the effect of 13-cis-retinoic acid on hamster buccal pouch carcinogenesis

oral pathology Editor: CHARLES E. TOMICH, D.D.S., M.S.D. American Academy of Orul Pathology Indiana University School qf Dentistry 1 I21 West Michigan Street Indianupolis, fndiana 46202

The effect of 13.cis-retinoic acid on hamster buccal pouch carcinogenesis William Gilmore, D.M.D., M.S.,* and John L. Giunta, D.M.D., MS.,** Boston, Mass.

TUFTS UNIVERSITY SCHOOL OF DENTAL MEDICINE

In order to determine whether 13-cis-retinoic acid, an analog of vitamin A, has antitumor activity in an oral cancer model system, the following study was undertaken. Fifty-three adult hamsters were divided into four groups. Group 1 was treated with a 0.5 percent solution of 9,10-dimethyl-1,2-benzanthracene (DMBA) in heavy mineral oil, which was painted on the right buccal pouch three times per week for 12 weeks. Group 2 received DMBA plus 13-cis-retinoic acid (RA) incorporated into gelatinized beadlets and mixed with a powdered commercial diet (dosage, 300 mg per kilogram of diet). Group 3 received only RA; Group 4 received a placebo. The animals were killed at 6, 12, and 18 weeks and tissues were studied clinically and histologically in a routine manner. Results show that all groups receiving DMBA developed epidermoid carcinomas. However, there were several other changes. In the RA-treated animals, particularly those treated with DMBA, there was an ingrowth of surface epithelium with development of ductal structures in the buccal pouch. There were changes in surface epithelium, and there were dense aggregates of lymphoid tissue with development of exophytic nodules suggestive of lymphoma. Animals fed RA showed a relative weight loss. The findings suggest that there was a hypervitaminosis A state yielding prominent epithelial metaplastic changes but not affecting the progression or production of carcinoma.

C hemoprevention is a new pharmacologic approach in the treatment of cancer in which agents are used at the preneoplastic stage in the hope of preventing neoplasia. Toward that goal, researchers have been exper- imenting with a family of compounds called retinoids,’ which include the natural forms and the synthetic analogs of vitamin A. 13-Cis-retinoic acid (RA) is one of the promising analogs which have shown antitumor activity in experimental tumor systems involving forestomach and esophageal carcinoma,* tracheobronchial carcinoma,” and transitional and squamous cell bladder cancer. J

Presented at the thirty-four annual meeting of the American Academy of Oral Pathology, San Antonio, Texas, May 12 to 16, 1980. *Assistant Professor of Oral Surgery. **Associate Professor of Oral Pathology.

256

The hamster buccal pouch tumor system for oral carcinogenesis is a well-recognized system for studying various chemicals and their effects on carcinogenesis.j* 6 The development and progression of histologic changes ranges from dysplasia (4 to 8 weeks) to squamous cell carcinoma (8 to 12 weeks). With this model, it has been demonstrated that oral chemical carcinogenesis can be altered by radiation, chemicals, and changes in the immune systems.‘-‘” It was theorized that feeding animals 13-cis-retinoic acid would cause a delay in latent period or, better yet, prevent tumor development. After com- pletion of this study, a similar study by Shklar and associates” reported that 13-cis-retinoic acid delayed carcinogenesis in the hamster buccal pouch system.

The effect of vitamin A on epithelium and on epithelial malignancies has been studied extensively and there are both claims and counterclaims concerning the ef-

0030-4220/81/030256+10$01.00/0 0 1981 The C. V Mosby Co.

Volume 51 Number 3

EfSect of 13-cis-retinoic acid on carcinogenesis 257

Fig. 1. At 6 weeks, a DMBA/RA-treated hamster shows a nodular area with a dense lymphocytic and well- vascularized infiltrate covered with an atrophic parakeratinized squamous epithelium which has extensions into the connective tissue. (Hematoxylin and eosin stain. Magnification, ~280.)

