Multipotential Carcinogenesis with Urethan in the Syrian Golden Hamster*

BELA TOTH, LORENZO TOMATIS, AND PHILIPPE SHUBIK

(Division of Om~ology, The Chicago Medical School, Chicago, Illinois)

SUMMARY 17rethan in the drinking water (0.~2-0.4 per cent) was administered to 5- to 7-week- old Syrian golden hamsters and continued for 4~ weeks with an 8-week interval between the 40th and 48th week. This treatment has been found to give rise to melanotic tumors of the skin, papillomas and carcinomas of the forestomach, adenomatous polyps of the cecum, pulmonary adenomatosis, mammary tumors, hepatolnas, helnangiosar- comas. Certain other tumors occurring in the controls appear to have bccn enhanced. After the 40th week, all the surviving animals in the urethan-treated groups developed one or more tumors.

Urethan has been found to be a multipotential denee of melanotic tumors of the skin and papillo- carcinogen in several strains of mice (17, 18). In mas of the forestomach. these long-term experiments, five separate lesions In the present experiments the effects of ure- were induced or potentiated by urcthan: pulmo- than in the Syrian golden hamster were further in- nary adenomas, malnmary carcinonms, malignant vestigated with a slightly modified experimental mescnchymal tumors in the interscapular fat, approach, with the use of younger animals and eystadenomas of the lacrimal gland, and blood higher concentrations of urethan. cysts in the liver. Urethan has been reported to have various MATERIALS AND METItODS roles in relation to the occurrence of leukemias Syrian golden hamsters from a colony originally and lymphomas in mice. It has been reported to obtained from Abrams Small Stock Breeders, potentiate the leukemogenic action of x-rays, Chicago, Illinois, and bred randomly in our labora- estrogens, and methylcholanthrene in mice (~, 10); tory since 1959, were used. They were housed in it has been found to accelerate the onset of leu- plastic cages with wood shavings in groups of five, kemia in AKR and C58 strains of mice (11) and according to sex, and fed Rockland mouse diet in to give rise to an early onset of lymphoma when pellets and tap water ad libitum. injected into Swiss mice at birth (6, 18). More In the experimental group, 30 females and 31 recently, we have observed that the incidence of males, 31-41 days of age, the females weighing lymphoma was increased and accelerated when 61.5 gm. (average), the males weighing 59.~ gm. large doses of urethan were administered to adult (average), were used and given 0.~ per cent ure- Swiss mice (19). In rats, the urethan treatment than (Fisher Scientific Co.) in the drinking water. resulted in pulmonary adenolnas, hepatomas, and At the beginning of the ~0th week of treatment, a few other tumors (9). However, in guinea pigs, the dose of urethan was increased to 0.4 per cent, rabbits, and chicks, urethan seems to be ineffec- and it was given until the 40th week, at which time tive (3, 16). the treatment was discontinued because of diar- Pietra and Shubik (14) reported that the ad- rhea. The treatment was resumed 8 weeks later ministration of urethan in the drinking water of and maintained for ~ weeks, which again resulted Syrian golden hamsters resulted in a high inci- in an outbreak of diarrhea. Therefore, it was de- * This investigation was supported by grants C-o~898 and cided to stop the treatment. CS-9~1o~ from the National Institutes of Health, National Can- The total dose of urethan administered during cer Institute, U.S. Public IIealth Service. the whole course of treatment to each hamster Received for publication June 12, 1961. could only bc estimated with a certain degree of 1537

approximation to be 10 gm. for each female and All the animals were allowed to die spontane- 7 gin. for each Inale. ously or were killed with ether when in poor con- The average daily water consumption per ani- dition. A complete necropsy was performed on mal was l~ ml. in females and 8 ml. in males, as every animal, with the exception of five females calculated from the water consumption for each and four males in the treated groups, which were cage, recorded throughout the treatment. lost through cannibalism. All organs were ex- ,,'ks a control, 47 females and 54 males were kept amined maeroseopieally, and all pathological le- untreated. sions were submitted to histological diagnosis. Both experimental and control animals were Sections were routinely stained with hematoxylin carefully checked and weighed at weekly intervals, and eosin and with special technics when needed. and skin changes were recorded on graph paper. Those pigmented lesions having a nodular ap- RESULTS pearance and extending for at least g ram. in their The survival rate and the number of different greatest diameter were classified as melanotic tu- types of tumors of both experimental and control mors. Other cutaneous lesions were also carefully groups are recorded in Tables 1 and ~. recorded. Tumors of the forestomach.--Eighteen females

