior to applying the to produce mild redness within the Spring, MD, or at the Office of the sunscreen product and sunscreen subsite border at 16 to 24 hours post- Federal Register, 800 North Capitol St. standard. Determine the final exposure. Calculate the SPF value of NW., suite 700, Washington, DC. It is unprotected MED on the same day that each sunscreen product and sunscreen advisable to measure the solar simulator UV radiation is administered to the standard using the MED of sunscreen- output before and after each phototest sunscreen-protected test sites. Use the protected skin (MED(PS)) and the final or, at a minimum, at the beginning and final unprotected MED (MED(US)) in unprotected skin MED (MED(US)) as end of each test day. This radiometer calculating SPF. follows: should be calibrated using side by side (3) Determination of individual SPF comparison with the spectroradiometer values. Administer a series of five UV (using the weighting factors determined radiation doses expressed as J/m2-eff according to paragraph (d)(1)(i) of this (adjusted to the erythema action (4) Determination of the test product section) at the time of the semiannual spectrum calculated according to SPF and PCD. Use data from at least 20 spectroradiometric measurement of the paragraph (d)(1) of this section) to the test subjects with n representing the solar simulator. subsites within each test site on a number of subjects used. First, compute (2) Determination of MED of subject using an accurately calibrated the SPF value for each subject as stated unprotected skin. Administer a series of solar simulator. The five UV doses will in paragraphs (d)(2) and (d)(3) of this five UV radiation doses expressed as J/ be a geometric series as described in section. Second, compute the mean SPF m2-eff (adjusted to the erythema action paragraph (d)(2) of this section, where value, x¯, and the standard deviation, s, spectrum calculated according to the middle exposure represents the for these subjects. Third, obtain the paragraph (d)(1) of this section) to the expected SPF. For products with an upper 5-percent point from Student’s t subsites within each test site on a expected SPF less than 8, use exposures distribution table with n-1 degrees of subject using an accurately calibrated that are the product of the initial freedom. Denote this value by t. Fourth, solar simulator. Use the series of five unprotected MED times 0.64X, 0.80X, compute ts/√n. Denote this quantity by exposures to the unprotected test site to 1.00X, 1.25X, and 1.56X, where X A (i.e., A = ts/√n). Fifth, calculate the determine the initial unprotected MED. equals the expected SPF of the test SPF value to be used in labeling as

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00050 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.008 EP27AU07.009 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49119