Table I. Major histologic observation by group and week of death

Week

Group 6 12 18

1 Mucous metaplasia Mucous metaplasia Mucous metaplasia (DMBA/RA) Epithelial atrophy and parakeratiniza- Epithelial atrophy and parakeratiniza- Well-developed carcinomas

(18) tion tion Lymphoid aggregates and nodules Hyperkeratosis Papillary carcinomas (6)

II (DMBA)

(18)

111 @A)

Lymphoid aggregates (6)

Hyperkeratosis

(6) Epithelial atrophy and parakeratiniza-

tion

Lymphoid nodules

Dysplasia Carcinomas in situ Papillary carcinomas

Normal Epithelial atrophy Lymphoid aggregates

Normal

(6)

(6)

Well-developed carcinomas

(6) Mucous metaplasia Epithelial atrophy and parakeratiniza-

(3)

(2)

tion Lymphoid aggregates

Normal (4)

(3)

(10)

IV Placebo

(7)

Lymphoid aggregates

(3) Normal (keratotic squamous epithe-

hum) (2)

Numbers in parentheses refer to number of animals in each group. DMBA = Dimethylbenzanthracene. RA = 13-Cis-retinoic acid.

258 Gilmore and Giunta Oral Surg. March, 1981

Fig. 2. At 6 weeks, a DMBA/RA-treated animal shows a transitional zone from a normal keratinized squamous epithelium on the left to a parakeratinized epithelium on the right as well as extensions of dysplastic-appearing epithelium into the connective tissue. One of the extensions has formed a ductlike structure with mucoid contents. (Hematoxylin and eosin stain. Magnification, x280.)

fectiveness of vitamin A. Examination of the literature indicates that the inconsistencies are related to the use of various forms of vitamin A by different investigators and to the lack of uniformity among various studies.

Vitamin A deficiency results in keratinization of epithelium.‘” In contrast, vitamin A in excess causes a metaplasia of epithelium from a keratinizing type to a nonkeratinizing variety with formation of mucus- producing cells.‘“, Ii

In experiments with vitamin A deficiency the epithelial changes resembled neoplastic characteristics, and this observation led to many studies investigating that relationship. Vitamin A-deficient animals treated with the carcinogen dimethylbenzanthracene developed sig- nificantly more tumors than did those that received vitamin A. ”

Other studies demonstrating the efficacy of vitamin A as an antitumor agent ranged from retardation of carcinogenesis of intrascapular tumors in miceIs to re- duction of dysplasia in the esophagus, forestomach, and small intestine.’ However, there was no change noted in perineal skin tumors in hamster? or in mouse prostate gland treated in vitro.“’

In contrast, some studies demonstrated an increase in tumor yield when vitamin A was used with the chemical carcinogen. These studies used a relatively high concentration of vitamin A (10 percent) which was applied topically to the carcinogen-treated site.“-‘” These inconsistencies led to the search for an analog of vitamin A which would have antitumor activity and less toxicity.

13-Cis-retinoic acid was found to be as effective in inhibiting keratinization as the natural vitamin A and was less toxic than other analogs. 2.’ Respiratory cancers in hamsters were reduced in incidence or prevented by long-term treatment with 13-cis-RA, depending on the dosage.2J Likewise, bladder carcinogenesis has been inhibited.24 Recently, Shklar and his co-authors’4. 26 reported a delay in the development of squamous carcinomas in experimental lingua1 carcinogenesis in hamsters receiving systemic 13-cis-RA and in buccal pouch carcinogenesis.

METHODS AND MATERIALS

Fifty-three 3-month-old Syrian golden hamsters (Cricetus auratus) of the LVG outbred strain, weighing

Volume 51 Number 3

Effect of 13-cis-retinoic acid on carcinogenesis 259

Fig. 3. At 6 weeks, a DMBA/RA-treated animal shows a large nodule of densely packed monotonous-appearing lymphocytes covered with an atrophic parakeratinized epithelium adjacent to the normal epithelium overlying striated muscle. (Hematoxylin and eosin stain. Magnification, x 120.)

90 to 120 grams each, were divided into four groups designated I, II, III, and IV in this project. Each group was then subdivided according to sex to prevent breed- ing during the study.

Group I was composed of eighteen hamsters that received topical application of 9,10-dimethyl-1,2- benzanthracene (DMBA) to the right buccal pouch three times per week. In addition, they received 13- cis-retinoic acid (RA) with an otherwise normal diet.

The eighteen animals of Group II received topical applications of DMBA to the right buccal pouch three times per week. A placebo consisting of the gelatinized beadlet vehicle was added to their regular diet.

Group III consisted of ten animals receiving an RA- supplemented diet.

Group IV consisted of seven animals and received placebo alone added to their diet.

