Vol. 76 Friday, No. 117 June 17, 2011

Part IV

Department of Health and Human Services

Food and Drug Administration 21 CFR Parts 201, 310, and 352 Sunscreen Drug Products for Over-the-Counter Human Use; Final Rules and Proposed Rules

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35620 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

DEPARTMENT OF HEALTH AND FOR FURTHER INFORMATION CONTACT: regulatory actions associated with OTC HUMAN SERVICES Reynold Tan, Center for Drug sunscreen products, please refer to our Evaluation and Research, Food and Web site: http://www.fda.gov/Drugs/ Food and Drug Administration Drug Administration, 10903 New DevelopmentApprovalProcess/ Hampshire Ave., Bldg. 22, rm. 5411, DevelopmentResources/Over-the- 21 CFR Parts 201 and 310 Silver Spring, MD 20993, 301–796– CounterOTCDrugs/StatusofOTC [Docket No. FDA–1978–N–0018] (Formerly 2090. Rulemakings/ucm072134.htm. In the Federal Register of May 12, Docket No. 1978N–0038) SUPPLEMENTARY INFORMATION: 1993 (58 FR 28194), we published a RIN 0910–AF43 Table of Contents proposed rule for OTC sunscreen products that identified active Labeling and Effectiveness Testing; I. Overview of Document A. Rulemaking History ingredients we tentatively considered to Sunscreen Drug Products for Over-the- be generally recognized as safe and Counter Human Use B. Scope of This Document C. Issues Outside the Scope of This effective (GRASE), as well as associated AGENCY: Food and Drug Administration, Document labeling and sun protection factor (SPF) HHS. D. Enforcement Policy testing to be required for these OTC E. Summary of Major Revisions to the sunscreen products (the 1993 proposed ACTION: Final rule. Labeling and Testing Requirements rule). The SPF test and corresponding Included in the 2007 Proposed Rule labeling reflect the level of protection SUMMARY: The Food and Drug II. Administrative and Other Issues Administration (FDA) is issuing this III. Principal Display Panel (PDP) Labeling against sunburn, which is caused document to address labeling and A. SPF Statement primarily by UVB radiation. The 1993 effectiveness testing for certain over-the B. Broad Spectrum Statement proposed rule also explained the counter (OTC) sunscreen products C. Water Resistance Statement importance of protection against UVA containing specified active ingredients D. UVB and UVA Educational Statement radiation (58 FR 28194 at 28232 and and marketed without approved IV. Drug Facts Labeling 28233). The proposed rule referenced A. Active Ingredients/Purpose applications. This document addresses published UVA test methods but did not B. Uses propose a specific method (58 FR 28194 labeling and effectiveness testing issues C. Warnings raised by the nearly 2,900 submissions at 28248 to 28250). Rather, the proposed D. Directions rule stated that a sunscreen product that we received in response to the E. Constitutionality of Labeling Statements sunscreen proposed rule of August 27, Regarding Skin Cancer and Skin Aging could be labeled as ‘‘broad spectrum,’’ 2007 (2007 proposed rule). The F. Other Information or labeled with a similar statement, if it document also identifies specific claims G. Reduced Labeling protected against UVA radiation as that render a product that is subject to V. Miscellaneous Labeling Outside Drug demonstrated by one of the published this rule misbranded or would not be Facts UVA tests or a similar test. VI. SPF Test Parameters In April 1994, we reopened the allowed on any OTC sunscreen product A. Solar Simulator administrative record to allow marketed without an approved B. Sunscreen Standards additional submissions concerning application. The document does not C. Test Subjects UVA-related issues. We also announced address issues related to sunscreen D. Test Sites and Subsites a public meeting to be held in May 1994 active ingredients or certain other issues E. Finger Cot to discuss UVA testing procedures (59 regarding the GRASE determination for F. Application Amount FR 16042, April 5, 1994). We held the sunscreen products. The document G. Water Resistance VII. SPF Test Issues (Other Than Test public meeting to gather more requires OTC sunscreen products to information to help us determine the comply with the content and format Parameters) A. Pass/Fail (Binomial) SPF Test most appropriate UVA test method and requirements for OTC drug labeling B. Photostability labeling. contained in the 1999 Drug Facts final C. In Vitro SPF Test In November 1997, Congress enacted rule (published in the Federal Register D. Anti-Inflammatory Ingredients the Food and Drug Administration of March 17, 1999, by lifting the delay VIII. Broad Spectrum Test Modernization Act of 1997 (FDAMA), of implementation date for that rule that A. In Vivo Test Method: Not Required which addressed OTC sunscreen we published on September 3, 2004). B. In Vitro Test Method: Critical products among other FDA issues. Wavelength DATES: Effective Date: This final rule is C. Critical Wavelength Test Parameters Section 129 of FDAMA stated that ‘‘not effective June 18, 2012. For additional IX. Analysis of Impacts later than 18 months after the date of information concerning this effective A. Final Regulatory Impact Analysis enactment of this Act, the Secretary of date, see section X in the preamble of B. Small Business Impact (Final Regulatory Health and Human Services shall issue this document. The incorporation by Flexibility Analysis) regulations for over-the-counter reference of a certain publication listed X. Paperwork Reduction Act of 1995 sunscreen products for the prevention in this rule is approved by the Director XI. Environmental Impact or treatment of sunburn.’’ We then of the Federal Register as of June 18, XII. Federalism determined that the GRASE active 2012. XIII. References ingredients, SPF testing requirements, Compliance Date: The compliance I. Overview of Document and related labeling were issues that we date for all products subject to this final could finalize within the timeframe set rule with annual sales less than $25,000 A. Rulemaking History by FDAMA. Because we had not is June 17, 2013. The compliance date This section of the document does not previously proposed specific UVA for all other products subject to this discuss every regulatory action testing and labeling requirements, we final rule is June 18, 2012. associated with OTC sunscreen did not have sufficient time to finalize Implementation date: FDA is lifting products. It highlights the major these UVA requirements within the the delay of implementation date for regulatory actions that are related to the FDAMA timeframe. § 201.66 as published at 69 FR 53801, regulatory actions being taken in this In the Federal Register of May 21, September 3, 2004. document. For a complete list of all 1999, we published a final rule for OTC

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35621

sunscreen products (64 FR 27666). The sunscreen products are appropriately products, this new section 21 CFR 1999 sunscreen final rule added the labeled and tested for both UVA and 201.327 includes requirements for sunscreen monograph (regulations) in UVB protection. In addition, the labeling and the effectiveness testing part 352 (21 CFR part 352) and included requirements in this final rule will help upon which it relies. Because we have an effective date of May 2001. The 1999 ensure the proper use of these not yet resolved all of the issues sunscreen final rule stated that we sunscreens and greater consumer regarding conditions under which would publish a proposed rule outlining protection from the damaging effects of sunscreens are GRASE and not UVA testing and labeling requirements UV radiation. This final rule also misbranded, the stay of 21 CFR part 352 at a future date. In 2000, we extended identifies claims that render a product remains in effect. Although we are not the effective date for the 1999 sunscreen that is subject to this rule misbranded or yet codifying these labeling and related final rule to December 2002 (65 FR are not allowed on any OTC sunscreen effectiveness testing provisions in the 36319, June 8, 2000). drug product marketed without an monograph regulation, they do embody In December 2001, we stayed the approved application. the agency’s current determination on December 2002 effective date of the As described in the previous section appropriate regulation of these aspects 1999 sunscreen final rule indefinitely. of this document, we issued the 2007 of sunscreens that were previously We took this action because we planned proposed rule as a proposed amendment identified as falling within the to revise part 352 to add UVA testing to the sunscreen monograph monograph in part 352, and supersede and labeling requirements so that OTC requirements in 21 CFR part 352 the prior approach embodied in the sunscreen products would be tested and primarily to establish UVA testing and never-effective provisions of 21 CFR labeled for both UVB and UVA radiation labeling requirements so that all OTC part 352 subparts C and D. While this protection. We included these revisions sunscreen products marketed under the rule does not lift the stay of part 352, we in a proposed rule that published in the sunscreen monograph would be tested are lifting the delay of implementation Federal Register of August 27, 2007 (72 and labeled for both UVB and UVA date for the Drug Facts labeling FR 49070). The 2007 proposed rule radiation protection. Sunscreen active requirements of 21 CFR 201.66. In identified UVA testing and labeling that ingredients, UVB testing, UVB labeling, we proposed should be required for all and other conditions under which addition, this rule codifies certain OTC sunscreen products. The proposed sunscreens would be considered GRASE specific claims that render a covered rule also revised SPF testing and and not misbranded had been addressed product misbranded or are not allowed corresponding labeling from the 1999 in the 1999 (stayed) final rule. In on any OTC sunscreen drug product final rule. The proposed rule did not lift response to the 2007 proposed rule, marketed in the United States without the existing stay of the effective date for however, we received submissions from an approved application. part 352. the public concerning all aspects of the We note that all provisions of new 21 On September 3, 2004 (69 FR 53801), sunscreen monograph (i.e., the CFR 201.327 and the amendments to we delayed until further notice the conditions specified in the 1999 final 310.545 included in this rule apply only implementation date for the Drug Facts rule and the 2007 proposed rule). As to the aforementioned covered products, final rule (64 FR 13254, March 17, 1999) discussed further in this section, some and references in this document to (21 CFR 201.66) for OTC sunscreen of the issues regarding the monograph ‘‘covered’’ products recognize this products. The Drug Facts final rule (21 conditions raised in the public limitation. Manufacturers of sunscreen CFR 201.66) establishes general labeling submissions will require further products that are already being format and content requirements for all evaluation by us. Therefore, we are not marketed pursuant to an approved OTC drugs. We explained that we issuing a final monograph with GRASE application can contact FDA’s Center for postponed the implementation date for conditions for sunscreens in this Drug Evaluation and Research to discuss general Drug Facts labeling document. Instead, we are publishing supplemental submissions that would requirements for sunscreens because we this final rule establishing labeling and enable them to include labeling on their did not expect to issue the sunscreen the effectiveness testing upon which it products like that specified in this final final rule containing UVA testing and relies, which applies to the same rule. product-specific labeling requirements sunscreens that were the subject of the (i.e., this document) by the Drug Facts 2007 proposed rule to amend the C. Issues Outside the Scope of This implementation date of May 2005. monograph, because it is in the best Document Therefore, we delayed the interest of public health to publish this implementation date until further notice final rule while we work on remaining There are a number of issues that to prevent sunscreen product issues that need to be addressed in order were raised in public submissions manufacturers from having to relabel to publish a final monograph. This responding to the 2007 proposed rule their products at two closely related labeling will help ensure that these that are outside the scope of this time intervals, as initially required by products are not misbranded by document. The issues fall into two the 1999 Drug Facts final rule and the providing specific indications, categories: 1999 sunscreen final rule. directions, warnings, and other • GRASE determination for sunscreen important information to help products and active ingredients B. Scope of This Document consumers select and use them • Issues affecting multiple OTC drug This final rule establishes the labeling appropriately. monographs and testing requirements for OTC In this final rule, then, we are sunscreen products containing specific codifying in 21 CFR part 201 As explained below, in this document, ingredients or combinations of requirements for OTC sunscreen we are not addressing these issues ingredients and marketed without an products containing specified active related to determining the GRASE status approved application under section 505 ingredients and marketed without of sunscreen products or sunscreen of the Federal Food, Drug, and Cosmetic approved applications under section active ingredients and are not Act (21 U.S.C. 355) (the FD&C Act). The 505 of the FD&C Act (21 U.S.C. 355) addressing the issues described below requirements in this final rule will help (hereafter referred to as ‘‘covered’’ affecting multiple OTC drug ensure that these currently marketed products). With respect to these covered monographs.

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35622 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

1. Issues Regarding GRASE reviewing the safety and effectiveness enforcement policy for OTC sunscreen Determination for Sunscreen Products data submitted for these and other products marketed without approved and Active Ingredients sunscreen active ingredients found applications. To clarify expectations for A large number of submissions on the eligible for potential addition to the industry, elsewhere in this issue of the 2007 proposed rule raised issues related monograph. When we complete our Federal Register, we are announcing the to the conditions that define what review, we will issue proposed rules availability of a draft guidance constitutes a GRASE finished OTC stating our tentative conclusions on the document, explaining the agency’s sunscreen product, irrespective of its safety and effectiveness of all of these intended enforcement policy for these active ingredients. These included over ingredients. products until a final sunscreen 1000 submissions requesting that we 2. Issues Affecting Multiple OTC Drug monograph becomes effective. limit the monograph to sunscreens that Monographs E. Summary of Major Revisions to the offer broad spectrum protection and This final rule also does not address Labeling and Effectiveness Testing have SPF values of 15 or higher. three issues raised in response to the Included in the 2007 Proposed Rule Because this final rule is a labeling rule, 2007 sunscreen proposed rule that are In response to the 2007 proposed rule, and not a monograph, we do not address not specific to sunscreen products. we received almost 2,900 submissions these issues here but plan to address Because these issues apply more from the public. Of these submissions, them in future rulemakings regarding generally to multiple categories of OTC over 2,500 expressed general support for the monograph and conditions for drug products, we are not addressing the proposed rule and urged us to general recognition of safety and these issues in this final rule, which is finalize and implement the new rule effectiveness. limited to OTC sunscreen products. This rule also does not address issues quickly. Three hundred twenty-five of The first issue concerns the inclusion the submissions raised approximately related to the GRASE status of of expiration dates on sunscreen labels. sunscreen active ingredients that are 90 specific issues related to the We received 12 submissions requesting proposed rule. We have addressed the included in the 2007 proposed rule that we require OTC sunscreen products (proposed 21 CFR 352.10 and 352.20). issues specifically relating to labeling to be labeled with an expiration date and effectiveness testing in this final We received 20 submissions raising (Ref. 1). Currently, regulations in 21 questions about the safety of ingredients rule. Based on the submissions received, CFR 211.137(h) do not require that an and the information and data included in sunscreens (Ref. 1). Ten of the expiration date be included in labeling submissions specifically asked that we in those submissions or otherwise if an OTC drug product does not have available to us, we have re-evaluated ensure that none of the ingredients are any dosage limitations and is stable for carcinogenic. Others asked that we our position on several issues in the at least 3 years. This regulation applies 2007 proposed rule and made several ensure that all ingredients in sunscreens to many OTC drug products, including are safe without citing a specific changes to our proposed labeling and sunscreen products. Any modification testing requirements. Tables 1, 2, 4, and concern. We intend to address of the existing regulations would require carcinogenicity and other safety 5 in this document summarize the publication of a proposed rule labeling and effectiveness testing considerations related to sunscreen addressing all OTC drug products active ingredients in a future requirements included in the 2007 affected by the expiration date proposed rule as well as the labeling rulemaking. regulations. We also received submissions and effectiveness testing required by The second issue concerns the term this final rule: requesting that we increase the GRASE ‘‘final monograph.’’ One submission • concentration of avobenzone from 3 Table 1: PDP Labeling (discussed in argued that we should not use this term section III) percent to 5 percent (Ref. 1). Another because it is inaccurate (Ref. 1). As the • submission points out that there are two Table 2: Drug Facts Labeling submission states, ‘‘FDA is to (discussed in section IV) USP 1 monographs for zinc oxide: continually evaluate products, so • • Table 4: SPF Test (discussed in Zinc oxide (Ref. 2) nothing is ever finalized.’’ This issue • Zinc oxide neutral (Ref. 3) section VI) applies to monographs representing all • Table 5: Broad Spectrum Test The submission would like us to clarify categories of OTC drug products. (discussed in section VIII) that zinc oxide in OTC sunscreen Therefore, we are not addressing the products can meet the specifications of issue in this document. Rather than summarizing all of the either USP monograph (Ref. 1). We The third issue concerns the country revisions to the labeling and testing intend to address all of these issues of origin listing for all ingredients (i.e., included in the 2007 proposed rule, we regarding GRASE determination for both active and inactive ingredients) on are highlighting what we consider to be sunscreen active ingredients in future a sunscreen drug product. We received the most important revisions in this rulemakings. a submission requesting that we provide section of the document. In addition, we received two the country of origin for each ingredient. We made the following changes to the submissions requesting that we classify The submission also requested that proposed labeling: three new ingredients not previously manufacturers be required to provide 1. The proposed UVA ‘‘star rating’’ is marketed in the United States as specific details about what each not required on the PDP. GRASE: bemotrizinol, bisoctrizole, and ingredient does in the product. This 2. A combined ‘‘Broad Spectrum SPF’’ octyl triazone (Ref. 1). We found these issue applies to all OTC drug products statement is required on the PDP for active ingredients eligible for review and, therefore, we are not addressing it sunscreen products that pass the broad under the OTC drug monograph system in this document. spectrum test established in new 21 CFR in 2003 (octyl triazone) and 2005 201.327(j). To pass the broad spectrum (bemotrizinol and bisoctrizole) (68 FR D. Enforcement Policy test, the amount of UVA protection must 41386, July 11, 2003, and 70 FR 72449, As noted, no final monograph is increase as the SPF value increases. December 5, 2005). We are currently currently in effect for OTC sunscreen 3. For sunscreen products that pass drug products, and in its absence, the broad spectrum test established in 1 United States Pharmacopeia. questions may arise regarding FDA’s new 21 CFR 201.327(j) and have SPF

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35623

values of 15 or higher in accordance • Revise the preemption language We are requiring that this final rule with the SPF test in 21 CFR 201.327(i): included in the 2007 proposed rule by become effective in 1 year, even though a. The ‘‘Sun Alert’’ warning proposed deleting any references regarding the we considered 18 months in the 2007 as the first warning in 2007 is not rule’s potential preemption of State tort proposed rule (72 FR 49070 at 49110). required (Warning proposed located in law. We are allowing products with annual 21 CFR 352.52(c)(1)). Our positions on these issues are sales less than $25,000 to comply with b. A new indication statement may be discussed in the remainder of this this rule in 2 years, as stated in the 2007 included to inform consumers that section of the document. proposed rule. In response to the using the product ‘‘as directed with All of the requests to extend the proposed rule, we received one other sun protection measures (see comment period were submitted before submission arguing that we should Directions [in bold italic font]) decreases the November 28, 2007 Federal Register require this final rule to become the risk of skin cancer and early skin notice in which we extended the effective in 1 year (Ref. 1). The aging caused by the sun.’’ comment period of the 2007 proposed submission stated that a later effective c. A new direction statement has been rule (72 FR 67264). In that notice, we date would have a negative public added informing consumers that extended the close of the comment health impact. We received eight exposure to the sun increases the risk of period from November 26, 2007, to submissions arguing that we should skin cancer and early skin aging and December 26, 2007. We have not extend the effective date from the providing a list of specific sun received any more requests to extend proposed 18 months to 3 years (Ref. 1). protection measures that can decrease the comment period since December The submissions listed the following this risk. 2007. reasons for allowing more than 18 4. For any OTC sunscreen product With regard to requests to lift the stay months: • that does not pass the broad spectrum of 21 CFR part 352 (the OTC sunscreen Repackaging • test in 21 CFR 201.327(j), or that are monograph), as already discussed, our Relabeling • broad spectrum with an SPF value less 2007 proposed rule anticipated Testing/retesting than 15, this final rule, like the 2007 • Removing products from market amending the testing and labeling • proposed rule, requires that the first provisions of that monograph and Impact on small businesses warning indicate the adverse subsequently lifting the stay. However, The most common argument was that consequences of spending time in the comments received on the 2007 more time would be needed to test/ sun. The wording of this warning has proposed rule not only addressed retest OTC sunscreen products for broad been revised to state, ‘‘Skin Cancer/Skin labeling and effectiveness testing for spectrum protection in accordance with Aging Alert [in bold font]: Spending final sunscreen formulations, but also both the in vitro and in vivo UVA test time in the sun increases your risk of raised other issues about the monograph methods included in the proposed rule. skin cancer and early skin aging. This conditions for OTC sunscreen products We agree with the submission which product has been shown only to help that require further consideration. As a stated that it would be beneficial for prevent sunburn, not [in bold font] skin result, we are not finalizing consumers to have this rule become cancer or early skin aging.’’ amendments to part 352 at this time nor effective within 1 year. As explained in We also made the following changes lifting the stay placed on that section as section VIII.A of this document, we are to the effectiveness testing proposed in enacted in 1999 (66 FR 67485). Rather, not requiring manufacturers to 2007: this final rule establishes in 21 CFR demonstrate broad spectrum protection 1. The number of subjects required in 201.327 labeling requirements and the by conducting in vivo and in vitro tests. the SPF test has been reduced from 20 effectiveness testing upon which it This final rule requires that subjects to 10 subjects. relies for covered OTC sunscreen drug 2. One in vitro test is required to manufacturers conduct only the simpler products. We intend to lift the stay on demonstrate broad spectrum protection and less expensive nonclinical in vitro part 352 when we reach our final rather than the two previously proposed test to demonstrate broad spectrum conclusions on the conditions under tests (an in vitro test and an in vivo test). protection. In vitro tests are 3. The broad spectrum test is a pass/ which sunscreen products are GRASE substantially shorter than in vivo tests. fail test based on the critical wavelength and not misbranded, including a Therefore, we are setting an effective value of 370 nm 2. determination regarding sunscreen date for this rule 1 year from the date active ingredients, and publish a revised of publication in the Federal Register. II. Administrative and Other Issues final monograph. In the interim, the However, we are providing two years for Some of the submissions that we labeling and effectiveness testing all products with annual sales less than received following publication of the provisions of this rule apply to covered $25,000 to comply with this rule. In 2007 proposed rule made the following OTC sunscreen products. addition, in order to ensure that limited requests involving administrative issues We received a request that we allow testing laboratory capacity does not (Ref. 1): interim marketing of avobenzone result in sunscreen shortages during the • Extend the comment period of the combinations in proposed § 352.20(a)(2) transition to the new rule, we intend to 2007 proposed rule. prior to issuing a final rule for part 352. exercise enforcement discretion for a • Lift the stay on 21 CFR part 352, Subject to our enforcement discretion, period of time with regard to the SPF imposed in 2001 (66 FR 67485). we will continue to allow the marketing test for certain OTC sunscreen products • Allow interim marketing of of avobenzone combinations provided on the market by June 17, 2011 (see our products containing avobenzone with for in the 1999 sunscreen final rule. draft guidance entitled ‘‘Guidance for ensulizole and avobenzone with zinc However, we are not allowing marketing Industry: Enforcement Policy—OTC oxide. of the additional avobenzone Sunscreen Drug Products Marketed • Set an effective date for this final combinations discussed in the 2007 Without An Approved Application’’ rule other than the 18 months proposed proposed rule until we reach a final announced elsewhere in this issue of in the 2007 proposed rule. conclusion on the GRASE determination the Federal Register). for sunscreen active ingredients and The submissions stating that 2 Nanometers. combinations of those ingredients. additional time is necessary for

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35624 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

repackaging and relabeling did not (Ref. 1). The submission claimed that The submission also requested that submit any information or data to we exceeded our authority when we we delete any references to implied support these arguments (Ref. 1). The stated that section 751(a) of the FD&C preemption. In this final rule, we have argument that more than 18 months is Act displaces both State legislative omitted any statement regarding needed to remove non-compliant requirements and State common law implied preemption because, although products from the market is not valid. duties. The submission argued that implied preemption may arise, such In the 2007 proposed rule, we indicated Congress intended to preserve State scenarios are necessarily case-specific. that sunscreen products which are common law claims by including Section XII of this document makes already distributed by the effective date section 51(e), which exempts State clear that the sole statutory provision of the final rule would not be expected product liability claims from express giving preemptive effect to the final rule to be relabeled or retested in conformity preemption under section 751(a) of the is section 751 of the FD&C Act. with the final rule conditions unless FD&C Act. The commenter appears to these products were subsequently have construed our statement in a way III. Principal Display Panel (PDP) relabeled or repackaged after the that would nullify section 751(e) of the Labeling effective date (72 FR 49070 at 49109). FD&C Act. We did not intend to suggest Consistent with this statement, we do that section 751(a) of the FD&C Act In response to the 2007 sunscreen not expect non-compliant products preempts State product liability claims, proposed rule, we received 45 introduced or delivered for introduction whether based on State legislative submissions requesting that we revise into interstate commerce prior to the enactments or common law, because the proposed principal display panel compliance dates specified for this final section 751(e) exempts such actions (PDP) labeling (Ref. 1). We are revising rule to be removed from the market. from the express preemption provision the PDP labeling based, in part, on these We received a submission that in section 751(a). However, it is submissions (see table 1 of this expressed concern about the agency’s important to note that section 751(e) of document). We have decided that the preemption discussion in the 2007 the FD&C Act exempts only those PDP labeling included in this document proposed rule (72 FR 49070 at 49109 common law claims that are based on will simplify the purchase decision for and 49110) and requested that we delete State product liability law. Our revised consumers by allowing them to more any discussion regarding the rule’s preemption discussion in section XII easily find important information potential preemption of State tort law remains consistent with applicable law. included on the PDP.

TABLE 1—SUMMARY OF PDP LABELING IN THE 2007 PROPOSED RULE AND THIS FINAL RULE USING A BROAD SPECTRUM SPF 30 WATER RESISTANT SUNSCREEN PRODUCT AS EXAMPLE A AND AN SPF 6 SUNSCREEN THAT IS NOT BROAD SPECTRUM AND NOT WATER RESISTANT AS EXAMPLE B

Labeled information 2007 Proposed rule This final rule

Effectiveness Rating 1 ...... Example A: Example A: ‘‘UVB SPF 30 High’’ ‘‘Broad Spectrum SPF 30’’ ‘‘UVA ★★★✰ High’’ Example B: Example B: ‘‘UVB SPF 6 Low’’ ‘‘SPF 6’’ ‘‘No UVA Protection’’ Water Resistance ...... Example A: Example A: ‘‘Water Resistant’’ ‘‘Water Resistant (40 minutes)’’ Example B: Example B: No statement on water resistance No statement on water resistance Educational Statement ...... Examples A & B: Examples A & B: ‘‘UV rays from the sun are made of UVB and UVA. It is No educational statement important to protect against both UVA and UVB rays.’’ 1 The UVA rating in the 2007 proposed rule is a four-tier rating (low, medium, high, highest). The UVA testing in this final rule is pass/fail—a product is either allowed or not allowed to include a broad spectrum statement depending on results of the test described in new 21 CFR 201.327(j) (see section VIII of this document).

A. SPF Statement We received 12 submissions regarding second requests, but are not granting the the SPF statement in response to the other three requests. In the 2007 sunscreen proposed rule, 2007 sunscreen proposed rule (Ref. 1). In this final rule, unlike the 2007 we proposed redefining the acronym Collectively, the submissions made the proposed rule, we have no express ‘‘SPF’’ as the ‘‘sunburn protection following requests: definitional section. However, we factor.’’ We also proposed that the term 1. Do not change the definition of SPF identify ‘‘SPF’’ as an abbreviation for ‘‘UVB SPF’’ would be required on the to ‘‘sunburn protection factor’’ ‘‘sun protection factor’’ in new 21 CFR PDP of all OTC sunscreen products 2. Remove UVB from ‘‘UVB SPF’’ 201.327(a)(1), and use it consistently in (proposed 21 CFR 352.50(a)). This term 3. Redefine the ‘‘highest’’ product this way throughout the rule. This use would be followed by the numerical category descriptor to include SPF 50 of the term SPF is identical to the value determined from SPF testing and 4. Require SPF values expressed in definition in the 1999 stayed sunscreen one of the following descriptors: ‘‘low,’’ multiples of 5 final rule (64 FR 27666). For products ‘‘medium,’’ ‘‘high,’’ or ‘‘highest.’’ For 5. Label SPF as the percent of UVB that are not broad spectrum, the term example, a sunscreen product could radiation screened ‘‘SPF’’ will appear on the PDP with the have contained the statement ‘‘UVB SPF As discussed in the remainder of this numerical SPF value calculated 40 High’’ on the PDP. section, we agree with the first and according to the test method in new 21

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35625

CFR 201.327(i). For broad spectrum that the term ‘‘sunburn protection followed by one of the descriptors sunscreen products, the term ‘‘Broad factor’’ may be misleading. In the 2007 ‘‘low,’’ ‘‘medium,’’ ‘‘high,’’ or ‘‘highest’’ Spectrum SPF’’ will appear on the PDP sunscreen proposed rule, we revised the (proposed 21 CFR 352.50(a)). The along with the numerical SPF value definition of SPF from ‘‘sun protection proposed descriptors were included to calculated according to the test method factor’’ to ‘‘sunburn protection factor’’ help consumers understand the SPF in new 21 CFR 201.327(i). because we thought that the new value because the label would have The term ‘‘UVB’’ will not be required definition was more descriptive of what included identical descriptors for the as part of the SPF statement. We are also an SPF value represents (72 FR 49070 at UVA star rating. As discussed in section not requiring the descriptor (e.g., ‘‘high’’ 49077). The SPF value is determined III.B. of this document, we are not or ‘‘low’’). We included these two from a clinical test with sunburn as the requiring a UVA effectiveness rating on requirements in the 2007 proposed rule endpoint. However, for broad spectrum the PDP. Therefore, descriptors are no because we had concluded that the sunscreen products, the SPF statement longer required to distinguish the SPF requirements would help consumers also serves as a relative measure of the value from the UVA rating on a understand the side-by-side SPF magnitude of broad spectrum protection sunscreen label. Because we are not numerical rating in conjunction with (Ref. 4). In this final rule, while we do requiring a descriptor after the SPF the UVA star rating, which included the not codify a separate definitional value on the PDP in this document, the term ‘‘UVA’’ and the same descriptors section, we continue to use the term request to include SPF 50 sunscreen (72 FR 49070 at 49084). As discussed in ‘‘SPF’’ to mean ‘‘sun protection factor,’’ products in the ‘‘highest’’ category is no section III.B of this document, the UVA as we had done in the 1999 final rule longer relevant. star rating is not being included in this (21 CFR 201.327(a)(1)). We received two other requests for final rule, and as discussed below, we In this final rule, we are also revising revision to the SPF statement with have concluded that neither the term the effectiveness statement so that the which we do not agree. First, a ‘‘UVB’’ nor the descriptor is necessary term ‘‘UVB’’ is not required before the submission stated that SPF values for consumers to understand the term ‘‘SPF,’’ as proposed in the 2007 should only be labeled in multiples of effectiveness statement. proposed rule (proposed 21 CFR five to be consistent with SPF labeling Neither the term ‘‘UVB’’ nor a 352.50(a)). We received six submissions recommendations by the European descriptor (e.g. ‘‘low’’ or ‘‘high’’) had requesting this revision (Ref. 1). These Commission (Ref. 7). Second, one been included on sunscreen labels prior submissions argued that ‘‘UVB SPF’’ is request from a submission suggested to our 2007 proposal, and consumers an incorrect representation of the SPF that SPF values should be expressed as had been able to make purchase and use value determined from a test using a the percent of UV absorption. The decisions based on SPF values alone. solar stimulator that emits both UVA submission argued that the current SPF Under this final rule, the SPF value will and UVB radiation. The submissions values are misleading because be expressed on the PDP by including point out that sunburn is not caused consumers believe an SPF 15 sunscreen the term ‘‘SPF,’’ followed by the solely by UVB radiation. It is well product is not very protective even numerical value determined from the known that UVA radiation contributes though it screens 93 percent of UV SPF test, similar to how it has appeared up to 20 percent of the skin’s sunburn radiation. on the labels of OTC sunscreen products response (Refs. 5 and 6). One We do not agree with either for more than 30 years. As described in submission points out that if a submission. Based on SPF test data we section III.B of this document, for sunscreen product blocked 100 percent have reviewed, we find that SPF values products passing the critical wavelength of the incident UVB radiation and none for sunscreen products generally can be test in new 21 CFR 201.327(j), the SPF of the erythemally effective UVA determined with a precision that allows value statement will be expressed as radiation, the sunscreen product would the products to be labeled with SPF ‘‘Broad Spectrum SPF’’ followed by the have SPF values no higher than 11 (if values in intervals of less than 5 units numerical SPF value calculated only 9 percent or 1/11 of UV radiation (Ref. 1). Therefore, there is no according to the test method in 21 CFR reaches the skin) (Ref. 4). mathematical or statistical basis for 201.327(i). We agree that UVA radiation restricting SPF values to multiples of We received five submissions contributes to the development of five. Contrary to the second request, objecting to the definition of SPF as sunburn. Although the contribution of consumers have relied on SPF values for ‘‘sunburn protection factor’’ and only UVA to sunburn is less than UVB, it is over 30 years and are familiar with this one submission supporting the still significant (Ref. 5). Further, as format. Therefore, expressing SPF definition (Ref. 1). The submissions stated in the submissions, protection values as percentages may be confusing. objecting to the definition argued that, against UVA radiation is necessary to It would imply that the stated if the term ‘‘sunburn protection factor’’ achieve higher SPF values (Ref. 5). We percentage of the entire UV spectrum is is used, consumers may mistakenly proposed including the term ‘‘UVB’’ in absorbed by a sunscreen. However, the assume that a higher SPF value means the SPF statement in the 2007 proposed SPF values only reflect protection a higher probability of sunburn. rule to help consumers understand that against the portion of the UV spectrum Additionally, they argued that the SPF effectiveness rating is different that causes sunburn. Additionally, the sunscreen products protect against from the UVA effectiveness (star) rating percentages of UV radiation screened various harmful effects of sun exposure, being proposed (72 FR 49070 at 49084). that the submission notes are such as early skin aging and skin cancer, However, as discussed elsewhere in this theoretical. The percentages are in addition to protecting against final rule we are not requiring a UVA determined in a laboratory setting and sunburn. Some submissions suggested effectiveness rating on the PDP (see not under actual use conditions. For that the term ‘‘sunburn protection section III.B.). Therefore, the term example, laboratory tests may show that factor’’ will lead consumers with darker ‘‘UVB’’ is not necessary as part of the an SPF 15 sunscreen absorbs 93 percent skin to assume that they do not need SPF statement. In this final rule, we are of UV rays, but, under actual use sunscreen products because they do not not requiring the term ‘‘UVB’’ be placed conditions, the level of protection burn easily (Ref. 1). before the term ‘‘SPF.’’ provided by an SPF 15 sunscreen We agree with the arguments In the 2007 sunscreen proposed rule, product may be significantly below 93 provided by the submissions suggesting we stated that the SPF value should be percent. There are a number of factors

