Effect of Diet Upon Radiation Response in Cervical Carcinoma of the Uterus: A Preliminary Report

The impact of ionizing radiation on cancer may be conveniently considered in two categories: (1) the direct effect on tumor cells, and (2) the indirect effect on tumor mediated through alterations in the host and tumor bed. Successful radiotherapy is the product of the direct and indirect effects on the cancer. Similarly, efforts to predict the radiocurability of cervical carcinoma can be considered under two broad headings: (1) studies of tumor characteristics, and (2) studies of host characteristics.¹

It is the purpose of this report to assess the state of radiation response in cervical carcinoma through dietary modification and supplementation. Specifically, an attempt will be made to answer the following three questions:

1. What is the radiation response in a random group of patients with biopsy-proven squamous cell carcinoma of the cervix uteri?
2. What is the radiation response in a similar age-paired group following the initiation of a relatively high-protein low-refined-carbohydrate diet and multivitamin-mineral supplementation?
3. Are there significant radiation response differences within the two groups which relate to the clinical stage of the malignancy?

Review of the Literature

The overall mean cervical cancer cure rate, notwithstanding improvements in detection,¹⁸ according to international statistical data, is only approximately 50 per cent.^7,11-13 When viewed according to the initial clinical staging, mean five year percentage survival rates approximate 75, 50, 25, and 5 respectively for Stages I through IV.^{7,11,12,29,34} Thus, from an observation of the salvage percentages, there is considerable room for improvement.

In the pursuit of acquiring prognostic data from the vaginal smear, Graham, in 1947, described specific cellular alterations after radiation therapy. Changes in the exfoliated benign epithelium consisted of one or more of the following features: increase in cell size, cytoplasmic vacuoles, multinucleation, and alteration in nuclear appearance.¹⁵ The radiation response or reaction of nonmalignant, vaginal, epithelial cells was expressed as the per cent showing the above cited cellular changes. Graham noted that the degree of cellular response to radiation therapy of cervical carcinoma varied considerably in different patients. Specifically, a favorable cytologic response was associated with survival and a poor response with poor longevity.¹⁶

There is substantial evidence in the medical literature that the radiation response or reaction of normal, exfoliated, vaginal cells has prognostic value. A good or high response has been associated with a better survival rate than a poor or low response.^{1,2,4,9,13,17,19,20,23,26-28,35} It has been recently noted that the radiosensitivity of cervical carcinoma cells is, likewise, of prognostic value. Those individuals exhibiting favorable radiation response in the malignant cells have a considerably higher recovery rate than those with a low response.¹⁰

Brief mention has been made that host radiosensitivity may be, in part, a function of oxygen tension,^8,26,30 estrogen level,^{8,22,24,31,32} anemia,³ infection,³ and carbohydrate metabolism.^5,6 However, these host factors are generally not considered to be dominant determinants.

In a summary of four-year results, Graham and associates have proposed that supplemental agents may be used to potentiate the radiation response and that the induced desired response enhances survival.¹⁴

It is possible to convert the cytologic responses from poor to good in the majority of patients with an initially poor cytologic response by the administration of testosterone propionate or alpha tocopherol. Patients with an induced good response seem to have a prognosis similar to those with a spontaneously good cytologic response.

Subsequent evaluation by Graham¹³ and Kjellgren,²¹ however, failed to denote improvement in the five-year salvage rate of tocopherol supplemented patients. These observations^13,14,21 served as a stimulus for the study reported here.

Materials and Method

Fifty-four female subjects with biopsy-proven squamous cell carcinoma of the uterine cervix participated in this experiment. Racially, there were 30 Caucasian and 24 Negro patients equally divided between the experimental and control groups. One-third were patients of local gynecologists from a middle to high socio-economic level. The remainder (two-thirds) were recipients of Tumor Clinic Aid from the State of Alabama and could be socio-economically classified as middle to low. The control and experimental groups shared equally between the private and state supported patients. Twenty-seven, subsequently referred to as the experimental group, underwent nutritional therapy. All were considered intellectually and financially able to participate. The nutritional approach was selected since adequate investigation in this field has not been conducted. Also, brief mention has been made that host radiosensitivity may be, in part, altered by anemia,³ infection,³ carbohydrate metabolism,^5,6 and tocopherol supplementation.^13,14,21