In this study, the dose of 13-cis-retinoic acid was 300 mg. per kilogram of diet. This was administered in the form of gelatinized beadlets (13-cis-retinoic acid supplied by Hoffman-LaRoche, Nutley, N. J.) and as- sayed at 12 percent of vitamin A (by weight). This was mixed with a standard powdered hamster diet.’ The

administration of the 13-cis-retinoic acid was begun 4 days prior to the first application of the carcinogen since the most efficacious results have occurred when the animal was given the vitamin prior to the carcinogenic insult4 An adult hamster consumes approximately 10 Gm. of food per day. The dose of vitamin A was approximately 3 .O mg. per day.

Both food and water were available to the animals ad libitum. The feed was placed in heavy glass containers with locking metal tops and perforated metal inserts to minimize wasting and soiling by the hamsters. Fresh food mixture was provided three times per week.

The carcinogen, DMBA, was prepared as a 0.5 percent solution dissolved over mild heat in heavy mineral oil. This was painted liberally into the right buccal pouch three times per week for 12 weeks with a No. 4 sable brush. Approximately 0.25 cc. (1.25 mg.) of DMBA was applied per application.27

The animals were kept in clear plastic cages with metal wire grid locking tops and cellulose air filter covers. The maximum number of animals per cage was six. Bedding consisted of at least % inch of finely chopped wood chips, changed weekly. The animal

Gilmore and Giunta Oral S urg March, 1 981

Fig. 4. At 12 weeks, a DMBA/RA-treated animal shows a nodule of densely packed lymphocytes with a prominent ductal structure with desquamated cells. The epithelium is atrophic and parakeratotic and has multiple extensions into the connective tissue. (Hematoxylin and eosin stain. Magnification, x280.)

room climate was controlled for temperature and humidity.

All animals were weighed at the beginning of the experiment and at death. Animals were killed at 6, 12, and 18 weeks by placement in a container with diethyl ether.

At death the animals were weighed, the right pouches were everted and examined, and clincal results were documented. The pouches were then excised and placed in 10 percent formalin for at least 24 hours. Occasional left pouches were taken. The specimens were prepared and section in a routine manner and stained with hematoxylin and eosin for histologic evaluation. Mucicarmine and periodic acid-Schiff stains were used on selected specimens.

RESULTS

A summary of the major histologic findings is listed by group in Table I. With regard to carcinogenesis, all groups receiving the carcinogen developed epidermoid carcinomas according to the standards expected for this model system. The latency and morphologic pattern of

tumors was not different in the experimental group. However, there were significant observations in the pouches of animals receiving retinoic acid. There was a metaplasia of epithelium into apparent mucus-producing epithelium (that is, mucous metaplasia). The mucous metaplasia recapitulated the embryology of epithelium glandular formation. There was a change in the surface of the epithelium from a keratinizing comeal layer to a parakeratinizing layer. The epithelium was atrophic, and there was an ingrowth of surface epithe- hum into the underlying pouch connective tissue. Duc- tal spaces developed, and many were lined with mucus-producing cells. Stains with periodic acid- Schiff (PAS) reagent and mucicarmine revealed material stained positively in the cytoplasm and lumen.

Also noted, only in animals receiving RA and more particularly in those receiving DMBA and RA, were prominent lymphocytic aggregates either beneath the epithelium or deeper in the tissue. In some instances these aggregates were nodular and tumorlike. A desig- nation of lymphocytic infiltrate suggestive of lymphoma could be applied.

Volume 51 Number 3


Fig. 5. At 12 weeks, a DMBA/RA-treated animal shows the periphery of a nodule demonstrating extension of epithelium into connective tissue with ductal or primitive glandlike formation. Two ducts contain material which stains positively for mucin (center, bottom third). (Mucicarmine stain. Magnification, x280.)

At 6 weeks, the pouches receiving DMBA were pink group there was prominent mucous metaplasia with epi- with a slight whitening. The RA and placebo groups thelial atrophy and parakeratosis in most pouches as were normal, pink, and glistening. Histologically, in well as areas of definite dysplasia. There were many the DMBA/RA group there were prominent mucous carcinomas-in-situ and well-differentiated papillary metaplasia and lymphocytic infiltrates, some in a nodu- epidermoid carcinomas. One pouch had a nodular lar pattern. Where there was metaplasia, the overlying spindle-cell tumor. There were dense lymphocytic epithelium was atrophic and parakeratotic. (The normal aggregates, one of which was a large nodule suggestive pouch epithelium is orthokeratotic.) Some pouches had of lymphoma (Figs. 4 to 6). The DMBA-only group dysplastic changes, whereas most were hyperorthoker- showed hyperorthokeratosis, dysplasia, carcinoma in atotic. In areas with mucous metaplasia, it was difficult situ, and well-differentiated epidermoid carcinomas. at times to distinguish between minimal dysplasia and The RA group was normal except for one pouch which atrophic change with hyperchromasia and pleomor- had lymphocytic infiltrates, epithelial atrophy, and phism (Figs. 1 to 3). The DMBA group had hy- subepithelial splitting (Fig. 7). The placebo group was perorthokeratosis, with only two of six pouches having normal. Of special interest was a control (left) pouch in questionable dysplasia. The RA group had lymphocytic the DMBA/RA group which showed mucous metapla- infiltrates with nodule formation. There was epithelial sia, dense lymphocytic reaction, and epithelial atrophy atrophy with surface parakeratosis in one pouch. The with a parakeratotic surface layer. (The left pouch was placebo group was normal. only randomly removed.)