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35626 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

that lead to this decreased protection, a star is filled or empty, thereby not comprehension study data to allow us to the most important being under- knowing the UVA protection level. determine consumer understanding of application of the sunscreen product (72 Other submissions argued that the terms ‘‘broad spectrum,’’ ‘‘UVA,’’ FR 49070 at 49092). Therefore, if SPF consumers are familiar with star ratings, and ‘‘UVB’’ (58 FR 28194 at 28243). values were expressed as percentages, but that the star rating for items such as Unfortunately, the data we received consumers might mistakenly believe movies and hotels are based on were not sufficient to allow us to that the sunscreen products they are recommendations and not rigorous data. determine the level of consumer using provide more protection than they They suggested several options for understanding of these terms (72 FR really do provide under actual use labeling UVA protection, such as a 49070 at 49081 through 49085), and we conditions. numerical rating or another symbol received no further consumer other than stars. comprehension data in response to the B. Broad Spectrum Statement Some submissions suggested that the 2007 proposal to require the UVA star In response to the 2007 proposed rule, UVA rating should be proportional to rating. While we acknowledge the value we received over 50 submissions the SPF value but requested that there of consumer comprehension data, for collectively making the following four not be two ratings on the PDP. The reasons explained below, we conclude requests regarding the UVA submissions cited the European that conducting consumer effectiveness rating (Ref. 1): Commission’s recommendation that comprehension testing is not necessary 1. Do not require UVA 4-star rating UVA protection increase as the SPF in this case in light of the labeling we system. value increases (Ref. 7). The European have selected for the final rule. 2. Do not require ‘‘no UVA Commission recommends a minimum First, submissions suggesting protection’’ statement if a product does UVA protection factor equal to at least consumer testing were responding to the not protect against UVA radiation. one-third of the labeled SPF or a critical UVA star rating in the proposed rule, 3. Do not require the UVA statement wavelength of at least 370 nm, but does the value of which would have been to be the same size as the SPF statement. not recommend that the actual value of based on the results of two tests (72 FR 4. Perform label comprehension the UVA protection factor or critical 49070 at 49081 through 49085). As studies prior to implementing proposed wavelength be displayed. The European noted, we agree with the submissions PDP labeling. Commission recommends that the main suggesting that the 2007 UVA labeling The submissions included arguments, indicator of sun protection be the SPF proposal was likely to be confusing. but no data, to support these requests. value. Broad spectrum protection is Elsewhere in the document, we also indicated by a symbol on sunscreen discuss our final choice of a pass-fail We agree with the first and second labels—the acronym ‘‘UVA’’ enclosed test for establishing UV protection requests. However, we are not granting within a circle the diameter of which (section VIII.B). As a result of these the third and fourth requests. Our should not exceed the height of the SPF changes in the underlying test method reasons for these decisions are value. and the submissions on the proposed explained below, but we first We agree with the submissions that labeling, we have incorporated a much summarize the related provisions of this the UVA star rating would likely be simpler labeling statement in this final final rule. We are not requiring a star confusing in conjunction with the rule. This statement designates as rating or descriptors to indicate the level numerical SPF rating. We also agree ‘‘broad spectrum’’ those products that of UVA protection as proposed. Instead, with the submissions requesting that are demonstrated to have a critical to indicate the level of UVA and UVB UVA protection should be proportional wavelength of at least 370 nm, using the protection, we are establishing a pass/ to the SPF value. We are requiring such test in new 21 CFR 201.327(j). fail broad spectrum test and a broad proportionality in the broad spectrum Second, unlike in 1993 when we first spectrum labeling statement. If a test described in this document. sought consumer data on the term sunscreen product passes the broad Because of this proportionality, there is ‘‘broad spectrum’’, and unlike the UVA spectrum test (see section VIII.B. of this no longer a need for a separate UVA star rating that we proposed in 2007, document), under this final rule, the rating. Instead of a rating, we are consumers are now likely to be familiar PDP of the product must include the requiring a ‘‘broad spectrum’’ statement with the term ‘‘broad spectrum’’ as statement ‘‘Broad Spectrum SPF [insert on the PDP if a product has a critical included in this document because numerical SPF value resulting from wavelength equal to or greater than 370 some sunscreen manufacturers have testing under paragraph (i) of this nm. This pass/fail ‘‘broad spectrum’’ labeled sunscreen products as ‘‘broad section],’’ without any ‘‘UVA’’ reference statement is consistent with the spectrum’’ for over 20 years. For (§ 201.327(a)(1)(i)). We are requiring the recommendations in the submissions example, the Johnson and Johnson Broad Spectrum SPF statement to citing the recommendations of the ‘‘Sundown Broad Spectrum’’ line of appear as continuous text with no European Commission. sunscreens was on the market in 1988 intervening text or graphics. We are also As noted, several submissions (Ref. 8). As already noted, in our 1993 requiring that the entire text be the same responding to our proposal for a proposed rule, we not only sought font style, size and color on the same separate UVA rating with stars consumer data, but in fact proposed that background color. (§ 201.327(a)(1)(ii)). suggested that consumer comprehension products be permitted to be labeled as With regard to the submissions testing should be performed before the ‘‘broad spectrum’’ if they contained received, nearly all of the 50+ proposed labeling is implemented. We sunscreen active ingredients that submissions argued against requiring agree with the submissions that absorbed UVA radiation, although we the 4-star rating system to display the consumer comprehension data can be did not at that time propose to require level of UVA protection on the PDP of very helpful in formulating labeling a specific test to demonstrate UVA OTC sunscreen products (Ref. 1). Many changes. In fact, in conjunction with our protection (58 FR 28194 at 28233). We submissions stated that the presence of 1993 proposal to allow products to be continued to allow this statement in the stars and a number (SPF) on the PDP labeled as ‘‘broad spectrum’’ if they 1999 sunscreen final rule (64 FR 27666 will lead to consumer confusion. Some contained sunscreen active ingredients at 27666 through 27667). submissions argued that consumers may that absorbed UVA radiation (58 FR Many consumers may also be familiar be confused when determining whether 28194 at 28233), we requested label with the term ‘‘broad spectrum’’ because

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35627

of public health campaigns and news decision concerning protection level which it is effective. Other submissions articles about the importance of broad primarily on the SPF value, and only suggested that we should require all spectrum UV protection over the last secondarily on indications of whether or sunscreen products to provide UVA and two decades. For example, an article not the sunscreen provides broad UVB protection, making this statement appearing in Working Woman magazine spectrum protection. For example, a unnecessary. in 1990 urged women to ‘‘make sure to consumer seeking lower protection may We have concluded that the ‘‘No UVA look for the term ‘broad spectrum’ on have chosen an SPF 15 sunscreen Protection’’ statement is not necessary the label of a sunscreen’’ because ‘‘it product, whereas a consumer seeking and could be misleading. Under this means you’re getting protection from higher protection may have chosen an final rule, the labeling on the PDP of both types of radiation’’ (Ref. 9). SPF 40 sunscreen product. By creating sunscreens no longer refers the type of For consumers not already familiar a clear and standardized ‘‘yes/no’’ UV radiation (UVA or UVB) protection with the term ‘‘broad spectrum,’’ the indicator regarding broad spectrum offered; rather, products that pass the additional indication statement allowed protection, these final labeling critical wavelength test in 201.327(j) are in this document for certain broad requirements will enable consumers to labeled with ‘‘Broad Spectrum SPF’’ spectrum sunscreen products should make better and more informed values. Under this labeling, consumers help consumers recognize the benefit of purchase decisions by looking to see if who see ‘‘UVA’’ on the PDP, even if it these products. Under ‘‘Uses’’ in Drug a product has a ‘‘Broad Spectrum SPF’’ is part of the statement ‘‘No UVA Facts, broad spectrum sunscreen value on the label. Thus, the ultimate Protection,’’ may mistakenly believe products with an SPF value of 15 or purchase decision would be based on that the product offers UVA protection. higher are allowed the following the numerical value associated with the To eliminate this potential indication statement: ‘‘if used as Broad Spectrum SPF statement. For misunderstanding, we are not including directed with other sun protection products offering broad spectrum the ‘‘No UVA Protection’’ statement on measures (see Directions [in bold italic protection, the Broad Spectrum SPF the PDP. font]), decreases the risk of skin cancer value on the PDP will not only indicate In contrast to four submissions and early skin aging caused by the sun’’ the relative level of protection against requesting that we make the UVA (new 21 CFR 201.327(c)(2)). UVB radiation but will also reflect the statement less prominent than the SPF In addition, educational campaigns level of UVA protection, with increasing statement, we are requiring the SPF and about sun protection will further inform SPF values indicating greater protection broad spectrum statements to be equally consumers about the benefits of using against both UVA and UVB radiation. prominent on the PDP by appearing as sunscreens that include the term ‘‘broad For broad spectrum products, linking a combined statement. The four spectrum’’ on their labels and have an the amount of UVA protection to the submissions stated that they believe SPF value of 15 or higher. We expect SPF value, is consistent with the UVB radiation contributes more to skin consumers to learn that a sunscreen approach taken in Europe (Ref. 7). cancer and photodamage than UVA labeled with the statement ‘‘Broad For broad spectrum products, we are radiation and argued that more Spectrum SPF’’ 15 or higher, when used requiring the broad spectrum statement prominence should be given to the SPF as directed with other sun protection on the PDP to appear in combination statement. However, none of the measures, offers more comprehensive with the SPF statement. For example, an submissions included data to support protection against sun-induced skin SPF 40 sunscreen product which passes this argument. Some submissions damage than that provided by a the broad spectrum test will be labeled suggested that consumers are familiar sunscreen that is not broad spectrum or ‘‘Broad Spectrum SPF 40’’ in a uniform with SPF ratings and that providing that are broad spectrum with an SPF font style, size, and color and with the another rating with similar prominence value less than 15. same background color. This placement may mislead and confuse consumers. It is important to note that the broad will help consumers recognize that the It is well known that both UVA and spectrum test required in this document particular sunscreen product is broad UVB radiation contribute to captures both UVB and UVA protection spectrum in conjunction with the SPF photodamage and skin cancer (Refs. 6– for the effectiveness of a sunscreen value. As previously explained, the 7 and 10–12). Therefore, in our view, product. The broad spectrum test is not broad spectrum statement and SPF providing consumers with information limited to UVA wavelengths as was the value together will provide a relative about the effectiveness of a sunscreen case with the proposed test (see section measure of both UVB and UVA product for UVA and UVB radiation VIII.B of this document). By requiring protection. Combining the broad protection is equally important. We are that a broad spectrum sunscreen spectrum and SPF statements will help requiring that the broad spectrum provide both UVB and UVA protection consumers become more aware of the statement be displayed in combination in a pass/fail test, the amount of UVA importance of broad spectrum with the SPF statement. The two protection for a sunscreen product that protection. statements must not be interrupted with passes the test must increase as the SPF Under the 2007 proposed rule, if an any graphics or text. In addition, the increases. For example, a Broad OTC sunscreen product was not tested broad spectrum statement must be the Spectrum SPF 40 sunscreen product for or did not protect against UVA same font style, size, and color as the provides greater protection against both radiation, the statement ‘‘No UVA SPF statement with the same UVB and UVA than a Broad Spectrum protection’’ would have been required background color. It is important for SPF 20 sunscreen product. In contrast, on the PDP (proposed 21 CFR consumers to evaluate both statements an SPF 40 sunscreen product that is not 352.50(b)(1)). Ten submissions argued when making a purchase decision. By broad spectrum provides more UVB against requiring this statement (Ref. 1). requiring this information to be protection than a SPF 20 sunscreen Some submissions argued that this presented with identical prominence on product that is not broad spectrum, but statement is misleading because all the PDP, consumers should be able to may not provide more UVA protection. sunscreen products provide some UVA quickly and easily identify sunscreen This proportionality between UVB protection. Submissions also stated that products that provide broad spectrum and UVA protection is important a negative statement is inconsistent with protection, as well as the SPF of all because consumers have been the OTC Drug Review because a drug sunscreen products. While we are not accustomed to basing their purchase should only describe the indications for requiring a negative statement on the

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35628 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

PDP of products that do not pass the sunscreen product is ‘‘very water concepts on sunscreen labeling is critical wavelength test in new resistant’’ does not give any indication potentially confusing and could raise 301.327(j), we caution that such of how much time a consumer can additional questions about their products may be misbranded if they safely spend in the water. Under the meaning. We could not determine a include statements regarding UVA 2007 proposed rule, a consumer would succinct educational statement that protection; such statements may have had to read either the ‘‘Uses’’ or would not also be potentially misleadingly imply that the product the ‘‘Directions’’ section of the Drug misleading. Therefore, we have provides benefits that are similar or Facts label to determine the duration of concluded that an educational statement superior to those of products labeled water resistance for a sunscreen product should not be required on the PDP. with Broad Spectrum SPF values. (proposed 21 CFR 352.52(b)(1)(vii) and As noted, submissions also requested (b)(1)(viii) and proposed 21 CFR that the proposed educational statement C. Water Resistance Statement 352.52(d)(2) and (d)(3), respectively; 72 be combined with proposed sun alert, In the 2007 sunscreen proposed rule FR 49070 at 49113). included in the proposed rule as a (proposed 21 CFR 352.52), we allowed Providing, on the PDP, specific warning. In section IV.C of this the PDP of OTC sunscreen products to information about the actual time (40 or document, we address submissions on contain the statement ‘‘water resistant’’ 80 minutes) a consumer can expect a the sun alert warning, and explain our if a sunscreen product was shown to sunscreen product to retain its labeled decision to incorporate the information retain the labeled SPF value after 40 SPF value is likely to be more helpful regarding the role of certain sunscreens minutes of water immersion, or ‘‘very to consumers because the information is in reducing the risk of skin cancer and water resistant’’ if a sunscreen product displayed in one place—on the PDP and early skin aging into a new indication was shown to retain the labeled SPF not on different parts of the labeling. and accompanying directions for value after 80 minutes of water The revised statements ‘‘water resistant sunscreens with Broad Spectrum SPF immersion, according to the test in (40 minutes)’’ or ‘‘water resistant (80 values of 15 or higher. We are retaining proposed 21 CFR 352.76. We minutes)’’ should make it clearer and a modified warning to be included as simultaneously proposed that the easier for consumers to understand the first warning on sunscreen products ‘‘Uses’’ section of labeling (not the PDP) water resistance as part of their that are either not broad spectrum or indicate specifically whether the purchase decision. This water resistance that are broad spectrum with an SPF product had been established to be information continues to be reinforced value less than 15. Because we are not water resistant for 40 minutes or 80 by information in the directions requiring an educational statement on minutes, and included specific regarding reapplication. the PDP and are either eliminating or directions addressing times for modifying the proposed sun alert reapplication of each product, D. UVB and UVA Educational warning, the request to combine these dependent on its level of water Statement two statements is no longer relevant. resistance (proposed 21 CFR In the 2007 sunscreen proposed rule, 352.52(b)(1)(vii), (b)(1)(viii), (d)(2), and we proposed that the following IV. Drug Facts Labeling (d)(3); 72 FR 49070 at 49113). In this educational statement be included on In September 2004 (69 FR 53801), we document, we are revising the PDP to the PDP of all OTC sunscreen products delayed the May 16, 2005, contain the statement ‘‘water resistant (proposed 21 CFR 352.50(c)): ‘‘UV rays implementation date for the Drug Facts (40 minutes)’’ or ‘‘water resistant (80 from the sun are made of UVB and final rule (21 CFR 201.66) for OTC minutes)’’ as determined by the water UVA. It is important to protect against sunscreen products until further notice). resistance test in new 21 CFR both UVB and UVA rays to prevent The Drug Facts final rule (21 CFR 201.327(i)(7). We are removing this sunburn and other skin damage.’’ 201.66) establishes general labeling information from the indications section We received four submissions format and content requirements for all of Drug Facts (section IV.B of this regarding the UVB and UVA OTC drugs. With the additional document). We continue to include educational statement in response to the exception of certain OTC drug products directions based on the duration of 2007 sunscreen proposed rule (Ref. 1). in ‘‘convenience size’’ packages (see 67 water resistance established under the The submissions made the following FR 16304 at 16306 (April 5, 2002), other new water resistance test (section IV.D requests: OTC drug products are already required of this document). • Do not require the educational to comply with 201.66. We delayed One submission stated that including statement on the PDP or implementation of 201.66 for information about water resistance in • Combine the educational statement sunscreens so as to avoid the potential the indications section as well as in the with the sun alert statement and include that sunscreen manufacturers would directions section is ‘‘redundant and the combined statement in the ‘‘Other have to relabel their products twice confusing’’ (Ref. 1). The submission Information’’ section of the Drug Facts within a short time period if a final rule recommended that we delete the label. specifying labeling for sunscreens indications statement. We agree with We considered including the published shortly after the original May the submission. To eliminate proposed educational statement on the 2005 implementation date for the redundancy and simplify the labeling PDP. We concluded that this general content and format requirements for consumers, we are relocating the information is not critical for effective of the Drug Facts final rule. We information formerly contained within use of sunscreen products, particularly published the notice of delay for OTC the indication statement to the PDP. since we are no longer requiring other sunscreens’ implementation of the Drug The content of the labeling as a whole PDP statements to refer separately to Facts final rule so that such products is the same as that included in the 2007 UVA and UVB protection. An could simultaneously implement both proposed rule. However the proposed understanding that the sun produces the general labeling provisions of that statement on the PDP did not clearly ultraviolet (UV) rays or that there are rule and the specific labeling provisions and accurately convey to consumers the two types of UV rays that reach the for sunscreens when we published a difference between ‘‘water resistant’’ earth’s surface is not necessary to ensure sunscreen labeling final rule. We are and ‘‘very water resistant’’ sunscreen the safe and effective use of sunscreen now lifting the stay on the products. For example, knowing that a products. The explanation of these implementation of the Drug Facts final

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35629

rule for OTC sunscreen products. In this they will achieve compliance with two status of individual sunscreen active document, we are requiring the same rules through one labeling revision ingredients, we are not lifting the stay implementation date for the regulations (rather than following the more of 21 CFR 347.20(d). set forth in this labeling and testing final expensive course of making two labeling This document requires much of the rule (21 CFR 201.327) and in the Drug changes at two different times). Drug Facts labeling included in the 2007 Facts final rule (21 CFR 201.66) as In 2003 (68 FR 33362, June 4, 2003), proposed rule. However, we have made applied to these sunscreen products. we also stayed the part of the skin This action will benefit both protectant monograph that describes several revisions to the proposed consumers and manufacturers. GRASE combinations of skin protectant labeling. These revisions are discussed Consumers will benefit by having and sunscreen active ingredients (21 in detail throughout the remainder of sunscreen labeling presented in the CFR 347.20(d)). Because this document this section. In addition, table 2 of this Drug Facts format that they are familiar addresses the labeling and testing of document summarizes these revisions with. Manufacturers benefit because sunscreen products and not the GRASE as follows:

TABLE 2—SUMMARY OF DRUG FACTS LABELING INCLUDED IN THE 2007 PROPOSED RULE AND THIS FINAL RULE

Drug facts section 2007 Proposed rule This final rule

Active Ingredients/ Name and amount of ingredient(s) followed by ‘‘sunscreen’’ Name and amount of ingredient(s) followed by ‘‘sun- Purpose. screen.’’ Uses ...... • [low, medium, high, or highest] UVB sunburn protection • for all sunscreen products: ‘‘helps prevent sunburn.’’ • [low, medium, high, or highest] UVA protection • Optional, for sunscreen products with Broad Spectrum • retains SPF after 80 minutes of activity in the water SPF values of 15 or higher, ‘‘if used as directed with other sun protection measures (see Directions), de- creases the risk of skin cancer and early skin aging caused by the sun.’’

Warnings ...... UV exposure from the sun increases the risk of skin can- For sunscreen products that are not broad spectrum or for cer, premature skin aging, and other skin damage. It is products that are broad spectrum with an SPF value less important to decrease UV exposure by limiting time in than 15, Skin Cancer/Skin Aging Alert [in bold font]: the sun, wearing protective clothing, and using a sun- Spending time in the sun increases your risk of skin can- screen. cer and early skin aging. This product has been shown only to help prevent sunburn, not [in bold font] skin can- cer or early skin aging. For all sunscreens: For external use only For external use only Do not use on damaged or broken skin Stop use and ask a doctor if skin rash occurs Stop use and ask a doctor if rash occurs When using this product keep out of eyes. Rinse with When using this product keep out of eyes. Rinse with water to remove. water to remove. Keep out of reach of children. If swallowed, get medical Keep out of reach of children. If swallowed, get medical help or contact a Poison Control Center right away. help or contact a Poison Control Center right away. Directions ...... Non-Water Resistant Product Non-Water Resistant Product • apply liberally [# minutes] before sun exposure • apply liberally 15 minutes before sun exposure • reapply at least every 2 hours and after towel drying, • use a water resistant sunscreen if swimming or sweating swimming, or sweating • reapply at least every 2 hours • apply and reapply as directed to avoid lowering protec- • children under 6 months: Ask a doctor tion • children under 6 months: Ask a doctor Water Resistant Product Water Resistant Product • apply liberally [# minutes] before sun exposure • apply liberally 15 minutes before sun exposure • reapply after 40 [or 80] minutes of swimming or sweating • reapply: and after towel drying. Otherwise, reapply at least every • after 40 [or 80] minutes of swimming or sweating 2 hours. • immediately after towel drying • apply and reapply as directed to avoid lowering protec- • at least every 2 hours tion • children under 6 months: Ask a doctor • children under 6 months: Ask a doctor Water Resistant and Non-Water Resistant Products Water Resistant and Non-Water Resistant Products No statement For sunscreens with Broad Spectrum SPF values of 15 or higher: • Sun Protection Measures [in bold font]. Spending time in the sun increases your risk of skin cancer and early skin aging. To decrease this risk, regularly use a sunscreen with a Broad Spectrum SPF value of 15 or higher and other sun protection measures including: • limit time in the sun, especially from 10 a.m.–2 p.m. • wear long-sleeved shirts, pants, hats, and sun- glasses. Inactive Ingredients .. List inactive ingredients in alphabetical order List inactive ingredients in alphabetical order. Other Information ..... No required statements • protect this product from excessive heat and direct sun. Questions? ...... No required statements No required statements.

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35630 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

A. Active Ingredients/Purpose the public concerning these statements increase in the consumer’s overall UV We received one submission following publication of the 2007 exposure. Therefore, if the indication regarding the listing of active proposed rule (Ref. 1). statement regarding decreasing risk of ingredients and one submission We are not requiring these indication skin cancer and early skin aging does requesting that we provide specific statements in this final rule. Instead, all not include the information about using details about what each ingredient does sunscreen products covered by this rule the product as directed, which includes in the product (Ref. 1). One of these will be required to include the following other sun protection submission also requested that we indication statement ‘‘helps prevent measures, the statement will be require listing of the percentage of each sunburn,’’ as required in the 1999 considered misleading (and thus make a active ingredient next to the ingredient sunscreen final rule (64 FR 27666; new sunscreen product misbranded) (new 21 name. 21 CFR 201.327(c)(1)). We are requiring CFR 201.327(c)(3)). Similarly, sunscreen We are not making any changes to the this statement instead of the first products covered by the rule that ‘‘Active ingredients/Purpose’’ section of proposed statement (level of UVB provide broad spectrum with SPF the Drug Facts label. The general OTC sunburn protection) because we agree values between 2 and 15 or do not labeling regulations specify that the with submissions arguing that sunburn provide broad spectrum protection ‘‘quantity of each active ingredient per is not caused solely by UVB radiation should not state or imply that the use of dosage unit’’ be listed with the (Ref. 1). We also agree with submissions a sunscreen product alone will reduce established name of each active arguing that the SPF value by itself on the risk of skin cancer or early skin ingredient (21 CFR 201.66(c)(2)). the PDP informs consumers of the level aging. Doing so would cause the product Therefore, every sunscreen product is of sunburn protection, so a separate to be misbranded. already required to include the active description of the level of sunburn We are not including the third ingredient names followed by the protection does not need to be included proposed indication statement percentage (weight per volume) in the as an indication. (regarding water resistance) in this ‘‘Active ingredients/Purpose’’ section, In addition, sunscreen products document. As already discussed, under as requested by the first submission. covered by this rule that provide broad this final rule, information about water We are not requiring specific details spectrum protection according to the resistance is included on the PDP, as about what each ingredient does in the test in new 21 CFR 201.327(j) and have well as under ‘‘Directions’’ in Drug product. The function of each active SPF values of 15 or higher, may include Facts (see sections III.C and IV.D of this ingredient in an OTC drug product is the following indication statement (new document). We conclude that already required to be listed by 21 CFR 21 CFR 201.327(c)(2)(i)): ‘‘if used as information about the water resistance 201.66(c)(3), which specifies that the directed with other sun protection of a sunscreen product is more ‘‘Active ingredients/Purpose’’ section of measures (see Directions), decreases the effectively and accurately presented on the label list the ‘‘general pharmacologic risk of skin cancer and early skin aging the PDP and as a direction than as an categories or principal intended actions caused by the sun.’’ This statement indication statement. The extent of of each active ingredient.’’ There is not replaces the second proposed indication water resistance informs a consumer currently a requirement to list the statement. We are allowing this about how long the SPF value is purpose of inactive ingredients on OTC statement for certain sunscreens covered retained following water exposure and, drug labels. This information is not by this rule based on available clinical therefore, how long an interval can needed to safely and effectively use studies, the fact that UV radiation from elapse before reapplying the sunscreen sunscreen products. Therefore, in this the sun is harmful, and the scientific product (40 or 80 minutes). In addition, document, we are not requiring the understanding that substantially the PDP requirements in this document purpose of inactive ingredients to be limiting overall UVB and UVA exposure include the time interval as part of the listed on sunscreen labels. reduces the risk of skin cancer and early water resistance statement, so that consumers can readily distinguish B. Uses skin aging. As discussed in the remainder of this between products on this basis when 1. Indications Statements Proposed in section of the document, it is critical making purchasing decisions. Because the 2007 Proposed Rule that the indication statement regarding we include water resistance on the PDP The 2007 proposed rule included skin cancer and early skin aging and under ‘‘Directions,’’ we are not three indication statements under includes information about using the including a separate indication ‘‘Uses’’ in Drug Facts: products as directed and following other statement about water resistance in this 1. Level of UVB sunburn protection sun protection measures (listed under document. (proposed 21 CFR 352.52(b)(1)(i)– the heading Directions). We have 2. Statement Regarding Skin Cancer and (b)(1)(iv)) concluded that the reference to other Early Skin Aging 2. Level of UVA protection (proposed sun protection measures is necessary to 21 CFR 352.52(b)(1)(v) and (b)(1)(vi)) ensure that the consumer’s overall UV a. Submissions Arguing For a Skin 3. Extent of water resistance exposure is substantially decreased. A Cancer and Early Skin Aging Indication (proposed 21 CFR 352.52(b)(1)(vii) and consumer who relies on the use of a As already stated, in this final rule we (b)(1)(viii)) sunscreen with Broad Spectrum SPF have adopted, for the first time, an The first statement would have value of 15 or higher alone may not indication for skin cancer and early skin appeared on all monograph sunscreen obtain a meaningful net decrease from aging for sunscreen products covered by products. The second statement would the risk of skin cancer or early skin the rule that have Broad Spectrum SPF only have appeared on monograph aging if, because he or she is wearing values of 15 or higher. In our 2007 sunscreen products providing UVA the sunscreen, the consumer spends proposed rule, we had included protection. The third statement would more time in the sun and/or wears less indication statements that indicated the only have appeared on monograph protective clothing. In fact, reliance on degree of protection against both UVB sunscreen products that are water sunscreen use alone, without also and UVA radiation but that linked UVB resistant for either 40 or 80 minutes. We employing other sun protection protection only to sunburn prevention received numerous submissions from measures, could actually result in an and did not expressly link UVA

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35631

protection to any specific health benefit sunscreens than for sunscreens covered b. Limiting Overall UV Exposure (proposed 21 CFR 352.52(a)). At the by the rule that are not broad spectrum Reduces Risk of Skin Cancer and Early same time, however, we had proposed or that provide broad spectrum with Skin Aging both an educational statement on the SPF values less than 15. In addition, we For drugs subject to OTC monographs, PDP stating that UV rays from the sun conclude that such an indication should like sunscreen products, indication are made of both UVB and UVA and not be included in the Warnings section statements about the effectiveness of the that it is important to protect against of Drug Facts. We have concluded that, drug products must be supported with both types of radiation to prevent as proposed in 2007, the second scientific data (21 CFR 330.10(a)(4)(ii)). sunburn and other skin damage sentence of the first warning (i.e., the In order for an OTC drug to be (proposed 21 CFR 352.50 (c)). We also ‘‘Sun Alert’’ warning) is an implied considered generally recognized as proposed a ‘‘sun alert’’ statement as the indication: ‘‘It is important to decrease effective (GRAE), there must be a first warning. This first warning read, UV exposure by limiting time in the reasonable expectation that, in a given ‘‘UV exposure from the sun increases sun, wearing protective clothing, and proportion of the target population, the the risk of skin cancer, premature skin using a sunscreen.’’ Because it follows drug will provide clinically significant aging, and other skin damage. It is relief of the type claimed (21 CFR important to decrease UV exposure by a warning that ‘‘UV exposure from the sun increases the risk of skin cancer, 330.14(a)(4)(ii)). Based on the available limiting time in the sun, wearing data concerning the harmful effects of premature skin aging, and other forms of protective clothing, and using a UV radiation and sunscreen UV skin damage,’’ the second sentence sunscreen.’’ (proposed 21 CFR protection, we have concluded that 352.52(c)(1)). implies that using any sunscreen, sunscreens, in conjunction with the In response to our proposed rule, we regardless of SPF value or broad critical behavioral steps of limiting time received a total of 12 submissions spectrum protection, and following in the sun particularly during the asking that we include a specific other sun protection measures will midday hours and wearing protective statement regarding reduction in risk of decrease the risks of skin cancer, early clothing (long sleeve shirt, pants, hat, skin cancer and early skin aging as an skin aging, and other consequences of and sunglasses), provide ‘‘clinically indication for covered sunscreens (Ref. UV exposure to the sun. We have significant relief’’ in reducing the risk of 1). The submissions asked that we allow concluded, based on a reconsideration skin cancer and early skin aging. Based an indication statement informing of data previously reviewed in the 2007 on the available data, we have limited consumers that the regular, consistent, proposed rule, that, if consumers use this claim to broad spectrum sunscreen or continued use of a sunscreen product broad spectrum sunscreens with SPF products with an SPF value of 15 or reduces or helps reduce the risk or values of 15 or higher and follow other higher. chance of developing skin damage, early sun protection measures, they can UV radiation from the sun has been skin aging, and some types of skin reduce their risk of skin cancer and associated with nonmelanoma skin cancer (Ref. 1). These submissions also early skin aging. For these products, we cancers since 1927 and with melanomas supported our proposed requirement of agree with the public submissions that since 1952 (Ref. 13). It is estimated that a ‘‘sun alert’’ on the labeling to inform this information is most appropriately as much as 90 percent of melanomas and nonmelanomas are caused by sun consumers of the need to limit time in placed as an indication (i.e., under Uses) exposure (Ref. 5). In 1992, the the sun and wear protective clothing. with a reference to the need to use the The submissions came from sunscreen International Agency for Research on product as directed with other sun Cancer (IARC), under the auspices of the manufacturers and public health protection measures. For these products, organizations including the American World Health Organization, identified we include under the heading UV radiation as a human carcinogen 3 Academy of Dermatology, the American Directions, specific reference not only to Cancer Society, and the Skin Cancer (Ref. 14). More recently, broad spectrum regularly use sunscreens with Broad Foundation. Many of the submissions UV radiation was listed as a human Spectrum SPF values of 15 or higher provided references to studies that they carcinogen in the National Toxicology (the subset of sunscreens for which the argued support the inclusion of this Program’s 11th Report on Carcinogens indication statement. One submission indication is allowed) but also to issued in 2005 (Ref. 15). It is important specifically requested that we allow an employ the other listed sun protection to note that this report indicates that anti-aging claim (without mention of measures listed under Directions. For UVB and UVA radiation across the skin cancer), and one other submission sunscreen products covered by this rule spectrum are known human argued that no sunscreen can claim to that are not Broad Spectrum or that are carcinogens, but that either UVB prevent cancer (Ref. 1). We received no broad spectrum with an SPF value less radiation alone or UVA radiation alone is ‘‘reasonably anticipated to be a new data to accompany these requests than 15, however, we conclude that human carcinogen.’’ This classification for a separate indication that the regular existing data are insufficient to support is due to the fact that the exact use of sunscreen decreases the risk of an indication for reducing risk of skin wavelengths of UV radiation that cause skin cancer and early skin aging. cancer or early skin aging. In the different harmful effects (e.g., DNA However, on reconsideration of the data sections that follow, we explain the damage or loss of skin elasticity) have reviewed prior to the 2007 proposed specific scientific basis for our not yet been identified. It is clear, rule, we agree with the argument that conclusion, as well as explain our though, that broad spectrum UV the data underpinning our proposed rationale for the specific framing of the radiation causes skin cancer. Broad education statement and warning are labeling, as included in the final rule, spectrum UV radiation has also been sufficient to support an appropriately for both subsets of the sunscreens shown to cause other types of skin qualified skin cancer and premature covered by the final rule—those that damage, including early skin aging skin aging indication for one subset of have Broad Spectrum SPF values of 15 (Refs. 6 and 16). Therefore, we agree sunscreens covered by this rule—those or higher and those that do not have that have Broad Spectrum SPF values of Broad Spectrum or that are Broad 3 A carcinogen is anything that is known to cause 15 or higher. As a result, our final rule spectrum with SPF values less than 15. the development of cancer. UV radiation is known provides different labeling for these to cause skin cancer.

