View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

MODIFICATION OF CARCINOGENESIS BY PHOTOACTIVE AGENTS*

PRELIMINARY REPORT FREDERICK URBACH, M.D.

That prolonged exposure to ultraviolet lightranged in parallel in a specially designed polished can produce cancer of the skin and of other ex-aluminum reflectort. The mice were kept in a wire posed organs (eye, mucosa, etc.) has long beenbasket cage constructed so as to prevent huddling suspected in humans (1) and has been experi-or movement, placed 42 cm. below the fluorescent mentally proven in animals (2—7). The detailsfixture. The entire unit was built into a standard of absorption of ultraviolet energy of variouslaboratory fume hood with a glass sliding door, partly to protect the laboratory personnel, partly wavelengths by the skin have been well re-to allow for evacuation of ionized air. (Fig. 1) viewed by Ellis and Wells (8), and the voluminousThe output of the fluorescent "sun" lamps was literature concerning ultraviolet carcinogenesismeasured continuously during each treatment brought up to date by Blum (9, 10). No attemptwith a Westinghouse Rentschler "click" meter will be made in this communication to discuss(Model SM200) equipped with a WL767 photo- the many and varied studies concerned withtube and precalibrated against a standard lamp this problem which have been carried out sincetested by the United States Bureau of Standards. Findlay (2) demonstrated that cutaneous cancerEach click of the meter represented 13.69 X 10 ergs/cm2, and each group of animals received 25 could be produced in the skin of mice by repeatedclick units (34.225 X 1O ergs/cm2) daily. The exposure to mercury arc radiation. relative energy distribution of the ultraviolet The development of a pure chemical compound,source used is shown in Figure 2. As can be seen, methoxsalen, capable of producing extremethe lamps emit a continuous spectrum from 2750 photosensitivity in the skin of humans andto 3600 A. animals, has reawakened interest in the relation- Except during the ultraviolet exposures, all ship of photoactive agents to ultraviolet carcino-mice were kept in plastic cages, eight animals to genesis. The experiments to be reported here arethe box, and given ground purina chow and water in part a repetition, and in part an extension ofad libitum. All animals were housed in the same the work of O'Ncal and Griffin (11) who notedroom under identical conditions of light and tem- perature. that methoxsalen given intraperitoncally po- The animals were weighed, sexed and randomly tentiated, while the given orally pro-divided into groups of 24 each, and then pretreated tected against ultraviolet induced skin cancer inas follows: albino mice. Group I—Topical Administration METHOD Ta. Methoxsalen—0.1% in absolute ethanol Groups of randomly selected female ICR Swiss Tb. Crude —3.0% in chloroform mice, obtained from the Roswell Park Memorial Ic. Atabrine—0.1% in absolute alcohol Institute breeding colony at six weeks of age, were Id. Hematoporphyrin—0.1% in absolute alco- exposed to whole body radiation from Westing- hol house fluorescent sunlamps five days weekly for le. Control—untreated a period of 175 days. The ultraviolet source con- Group 11—0 rot Administration sisted of four FS4OT12 Westinghouse lamps, ar- ha. Methoxsalen—0.5 gm/kg diet *Fromthe Departments of Dermatology and JIb. Atabrine—50 mg/kg diet Physiology, Roswell Park Memorial Institute, Buffalo, New York. Present address: Skin and t We are indebted to Dr. Rudolph Nagy, Ad- Cancer Hospital of Philadelphia, Philadelphia, Pa.visory Engineer of the Research Department of Supported in part by USPHS Grant C-2S18-C. the Lamp Division of Westinghouse Electric Cor- Presented at the Brook Lodge Invitationalporation for designing this unit, and for much Symposium on the , sponsored by Theinvaluable information and many helpful sugges- Upjohn Company, Kalamazoo, Michigan, Marchtions. The Atlantic Ultraviolet Company built the 27—28, 1958. ultraviolet unit. 373 374 THEJOURNAL OF INVESTIGATIVE DERMATOLOGY

FIG. 1. Ultraviolet unit (a) hood (b) ultraviolet fluorescent lamps in reflectors (c) wiremesh cage (d) phototube in housing. Distance from lamps to animals, 42 cm.

WAVELENGTH IN ANGSTROM UNITS FIG. 2. Relative energy distribution of Westinghouse fluorescent sun lamps.

