Adrenal Cortical Hyperplasia Accompanying Cancer
THOMAS G. PARKER, M.D. and SHELDON C. SOMMERS, M.D., Boston
The adrenal cortex has become a focus of interest in studies of human cancer since Huggins and co-workers1 in the last 10 years demonstrated the therapeutic effects of surgical adrenalectomy. Recently, so- called “medical adrenalectomy” after corti- sone administration also has been thought to influence favorably the courses of certain patients with metastatic cancer.2
No significant morphologic alterations of the adrenal have been observed accompany- ing the common types of cancer, although a variety of abnormal cortical secretion prod- ucts has been identified, particularly by Dobriner and his associates.3 The present analysis is part of a pathologic study of more than 1000 autopsied cancer patients, with particular attention to persons with mul- tiple primary cancers. The findings con- cerning adrenal cortical hyperplasia demon- strate (1) the types of neoplasms more and less frequently associated, (2) the higher incidence of the adrenal cortical hyperplasia accompanying multiple primary cancers, and (3) the lower incidence in noncancer as com- pared with cancer autopsies.
Submitted for publication August 3, 1955.
Aided by grants C-1413 (C) and C-1754 (R), United States Public Health Service.
From the Departments of Pathology and Surgery, Massachusetts Memorial Hospitals; the Department of Pathology, Harvard Medical School; The Cancer Research Institute, New Eng- land Deaconess Hospital, Boston; and the Depart- ment of Pathology, Pondville Hospital (Massa- chusetts Division of Public Health), Walpole, Mass.
MATERIAL AND METHODS
Clinical histories, gross autopsy protocols, and microscopic slides were examined from 1012 cancer cases provided from six hospitals in the Boston area. All available instances of multiple primary cancer + were sought, and 318 acceptable instances were included. Component cancer control cases were selected with the same sex, age within ± 10 years, and of a cancer site and histologic type corresponding to each neoplasm observed in a multiple cancer case. Six hundred ninety-four such cases with single cancers were examined. Among the cancer autopsy cases, 66.0% were men and 34.0% women. Three hundred ninety-three autopsies without cancer, of representative ages and comparable sex representations, were studied as noncancer controls.
Particular scrutiny was made for changes in endocrine glands and target organs. Certain find- ings concerning the pancreas,5 pituitary, and other organs have been reported, and additional observa- tions are to be published later.
A total of 72 cases of adrenal cortical hyper- plasia was collected, 60 in the cancer groups. Gross adrenal weights were often not recorded or were considered unreliable. All 14 adrenal hyperplasia cases with accurate complete records among the 60 cancer autopsies just cited had aggregate adrenal weights exceeding the normal maximum of 10 to 12 gm. Six cases had total adrenal tissue weighing 15 to 20 gm .; three, 21 to 25 gm. ; one, 26 to 30 gm .; two, 31 to 35 gm., and two more than 40 gm. Twelve others among this cancer group had ad- renals described as thickened or enlarged, and the remaining hyperplastic glands were referred to grossly merely as negative for metastases. Hence, hyperplasia of adrenal cortex was defined as a thickening of the cortex, so that it measured 2.0 mm. or more from capsule to medulla. The measurements were made directly on the micro- scope stage, with a translucent millimeter rule. Care was taken to identify the three cortical zones, capsule, and medulla, so that oblique sections simulating hyperplastic thickening would be ex- cluded. The definition of hyperplasia employed would exclude all but outspoken instances of adrenal cortical hyperplasia.6 All cases observed were bilateral hyperplasias of the zona fasciculata,
with abundant cytoplasmic lipid present, as judged by fine vacuoles in paraffin sections. No cases of Cushing’s or adrenogenital syndromes were recog- nized.
RESULTS
In the multiple cancer cases 24 cases of hyperplasia were found (7.5%); in the component single cancer group there were 36 cases (5.2%), and among the noncancer controls, 12 cases (3.1%). The probability of this occurring by accident is only 15 in 1000. Taken as a whole, cancer cases were accompanied by adrenal cortical hyperplasia
| 60 Cases of Adrenal Cortical Hyperplasia | ||||
|---|---|---|---|---|
| Women, 18 (30%) | Men, 42 (70%) | |||
| Age Group | Number | % of Total | % of | |
| Number | Total | |||
| 30-39 | 0 | 0 | 1 | 1.7 |
| 40-49 | 1 | 1.7 | 2 | 3.3 |
| 50-59 | 3 | 5.0 | 5 | 8.3 |
| 60-69 | 7 | 11.7 | 8 | 13.3 |
| 70-79 | 6 | 10.0 | 24 | 40.0 |
| 80-89 | 1 | 1.7 | 2 | 3.3 |
| 952 Cases Without Adrenal Cortical Hyperplasia | ||||
| Women, | 324 (34.0%) | Men, 628 | (65.9%) | |
| 20-29. | 4 | 0.4 | 3 | 0.3 |
| 30-39 | 4 | 0.4 | 2 | 0.2 |
| 40-49 | 24 | 2.5 | 25 | 2.6 |
| 50-59 | 82 | 8.6 | 102 | 10.7 |
| 60-69 | 111 | 11.7 | 204 | 21.4 |
| 70-79 | 84 | 8.8 | 222 | 23.3 |
| 80-89 | 15 | 1.6 | 67 | 7.0 |
| 90-99 | 0 | 0 | 3 | 0.3 |
| Average age: 63.1 yr. 67.8 yr. | ||||
| Average age of total group: 65.4 yr. | ||||
in 5.9%, compared with 3.1% for noncancer controls.