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35632 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

with the principle that a reduction, of proposed rule (72 FR 49070 at 49086). study data, we have concluded that if sufficient magnitude, in broad spectrum Reynolds et al. published a study in consumers use sunscreens with Broad UV exposure should reduce the risk of 1996 finding, in a study of 509 sixth- Spectrum SPF values of 15 or higher, harmful effects to the skin, including graders, that adolescents who used limit time in the sun especially during skin cancer and early skin aging. sunscreen on both Saturday and Sunday the midday hours, and wear protective Broad spectrum sunscreens, by of a Labor Day weekend spent clothing when exposed to the sun, the absorbing UVA and UVB radiation, significantly more time in the sun than resulting reduction in overall UV decrease consumer exposure to both those who used sunscreen only one day exposure will reduce the risk of skin types of UV radiation from the sun that or not at all (Ref. 17). In the second cancer and early skin aging. Therefore, reach the earth’s surface. Other critical study, parents of 503 children, aged less there is sufficient evidence of behavioral steps, such as limiting time than 2 to 12 years, were surveyed as to ‘‘clinically significant relief’’ to justify in the sun and wearing protective parental attitudes about their children’s the indication and related directions for clothing, also decrease consumer sun exposure (Ref. 18). The authors this subset of products, as set forth in exposure to UVA and UVB radiation. reported that ‘‘sunscreen use in children the rule. However, we conclude that the After considering the submissions and was significantly associated with longer omission of prominent information in other available data, we have concluded duration of sun exposure’’ (Ref. 18). the indication regarding the need for that a claim for the reduction in risk of Increased overall UV exposure might, other sun protection measures would skin cancer and early skin aging is in fact, increase the risk of skin cancer misbrand the product, as would the appropriate for certain sunscreen and early skin aging, despite the proper omission of the associated direction products, when the claim also includes use of sunscreens. Likewise, if specifying these measures. Indeed, it the requirement that consumers use the consumers limit time in the sun, would suggest a different indication product as directed and the Directions especially during midday, and wear than that which available evidence specify other sun protection measures more protective clothing (such as broad supports. Consequently, we have be followed (see section IV.D of this brimmed hats, long pants, and long included in this final rule a new document). We are basing this claim on sleeve shirts) while outside, but do not provision indicating that ‘‘Any labeling the scientific understanding of the harm use sunscreens for areas of the skin or promotional materials that suggest or from UVA and UVB radiation and the exposed to the sun (such as parts of face imply that the use, alone, of any absorption and/or reflection of that UV and neck), then the consumer may not sunscreen reduces the risk of or radiation by broad spectrum sunscreens, decrease the risk of skin cancer and prevents skin cancer or early skin aging as well as data from studies concerning early skin aging for sun-exposed areas. will cause the product to be misbranded sunscreen use and the development of For these reasons, for products that are under section 502 of the FD&C Act (21 skin cancer or precursors of skin cancer entitled to include an indication for U.S.C. 352).’’ (new 21 CFR (section IV.B.2.c of this document). reducing the risk of skin cancer and 201.327(c)(3)). For a sunscreen to be effective (i.e., early skin aging, we continue to direct provide ‘‘clinically significant relief’’) in consumers to follow a comprehensive c. Available Scientific Data reducing the risk of skin cancer and sun protection program that includes We are not aware of any data other early skin aging, consumers must not use of sunscreens with Broad Spectrum than what we reviewed in the 2007 increase their overall exposure to UV SPF values of 15 or higher, limiting time proposed rule that evaluate the radiation by overreliance on sunscreen in the sun, and wearing protective effectiveness of sunscreens in reducing use. Other behavioral factors could clothing, similar to the sun protection the risk of skin cancer or early skin account for such an increase, such as measures discussed in the 2007 aging for healthy subjects. One more the amount of time spent in the sun and proposed rule (72 FR 49070 at 49089). recent study, published in 2009, found the use of protective clothing. If Nearly identical multi-step behavioral that regular use of Broad Spectrum SPF consumers rely on sunscreen use to sun protection programs are advocated 50+ sunscreen ‘‘may prevent’’ the spend more time in the sun and/or to by a number of medical and public development of actinic keratoses and wear less protective clothing, then health organizations, including the non-melanoma skin cancer in immune- consumers could actually increase their American Academy of Dermatology, the compromised organ transplant overall UV exposure, which would Skin Cancer Foundation, and the recipients (Ref. 19). We have not relied eliminate the effectiveness of sunscreen American Cancer Society. on this study in reaching our use in reducing the risk of skin cancer We have concluded that a conclusions regarding OTC sunscreens, and early skin aging. comprehensive sun protection approach because we do not consider the To illustrate this point, it is helpful to is critical to ensure that consumers who immune-compromised study population consider what has been termed the are seeking to obtain a reduction in the to be representative of the general ‘‘compensation hypothesis.’’ As we risk of skin cancer and early skin aging population. noted in the 2007 proposed rule, the limit their overall sun exposure. We have re-evaluated the data compensation hypothesis states that Without the reduction in consumers’ originally reviewed in preparing the consumers who wear high SPF overall UV exposure, even a sunscreen 2007 proposed rule to determine sunscreens generally spend more time with Broad Spectrum SPF value of 15 or whether those data support allowing the in the sun and/or wear less protective higher may not be effective in indication for all sunscreen products or clothing (72 FR 49070 at 49086). If the decreasing the risk of skin cancer and only for certain sunscreen products. hypothesis is true, consumers would not early skin aging. As discussed below, Based on our re-evaluation, we have reduce their risk of skin cancer or early the available clinical studies do not concluded that the data is supportive of skin aging because their overall UV control for these behavioral factors and, an indication for broad spectrum exposure increases, even though a therefore, do not demonstrate that even sunscreens having SPF values of at least properly applied (and reapplied) this subset of sunscreens alone reduce 15. Further, we have determined that, sunscreen absorbs UV radiation and the risk of skin cancer and early skin while the existing evidence does not helps prevent sunburn. We cited two aging. However, based on the scientific support a claim for the use of any retrospective studies which support the understanding of the harm from UV sunscreen alone, it does support an compensation hypothesis in the 2007 exposure and our assessment of the indication that the combination of using

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35633

a sunscreen with Broad Spectrum SPF broad spectrum with SPF values less cancer and early skin aging. UVA and value of 15 or higher along with other than 15. UVB radiation is the only known sun protection measures, reduces the Two prospective studies have shown external risk factor for skin cancer and risk of skin cancer and early skin aging, that regular use of a Broad Spectrum early skin aging. Therefore, measures consistent with other positions in the SPF 30 sunscreen reduces the risk of that significantly reduce both UVA and 2007 proposed rule (72 FR 49070 at developing melanocytic nevi, which can UVB exposure should decrease the risk 49087 through 49090). progress into melanomas (Ref. 22). In a of skin cancer and early skin aging. To date, there are no clinical studies 2000 study, Gallagher et al. examined Based on this understanding, limiting demonstrating that use of any sunscreen the formation of new melanocytic nevi time in the sun, wearing protective alone can prevent skin cancer. There are in 393 Canadian school children. The clothing and using a broad spectrum two prospective 4 studies that directly group of children given Broad Spectrum sunscreen with an SPF value of 15 or examine the role of sunscreen products SPF 30 sunscreen product had fewer higher should decrease the risk of skin in preventing skin cancer. Although it new nevi over the course of the three cancer and early skin aging. Using a did not show any difference in primary year study than did children not given broad spectrum sunscreen with an SPF endpoints, a large 1999 study conducted sunscreen products or advice on value of 15 or higher ensures adequate in Australia demonstrated that people sunscreen use (Ref. 27). The difference breadth and magnitude of UVA and who applied a Broad Spectrum SPF 15 was small (24 v. 28 nevi, respectively), UVB protection. For these products, the sunscreen product on a daily basis over but statistically significant (p = 0.048). broad spectrum test measures breadth a 4.5 year period had a lower overall In a follow-up study published in 2005, and SPF test measures magnitude of UV incidence of one type of skin cancer, Lee et al. evaluated the same group of protection. Consistent with this squamous cell carcinoma, on the head, children for differences in melanocytic scientific principle, the short-term neck, arms, and forearms than study nevi by location on the body and effectiveness studies demonstrate a participants who did not apply demographic factors (Ref. 28). These decrease in the development of investigators found that the sunscreen sunscreen (28 cases in the broad surrogates for skin cancer and early skin group had significantly fewer new nevi spectrum sunscreen group vs. 46 cases aging. Thus, we have concluded that the on the trunk than the control group in the group not using broad spectrum available evidence supports our finding (p = 0.05). that sunscreen products, in conjunction sunscreen) (Ref. 20). In an extension of With respect to the role of sunscreen that study, van der Pols et al. evaluated with limiting time in the sun and products in decreasing the risk of early wearing protective clothing, reduce the the same population of subjects over an skin aging, we are aware of only indirect additional 8 years, and found that the risk of developing skin cancer or early evidence that sunscreen use decreases skin aging. sunscreen users continued to have a early skin aging. One recent study statistically significant lower incidence demonstrated that a broad spectrum d. Indication Limited to Covered of squamous cell carcinoma over the sunscreen product can reduce the extent Sunscreens With Broad Spectrum SPF entire 12.5 year period (Ref. 21). Neither of solar UV-induced damage to factors Values of 15 or Higher study found that daily sunscreen use associated with early skin aging even In light of the submissions requesting had any measurable effect on the most when the SPF value is less than 10 (Ref. that we reframe our labeling information common form of skin cancer, basal cell 29). Although this study was small, regarding sunscreen use and reduced carcinoma. Further, we are not aware of evaluating only 12 Caucasian subjects, it risk of skin cancer and premature skin any studies examining the effect of shows the importance of broad aging as an indication, we re-evaluated sunscreen use on the development of spectrum protection. These findings skin cancer and aging studies discussed melanoma, which is the deadliest form have been corroborated in a large in the 2007 proposed rule to determine of skin cancer. number of studies using broad spectrum whether the skin cancer and early skin Although data from clinical studies sunscreens with SPF values ranging aging indication should apply to all addressing the specific end points of from 19 to 50, as reported by Fourtanier sunscreen products or be limited to cancer is limited, some prospective et al. in two recent reviews (Refs. 10 and certain sunscreen products. Available studies have evaluated the effects of 30). data support this indication only for regular sunscreen use on the Neither those studies evaluating the broad spectrum sunscreens with SPF development of surrogate skin lesions long term effect of regular sunscreen use values of 15 or higher. Several reports that can be precursors to cancer: actinic on the development of skin cancer and have indicated that UV-induced skin keratoses and melanocytic nevi. A small early skin aging nor those evaluating the damage associated with both skin percentage of actinic keratoses progress long term effect of sunscreen use on cancer and early skin aging can be to squamous cell carcinomas (Ref. 22). surrogate markers for these conditions reduced by the use of broad spectrum At least four studies have demonstrated were adequately controlled. Such sunscreens (Refs. 10 and 29 through 31). that the number of actinic keratoses is studies, which must take place over In a direct comparison of a broad lower for individuals regularly using many years, make adequate controls spectrum sunscreen and a non-broad sunscreens with Broad Spectrum SPF extremely difficult, if not impossible to spectrum sunscreen with the same SPF, values of 15 or higher (Refs. 23 through implement. For example, one cannot Moyal and Fourtanier found that the 26). We are not aware of any studies control for time and duration of broad spectrum sunscreen provided examining the potential effects on exposure, application and re-application significantly better protection from UV surrogate skin lesions of sunscreens that amounts, or use of supplemental radiation-induced immunosuppression, either are not broad spectrum or are behavioral measures such as wearing a factor associated with both skin cancer protective clothing for a study which and early skin aging (Ref. 32). 4 A prospective study is designed to study takes place over several years. Furthermore, the National Toxicology subjects under pre-specified conditions. These Despite their limitation, the results of Program classified broad spectrum UV studies differ from retrospective studies that try to the short-term effectiveness studies are radiation as a known human carcinogen prove hypotheses by assessing past experiences. Generally, prospective studies are superior to consistent with our understanding that because it is not clear which UVB and/ retrospective studies in demonstrating drug measures which significantly reduce UV or UVA wavelengths contribute to the effectiveness. exposure decrease the risk of skin development of cancer (Ref. 15).

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35634 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

Therefore, available data indicate that a is to be used ‘‘for weight loss in plus cholesterol absorbed from foods broad spectrum sunscreen is necessary overweight adults, 18 years and older, containing fats. Therefore, if consumers to reduce the risk of skin cancer. when used along with a reduced-calorie use a statin and minimize the amounts Likewise, we do not know which UVB and low-fat diet’’ (Ref. 33). The of food containing fats in their diet, then and/or UVA wavelengths contribute to behavioral measure of reduced caloric they will reduce the total cholesterol early skin aging. Therefore, it is intake is necessary for consumers to level in the blood. However, if reasonable to conclude that reducing the experience weight loss. A low-fat diet is consumers do not minimize the risk of early skin aging also requires a necessary for consumers to avoid the amounts of food containing fats in their broad spectrum sunscreen (in undesirable side effect of diarrhea diet, they may not reduce the total conjunction with limiting time in the caused by consuming a high-fat diet cholesterol in the blood. The decreased sun and wearing protective clothing). while taking orlistat. cholesterol production in the body With regard to SPF value, the The need to include reduced caloric caused by the statin may not be available study data concerning the use intake as part of the indication significant compared to the high amount of sunscreens in reducing the risk of statement for orlistat is similar to the of cholesterol derived from food eaten skin cancer is based on products with need for including the use of other sun by consumers. SPF values of 15 or higher. The protection measures as part of the In the same way that regularly taking sunscreen product used in the 1999 indication statement for sunscreens. an OTC weight loss aid or a prescription Australian study on skin cancer Orlistat increases the likelihood of cholesterol-lowering drug product (squamous cell and basal cell weight loss by preventing fat from being without also following a healthy diet carcinomas) had a Broad Spectrum SPF absorbed as food is digested in the may not result in the intended health value of 16, and those that were found stomach and intestines. If consumers effect, use of a sunscreen with Broad to reduce actinic keratoses and nevi had take orlistat and decrease their caloric Spectrum SPF value of 15 or higher SPF values ranging from 16 to 46. The intake, they increase the likelihood of without also limiting time in the sun studies on early skin aging make it losing weight. However, if consumers and covering sun-exposed areas may not difficult to know for certain whether increase their caloric intake while result in a net reduction in the risk of Broad Spectrum SPF values of 15 or taking orlistat, they are less likely to developing skin cancer or early skin higher are necessary to reduce the risk lose weight. Orlistat’s effect of aging. For this reason, we are requiring of early skin aging. However, we preventing fat absorption could be offset that the indication statement allowed on conclude that the data regarding the by the high number of calories being sunscreens with Broad Spectrum SPF minimum sunscreen protection eaten. Similarly, the reduction in UV values of 15 or higher include all parts necessary to reduce the risk of skin exposure afforded by use of broad of the sun protection program and not cancer can be extrapolated to early skin spectrum sunscreens with SPF values of suggest or imply that use of a sunscreen aging. In many ways, the biological 15 or higher can be offset if consumers alone reduces the risk of skin cancer or processes that take place in response to increase their UV exposure by spending early skin aging. UV radiation are similar for both more time in the sun and/or wearing C. Warnings conditions. For both skin cancer and less protective clothing. This increased early skin aging, UV radiation causes overall exposure could eliminate the We received submissions requesting damage in the skin that is not effectiveness of sunscreen use in that we revise warnings included in the completely repaired and leads to cancer, reducing the risk of skin cancer and 2007 proposed rule and that we add fine lines, wrinkles, etc. Because the early skin aging. new warnings not included in the 2007 supporting data for a skin cancer claim The labeling of prescription proposed rule (Ref. 1). In section IV.C.1 are based on products with SPF values cholesterol-lowering drug products (i.e., of this document, we discuss one new of 15 or higher, we are only allowing the statins) follows a similar principle by and one revised warning included in skin cancer and early skin aging claim emphasizing that reduction of this final rule. We are adding the new for covered sunscreen products that are cholesterol levels requires not only use warning ‘‘Do not use on damaged or broad spectrum and have SPF values of of the drug product but also a healthy broken skin’’ (new 21 CFR at least 15. This rule does not preclude diet. The National Institutes of Health 201.327(d)(1)). We are revising the approval of a new drug application (NIH) specifies therapeutic lifestyle warning about skin rash (proposed 21 including an indication for reduction in changes that can be followed to lower CFR 352.52(c)(3)): ‘‘Stop use and ask a risk of skin cancer and early skin aging levels of cholesterol in the blood (Ref. doctor if skin rash occurs’’ to read ‘‘Stop for any sunscreen product. To be 34). These changes include following a use and ask a doctor if rash occurs.’’ approved, such an application must be diet restricted in saturated fat and In section IV.C.2 of this document, we supported by the submission of cholesterol, exercising regularly, and discuss our revision to the proposed adequate data. This rule also does not managing weight. Used in conjunction ‘‘Sun Alert’’ warning. Under this final preclude future amendment of the with cholesterol reducing drugs rule, the warning proposed for all sunscreen monograph in 21 CFR part (currently available only by monograph sunscreens is replaced with 352, if additional data are provided to prescription), these lifestyle changes an optional indication and required support a similar indication for other improve the chance of effectively direction on covered sunscreens with types of sunscreens. treating high cholesterol levels. Broad Spectrum SPF values of 15 or Prescription cholesterol-lowering higher, while covered sunscreens that e. Precedent for an Indication Statement drug products include the behavioral are broad spectrum with SPF values less That Includes Behavior Modification step of following a low fat diet in the than 15 or that do not provide broad There is at least one other OTC drug indication statement (Ref. 35). The body spectrum protection will bear a revised product with an indication statement produces cholesterol, which the drug warning, called the ‘‘Skin Cancer/Skin that describes not only the drug’s product inhibits to produce the desired Aging Alert.’’ (new 21 CFR intended effect but also one or more drug effect of lowering cholesterol being 201.327(d)(2)). behavioral measures to ensure the effect. made by the body. However, the total In section IV.C.3 of this document, we The indication statement on the weight cholesterol circulating in the blood discuss three new warnings that were loss aid orlistat states that the product reflects cholesterol made by the body requested in submissions, but are not

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35635

being included in this document. CFR 352.52(c)(3): ‘‘Stop use and ask a Broad Spectrum SPF value of 15 or Submissions argued that we should add doctor if skin rash occurs.’’ We are higher, are necessary for the consumers’ warnings that the regular use of deleting the word ‘‘skin’’ so that the sun protection as part of a sunscreen products may cause vitamin new warning reads: ‘‘Stop use and ask comprehensive program. D deficiency and may reduce the a doctor if rash occurs’’ (new 21 CFR For covered sunscreen products that photoprotective effects of tanning. We 201.327(d)(1)(iii)). We received two do not provide broad spectrum also considered adding a warning submissions arguing that the word protection or those that do provide concerning sunscreen products ’’skin’’ is unnecessary in this warning broad spectrum protection with SPF containing alpha hydroxy acids (AHAs). because every rash is a skin rash (Ref. values less than 15, we conclude that a We are not adding any of these warnings 1). We agree and are removing the word warning regarding the risks of skin because the available data do not to make the warning more concise. cancer and skin aging remains support the need for these warnings. Consumers will likely understand the necessary. In light of comments received In summary, this document requires warning without the word ‘‘skin.’’ on the ‘‘Sun Alert’’ warning proposed in the following warnings on all covered 2. Revision of the Proposed ‘‘Sun Alert’’ OTC sunscreen products (new 21 CFR 2007, however, we are revising the text Warning 201.327(d)): to read as follows: ‘‘Skin Cancer/Skin • ‘‘Do not use on damaged or broken In 2007, we proposed a warning, Aging Alert [in bold font]: Spending skin’’ based on the ‘‘Sun Alert’’ statement time in the sun increases your risk of • ‘‘Stop use and ask a doctor if rash cited in the 1999 stayed sunscreen final skin cancer and early skin aging. This occurs’’ rule (64 FR 27666 at 27679), as the first product has been shown only to help • ‘‘When using this product keep out statement under the heading Warnings prevent sunburn, not [in bold font] skin of eyes. Rinse with water to remove.’’ for all monograph sunscreen products cancer or early skin aging.’’ (new 21 CFR 201.327(d)(2). The title ‘‘Skin For all covered sunscreen products that regardless of SPF value or broad Cancer/Skin Aging Alert’’ more either are not broad spectrum or are spectrum protection (proposed 21 CFR broad spectrum with SPF values less 352.52(c)(1)). As proposed, this warning accurately and specifically conveys the than 15, this final rule also requires a would have stated, ‘‘UV exposure from nature of the warning that follows than ‘‘Skin Cancer/Skin Aging Alert’’ as the the sun increases the risk of skin cancer, the proposed ‘‘Sun Alert’’ warning, first statement under the heading premature skin aging, and other skin particularly since the products that will Warnings. In addition to these warnings, damage. It is important to decrease UV bear this statement are indicated to help all sunscreen products are required to exposure by limiting time in the sun, prevent sunburn, one consequence of include the ‘‘external use’’ and ‘‘keep wearing protective clothing, and using a sun exposure. The first sentence of this out of reach of children’’ warning sunscreen.’’ Submissions regarding this warning is a factual statement similar in statements required on all topical OTC proposed warning are discussed in content to the opening statement of the drug products (21 CFR 201.66(c)(5)(i) section IV.B.2 of this document. As warning proposed in 2007. Like the and (c)(5)(x)). noted there, we agree that, as proposed, proposed ‘‘Sun Alert’’ warning, this this warning included an implied statement alerts consumers to risks they 1. New and Revised Warnings for indication that all sunscreens reduce the continue to incur from sun exposure, Damaged or Broken Skin and Rash risk of skin cancer and skin aging. the conditions under which they will The new warning that we are Under this final rule, we are no longer make use of the product. The second requiring on all covered sunscreen drug requiring a ‘‘Sun Alert’’ or similar sentence clarifies for users the limits on products reads, ‘‘do not use on damaged warning on broad spectrum sunscreens the benefits that the product in hand has or broken skin.’’ We require this with SPF values of 15 or higher covered been established to provide, specifying warning or a similar warning for other by the rule. This decision is based on that these products have been shown to topical OTC drug products: our re-evaluation of the available help prevent sunburn but have not been • Acne treatments (21 CFR scientific data. We are now permitting shown to reduce the risk of skin cancer 333.350(c)(3)) an indication stating that, used as or early skin aging. Inclusion of this • Skin protectants (21 CFR directed with other sun protection warning is critical to help ensure that 347.50(c)(6)) measures, these sunscreens reduce the consumers do not mistakenly conclude • Antiperspirants (21 CFR risk of skin cancer and premature skin that all sunscreens have been 350.50(c)(1)) aging (new 21 CFR 201.327 (c)(2)). demonstrated to provide the same For these products we are also benefits. It will reinforce the distinction The safety data for these ingredients are requiring a new direction statement based on application to intact (i.e., between sunscreens indicated only for (new 21 CFR 201.327(e)(1)(iv)). The preventing sunburn (those that have unbroken or undamaged) skin. We do direction states: not have data of the safe use of these broad spectrum with SPF values below Sun Protection Measures. [in bold font] ingredients if the skin is not intact. For 15 or that are not broad spectrum) and Spending time in the sun increases your risk sunscreens that have also been shown to the same reason, the warning appears on of skin cancer and early skin aging. To sunscreen products marketed under decrease this risk, regularly use a sunscreen reduce the risk of skin cancer and early new drug applications (NDAs).5 with a Broad Spectrum SPF of 15 or higher skin aging when used as directed with Therefore, in this document, we are and other sun protection measures including: other sun protection measures (those requiring this warning for all covered [bullet] limit time in the sun, especially from with Broad Spectrum SPF values of 15 OTC sunscreen products, which are 10 a.m.–2 p.m. [bullet] wear long-sleeved or higher). This warning serves a similar marketed without approved shirts, pants, hats, and sunglasses purpose to one required on cosmetic applications (new 21 CFR We have concluded that information suntanning preparations that do not 201.327(d)(1)(i)). about decreasing sun exposure and contain a sunscreen ingredient, which In addition to the new warning, we wearing protective clothing is more likewise is intended to assist consumers are revising the warning in proposed 21 appropriate in ‘‘Directions’’ than in in distinguishing among products that ‘‘Warnings.’’ These measures, in they might otherwise confuse. (See 21 5 NDAs 21–501, 21–502, 21–471, and 22–009. addition to use of a sunscreen with CFR 740.19).

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35636 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

3. Warnings Requested in Submissions product contains an alpha hydroxy acid (3) clothing, (4) season, (5) latitude, (6) But Not Included in This Final Rule (AHA) that may increase your skin’s time of day, (7) weather conditions, and We considered adding the following sensitivity to the sun and particularly (8) sunscreen application (Refs. 40–43). three warnings: the possibility of sunburn. Use a Vitamin D deficiency has long been • Sunscreens may reduce the sunscreen and limit sun exposure while associated with Ricketts, but recent photoprotective effects of tanning using this product and for a week research suggests that vitamin D • Increased sun sensitivity caused by afterwards.’’ deficiency may also be associated with alpha hydroxy acids (AHAs) in Many cosmetic products containing other diseases (Ref. 38). However, the sunscreen products alpha hydroxy acids also contain threshold of vitamin D blood levels that • Regular use of sunscreen products sunscreens because the sunscreen helps constitutes a deficiency is currently may cause vitamin D deficiency protect the skin made sensitive to the being re-evaluated by scientific experts However, as discussed in this section of sun by the alpha hydroxy acids. The (Refs. 40, 44, and 45). the document, we conclude that these guidance does not address products To determine whether sunscreen use warnings are not needed for the safe and combining alpha hydroxy acids and causes vitamin D deficiency, we effective use of sunscreen products. sunscreens. reviewed four clinical studies included Two submissions stated that in the submissions that explored the We received a submission arguing additional warnings are not necessary that we should require the following effect of sunscreen use on vitamin D on these products (Ref. 1). We agree levels as well as three studies that we warning on all OTC sunscreen products with these submissions. We considered containing UVA-protective active identified in a literature search: added a warning or other labeling to • Matsuoka et al 1987 (Ref. 46) ingredients (Ref. 1): ‘‘The use of this inform consumers that AHAs contained • Matsuoka et al 1988 (Ref. 47) product will prevent the development of in some sunscreen products may make • Marks et al. 1995 (Ref. 48) photo-protective facultative the consumer more likely to sunburn. • Farrerons et al. 1998 (Ref. 49) pigmentation, a.k.a., a tan.’’ The However, the sunscreen component of • Kimlin et al., 2007 (Ref. 50) submission implies that UVA protection such products would, in fact, protect • Cusack et al., 2008 (ref. 51) is not only unnecessary but harmful to consumers from sunburn. Furthermore, • Hoesl et al., 2010 (Ref. 52). consumers. No data were included in we have concluded that the addition of All but one of these studies assessed 25- the submission. sunscreen active ingredients to AHA- hydroxyvitamin D levels because 25- We agree that tanning caused by UVA containing cosmetic products provides hydroxyvitamin D is typically used as radiation offers some protection against valuable UV protection for consumers. the biological marker for vitamin D (in sunburn. However, tanning, particularly Therefore, at this time, we have the D2 or D3 form) (Ref. 53). Much of the when attributable to prolonged exposure concluded that a warning about AHA is data available in the literature involves to UVA radiation in tanning beds or not necessary on OTC sunscreen nonclinical studies, which can be booths, may also have harmful effects products. difficult to extrapolate to consumer on the skin (Refs. 36 and 37). In The other new warning requested in (human) actual use conditions. Studies addition, one study suggests that the submissions relates to vitamin D with clinical data provide more protective effects of tanning are small, deficiency. We received six submissions meaningful results because, if as a tan only appears to provide an SPF arguing that consumers should be adequately designed, they can be more value of approximately 4 (Ref. 36). As warned that frequent sunscreen use may easily extrapolated to consumer actual stated in the 2007 proposed rule (72 FR result in vitamin D deficiency (Ref. 1). use conditions. Therefore, we are 49070 at 49083), we do not know which The submissions cite articles discussing focusing discussion in this document on UVA wavelengths cause specific types the negative effects of vitamin D the clinical studies. of damage (e.g., skin cancer or early skin deficiency, such as growth retardation, In the 1987 study by Matsuoka et al., aging). We continue to assert, however, rickets, and osteoporosis (Ref. 38). The four subjects applied a sunscreen that protection against UVA radiation is submissions include numerous product with an unknown SPF to the important for consumers’ health (72 FR published articles concerning vitamin entire body, while four control subjects 49070 at 49083). We have concluded D, but only four clinical studies that did not apply any topical product (Ref. that the warning suggested in the directly examine the effect of sunscreen 46). All of the subjects were exposed to submission is not in the best interest of use on vitamin D levels. In the 1 MED 6 of UV radiation (260–330 nm 7) public health because the warning remainder of this section, we discuss and then vitamin D3 levels were discourages consumers from using the four studies included in monitored for 15 days. The subjects broad spectrum sunscreen products. submissions, as well as three additional using sunscreen product applied the Therefore, we are not requiring any studies that we located through a sunscreen product 1 hour before UV warning related to tanning. literature search. Collectively, the exposure. The level of vitamin D was We are not adding any additional 3 studies do not demonstrate that the use determined one day before UV exposure warnings to sunscreen products of sunscreen causes vitamin D to serve as the baseline measure. containing AHAs. In the 2007 proposed deficiency. The level of vitamin D in the control rule, we requested comment on the need The term ‘‘vitamin D’’ refers to several 3 group (no sunscreen) increased for additional warnings or directions on forms of the vitamin, but the two forms significantly over baseline 1 day after sunscreen products combined with important to humans are vitamin D 2 UV exposure (from ∼2 ng/ml 8 to 25 ng/ AHAs (72 FR 49070 at 49110). We made (ergocalciferol) and vitamin D3 ml) and then decreased gradually, this request in response to a 2005 (cholecalciferol) (Ref. 39). Vitamin D2 is returning to baseline 15 days after UV guidance that we issued for cosmetic obtained by eating vitamin D-rich foods exposure. In contrast, the levels of products containing alpha hydroxy such as fish or food fortified with vitamin D3 in the sunscreen group did acids (70 FR 1721, January 10, 2005). vitamin D. The skin makes vitamin D3 The guidance recommends the when it is exposed to sunlight (Ref. 40) 6 MED refers to the lowest dose of UV radiation following warning be included on and, therefore, vitamin D production that produces perceptible reddening of the skin. cosmetic products containing alpha may vary depending on the following 7 Nanometers. hydroxy acids: ‘‘Sunburn Alert: This factors: (1) Skin pigmentation, (2) age, 8 Nanograms per milliliter.