Group III—Parenteral Each group of animals was exposed to ultra- lila. Methoxsalen—O.4 mg. intraperitoneally,violet simultaneously in a single wire cage, 5 daily 1 hour pre-treatment days each week. The length of exposure was the 11Th. Hematoporphyrin—2.5 mg. intraperi-amount of time needed to deliver 25 "click" units toneally daily 1 hour pre-treatment as measured by the photo tube placed during each Tile. Control—Acacia intraperitoneally dailyexperiment at the level of the wire mesh mouse 1 hour pre-treatlnent cage. (Fig. 1) PHOTOACTIVE AGENTS AND ULTRAVIOLET CARCINOGENESIS 375

The topical preparations were freshly preparedtion quantitatively transferred to a 200 ml. every two weeks, and applied to the top of eachvolumetric flask and the final volume made up to ear with one stroke of a camels hair brush. Me-200 ml. by addition of isotonic saline. The solu- thoxsalen was obtained in pure form from P. B.tion was filtered through a Selas filter (0.03 pore Elder Company, atabrine powder from Winthropsize), transferred to sterile serum bottles, capped Laboratories, hematoporphyrin from Mannand stored at 4°C. The dose (2.5 mg. per 20 gm. Laboratories, and tar from the Stock Crude Coalmouse) was arrived at on basis of unpublished Tar (McKesson & Robbins Co.) carried by thetoxicity experiments carried out by a member of Institute pharmacy. the Institute staff (12). For oral administration, exactly weighed out The mice were weighed and examined at two quantities of methoxsalen and of atabrine wereto three week intervals for degree of searing and dissolved and then mixed with pulverized purinaerythema of ears, nose and tail, and for tumor chow in a standard laboratory food grinder. Forformation. Histologic study of the tumors is being intraperitoneal injection, the psoralen was sus-carried out, and will be reported at a later date. pended in a diluent consisting of 10% ethanol, All groups received ultraviolet irradiation for 5% glycerin, 1% pectin and water q.s. The he-a total of 175 days, for a total dose of 5.99 X 10 matoporphyrin (HP) solution was prepared asergs/cm2 of mid ultraviolet (2800—3600 A) energy. follows: 2 gm. HP were weighed, dissolved in 80 ml. of 1% aqueous NaOH, and 50 ml. sterile iso- RESULTS tonic saline added. The pH was then adjusted to The results of the experiments reported here 7.4 by slow addition of 0.2 normal HCI, the solu-are preliminary only, and are compiled for pre-

I00 END OF RADI.ATI ON 90

80

7C I apical

60

(I, 0 Sc

I— 40 IMP oral

LU I 30 I—

2C C') 4-J 0 4z

32 34 TIME IN WEEKS FIG.3. Incidence of ear tumors in Swiss mice treated with methoxsalen (8MP) administered locally, intraperitoneally and orally. The difference in the number of tumors present at the end of radiation is statistically significant between the controls and the groups given methoxsalen locally and intraperi- toneally, but not significant between the controls and the group given the psoralen orally. 376 THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