follows: The label SPF equals the largest should be set to provide a bandwidth reference in accordance with 5 U.S.C. whole number less than x¯ - A. Sixth and that is less than or equal to 5 552(a) and 1 CFR part 51. Copies are last, the sunscreen product is classified nanometers. available from CIE Central Bureau, into a PCD as follows: If 50 + A < x¯, the (e) Sunscreen drug product Kegelgasse 27, A–1030, Vienna, Austria, PCD is Highest; if 30 + A ≤ x¯ ≤ 50 + A, application to substrate. The accuracy or may be examined at the Center for the PCD is High; if 15 + A ≤ x¯ < 30 + of the test depends upon the application Drug Evaluation and Research, 10903 A, the PCD is Medium; if 2 + A ≤ x¯ < of a precisely controlled amount of New Hampshire Ave., Bldg. 22, Silver 15 + A, the PCD is Low; if x¯ < 2 + A, sunscreen product with a uniform Spring, MD 20993, or at the Office of the the product shall not be labeled as an distribution over the application area of Federal Register, 800 North Capitol St. OTC sunscreen drug product and may the substrate. The product is applied at NW., suite 700, Washington, DC. not display an SPF value. 2 milligrams per square centimeter to (g) Calculation of the spectral the substrate. To achieve uniform transmittance at each wavelength § 352.71 UVA in vitro testing procedure. distribution over the substrate, the interval. The dynamic range of the (a) Light source for transmittance/ sunscreen product should be applied in measurement system and the intensity absorbance measurements. The light a series of small dots over the of the light source should be sufficiently source should satisfy the requirements application area of the substrate and high that signals measured at all UV for solar simulators described in then spread evenly using a gloved wavelengths (290 to 400 nanometers) § 352.70(b). finger. A very light spreading action for through a highly absorbing sunscreen (b) Substrate. Use optical-grade quartz a short period of time (approximately 10 product are above the noise level of the plate suitable for substrate seconds) should be used when measurement system. Spectral spectrophotometry that has been distributing the product to ensure irradiance will be measured at 5 roughened on one side. complete coverage without excessive nanometer intervals, from 290 to 400 (c) Sample holder. The sample holder buildup of product in the troughs of the nanometers. At least 12 measurements should hold the substrate in a horizontal substrate. of spectral irradiance transmitted position to avoid flowing of the (f) Pre-irradiation to account for through the substrate without sunscreen sunscreen drug product from one edge differences in photostability. To account drug product present will be obtained of the substrate to the other. It should for potentially varying degrees of from different locations on the substrate be mounted as close as possible to the photostability between sunscreen drug surface (C(λ)1, C(λ)2, C(λ)3, . . . C(λ)12). input optics of the spectroradiometer to products, irradiate the sunscreen In addition, a minimum of 12 maximize capture of forward scattered product on the substrate with a dose of measurements of spectral irradiance radiation. The sample holder should be UV radiation equal to the SPF of the transmitted through the substrate with a thin, flat plate with a suitable aperture sunscreen product multiplied by 200 J/ the sunscreen drug product present will through which UV radiation can pass. m2-eff multiplied by 2/3. A UV be similarly obtained after pre- The substrate will be placed on the radiation dose of 200 J/m2-eff is irradiation of the sunscreen drug upper surface of the sample holder. equivalent to one minimal erythema product (P(λ)1, P(λ)2, P(λ)3, . . . P(λ)12). (d) Spectroradiometer input optics. dose (MED). The UV dose to be The mean transmittance for wavelength Unless the spectroradiometer is delivered is determined by multiplying λ, T(λ), is the ratio of the mean of the equipped with an integrating sphere, an the light source spectral irradiance C(λ) values to the mean of the P(λ) ultraviolet radiation diffuser should be action spectrum for erythema in ‘‘CIE S values, as follows: placed between the sample and the 007/E Erythemal Reference Action input optics of the spectroradiometer. Spectrum and Standard Erythema The diffuser will be constructed from Dose,’’ at each wavelength, integrating any UV radiation transparent material over wavelength, and multiplying the (e.g., Teflon or quartz). The diffuser integral by the exposure time. ‘‘CIE S The standard deviation, s, associated ensures that the radiation received by 007/E Erythemal Reference Action with the spectral transmittance is the spectroradiometer is not collimated. Spectrum and Standard Erythema evaluated using Taylor’s approximation, The spectroradiometer input slits Dose,’’ dated 1998, is incorporated by as follows:

where C(λ) = mean of the measurements 290 to 400 nanometers. The standard of C at wavelength λ. deviation values will provide an indication of the uniformity of P(λ) = mean of the measurements of P sunscreen drug product spreading at wavelength λ. s(P(λ)) is calculated as follows: during application to the substrate. The s(C(λ)) = standard deviation of the coefficient of variation, which is the measurements of C at wavelength λ. standard deviation divided by the mean, and expressed as a percentage, should s(P(λ)) = standard deviation of the be less than 10 percent. measurements of P at wavelength λ. λ This calculation gives 23 spectral (h) Calculation of the UVA I/UV ratio. s(C( )) is calculated as follows: transmittance values with associated (1) Spectral transmittance values, T(λ), standard deviations, one for each 5 are converted into absorbance values, nanometer wavelength increment from A(λ), by taking the negative logarithm of

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00051 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.010 EP27AU07.011 EP27AU07.012 EP27AU07.013 49120 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