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35637

not change significantly from the The control group of 19 subjects did not sun protection applied by these baseline level (5 ng/ml) at each time use sunscreen product, but had the patients’’ (Ref. 52). point. same skin characteristics. Mean serum These seven clinical studies are Based on this preliminary study, levels of 25-hydroxyvitamin D were inconclusive because the results were Matsuoka et al. conducted another measured at eight different time points contradictory. Two studies suggest that study in 1988 (Ref. 47). This study (four in the autumn and four in the sunscreens decrease vitamin D levels enrolled 40 subjects from Illinois and spring) over the two-year study period. and the other five studies suggest that Pennsylvania with 20 subjects in the The mean serum levels of 25- sunscreens do not decrease vitamin D control group and 20 subjects in the hydroxyvitamin D were statistically levels. In addition, the studies were sunscreen group. Each time they went lower in the sunscreen group as relatively small, only enrolling 8 to 126 outdoors for 1 year, the subjects in the compared to the control group at one subjects. The study with the greatest sunscreen group, who had a history of spring and one autumn time point number of participants was skin cancer, applied a sunscreen (p < 0.05). However, the mean serum inconclusive showing that people who product with an unknown SPF to all levels of 25-hydroxyvitamin D were not regularly used sunscreens and wore hats sun-exposed areas of the body. statistically different between the had increased levels of vitamin D, Serum 25-hydroxyvitamin D levels groups at the other 6 spring and autumn whereas people who regularly wore were measured in each group at the time points. pants outside had decreased levels conclusion of the study and were In 2007, Kimlin et al. reported that (Ref. 50). significantly lower in the sunscreen there was ‘‘no association’’ between use Because the preponderance of group than the control group: 40.2 and of sunscreens with SPF values higher currently available data suggests that 91.3 nmol/L,9 respectively. The than 15 and blood levels of 25- sunscreen use does not cause clinically difference in 25-hydroxyvitamin D hydroxyvitamin D in a study of 126 meaningful decreases in vitamin D levels between the two groups was Australian adults 18–87 years of age levels (i.e., decreases that lead to statistically significant (p < 0.001). (Ref. 50). However, the authors stated vitamin D deficiency and/or disease Marks et al. conducted a randomized, that mean levels of 25-hydroxy vitamin caused by low levels of vitamin D), we double-blind controlled clinical study D increased with increasing frequency are not including a warning regarding over a summer period in Australia (Ref. of sunscreen use. Interestingly, study vitamin D deficiency on OTC sunscreen 48). In this study, 113 subjects over 40 ‘‘participants who ‘usually’ or ‘almost products. In addition, determining years old who exhibited at least one always’ wore a hat when outdoors’’ whether decreases in vitamin D levels solar keratosis (a precursor of carcinoma were significantly more likely to have result in vitamin D deficiency is of the skin) were recruited and divided higher serum 25-hydroxy vitamin D especially difficult because the into two groups. The first group of 56 levels than those who wore hats less threshold of vitamin D blood levels that subjects applied an SPF 17 sunscreen often (Ref. 50). On the other hand, study constitutes a deficiency is currently cream. Fifty-five subjects in the control participants who usually or almost being re-evaluated by scientific experts group applied a placebo cream. Subjects always wore long sleeve shirts or pants (Refs. 38, 44, and 45). We recognize that in both groups were asked to apply their while outside were statistically more certain subpopulations may be at cream on the head, neck, forearm and likely to have lower serum 25- increased risk of vitamin D deficiency, dorsal side of each hand once a day in hydroxyvitamin D levels than those who as pointed out in one submission. the morning and more frequently if wore these types of protective clothing However, there are many factors that sweating, swimming, or involved in less often (Ref. 50). determine the amount of sun exposure activities that might rub off the cream. In 2008, Cusack et al. reported that necessary to ensure adequate vitamin D The mean levels of 25- decreased levels of 25-hydroxyvitamin levels (e.g., geographical location, hydroxyvitamin D rose significantly by D levels were only ‘‘weakly correlated’’ season, skin pigmentation, dietary almost the same amount in both groups with sunscreen usage in 52 Irish vitamin D intake). Because of these over the period of the study. The mean patients with cutaneous lupus many other factors, it is difficult for us level in the placebo group increased by erythematosus (Ref. 51). This study to determine a meaningful message in 12.8 mmol/L, whereas the mean level in population was specifically selected sunscreen product labeling for the sunscreen group increased by 11.8 because patients with lupus are consumers, especially in the absence of mmol/L. The difference between these particularly sensitive to exposure to the conclusive data. If we become aware of increases from baseline values was not sun. While an analysis of the effects of data from adequate and well-controlled statistically significant. daily sunscreen use on serum levels of studies demonstrating that regular use In 1998, Farrerons et al. carried out a 25-hydroxyvitamin D showed the of sunscreen causes vitamin D study to examine the effects of relationship between these two deficiency, we will re-evaluate this sunscreen use on vitamin D levels in parameters to be significant, a issue. elderly individuals (Ref. 49). In this 2- multivariate analysis of the same data year study, 24 subjects (10 men and 14 was not significant (Ref. 51). D. Directions women with a mean age of 71 years) Most recently, in 2010, Hoesl et al. We received numerous submissions were enrolled in the sunscreen group. reported ‘‘no statistically significant requesting that we revise directions The subjects had actinic keratosis, basal association’’ between serum levels of included in the 2007 proposed rule (Ref. cell carcinoma, or squamous cell 25-hydroxyvitamin D and use of the 1). In response to those requests and our carcinoma. None of the subjects had sunscreen drometrizole trisiloxane in a reevaluation of OTC sunscreen labeling, previously used sunscreen products, but cohort of 15 patients with Xeroderma we are revising the following directions: were instructed to apply an SPF 15 pigmentosum (Ref. 52). Like those with • ‘‘Reapply after [select one of the sunscreen product to sun-exposed areas lupus ertythematosus, patients with following: ‘40 minutes of’ or ‘80 minutes of the body each morning, avoid mid- Xeroderma pigmentosum are extremely of’ ’’ for products that satisfy either the day sun, and wear UV-protective sensitive to the sun. The authors water resistant or very water resistant clothing during the spring and autumn. reported that reductions in serum levels test procedures in proposed paragraphs of 25-hydroxyvitamin D are ‘‘not 352.76(a) and (b), respectively] 9 Nanomoles per liter. associated with any type or duration of swimming or [select one of the

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35638 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

following: ‘sweating’ or ‘perspiring’] and if swimming or sweating.’’ These product, such as swimming, toweling, after towel drying. Otherwise, reapply at revisions should clarify the directions. excessive sweating, or rubbing. No other least every 2 hours’’ (proposed 21 CFR We are removing reapplication reapplication time is suggested. The 352.52(d)(2)). directions concerning swimming and usefulness of this study in determining • ‘‘Reapply at least every 2 hours after sweating from non-water resistant whether to revise the directions is towel drying, swimming, or sweating’’ products because these products should limited. In particular, we do not know (proposed 21 CFR 352.52(d)(3)). not be used when swimming or whether this simulation was validated, These two directions are the sweating. Instead, we are requiring more because it has not been confirmed with reapplication instructions for water accurate directions instructing clinical studies. Until we receive resistant and non-water resistant consumers to use a different sunscreen clinical studies demonstrating that products, respectively. We also received product—a water resistant sunscreen consumers do not experience skin five submissions requesting that we product—if swimming or sweating. damage when sunscreen is reapplied at revise the direction: ‘‘Apply [select one We considered revising the 2-hour longer timeframes, we will continue to of the following: ‘liberally’ or reapplication timeframe because some require the 2-hour reapplication ‘generously’] [and, as an option: ‘and of the submissions objected to this timeframe. As discussed in the 1999 evenly’] [insert appropriate time specific timeframe (Ref. 1). The final rule, manufacturers may seek interval, if a waiting period is needed] submissions argued that we should approval of different reapplication before sun exposure’’ (proposed 21 CFR require the word ‘‘often’’ instead of a 2- directions by submitting specific and 352.52(d)(1)(i)). As discussed in this hour reapplication timeframe because substantive supporting data to us under section, we are not revising this there are no data supporting this an NDA deviation (described in 21 CFR direction statement. timeframe. The submissions also point 330.11). out that the American Academy of In addition to the revisions to these Dermatology (AAD) no longer supports 2. Proposed Directions Not Being provisions (described in more detail in a 2-hour timeframe, even though we Revised this section of the document), we are no cited AAD as supporting the 2-hour We are not revising proposed 21 CFR longer requiring the following proposed timeframe in the 2007 proposed rule (72 352.52(d)(1)(i): ‘‘Apply [select one of the direction: ‘‘Apply and reapply as FR 49070 at 49093). following: ‘Liberally’ or ‘generously’] directed to avoid lowering protection’’ In its submission following the 2007 [and, as an option: ‘And evenly’] [insert (proposed 21 CFR 352.52(d)(1)(ii)). proposed rule, the AAD does not state appropriate time interval, if a waiting As already discussed, for covered its support for the 2-hour timeframe. period is needed] before sun exposure.’’ sunscreen products with Broad However, all of the public education Several submissions requested that we Spectrum SPF values of 15 or higher, we materials from AAD instruct consumers allow ‘‘smoothly’’ to be included in this are requiring the following direction: to reapply sunscreen at least every 2 statement (Ref. 1). However, we ‘‘Sun Protection Measures. [in bold font] hours (Refs. 54 through 58). In addition, continue to consider this word to be Spending time in the sun increases your risk other public health organizations such vague (72 FR 49070 at 49072 and of skin cancer and early skin aging. To as the Centers for Disease Control and 49092). Some submissions also decrease this risk, regularly use a sunscreen Prevention (CDC) and the U.S. requested that we include a specific with a Broad Spectrum SPF of 15 or higher Environmental Protection Agency (EPA) application amount in place of the terms and other sun protection measures including: recommend reapplication at least every ‘‘generously’’ and ‘‘liberally’’ (Ref. 1). [bullet] limit time in the sun, especially from 10 a.m.–2 p.m. [bullet] wear long-sleeved 2 hours (Refs. 59 and 60). For example, the submissions suggested shirts, pants, hats, and sunglasses’’ We disagree with the submissions that the statement could read ‘‘apply 2 stating that data do not support this tablespoonsful.’’ The submissions (new 21 CFR 201.327(e)(1)(iv)). For timeframe. In the 2007 proposed rule, argued that more specific directions these products, this direction most we described two studies demonstrating would lead to consumers applying more appropriately conveys the information a significantly decreased sunburn risk if sunscreen product, reflecting the 2 proposed in the ‘‘Sun Alert’’ warning sunscreen product were applied at least milligrams per square centimeter (mg/ included in the 2007 proposed rule, and every 2 hours (72 FR 49070 at 49092 cm2) used during the SPF test. However, provides the necessary directions to through 49093). Wright et al. found that specifying a certain amount in the complement the new indication subjects who reapplied sunscreen every directions will not accomplish this goal. permitted for these products. 1 to 2 hours and after swimming were The amount of sunscreen product that In addition to the required directions, not sunburned (Ref. 61). Similarly, Rigel needs to be applied to reach 2 mg/cm2 we will allow the optional direction et al. reported that people who varies for each sunscreen product and heading ‘‘for sunscreen use’’ (new 21 reapplied sunscreen every two hours or depends on the amount of skin surface CFR 201.327(e)(1)(i)). sooner were five times less likely to area being covered. For example, the sunburn compared to those who 1. Revised Directions volume of sunscreen oil applied to the reapplied sunscreen only after 2.5 hours neck and face will differ greatly from We are revising the directions for or longer (Ref. 62). the amount needed to apply a sunscreen water resistant sunscreen products (new One of the submissions following the lotion to every sun-exposed area of the 21 CFR 201.327(e)(2)) to read: 2007 proposed rule included results body. Therefore, we are continuing to • Reapply: from a computer-simulation of require the terms ‘‘generously’’ and • After 40 [or 80] minutes of sunscreen product reapplication based ‘‘liberally.’’ swimming or sweating on a mathematical model (Ref. 1). The • Immediately after towel drying results of this simulation suggested that 3. Proposed Directions Not Being • At least every 2 hours sunscreen products should be reapplied Required We are also revising the directions for 15 to 30 minutes after sun exposure We are not requiring the proposed non-water resistant sunscreen products begins. The results also suggested that statement ‘‘apply and reapply as (new 21 CFR 201.327(e)(3)) to read: further reapplication of sunscreen directed to avoid lowering protection’’ ‘‘[Bullet] reapply at least every 2 hours product is necessary after vigorous (proposed 21 CFR 352.52(d)(1)(ii)). We [bullet] use a water resistant sunscreen activity that could remove sunscreen included this statement in the 2007

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35639

proposed rule because reapplication directly overhead). In our 1993 be permissible under the First time appears to be critical to achieve proposed rule, we stated that, ‘‘on any Amendment, in part, because, at that proper sun protection (72 FR 49070 at day of the year, the intensity of the UV time, we tentatively concluded that 49093). However, we have concluded energy of sunlight is greatest between 10 there were insufficient scientific data to that this statement is redundant with a.m. and 2 p.m.’’ (58 FR 28194 at support inclusion of such claims in the more specific reapplication directions 28199). We have concluded that this sunscreen monograph. As described and may confuse consumers. It is not information is important to consumers elsewhere in this document, we clear that consumers will understand trying to protect themselves from the received numerous submissions in the intent of this statement to emphasize sun and are including the information response to the 2007 proposed rule, the need to follow reapplication in the new direction statement. This some of which contained references to instructions. Therefore, we are not change is also responsive to the clinical studies we had reviewed in requiring the statement in this concerns of two submissions on the preparing the 2007 proposed rule about document. portion of the proposed sun alert that the effectiveness of sunscreens in referred to ‘‘limiting time in the sun,’’ 4. New Directions Resulting From protecting against the harmful effects of both of which suggested alternatives Submissions on the Proposed Rule UV radiation. As already described in intended to provide more concrete section IV.B.2, based in part on our re- For covered sunscreens with Broad information for consumers to act on evaluation of the data from these Spectrum SPF values of 15 or higher, we (Ref. 1). studies, as well as the scientific fact that are requiring a new Directions statement Several submissions argued that we reducing exposure to both UVB and that emphasizes the need not only to should allow different Drug Facts UVA radiation by a substantial amount regularly use such a sunscreen, but also labeling for cosmetics containing (i.e., equivalent to that provided by a to follow other sun protection measures. sunscreens so that consumers will apply broad spectrum sunscreen with an SPF For these sunscreens, the statement will the product appropriately for its value of 15 or higher) decreases the risk read, ‘‘[bullet] Sun Protection Measures. intended cosmetic use (Ref. 1). For of damaging the skin, we find that the [in bold font] Spending time in the sun example, the submissions argued that science supports the conclusion that increases your risk of skin cancer and reapplication every 2 hours may not be one subset of sunscreens covered by this early skin aging. To decrease this risk, appropriate for cosmetic-sunscreen rule, broad spectrum sunscreen regularly use a sunscreen with a Broad products. We disagree with these products with an SPF value of 15 or Spectrum SPF of 15 or higher and other submissions. Cosmetic-sunscreen higher, in conjunction with limiting sun protection measures including: combinations that are intended for use time in the sun and wearing protective [Bullet] limit time in the sun, especially as drugs require adequate labeling for clothing, reduce the risk of developing from 10 a.m.–2 p.m. [bullet] wear long- their drug use. (See 21 CFR 700.35). The skin cancer and early skin aging. Our sleeved shirts, pants, hats, and Drug Facts label communicates conclusion is reflected in the sunglasses (new 21 CFR information to the consumer so that the permissible indication described in this 201.327(e)(1)(iv)). This statement is cosmetic-sunscreen product can be used final rule for covered products with taken from the proposed warning ‘‘UV safely and effectively. To help Broad Spectrum SPF values of 15 or exposure from the sun increases the risk consumers understand that the higher. Although we have decided to of skin cancer, premature skin aging, sunscreen directions apply to the use of permit a claim about the prevention of and other skin damage. It is important the product as a drug, for sun skin cancer and early skin aging for to decrease UV exposure by limiting protection, we are allowing the optional certain covered sunscreens, as requested time in the sun, wearing protective statement ‘‘for sunscreen use:’’ to appear in the submissions, we have clothing, and using a sunscreen.’’ as the first line under ‘‘Directions.’’ nevertheless conducted a First (proposed 21 CFR 352.52(c)(1)). As Consumers who are using these Amendment analysis of our discussed in section IV.C. of this products primarily for cosmetic use will requirements concerning the skin document, this warning is no longer be more likely to understand that they cancer/early skin aging claim in this being required for sunscreens with might not receive the intended sun final rule (hereinafter ‘‘skin cancer/early Broad Spectrum SPF values of 15 or protection if they do not follow the aging indication’’), as well as the ‘‘Skin higher. Rather, as discussed in section directions in the Drug Facts label. IV.B of this document, submissions Cancer/Skin Aging Alert’’ required as a suggested that the information proposed E. Constitutionality of Labeling warning for covered products that do as a warning is better understood as an Statements Regarding Skin Cancer and not provide broad spectrum protection indication, with the supporting Skin Aging with an SPF value of 15 or higher. For conditions for achieving effectiveness. Two submissions questioned the the following reasons, we have As described in section IV.B, on constitutionality of the labeling concluded that these requirements do reexamination of the scientific data, we provisions in the 2007 sunscreen not violate the First Amendment. agree that an appropriately limited proposed rule. Specifically, the This rule establishes effectiveness indication for reduction in risk of skin submissions contended that our testing methods and labeling that are cancer and early skin aging is supported proposed restriction on any claims appropriate for the safe and effective use for sunscreens with Broad Spectrum about the prevention of skin cancer, of OTC sunscreen products covered by SPF values of 15 or higher. For these early skin aging, and related skin this rule. Any covered sunscreen products, the direction instructs users damage would violate the sunscreen product that deviates from the how to use the product in a manner that manufacturers’ commercial speech requirements set forth in this labeling supports that indication. rights under the First Amendment to the regulation and any other applicable In this final rule, we are being more U.S. Constitution. labeling regulation would be considered specific about the need to limit time in In the 2007 proposed rule preamble, misbranded under section 502 of the the sun especially during the midday we had concluded that our proposed FD&C Act. In particular, sunscreen hours of 10 a.m. to 2 p.m. when the restriction on claims about the products covered by this rule would be intensity of solar radiation is greatest prevention of skin cancer, early skin misbranded if they are labeled with a because the sun is at its zenith (i.e., aging, and related skin damage would skin cancer/early aging indication but

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35640 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

do not provide broad spectrum For covered sunscreen products that between means and ends) (Board of protection with an SPF value of 15 or do provide broad spectrum protection Trustees of the State Univ. of N.Y. v. higher. Such products would also be with an SPF value of 15 or higher, we Fox, 492 U.S. 469, 480 (1989)). In misbranded if they do not include the are requiring that the additional subsequent decisions, the Court has also ‘‘Skin Cancer/Skin Aging Alert’’ information about using the product as clarified that ‘‘misleading’’ in the first described in this rule (see 21 CFR directed with other sun protection element of the test refers to speech that 201.327(d)(2)). Covered sunscreen measures be included in the indication is inherently or actually misleading. products that do provide broad so that consumers are not misled about Based on the data currently available, spectrum protection with an SPF value how to use these sunscreens effectively we have concluded that the following of 15 or higher would be misbranded if or about the conditions under which statements or omissions would be false they are labeled with the permissible these sunscreens are effective. Use of a or inherently misleading: (1) Use of the skin cancer/early aging indication but sunscreen alone—even a broad skin cancer/early aging indication on do not include reference to the need to spectrum sunscreen with an SPF value the labeling of a sunscreen product that use the product as directed with other of 15 or higher—has not been shown to does not provide broad spectrum sun protection measures (21 CFR reduce the risk of skin cancer or early protection with an SPF value of 15 or 201.327(c)(3)). Manufacturers of covered skin aging if a consumer increases higher, (2) the omission of the ‘‘Skin sunscreen products that comply with overall UV exposure by spending greater Cancer/Skin Aging Alert’’ under the the labeling requirements in this time in the sun and/or wearing less ‘‘Warnings’’ heading of the labeling for document would not be subject to protective clothing. The additional sunscreen products that do not provide enforcement actions on the basis that information required in the skin cancer/ broad spectrum protection with an SPF the products are misbranded, provided early aging indication about using the value of 15 or higher, and (3) use of the they comply with all other requirements product as directed with additional sun skin cancer/early aging indication that under section 502 of the FD&C Act. protection measures clarifies how the omits the required information about Because this rule applies only to use of sunscreens is part of a using the product as directed with other products marketed without approved comprehensive sun protection program. sun protection measures. applications, manufacturers who wish Displaying this information elsewhere Use of the skin cancer/premature to deviate from the testing or labeling would underemphasize its importance aging indication on the labeling of requirements in this document may do in relation to the use of these sunscreens covered sunscreen products that do not so by means of a new drug application for protection against skin cancer and provide broad spectrum protection with (NDA) under section 505 of the FD&C early skin aging (see N.Y. State Rest. an SPF value of 15 or higher would be Act. Ass’n v. N.Y. City Bd. of Health, 556 false or inherently misleading for several reasons. As discussed elsewhere We have concluded that the labeling F.3d 114 (2d Cir. 2009); see also 21 in this document, only broad spectrum requirements in this rule satisfy the U.S.C. 352(c)). Thus, these disclosure UV radiation is classified as a known applicable tests governing commercial requirements will promote the proper human carcinogen, according to the speech, as set forth by the Supreme use of covered sunscreens and are, National Toxicology Program. Court. The requirements for the ‘‘Skin therefore, reasonably related to the Therefore, covered sunscreen products Cancer/Skin Aging Alert’’ and the Government’s interest in protecting that do not provide broad spectrum UV information in the skin cancer/early public health. aging indication about using the product protection may not reduce the risk of as directed with other sun protection Our requirements concerning the skin skin cancer. Furthermore, since the measures, are permissible under the cancer/early aging indication would precise wavelengths of UV radiation First Amendment because they are also be permissible under the First that cause skin cancer and early skin reasonably related to the Government’s Amendment using the analytical aging are unknown, a covered sunscreen interest in protecting public health (see framework provided in Central Hudson product that only provides protection Zauderer v. Office of Disciplinary Gas & Electric Corporation v. Public against part of the UV spectrum may not Counsel, 471, U.S. 626, 651 (1985)). Service Commission, 447 U.S. 557 ensure a reduction in the risk of We are requiring covered sunscreen (1980). Under Central Hudson, developing skin cancer or early skin products that do not provide broad commercial speech that is false, aging. In addition, all of the scientific spectrum protection with an SPF value misleading, or concerns unlawful data that support the skin cancer/early of 15 or higher to include the ‘‘Skin activity is not entitled to protection aging indication for certain covered Cancer/Skin Aging Alert’’ under the under the First Amendment. While sunscreens were derived from studies ‘‘Warnings’’ heading on the label to commercial speech that concerns lawful that used sunscreen products with an ensure that consumers are aware of the activity and is not misleading receives SPF value of 15 or higher. Therefore, the continued risks of skin cancer and early some protection under the First skin cancer/early aging indication skin aging that occur from sun exposure, Amendment, it may nonetheless be would be false or inherently misleading the conditions under which they will be regulated by the Government if the on covered sunscreen products that do using the product, and that they following conditions are met: (1) The not provide this level of protection, understand that the product has been asserted governmental interest is because there is a lack of any evidence shown only to help protect against substantial; (2) the regulation directly demonstrating that these products sunburn. Without this warning, advances the asserted governmental would reduce the risk of skin cancer or consumers could fail to distinguish interest; and (3) the regulation is not early skin aging. Similarly, omitting the between these sunscreen products and more restrictive than necessary to serve ‘‘Skin Cancer/Skin Aging Alert’’ on other sunscreen products that have been that interest (Id. at 566). The Supreme these products, which are identified on proven to help provide protection Court has explained that the last their labels as ‘‘sunscreens,’’ would be against skin cancer and early skin aging. element of the Central Hudson test is inherently misleading because Providing this information is important not a ‘‘least restrictive means’’ consumers who are using these products for consumers to be able to make requirement but, rather, requires narrow for sun protection would not be informed choices about the selection tailoring (i.e., ‘‘a fit that is not sufficiently alerted to the fact that these and use of sunscreens. necessarily perfect, but reasonable’’ products have been shown only to

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35641

protect against sunburn, while sun supports the skin cancer/premature covered by this rule have several exposure also increases the risks of skin aging indication for sunscreen products alternatives for adding labeling cancer and early skin aging. that provide broad spectrum protection information that is not included in this A skin cancer/early aging indication with an SPF value of 15 or higher. labeling regulation. For example, such on a covered product with Broad Allowing the skin cancer/early aging manufacturers can file an NDA under Spectrum SPF value of 15 or higher that indication on sunscreen products for section 505 of the FD&C Act or submit omits the required information about which it is not scientifically supported a petition under 21 CFR 10.30 to amend using the product as directed with other would lead to consumers unjustifiably the labeling regulation. In either case, sun protection measures would also be relying on such products for protection the manufacturer need only submit the false or inherently misleading because against skin cancer and early skin aging. requisite evidence to support the sunscreen use alone has not been shown Furthermore, the ‘‘Skin Cancer/Skin indication or other labeling for the to reduce the risk of skin cancer or early Aging Alert’’ allows consumers to be product that differs from that addressed skin aging if a consumer increases aware that spending time in the sun by the regulation. Therefore, we are not overall UV exposure by spending greater increases their risk of skin cancer and being more restrictive than necessary time in the sun and/or wearing less early skin aging, and that products on when these viable alternatives are protective clothing. As discussed above which this alert appears have not been available for manufacturers. in this section and elsewhere in this shown to provide this type of Reacting to the fact that our proposed document, without the reduction in protection. The requirement for rule did not permit any indication consumers’ overall UV exposure, a information in the skin cancer/early statement for any sunscreen regarding covered sunscreen product may not be aging indication about using sunscreens prevention of skin cancer and early skin effective in reducing consumers’ risk of as directed with sun protection aging, one submission asserted that we skin cancer and early skin aging. measures also directly advances our must consider use of a disclaimer as an We also conclude that the labeling interest in protecting public health alternative means of addressing the claims and omissions described above because these elements are essential for limits of the product’s effectiveness. As would cause the product to be consumers to reduce their overall UV noted previously in this document, this misbranded and, therefore, relate to an exposure and, consequently, their risk final labeling regulation permits an unlawful activity. As described earlier of developing skin cancer and early skin appropriately limited indication for in this section and elsewhere in this aging. Thus, these requirements directly broad spectrum sunscreens with SPF document, labeling regulations establish advance the Government’s interest in values of 15 or higher—one stating that certain requirements that help ensure protecting public health through the when used as directed with other sun the safe and effective use of OTC drug safe and effective use of sunscreens. protection measures, such products products. The false or misleading Under the final remaining step of the decrease the risk of skin cancer and labeling described above would cause Central Hudson test, our requirements early skin aging caused by the sun. The covered products to be misbranded concerning the skin cancer/early aging claim is authorized for this subset of under section 502 of the act. Therefore, indication are not more restrictive than covered sunscreen products because such labeling would concern the illegal necessary, because there are not available scientific data discussed sale of misbranded drugs. Under the numerous and obvious alternatives elsewhere in this document are Central Hudson test, then, we have not (Cincinnati v. Discovery Network, 507 sufficient to substantiate the claim for violated the First Amendment with U.S. 410, 418 n. 13 (1993)) to achieve these products. Because we have these requirements, which simply the Government’s substantial interests. included a skin cancer/early skin aging prohibit false or inherently misleading By permitting the skin cancer/early claim in these labeling regulations, we labeling. aging indication only for covered no longer view the submission’s request Although we conclude that the sunscreen products with Broad as being applicable. labeling described above would not be Spectrum SPF values of 15 or higher, In any event, we note that the use of entitled to First Amendment protection and requiring the ‘‘Skin Cancer/Skin disclaimers on drug labeling to qualify under the threshold inquiry of the Aging Alert’’ for products that do not inadequately supported or unapproved Central Hudson test, we conclude that offer this level of protection, we are indications is not an effective, less our regulation directly advances a ensuring that consumers do not restrictive means of achieving FDA’s substantial Government interest and is mistakenly rely on sunscreen products substantial interests in protecting public no more extensive than necessary, and that have not been demonstrated to be health and preserving the integrity of its therefore would also pass muster under effective for protection against skin premarket approval systems. Indeed, the test’s three remaining steps. Under cancer and early skin aging. In addition, disclaimers on drug labeling would the first remaining step, we have a labeling that omits a statement regarding severely undermine the Government’s substantial interest in protecting public the use of other sun protection measures interests here. For over 100 years, health (see Pearson v. Shalala, 164 F.3d as directed from the skin cancer/early Congress has charged FDA with 650, 656 (DC Cir. 1999) (citing Rubin v. aging indication could lead to enforcing misbranding laws to protect Coors Brewing Co., 514 U.S. 476, 484– consumers foregoing other sun public health. In 1962, Congress 485 (1995)). protection measures, thereby negating amended the FD&C Act to require that Under the second remaining step of the protective effect of the sunscreen. all new drugs be approved as both safe the Central Hudson test, our labeling Including a statement in the skin and effective prior to marketing. requirements discussed in this section cancer/early aging indication regarding Congress found that a premarket directly advance the Government’s the need to follow other sun protection approval system, requiring specific interests in protecting public health measures as well as the related types of supporting evidence (see 21 because they help ensure that covered directions ensures that consumers U.S.C. 355(d)), and misbranding sunscreen products are adequately understand how to use sunscreens to provisions, among other requirements, labeled for safe and effective use by reduce their risk of skin cancer and were necessary to avoid further consumers. early skin aging. tragedies involving unsafe and As stated previously in this It is important to note that ineffective drugs. Using disclaimers for document, scientific evidence only manufacturers of OTC sunscreens drugs would completely undermine the

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35642 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

regulatory framework established by covered by the rule—those that provide which meet the requirements in 21 CFR Congress for the protection of public broad spectrum protection with an SPF 201.66(d)(10)) that are labeled for use health. FDA’s labeling regulations help value of 15 or higher, and those that do only on small areas of the face. One ensure the safety and effectiveness of not. We do not agree that this submission stated that all cosmetic OTC drugs and establish the conditions information belongs under the heading products labeled with sunscreen under which a drug is not misbranded ‘‘Other information’’ but have included indications should be required to under the FD&C Act. If a manufacturer it in modified form under the headings include all sunscreen product labeling. of a covered sunscreen would like to Uses and Directions for products with After reassessing the criteria for label its sunscreen product in a way that Broad Spectrum SPF values of 15 or reduced labeling, we are not allowing does not conform to this labeling higher (new 201.327(c)(2) and (e)(2), the reduced labeling included in the regulation, it cannot circumvent the and under a revised ‘‘Skin Cancer/Skin 2007 proposed rule. OTC drug labeling premarket NDA process. Aging Alert’’ under the heading regulations (21 CFR 201.66(d)(10)) allow In summary, we conclude that the Warnings for other sunscreens (new reduced labeling for any OTC drug labeling requirements provided in this 201.327(d)(2)). product sold in a small package, document do not violate the First In this document, we are also including sunscreen products. In the Amendment. removing the optional ‘‘Other 2007 proposed rule, we proposed additional reductions in labeling for F. Other Information information’’ statements in proposed 21 CFR 352.52(e): three types of sunscreen products sold We received submissions requesting 1. ‘‘Low,’’ ‘‘medium,’’ ‘‘high’’ or in small packages and intended for use that we add a new statement about ‘‘highest’’ ‘‘sunburn protection product’’ on small areas of the face: • storage conditions under ‘‘Other 2. ‘‘Higher SPF products give more Proposed 21 CFR 352.52(f)(1)(i)– information’’ in the Drug Facts label sun protection, but are not intended to (f)(1)(iv): Sunscreen products sold in (Ref. 1). The submissions argued that extend the time spent in the sun.’’ small packages and labeled for use sunscreen products in containers are specifically on the lips, nose, ears, and/ According to the 2007 proposed rule, often exposed to heat when used at the or around the eyes (i.e., small areas of these statements could appear in ‘‘Other beach, swimming pools, etc. The the face) concern expressed in the submissions information’’ or anywhere outside Drug • Proposed 21 CFR 352.52(f)(1)(v): was that heat could cause sunscreen Facts. However, in this rule, we have Sunscreen-lip protectant combination formulations inside containers to revised the labeling and are no longer products sold in small packages change, resulting in less sun protection. requiring the principal display panel to • Proposed 21 CFR 352.52(f)(1)(vi): We agree with the submissions. characterize the level the sunburn Sunscreen products formulated as Sunscreen products within containers protection. Rather, for broad spectrum lipsticks, lip products that prolong wear should not be exposed to direct sun and products, the rule requires only the of lipstick, lip gloss, and lip balms can be protected by wrapping them in statement ‘‘Broad Spectrum SPF [fill in Three submissions argued that we towels and/or keeping them in shaded tested SPF value]’’ to appear on the should not restrict labeling exemptions environments (e.g., under an umbrella principal display panel. In light of this only to sunscreen products sold in small and/or in a purse or bag). Consumers revised approach to labeling, we are packages and labeled for use on small could also store sunscreen product concerned that including the areas of the face. The submissions stated containers in coolers while outside characterizations of the product as that reduced labeling provisions should during hot periods. In this final rule we providing ‘‘low,’’ ‘‘medium,’’ ‘‘high’’ or apply to all sunscreen products sold in are requiring the following statement in ‘‘highest’’ ‘‘sunburn protection would be small packages whether or not they are the ‘‘Other information’’ section of the confusing or misleading, and are no labeled for use on small parts of the Drug Facts label: ‘‘[Bullet] protect the longer including it as an option. face. Two of the submissions argued product in this container from excessive We have concluded that the second that such a restriction violates the heat and direct sun’’ (new 21 CFR statement, although truthful, is not Administrative Procedures Act (APA). 201.327(f)). necessary. Consumers likely understand The submissions cite Bracco In addition to the statement about the first part of this statement (higher Diagnostics, Inc., v. Shalala 963 F. storage conditions, we received SPF values represent more sun Supp. 20, 27–28 (D.D.C. 1997) as numerous submissions requesting that protection) based on the long-standing evidence that the courts oppose we relocate the proposed ‘‘sun alert’’ inclusion on SPF values on OTC regulations requiring ‘‘two sets of warning to the ‘‘Other information’’ sunscreen products. The second part of similar products to run down two sets section of the Drug Facts label. The the statement (higher SPF products are of separate [regulatory] tracks * * * for submissions argued that the ‘‘sun alert’’ not intended to extend time spent in the no apparent reason.’’ is an educational statement and not a sun) is redundant with the information In this document, we continue to warning: ‘‘UV exposure from the sun already provided under ‘‘Uses’’ and allow the reduced labeling specified in increases the risk of skin cancer, ‘‘Directions,’’ particularly concerning 21 CFR 201.66(d)(10). Therefore, if the premature skin aging, and other skin the need for limiting time in the sun information listed under Drug Facts damage. It is important to decrease UV (see sections IV.B and IV.D). Although requires more than 60 percent of the exposure by limiting time in the sun, we are not requiring inclusion of the total available surface area, the Drug wearing protective clothing, and using a second statement under ‘‘Other Facts labeling can be reduced by making sunscreen.’’ information,’’ the statement may appear the formatting changes specified in 21 As already discussed, in light of our outside the Drug Facts label because it CFR 201.66(d)(10)(i)–(d)(10)(v). re-evaluation of the evidence supporting is truthful and nonmisleading. However, in contrast to the 2007 the indications for sunscreens, we have proposed rule, we are not allowing G. Reduced Labeling made changes to the labeling to more additional reductions in labeling for any accurately convey appropriate Five submissions requested changes sunscreen products. information to consumers about the to our proposed regulations allowing When we proposed the additional benefits, directions, and limitations of reduced labeling for sunscreen products reduced labeling, we recognized that two different groups of products sold in small packages (i.e., packages many of the sunscreen products sold in