TABLE 1 SURVFVAL RATE OF HAMSTERS TREATED WITIt URFTnAN AND OF UNTREATEI) IIAMSTERS

--~'O. SLRVIVORS AT VCEI~,;.S OI" AGE INI)ICATEI) INITIAL HAMSTEHg ~" i I I AXiX~ALS 10 '20 30 40 50 60 __70 __8.O0 __91) __10~0 __110~ __17_0.__ 1_30______140

Ure tlm n- 17 5 I 1 treated* 31 C 3{i) 30 ~9 [ ~6 ~1 ] 14 Untreated ~7~ I 47 I ,7 ~7 46 46 4~ 37 04 9 5 o 5 t c? 4~ 39 39 37 37 36

* 0.~-0.4 per cent in (Mnking water.

TABLE 2

TI;M()R INCIDI",N(?E IN ]I kMSTI:ItS TPd';ATE1) WITH URETHAN AND IN UNTREATED IIAMSTERS I No. ANI.MAL$ VVITtl NO. ANI- ~-~'O, A-NI- No. 'roxAL MALS/NO. MA LS/NO. [NI TIA L AN LMALS NO. TIT- AI) EN(.IM- MEL&NO'I'- HAMSTERS _NO. ANI- ~vr TIt MOR-BEAR- Tu m or s M alig- Pll I- ATOUS I f' TU- ~- ~till- i tIenlan- I 31A L.~ COMPLNTE ING of tile }Icpa- nemail- moIlary Other POLS'ps mary glosar- NECROPSY ANIMALN XI()II8 OF fore- |ympho- tomas gioma.s adcno- tumors ()P ('JE('UM SKIN tumors eomtls stomach in ~/s ! ma.tosis I Urethan- 3O ~5 ~1 11/11 18t 1w 4# ++ treated* 31C ~/~1 ~11 5** l?t 4++ Untrea ted 47 9 47 8 0 0 1w167 0 0 1 54C : 54 9 0 0 0 ~1111 0 I 0 I * 0r per cent in drinking water. t Eighteen animals with multiple papillomas; two animals also with squamous-ccll carcinomas. :~ Hemangioma of spleen. w Hemangiosarcoma of liver. # One squamous-cell carcinoma of vagina; one mastocytoma; one thecoma; one malignant neurilemmoma. [1 Twenty-two animals with multiple papillomas; three animals also with squamous-cell carcinomas. ** One animal with hemangioma of spleen; two animals with hemangioma of liver; one animal with hemangioendothelioma of spleen; one animal with hemangioma of spleen, liver, and cecum. tt Hemangiosarcoma of liver. +:~ One malignant melanoma; one cholangioma; one clear-cell adenoma of kidney; one adrenal cortical carcinoma. ww One papilloma. ## One adenoma of thyroid; one sarcoma of s.c. tissue. H[[ One animal with hemangioma of liver; one animal with hemangioma of spleen and liver. *** One cholangioma; four adrenal cortical adenomas.