The ages and sexes of the 60 cancer cases accompanied by adrenal hyperplasia are analyzed in Table 1. The representation by sex and age was about the same as in the entire cancer series. The greatest frequencies occurred at 60 to 69 years, with 25% of all cases, and 70 to 79 years, with 50% of the total. The average age of these cases was 67.5 years, 66 for women and 68 for men, and 65.4, 63.1, and 67.8 years respectively for the comparable cancer groups without ad- renal hyperplasia. Among the 952 cases without adrenal cortical hyperplasia, 33.1% were 60 to 69 years old, and 32.1% were 70 to 79 years of age (Table 1).
Body weights of the 60 cancer cases with adrenal hyperplasia indicated the patients to be emaciated in 20, of normal weight in 26, and obese in 14 instances. Two-thirds of the group thus were well nourished or overweight, and inanition accompanied ad- renal hyperplasia in the remaining third.
The contingency of occurrence of adrenal cortical hyperplasia accompanying single cancers is shown in Table 2. For various sites urinary bladder cancers had the high- est incidence, 23.5% ; endometrium second highest, with 17.6%, and kidney third, with 9.5% adrenal cortical hyperplasia, but these subgroups were too small for detailed analy- sis. Respiratory tract cancers with adrenal
| Female | Male | Combined Cases | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total Cases | Cases with ACH | Inci- dence, % | Total Cases | Cases with ACH | Inci- dence, % | No. Cases | Cases with ACH | Total Inei- dence, % | ||
| 1 | Bladder | 2 | 1 | 50.0 | 15 | 3 | 20.0 | 17 | 4 | 23.5 |
| 2 | Endometrium | 17 | 3 | 17.6 | .. | 17 | 3 | 17.6 | ||
| 3 | Kidney | 6 | 0 | 0 | 15 | 2 | 13.3 | 21 | 2 | 9.5 |
| 4 | Respiratory | 7 | 1 | 14.3 | 27 | 2 | 7.4 | 34 | 3 | 8.8 |
| 5 | Prostate | 107 | 6 | 5.6 | 107 | 6 | 5.6 | |||
| 6 | Gastrointestinal. | 64 | 3 | 4.7 | 188 | 7 | 5.1 | 202 | 10 | 5.0 |
| 7 | Skin | 11 | 0 | 0 | 53 | 3 | 5.6 | 64 | 3 | 4.7 |
| 8 | Breast | 42 | 2 | 4.8 | 2 | 0 | 0 | 44 | 2 | 4.5 |
| 9 | Cervix | 24 | 1 | 4.1 | 24 | 1 | 4.1 | |||
| 10 | Pancreas | 10 | 0 | 0 | 9 | 0 | 0 | 19 | 0 | 0 |
| 11 | Ovary | 10 | 0 | 0 | .. | 10 | 0 | 0 | ||
| 12 | Ureter & Urethra | 1 | 0 | 0 | 1 | 0 | 0 | 2 | 0 | 0 |
| 13 | Testis | .. | 1 | 0 | 0 | 1 | 0 | 0 | ||
| Totals | 194 | 11 | 5.7 | 368 | 23 | 6.3 | 562 | 34 | 6.0 | |
hyperplasia in 8.8%, prostate cancers with 5.6%, and gastrointestinal cancers with 5.0% were larger homogeneous groups with high incidences. Skin, breast, and cervix all had somewhat more than 4% of cases with adrenal cortical hyperplasia. Adrenal hyperplasia was not found with 19 pan- creatic or 10 ovarian cancers.
Multiple as well as single primary cancer cases with adrenal hyperplasia are con- sidered in Table 3. Urinary bladder, endo- metrium, respiratory tract, prostate, and kidney still had the highest incidences of adrenal hyperplasia accompanying cancer of these sites. Except for pancreas, cervix, and ovary as sites, incidences of adrenal hyperplasia were still about twice the figure observed in noncancer control material.
Single and multiple cancer groups are separately analyzed in Table 4. Adrenal cortical hyperplasia was observed in about one of every 6 multiple primary cancer cases involving respiratory tract, one of 8 in pros- tate, one of 11 in gastrointestinal tract, and one of 13 in breast. There was adrenal hyper- plasia in four of eight men with both respira- tory and prostatic cancers.