The nutritional regimen began one week prior to the initiation of radiation therapy and was continued until approximately three weeks following its termination. Specifically, each patient was instructed to consume a relatively high-protein low-refined-carbohydrate diet. Animal protein [meat, fish, fowl, eggs] was encouraged at each meal, and carbohydrates of low nutritional value [desserts, sweet snacks of all kinds, sucrose] were virtually eliminated. This dietary was selected not only because it provides an optimal intake of vitamins, minerals, and protein³³ but to compensate for the observation that a reduced glucose tolerance may accompany cervical carcinomatosis.⁶ A seven-day diet record, immediately prior to the initiation of the experimental diet, substantiated that these patients were consuming a high-carbohydrate relatively low-protein diet. One Optilets-M film tab [multivitamin-mineral by Abbott Laboratories] and two Duo-C.V.P. capsules [ascorbic acid-citrus bioflavonoid by United States Vitamin and Pharmaceutical Corporation] were taken with the morning, noon, and evening meals. These preparations were utilized to assure a vitamin-mineral excess. The supplements were provided to the patients. Intake records at weekly intervals demonstrated that all were following the dietary instructions and taking the prescribed supplements.

A like group of patients, paired by age and cancer stage, was obtained from the current records of the University of Alabama Hospital Cancer Registry. This group was not provided with dietary counsel or supplementation and will be referred to as the control group.

Radiation therapy consisted of intracavitary and external therapy. The mean dosage was 5952 r with a standard deviation of 995 r. This was administered over a period of 37 ± 7 days.

Approximately three weeks following the termination of radiation, the response in both groups for benign, exfoliated, vaginal cells was assessed. This evaluation of the radiation response in the vaginal smear was conducted according to the Graham technique.¹⁶ Each smear was read twice. Reproducibility of the radiation response is reflected by a highly significant [P < 0.001] correlation coefficient [r = 0.9040] between the first and second readings. The mean radiation response scores [the average of the two readings] are listed in Table I.

Results

Question One. Table I lists and Figure 1 pictorially emphasizes that the radiation response scores in the control group range from 0 to 100 per cent with a mean and standard deviation of 63.3 ± 34.8. On the assumption that a desirable radiation response score is reflected by a value of 70 per cent or higher,^4,26,35 16 of the 27 subjects [59%] show a favorable radiation response. Two points may be made in response to the first question. First, on a mean basis, the radiation response is poor [<70]. Second, on a more quantitative basis, approximately 40 per cent of the group displays unfavorable scores. 

Table I. Radiation Response in Age and Cancer Stage Paired Patients with Cervical Carcinoma

Fig. 1. The radiation response in a group with nutritional therapy [experimental] versus a control group.

Question Two. A study of Table I and Figure 1 also discloses the radiation response scores for the experimental group. The values range from a low of 91 to a high of 100 per cent with a mean and standard deviation of 97.5 ± 2.8. Two points may be made in answer to the second question. First, on a mean basis and utilizing the same cutoff point [>70], the radiation response is good. Second, on a quantitative basis, every subject demonstrates a favorable score.

Question Three. Table II attempts to determine whether the radiation response values are significantly different among the four subgroups within the control series. It would appear [Table I], within the limits of these small samples, that the mean radiation response scores tend to be progressively lower [and therefore poorer] with an increase in clinical state [Fig. 2]. However, none of the relationships is statistically significant; two of the combinations [I versus III and II versus III] border on significance at the 5 per cent confidence level. The lack of statistical significance may be a function of sample size.

Table II. Initial Radiation Response in Control Subjects

Fig. 2. Radiation response according to tumor stage in groups with [experimental] and without [control] nutritional therapy.

Table III is designed to analyze the statistical significance of the four subgroups [Stages I, II, III, and IV] which comprise the experimental sample. It is abundantly clear from the means and standard deviations [Table I] and statistical analysis [Table III] that the radiation responses of the subgroups are unequivocally very similar.

Finally, Table IV summarizes the statistical relationships between the control and dietary groups and their respective subgroups. It will be observed that, for the entire sample of 54 patients, the mean radiation response differences [97.5 ± 2.8 for the experimental subjects, 63.3 ± 34.8 for the control series] are statistically significantly different [P < 0.001]. Furthermore, each of the subgroups yielded significance in a comparison of the experimental and control series. The only one lacking significance [Stage IV] is obviously the result of sample size.

Table III. Initial Radiation Response in Experimental Subjects

Table IV. Initial Radiation Response in Postdietary and Control Subjects

Discussion 

From these preliminary observations, there are four points which seem abundantly clear. First, the radiation response, following routine radiotherapeutic procedures, is quite variable. This is underscored by a mean and standard deviation of 63.3 ± 34.8 per cent for the entire control sample. If 70 per cent is the delineation between a good and poor radiation response, the mean response for this group is poor. Second, it will be observed that only slightly over one-half of the control subjects showed desirable radiation responses according to conventional criteria. Third, it has been suggested that, on a mean basis, the radiation response may be inversely related to the clinical stage. Finally, the findings following dietary change are significantly different from the observations in the control series. This is true for the entire experimental versus control sample and for each of the subgroups in terms of stage except Stage IV. This subgroup was obviously too small [2 patients].