(60 per cent) and ~ males (70 per cent) developed pigment were also recognizable and were classified tumors of the forestomach. as melanocytes. In other instances, the absence Most of these animals had more than one squa- of dendritic processes and the presence of large mous-cell papilloma of the forestomach (Fig. 7). coarse amounts of pigment helped us to identify In addition, in two females killed at the 28th and some of the cells as macrophages, mainly in the 64th week from the beginning of the experiment, peripheral areas of the tumor. The neoplastic respectively, and in three males killed at the 84th, growth appeared well encapsulated and had no 85th, and 85th week, respectively, a squamous- relationship to the overlying epithelium. No junc- cell carcinoma was found (Fig. 8). In spite of the tional changes were seen (Fig. 9). A metastasis fact that the tumors invaded the muscular wall, was found in the lungs. we were unable to find metastases. Hemangiomas and hemangiosarcomas.--In sev- No tumor was seen in the glandular part of the eral animals, both females and males, a marked stomach. dilation of liver sinusoids was noted. Two females Intestinal tumors.--Adenomatous polyps of the showed large blood lakes in the liver and one a intestine were found in five males (16 per cent) hemorrhagic cyst in the salpinx. One male also (Figs. 1 and 2). In total, eighteen tumors were had blood lakes in the liver. Only those lesions in observed, three or four per animal. Only one which a real proliferation of new vessels and of female had two adenomatous polyps of the mu- endothelial tissue was observed were classified as cosa of the cecum. All these tumors were pedun- hemangiomas. In five males vessel tumors were culated. Their size ranged between 2 X 2 X 1 and seen in different organs, in two animals in the liver, 5 X 3 X 2 mm. They were supported by a central in two other animals in the spleen, and in one connective stalk in which an incomplete muscu- animal in the liver, in the spleen, and in the intes- laris mucosae could always be distinguished.Some- tine (Fig. 5). Only one hemangioma of the spleen times the glandular formations were dilated up to was observed in female hamsters. Grossly, the a cystic appearance, and they were often observed tumors appeared as slightly elevated, reddish- in the thickness of the stalks simulating an early black areas on the liver surface, ranging from 4 to invasion of the stroma. Many mucus-producing 9 ram. in diameter. Microscopically, most had the cells were noted, several of them exhibiting a high appearance of hemangioma cavernosum, and one degree of activity. No mitoses and no nuclear had the histological pattern of benign hemangio- abnormalities were seen. The lumen of the dilated endothelioma. Thrombosis was a common finding. glands generally contained mucous material. Two The surrounding parenchyma was compressed by lesions that appeared grossly as sessile masses with the tumorous growth, and some areas of necrosis broad bases were not included in the total number were seen. of tumors: they showed an extensive infiltration Furthermore, a hemangiosarcoma of the liver with inflammatory cells that masked the histo- was found in one female and in one male, killed at logical appearance of the original structure. the 46th and the 73d week from the beginning of Melanotic tumors.--Increased pigmentation of the experiment, respectively (Fig. 6). A metasta- the skin, with the appearance of minute black sis in the spleen was seen in the male. spots spread over the back and flanks, was seen It is noteworthy that all the above-mentioned in many animals. Eleven melanotic tumors were findings of dilated vessels or blood lakes were seen found in eleven females (38 per cent) and 21 in animals which died or were killed between the melanotic tumors in twelve males (89 per cent). 40th and the 60th week, whereas the hemangiomas The histological characteristics were those de- were seen between the 60th and the 80th week, scribed by Della Porta et al. (5) resembling the with the exception of one female that died at the blue nevi of man. 50th week. One malignant melanoma was recorded in a Pulmonary adenomatosis.--Areas of hyperplasia male hamster killed at the 84th week. The tumor of the bronchiolar epithelium with some adenoma- appeared as a brownish nodular mass, 25 X 15 X tous structures were observed in the lungs of six 5 ram. in size, growing on the skin of the low back, females and nine males. In addition, three males between the two pigmented costovertebral spots, and five females exhibited pulmonary lesions with but having no relationship to them. Microscopi- the following characteristics. Groups of alveoli cally, the tumor was composed of polyhedral, constituting a focal area without any demarcation round, or spindle-shaped cells, frequently ar- from the surrounding parenehyma were lined by ranged in alveolar patterns. The cytoplasm of one or two layers of eylindroid or columnar cells, most of the cells was loaded with melanotic pig- several of which secreted mucus. Only in very ment. Several cells showed the presence of den- limited areas did the alveolar structure appear to dritie processes in which minute granules of the be destroyed; otherwise, the neoplastic cells formed