At 12 weeks the pouches receiving DMBA were whitened and had multiple 1 to 3 mm. nodules and some ulcerations. Animals of the RA and placebo group were normal. Histologically, in the DMBA/RA

At 18 weeks, the pouches receiving DMBA were whitened and had multiple 2 to 3 mm. exophytic massses and some 5 to 10 mm. exophytic, ulcerative tumors. The RA and placebo groups were normal.

262 Gilmore and Giunta Oral Surg.

March, 1981

Fig. 6. At 12 weeks, a DMBAiRA-treated animal shows a typical low-grade, exophytic carcinoma. Similar lesions were noted in all DMBA groups at 12 and 18 weeks. (Hematoxylin and eosin stain. Magnification, x 100.)

Histologically, in the DMBA/RA animals there was prominent mucous metaplasia, even associated with carcinoma. There were dysplasia, multiple well- differentiated papillary and nonpapillary epidermoid carcinomas, and dense lymphoid aggregates (Fig. 8). The DMBA-only group had epithelial dysplasia and well-differentiated papillary carcinomas. There were scattered infiltrates of lymphocytes, and none were in a nodular pattern. The RA group had dense lymphocytic aggregates and epithelial atrophy with a parakeratotic surface in half of the pouches. One pouch had mucous metaplasia. The placebo ‘group was normal.

The weights of the animals varied. Those animals receiving retinoic acid weighed 43 percent less than those that did not receive the acid.

DISCUSSION

The systemic administration of 13-cis-retinoic acid in the hamster buccal pouch tumor model resulted in no change in the development of the tumors but did yield epithelial changes and lymphocytic changes.

The mucous metaplasia noted in this experiment is similar to that observed by Mock and Mainli in hamster

buccal pouch explants which were treated with an excess of vitamin A. Similar results have been seen in experiments with skin explants receiving retinoic acid and carcinogen. l6 The change from a keratinizing to a parakeratinizing squamous epithelium which develops into ductal structures reinforces the role of vitamin A in the differentiation of epithelial cells into specialized tissues. Such a change suggests that there may have been a toxic dose of vitamin A in this study. Despite the mucous metaplasia, no adenomas or mucoepidermoid carcinomas were observed. It is interesting to note that mucous metaplasia was not reported in the studies of Shklar and associates. 14* 26

In a similar study on hamster buccal pouch carcinogenesis,LJ Shklar and his co-workers reported a delay of from 2 to 4 weeks in the development of epidermoid carcinomas. As in the present study, no differences in the histologic character of the tumors were noted between the experimental and control groups. Comparison of the two studies reveals both similarities and differences. Both used a similar dosage of retinoic acid per week (20 mg.), given via the gastro- intestinal system, however, the medium for delivery

Volume 51 Number 3

Effect of I3-cis-retinoic acid on carcinogenesis 263

Fig. 7. At 12 weeks, an RA-treated hamster shows focus of dense lymphocytic infiltrate resembling lymph node with follicular pattern at right. The epithelium changes from normal at upper right to atrophic and parakeratotic overlying the area of lymphoid aggregate. (Hematoxylin and eosin stain. Magnification x280.)

and rate of delivery were different. Shklar and col- leagues used the acid in peanut oil and administered two times per week. In our study the acid was given each day using beadlets in the regular diet. Perhaps the constant exposure to the acid helped contribute to an apparent overdose, since there would be less chance for clearance of the acid if the acid were given daily. How- ever, other studies have used a similar pattern of delivery of the retinoid in rats with a resultant inhibition of transitional cell and squamous cell cancer of the bladder. 24