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35643

small packages and labeled for use on through 27680). These terms are also did not receive any data regarding small areas of the face could not essentially exaggerations of performance claims for extended periods of use, so it accommodate full Drug Facts labeling. that FDA does not think can be is not clear whether these claims are However, in the last several years, substantiated. Accordingly, in this final truthful. Claims that a product provides manufacturers have introduced new rule, we codify these as terms or phrases for an extended period of protection label designs that permit full Drug Facts that would be false or misleading on between applications or immediately labeling on very small packages. For covered products, and are therefore upon application would have to be example, some stick products, including prohibited (21 CFR 201.327(g)). supported by data. Therefore, these lip protectant-external analgesic In addition to submissions requesting claims could be made only under combinations marketed in 0.15 oz. that we allow certain labeling outside approved new drug applications (NDAs) amounts, have been labeled with wrap- Drug Facts, we also received a with the required data. around labels that contain full Drug submission requesting that we require In this document, we are specifically Facts labeling. If these products can be information about the UV index (UVI). identifying these claims as not allowed labeled to accommodate full Drug Facts A stated in the 2007 proposed rule, we on any OTC sunscreen product, labeling, then all sunscreen products have determined that the usage regardless of SPF value or broad should be able to accommodate full information provided on OTC sunscreen spectrum protection, without an Drug Facts labeling. Requiring full Drug products applies regardless of the UVI approved application containing Facts labeling should not discourage value (72 FR 49070 at 49073). Therefore, sufficient substantiation to support the manufacturers from including sunscreen we will allow but do not require claim. (new 21 CFR 310.545(a)(29)(ii)): ingredients because of limited labeling information about the UV index to be • Instant protection or protection space, as stated in the 2007 proposed included on sunscreen products outside immediately upon application rule (72 FR 49070 at 49075 through the Drug Facts label. • Claims for ‘‘all-day’’ protection or 49077). Therefore, in this document, we A submission requested that we extended wear claims citing a specific are eliminating all of the allowances for require that the UV index appear on number of hours of protection that are reduced labeling in proposed 21 CFR sunscreen product labels because this inconsistent with the directions for 352.52(f). Sunscreen products can only information would help consumers application in 21 CFR 201.327. have reduced labeling for formatting if understand and use the UV index to In addition, we are identifying the they meet the criteria in 21 CFR determine their risk of sunburn. The UV terms ‘‘sunblock’’ ‘‘waterproof,’’ and 201.66(d)(10). index was developed in 1995 by the ‘‘sweatproof’’ as false and misleading, as National Weather Service, we have stated in previous sunscreen V. Miscellaneous Labeling Outside Drug Environmental Protection Agency, and Facts rulemakings: Centers for Disease Control and • Sunblock (64 FR 27666 at 27679 We received several submissions Prevention to provide a forecast of the and 27680) regarding various performance claims, expected risk of overexposure to UV • Sweatproof (58 FR 28194 at 28227 including comments asking us to allow rays. The UV index is calculated using through 28228) claims for protection immediately upon ozone data, atmospheric pressure, • Waterproof (58 FR 28194 at 28227 application (instant protection) and for temperature, and cloudiness. As stated through 28228). extended duration between applications in the 2007 proposed rule, we are not We have previously identified these (extended wear) and comments asking requiring labeling of UV index claims as ones that would render a us not to allow terms such as information because it is not necessary product misbranded but are addressing ‘‘sunblock,’’ ‘‘waterproof,’’ and for consumers to understand this index them again in this document because ‘‘sweatproof’’ (Ref. 1). These kinds of in order to safely and effectively use OTC sunscreen products currently claims were not included in the 2007 OTC sunscreen products (72 FR 49070 marketed without approved proposed rule (Ref. 1). at 49073). However, manufacturers may applications continue to contain the We are not including labeling in 21 include truthful and nonmisleading claims. In this final rule, we are listing CFR 201.327 permitting these claims on information about the UV index in the these false and misleading terms in 21 OTC sunscreen products covered by the labeling outside of Drug Facts if they CFR 201.327(g). These terms may not be rule. The current record does not choose. included on any OTC sunscreen contain support for any of these kinds We also received a submission products covered by the rule. of claims. To clarify the status of these requesting that we allow a claim of Finally, in the 2007 proposed rule, we kinds of claims, we are finalizing two ‘‘instant protection’’ and to allow claims proposed to specify other optional provisions. We include instant for extended periods of protection statements that could be included protection and extended wear claims, between applications (i.e., longer than outside of Drug Facts in proposed 21 which are claims that we think may be the 2 hours specified in ‘‘Directions’’ in CFR 352.52(e)(3): capable of substantiation, in 21 CFR the 2007 proposed rule). The • ‘‘Broad spectrum sunscreen’’ 310.545(a)(29)(ii). While these claims submission argued that several • ‘‘Provides [select one of the may not be included on products marketed products provide sunburn following: ‘UVA and UVB’ or ‘broad marketed without approved protection immediately upon spectrum’] protection’’ applications, including them in this application, as demonstrated by test • ‘‘Protects from UVA and UVB provision makes it clear that these results included in the submissions. In [select one of the following: ‘rays’ or claims may be substantiated for an this document, SPF testing requires a ‘radiation’]’’ individual product by the submission of 15-minute waiting period between • ‘‘[Select one of the following: adequate data in an NDA. sunscreen application and UV exposure ‘absorbs’ or ‘protects’] within the UVA We agree with the submissions that of the test site. It appears that the spectrum.’’ argue that ‘‘sunblock,’’ ‘‘waterproof,’’ submitted test method included the This final rule is not a monograph, and ‘‘sweatproof’’ claims are false or same 15-minute waiting period. and we do not consider it necessary in misleading, as we have stated in Therefore, the assertion that this this rule to codify optional statements previous sunscreen rulemakings (58 FR product provides ‘‘instant protection’’ for use outside of ‘‘Drug Facts.’’ The 28194 at 28228; 64 FR 27666 at 27676 does not appear to be substantiated. We labeling required in this document

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35644 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

should provide consumers with the would misbrand that product. We requesting that we revise additional test information that they need to safely and likewise caution against references to parameters (Ref. 1). In this document, effectively use the sunscreen products ‘‘UVA’’ (or ‘‘UVA/UVB’’) protection on we have rewritten the regulations that it addresses. Under this final rule, products that do not provide broad describing the SPF test in an effort to products marketed without approved spectrum protection as demonstrated by make it easier to read and understand applications that provide broad the test in 21 CFR 201.327(j). Such and to more closely follow the order in spectrum protection according to the labeling would misbrand the products if which steps of the SPF testing test in new 21 CFR 201.327(j) of this it misleadingly suggests that the procedure are conducted. We have also document will be identified on the PDP products provide protection that is made several revisions to the test equivalent or greater to that provided by by use of the term ‘‘Broad parameters. However, we did not make products labeled with ‘‘Broad Spectrum all of the revisions requested in the Spectrum SPF.’’ In light of this SPF’’ values or is otherwise false or requirement in the rule for use of the misleading. submissions. Table 4 of this document term ‘‘broad spectrum’’ on these summarizes test parameters that we particular products, including a VI. SPF Test Parameters considered revising. The table identifies statement anywhere in the labeling of a The 2007 proposed rule included the the parameters that we are changing in product that does not pass the broad SPF test from the 1999 final rule with this document as well as those that we spectrum test in 21 CFR 201.327(j) that revisions to a few test parameters. In are not changing. Detailed discussion of suggests or implies that the product response to the 2007 proposed rule, we each test parameter appears throughout provides broad spectrum protection received numerous submissions the remainder of this section.

TABLE 4—SUMMARY OF SPF TEST PARAMETERS INCLUDED IN THE 2007 PROPOSED RULE AND THIS FINAL RULE

2007 Proposed rule This final rule

21 CFR 352.70(a). Standard sunscreens 21 CFR 201.327(i)(2). SPF standard Two standards: One standard: 8% homosalate (SPF 2— ≤ 15) 7% padimate, 3% oxybenzone (all SPFs) 7% padimate, 3% oxybenzone (SPF > 15) HPLC reference standard: HPLC reference standard: no limits set for accuracy of oxybenzone & padimate O limit set to within 5% of theoretical for accuracy of oxybenzone & padimate O 21 CFR 352.70(b). Light source (solar simulator) 21 CFR 201.327(i)(1). UV source (solar simulator) Emission spectrum specifications: Emission spectrum specifications: (1) COLIPA 1 1994 (Ref. 63) (1) COLIPA 1 2006 (Ref. 64) (2) no specifications for UVA (2) specifications for UVA I and UVA II percentages of total UV Calibration: Calibration: every 6 months at least annually Total irradiance: Total irradiance: 1500 Watts/square meter (W/m2) 1500 Watts/square meter (W/m2) Beam uniformity: Beam uniformity: within 20 percent within 20 percent 21 CFR 352.70(c)(7). Number of subjects 21 CFR 201.327(i)(3). Test subjects SPF < 30: All SPFs: 20–25 subjects; ≥ 20 valid results • 10–13 subjects; ≥ 10 valid results SPF ≥ 30: 25–30 subjects; ≥ 25 valid results 21 CFR 352.70(c)(4). Test site delineation/subsite 21 CFR 201.327(i)(4)(i) and (ii). Test site/subsite test site area: test site area: ≥ 50 cm2 ≥ 30 cm2 test subsite area: test subsite area: ≥ 1 cm2 ≥ 0.5 cm2 Distance between subsites: Distance between subsites: ≥ 1 cm ≥ 0.8 cm 21 CFR 352.70(c)(5). Application of test materials 21 CFR 201.327(i)(4)(iii). Applying test materials Application amount: Application amount: 2 milligrams per square centimeter (mg/cm2) 2 milligrams per square centimeter (mg/cm2) Presaturation of finger cot: Presaturation of finger cot: Required not required Water-resistant statement requirements: Water-resistant statement requirements: 20 minute water immersion times 20 minute water immersion times 20 minute drying times 15 minute drying times 21 CFR 352.70(d)(3). Determination of individual SPF values 21 CFR 201.327(i)(5). UV exposure Definitions of MED: Definitions of MED: (1) MED(PS) = MED for protected skin (1) ssMEDp = MED for skin protected by sunscreen standard (2) MED(US) = MED for unprotected skin (2) tpMEDp = MED for skin protected by test product (3) initial MEDu = MED for unprotected skin prior to testing test product (4) final MEDu = MED for unprotected skin determined when test- ing test product UV doses for MED(US): UV doses for initial MEDu:

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35645

TABLE 4—SUMMARY OF SPF TEST PARAMETERS INCLUDED IN THE 2007 PROPOSED RULE AND THIS FINAL RULE— Continued

2007 Proposed rule This final rule

five doses number of doses not specified 21 CFR 352.70(c)(8) Response criteria 21 CFR 201.327(i)(5). UV exposure Maximal UV exposure: Maximal UV exposure: ‘‘no more than twice the total energy of the minimal exposure’’ not specified 1 Draft test method entitled ‘‘International Sun Protection Factor (SPF) Test Method’’ developed by the European Cosmetic, Toiletry and Per- fumery Association (COLIPA).

We are not making some of the specifications in the COLIPA SPF test erythema. One submission argued that requested changes to certain test (Ref. 1). We are revising solar simulator no data indicate that exceeding 1500 W/ parameters because we lack adequate specifications to: m2 causes excessive heat or affects SPF data to determine whether these • Allow the use of smaller beam, test results. The submission argued that changes would change the accuracy or multiport simulators higher intensities should be allowed as reproducibility of the SPF test. We are • Adjust the relative cumulative long as they are thermally tolerated by making changes to some test parameters erythemal effectiveness (RCEE) range test subjects, because allowing higher based on the following developments specifications for each wavelength band intensities enables faster SPF testing. since the 2007 proposed rule published: • Specify that UVA II (320–340 nm) We are not changing the 1500 W/m2 • New data (submitted by the public and UVA I (340–400 nm) irradiance total irradiance limit. We do not have or published in the scientific literature) should equal or exceed 20 percent and data demonstrating that exceeding 1500 • Technical improvement of SPF 60 percent, respectively, of the total UV W/m2 leads to loss of dose reciprocity. testing equipment (290–400 nm) irradiance However, we conclude that the limit • Accumulating experience in the • Change the regular calibration should be retained to protect test performance of SPF testing period from every 6 months to at least subjects. The COLIPA SPF test cites a • Efforts towards international once a year study showing that total irradiance of harmonization of SPF testing These changes are consistent with the 1600 W/m2 induces heat and pain in a procedures COLIPA SPF test. More importantly, majority of test subsites, and In support of the requested changes, these revisions will allow the SPF test recommends keeping total irradiance several submissions (Ref. 1) cited to continue to be accurate and below 1600 W/m2 (Ref. 64). Therefore, differences between the SPF test in the reproducible. For example, we received we are keeping the 1500 W/m2 total 2007 proposed rule and the COLIPA calibration data demonstrating that solar irradiance limit (new 21 CFR SPF test (Ref. 64). The COLIPA SPF test simulators and their UV lamps are 201.327(i)(1)(i)). is a joint effort by the cosmetic industry stable for periods longer than 1 year. One submission also objected to the trade associations in Europe, Japan, Therefore, the requirement in the 2007 250–1400 nm range over which total South Africa, and the United States to proposed rule to calibrate every 6 irradiation should be monitored (Ref. 1). harmonize SPF test procedures. The months is unnecessary. The test results The submission argued that portable International Organization for should be the same whether calibration spectroradiometers are typically Standardization (ISO) is currently is done annually or every 6 months. incapable of measuring wavelengths out developing an SPF test method. Because to 1400 nm. According to the In contrast, we are not changing the harmonization of testing methods is submission, emissions from longer following solar simulator specifications important, we are actively involved in wavelengths have not been shown to because changes to these specifications the ISO working group responsible for affect SPF testing. could reduce test accuracy and/or developing methods for assessing the We are not changing the requirement reproducibility: efficacy of sun protection products. that total irradiation be monitored over • Total irradiance limit of 1500 W/m2 We are revising our proposed SPF test a range of 250–1400 nm. We have • Total irradiance range of 250–1400 method to be as consistent as possible concluded that monitoring over this nm with the COLIPA SPF test. We range of wavelengths helps protect SPF • 20 percent beam uniformity acknowledge the merits of harmonizing test subjects from being exposed to requirement. test methods and are an active undesirable, unnecessary radiation. The participant in ongoing harmonization These test specifications differ from the requirement should not impose undue efforts. However, some of the test COLIPA SPF test, which recommends a hardship, because longer wavelengths 2 parameters in this document differ from 1600 W/m limit and a 10 percent beam can be monitored using a thermopile, comparable parameters in the COLIPA uniformity requirement. pyroelectric, or similar detectors. SPF test because we have concluded Two submissions (Ref. 1) objected to We received two submissions that the data do not support using the limiting total solar simulator irradiance addressing the requirement in proposed COLIPA SPF test parameters. to 1500 W/m2 for all wavelengths 21 CFR 352.70(b)(2) that a solar Throughout the remainder of this between 250 and 1400 nm (proposed simulator have ‘‘good beam uniformity section, we discuss whether test 21 CFR 352.70(b)(1)). We proposed the (within 20 percent) in the exposure parameters in this document match or 1500 W/m2 limit because we were plane’’ (Ref. 1). One submission argued do not match those in the COLIPA SPF concerned that solar simulators that advances in equipment and methods. operating above this limit could cause monitoring allow for a stricter beam excessive heat. Excessive heat could uniformity requirement (<20 percent), A. Solar Simulator harm test subjects and/or cause loss of which would result in less variability in Several submissions recommended dose reciprocity, the correlation SPF test results. Another submission adopting the solar simulator between UV dose and resulting argued that the beam uniformity

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35646 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

requirement is only important for large COLIPA SPF test, but we are not oxybenzone standard is used much diameter beams and has no impact on including the ‘‘P3’’ standard in this more frequently than the homosalate SPF testing using small beams. document. In the 2007 proposed rule standard. We are not changing the 20 percent (72 FR 49070 at 49095 to 49095), we We received one submission beam uniformity requirement because requested further data to show that identifying errors in the ‘‘Composition accurate determination of SPF values testing using the ‘‘P3’’ standard could be of the Padimate O/Oxybenzone relies upon good beam uniformity for all performed with: Standard Sunscreen’’ table that appears beam sizes. In the 2007 proposed rule, • Low level interlaboratory variation in the 2007 proposed rule. As suggested we described how small diameter beams • Sufficient sensitivity to detect by the submission, we are moving the can be tested for beam uniformity (see experimental error inactive ingredient ‘‘propylparaben’’ 72 FR 49070 at 49098). The submission • A reasonable degree of accuracy from ‘‘Part A’’ to ‘‘Part B,’’ as it appears requesting stricter requirements did not The submitted data (i.e. the table of SPF in the COLIPA SPF test. We are not include data showing that current solar values) fail to show that the ‘‘P3’’ revising the listing of the inactive simulators can reasonably be expected standard meets these performance ingredient ‘‘glyceryl monostearate’’ to to have beam uniformity less than requirements because they do not show: read ‘‘glyceryl monostearate (Glyceryl 20 percent. We conclude that a 20 • Individual lab results Stearate SE),’’ as suggested. The United percent beam uniformity requirement is • The number of tests conducted in States Pharmacopeia defines ‘‘glyceryl adequate to produce reliable SPF each lab monostearate’’ as an ‘‘emulsifying and/ results. Therefore, we are keeping the • The number of test subjects used in or solubilizing agent,’’ which adequately requirement that solar simulators each test describes the ingredient that is demonstrate good beam uniformity • Calculated standard errors for each appropriate for use in the formulation. (within 20 percent) in new 21 CFR test C. Test Subjects 201.327(i)(1) (iii). Without these data, we cannot assess In the 2007 proposed rule, we interlaboratory variability, sensitivity to B. Sunscreen Standards proposed requiring the following experimental error, or test result The 2007 proposed rule include two numbers of test subjects providing valid sunscreen standards for use in SPF accuracy. In addition, the advantage of results: testing. The two proposed sunscreen using the ‘‘P3’’ standard instead of the • 20 to 25 subjects for sunscreen standards were a 7 percent padimate padimate O/oxybenzone standard is products with SPF less than 30 O/3 percent oxybenzone standard (mean unclear, because both these standards • 25 to 30 subjects for sunscreen SPF value of 16.3) and an 8 percent have approximately the same SPF value products with SPF value of 30 or more homosalate standard (mean SPF value of of 16. Therefore, we are not including We explained that a minimum of 20 4.47). For SPF testing of sunscreen the ‘‘P3’’ standard in this document. subjects would be required to provide products with SPF values of 2 to 15, We are also eliminating the proposed an acceptably accurate SPF result (i.e., either the padimate O/oxybenzone homosalate standard with an SPF value low standard error of the mean). We had standard or the homosalate standard of 4.47 because the padimate O/ concluded that sunscreen products with would have been required to be tested oxybenzone standard with an SPF value SPF values of 30 or more required a along with the test sunscreen product. of 16.3 is adequate for validating all test greater number of test subjects because Tests for sunscreen products with SPF methodologies. In the 2007 proposed we suspected higher test result values over 15 would have required use rule, we stated that the sunscreen variability for these sunscreen products. of the padimate O/oxybenzone standard. standards were ‘‘method controls rather However, the data used for determining We received two requests to include than calibration tools.’’ As a method appropriate test subject numbers were an additional sunscreen standard with control, the purpose of the sunscreen limited and dated. Therefore, we invited an SPF value of 30 or higher to test standard is verifying proper and submission of additional data sunscreen products with SPF values of consistent performance of test demonstrating what subject numbers 30 or more (Ref. 1). Neither request equipment and procedures, rather than would be adequate. specified any particular sunscreen verifying the accuracy of the SPF value Several submissions recommend standard formulation with an SPF in determined for sunscreen test products. requiring 10 to 25 test subjects as in the this range. If a particular sunscreen Therefore, we conclude that it is not COLIPA SPF test (Ref. 1). These standard formulation were specified, we critical for the SPF value of the submissions include data demonstrating would also need validation data to sunscreen standard to be close to the that SPF testing can be performed with support including the additional SPF value of the sunscreen test product. suitable accuracy and precision with as sunscreen standard in the monograph. It is more important that the sunscreen few as 10 test subjects. The submissions Therefore, we are not including a standard demonstrate consistency of test further argued that SPF testing using a sunscreen standard with an SPF value performance. Consequently, we have minimum of 10 test subjects has been of 30 or more in this document. concluded that including multiple practiced globally for many years, even We also received a request to include sunscreen standards is unnecessary, and for sunscreen products with high SPF the JCIA SPF 15 ‘P3’ sunscreen standard that the padimate O/oxybenzone values. containing 0.5-percent avobenzone, 3- standard is a suitable sunscreen We agree with the submissions and percent octyl methoxycinnamate, and standard for all sunscreen products. We are lowering the number of test subjects 2.78-percent phenylbenzimidazole favor including the padimate O/ required for SPF testing. We are sulfonic acid. To support including the oxybenzone standard over the requiring that a test panel produce a ‘‘P3’’ standard, the request included a homosalate standard because the minimum of 10 valid test results. A table showing mean, maximum, and homosalate standard was only proposed maximum of three subjects may be minimum SPF values from tests for use for SPF testing of sunscreen rejected from the panel. Therefore, if 3 conducted in labs in Europe, Japan, products with SPF values lower than 15. subjects would be rejected, a test panel Australia, and South Africa. We Because most currently marketed would have had to include 13 subjects. recognize that the ‘‘P3’’ standard has sunscreen products have SPF values of We are reducing the number of test been widely used and is included in the 15 or higher, the padimate O/ subjects in this document because the

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35647

data we received demonstrate that SPF size specifications (Ref. 1). Therefore, product application at 2 mg/cm 2 testing can be conducted with adequate we are revising the test site/subsite size (100 mg sunscreen product applied over accuracy and precision using as few as specifications to accommodate new 50 cm 2). However, the study did not 10 test subjects, even when testing high equipment and to harmonize our include a comparison to presaturated SPF products. The submissions include specifications with global SPF test finger cots. Therefore, it is difficult to SPF test results for several sunscreen methods. determine the effect of presaturation on formulations using panels of 20 to 25 residual sunscreen amounts. E. Finger Cot test subjects. We randomly selected 10 In addition, we reassessed the basis subjects within each of these panels to In the 2007 proposed rule, we for presaturation. We are now determine if using fewer subjects proposed that a finger cot, presaturated concerned that performing the significantly decreased test accuracy with sunscreen, be used to apply the presaturation step may lead to and precision. For each of these panels, sunscreen in the SPF test (proposed overestimation of SPF values, because the mean SPF value and standard error 21 CFR 352.70(c)(5)): the residual amount normally left on a calculated from a randomly selected Use a finger cot compatible with the finger cot with presaturation may subset of 10 subjects were not sunscreen to spread the product as evenly as increase the amount of sunscreen significantly different from those possible. Pretreat the finger cot by saturating applied to the skin This could lead to calculated from all 20 to 25 subjects in with the sunscreen and then wiping off overestimation of SPF values. the panel. Therefore, these data indicate material before application. Preteatment is Overestimation of SPF may, in turn, that using as few as 10 test subjects will meant to ensure that sunscreen is applied at lead to increased incidence of sunburn the correct density of 2 mg/cm 2. not compromise SPF test accuracy or because consumers may anticipate precision. Consequently, fewer test sites We received one submission that greater protection than a sunscreen and subsites need to be tested and fewer objected to the use of finger cots because product actually provides. This test results need to be rejected, thereby consumers do not typically use finger overestimation risk is a sufficient basis decreasing the number of test subjects cots when applying sunscreens (Ref. 1). to remove the presaturation step from needed. Our revised SPF test subject Other submissions argued that the the proposed SPF test method. number requirement is similar to the presaturation requirement for finger cots We also received data showing that COLIPA SPF test requirement. The only is unnecessary and introduces testing without the presaturation step significant difference related to test variability in applied amounts (Ref. 1). can produce highly reproducible results subject number is that we are not Other submissions requested the (Ref. 1). In a test of 20 subjects without including a statistical requirement or optional use of sponge applicators for the presaturation step, a control allowing individual subjects to be added testing powder formulations, because sunscreen product yielded a mean SPF incrementally to a test panel as allowed they argued that sponge applicators value of 4.19 with a standard error of under the COLIPA SPF test. distribute powder formulations more 0.06 (i.e., 1.4 percent error), while a test evenly than finger cots (Ref. 1). We are sunscreen product yielded a mean SPF D. Test Sites and Subsites not addressing issues regarding the use value of 15.54 with a standard error of Several submissions requested the of sponge applicators for the testing of 0.22 (i.e., 1.4 percent error). These errors following revisions of the minimum size powders in this rule. Elsewhere in this are small, suggesting that the calculated specifications for test sites and subsites issue of the Federal Register, we SPF values did not vary significantly proposed in the 2007 proposed rule publish an advance notice of proposed between test subjects. If lack of (Ref. 1): rulemaking that discusses sunscreen presaturation increased variability, then • Test site: proposed 50 cm2 revised dosage forms, including powders. We the errors would be expected to be to 30 cm2 may address this issue in a future larger. Therefore, we are removing the • Test subsite: proposed 1 cm2 rulemaking.. presaturation requirement because of revised to 0.5 cm2 While we acknowledge that the risk of overestimation of SPF values • Subsite separation: proposed 1 cm consumers do not use finger cots to and our conclusion that the removal of revised to 0.8 cm apply sunscreens, we are continuing to the presaturation step will not affect the According to the submissions, these require the use of finger cots in the SPF reproducibility of SPF test results. smaller revised minimum sizes would test. The use of finger cots seems to allow multiport solar simulators to be increase reproducibility of test results, F. Application Amount used, while the larger proposed sizes which was why we originally proposed We are continuing to require that 2 would not. These revised specifications requiring use of finger cots (72 FR 49070 mg/cm2 sunscreen product be applied have also been adopted in the COLIPA at 49100 through 49101). We agree with for the SPF test (proposed 21 CFR SPF test (Ref. 64). the submissions that the presaturation 352.70(c)(5); new 21 CFR We are revising the test site and requirement is unnecessary and are 201.327(i)(4)(iii)). Several submissions subsite size specifications as requested removing this requirement. We argued for a lower application amount by these submissions. Our previously proposed requiring finger cot that better reflects the actual amount proposed specifications were based on presaturation to prevent sunscreen used by consumers, which they argued single port solar simulators. Some new product from adhering to the finger cot is commonly 1 mg/cm2 or less (Ref. 1). multiport solar simulators cannot meet instead of being transferred to the test These submissions argued that the these proposed specifications. In the subject’s skin, resulting in sunscreen unrealistically high 2 mg/cm2 2007 proposed rule, we stated that product being applied at less than the application amount results in SPF reducing test site/subsite size intended 2 mg/cm 2. We received study values that overstate the actual sun specifications would be considered if results showing that a residual amount protection provided by the amounts data were submitted showing that these of sunscreen product may adhere to consumers typically apply. Other reductions would not compromise non-presaturated finger cots, but the submissions supported the 2 mg/cm2 testing accuracy (72 FR 49070 at 49100). amount was small (approximately application amount (Ref. 1). These New data show that SPF testing can still 2 percent) (Ref. 1). In this study, each of submissions argued that SPF values are be accurately performed using the 100 finger cots (without presaturation) relative, not absolute, values that allow recommended reduced test site/subsite was weighed before and after sunscreen comparison of sun protection provided

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35648 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

by different sunscreen products. We received various requests to revise because it would expose fewer subjects According to the submissions, changing the test (Ref. 1). One submission to UV irradiation, cost less, and save the application amount will affect the recommended longer water immersion time. The pass/fail test is based on the ability of consumers to make this times equal to those in water resistance hypothesis that a sunscreen product of comparison. tests used in Australia and New a certain SPF has a 50:50 probability of We are not changing the sunscreen Zealand. Another submission included preventing the MED response when product application amount because we data from an in vitro water resistance irradiated with a UV dose correlated have concluded that the advantages of test to support removing the in vivo with that SPF. For example, a sunscreen continuing to require 2 mg/cm2 exceed water resistance test. A third submission product with an expected SPF value of the disadvantages of lowering the stated the test should be eliminated 30 or more should prevent the MED amount. Requiring the 2 mg/cm2 because it is not validated and requires response in greater than 50 percent of sunscreen product application amount too much time. Further, the submission test subsites irradiated with a UV dose is consistent with SPF test methods argued that directions for frequent equivalent to 30 times the UV dose that used in other countries. The 2 mg/cm2 reapplication make the test unnecessary. causes the MED response on application amount is being used in We are continuing to include a water unprotected skin. If a test sunscreen Europe, Australia, Canada, Korea, and resistance test because water resistance product prevents the MED response in Japan (Refs. 65–67). If we lower the is an important property of sunscreen a significant number of the subsites (i.e., application amount, sunscreen products products that can benefit consumers. significantly more subsites that ‘‘pass’’ available in the United States will have The water resistance test indicates that versus ‘‘fail’’), then the test sunscreen significantly lower SPF values than a sunscreen product’s labeled SPF product would be allowed to be labeled similar products available in other protection is retained for a certain with the SPF correlated to the UV dose. countries. This discrepancy in SPF period of time after immersion in water. We are not including the optional use values is counterproductive to our This is useful information to consumers. of a pass/fail test for SPF testing. We global harmonization efforts and would Therefore, we conclude that a water considered a pass/fail SPF test in the likely mislead consumers traveling to resistance statement based on the test 2007 proposed rule (72 FR 49070 at other countries about the SPF protection should be allowed (see section III.C of 49094 to 49095). We stated that a pass/ of foreign sunscreen products. this document). fail test could be a reasonable substitute Another advantage of continuing to We are not changing the 20-minute for our proposed SPF test for sunscreen 2 require a 2 mg/cm sunscreen product water immersion periods or the number products with SPF values of 30 or more application amount is greater of immersion periods required. We if certain modifications were made and reproducibility of SPF test results. based these time periods on marketing validation data demonstrated that the Bimczok et al. compared the SPF values data indicating that individuals at the test could be performed similarly determined using sunscreen product beach or the pool spend an average of between labs. application amounts of 0.5, 1, and 2 mg/ 21 minutes in the water and go into the In response to our invitation for 2 cm (Ref. 68). The SPF values water an average of 3.6 times (43 FR public comment, one submission 2 determined using 2 mg/cm sunscreen 38206 at 38263, August 25, 1978). We included two studies comparing a pass/ product were more reliable and have not received any other data fail SPF test to the proposed SPF test: reproducible than SPF values supporting different time periods. We (1) A single center study of four determined using the lower application have concluded that more or longer sunscreen products with different SPF amounts. A sunscreen product values and (2) a multicenter (four 2 water immersion periods are not application amount of 2 mg/cm is a needed. laboratories) study of two high SPF large enough amount to allow We are, however, reducing the drying sunscreen products. After reviewing visualization of the distribution of period from 20 minutes to 15 minutes. these data, we have determined that the sunscreen product as it is applied. This We are making this change to decrease pass/fail test has the following allows for more consistent and uniform the time required for testing. Shorter drawbacks: application of the sunscreen used in • Each test subsite evaluation is 2 testing time may increase test accuracy testing. Therefore, the 2 mg/cm and reproducibility, especially for high biased towards ‘‘pass’’ because the sunscreen product application amount SPF sunscreens that retain their water evaluator expects that no skin reaction is more likely to generate reproducible resistance for 80 minutes. In addition, should occur on subsites protected by results. 15 minutes is adequate time to allow for the test sunscreen product. • G. Water Resistance drying. It is possible that sunscreens The test fails to reject test sites where all of the subsites show positive In the 2007 proposed rule, sunscreen may lose water resistance with repeated wetting and drying. However, we have responses or all of the subsites show products tested with two 20-minute negative responses. immersion periods (i.e., 40 minutes concluded that a 15-minute drying • period mimics consumer behavior and The validity of treating each subsite total) would be allowed to include a as an independent variable is ‘‘water resistant’’ statement and ensures that the water resistant properties of a sunscreen do not change questionable. sunscreen products tested with four 20- • The test endpoint (any observed minute immersion periods (i.e., 80 with multiple cycles of water immersion and drying. reaction) differs from the endpoint in minutes total) would be allowed to the proposed SPF test (clearly defined include a ‘‘very water resistant’’ VII. SPF Test Issues (Other than Test erythema). statement. There is a 20-minute drying Parameters) • A passing test result for the period between each immersion period. sunscreen standard does not A. Pass/Fail (Binomial) SPF Test For example, a ‘‘water resistant’’ demonstrate that the test is being sunscreen product would be tested by Several submissions requested the performed correctly. having test subjects in the water for 20 optional use of a pass/fail (binomial) • Test results do not include data for minutes, out of the water for 20 test to determine the SPF value of a water resistant sunscreen products. minutes, and in the water for 20 sunscreen product (Ref. 1). These • Allowing this test as an option minutes. submissions promote the pass/fail test would yield products with different UV

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35649

protection levels labeled with the same product is significantly diminished response to UV radiation (Ref. 1). The SPF. because of photodegradation. We submission did not identify specific • SPF test methods developed by maintain that the proposed SPF test anti-inflammatory ingredients. The various standards-setting organizations procedure does account for submission argued that, by decreasing do not include a pass/fail test. photostability to some extent, because the erythemal response, these • The study report includes statistical the SPF test exposes sunscreen products ingredients could falsely inflate SPF errors that overstate the statistical power to UV radiation before an SPF value is values determined in SPF testing. In of the test to distinguish whether a test determined. Consequently, sunscreen addition, these anti-inflammatory sunscreen product provides significant products susceptible to ingredients may increase the likelihood UV protection. photodegradation have correspondingly of unwanted harmful effects from sun Therefore, we are not including a pass/ lower SPF values. One submission exposure because sunburn, a cue to fail test in the SPF test procedure, argued that the SPF test does not fully avoid sun exposure, would be less because including a pass/fail test would account for photostability because the evident. present numerous complications and solar simulator emission spectrum is Although the submission raises a the available data indicate that a pass/ different than natural sunlight. serious concern, we are not aware of any fail test has disadvantages compared to However, this difference is an data confirming that this problem exists. the SPF test included in this document. unavoidable limitation in testing Therefore, a test to show that anti- because solar simulators cannot inflammatory ingredients may be B. Photostability perfectly replicate natural sunlight. decreasing erythemic response to UV Several submissions expressed We acknowledge that UV radiation radiation is not required at this time. It concern about the loss of UV protection can change the composition of seems unlikely that anti-inflammatory by sunscreen products due to sunscreen products if the products are ingredients will affect SPF values breakdown of ingredients from exposure not photostable, as demonstrated by the because their anti-erythemic effect is to sunlight (Ref. 1). These submissions submitted data. However, we are not relatively short-lived compared to the recommended a test to ensure that certain that these data are applicable 16–24 hour interval between UV sunscreen products exposed to sunlight under actual use conditions. The data exposure and erythema observation in retain sufficient UV protection. regarding the effects of UV radiation on the SPF test. the protection provided by sunscreen Submitted data show that the VIII. Broad Spectrum Test composition of sunscreen products can active ingredients are limited and change from exposure to UV radiation. inconclusive. Therefore, we are not In this document, we are referring to The submissions argue that the creating a photostability test as part of testing involving the UVA part of the published photostability studies are the SPF test procedure in this spectrum as ‘‘broad spectrum testing.’’ inconclusive because the studies document. The term ‘‘broad spectrum’’ more accurately describes the test as covering employ artificial test conditions that C. In Vitro SPF Test may not be appropriately extrapolated the full extent of the terrestrial solar UV to actual use of sunscreens: One submission suggested replacing spectrum (i.e., UVA and UVB radiation). • Tested sunscreen active ingredients the proposed in vivo SPF test with an Section VIII.A. of this document were contained in solutions rather than in vitro SPF test (Ref. 1). An in vitro SPF provides our rationale for no longer in typical sunscreen product test would have advantages of faster requiring an in vivo test assessing the formulations performance, lower expense, and no persistent pigment darkening associated • Tested sunscreen products exposure of subjects to UV radiation. with UVA radiation. Section VIII.B. of contained active ingredients that are not We agree that an in vitro SPF test has this document explains why the in vitro representative of the active ingredients these advantages. However, we are not test should be changed from a modified included in typical sunscreen products replacing the in vivo SPF test with an Diffey-Robson ratio to the critical • Products were tested over a limited in vitro SPF test for the same reasons we wavelength test. Section VIII.C. defines range of the UV spectrum stated in the 2007 proposed rule (72 FR the testing parameters to be employed in 49070 at 49095). One shortcoming of an The submissions argue that evaluating the critical wavelength of an in vitro test is the lack of data on the understanding the photostability of OTC sunscreen product. performance characteristics of in vitro sunscreen active ingredients alone is not A. In Vivo Test Method: Not Required test substrates, such as quartz or useful. Rather, the submissions argue artificial skin. In the 2007 proposed We stated in the 2007 proposed rule that it is critical to understand the rule, we stated that data failed to show that an assessment of UVA protection photostability of sunscreen active that a substrate adequately mimicked should include determination of both ingredients as part of an overall the physiological characteristics of the magnitude and breadth of sunscreen product. human skin. We stated that we would absorption in the UVA part of the We agree that the available data have consider an in vitro test if validating spectrum (72 FR 49070 at 49102 through limitations. Although the submissions data demonstrated that the performance 49106). We proposed that an in vivo argue that the inconclusive data support of the in vitro test was equivalent to the Persistent Pigment Darkening (PPD) test including a test for photostability, we in vivo test. We have not received be used to evaluate the magnitude of have concluded that the data do not adequate data to validate an in vitro SPF absorption and an in vitro test be used justify requiring a photostability test at test. Therefore, we are not including an to evaluate the breadth of absorption. this time. We are not able to establish in vitro test in this document. The PPD test, a modification of the PPD specific photostability test procedures test accepted by JCIA 10 since 1996, is or specifications based on the available D. Anti-Inflammatory Ingredients almost identical to the SPF test. It is data. We have not received data One submission recommended recognized as a standard for the in vivo validating the performance of a requiring a test to verify that sunscreen assessment of UVA protection by the photostability test, nor have we received products do not contain anti- JCIA and the European Commission data demonstrating that the inflammatory ingredients that effectiveness of any particular sunscreen significantly decrease erythemic 10 Japanese Cosmetic Industry Association.