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1961 American Association for Cancer Research. 1540 Cancer Research Vol. ~1, December 1961 a regular lining coat. The supporting stroma was the thyroid, and one sarcoma of the subcutaneous formed by the alveolar septa, often thickened tissue were seen. Four cortical adenomas, two through an inflammatory reaction. The presence malignant lymphomas of the histioeytie type, one of abundant mucous material in the alveoli lined papilloma of the forestomaeh, one eholangioma, by the neoplastic cells simulated the presence of and three hemangiomas were observed in the glandular structures. Mitoses were extremely rare. males. In a few animals focal areas of bronehiolar These lesions were always in close proximity to a hyperplasia were found, and areas of puhnonary bronehiole, but continuity between the bronehi- adenomatosis were noted in one female. olar epithelium and the cells lining the alveoli could not be established histologically. We have DISCUSSION called these lesions pulmonary adenomatosis, from The finding of a high incidence of melanotic their similarity to the lesions described in human tumors of the skin and of squamous-eell papillo- pathology. mas of the forestomaeh in urethan-treated ham- Other tumors and lesions.--Besides those already stets was expected from previous results obtained mentioned, nine tumors were observed in the fe- in this laboratory by Pietra and Shubik (14). In males. One malignant lymphoma of the lympho- addition, we observed the occurrence of squamous- eyrie type (lymphosareoma) was seen in one fe- cell carcinomas of the forestomach in five animals, male that died at the ~8th week from the begin- which was not previously reported. ning of the experiment. One malignant neurilem- The large number of adenomatous polyps of the morea (Schwannoma) was found in a female that cecum is noteworthy, since we did not find any in was killed at the 29th week, The tumor, :30 X our present control groups and since these tumors ~0 X i0 mm. in size, was solid, well encapsulated, have not been reported in other species following whitish-pink in color. Fragments of the tumor urethan administration. However, in our labora- were successfully transplanted to four male young tory, small inflammatory polyps of the cecum were hamsters, and presently the transplant is in its observed in another control group (4), and Fortner seventh generation. Microscopically the tumor (7) reported that he found a number of intestinal was composed of intermingled bands of elongated polyps and adenoeareinomas in untreated golden cells. Palisading of the nuclei was evident in inany hamsters. fields, and characteristic Veroeay bodies were also The hemorrhagic lesions in the liver of urethan- found (Fig. 10). treated mice have been found by several investiga- In addition, one squamous-eell carcinoma of tors and differently interpreted (8, 11, 1~, 15, 17, the vagina, one theeoma, one malignant masto- 18). Recently, Kawamoto et al. (11), using AKR eytoma, two mammary adenocareinomas, and one and C58 strains of mice, reported that urethan mammary fibroadenoma were recorded in the fe- induced the formation of hemangiomas of liver, males. intestine, and pancreas. In our present study we Four malignant lymphomas, all of them of the observed dilated vessels and blood lakes in the histioeytie type (retieulum-eell sarcoma) (Fig. liver of several urethan-treated hamsters. Further- 11), two hepatomas, one cortical carcinoma of more, we have found hcmangiomas and hemangio- the adrenal gland, one clear-cell adenoma of the sarcomas in different organs of the urethan-treated kidney (Fig. 1~), and one eholangioma were found animals. However, in our control group wc have in the males. seen two animals with hemangiomas in the liver Lesions in the control group.--Microcysts, bile and spleen. duct formation, and proliferation of oval cells were With regard to pulmonary lesions, eight ani- observed frequently in the liver. In the females mals of the urethan-treated groups developed one papilloma of the forestomaeh, one adenorna of pulmonary adenomatosis, as compared with one

Fro. 1.--Adenomatous polyp of tile cecum, in a male hamster killed 6! weeks after beginning of the treatment. H & E., X70. FI(~. P2.--Same tumor as in Figure 1. Dilated glandular structures, lined by regular cylindric epithelium. II. & E., X 1~20. FI(~. 8.--Pulmonary adeuomatosis in a male hamster killed 84 weeks after beginning of the treatment. H. & E., Xl15. FIG. 4.--Same tumor as in Figure 3. Glandular-like structures, lined by euboidal or cylindric cells, many of them coll- taining vacuoles. Desquamated cells and inflaminatory cells are present in the lumen. H. & E., X'280.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1961 American Association for Cancer Research. ~i~!~i~~ii~iil i~i~i~ ~iiiiiii~ ~ii~iiiii~ ~ii~i!i~!iii~i~iii~i~ii~iii~ i~ii~i~ii~i~ ~iii~il~,i~iii~ii~i~ili~iill iiiii~ii~i