If one may assume that there is a correlation between radiocurability and radiation response, then it would follow that the experimental group should fare better both in terms of morbidity and mortality. The controls should, however, experience no better survival than has been reported for various populations.7,11,12,29,34

The question which comes to mind is the mechanism by which the radiation response is altered through dietary means. The answer is not forthcoming. However, there is abundant information in the literature to suggest that nutrition influences the endocrine system. Additionally, it is well-known that diet is an important ingredient in the enzymes vital to physiologic intermediary metabolism.

This preliminary study involves, admittedly, multiple changes [see “Materials and Method”] in the dietary regime. It should be recalled that the subjects were instructed to consume more animal protein. They were advised to eliminate sucrose and sucrose-rich products. Finally, all patients were given a multivitamin-mineral supplement and a vitamin C-water soluble bioflavonoid preparation. It might be contributory to repeat this study using isolated nutrients to determine the relative effects of protein, carbohydrates, vitamins, and minerals, in combination and singly, upon the radiation response. Also, it would be desirable if such a study could be repeated under conditions which would provide a larger sample with more sophisticated supervision of the diet [e.g. metabolic ward].

Summary

The radiation response was studied in 54 female subjects with biopsy-proven squamous cell carcinoma of the uterine cervix. Twenty-seven underwent nutritional therapy prior to irradiation; the remaining 27 served as controls. Two points deserve special mention. First, the radiation response is significantly higher following dietary changes than observed under control conditions. Second, there is the suggestion of an inverse relationship between clinical stage and the radiation response. It is hoped that this preliminary report will catalyze additional study of the effect of diet upon radiation response and radiocurability.

Acknowledgements

Grateful acknowledgement is made to Abbott Laboratories for supplying the Optilets-M film-tabs and to the United States Vitamin and Pharmaceutical Corporation for furnishing the Duo-C.V.P. capsules.

Editor’s note: Since the era in which this article was written, society’s understanding of respectful terminology when referring to ethnic and cultural groups has evolved, and some readers may be offended by references to “Negro” people and other out-of-date terminology. However, this article has been archived as a historical document, and so we have chosen to use Cheraskin’s exact words in the interest of authenticity.  No disrespect to any cultural or ethnic group is intended.

References Cited:

1. Agnew, A. M., H. K. Fidler, and D. A. Boyes: “Evaluation of radiation response.” Amer. J. Obstet. Gynec. 79: 698-699, 1960.
2. Antos, J., and O. Nyklicek: “Experience with the testing of radiosensitivity in cancer or the uterine cervix.” Neoplasma 9: 555-562, 1962.
3. Bourne, R. G.: “Factors influencing radiocurability of carcinoma of the cervix.” J. Coll. Radiol. Austral. 7: 126-127, 1963.
4. Chanen, W.: “Cytological prediction of radiosensitivity in carcinoma of the cervix.” Austral. New Zeal. J. Obst. Gynec. 3: 14-20. 1963.
5. Cheraskin, E., and W. M. Ringsdorf, Jr.: “Carbohydrate metabolism and carcinomatosis.” Cancer 17: 159-162, 1964.
6. Cheraskin, E., W. M. Ringsdorf, Jr., and K. Hutchins, A. T. S. H. Setyaadmadja, and G. L. Wideman: “Carbohydrate metabolism and cervical [uterine] carcinoma.” Amer. J. Obstet. Gynec. 97: 817-820, 1967.
7. Clark, D. G. C.: “Gynecologic cancer. Part I.” New York State J. Med. 65: 2218-2228, 1965.
8. Davis, H. J.: “Radiosensitivity and cervical cancer.” Obstet. Gynec. Survey 15: 301-313, 1960.
9. Davis, H. J., H. W. Jones, and R. J. Dickson: “The bioassay of host radiosensitivity: an index of radiocurability applied to cervical carcinoma.” Cancer 13: 358-361, 1960.
10. Dubrauszky, V.: “Assessment of radiosensitivity of carcinoma of the cervix.” J. Obst. Gynaec. Brit. Cwlth. 73: 41-43, 1966.
11. Dunn, J. E., and G. Linden: “Cancer of the Cervix–End Results Report.” Fifth National Cancer Conference Proceedings. Philadelphia, J. B. Lippincott Company, 1965, pp. 253-257.
12. Friedman, E. A., and H. C. Taylor: “Comparative cancer cure statistics.” Amer. J. Obstet. Gynec. 93: 758-763, 1965.
13. Graham, J., R. Graham, and M. Schulz: “Cancer of the uterine cervix; Harvard study, 1954 through 1956.” Amer. J. Obstet. Gynec. 89: 421-431, 1964.
14. Graham, J. B., R. M. Graham, and H. L. Kottmeier: “Potentiation of radiotherapy by supplemental agents in cancer of the uterine cervix [four-year results].” Acta Union Internat. contre Cancrum 16: 1291-1293, 1960.
15. Graham, R. M.: “The effect of radiation on vaginal cells in cervical carcinoma. I. Description of cellular changes.” Surg., Gynec. Obstet. 84: 153-165, 1947.
16. Graham, R. M.: “The effect of radiation on vaginal cells in cervical carcinoma. II. The prognostic significance.” Surg., Gynec. & Obstet. 84: 166-173, 1947. 
17. Graham, R. M., and J. B. Graham: “Cytologic prognosis in cancer of the cervix: two-year survival rates in a randomized series.” Amer. J. Roentgenol., Rad. Therap. & Nucl. Med. 87: 56-59, 1962.
18. Hammond, E. C., and H. Seidman: “Progress in control of cancer of the uterus.” Arch. Environ. Health 13: 105-116, 1966. 
19. Jones, H. W., B. Goldberg, H. J. Davis, and B. C. Burns: “Cellular changes in vaginal and buccal smears after radiation: an index of the radiocurability of carcinoma of the cervix.” Amer. J. Obstet. Gynec. 78: 1083-1100, 1959.
20. Kangas, S.: “Radiation response in patients with cervical cancer.” Acta Pathol. Microbiol. Scandinav. Suppl. 154: 179-180, 1962.
21. Kjellgren, O.: “Enhanced radiation reaction in the vaginal smear by humoral agents and surgical spaying.” Acta Cytol. 6: 455-458, 1962.
22. Kjellgren, O.: “Radiation dose and radiation reaction in the vaginal smear in carcinoma of the uterine cervix.” Acta Radiol. 58: 435- 444, 1962.
23. Kjellgren, O.: “The radiation reaction in the vaginal smear and its prognostic significance.” Acta Radiol. Suppl. 168: 1-170, 1958.
24. Koga, K., M. Yamada, and H. Tanaka: “Cytological studies on the influence of estrogen to the radiosensitivity of cervical cancer.” Kyushu J. Med. Sc. 14: 379-383, 1963.
25. Kolstad, P.: “Oxygen tension and radiocurability in cancer of the cervix.” Acta Obst. & Gynec. Scandinav. 43: Suppl. 7, 100-102, 1964.
26. McLennan, M. T., and C. E. McLennan: “Cytologic radiation response in cervical cancer: a critical appraisal, including the effect of super-voltage radiation.” Obstet. Gynec. 24: 161-168, 1964.
27. Merrill, J. A.: “Radiosensitivity studies in treatment of cancer of the cervix.” Cancer 19: 143-148, 1966.
28. Morton, D. G., N. Chang, and J. G. Moore: “Radiation response as an aid in the treatment of cervical cancer.” Pac. Med. & Surg. 73: 157-160, 1965.
29. Nielson, I. C., R. R.Smith, J. R. McLaren, and J. E. Scarborough: “Carcinoma of the uterine cervix: a study of 864 patients.” Cancer 20: 86-92, 1967.
30. Oak Ridge Institute of Nuclear Studies, Inc. Sixteenth Annual Report, 1962.
31. Sugimori, H.: “A study on the influence of estrogen on radiosensitivity of patients with cervical cancer; its clinical and histological aspects.” Kyushu J. Med. Sc. 14: 355-368, 1963.
32. Takiguchi, H.: “Studies on the effect of estrogen on the radiosensitivity of cervical cancer, with special reference to the patterns of serum lactic dehydrogenase and beta glucuronidase.” Kyushu J. Med. Sc. 14: 403-419, 1963.
33. Watt, B. K., and A. L. Merrill: “Composition of Foods, Raw, Processed, Prepared.” Agriculture Handbook No. 8. U. S. Government Printing Office, Washington, D. C., Revised December 1963.
34. Wentz, W. B., and G. C. Lewis: “Correlation of histologic morphology and survival in cervical cancer following radiation therapy.” Obstet. Gynec. 26: 228-232, 1965.
35. Zerne, S. R. M., and J. M. Morris: “Prognostic significance of cytologic response in radiation of gynecologic cancer.” Obstet. Gynec. 19: 145-155, 1962.