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35650 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

(Ref. 7). The most significant differences Other disadvantages are pointed out by longer wavelengths of UVA radiation. in the PPD test compared to the SPF test Nash et al. (Ref. 4): Therefore, we modified the ratio to give are (1) the light source emits only UVA • The physical properties of more emphasis to the UVA I area under radiation (320–400 nm) and (2) the sunscreen products may differ when the absorbance curve. endpoint is darkening of the skin sunscreen products are exposed to UVA Many submissions argued that we (tanning) rather than reddening of the radiation alone. should require a determination of skin (erythema). • The PPD test is expensive, time critical wavelength rather than the We have concluded that the PPD test consuming, and labor intensive. proposed ratio to determine broad is not necessary to establish that a • The ability to identify small spectrum protection (Ref. 1). We agree sunscreen product provides protection differences in pigmentation requires a with the arguments made in the against UVA radiation. The magnitude high degree of expertise and submissions. Therefore, in this of absorption over the solar terrestrial interpretation of pigmentation changes document, we are requiring that broad UV portion of the spectrum (both UVA will be dependent on the examiner. spectrum protection be assessed by and UVB) can be effectively assessed • There may be a high degree of determining the critical wavelength of a based on the SPF test in combination variability in test results between sunscreen formulation. The submissions with a pass/fail broad spectrum in vitro noted the following disadvantages with test (see Section VIII.B of this subjects in the same test panel as well as between different test panels for the the proposed ratio: document). If sunscreen products pass • same sunscreen product. The proposed ratio places too much the in vitro broad spectrum test, then • emphasis on the UVA I region, which is the amount of UVA radiation protection, The test results may not be reproducible between labs. not generally considered to contribute as well as UVB radiation protection, significantly to the harmful effects of must increase as the SPF value Because of these disadvantages of conducting the PPD test, and the fact exposure to UV radiation. increases. For example, a Broad • A large ratio could result if one or Spectrum SPF 40 sunscreen product that information obtained from such tests is already provided by SPF testing more ingredients absorb radiation in the must provide more UVB and UVA shorter wavelength UVA II region but radiation protection than a Broad for sunscreen products that pass the in vitro broad spectrum test, we are not at all or only minimally in the Spectrum SPF 20 sunscreen product. longer wavelength UVA I region. For For sunscreen products that pass the eliminating the requirement to conduct example, oxybenzone absorbs radiation in vitro broad spectrum test, we have a PPD or any other in vivo UVA test in at 340–360 nm, and inclusion of this concluded that the SPF and PPD tests this final rule. ingredient at higher concentrations are redundant of each other, but we B. In Vitro Test Method: Critical might result in a high ratio even though have reasons to prefer the SPF test. The Wavelength SPF and PPD tests are both clinical and it does not provide true broad spectrum Many submissions objected to our protection. indicative of the magnitude of • absorbance of UV radiation. proposal to use a modification of the The proposed ratio is not a Furthermore, both tests depend on the Boots adaptation of the Diffey/Robson validated measure of UVA protection skin type of the individual. The SPF test ratio as an in vitro measure of UVA and is not used anywhere else in the protection (Ref. 1). The Diffey/Robson world. measures skin reddening, which is due • primarily to UV radiation in the UVB ratio evaluates UVA protection relative To achieve high ratios with existing and UVA II regions (290–340 nm). The to UVB protection. The ratio is GRASE active ingredients, the PPD test measures skin darkening, calculated as the area under the concentrations of ingredients that which is due primarily to UV radiation absorbance curve in the UVA region absorb in the UVB and UVA II parts of in the UVA II part of the spectrum (320– (320–400 nm) divided by the area under the spectrum have to be reduced, 340 nm). Therefore, the UV radiation the absorbance curve in the UVB region lowering protection in these parts of the range covered by the PPD test is also (290–320 nm). As the degree of spectrum (i.e., the SPF has to be lowered covered by the SPF test. In both tests, protection against UVA radiation to increase the ratio). the endpoint is indicative of how much increases, the ratio increases. We agree that our proposed ratio is UV radiation is absorbed. As the We proposed a modification of this not the most appropriate in vitro magnitude of UV radiation absorbance ratio to be calculated as the area under measure of broad spectrum protection. increases for a sunscreen product, both the absorbance curve in the UVA I In agreement with many of the the SPF and PPD ratings increase. region (340–400 nm) divided by the area submissions, we have concluded that We have identified several under the absorbance curve over total the ratio places too much emphasis on disadvantages of the PPD test as UVB and UVA range (290–400 nm). We absorption in the UVA I part of the described in the proposed rule (72 FR indicated that this modification was spectrum. Although there is some 49070 at 49103): necessary because we were concerned evidence that UVA I radiation • Human subjects are exposed to high that a sunscreen product absorbing contributes to immune suppression and doses of UVA radiation with unknown strongly in the UVA II region (320–340 an increase in p53-positive cells, the health consequences. nm), but not absorbing strongly in the effects of UVA I radiation on these • Exposure to UVA radiation alone UVA I region, might produce a processes are 100 to 1000 times less (i.e., in the absence of UVB radiation) is disproportionately high ratio value (72 than the effects attributed to UVB and never encountered in nature, and the FR 49070 at 49105). We would not UVA II radiation (Ref. 4). We also biological effects of such exposure may consider this sunscreen product to be a acknowledge that there is no experience differ greatly from those due to exposure good broad spectrum sunscreen product using the proposed ratio. Further, we to natural sunlight. even though it has a high ratio value. received some data in the submissions • Because it is unclear how tanning We noted the importance of ensuring that demonstrate the need to reduce SPF relates to the harmful effects of sunlight, that protection extends well into the values in order to achieve high ratio it is unclear whether persistent pigment UVA I region (340–400 nm), because values. We are concerned that, in an darkening represents a clinically neither SPF nor PPD measurements effort to gain UVA protection, meaningful endpoint. provide much information about the consumers may be more susceptible to

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35651

sunburn because SPF values could be a better measure of broad spectrum under the absorbance curve represents lower in products with higher ratios. protection. The critical wavelength (λc) 90 percent of the total area under the In agreement with many of the is derived from the same data as the curve in the UV region. This is submissions, we have concluded that modified ratio. The critical wavelength expressed mathematically as: the critical wavelength method provides is the wavelength at which the area

In this expression, A(λ) is the mean In this document, we are requiring value (magnitude of UVB and UVA absorbance at each wavelength, and dλ that sunscreen products have a critical protection) provides a complete is the wavelength interval between wavelength of at least 370 nm (the mean measure of broad spectrum protection measurements. value must be equal to or greater than provided by a sunscreen product. Like the proposed ratio, the critical 370 nm) to be labeled as providing C. Critical Wavelength Test Parameters wavelength measures the breadth of the broad spectrum protection (see section UV absorbance curve. Unlike the VIII.B.). This differs from the tiered Although the proposed ratio and proposed ratio, the critical wavelength rating (low, medium, high, and highest) critical wavelength calculations are does not emphasize certain parts of the that we included in the 2007 proposed different, both tests are based on the UV spectrum, but is a measure of rule (proposed 21 CFR 352.50(b)(2)). We construction of a transmittance curve absorbance across the entire solar have concluded that the threshold over the range of UV wavelengths from terrestrial UV spectrum (UVB and UVA critical wavelength for a broad spectrum 290 to 400 nm. We received several radiation). Sunscreen products offering statement should be 370 nm. This submissions requesting that we change primarily UVB protection would have a wavelength is sufficiently difficult to or, in some cases, better define aspects critical wavelength less than 320 nm, achieve and will ensure that sunscreen of the methodology used to measure whereas those providing both UVB and products meeting this threshold provide transmittance over these wavelengths UVA protection would have critical a significant amount of broad spectrum (Ref. 1). Although the submissions, in wavelengths between 320 and 400 nm. protection. On the other hand, it is not most cases, referred specifically to the The critical wavelength method is so difficult to formulate sunscreen proposed ratio test, the points made simple, reproducible, and inexpensive. products to achieve this critical regarding methodology apply equally to It has been used by sunscreen wavelength that manufacturers cannot the critical wavelength test. manufacturers to evaluate UVA develop broad spectrum sunscreen We are making several revisions to the protection for over a decade and is one products. We have concluded that UV section we referred to as the ‘‘UVA in of the most commonly used UVA tests. radiation in the range of 370—400 nm vitro testing procedure’’ in the 2007 This is evidenced by the organizations is not very harmful based on the proposed rule (proposed 21 CFR that recommend its use for determining available action spectra for sunburn and 352.71). To more accurately describe the broad spectrum protection, including skin cancer. We conclude that most of test as covering both the UVB and UVA the European Commission, the the harmful effects from the sun are regions of the spectrum, we now refer to American Academy of Dermatology, the caused by UV radiation in the range of the test as the ‘‘broad spectrum test.’’ American Society for Dermatologic 290—370 nm. Further, we conclude that The revisions are listed in Table 5 in the Surgery, and the Skin Cancer critical wavelength (breadth of UVB and order in which they appear in this Foundation (Ref. 1). UVA protection) coupled with the SPF section of the document.

TABLE 5—SUMMARY OF REVISIONS TO THE PROPOSED IN VITRO BROAD SPECTRUM TEST INCLUDED IN THIS FINAL RULE

Revised test parameter 2007 proposed rule This final rule

Plate ...... Quartz plate (21 CFR 352.71(b)) PMMA1 plate (21 CFR 201.327(j)(1)(i)) Term ‘‘spectroradiometer’’ ... Spectroradiometer Spectrometer (21 CFR 352.71(c) and (d)) (21 CFR 201.327(j)(1)(ii), (iv), and (v)) Light source for transmit- Solar simulator Produce a continuous spectral distribution of UV radi- tance measurements. (21 CFR 352.71(a)) ation from 290 to 400 nanometers (21 CFR 201.327(j)(1)(iii) Input optics: Bandwidth ...... 5 nanometers 1 nanometer (21 CFR 352.71(d)) (21 CFR 201.327(j)(1)(iv)) Dynamic range of the spec- Not specified Sufficient to measure transmittance accurately through trometer. highly absorbing sunscreen (21 CFR 201.327(j)(1)(v)) Application of sunscreen 2.0 mg/cm2 with single-phase spreading (21 CFR 0.75 mg/cm2 with 2-phase spreading (21 CFR drug product to plate. 352.71(e)) 201.327(j)(2)) Pre-Irradiation dose ...... Proportional to SPF value (21 CFR 352.71(f)) Fixed at 800 J/m2-eff (21 CFR 201.327(j)(3)) Number of transmittance 12 measurements of mean transmittance on 5 different 5 measurements of mean transmittance on 3 different measurements. plates (21 CFR 352.71(g) and (i)) plates (21 CFR 201.327(j)(4) and (6)) Calculation of critical wave- Not applicable 21 CFR 201.327(j)(7)) length. 1 Polymethylmethacrylate

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.001 35652 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

We re-organized the broad spectrum use of PMMA plates primarily because 3. Light Source for Transmittance test parameters in this final rule so that the vast majority of validation data we Measurements they are listed in the order that the test have reviewed was collected using Four submissions (Ref. 1) asserted is done. This section of the document PMMA rather than quartz plates. that it is inappropriate to specify a solar begins with a description of the plates Further, we agree with the submissions simulator as the light source for to be used and the requirements for UV noting that PMMA plates are less measuring transmittance (proposed 21 spectrometry. The next section expensive than quartz and, therefore, CFR 352.71(a)). Three of the addresses application of the sunscreen can be disposable. The disposability of submissions argued that radiation product to the plate, and the following the PMMA plates will eliminate the emitted from a solar simulator is filtered section addresses the pre-irradiation requirements for cleaning and re- such that there is very low energy procedure. The last sections included roughening the surface characteristic of output in the UV region below 300 nm under broad spectrum test parameters quartz plates. (Ref. 1). One submission noted that a address measuring the amount of Consistent with COLIPA, we are also light source filtered in this way cannot radiation transmitted through the specifying the degree of roughness and provide sufficient energy to measure sunscreen product, converting these size of the application area on these transmittance through highly absorbing measurements to absorbance values, and plates. Plates should be roughened on sunscreen products. The same calculating the critical wavelength of a one side to a three-dimensional surface submission suggested that there may not sunscreen product. topography measure (Sa) between 2 and be enough transmittance at wavelengths All of the proposed test parameters 7 micrometers. These Sa values are less than 300 nm to exceed the noise were re-evaluated in the preparation of supported by validation studies (Ref. 70) level of the system even in the absence this document. Some of the parameters of a sunscreen product (when did not require revision. Test and are comparable to those recommended by COLIPA (Ref. 69). The transmittance should be maximal). parameters not revised include: We agree with the submissions and, • Sample holder application area must be at least 16 in 21 CFR 201.327(j)(1)(iii) of this • Input optics (other than slit width) square centimeters with no side shorter • Light source for pre-irradiation than 4 centimeters. We are also document, are specifying that the light • Calculation of mean transmittance replacing the word ‘‘substrate’’ with the source for transmittance measurements values simpler and more widely used term provide continuous, full spectrum • Calculation of mean absorbance ‘‘plate.’’ radiation from 290 to 400 nanometers. The use of such a light source should values These changes are included in 21 CFR maximize instrument transmission The parameters defined in this section 201.327(j)(1)(i) of this document. properties while retaining full are based on our review of submitted Specifying standardized roughness and sensitivity. We note that this type of data (Ref. 1) and peer-reviewed size parameters will result in more light source is recommended by literature. Wherever possible and accurate and reproducible intra- and COLIPA (Ref. 69). consistent with sound science, we have inter-laboratory measurements of broad attempted to harmonize the parameters spectrum photoprotection. Because 4. Wavelength Interval Between with existing standards, including those these PMMA plates of specified Transmittance Measurements of the European Commission (Ref. 7) roughness and size are already being and COLIPA (Ref. 69). As stated earlier Two submissions argued that we used in many parts of the world and are should reduce the wavelength intervals in this document, we are also actively recommended by COLIPA, we have involved in the ISO working group between transmittance measurements concluded that they can be employed in from the proposed 5 nm to 1 nm (Ref. responsible for developing broad spectrum testing in this country methodologies for assessing sun 1). The submissions stated that with minimal expense or training of specifying a smaller interval would protection (both UVB and UVA personnel. protection). produce more accurate results and 2. ‘‘Spectroradiometer’’ vs. noted that current spectrometers are 1. Plate ‘‘Spectrometer’’ capable of making measurements at 1 Many submissions argued that we nm intervals. We agree with the should specify that roughened PMMA Four submissions asked us to replace submissions. Additionally, we are aware (polymethylmethacrylate) plates be used the term ‘‘spectroradiometer’’ with the that the COLIPA guideline (Ref. 69) as a substrate rather than roughened more generally used term specifies that transmittance quartz included in the 2007 proposed ‘‘spectrophotometer’’ (Ref. 1). We measurements are to be taken at 1 nm rule (Ref. 1). The submissions stated originally chose the term intervals. Therefore, we are revising the that they prefer PMMA plates because ‘‘spectroradiometer’’ because UV required input slit bandwidth in this these plates are: radiation is not detectable by the human document to specify that it be less than • Less expensive than quartz eye and, therefore, is not gauged by or equal to 1 nm (new 21 CFR • Disposable—no need to clean or re- photometry (which measures visible 201.327(j)(1)(iv)). We are also revising roughen light). However, the term the measurement interval (new 21 CFR • Readily available with roughened ‘‘spectrophotometer’’ is often used 201.327(j)(4)) to state that transmittance surface interchangeably with the term values should be measured at 1 nm • Validated in COLIPA ring tests and ‘‘spectroradiometer.’’ In this document, intervals. in widespread use for more than a we are replacing the term 5. Dynamic Range of the Spectrometer decade ‘‘spectroradiometer’’ with the more • Recommended by the European inclusive term ‘‘spectrometer.’’ Use of We are adding new 21 CFR Commission and COLIPA the term ‘‘spectrometer’’ allows the use 201.327(j)(1)(v) to specify that the We agree with these submissions and of either a spectroradiometer or dynamic range of the spectrometer be are specifying, in this document, that spectrophotometer and will make the ‘‘sufficient to measure transmittance PMMA plates be used as the substrate language more consistent with current accurately through a highly absorbing in this document. We are specifying the COLIPA guidelines (Ref. 69). sunscreen product at all UV

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35653

wavelengths (between 290 and 400 application of 0.75 to 1.0 mg/cm2 results dose be proportional to the SPF value of nm).’’ The information in this section in good transmission within the a sunscreen product (proposed 21 CFR had been included in the section dynamic range of UV detectors (Ref. 1). 352.71(f)). This was to account for the entitled ‘‘Calculation of the spectral Therefore, in this document, we are possibility that consumers may spend transmittance at each wavelength reducing the application amount to 0.75 more time in the sun with higher SPF interval’’ in the proposed rule (proposed mg/cm2 to ensure the UV radiation products. Proportional pre-irradiation 21 CFR 352.71(g)). We considered transmitted through sunscreens is dosing is also recommended in the requiring a minimum dynamic range of within the dynamic range of UV testing procedure published by COLIPA 2.2 absorbance units, as specified in the detectors (21 CFR 201.327(j)(2)). (Ref. 69). In these documents, the pre- COLIPA guidelines (Ref. 69). However, We are also specifying the type of irradiation dose is determined relative we have concluded that it is not spreading action to be employed when to the UVA protection factor. Pre- necessary to include this requirement applying sunscreen product to a plate. irradiation dose increases as the UVA because nearly all current spectrometers One submission noted that the type of protection factor increases. are capable of measuring a dynamic spreading action employed would Two submissions suggested that we range of 2.2 absorbance units or better. depend on the type of product being use a fixed or absolute dose rather than applied. The submission argued that it a relative dose proportional to the SPF 6. Application of Sunscreen Product to might take 30 seconds to evenly spread value of a sunscreen product (Ref. 1). PMMA Plate thicker water resistant creams, but only The submissions noted that, at the same Thirteen submissions (Ref. 1) 10 seconds to evenly spread lotions or time and location on the earth’s surface, expressed one or more concerns over oils. We recognize that the very light all sunscreen products are exposed to the method by which we proposed spreading action for 10 seconds we the same intensity of sunlight. applying sunscreen product to the plate proposed may not be sufficient to Therefore, sunscreen products with (proposed 21 CFR 352.71(e)). Eleven of evenly distribute all dosage forms on a higher SPF values or UVA protection the thirteen submissions recommended plate (proposed 21 CFR 352.71(e)). One factors should not be exposed to higher we reduce the amount applied from 2 submission provided data from a ring pre-irradiation doses. milligrams per square centimeter (mg/ test involving 7 different laboratories We agree with these two submissions. cm2) to between 0.75 and 1.2 mg/cm2. showing that the UVAI/UV absorbance It is appropriate to evaluate sunscreen Three submissions suggested we specify ratio is affected by the amount of product photostability using a fixed that the sunscreen product be applied pressure applied during application. A exposure intensity. We have data with a better defined spreading action. second submission referenced a paper demonstrating that avobenzone- Two submissions requested we consider by Ferrero et al. which shows that light containing sunscreen products undergo requiring that a saturated fingertip be pressure applied to some sunscreen almost complete photodegradation used to apply the product rather than a products results in different ratios than when exposed to doses between 2 and gloved finger. application with greater pressure (Ref. 3 MEDs 11 (Ref. 71). At a dose of 4 We are reducing the application 70). Both submissions recommended MEDs, there were no further decreases amount in this document because adopting a two-phase application in UVB and UVA absorption of five transmittance of UV radiation through a process like that recommended by different sunscreen products containing film of 2 mg/cm2 thickness is low and, COLIPA (Ref. 69). 2.5- to 3- percent avobenzone. These therefore, can result in inaccurate and/ We agree that a two-phase spreading data reflect the worst case scenario for or irreproducible measures of UVA action is a more effective means of photodegradation because avobenzone protection. UV detectors have a range of achieving a film of uniform thickness appears to be the least photostable UV radiation that they can accurately and distribution for a variety of active ingredient in the sunscreen measure referred to as the dynamic sunscreen dosage forms than is the monograph. Therefore, all sunscreen range. If UV radiation is outside the proposed 10 seconds of light spreading. products marketed under the dynamic range (either lower or higher), This type of spreading action is more monograph are likely to be completely measurements from the detector become reflective of actual use than the method degraded after 4 MEDs. Based on this less accurate and often less we proposed. Therefore, we are data, we are specifying a fixed pre- reproducible. We received validation harmonizing the standard with the irradiation dose equivalent to 4 MEDs. data demonstrating that application COLIPA guidelines by specifying that a As we noted in the 2007 proposed rule, 2 amounts lower than 2 mg/cm are more two-phase process be used. Section one MED for a skin type II individual is accurate and reproducible than an 201.327(j)(2) in this document specifies 2 2 200 J/m -eff (72 FR 49070 at 49107). application of 2 mg/cm (Ref. 1). The that ‘‘spreading should be done with a Therefore, in this document, we are 2007 proposed rule required an very light spreading action for 2 specifying a pre-irradiation dose of 4 application amount of 2 mg/cm approximately 30 seconds followed by times 200 J/m2-eff (800 J/m2-eff). because this is the amount specified in spreading with greater pressure for the proposed in vivo SPF and PPD tests. approximately 30 seconds.’’ 8. Number of Transmittance We are not including the PPD test in Two submissions argued that we Measurements this document and we have concluded should specify a saturated fingertip be Two submissions (Ref. 1) stated that that consistency with the SPF test is not used rather than a gloved finger. We do requiring 12 transmittance warranted given the concerns about not agree for the reasons specified in measurements on each plate as inaccurate and/or irreproducible results section VI.E of this document. proposed is excessive and not with an application amount of 2 mg/cm2 7. Pre-Irradiation Dose statistically warranted (proposed 21 in the in vitro UVA method. A reduced CFR 352.71(g)). One submission application amount is consistent with Several submissions expressed provided data showing that there are no the COLIPA guidelines (Ref. 69). Both of concern that the pre-irradiation dose we significant differences in UVAI/UV these documents specify an application proposed to account for differences in ratios calculated based on 3, 5, 8, or 12 amount of 0.75 mg/cm2. Data we have photostability is too high, particularly if reviewed from the Personal Care we reduce the application amount (Ref. 11 Minimal erythema dose—the lowest UV dose Product Council demonstrate that 1). We proposed that the pre-irradiation that produces skin reddening (erythema).

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35654 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

sub-sites per plate. The submission has determined that this final rule is a sunscreens will also incur SPF testing argued that we should reduce the significant regulatory action under costs (see section IX.A.5 of this number of required test sites per sample Executive Order 12866. Consistent with document). Some manufactures will to 6. The other submission proposed Executive Order 13563, the approach also have to relabel products that are that we require only one transmittance taken here maintains ‘‘flexibility and currently labeled with claims that are measurement per plate. The submission freedom of choice for the public,’’ above not allowed under this final rule suggested that, rather than taking all by providing ‘‘information for the (§ 201.327(g) and § 310.545(a)(29)(ii)). public in a form that is clear and multiple measurements from several 2. Benefits small areas on the plate, one intelligible. measurement could be made over a The Regulatory Flexibility Act As discussed in section IV.B of this relatively broad area. requires agencies to analyze regulatory document, the regular use of a Broad One of the submissions also argued options that would minimize any Spectrum SPF 15 or higher sunscreen that it is not necessary to evaluate significant impact of a rule on small product, when combined with limiting transmittance on five different plates entities. Because we lack information time in the sun and wearing clothing to (proposed 21 CFR 352.71(j)). The characterizing the number of products protect sun-exposed areas, reduces the submission provided data showing that by firm-size and because most affected risk of skin cancer and early skin aging. the UVAI/UV ratio for an SPF 15 entities are considered small, we The National Cancer Institute estimates sunscreen product is not significantly conclude that this final rule will have a that there are more than one million different whether it is measured on 1, 2, significant economic impact on a new cases of non-melanoma skin cancer 3, or 5 plates (with 12 measurements per substantial number of small entities. and more than 68,000 new cases of plate). We note that the COLIPA Section 202(a) of the Unfunded melanoma per year in the United States guidelines (Ref. 69) recommend that 3 Mandates Reform Act of 1995 requires (Refs. 72 and 73). According to the separate plates be used. that agencies prepare a written National Cancer Institute, about 8,700 We agree with the submissions that statement, which includes an persons will die of melanoma in 2010. requiring 12 discrete measurements on assessment of anticipated costs and Fatal cases of non-melanoma skin each plate is not necessary to obtain an benefits, before proposing ‘‘any rule that cancer are less common but nonetheless accurate transmittance spectrum. The includes any Federal mandate that may number several hundred per year. The submitted data demonstrate that there result in the expenditure by State, local, labeling requirements in this rule, in are no significant differences in UVAI/ and tribal governments, in the aggregate, conjunction with implementing the UV ratios based on 3, 5, 8, or 12 test or by the private sector, of $100,000,000 format and content requirements in 21 sites. Similarly, we agree with the or more (adjusted annually for inflation) CFR 201.66, which were stayed for submissions that requiring in any one year.’’ The current threshold sunscreens but are being lifted in this measurements for five plates is not after adjustment for inflation is $136 rule, will provide consumers with clear necessary to obtain an accurate million, using the most current (2010) and concise information about sunscreen use and protection, and about transmittance spectrum. Determining 12 Implicit Price Deflator for the Gross the role of sun exposure in increasing transmittance measurements on five Domestic Product. We do not expect the risk of skin cancer and early skin plates, as proposed, results in a total of this final rule to result in any 1-year expenditure that would meet or exceed aging. Consumers will be able to more 60 transmittance measurements. Based this amount. easily identify products that reduce the on the submitted data, a total of 15 risks of skin cancer and early skin aging, transmittance measurements should 1. Background when used as directed. The new produce an accurate transmittance The purpose of this rule is to finalize requirements for product testing will spectrum. Therefore, we are requiring 5 labeling and testing conditions under ensure the accuracy of the SPF value or more measurements on at least 3 which OTC sunscreen drug products and broad spectrum claim on the different plates (21 CFR 201.327(j)(6) in marketed without approved product label. this document. applications are not misbranded. This Although we are unable to quantify 9. Determination of Critical Wavelength rule addresses labeling and testing the effects of clear and concise requirements for both UVB and UVA information, the final rule will provide Critical wavelength is to be radiation protection. The rule modifies clearer and more consistent information determined as described in section the existing SPF test, specifies a test for on the benefits of certain sunscreens in VIII.B of this document. broad spectrum protection, and requires regard to skin cancer risk reduction than IX. Analysis of Impacts changes to the product label that affect is available on current labels. By both the front of the package (the requiring better information on levels of A. Final Regulatory Impact Analysis principal display panel or PDP) and the protection, the rule should contribute to We have examined the impacts of the Drug Facts section. In addition, the rule reduced exposure to UVB and UVA final rule under Executive Order 12866, lifts the stay of effective date applied to radiation and thereby reduce the Executive Order 13563, the Regulatory the 1999 Drug Facts labeling final rule incidence of skin cancer. Flexibility Act (5 U.S.C. 601–612), and (64 FR 13254) specifically for sunscreen The benefits from reduced incidence the Unfunded Mandates Reform Act of products (66 FR 67485). All of skin cancer will equal the value of the 1995 (Pub. L. 104–4). Executive Orders manufacturers of sunscreens will incur illnesses averted. The most appropriate 12866 and 13563 direct agencies to some labeling costs due to revisions to measure of that value is based on the assess all costs and benefits of available both the PDP and the Drug Facts section average willingness to pay to reduce the regulatory alternatives and, when of the product label (see section IX.A.4 probability of skin cancer. We would regulation is necessary, to select of this document). In addition, many then multiply the value per illness regulatory approaches that maximize manufacturers will incur additional averted by the likely number of illnesses net benefits (including potential broad spectrum testing costs unless they averted to determine the benefits of this economic, environmental, public health have already tested their products final rule. Because we lack estimates of and safety, and other advantages; according to the broad spectrum test the likely numbers of illnesses averted, distributive impacts; and equity). OMB required in this rule. Manufacturers of we present estimates of the value per

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00036 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35655

illness averted to illustrate the gains per exposure to sunlight, as well as any System does not have accurate averted case. preventative effects of sunscreen (or any information on the number of marketed We estimated the value per case of other intervention), occur with a long OTC sunscreen products. In the 2007 preventing skin cancer for fatal and non- lag. To estimate the monetary value of proposed rule (72 FR 49070 at 49108), fatal cases of melanoma and non- an averted case of melanoma or non- we estimated that there were about melanoma skin cancer. The estimated melanoma skin cancer through 3,000 OTC sunscreen drug products, average medical cost of treatment, lost combining other protective measures including cosmetic products containing productivity, and willingness to pay to with increased broad spectrum and at sunscreen, with about 12,000 SKUs.12 avoid some symptoms and other effects least SPF 15 protection, we adjust for In response to the 2007 proposed rule, represents a plausible lower bound on the lag between increased protection we received a submission arguing that willingness to pay to avoid a non-fatal and a decrease in the incidence of non- our estimates of the number of products case of skin cancer. For melanoma, the melanoma skin cancer. The only and SKUs were low but the submission estimated total cost is about $2,860 per available long-term study finds a did not suggest a corrected value. We non-fatal case; for non-melanoma skin minimum lag of 5 years before any contracted with the consulting firm cancer, the total cost is about $1,400 per significant risk reduction would occur Eastern Research Group (ERG) to profile non-fatal case; (Refs. 74 and 75). (Refs. 20 and 21). Substantial reductions the sunscreen market and assess the cost The largest potential public health occur with a much longer lag, probably to reformulate a sunscreen product. gains from this final rule would likely 15 to 25 years; we use a 20-year lag in ERG’s full report can be found in Docket come from averted deaths. We can this illustrative analysis. With a 20-year No. FDA–1978–N–0018 (Ref. 80). ERG calculate the monetary value of averted lag discounted at 3 percent, the value did an extensive search using the fatal cases as either the value of per averted statistical case of non-fatal internet and other sources and found statistical lives saved or the value of melanoma is $1,586; if we discount for fewer dosage forms and SKUs than we statistical life-years saved. Although at 7 percent, the value per averted case had estimated. ERG estimates that there skin cancers occur at all ages, most is $740. With a 20-year lag discounted are about 3,065 to 3,600 SKUs. More cases occur at older ages. For that at 3 percent per year, the monetary recently, the new FDA labeling cost reason, we estimate the benefit from value per averted statistical case of non- model estimates that about 3,591 preventing fatal cases using the value of melanoma skin cancer is $773; if we sunscreen SKUs are marketed, with up life years saved. According to the discount at 7 percent, the value per to 2,348 different formulations. Because National Cancer Institute, the average averted case is $361. these data are based on a recent survey age of death from melanoma is 68 (Ref. For fatal cases, with the 20-year lag of the market, we conclude that they are 73); life expectancy for a person discounted at 3 percent per year, the more representative of the number of between the ages of 68 and 69 is about monetary value per averted statistical products affected than the estimates in 16 years (Ref. 76). If we discount the case of fatal melanoma is $1.49 million; the proposed rule. For this analysis, we average years of life saved for averted discounted at 7 percent, the value per therefore use 3,591 SKUs to represent fatal melanoma with rates of 3 and 7 averted fatal case is $520,000. With a the number of affected sunscreen labels percent, we get discounted statistical 20-year lag and a 3 percent rate of and 2,348 for the number of life-years saved equal to 12.6 and 9.4 discount, the discounted value per formulations. years. The various studies of fatal cases averted case of non-melanoma skin To comply with the rule, sunscreen of non-melanoma skin cancer find mean cancer is $920,000 million; with a 7 products currently marketed as or median ages of death in the 77 to 82 percent rate of discount, value per providing broad spectrum protection range (Refs. 77–79). The life expectancy averted fatal case is $360,000. that were already tested using the test for someone between the ages of 79 and We have four estimates of the method in this rule will have to be re- 80 is about 9 years (Ref. 76). If we discounted value per averted cases of labeled but will not have to be retested discount the average years of life saved melanoma and non-melanoma skin for broad spectrum protection. Other for fatal non-melanoma skin cancers cancer, with values corresponded to products will be tested for broad with discount rates of 3 and 7 percent, non-fatal and fatal cases. The annual spectrum protection and, if they pass we get discounted years saved equal to benefits of this final rule will be the and, will be relabeled with the broad 7.9 and 6.5 years. numbers of cases of each type averted spectrum protection claim. In other analyses of life-years saved, multiplied by the value of each type. We Manufacturers may also choose to we have used values for a statistical life- do not, however, have estimates of the reformulate their products to pass the year in the $107,000 to $322,000 range numbers of actual or statistical cases test or discontinue production of the (74 FR 33030, July 9, 2009; updated to that will be averted. Although there is products. current prices). For this illustrative wide agreement among experts that the We have not attributed any analysis, we use a medium value of use of more effective sunscreens reduces reformulation costs to this final rule but $214,000 per statistical life-year. We the risk of sun-related skin cancer, we realize that some manufacturers may multiply the value of a statistical life- are unaware of any studies that quantify choose to reformulate their product if it year by the discounted life-years saved the reduced risk. Without quantitative does not pass the broad spectrum test. per fatal case of melanoma, which estimates of the risk reduction yields $2.69 million using a 3 percent associated with broad spectrum 4. Cost To Relabel Sunscreen Products rate of discount and $2.02 million using protection, we are unable to quantity the The cost to relabel varies greatly a 7 percent rate of discount. If we overall effects of this final rule on depending on the printing method and multiply the value of a statistical life- public health. number of colors used. In the 2007 year by discounted life-years saved per 3. Number of Products Affected proposed rule, we stated that the fatal case of non-melanoma skin cancer, majority of sunscreen products are we get $1.67 million using a 3 percent Estimating the number of products packaged in plastic bottles or tubes with rate of discount and $1.39 million using affected by this rule is difficult because the label printed directly on the a 7 percent rate of discount. we do not have complete data on the The development of melanoma and number of OTC sunscreen products 12 SKUs refers to ‘‘stock keeping units,’’ which are non-melanoma skin cancer from chronic currently marketed. Our Drug Listing individual products, packages, and sizes.