~ !!!!!!!!!!!~iiiiiiiii iiiiiiiii 84iiiiiiiiiii 84iiiiiiii iiii'iiii 84iiiiiiiii 84iii!i!i!iii iiiii!~i!i 84i!i!i!i!!!!! !!!!!!!!!!! !!!!!!!!!!~ !!!!!!!!!~ !!!!!!!!! ~!~!!~!~ ~!~!~ !~ii~ ~ ~i~i~~

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1961 American Association for Cancer Research. FIG. 5.--Ilemangioma of the intestine, in a male hamster dying 78 weeks after beginning of the treatment. H. & E., Xl15. FIG. 6.--IIemangiosarcoma of the liver in a male hamster killed 73 weeks after beginning of the treatment. H. & E., X~00. FIG. 7.--Papilloma and ear(!inonm of the forestomach in a male hamster killed 85 weeks after beginning of the treatment. II. & E., XS0. F~G. 8.--Enlargement of Figure 7. Carcinomatous portion of the tumor: undifferentiated carcinoma. H. & E., )< 180.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1961 American Association for Cancer Research. FIG. 9.--Malignant melanonm, poorly pigmented, in a male hamster killed 84 weeks after beginning of the treatment. H. & E., X300. FIG. 10.--Malignant neurilemmoma (Schwannoma) with so-called "Veroeay bodies" in a female hamster killed ~29 weeks after beginning of the treatment. H. & E., Xl15. Fro. l l.--M~alignant lymptloma, histiocytie type (retieu- lure-cell sarcoma), lymph node in a male hamster killed 80 weeks after beginning of the treatment. H. & E., X450. FIG. l~.--Clear-cell adenoma of the kidney in a male hamster killed 85 weeks after begitming of the treatment. H. & E., Xll0.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1961 American Association for Cancer Research. TOTH et al.--Carcinogenesis with Urethan 1541 animal in the untreated groups. These findings ~. BERENBLUM, I., and TRAININ, N. Possible Two Stage suggest that the growth of these tumors is poten- Mechanism in Experimental Leukemogenesis. Science, 132: 40-41, 1960. tiated by urethan. 3. COWEN, P. N. Strain Differences in Mice to the Carcino- The finding of mammary tumors and hepato- genic Action of Urethane and Its Non-carcinogenicity in mas appears significant, because of their extreme Chicks and Guinea Pigs. Brit. J. Cancer, 4: ~45-53, 1950. rarity in the untreated animals. 4. DELLA PORTA, G. Induction of Intestinal, Mammary, and One of the questions to be asked is why the Ovarian Tumors in Hamsters with Oral Administration of ~0-Methylcholanthrene. Cancer Research, 21: 575-79, results of the present study with urethan differ 1961. from the previous report from this laboratory in 5. DELLA PORTA, G.; RAPPAPORT, H.; SAFFIOTTI, U., and which fewer tumors of different types were report- SHUBIK, P. Induction of Melanotic Lesions during Skirl ed in hamsters (14). The first factor that may be Carcinogenesis in Hamsters. A.M.A. Arch. Path., 61:305- implicated is the age of the animals, It has been 13, 1956. 6. FlOnE-DO~ATI, L.; CHIEco-BIANCHI, L.; DE BENEDICTIS, shown recently that newborn mice are particularly G.; and MAIORANO, G. Leukaemogenesis by Urethan irL sensitive to small doses of different chemical Newborn Swiss Mice. Nature, 90:~78, 1961. carcinogens (13). Not only are tumors induced in 7. FORTNER, J. G. Spontaneous Tumors, including Gastro- newborn mice with very small doses of carcinogen, intestinal Neoplasms and Malignant Melanomas in the as compared with those required by the adult, but Syrian Hamster. Cancer, 10:1153-56, 1957. 