END OF RADIATION 00

9( CRUDE COAL TAR 80 topical

70

60

50

40

° 30 CONTROL I ATABR NE

..EMATOPORPHYRIN

ATABRINE .T.1T ...' topical 4 16 18 20 22 24 26 28 3032 34 36 38 40 TIME IN WEEKS Fin. 4. Incidence of ear tumors in Swiss mice treated with hematoporphyrin, atabrine and crude coal tar. Local application of atabrine causes complete protection. Local application of crude coal tar causes highly significant increase in number of tumors. Oral administration of atabrine and local administration of hematoporphyrin results in no significant difference from the controls. sentation to this conference. A complete final Group ha (methoxsalen orally) showed about report will be prepared for publication at a lateras much searing and erythema as the control date. groups (Ic, Ilic) and a slight increase in the At the time radiation was discontinued, a totalnumber of ear tumors, which at the time of dose of 5.99 X 108 ergs/cm2 of ultraviolet with aevaluation was statistically not significant. wavelength of 2800—3600 A had been given, Local application of HP (Group Id) and in- and 20% of the control animals had developedgestion of atabrine (Group lib) resulted in no ear tumors. These values are in close agreementdeviation fron the changes observed in the with those published by Rush et al. (13), Griffincontrols, nor in a significant alteration in ear (3) and O'Neal et al. (11). tumor incidence. Local applications of atabrine Of the animals pre-treated with intraperitoneal (Group Ic) to the ears resulted in complete methoxsalen, 68% had developed tumors, and allprotection against erythema and tumor forma- showed marked erythema, scarring, alopecia andtion (Fig. 4). thickening of face, ears and tail. The group The most spectacular changes were noted in painted locally with 0.1% methoxsalen on thethe animals treated locally with crude coal tar ears showed as severe skin changes, and an(Group Ib). Papillomas began to form about incidence of ear tumors of 52% which rose rapidlythe fourteenth week, and typical carcinomas in the ensuing two weeks. These two groups16 weeks after start of therapy. Even before (Ta and lila) showed the most marked skincessation of radiation, 100% of the animals had changes, and a statistically significant increaseeither papillomas or carcinomas or both (Fig. 4). in ear tumor formation (Fig. 3). A control group treated with locally applied PHOTOACTIVE AGENTS AND ULTRAVIOLET CARCINOGENESIS 377 tar alone was started later, and while it is tooet at. (11). Until the intermediary pathways early to be absolutely certain, it appears thatof methoxsalen are better defined, such an tar application without ultraviolet causes papil-explanation must remain highly speculative, lomas to develop much later, and in lesser num-and subject to further experimentation. The ber. increase in searing and in ear tumor formation The data obtained on the animals treated withfollowing local psoralen application is a reasonable intraperitoneal HP could not be evaluated, sincecorollary to its known, potent photosensitizing the majority of the experimental mice died 14capacity following external use. to 18 weeks after the onset of the experiment, The complete protection afforded by local apparently of chronic intoxication. application of atabrine is not surprising in view of the strong absorption of this compound in the DISCUSSION medium ultraviolet. The lack of protection afforded by oral administration of atabrine The results of these experiments confirm againmay be due to insufficient dosage. It is however the previously quoted findings of Rush et at. (13),interesting in light of the observations of Griffin (3) and O'Neal et at. (11) as to the amountCahn et at. (15) that atabrine, given to humans of mid-ultraviolet energy necessary to producein therapeutic doses, does not affect the normal skin tumors in albino mice. erythema threshold, but is able to suppress There is however a marked difference betweenabnormal reactions to ultraviolet light. our results, and those obtained by O'Neal et at. The spectacular carcinogenic capacity of (11) as far as the demonstration of a protectivelocally applied crude coal tar combined with effect of orally administered methoxsalen isultraviolet irradiation is worth singling out. Crude concerned. We have been unable to observe suchcoal tar has long been suspected of being a an effect. If anything, there appears to be a slightphotosensitizer and used as such in the medical sensitization to ultraviolet at this time. Themanagement of certain skin diseases (e.g., reason for this discrepancy is not immediatelyGoeekermann treatment of ). If the apparent. An attempt was made to reproducedifference in number of tumors and in the rate O'Neal's experiments as closely as possible,of their appearance that is suggested by our except for the type of ultraviolet light source.preliminary control experiments holds up in the The major differences between the mercury arefinal analysis, it might turn out that the combina- lamps used by O'Neal and the fluorescent "sun"tion of crude coal tar and ultraviolet constitutes lamps used by us lie in their emission spectra.one of the most potent carcinogenic agents for Approximately 75% of the energy output of therodents, and may be of interest in the study of mercury ares lies below 2800 A and the mercuryhuman skin eareinogenesis as well. spectrum consists of discrete lines, while our Finally, these experiments seem to show, as fluorescent lamps have a continuous spectrumhas been previously suggested (11), that com- entirely in the mid-ultraviolet with a markedpounds which cause a marked increase in the peak from 3000 to 3250 A (i.e., in the area ofchemical reaction of mouse skin to ultraviolet maximum erythema production) (Fig 2). It islight also increase the careinogenicity of this known that short ultraviolet radiation is abletype of radiation, while those which interfere to destroy a variety of chemical compoundswith the burning effect decrease the rate of (e.g., destruction of sterols by ultraviolet ofultraviolet cancer production. 2537 A), and Lerner (14) has shown that pro- longed irradiation of methoxsalen in vitro with SUMMARY mercury ares changes its absorption spectrum. It is tempting to suggest the possibility that 1. The effect of several known photoaetive methoxsalen given orally may be changed duringcompounds (methoxsalen, HP, crude coal tar, its passage through the gut and liver into anand atabrine) on ultraviolet skin eareinogenesis intermediate compound, still able to photo-in albino mice has been studied. These com- sensitize, but perhaps more easily destroyed bypounds were administered locally, orally and the chemical effects of short ultraviolet. Such aintraperitoneally in appropriate concentrations. hypothesis could explain the marked differenceMid range ultraviolet (wavelengths 2800—3600 between our observations and those of O'NealA) was administered daily by means of a bank of 378 THEJOURNAL OF INVESTIGATIVE DERMATOLOGY