the spectral transmittance value as (i) Category determination of the UVA V: Rarely burns; tans profusely (dark follows: I/UV ratio. Perform at least 5 separate brown) (insensitive). A(λ) = -log T(λ) determinations of the UVA I/UV ratio, VI: Never burns; deeply pigmented The calculation yields 23 from which the mean can be calculated. (insensitive). monochromatic absorbance values in 5 Using the mean, the sunscreen drug Determine that the subject is not taking nanometer increments from 290 to 400 product is classified by in vitro UVA I/ topical or systemic medication that is nanometers. UV ratio as follows: known to alter responses to ultraviolet radiation and that the subject has no (2) The index of UVA I protection is UVA I/UV Ratio Category calculated as the area (per unit history of sensitivities to topical wavelength) under the UVA I portions 0.20 to 0.39 Low products and/or abnormal responses to of a plot of wavelength versus A(λ), 0.40 to 0.69 Medium sunlight, such as a phototoxic or divided by the area (per unit 0.70 to 0.95 High photoallergic response. wavelength) under the total curve, as greater than 0.95 Highest (2) Physical examination. The follows: physical examination shall be § 352.72 UVA in vivo testing procedure. conducted as specified in § 352.70(c)(1). (a) Standard sunscreen. A standard (3) Informed consent. Obtain legally sunscreen shall be tested concomitantly effective written informed consent from in the procedure for determining the all subjects. UVA protection factor (UVA–PF) value (4) Test site delineation—(i) Test site. λ UVA I area per unit is given as: by means of persistent pigment A test site is the location on the back for darkening to ensure the uniform determining the subject’s initial and evaluation of sunscreen drug products. final minimal pigmenting dose (MPD) UV area per unit λ is given as: The standard sunscreen shall be a for unprotected skin and for preparation containing 7 percent determining UVA–PF values after padimate O and 3 percent oxybenzone application of the sunscreen standard as specified in § 352.70(a)(3). For the and the test sunscreen product(s). There where: A(λ) = effective absorbance given test to be valid, the measured mean typically are 4 to 6 test sites for each as -log T(λ) UVA–PF value of the standard subject. Test sites should be located on d(λ) = wavelength interval between preparation shall be 3.2 with a standard the back between the beltline and the measurements deviation less than or equal to 0.5. shoulder blades (scapulae) and lateral to B(λ) = any biological action spectrum (b) Light source. The light source used the midline. Each test site shall be a factor for determining the UVA–PF value of a minimum of 50 square centimeters sunscreen drug product shall provide a (cm2) (i.e., 5 x 10 centimeters). Outline Because no appropriate biological action continuous emission spectrum in the the test sites to which the sunscreen spectrum for UVA radiation damage has range of 320 to 400 nanometers. The standard and the test sunscreen been universally accepted, no action λ ratio of UVA I (340 to 400 nanometers) product(s) will be applied with spectrum is specified. The value of B( ) to UVA II (320 to 340 nanometers) in the indelible ink. If the subject is to receive is, therefore, equal to 1.0 for all final beam shall be close to that of the doses of ultraviolet radiation in an wavelengths. sunlight, i.e., emitted UVA II shall be 8 upright (seated) position, draw the lines (3) The integrals in the formulae in to 20 percent of the total UVA radiation. on the skin with the subject upright paragraphs (h)(1) and (h)(2) of this Optical radiation from 250 to 320 (seated). If the subject is to receive the section are evaluated using Simpson’s nanometers shall be less than 0.1 doses of ultraviolet radiation while Rule for irregular areas, which states: percent of the optical radiation between prone, draw the lines with the subject Area = h/3 x [Y0 + Y2m + 4(Y1 + Y3 . . . + prone. Y ) + 2(Y + Y + . . . Y )] 320 to 400 nanometers. Exclude visible 2m-1 2 4 2m-2 and infrared light to avoid the darkening (ii) Test subsite. Test subsites are the In this equation, Y0, Y1, Y2, . . . Y2m are effects of visible light and the effect of locations to which ultraviolet radiation the lengths of 2m parallel lines drawn heat. Perform monitoring and is administered within a test site. At vertically to divide the area under the maintenance of the light source as least 5 test subsites will receive UV curve of a graph into 2m-1 segments of specified in § 352.70(b)(3). doses within each test site. Test subsites equal width, h. In practice, the values of (c) General testing procedures—(1) will be at least 1 cm2 in area and will λ Y0, Y1, Y2, . . . Y2m are the A( ) values Medical history. Obtain a medical be separated from each other by at least determined and h is the wavelength history from each subject with emphasis 1 cm. Mark the location of each test interval at which the spectral on the effects of sunlight on his/her subsite with indelible ink. transmittance is determined (i.e., 5 skin. Determine that each subject is in (5) Application of test materials. nanometers). good general health and has skin type II Apply the test sunscreen product and (4) UVA I area per unit wavelength or III (as described in this paragraph). the standard sunscreen as specified in λ (aUVA I/ ) is calculated as follows: Skin Type and Sunburn and Tanning § 352.70(c)(5). aUVA I/λ = 5/3 x [A + A + 4(A + ... 340 400 345 History (Based on first 30 to 45 minutes (6) Waiting period. Before exposing + A395) + 2(A350 + A360 + A370 + ... + A390)]/ the test site areas after applying a 60 of sun exposure after a winter season of no sun exposure). product, wait at least 15 minutes. UV area per unit wavelength (aUV/λ) is (7) Number of subjects. A test panel calculated as follows: I: Always burns easily; never tans shall consist of 20 to 25 subjects with λ (sensitive). aUV/ = 5/3 x [A290 + A400 + 4(A295 + A305 at least 20 subject who produce valid + A315 + ... + A395) + 2(A300 + A310 + ... + II: Always burns easily; tans minimally data for analysis. Data is valid unless A390)]/110 (sensitive). rejected in accordance with UVA I/UV ratio is calculated as follows: III: Burns moderately; tans gradually § 352.70(c)(9). If more than 5 subjects (light brown) (normal). are rejected based on § 352.70(c)(9), the IV: Burns minimally; always tans well panel is disqualified, and a new panel (moderate brown) (normal). must be created.