8. HESTON, W. E.; VLAtlAKIS, G.; and DERINGER, M. K. the type of tumor occurring is different. In the High Incidence of Spontaneous Hepatomas and the In- present study 5- to 7-week-old hamsters have been crease of This Incidence with Urethan in C3H, C~Hf, and found to develop many different tumors following C3He Male Mice. J. Nat. Cancer Inst., 24:4~5-36, 1960. administration of urethan, as compared with 8- 9. JAFFE, W. G. Carcinogenic Action of Ethyl Urethane oi~ to 10-week-old animals. Unfortunately, it is im- Rats. Cancer Research, 7:107-11, 1947, with an appendix by JAFFE, R. Histological Findings in Lungs and Livers of possible to attribute this difference entirely to the Rats Treated with Ethyl Urethane. Ibid., pp. 111-1~. age of the animals, since the dosage of carcinogen 10. KAWAMOTO,S. ; IDA, N. ; KIRSCHBAUM, A. ; and TAYLOR, G. differed from the previous study. In the original Urethan and Leukemogenesis in Mice. Cancer Research, study, 0.~ per cent urethan was administered to 18 : 7~5-~9, 1958. hamsters continuously; in the present study, ure- 11. KAWAMOTO, S.; KIRSCHBAU),I, A.; IBANEZ, M. L.; TREN- TIN, J. J.; and TAYLOR, H. G. Influence of Urethan Oll the than was given at a dosage of 0.~ per cent in the Development of Spontaneous Leukemia and on Induction drinking water for the first ~0 weeks, and then at of Hemangioma in AKR and C 58 Strains of Mice. Cancer a level of 0.4~ per cent for ~ weeks with an inter- Research, 21:71-75, 1961. ruption between the 40th and 48th weeks. It is 1~. KIRSCHBAUM, A.; BELL, E. T.; and GORDON, J. Spontane- thus impossible to say which of these variants is ous and Induced Glomerulonephritis in an Inbred Strain of Mice. J. Lab. & Clin. Med., 34:~209-~0, 1949. the more important. It is, however, of some inter- 18. PIETRA, G.; RAPPAPORT, H. ; and SItUBIK, P. The Effects of est to note once again that the full range of ac- Carcinogenic Chemicals in Newborn Mice. Cancer, 14: tivity of a carcinogen may be determined only 803-17, 1961. after many complex studies. This has been par- 14. PIE~RA, G., and SHUBIK, P. Induction of Melanotic Tu- tieularly notable in the instance of urethan, a mors in the Syrian Golden Hamster after Administration of Ethyl Carbamate. J. Nat. Cancer Inst., 25:6~7-30, 1960. compound originally thought only to give rise to 15. RoF~, F. J. C., and SALAMAN, M. H. A Quantitative Study lung adenomas and now kno~ to have perhaps of the Power and Resistance of the Tumor-Initiating Effect the most widespread activity of any chemical of Ethyl Carbamate (Urethane) on Mouse Skin. Brit. J. carcinogen. Cancer, 8: 666-76, 1954. 16. SHUBIK, P., and HARTWELL, J. L. Survey of Compounds ACKNOWLEDGMENTS Which Have Been Tested for Carcinogenic Activity. Pub. The authors wish to acknowledge their gratitude to Dr. Health Service Publ. No. 149, Suppl. 1, pp. 39-47. Wash- Henry Rappaport for his help in diagnosing the tumors; to ington, D.C.: U.S. Gov"t. Print. O/tlce, 1957. Mrs. Shelagh Maxwell and Mrs. Binlte Zidonis for their techni- 17. TANNENBAVM, A. Studies on Urethan Carcinogenesis. cal assistance; and to Mr. Manuel Sueiro for the histological Aeta Unio Internat. contra eanerum, 17:7~-85, 1961. preparations. 18. TANNENBAV~, A., and SILVERSTONE, H. Urethan (Ethyl Carbamate) as a Multipoteutial Carcinogen. Cancer Re- REFERENCES search, 18:1~5-31, 1958. 1. BERENBLUM, I., and HARAN, N. The Initiating Action of 19. TOTH, B.; DELLA PORSA, G. ; and SHVBIK, P. The Occur- Ethyl Carbamate (Urethan) on Mouse Skin. Brit. J. Can- rence of Malignant Lymphomas in the Urethan-treated eer, 9:453-56, 1955. Swiss Mice. Brit. J. Cancer, 15:3~-~o6, 1961.

Bela Toth, Lorenzo Tomatis and Philippe Shubik

Cancer Res 1961;21:1537-1541.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/21/11/1537

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/21/11/1537. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.