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35656 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

container or applied as a decal or paper We agree that cosmetic packaging and million to $35.1 million, with a medium label during the packaging process. labeling is generally more costly than estimate of $21.8 million (1,796 × The labeling requirements in this rule OTC drug labeling. We also agree that $11,572 + 1,796 × $550). will change both the PDP and the Drug manufacturers of sunscreen-cosmetic 5. Cost To Test or Retest Products To Facts section of the package and are products would use the packaging norm Determine SPF Values considered a major redesign. Frequent of the cosmetic industry because those label redesigns are typical for OTC are the products they are competing Manufacturers will incur SPF testing sunscreen products, with redesigns with. The cost estimates we are using costs because the rule requires labeling generally implemented every 1 to 2 now demonstrate a large variation in the for OTC sunscreen products to include years. If a scheduled redesign coincides price per SKU to account for the SPF values determined in accordance with relabeling required by this rule, the differences in packaging. If the standard with the specific test method that it incremental labeling cost will be lower content and format changes required by describes. We will publish draft than if the labeling change takes place the OTC labeling final rule (64 FR guidance entitled ‘‘Guidance for before scheduled changes. To estimate 13254) are being implemented for the Industry: Enforcement Policy—OTC the cost to relabel, we are assuming that first time, there could be increases in Sunscreen Drug Products Marketed all products will be relabeled and none the size of container and carton labels. Without An Approved Application’’ are discontinued. Since we are allowing, in this rule, for that describes our intended enforcement In the 2007 proposed rule, we used a a compliance period of 1 year for most policy regarding these OTC sunscreen model developed for us by the products but 2 years for products with products. In the draft guidance, we consulting firm RTI International to low sales volume ($25,000 annually), propose to exercise enforcement derive an estimate of the cost to relabel inventory losses for unused packaging discretion for a period of 2 years after sunscreen products (Ref. 81). The model and labels are minimized and accounted the publication of this final rule with was developed to estimate the cost of for in this analysis. regard to the SPF testing requirements food labels, which are similar to the For this final rule, we use the new for certain OTC sunscreen products on labels on the products affected by this FDA labeling cost model developed by the market prior to June 17, 2011. We final rule. In response to the 2007 RTI International, which includes estimate that 65 to 75 percent of proposed rule, we received a estimates for changing sunscreen labels. sunscreen reformulations, or 1,526 to submission disagreeing with our The one-time costs for a major labeling 1,761 will require SPF retesting. The estimates of how sunscreens are change to sunscreen labels are $7,454 to cost of an SPF test depends on whether packaged and the cost to relabel these $18,785, depending on the type of the product is also making water products (Ref. 1). The submission labeling and packaging. The medium resistance claims and the SPF value argued that many sunscreen products, estimate is $11,572 per major labeling being tested; the cost of water resistant particularly sunscreen-cosmetic changes. These costs include mostly testing is much higher than static testing combinations, have a secondary labor and materials, with some cost for (see Table 6). In their analysis of the container and, therefore, an additional lost inventory. sunscreen market ERG found that about label. The submission also argued that We estimate that the timing of 5 percent of products claimed water some sunscreen products would require scheduled relabeling will coincide with resistance and SPF values less than 30, a fold-out label or new secondary carton the relabeling required by this rule for 3 percent of products claimed water to accommodate the labeling required in 50 percent of the 3,591 SKUs . We resistance with SPF greater than 30, this rule. Furthermore, the submission estimate the total labeling cost for the while the remaining 92 percent could argued that relabeling these products SKUs with coinciding scheduled use the static SPF test. We use those would cost $15,000 to $17,000 per SKU. redesign would be minimal percentages to estimate total SPF testing The submission did not include any administrative costs or about $550 ($310 costs of $3.2 to $5.9 million (see Table data or information to support its to $790). Therefore, the total one-time 6). The midpoint of estimated SPF estimate. cost for relabeling would be about $13.9 testing costs is $4.6 million.

TABLE 6—COST OF SPF TESTING

Estimated number of Cost of test Total cost Type of test formulations Low High Low High Low High

Water resistant, SPF < 30 ...... 76 88 $4,500 $4,860 $343,395 $427,923 Water resistant, > 30 ...... 46 53 4,500 5,130 260,037 271,018 SPF static test...... 1,404 1,620 1,900 3,240 2,667,798 5,249,189

Total Cost for SPF testing ...... 3,217,230 5,948,130

6. Cost to Test or Retest Products for (Ref. 1). The argument in the submission broad spectrum statement. Over the Broad Spectrum Protection was based on the four-tier UVA star years, there has been a steady increase rating in the proposed rule. The in the number of products with claims In the proposed rule, we estimated that about 75 percent of sunscreen submission stated that sunscreen of broad spectrum protection. A recent products would need to be tested for products with ‘‘low,’’ one-star survey of marketed products found that broad spectrum protection. We received protection would need to be tested. We 65 percent of the products surveyed met a submission arguing that our estimate have now changed the rating criteria to the criteria for the broad spectrum was too low and that at least 90 percent pass-fail, where a critical wavelength of statement (Ref. 82). Products that were of products would need to be tested at least 370 nm is necessary to make the tested in accordance with the broad

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00038 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35657

spectrum test in this rule would not package size to accommodate the Drug entities are considered small, we need to be re-tested. Facts format. conclude that this final rule will have a Because the broad spectrum test in The estimated total one-time significant economic impact on a this rule is different than the proposed incremental cost of this rule range $17.6 substantial number of small entities. test, we assume that all affected to 42.5 million [($13.9 million labeling Consequently, this analysis, together products would need to be tested. In the cost + $3.2 million SPF testing cost + with other relevant sections of this 2007 proposed rule, we estimated a one- $0.5 million broad spectrum testing document, serves as the Final time testing cost of approximately $5.4 cost) to ($35.1 million labeling cost + Regulatory Flexibility Analysis, as million for products that have broad $5.9 million SPF testing cost + $1.4 required under the Regulatory spectrum protection claims. This million broad spectrum testing cost)]. Flexibility Act. estimate was based on 2,250 sunscreen The medium estimated one-time The average one-time incremental products (75 percent of marketed incremental costs are $27.3 million. cost per firm will be about $185,000 to products) being tested with a test cost of Annualized over 10 years, the costs are $445,000, with a medium of about $2,400. The test costs were estimated as $2.1 to $5 million using a 3 percent rate $285,000. This burden, described in $2,200 for the proposed in vivo test and of discount and $2.5 to $6.1 million more detail in section IX.A of this $200 for the proposed in vitro test. In using a 7 percent rate of discount. document, includes labeling costs, SPF this rule, we are not requiring the in Annualized medium costs are $3.2 testing costs, and broad spectrum testing vivo test. million using a 3 percent rate of costs. The economic impact will vary by In response to the proposed rule, we discount and $3.9 million using a 7 firm, depending on the number of received two submissions arguing that percent rate of discount. If some products requiring testing and the our estimate of $200 for the cost of the manufacturers of sunscreen products number of SKUs requiring labeling. in vitro test was too low (Ref. 1). The have already complied with the 1999 Also, firm-specific impact will vary first submission states that the cost of an final rule and would not otherwise have inversely with the product sales; the per in vitro test is $500, and the second to relabel products as a result of this firm burden will be lower for firms with states that the cost is $800. The first final rule, then these estimates may products with high sales volumes. submission, from a sunscreen overstate actual total costs. Because the relative economic impact of product retesting is greater for products manufacturer, states that $500 is the 8. Analysis of Alternatives price charged by an independent testing with lower sales volume, which could laboratory to test its product. The The principal alternatives we disproportionately affect smaller firms, second submission does not provide any identified were the inclusion of several we are providing a longer basis for its estimate. Although the in provisions from the 2007 proposed rule. implementation period (2 years) for vitro test in this rule is different than In the 2007 proposed rule, we required products with annual sales of less than the in vitro test in the 2007 proposed in vivo and in vitro tests for determining $25,000. Because the OTC drug industry rule, the cost to conduct the tests is the UVA protection. In this rule, we have is highly regulated, all firms are expected to have access to the necessary same. ERG found that the cost of the test eliminated the in vivo test requirement, professional skills on staff or to have ranges from $300 to $800 (Ref. 80). reducing compliance costs by about $5 contractual arrangements to comply Assuming all affected marketed product million. We also proposed labeling on the PDP that would indicate the level of with the testing requirements of this formulations (1,526 to 1,761 UVA protection. In this rule, we rule. formulations) will be tested for broad changed the in vitro test to one that spectrum protection at a cost ranging X. Paperwork Reduction Act of 1995 measures both UVB and UVA protection from $300 to $800, the total cost to test (i.e., broad spectrum protection). We This final rule contains certain sunscreen products for broad spectrum also established a pass/fail broad information collection provisions that protection is estimated to be $457,860 to spectrum protection statement on the are subject to review by the Office of $1,408,800 [(1,526 × $300) to (1,761 × PDP in place of a UVA rating. Management and Budget (OMB) under $800)]. We considered requiring a negative the Paperwork Reduction Act of 1995 7. Total Incremental Costs statement on the PDP indicating that a (44 U.S.C. 3501–3520). Specifically, the product did not have broad spectrum final rule establishes requirements for Because we took steps earlier to protection if it failed the in vitro test. SPF labeling based on specified testing mitigate the impact of labeling changes Numerous submissions from of covered products, (21 CFR on the sunscreen industry by staying the manufacturers opposed this 201.327(a)(1) and (i)). This rule also lifts requirements in earlier rules, the requirement, and we are concerned that the delay of implementation date for labeling costs in this rule incorporate the statement could be misinterpreted § 201.66 (21 CFR 201.66), the general the labeling costs from three final rules: by consumers. Moreover, as noted OTC Drug Facts labeling format 1. 1999 OTC drug labeling final rule previously, this alternative is beyond regulation, which has applied to all (64 FR 13254) the scope of this final rule, which OTC sunscreen products (69 FR 53801). 2. 1999 Sunscreen final rule (64 FR applies only to products that do provide The information collections associated 27666) broad spectrum protection. with § 201.66 have been approved in 3. This rule. accordance with the PRA under OMB Manufacturers were able to postpone B. Small Business Impact (Final Control Number 0910–0340, but this compliance costs when we chose to stay Regulatory Flexibility Analysis) approval does not currently include the labeling requirements for the 1999 We estimate that about 78 percent of application of these provisions to OTC final rule that standardized the format the approximately 100 domestic sunscreens. (76 FR 9022, February 16, and content requirements for labeling companies that manufacture OTC 2011). The lifting of the stay of effective OTC drug products (21 CFR part 201), sunscreen products would be date of § 201.66 for OTC sunscreens will which would have become effective for considered small business entities modify this information collection. all sunscreens by 2005 (69 FR 53801). (defined by the Small Business Elsewhere in this issue of the Federal We include, as part of labeling costs, the Administration as having fewer than Register, in accordance with section cost of increased container labels and 750 employees). Because most affected 3506(c)(2)(A) of the PRA (44 U.S.C.

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00039 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35658 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

3506(c)(2)(A)), we are publishing a 60- with all of the applicable requirements in Environmental UV Radiation: Impact day notice soliciting public comment on under the Executive order and have on Ecosystems and Human Health and the collections of information resulting determined that the preemptive effects Predictive Models, Springer, from this final rule and will then submit of this rule are consistent with Netherlands, 249–258, 2006. these information collection provisions Executive Order 13132. 12. Lucas, R. et al., ‘‘Solar Ultraviolet to OMB for approval. These Radiation. Global Burden of Disease requirements will not be effective until XIII. References from Solar Ultraviolet Radiation’’, we obtain OMB approval. We will The following references are on Environmental Burden of Disease Series publish a notice concerning OMB display in the Division of Dockets No. 13, World Health Organization, approval of these requirements in the Management (HFA–305), Food and Drug 2006. Federal Register prior to the effective Administration, 5630 Fishers Lane, rm. 13. Armstrong, B. K. et al., ‘‘Sun date of this final rule. 1061, Rockville, MD 20857, under Exposure and Skin Cancer,’’ With the exceptions noted above, we Docket No. FDA–1978–N–0018 Australasian Journal of Dermatology, conclude that the other provisions of (formerly 1978N–0038) and may be seen 38:S1–S6, 1997. this rule are not subject to OMB review by interested persons between 9 a.m. 14. World Health Organization, IARC, under the PRA. Section 201.327 and 4 p.m., Monday through Friday. ‘‘IARC Monographs on the Evaluation of contains specific labeling information, (FDA has verified all Web site Carcinogenic Risks to Humans. Vol. 55. including directions and warnings, addresses, but FDA is not responsible Solar and Ultraviolet Radiation,’’ http:// which are a ‘‘public disclosure of for any subsequent changes to the Web monographs.iarc.fr/ENG/Monographs/ information originally supplied by the sites after this document publishes in vol55/volume55.pdf, September 1, 2009. Federal Government to the recipient for the Federal Register. 15. ‘‘Report on Carcinogens, Eleventh the purpose of disclosure to the public’’ 1. FDA List of Docket Submissions Edition,’’ http://ntp.niehs.nih.gov/ (5 CFR 1320.3(c)(2)) and, therefore, are Addressed in This Rule; Organized by index.cfm?objectid=32BA9724–F1F6– not collections of information. The Issue. 975E–7FCE50709CB4C932, October 13, requirements for obtaining certain 2. The United States Pharmacopeia 2009. medical history information and 31–National Formulary 26, The United 16. Rabe, J. et al., ‘‘Photoaging: informed consent from test subjects (21 States Pharmacopeial Convention, Inc., Mechanisms and Repair,’’ Journal of the CFR 201.327(i)(3)(ii) and (i)(3)(iv)) are MD, p., 3547, 2009. American Academy of Dermatology, not collections of information because 3. The United States Pharmacopeia 55:1–19, 2006. information collected from subjects of 31–National Formulary 26, The United 17. Reynolds, K.D., et al., ‘‘Predictors clinical testing does not constitute States Pharmacopeial Convention, Inc., of Sun Exposure in Adolescents in a information under 5 CFR 1320.3(h)(5). MD, p., 3548, 2009. Southeastern U.S. Population,’’ Journal There are no recordkeeping provisions 4. Nash, J. F. et al., ‘‘Ultraviolet A of Adolescent Health, 19:409–415, 1996. associated with the SPF and broad Radiation: Testing and Labeling for 18. Robinson, J.K., D.S. Rigel, and spectrum testing (i.e., effectiveness Sunscreen Products,’’ Dermatologic R.A. Amonette, ‘‘Summertime Sun testing) described in this rule. The Clinics, 24:63–74, 2006. Protection Used by Adults for Their burdens of SPF testing as relevant to 5. Sayre, R. M. et al., ‘‘Commentary on Children,’’ Journal of the American labeling (third party disclosures) are ‘UVB–SPF’: the SPF Labels of Sunscreen Academy of Dermatology, 42:746–753, addressed in the notice published Products Convey More Than Just UVB 2000. elsewhere in this issue of the Federal Protection,’’ Photodermatology 19. Ulrich, C., et al., ‘‘Prevention of Register. Photoimmunology Photomedicine, Non-Melanoma Skin Cancer in Organ 24:218–220, 2008. Transplant Patients by Regular Use of a XI. Environmental Impact 6. Vainio, H. and F. Bianchini, Sunscreen: A 24 Months, Prospective, FDA has determined under 21 CFR ‘‘Cancer-Preventive Effects of Case-Control Study,’’ British Journal of 25.31(a) that this action is of a type that Sunscreens’’ in IARC Handbooks of Dermatology, 161:78–84, 2009. does not individually or cumulatively Cancer Prevention: Volume 5 20. Green, A. et al., ‘‘Daily Sunscreen have a significant effect on the human Sunscreens, IARC, Lyon, FR, 69–124, Application and Betacarotene environment. Therefore, neither an 2001. Supplementation in Prevention of Basal- environmental assessment nor an 7. ‘‘Commission Recommendation of Cell and Squamous-Cell Carcinomas of environmental impact statement is 22 September 2006 on the Efficacy of the Skin: A Randomized Controlled required. Sunscreen Products and the Claims Trial,’’ The Lancet, 354:723–729, 1999. Made Relating Thereto,’’ Official Journal 21. van der Pols, J. C. et al., XII. Federalism of the European Union, L265, 49:39–43, ‘‘Prolonged Prevention of Squamous FDA has analyzed this final rule in 2006. Cell Carcinoma of the Skin by Regular accordance with the principles set forth 8. Document No. FDA–1978–N–0018– Sunscreen Use,’’ Cancer Epidemiology, in Executive Order 13132. Section 4(a) 0692 in Docket No. FDA–1978–N–0018. Biomarkers, and Prevention, 15:2546– of the Executive order requires agencies 9. Calvacca, L., ‘‘Screening the New 2548, 2006. to ‘‘construe * * * a Federal statute to Sunscreens,’’ Working Woman, 15:120– 22. Habif, T. P. et al., ‘‘Premalignant preempt State law only where the 123, 1990. and Malignant Nonmelanoma Skin statute contains an express preemption 10. Fourtanier, A. et al., ‘‘Sunscreens Tumors’’ in Skin Disease Diagnosis and provision or there is some other clear Containing the Broad-Spectrum UVA Treatment, Mosby, St. Louis, MO, 360– evidence that the Congress intended absorber, Mexoryl SX, Prevent the 395, 2001. preemption of State law, or where the Cutaneous Detrimental Effects of UV 23. Araki, K. et al., ‘‘Incidence of Skin exercise of State authority conflicts with Exposure: A Review of Clinical Study Cancers and Precancerous Lesions in the exercise of Federal authority under Results,’’ Photodermatology, Japanese. Risk Factors and Prevention,’’ the Federal statute.’’ The sole statutory Photoimmunology, and Photomedicine, Journal of Epidemiology, 9:S14–S21, provision giving preemptive effect to the 24:164–174, 2008. 1999. final rule is section 751 of the FD&C Act 11. De Gruijl, F., ‘‘UV Radiation, DNA 24. Darlington, S. et al., ‘‘A (21 U.S.C. 379r). We have complied Damage, Mutations, and Skin Cancer’’ Randomized Controlled Trial to Assess

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00040 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35659

Sunscreen Application and Beta 37. Dowdy, J. C. et al., ‘‘Indoor 50. Kimlin, M. et al., ‘‘Does a High UV Carotene Supplementation in the Tanning Injuries: An Evaluation of FDA Environment Ensure Adequate Vitamin Prevention of Solar Keratoses,’’ Archives Adverse Event Reporting Data,’’ D Status?’’ Journal of Photochemistry of Dermatology, 139:451–455, 2003. Photodermatology Photoimmunology and Photobiology B. Biology, 89:139– 25. Naylor, M. F. et al., ‘‘High Sun Photomedicine, 25:216–220, 2009. 147, 2007. Protection Factor Sunscreens in the 38. Holick, M. F. and T. C. Chen, 51. Cusack, C et al., ‘‘Photoprotective Suppression of Actinic Neoplasia,’’ ‘‘Vitamin D Deficiency: A Worldwide Behavior and Sunscreen Use: Impact on Archives of Dermatology, 131:170–175, Problem with Health Consequences,’’ Vitamin D Levels in Cutaneous Lupus 1995. The American Journal of Clinical Erythematosus,’’ Photodermatology, 26. Thompson, S. C. et al., ‘‘Reduction Nutrition, 87:1080S–1086S, 2008. Photoimmunology, and Photomedicine, of Solar Keratoses by Regular Sunscreen 39. Cable News Network, ‘‘Vitamin D 24:260–267, 2008. Use,’’ New England Journal of Medicine, is Hot! Here’s How to Get it,’’ http:// 52. Hoesl, M. et al., ‘‘Vitamin D Levels 329:1147–1151, 1993. www.cnn.com/2008/HEALTH/ of XP–Patients Under Stringent Sun 27. Gallagher, R. P. et al., ‘‘Broad- diet.fitness/05/20/cl.vitamin.d/ Protection,’’ European Journal of Spectrum Sunscreen Use and the index.html, October 31, 2008. Dermatology, 20:457–460, 2010. Development of New Nevi in White 40. Holick, M. F., ‘‘Sunlight, UV– 53. Zerwekh, J. E., ‘‘Blood Biomarkers Children: A Randomized Controlled Radiation, Vitamin D and Skin Cancer: of Vitamin D Status,’’ American Journal Trial,’’ The Journal of the American How Much Sunlight do We Need?,’’ of Clinical Nutrition, 87:1087S–1091S, Medical Association, 283:2955–2960, Advances in Experimental Medicine 2008. 2000. and Biology, 624:1–15, 2008. 54. The American Academy of 28. Lee, T. K. et al., ‘‘Site-Specific 41. Maxwell, J. D., ‘‘Seasonal Dermatology, ‘‘Sun Protection for Protective Effect of Broad-Spectrum Variation in Vitamin D,’’ Proceedings of Children,’’ http://www.aad.org/public/ Sunscreen on Nevus Development the Nutrition Society, 53:533–543, 1994. publications/pamphlets/sun_ Among White Schoolchildren in a 42. Salih, F. M., ‘‘Effect of Clothing sunprotection.html, February 1, 2009. Randomized Trial,’’ Journal of the Varieties on Solar Photosynthesis of 55. The American Academy of American Academy of Dermatology, Previtamin D3: An in Vitro Study,’’ Dermatology, ‘‘Actinic Keratoses,’’ 52:786–792, 2005. Photodermatology, Photoimmunology, http://www.aad.org/public/ 29. Seite, S. and A. Fourtanier, ‘‘The and Photomedicine, 20:53–58, 2004. publications/pamphlets/sun_actinic. Benefit of Daily Photoprotection,’’ 43. Webb, A. R., ‘‘Who, What, Where html, February 1, 2009. Journal of the American Academy of and When-Influences on Cutaneous 56. The American Academy of Dermatology, 58:S160–S166, 2008. Vitamin D Synthesis,’’ Progress in Dermatology, ‘‘The Sun and Your Skin,’’ 30. Fourtanier, A. et al., ‘‘Protection of Biophysics and Molecular Biology, http://www.aad.org/public/ Skin Biological Targets by Different 92:17–25, 2006. publications/pamphlets/sun_sun.html, Types of Sunscreens,’’ 44. Institute of Medicine, ‘‘Vitamin D’’ February 1, 2009. Photodematology, Photoimmunology, in Dietary Reference Intakes For 57. The American Academy of and Photomedicine, 22:22–32, 2006. Calcium, Phosphorus, Magnesium, Dermatology, ‘‘The Darker Side of 31. Young, A. R. et al., ‘‘The Vitamin D, And Flouride, National Tanning,’’ http://www.aad.org/public/ Detrimental Effects of Daily Sub- Academies Press, Washington, DC, 250– publications/pamphlets/sun_darker. Erythemal Exposure on Human Skin In 287, 1997. html, February 1, 2009. Vivo Can Be Prevented by a Daily-Care 45. Wolpowitz, D. and B. A. Gilchrest, 58. The American Academy of Broad-Spectrum Sunscreen,’’ Journal of ‘‘The Vitamin D Questions: How Much Dermatology, ‘‘Skin Cancer,’’ http:// Investigative Dermatology, 127:975–978, Do You Need And How Should You Get www.aad.org/public/publications/ 2007. It?,’’ Journal of the American Academy pamphlets/sun_skin.html, February 1, 32. Moyal, D. D. and A. M. Fourtanier, of Dermatology, 54:301–17, 2006. 2009. ‘‘Broad-Spectrum Sunscreens Provide 46. Matsuoka, L. Y. et al., 59. Centers for Disease Control and Better Protection From Solar Ultraviolet- ‘‘Sunscreens Suppress Cutaneous Prevention, ‘‘Sunscreen for Your Sun Simulated Radiation and Natural Vitamin D3 Synthesis,’’ Journal of Day,’’ http://www.cdc.gov/cancer/skin/ Sunlight-Induced Immunosuppression Clinical Endocrinology and Metabolism, chooseyourcover/index.htm, October 23, In Human Beings,’’ Journal of the 64:1165–1168, 1987. 2009. American Academy Dermatology, 47. Matsuoka, L. Y. et al., ‘‘Chronic 60. ‘‘Sunwise Program: Action Steps 58:S149–54, 2008. Sunscreen Use Decreases Circlulating for Sun Safety,’’ http://www.epa.gov/ 33. ‘‘Label Information for AlliTM Concentrations of 25-hydroxyvitamin D. sunwise/actionsteps.html, February 1, Weight Loss Aid,’’ http://www.access A preliminary Study,’’ Archives of 2009. data.fda.gov/drugsatfda_docs/label/ Dermatology, 124:1802–1804, 1988. 61. Wright, M.W. et al., ‘‘Mechanisms 2007/021887lbl.pdf, November 23, 48. Marks, R. et al., ‘‘The Effect of of Sunscreen Failure,’’ Journal of the 2009. Regular Sunscreen Use on Vitamin D American Academy of Dermatology, 34. ‘‘Your Guide to Lowering Your Levels in an Australian Population. 44:781–784, 2001 Cholesterol with TLC. Therapeutic Results of a Randomized Controlled 62. Rigel, D., ‘‘American Academy of Lifestyle Changes,’’ NIH Publication No. Trial,’’ Archives of Dermatology, Dermatology’s Melanoma/Skin Cancer 06–5235, 2005. 131:415–421, 1995. Detection and Prevention Month Press 35. ‘‘Lipitor’’ in Physicians’ Desk 49. Farrerons, J. et al., ‘‘Clinically Release,’’ April 25, 2001. Reference, ed., Thomson Healthcare Prescribed Sunscreen (Sun Protection 63. European Cosmetic, Toiletry and Inc., Montvale, NJ, 2457–2462, 2008. Factor 15) Does not Decrease Serum Perfumery Association (COLIPA); JCIA; 36. ‘‘The Risks of Tanning,’’ http:// Vitamin D Concentration Sufficiently CTFA–SA; CTFA, ‘‘Sun Protection www.fda.gov/Radiation-Emitting Either to Induce Changes in Parathyroid Factor Test Method,’’ 1994. Products/RadiationEmittingProducts Function or in Metabolic Markers,’’ 64. European Cosmetic, Toiletry and andProcedures/Tanning/ucm116 British Journal of Dermatology, Perfumery Association (COLIPA); JCIA; 432.htm, October 21, 2009. 139:422–427, 1998. CTFA–SA; CTFA, ‘‘International Sun

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00041 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 35660 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