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00052 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.014 EP27AU07.015 EP27AU07.016 EP27AU07.017 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules 49121

(8) Response criteria. In order that the determine the initial unprotected MPD. follows: If 12 + A ≤ x, the PCD is person who evaluates the MPD Select the doses that are a geometric Highest; if 8 + A ≤ x < 12 + A, the PCD responses is not biased, he/she must not series represented by (1.25n), wherein is High; if 4 + A < x < 8 + A, the PCD be the same person who applied the each exposure dose is 25 percent greater is Medium; if 2 + A ≤ x < 4 + A, the sunscreen drug product to the test site than the previous exposure dose to PCD is Low; if x < 2 + A, the product or administered the doses of UV maintain the same relative uncertainty shall not display a UVA–PF value. radiation. After UV radiation exposure (expressed as a constant percentage), from the solar simulator is completed, § 352.73 Determination of the labeled UVA independent of the subject’s sensitivity protection value. all immediate responses shall be to UV radiation. Usually, the UV Test the sunscreen product in recorded. These may include an radiation for determining the initial accordance with §§ 352.71 and 352.72. immediate darkening or tanning, unprotected MPD is administered the The UVA category on the principal typically grayish or purplish in color, day prior to applying the sunscreen display panel (PDP) of the tested which fades in 30 to 60 minutes; an product and standard sunscreen, and sunscreen product, as specified in immediate reddening at the subsite, due the responses are then evaluated § 352.50, shall be the lower of either the to heating of the skin, which fades immediately prior to applying the UVA I/UV ratio category determined in rapidly; and an immediate generalized sunscreen product and sunscreen § 352.71(j) or the UVA–PF product heat response, spreading beyond the standard. Determine the final category designation (PCD) determined subsite, which fades in 30 to 60 unprotected MPD on the same day that in § 352.72(d)(3). If the product does not minutes. After the immediate responses UV radiation is administered to the are noted, each subject shall shield the attain at least a ‘‘low’’ category rating for sunscreen-protected test sites. Use the both the UVA–PF and the UVA I/UV exposed area from further UV radiation final unprotected MPD (MPD(US)) in until the MPD response is evaluated. ratio, the product shall not display a calculating UVA–PF. UVA claim. State the final combined Determine the MPD 3 to 24 hours after (2) Determination of individual UVA– exposure. Evaluate the pigmentation category rating (i.e., the lower of either PF values. Administer a series of five the UVA I/UV ratio or UVA–PF PCD responses of each test site using either UV radiation doses expressed as Joules tungsten or warm white fluorescent categories) on the PDP of the product per square meter to the subsites within along with the corresponding number of lighting that provides 450 to 550 lux of each test site on a subject using the light illumination at the test site. For the stars for that combined category rating source described in paragraph (b) of this as follows: evaluation, the test subject should be in section. The five UV doses will be a the same position used when the test geometric series as described in Combined Category Rating Star Rating site was irradiated. For each test site, paragraph (d)(1) of this section, where determine the smallest UV dose that the middle exposure represents the Low ✰✰✰ produced mild pigmentation reaching Medium ✰✰ expected UVA–PF. Use exposures that ✰ the borders of the test subsite. The MPD are the product of the initial High Highest is the smallest UV dose required to unprotected MPD times 0.64X, 0.80X, produce the first perceptible pigment 1.00X, 1.25X, and 1.56X, where X 11. Section 352.76 is amended by darkening at 3 to 24 hours post- equals the expected UVA–PF of the test revising the introductory paragraph and exposure. To determine the MPD, there product. The MPD is the smallest