Westinghouse fluorescent "sun" lamps to a total 2. FINDLAY, G.M.: Ultraviolet light and cancer. Lancet, 2: 1070, 1928. dose of 5.99 X 10 ergs/em1, divided over a period 3. GEIFFIN,A.C.,DOLMAN,V.S.,BAHLKE,E.B., of 175 days. BOUvART, D. AND TATU, E. D.: The effect 2. Local aRd intraperitoneal administration of visible light on the carcinogcnicity of ultraviolet light. Cancer Research, 15: 532, of methoxsalen resulted in a significant increase 1955. in the number of ear tumors present at the time of 4. PUT5CHAR, W. AND HOLTZ, F.: Erzeugung von Hautkrebsen hei Ratteu dureh langdauernde discontinuation of the radiation. Contrary to Ultraviolett Bestrahlung. Ztsch. f. Krebs- previous reports, no evidence of any protective forsch., 33:219, 1930. effect of orally administered psoralen could 5.Rono, A. H.: Cancer et soleil. Carcinoma et sarcoma provoqué par l'actioD du soleil in be demonstrated. toto. Bull. Assoc. Franc. étude Cancer, 23: 3. Local administration of atabrine com- 59,1933. 6.HERLITZ, C. W., JUNDELL, I. AND WAHLGEEN, pletely protected the animals against the effects F.: Durch Ultraviolett Bestrahluug er- of intense ultraviolet radiation, and completely zengte maligne Neubildungen bei Weissen suppressed tumor formation. Oral atabrine and Mausen. Acta Poediat., 10: 321, 1931. 7. BAIN, J. A. AND Ruscu, H. P.: Carcinogenesis local application of crude coal tar to the ears with ultraviolet radiation of wave length of the animals caused rapid development of 2800—3400 Angstrom. Cancer Research, 3: papillomas and carcinomas. The result of in- 435, 1943. S. ELris, C., WELLS, A. A. AND HEGRATH, F. F.: jected hematoporphyrin could not be evaluated The Chemical Action of Ultraviolet Rays. because of excessive early mortality, presumably ReinholdPub!. Co., N. Y., 1941. due to the toxic effects of the drug. 9.BLUM, H. F.: Photodynamic Action and Dis- eases Caused By Light.N. Y., Reinhold 4. The findings are discussed, particularly in PubI.Co., 1941. respect to the possible causes for the lack of 10.BLUR, H. F.:Ultraviolet Radiation and Can- cer. In: A. Hollaender (ed) Radiation Biol- protective effect of oral methoxsalen, which ogy, Vol. 2. N. Y. McGraw-Hill, 1955. may be due to the absence of short ultraviolet 11. O'NEAL, M. A. AND GRIFFIN, A. C.: The effect radiation inherent in the emission spectrum of oxypsnralen upon ultraviolet carcinogen- esis in albino mice. Cancer Research, 17: of fluorescent "sun" lamps. 911,1957. 12. HOFFMETSTEE,F. S. AND LAPLATNEY, D.: ACKNOWLEDGMENTS Influenceof Hematopnrphyrin on X-Ray Damageto Ehrlich-Lettre Ascites Tumor. The author is indebted to Mr. J. E. Freeman RPMI Interdepartmental Report, 1956. 13. Ruscu, H. P., KLINE, B. E. AND BAUMANN, and Mrs. R. Simons for splendid technical C. A.: Carcinogenesis by ultraviolet rays assistance. The generosity of Paul B. Elder Co. with reference to wave length energy. Arch. and Winthrop Laboratories for supplying large Path.,31: 135, 1941. 14.LERNER, A. A., DENTON, C. H. AND FITZFAT- quantities of methoxsalen and atabrinc is grate- RICK, T. B.: Clinical and experimental fully acknowledged. studies with 8-methnxypsnralen in vi- tiligo. J. Invest. Dermat., 20: 299, 1953. REFERENCES 15. CAHN, M. H., LEVY, E. J. AND SHAFFER, B.: Polymorphous lighteruption.The effect of 1. UNNA, P. G.: Die Histopathologie der Haut- chloroquine phosphate in modifying reac- krankheiten. A. Hirschwald, Berlin, p. 725, tions to ultraviolet light. J. Invest. Dermat., 1894. 26: 201, 1956.