In both studies, the tumorigenesis proceeded in an expected pattern histologically. Neither study inhibited tumor formation; however, that of Shklar and associ- ates14 demonstrated a delay. Some explanation might be inferred from the prominent lymphocytic response in this study. Although dense aggregates of lymphocytes were seen at weeks 10 and 12 in the study by Shklar and his group, they were not of the extent and pattern seen in this study, where it has been seen in all groups that received retinoic acid. No mention is made in the Shklar study of histologic characteristics seen in the group receiving only retinoic acid. Such aggre-

gates suggest a nonspecific hyperimmune response. The dense lymphocytic and sometimes lymphoma-

tous infiltrates were noted only in animals receiving retinoic acid. This suggests that perhaps the retinoic acid stimulated the immune system which has been shown in other studies. Lymph nodes draining the site of injection of vitamin A developed a hyper- cellularity. 2H Systemic administration of vitamin A acid in mice resulted in a two- to fourfold faster rejection of skin homograft. 2g Vitamin A and retinoic acid have been shown to be an immune adjuvant suggesting T-cell activity. 3o The production of lymphocytes as a reaction to neoplasia is common, but in this model system it is usually not so pronounced. Such a response does not necessarily protect against carcinogenesis since it can be nonspecific or the tumor may be weakly antigenic (most likely with DMBA). If the lymphocytes do not recognize tumor antigens, then tumors can de- velop in a normal pattern as noted in this study.

The loss of weight or the lack of appropriate weight gain in those animals receiving retinoic acid suggests a hypervitaminosis. Signs of vitamin A toxicity include loss of appetite, weight loss, and changes of the skin,

264 Gilmore and Giunta Oral Sure.

March, 198y1

Fig. 8. At 18 weeks, a DMBA/RA-treated animal shows epithelial proliferation with multiple extensions into the connective tissue and the formation of multiple ductal structures with mucus-producing cells (mucous metaplasia). (Hematoxylin and eosin stain. Magnification x 180.)

among other changes. It appears that the calculated dose of 300 mg. per kilogram of diet, given without interruption, was within the toxic range. Future experiments will test varying doses of vitamin A.

REFERENCES

1. Sporn, M. B., Dunlop, N. M., Newton, D. L., and Smith, J: M.: Prevention of Chemical Carcinogenesis by Vitamin A and Its Svnthetic Analonues. Fed. Proc. 35: 1332-1338, 1976.

2. Chu,.E., and MaImiran, R.: An Inhibitory Effect of Vitamin A on the Induction of Tumors of Forestomach and Cervix in the Syrian Hamster by Carcinogenic Polycyclic Hydrocarbons, Cancer Res. 25: 884-895, 1965.

3. Saffiotti, U., Montesano, R., Sellakumar, A. R., and Borg, S. A.: Experimental Cancer of the Lung: Inhibition by Vitamin A of the Induction of Tracheobronchial Squamous Metaplasia and Squamous Cell Tumors, Cancer 20: 857-864, 1967.

4. Spom, M. B.. Squire, R. A., Brown, C. C., Smith, J. C., Wenk, M. L., and Springer, S.: 13-Cis-Retinoic Acid: Inhibi- tion of Bladder Carcinogenesis in the Rat, Science 195: 487- 489, 1977.

5. Salley, J. J.: Experimental Carcinogenesis in the Hamster Cheek Pouch of the Syrian Hamster, J. Dent. Res. 33: 253-262, 1954.

6. Morris, A. L.: Factors Influencing the Development of Oral Carcinogenesis, J. Dent. Res. 40: 3-15, 1954.

7. Siene], W. V., and Shklar, G.: The Effect of Dimethyl Sulf- oxide and Topical Triamcinolone on Chemical Carcinogenesis of Hamster Buccal Pouch. ORAL SURG. 27: 772-779, 1969.

8. Shklar, G., Turbiner, S., and Siegel, W.: Chemical Car-

cinogenesis of Hamster Mucosa: Reaction to Dimethyl Sulf- oxide, Arch. Pathol. 87: 637-642, 1969.

9. Sbklar, G.: Recent Advances in Experimental Oral and Salivary Gland Tumors, J. Oral Surg. 28: 495-500, 1970.

10. Shklar, G.: The effect of Cortisone on the Induction and Devel- opment of Hamster Pouch Carcinomas, ORAL SURG. 23: 241- 248, 1967.

11. Sheehan, R., Shklar, G., and Tennenbaum, R.: Azathioprine Effects on the Development of Hamster Pouch Carcinomas, Arch. Pathol. 21: 264-270, 1971.