UV paragraphs (a) introductory text, (a)(6), must be at least one subsite that dose required to produce pigmentation (b) introductory text, and (b)(10) to read received a smaller UV dose and does not at 3 to 24 hours post-exposure. Calculate as follows: produce pigmentation as well as a the UVA–PF value of each sunscreen subsite(s) with somewhat more intense § 352.76 Determination if a product is product and sunscreen standard using water resistant or very water resistant. pigmentation. For subsites showing MPD of sunscreen-protected skin pigmentation, the maximal exposure The general testing procedures in (MPD(PS)) and the final unprotected § 352.70(c) shall be used as part of the should be no more than twice the total MPD (MPD(US)) as follows: energy of the minimal exposure. following tests, except where modified (9) Rejection of test data. Reject test in this section. An indoor fresh water data if the exposure series fails to elicit pool, whirlpool, and/or jacuzzi an MPD response on either the treated maintained at 23 to 32 °C shall be used or unprotected skin sites, or all subsites (3) Determination of test product in these testing procedures. Fresh water within a test site show more intense UVA–PF and UVA product category is clean drinking water that meets the responses than the threshold designation (PCD). Use data from at standards in 40 CFR part 141. The pool pigmentation response, or the responses least 20 test subjects with n representing and air temperature and the relative are inconsistent with the series of UV the number of subjects used. First, humidity shall be recorded. doses administered, or the subject was compute the UVA–PF value for each (a) Procedure for testing the water noncompliant, e.g., the subject subject as stated in paragraph (d)(2) of resistance of a sunscreen product. For withdraws from the test due to illness this section. Second, compute the mean sunscreen products making the claim of or work conflicts or does not shield the UVA–PF value, x, and the standard ‘‘water resistant,’’ the label SPF and, if exposed testing sites from further UV deviation, s, for these subjects. Third, appropriate, UVA values shall be the radiation until the MPD is read. obtain the upper 5-percent point from label SPF and UVA values determined (d) Determination of UVA–PF Student’s t distribution table with n-1 after 40 minutes of water immersion values—(1) Determination of MPD of degrees of freedom. Denote this value by using the following procedure for the unprotected skin. Administer a series of t. Fourth, compute ts/√n. Denote this water resistance test: five UV radiation doses expressed as quantity by A (i.e., A = ts/√n). Fifth, * * * * * Joules per square meter to the subsites calculate the UVA–PF value to be used (6) Begin light source exposure to test within each test site on a subject using in labeling as follows: The label UVA– site areas as described in § 352.70(b) the light source described in paragraph PF equals the largest whole number less and, if appropriate, § 352.72(b). (b) of this section. Use the series of five than x - A. Sixth and last, the drug (b) Procedure for testing a very water exposures to the unprotected test site to product is classified into a PCD as resistant sunscreen product. For

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00053 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3 EP27AU07.018 49122 Federal Register / Vol. 72, No. 165 / Monday, August 27, 2007 / Proposed Rules

sunscreen products making the claim of immersion using the following Dated: August 10, 2007. ‘‘very water resistant,’’ the label SPF procedure for the water resistance test: Jeffrey Shuren, and, if appropriate, UVA values shall be * * * * * Assistant Commissioner for Policy. the label SPF and UVA values (10) Begin light source exposure to [FR Doc. 07–4131 Filed 8–23–07; 8:45 am] determined after 80 minutes of water test site areas as described in § 352.70(b) BILLING CODE 4160–01–S and, if appropriate, § 352.72(b).

VerDate Aug<31>2005 15:51 Aug 24, 2007 Jkt 211001 PO 00000 Frm 00054 Fmt 4701 Sfmt 4702 E:\FR\FM\27AUP3.SGM 27AUP3 rmajette on PROD1PC64 with PROPOSALS3