Protection Factor (SPF) Test Method,’’ Australia,’’ Cancer Causes & Control, test in paragraph (j) of this section. (A) 2006. 19:879–885, 2008. The labeling states ‘‘Broad Spectrum 65. Standards Australia and Standards 80. Eastern Research Group (2010) SPF [insert numerical SPF value New Zealand, ‘‘Australian/New Zealand ‘‘Sunscreen Drug Formulations for Over- resulting from testing under paragraph Standard Sunscreen Products the-Counter Human Use,’’ Task Order (i) of this section]’’. Evaluation and Classification,’’ AS/NZS No. 21, Contract No. 223–03–8500. (B) Prominence. The Broad Spectrum 2604:1998,’’ 1998. 81. RTI International, ‘‘FDA Labeling SPF statement shall appear as 66. Health Canada, ‘‘Category IV Cost Model, Final Report,’’ prepared by continuous text with no intervening text Monograph Sunburn Protectants. Mary Muth, Erica Gledhill, and Shawn or graphic. The entire text shall appear Procedure for Determining an SPF,’’ Karns, RTI. Prepared for Amber Jessup, in the same font style, size, and color 2002. FDA Center for Food Safety and with the same background color. 67. Korea Food and Drug Applied Nutrition, Revised January (ii) For sunscreen products that do Administration, ‘‘Korean Measurement 2003. not pass the broad spectrum test in Standards for UV Protection Efficacy,’’ 82. ‘‘Sunscreen Market Analysis: The paragraph (j) of this section. The 2001. Evolution and Use of UVA–1 Actives,’’ labeling states ‘‘SPF [insert numerical 68. Bimczok, R. et al., ‘‘Influence of The Proctor and Gamble Company, SPF value resulting from testing under Applied Quantity of Sunscreen Products Sharon Woods Technical Center, paragraph (i) of this section]’’. The on the Sun Protection Factor—A Cincinatti, OH, 2004. entire text shall appear in the same font Multicenter Study Organized by the List of Subjects style, size, and color with the same DGK Task Force Sun Protection,’’ Skin background color. Pharmacology and Physiology, 20:57– 21 CFR Part 201 (2) Water resistance statements. (i) For 64, 2007. Drugs, Incorporation by reference, products that provide 40 minutes of 69. COLIPA (The European Cosmetics Labeling, Reporting and recordkeeping water resistance according to the test in Association), ‘‘Method for the In Vitro requirements. paragraph (i)(7)(i) of this section. The Determination of UVA Protection labeling states ‘‘Water Resistant (40 21 CFR Part 310 Provided by Sunscreen Products,’’ minutes)’’. COLIPA Guidelines, 2007a. Administrative practice and (ii) For products that provide 80 70. Ferrero, L. et al., ‘‘Importance of procedure, Drugs, Labeling, Medical minutes of water resistance according to Substrate Roughness for In Vitro Sun devices, Reporting and recordkeeping the test in paragraph (i)(7)(ii) of this Protection Assessment,’’ IFSCC requirements. section. The labeling states ‘‘Water Magazine, 9:97–108, 2006. Therefore, under the Federal Food, Resistant (80 minutes)’’. 71. Sayre, R. M. and J. C. Dowdy, Drug, and Cosmetic Act and under (b) Statement of identity. The labeling ‘‘Photostability Testing of Avobenzone,’’ authority delegated to the Commissioner of the product contains the established Cosmetics and Toiletries, 114:85–91, of Food and Drugs, 21 CFR part 201 is name of the drug, if any, and identifies 1999. amended as follows: the drug as a ‘‘sunscreen.’’ 72. National Cancer Institute, ‘‘Skin (c) Indications. The labeling of the Cancer,’’ http://www.cancer.gov/cancer PART 201—LABELING product states, under the heading topics/types/skin, April 17, 2009. ■ 1. The authority citation for 21 CFR ‘‘Uses,’’ the phrases listed in this 73. National Cancer Institute, part 201 continues to read as follows: paragraph (c), as appropriate. Other ‘‘Surveillance Epidemiology and End truthful and nonmisleading statements, Results: Melanoma of the Skin,’’ Authority: 21 U.S.C. 321, 331, 351, 352, describing only the uses that have been http://seer.cancer.gov/statfacts/html/ 353, 355, 358, 360, 360b, 360gg–360ss, 371, established and listed in this paragraph melan.html, November 10, 2010. 374, 379e; 42 U.S.C. 216, 241, 262, 264. (c), may also be used, as provided in ■ 74. Bickers, D.R. et al., ‘‘The Burden 2. Section 201.327 is added to subpart § 330.1(c)(2) of this chapter, subject to of Skin Diseases: 2004,’’ Journal of the G to read as follows: the provisions of section 502 of the American Academy of Dermatology, § 201.327 Over-the-counter sunscreen Federal Food, Drug, and Cosmetic Act 55:490–500, 2006. drug products; required labeling based on (the FD&C Act) relating to misbranding 75. Freedberg, K.A. et al., ‘‘Screening effectiveness testing. and the prohibition in section 301(d) of for Mealignant Melanoma: A Cost- The following provisions apply to the FD&C Act against the introduction Effective Analysis,’’ Journal of the sunscreen products containing or delivery for introduction into American Academy of Dermatology, aminobenzoic acid, avobenzone, interstate commerce of unapproved new 41:738–745, 1999. cinoxate, dioxybenzone, ensulizole, drugs in violation of section 505(a) of 76. Arias, E., B.L. Rostron, and B. homosalate, meradimate, octinoxate, the FD&C Act. Tejada-Vera, ‘‘United States Life tables, octisalate, octocrylene, oxybenzone, (1) For all sunscreen products, the 2005,’’ National Vital Statistics Reports, padimate O, sulisobenzone, titanium following indication statement must be 58:1–132, 2010. dioxide, trolamine salicylate, or zinc included under the heading ‘‘Uses’’: 77. Lewis, K.G. and M.A. Weinstock, oxide, alone or in combination. The ‘‘[Bullet] helps prevent sunburn’’. See ‘‘Nonmelanoma Skin Cancer Mortality provisions do not apply to sunscreen § 201.66(b)(4) of this chapter for (1988–2000),’’ Archives of Dermatology, products marketed under approved new definition of bullet. 140:837–842, 2004. drug applications or abbreviated new (2) For sunscreen products with a 78. Nolan, R.C., M. T.-L. Chan, and drug applications. Broad Spectrum SPF value of 15 or P.J. Heenan, ‘‘A Clinicopathologic (a) Principal display panel. In higher according to the tests in Review of Lethal Nonmelanoma Skin addition to the statement of identity in paragraphs (i) and (j) of this section, the Cancers in Western Australia,’’ Journal paragraph (b) of this section, the labeling may include the following of the American Academy of following labeling shall be prominently statement in addition to the indication Dermatology, 52:101–108, 2005. placed on the principal display panel: in § 201.327(c)(1): ‘‘[Bullet] if used as 79. Girschik, J. et al., ‘‘Deaths from (1) Effectiveness claim. (i) For directed with other sun protection Non-Melanoma Skin Cancer in Western products that pass the broad spectrum measures (see Directions [in bold italic

VerDate Mar<15>2010 20:46 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00042 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35661

font]), decreases the risk of skin cancer a.m.–2 p.m. [bullet] wear long-sleeved SOLAR SIMULATOR EMISSION and early skin aging caused by the sun’’. shirts, pants, hats, and sunglasses’’. SPECTRUM—Continued (3) Any labeling or promotional (3) For products that satisfy the water resistance test in paragraph (i)(7) of this materials that suggest or imply that the Wavelength range (nm) Percent erythemal use, alone, of any sunscreen reduces the section. The labeling states ‘‘[bullet] contribution 1 risk of or prevents skin cancer or early reapply: [Bullet] after [select one of the skin aging will cause the product to be following determined by water 290–310 ...... 49.0–65.0 resistance test: ‘40 minutes of’ or ‘80 290–320 ...... 85.0–90.0 misbranded under section 502 of the 290–330 ...... 91.5–95.5 FD&C Act (21 U.S.C. 352). minutes of’] swimming or sweating 290–340 ...... 94.0–97.0 (d) Warnings. The labeling of the [bullet] immediately after towel drying 290–400 ...... 99.9–100.0 product contains the following warnings [bullet] at least every 2 hours’’. 1 under the heading ‘‘Warnings’’. (4) For products that do not satisfy the Calculation of erythema action spectrum described in § 201.327(i)(1)(ii) of this section. (1) For all sunscreen products. (i) The water resistance test in paragraph (i)(7) labeling states ‘‘Do not use [bullet] on of this section. The labeling states (B) In addition, UVA II (320–340 nm) damaged or broken skin’’. ‘‘[bullet] reapply at least every 2 hours irradiance should equal or exceed 20 (ii) The labeling states ‘‘When using [bullet] use a water resistant sunscreen percent of the total UV (290–400 nm) this product [bullet] keep out of eyes. if swimming or sweating’’. irradiance. UVA I (340–400 nm) Rinse with water to remove.’’ (f) Other information. The labeling of irradiance should equal or exceed 60 (iii) The labeling states ‘‘Stop use and the product contains the following percent of the total UV irradiance. ask a doctor if [bullet] rash occurs’’. statement under the heading ‘‘Other (ii) Erythema action spectrum. (A) (2) For sunscreen products that are information:’’ ‘‘[bullet] protect the Calculate the erythema action spectrum broad spectrum with SPF values of at product in this container from excessive weighting factor (Vi) at each wavelength λ least 2 but less than 15 according to the heat and direct sun’’. : (g) False and misleading claims. (1) Vi (λ) = 1.0 (250 < λ ≤ 298 nm) SPF test in paragraph (i) of this section λ or that do not pass the broad spectrum There are claims that would be false (2) Vi ( ) = 0.094 * (298 ndash; lambda;) λ test in paragraph (j) of this section. The and/or misleading on sunscreen 10 (298 < products. These claims include but are ≤ 328 nm) first statement under the heading λ ‘‘Warnings’’ states ‘‘Skin Cancer/Skin not limited to the following: (3) Vi ( ) = 0.015 * (140 ndash; lambda;) λ Aging Alert [in bold font]; Spending ‘‘Sunblock,’’ ‘‘sweatproof,’’ and 10 (328 < time in the sun increases your risk of ‘‘waterproof.’’ These or similar claims 400 nm) skin cancer and early skin aging. This will cause the product to be misbranded (B) Calculate the erythema-effective product has been shown only to help under section 502 of the FD&C Act (21 UV dose (E) delivered by a solar prevent sunburn, not [in bold font] skin U.S.C. 352). simulator as follows: cancer or early skin aging.’’ (h) Labeling of products containing a (e) Directions. The labeling of the combination of sunscreen and skin product contains the following protectant active ingredients. Statements of identity, indications, Where Vi(λ) = erythema action spectrum statements, as appropriate, under the λ heading ‘‘Directions.’’ More detailed warnings, and directions for use, weighting factor at each wavelength I(λ) = irradiance (Watts per square meter) directions applicable to a particular respectively, applicable to each ingredient in the product may be at each wavelength λ product formulation may also be t = exposure time (seconds) included. combined to eliminate duplicative (1) For all sunscreen products. (i) As words or phrases so that the resulting Erythema-effective dose (E) is expressed an option, the labeling may state ‘‘For information is clear and understandable. as effective Joules per square meter (J/ sunscreen use:’’. Labeling provisions in § 347.50(e) of this m2-eff). (ii) The labeling states ‘‘[bullet] apply chapter shall not apply to these (C) The emission spectrum must be [select one of the following: ‘Liberally’ products. determined using a handheld or ‘generously’] [and, as an option: ‘And (i) SPF test procedure. (1) UV source radiometer with a response weighted to evenly’] 15 minutes before sun (solar simulator). (i) Emission spectrum. match the spectrum in ISO 17166 CIE S exposure’’. A single port or multiport solar 007/E entitled ‘‘Erythemal reference (iii) As an option, the labeling may simulator should be filtered so that it action spectrum and standard erythema state ‘‘[bullet] apply to all skin exposed provides a continuous emission dose,’’ dated 1999 (First edition, 1999– to the sun’’. spectrum from 290 to 400 nanometers 12–15; corrected and reprinted 2000– (nm) with a limit of 1,500 Watts per 11–15), which is incorporated by (iv) The labeling states ‘‘[bullet] 2 children under 6 months of age: Ask a square meter (W/m ) on total irradiance reference in accordance with 5 U.S.C. doctor’’. for all wavelengths between 250 and 552(a) and 1 CFR part 51. You may (2) For sunscreen products with a 1,400 nm. obtain a copy from the ISO Copyright (A) The solar simulator should have Broad Spectrum SPF value of 15 or Office, Case Postale 56, CH–1211, the following percentage of erythema- higher according to the tests in Geneva 20, Switzerland, telephone +41– effective radiation in each specified paragraphs (i) and (j) of this section. 22–749–01–11 or fax +41–22–74 -09–47. range of wavelengths: The labeling states ‘‘[bullet] Sun http://www.iso.org. You may inspect a Protection Measures. [in bold font] copy at the Center for Drug Evaluation SOLAR SIMULATOR EMISSION Spending time in the sun increases your and Research, 10903 New Hampshire risk of skin cancer and early skin aging. SPECTRUM Ave., Bldg. 22, Silver Spring, MD 20993, To decrease this risk, regularly use a call 301–796–2090, or at the National Wavelength range (nm) Percent erythemal sunscreen with a Broad Spectrum SPF contribution 1 Archives and Records Administration value of 15 or higher and other sun (NARA). For information on the protection measures including: [Bullet] < 290 ...... < 0.1 availability of this material at NARA, limit time in the sun, especially from 10 290–300 ...... 1.0–8.0 call 202–741–6030, or go to: http://

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00043 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.002 35662 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

www/archives.gov/federal_register/ COMPOSITION OF THE PADIMATE O/ (2) Use HPLC grade reagents for code_offederal_regulations/ibr_ OXYBENZONE SPF STANDARD mobile phase. locations.html. The solar simulator (B) Preparation of the HPLC reference output should be measured before and Ingredients Percent by standard. (1) Weigh 0.50 gram (g) of after each phototest or, at a minimum, weight oxybenzone USP reference standard into a 250-mL volumetric flask. Dissolve and at the beginning and end of each test Part A: day. This radiometer should be Lanolin ...... 4.50 dilute to volume with isopropanol. Mix calibrated using side-by-side Cocoa butter ...... 2.00 well. comparison with the spectroradiometer Glyceryl monostearate ...... 3.00 (2) Weigh 0.50 g of padimate O USP (using the weighting factors determined Stearic acid ...... 2.00 reference standard into a 250-mL according to paragraph (i)(1)(ii)(A) of Padimate O ...... 7.00 volumetric flask. Dissolve and dilute to Oxybenzone ...... 3.00 this section) at the time of the annual volume with isopropanol. Mix well. Part B: (3) Pipet 3.0 mL of the oxybenzone spectroradiometric measurement of the Purified water USP ...... 71.60 solar simulator as described in Sorbitol solution ...... 5.00 solution and 7.0 mL of the padimate O paragraph (i)(1)(iv) of this section. Triethanolamine, 99 percent 1.00 solution into a 100-mL volumetric flask. Dilute to volume with isopropanol and (iii) Operation. A solar simulator Methylparaben ...... 0.30 Propylparaben ...... 0.10 mix well. should have no significant time-related Part C: (C) HPLC system suitability. (1) Make fluctuations (within 20 percent) in Benzyl alcohol ...... 0.50 three replicate 10-microliter injections radiation emissions after an appropriate Part D: 1 of the HPLC reference standard warm-up time and demonstrate good Purified water USP ...... QS (described in paragraph (i)(2)(ii)(B) of beam uniformity (within 20 percent) in 1 Quantity sufficient to make 100 grams. this section). The relative standard the exposure plane. The delivered dose deviation in peak areas should not be to the UV exposure site must be within Step 1. Add the ingredients of Part A into a suitable stainless steel kettle more than 2.0 percent for either 10 percent of the expected dose. equipped with a propeller agitator. Mix oxybenzone or padimate O. (iv) Periodic measurement. To ensure at 77 to 82 °C until uniform. (2) Calculate the resolution (R) that the solar simulator delivers the Step 2. Add the water of Part B into between the oxybenzone and padimate appropriate spectrum of UV radiation, a suitable stainless steel kettle equipped O peaks from one chromatogram as the emission spectrum of the solar with a propeller agitator and begin follows: simulator should be measured at least mixing at 77 to 82 °C. Add the annually with an appropriate and remaining ingredients of Part B and mix accurately calibrated spectroradiometer until uniform. system (results should be traceable to Step 3. Add the batch of Step 1 to the ° the National Institute for Standards and batch of Step 2 and mix at 77 to 82 C Where to = retention time for oxybenzone Technology). In addition, the solar until smooth and uniform. Slowly cool tp = retention time for padimate O ° simulator must be recalibrated if there is the batch to 49 to 54 C. Wo = oxybenzone peak width at baseline Step 4. Add the benzyl alcohol of Part Wp = padimate O peak width at baseline any change in the lamp bulb or the C to the batch of Step 3 at 49 to 54 °C. optical filtering components (i.e., filters, Mix until uniform. Continue to cool If the resolution (R) is less than 3.0, mirrors, lenses, collimating devices, or batch to 35 to 41 °C. adjust the mobile phase or replace the focusing devices). Daily solar simulator Step 5. Add sufficient water of Part D column. radiation intensity should be monitored to the batch of Step 4 at 35 to 41 °C to (D) SPF standard assay. with a broadband radiometer with a obtain 100 grams of SPF standard. Mix (1) The SPF standard is diluted to the response weighted to match the until uniform. Cool batch to 27 to 32 °C. same concentration as the HPLC erythema action spectrum in ISO 17166 (ii) HPLC assay. Use the following reference standard according to the CIE S 007/E entitled ‘‘Erythemal high performance liquid following steps: reference action spectrum and standard chromatography (HPLC) procedure to (i) Step 1. Weigh 1.0 g of the SPF erythema dose,’’ which is incorporated verify the concentrations of padimate O standard (described in paragraph (i)(2)(i) by reference in paragraph (i)(1)(ii)(C) of and oxybenzone in the SPF standard: of this section) into a 50-mL volumetric this section. If a lamp must be replaced (A) Instrumentation. (1) Equilibrate a flask. due to failure or aging during a suitable liquid chromatograph to the (ii) Step 2. Add approximately 30 mL phototest, broadband device readings following or equivalent conditions: of isopropanol and heat with swirling consistent with those obtained for the until contents are evenly dispersed. original calibrated lamp will suffice (i) Column ..... C–18, 250 millimeters (mm) (iii) Step 3. Cool to room temperature length, 4.6 mm inner di- (15 to 30 °C) and dilute to volume with until measurements can be performed ameter (5 microns) with the spectroradiometer at the (ii) Mobile 85:15:0.5 methanol: water: isopropanol. Mix well. earliest possible opportunity. Phase. acetic acid (iv) Step 4. Pipet 5.0 mL of the preparation into a 50-mL volumetric (2) SPF standard. (i) Preparation. The (iii) Flow Rate 1.5 milliliters (mL) per flask and dilute to volume with SPF standard should be a formulation minute (iv) Tempera- Ambient isopropanol. Mix well. containing 7-percent padimate O and 3- ture. (2)(i) Inject 10-microliter of diluted percent oxybenzone. (v) Detector ... UV spectrophotometer at SPF standard from paragraph (i)(2)(D)(1) 308 nanometers of this section and calculate the amount (vi) Attenu- As needed ation. of oxybenzone and padimate O as follows:

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00044 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.003 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35663

(ii) The percent of oxybenzone and temporarily alter the skin’s response to An MEDp is determined for the SPF padimate O in the SPF standard should UV radiation. standard (ssMEDp). An MEDp is be between 95 and 105. (iv) Informed consent. Obtain legally determined for the sunscreen test (3) Test subjects. (i) Number of effective written informed consent from product (tpMEDp). subjects. A test panel should include all test subjects. (ii) UV exposure for initial MEDu. For enough subjects to produce a minimum (4) Sunscreen application. (i) Test each test subject, administer a series of of 10 valid test results. A maximum of site. Test sites are locations on each UV radiation doses expressed as J/m2-eff three subjects may be rejected from this subject’s back, between the beltline and (as determined according to paragraph panel based on paragraph (i)(5)(v)) of the shoulder blades (scapulae) and (a)(2) of this section) to the test subsites this section. lateral to the midline, where skin within an unprotected test site using an (ii) Medical history. (A) Obtain a responses to UV radiation are accurately calibrated solar simulator. medical history from each subject with determined. Responses on unprotected Select doses that are a geometric series emphasis on the effects of sunlight on skin (no test material applied) and represented by 1.25n (i.e., each dose is the subject’s skin. Determine that each protected skin (sunscreen test product(s) 25 percent greater than the previous subject is in good general health with or SPF standard applied) are determined dose). skin type I, II, or III as follows: at separate unprotected and protected (iii) UV exposure for final MEDu, (1) Always burns easily; never tans test sites, respectively. Test sites should ssMEDp, and tpMEDp. For each subject, (sensitive). be randomly located in a blinded determine the final MEDu, ssMEDp, and (2) Always burns easily; tans manner. Each test site should be a tpMEDp by administering a series of five minimally (sensitive). minimum of 30 square centimeters and UV doses to the appropriate test sites. (3) Burns moderately; tans gradually outlined with indelible ink. The middle dose (X) in each of these (light brown) (normal). (ii) Test subsite. Test subsites are the dose series (i.e., the third dose) should (4) Burns minimally; always tans well locations to which UV radiation is equal the initial MEDu times the (moderate brown) (normal). administered within a test site. At least expected SPF. Note that the expected (5) Rarely burns; tans profusely (dark five test subsites should receive UV SPF equals 1 and 16.3 for the final brown) (insensitive). doses within each test site. Test subsites MEDu and ssMEDp, respectively. The (6) Never burns; deeply pigmented should be at least 0.5 square centimeters remaining UV doses in the series (insensitive). (cm2) in area and should be separated depend upon the expected SPF value of (B) Skin type is based on first 30 to from each other by at least 0.8 cm. Each the sunscreen test product(s). 45 minutes of sun exposure after a test subsite should be outlined with For products with an expected SPF winter season of no sun exposure. indelible ink. less than 8, administer UV doses that Determine that each subject is not taking (iii) Applying test materials. Apply increase by 25 percent with each topical or systemic medication that is the sunscreen test product and the SPF successive dose (i.e., 0.64X, 0.80X, known to alter responses to UV standard at 2 milligrams per square 1.00X, 1.25X, and 1.56X). For products radiation. Determine that each subject centimeter (mg/cm2) to their respective with an expected SPF from 8 to 15, has no history of sensitivities to topical test sites. Use a finger cot compatible administer UV doses that increase by 20 products and/or abnormal responses to with the sunscreen to spread the percent with each successive dose (i.e., sunlight, such as a phototoxic or product as evenly as possible. 0.69X, 0.83X, 1.00X, 1.20X, and 1.44X). photoallergic response. (iv) Waiting period. Wait at least 15 For products with an expected SPF (iii) Physical examination. Conduct a minutes after applying a sunscreen higher than 15, administer UV doses physical examination to determine the product before exposing the test sites to that increase by 15 percent with each presence of sunburn, suntan, scars, UV radiation as described in paragraph successive dose (i.e., 0.76X, 0.87X, active dermal lesions, and uneven skin (i)(5)) of this section. For water resistant 1.00X, 1.15X, and 1.32X). tones on the areas of the back to be sunscreen products, proceed with the (iv) Evaluation of test subsites. In tested. A suitable source of low power water resistance testing procedure order that the person who evaluates the UVA, such as a Woods lamp, is helpful described in paragraph (i)(7) of this test subsites is not biased, he/she should in this process. If any of these section after waiting at least 15 minutes. not be the same person who applied the conditions are present, the subject is not (5) UV exposure. (i) Definition of sunscreen drug product to the test site qualified to participate in the study. The minimal erythema dose (MED). The or administered the UV doses. After UV presence of nevi, blemishes, or moles minimal erythema dose (MED) is the doses are administered, all immediate will be acceptable if, in the physician’s smallest UV dose that produces responses should be recorded. These judgment, they will neither compromise perceptible redness of the skin may include an immediate darkening or the study nor jeopardize a subject’s (erythema) with clearly defined borders tanning, typically grayish or purplish in safety. Subjects with dysplastic nevi at 16 to 24 hours after UV exposure. The color, which fades in 30 to 60 minutes; should not be enrolled. Excess hair on MED for unprotected skin (MEDu) is an immediate reddening at the subsite, the back is acceptable if the hair is determined on a test site that does not due to heating of the skin, which fades clipped. Shaving is unacceptable have sunscreen applied. The MED for rapidly; and an immediate generalized because it may remove a significant protected skin (MEDp) is determined on heat response, spreading beyond the portion of the stratum corneum and a test site that has sunscreen applied. subsite, which fades in 30 to 60

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00045 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.004 35664 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations

minutes. After the immediate responses (i) Water resistance (40 minutes). The to maximize capture of forward are noted, each subject should shield labeled SPF should be determined after scattered radiation. The sample holder the exposed area from further UV 40 minutes of water immersion using should be a thin, flat plate with a radiation until the MED is determined. the following procedure: suitable aperture through which UV Determine the MED 16 to 24 hours after (A) Step 1: Apply the sunscreen as radiation can pass. The PMMA plate UV exposure. Because erythema is described in paragraph (d) of this should be placed on the upper surface evaluated 16 to 24 hours after UV section. of the sample holder with the exposure, the final MEDu, ssMEDp, and (B) Step 2: Perform moderate activity roughened side facing up. tpMEDp are typically determined the in water for 20 minutes. (iii) Light source. The light source day following determination of the (C) Step 3: Rest out of water for 15 should produce a continuous spectral initial MEDu. Evaluate the erythema minutes. Do not towel test site(s). distribution of UV radiation from 290 to responses of each test subsite using (D) Step 4: Perform moderate activity 400 nanometers. either tungsten or warm white in water for 20 minutes. (iv) Input optics. Unless the fluorescent lighting that provides at (E) Step 5: Allow test sites to dry spectrometer is equipped with an least 450 lux of illumination at the test completely without toweling. integrating sphere, an ultraviolet site. For the evaluation, the test subject (F) Step 6: Apply the SPF standard as radiation diffuser should be placed should be in the same position as when described in paragraph (d) of this between the sample and the input optics the test site was irradiated. section. of the spectrometer. The diffuser will be (v) Invalid test data. Reject test data Step 1. Expose test sites to UV doses constructed from any UV radiation ® for a test subject if erythema is not as described in paragraph (e) of this transparent material (e.g., Teflon or present on either the unprotected or section. quartz). The diffuser ensures that the protected test sites; or erythema is (ii) Water resistance (80 minutes). The radiation received by the spectrometer present at all subsites; or the responses labeled SPF should be determined after is not collimated. The spectrometer are inconsistent with the series of UV 80 minutes of water immersion using input slits should be set to provide a doses administered; or the subject was the following procedure: bandwidth that is less than or equal to noncompliant (e.g., the subject (A) Step 1: Apply the sunscreen as 1 nanometer. withdraws from the test due to illness described in paragraph (d) of this (v) Dynamic range of the or work conflicts or does not shield the section. spectrometer. The dynamic range of the exposed testing sites from further UV (B) Step 2: Perform moderate activity spectrometer should be sufficient to radiation until the MED is determined). in water for 20 minutes. measure transmittance accurately (6) Determination of SPF. (i) Calculate (C) Step 3: Rest out of water for 15 through a highly absorbing sunscreen product at all terrestrial solar UV an SPF value for each test subject (SPFi) minutes. Do not towel test site(s). as follows: (D) Step 4: Perform moderate activity wavelengths (290 to 400 nm). in water for 20 minutes. (2) Sunscreen product application to (E) Step 5: Rest out of water for 15 PMMA plate. The accuracy of the test minutes. Do not towel test site(s). depends upon the application of a (F) Step 6: Perform moderate activity precisely controlled amount of (ii) Calculate the mean in water for 20 minutes. sunscreen product with a uniform (G) Step 7: Rest out of water for 15 distribution over the PMMA plate. The minutes. Do not towel test site(s). product is applied at 0.75 mg per square and the standard deviation (s) from the (H) Step 8: Perform moderate activity centimeter to the roughened side of the in water for 20 minutes. PMMA plate. The sunscreen product SPFi values. Calculate the standard error (SE), which equals s/√n (where n equals (I) Step 9: Allow test sites to dry should be applied in a series of small the number of subjects who provided completely without toweling. dots over the entire PMMA plate and (J) Step 10: Apply the SPF standard as valid test results). Obtain the t value then spread evenly using a gloved described in paragraph (d) of this from Student’s t distribution table finger. Spreading should be done with section. corresponding to the upper 5-percent a very light spreading action for (K) Step 11: Expose test sites to UV point with n—1 degrees of freedom. approximately 30 seconds followed by doses as described in paragraph (e) of Determine the labeled SPF value, which spreading with greater pressure for this section. equals the largest whole number less approximately 30 seconds. The plate (j) Broad spectrum test procedure. (1) than should then be allowed to equilibrate UV Spectrometry. (i) Plate. Use optical- for 15 minutes in the dark before the grade polymethylmethacrylate (PMMA) pre-irradiation described in paragraph plates suitable for UV transmittance (c) of this section. In order for the SPF determination of measurements. The plate should be (3) Sunscreen product pre-irradiation. a test product to be considered valid, roughened on one side to a three To account for lack of photostability, the SPF value of the SPF standard dimensional surface topography apply the sunscreen product to the should fall within the standard measure (Sa) between 2 and 7 PMMA plate as described in paragraph deviation range of the expected SPF micrometers and must have a (b) of this section and then irradiate (i.e., 16.3 ± 3.43). rectangular application area of at least with a solar simulator described in (7) Determination of water resistance. 16 square centimeters (with no side section 352.70(b) of this chapter. The The following procedure should be shorter than 4 cm). irradiation dose should be 4 MEDs performed in an indoor fresh water (ii) Sample holder. The sample holder which is equivalent to an erythemal pool, whirlpool, and/or hot tub should hold the PMMA plate in a effective dose of 800 J/m2 (i.e., 800 J/m2- maintained at 23 to 32 °C. Fresh water horizontal position to avoid flowing of eff). is clean drinking water that meets the the sunscreen drug product from one (4) Calculation of mean transmittance standards in 40 CFR part 141. The pool edge of the PMMA plate to the other. It values. After pre-irradiation described and air temperature and the relative should be mounted as close as possible in paragraph (c) of this section, mean humidity should be recorded. to the input optics of the spectrometer transmittance values should be

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00046 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.005 ER17JN11.009 ER17JN11.012 Federal Register / Vol. 76, No. 117 / Friday, June 17, 2011 / Rules and Regulations 35665

determined for each wavelength λ over DEPARTMENT OF HEALTH AND the full UV spectrum (290 to 400 HUMAN SERVICES nanometers). The transmittance values should be measured at 1 nanometer Food and Drug Administration intervals. Measurements of spectral Where λc = critical wavelength 21 CFR Parts 201 and 310 irradiance transmitted for each λ wavelength λ through control PMMA A( ) = mean absorbance at each wavelength dλ = wavelength interval between [Docket No. FDA–2010–D–0509] plates coated with 15 microliters of measurements glycerin (no sunscreen product) should Draft Guidance for Industry on be obtained from at least 5 different A mean critical wavelength of 370 nm Enforcement Policy for Over-the- λ locations on the PMMA plate [C1( ), or greater is classified as broad spectrum Counter Sunscreen Drug Products λ λ λ λ C2( ), C3( ), C4( ), and C5( )]. In protection. Marketed Without an Approved addition, a minimum of 5 measurements Application; Availability of spectral irradiance transmitted for PART 310—NEW DRUGS each wavelength λ through the PMMA AGENCY: Food and Drug Administration, plate covered with the sunscreen ■ 4. The authority citation for 21 CFR HHS. product will be similarly obtained after part 310 continues to read as follows: ACTION: Notice of availability. pre-irradiation of the sunscreen product λ λ λ λ λ Authority: 21 U.S.C. 321, 331, 351, 352, SUMMARY: The Food and Drug [P1( ), P2( ), P3( ), P4( ), and P5( )]. 353, 355, 360b–360f, 360j, 361(a), 371, 374, The mean transmittance for each Administration (FDA) is announcing the 375, 379e; 42 U.S.C. 216, 241, 242(a), 262, availability of a draft guidance for wavelength, 263b–263n. industry entitled ‘‘Enforcement Policy— ■ 5. Section 310.545 is amended by OTC Sunscreen Drug Products Marketed revising paragraphs (a)(29) and (d)(31) Without an Approved Application.’’ is the ratio of the mean of the C(λ) and by adding new paragraph (d)(40) to The draft guidance is intended to inform values to the mean of the P(λ) values, as read as follows: manufacturers of over-the-counter follows: (OTC) sunscreen products about our § 310.545 Drug products containing enforcement policy for certain OTC certain active ingredients offered over-the- sunscreen products marketed without counter (OTC) for certain uses. an approved new drug application. The (a) * * * draft guidance describes our intended (29) Sunscreen drug products. approach to enforcement for certain ≥ Where n 5 (i) Ingredients. OTC sunscreen products prior to an effective final monograph. (5) Calculation of mean absorbance Diethanolamine methoxycinnamate DATES: values. (i) Mean transmittance values, Although you can comment on Digalloyl trioleate any guidance at any time (see 21 CFR Ethyl 4-[bis(hydroxypropyl)] 10.115(g)(5)), to ensure that the Agency aminobenzoate considers all comments on this draft are converted into mean absorbance Glyceryl aminobenzoate guidance before it begins work on the values, final version of the guidance, submit Lawsone with dihydroxyacetone either electronic or written comments Red petrolatum on the draft guidance by August 16, 2011. Submit written comments on the at each wavelength by taking the (ii) Any ingredients labeled with any of the following or similar claims. proposed collection of information by negative logarithm of the mean August 16, 2011. transmittance value as follows: Instant protection or protection immediately upon application. ADDRESSES: Submit written requests for Claims for ‘‘all-day’’ protection or single copies of the draft guidance to the extended wear claims citing a specific Division of Drug Information, Center for (ii) The calculation yields 111 number of hours of protection that is Drug Evaluation and Research, Food monochromatic absorbance values in 1 inconsistent with the directions for and Drug Administration, 10903 New nanometer increments from 290 to 400 application in 21 CFR 201.327. Hampshire Ave., Bldg. 51, rm. 2201, nanometers. Silver Spring, MD 20993–0002. Send * * * * * (6) Number of plates. For each one self-addressed adhesive label to sunscreen product, mean absorbance (d) * * * assist that office in processing your values should be determined from at (31) December 31, 2002, for products requests. See the SUPPLEMENTARY least three individual PMMA plates. subject to paragraph (a)(29)(i) of this INFORMATION section for electronic Because paragraph (d) of this section section. access to the draft guidance document. requires at least 5 measurements per * * * * * Submit electronic comments on the draft guidance to http:// plate, there should be a total of at least (40) June 18, 2012, for products www.regulations.gov. Submit written 15 measurements. subject to paragraph (a)(29)(ii) of this comments to the Division of Dockets (7) Calculation of the critical section. June 17, 2013, for products with Management (HFA–305), Food and Drug wavelength. The critical wavelength is annual sales less than $25,000. identified as the wavelength at which Administration, 5630 Fishers Lane, rm. the integral of the spectral absorbance Dated: June 9, 2011. 1061, Rockville, MD 20852. curve reaches 90 percent of the integral Leslie Kux, FOR FURTHER INFORMATION CONTACT: over the UV spectrum from 290 to 400 Acting Assistant Commissioner for Policy. Reynold Tan, Center for Drug nm. The following equation defines the [FR Doc. 2011–14766 Filed 6–14–11; 8:45 am] Evaluation and Research, Food and critical wavelength: BILLING CODE 4160–01–P Drug Administration, 10903 New

VerDate Mar<15>2010 18:47 Jun 16, 2011 Jkt 223001 PO 00000 Frm 00047 Fmt 4701 Sfmt 4700 E:\FR\FM\17JNR3.SGM 17JNR3 emcdonald on DSK2BSOYB1PROD with RULES3 ER17JN11.010 ER17JN11.006 ER17JN11.010 ER17JN11.011 ER17JN11.008 ER17JN11.007