12. Shklar, G., Cataldo, E., and Fitzgerald, A.: The Effect of Methotrexate on Chemical Carcinogenesis of Hamster Buccal Pouch, Cancer Res. 26: 2218-2224, 1966.

13. Giunta, I. L., and Shklar, G.: The Effects of Antilymphocyte Serum oh Experimental Hamster Buccal Pouch Carcinogenesis, ORAL SURG. 31: 344-353, 1971.

14. Shklar, G., Schwartz, J., Grau, D., Trickier, D., and Wallace, K.: Inhibition of Hamster Buccal Pouch Carcinogenesis by 13-Cis-Retinoic Acid, ORAL SURG. 50: 45-52, 1980.

15. Wolbach, S. B., and Howe, P. R.: Tissue Changes Following Deprivation of Fat-Soluble A Vitamin, J. Exp. Med. 42: 753- 777, 1925.

16. Fell, H. B.: The Effect of Excess Vitamin A on Cultures of Embryonic Skin Explanted at Different Stages of Differentia- tion, Proc. Roy. Sot. London 146: 242-256, 1957.

17. Mock, D., and Main, J. H. P.: The Effect of Vitamin A on Hamster Cheek Pouch Mucosa in Organ Culture, J. Dent. Res. 58: 635-638, 1979.

18. Rowe. N. H., and Gorlin, R. J.: The Effect of Vitamin A Deficiency Upon Experimental Carcinogenesis, J. Dent. Res. 38: 72-83, 1959.

Volume 51 Number 3

19. Rosicky, .I., and Halschek, R.: Uber die BeeinfluBbarkeit der Kresbildung nach Benzpyrenpinselung, 2. Krebsforsch. 54: 26-30, 1943.

20. Lasnitzki, I.: The Influence of a Hypervitaminosis on the Effect of 20-Methylcholanthrene on Mouse Prostate Glands Grown In Vitro, Br. J. Cancer 9: 434441, 1955.

21. Levij, I. A., and Polliack, A.: Potentiating Effect of Vitamin A on 9, IO-Dimethyl- ,2-Benzanthracene Carcinogenesis in Ham- ster Cheek Pouch, Cancer 22: 300-306, 1968.

22. Calvaris, C. J., Acomb, D., and Velme, M.: The Effect of Vitamin A on DMBA-Induced Tumorigenesis in the Epithelium of the Golden Hamster Cheek Pouch, Pharmacol. Ther. Dent. 1: 96-101. 1971.

23. Polliack, A., and Levij, 1. S.: The Effect of Topical Vitamin A on Papillomas and Intraepithelial Carcinomas Induced in Ham- ster Cheek Pouches With 9, IO-Dimethyl-l,2-Benzanthracene, Cancer Res. 29: 321-332, 1969.

24. Wilkoff, L., and Peckham, J.: Evaluation of Vitamin A Analogs in Modulating Epithelial Differentiation of I3-Day Chick Em- bryo Metatarsal Skin Explants, Cancer Res. 36: 964-972, 1976.

25. Port, C. C., Spom, M. B., and Kaufman, D. G.: Prevention of Lung Cancer in Hamsters by 13-Cis-Retinoic Acid, Proc. Am. Assoc. Cancer Res. 16: 21-26, 1975.


26. Shklar, G., Marefat, P., Komhauser, A., Trickier, D. P., and Wallace, K. D.: Retinoid Inhibition of Lingual Carcinogenesis, ORAL SURG. 49: 325-332, 1980.

27. Santis, H., Shklar, G., and Chauncy, H. H.: Histochemistry of Experimentally Induced Leukoplakia and Carcinoma of the Hamster Buccal Pouch, ORAL SURG. 17: 207-218. 1964.

28. Jurin, M., and Tannock, I. F.: Influence of Vitamin A on Immu- nological Response, Immunology 23: 283-290, 1972.

29. Florsheim, G. L., and Bollag, W.: Accelerated Rejection of Skin Homografts by Vitamin A, Transplantation 1.5: 564-570, 1972.

30. Dresser, D. W.: Adjuvanticity of Vitamin A, Nature 271: 527- 530, 1968.

Reprint requests to: Dr. William Gilmore Department of Oral Surgery Tufts University School of Dental Medicine One Kneeland St. Boston, Mass. 02111

the effect of 13-cis-retinoic acid on hamster buccal pouch carcinogenesis

Documents