Adrenal cortical hyperplasia accompanied single urinary bladder, kidney, and endo- metrial neoplasms more frequently than when multiple cancers were found. The groups available for analysis were, how- ever, too small to be significant. Collection of urogenital cancers in greater numbers would help determine whether there is regu- larly an associated alteration of the adrenal
| Single Cancers | Multiple Cancers | |
|---|---|---|
| Sites | ||
| Respiratory | 3/34 (8.8%) | 5/31 (16.1%) |
| Prostate | 6/107 (5.6%) | 14/110 (12.7%) |
| Bladder | 4/17 (23.5%) | 2/17 (11.7%) |
| Pancreas | 0/19 (0%) | 2/19 (10.5%) |
| Gastrointestinal. | 10/202 (5.0%) | 15/160 (9.4%) |
| Breast | 2/44 (4.5%) | 3/39 (7.7%) |
| Endometrium | 3/17 (17.6%) | 1/14 (7.1%) |
| Kidney | 2/21 (9.5%) | 1/19 (5.2%) |
| Cervix | 1/24 (4.1%) | 1/24 (4.1%) |
| Ovary | 0/10 (0%) | 0/10 (0%) |
cortex unlike that found accompanying other cancers.
COMMENT
Despite the collection of more than 1000 pathologically proved cases of cancer, the various individual types of neoplasm found were numerically insufficient for any valid statistical conclusions concerning subgroups. The observations made are considered chiefly suggestive and worthy of trial by others in attempts to establish or refute their validity.
Adrenal cortex hyperplasia did not appear to show any age or sex distribution strik- ingly different from the entire cancer case material collected. The increased frequency with age was not of necessity related to cancer. Old experimental animals, such as mice, have shown adrenal cortical hyper- plasia development without evident clear-cut connection to neoplastic tendencies.7
It was quite unexpected to find in both single and multiple primary cancer cases that
| Group | Total Cases | Per Cent of All Cancer | Cases of Multiple Cancers | Cases with Adrenal Cortical Hyperplasia | |||
|---|---|---|---|---|---|---|---|
| No. | % of Group | No. | % of Group | ||||
| 1 Bladder | 34 | 3.4 | 17 | 50.0 | 6 | 17.6 | |
| 2 Endometrium | 31 | 3.1 | 14 | 45.1 | 4 | 12.9 | |
| 3 Respiratory | 65 | 6.4 | 31 | 49.6 | 8 | 12.3 | |
| 4 Prostate | 217 | 21.4 | 110 | 50.7 | 20 | 9.2 | |
| 5 Kidneys. | 41 | 4.1 | 20 | 48.7 | 3 | 7.3 | |
| 6 Gastrointestinal. | 362 | 35.8 | 160 | 44.2 | 23 | 6.4 | |
| 7 Breast. | 83 | 8.2 | 39 | 46.9 | 5 | 6.0 | |
| 8 Pancreas | 38 | 3.7 | 19 | 50.0 | 2 | 5.2 | |
| 9 Cervix | 48 | 4.7 | 24 | 50.0 | 2 | 4.2 | |
| 10 | Ovary | 20 | 1.9 | 10 | 50.0 | 0 | 0 |
respiratory-tract, prostate, and gastroin- testinal sites were the largest homogeneous neoplastic groups with the highest incidences of accompanying adrenal cortical hyperplasia. That the frequencies in each instance fol- lowed in the same order likewise appeared notable. In cancers of bladder, kidney, and endometrium the frequency of adrenal cor- tical hyperplasia was even higher, but the small number of cases and selection factors hindered interpretation of the significance.
Persons with multiple primary cancers had adrenal hyperplasia on the order of twice as frequently as those with single cancers of similar sites. This was evident and statistically significant to a high degree for respiratory tract, prostatic, and gastro- intestinal neoplasms, but not for the small urogenital subgroup. Adrenal participation in a complex endocrine imbalance in many persons harboring multiple separate malig- nant tumors is thus suggested. From analy- sis of autopsy material it is difficult to estab- lish whether the adrenal enlargement was a forerunner or an accompaniment of an established cancer.
Pertinent experimental procedures 1” lead- ing to adrenal cortical hyperplasia in ani- mals are, in particular, the following: 1. Estrogen effects. Study of ovaries from human cases of lung and stomach cancer makes it evident that stromal hyperplasia and inferred continuous ovarian estrogen secretion characterize a large majority of women with these two types of cancer.11 Evidence recently collected also favors long- continued estrogenic stimuli in the back- ground of about two-thirds of men with prostatic cancers. 2. Hypermetabolic and protein catabolic states. These are sometimes lumped under the term “chronic stress” and might theoretically be implicated as cocar- cinogenic influences, but there does not ap- pear to be any reliable evidence that “stress” is a characteristic part of the background of human cancers.
In experimental situations starvation and inanition associated with a loss of tissue protein are sometimes accompanied by an
enlargement of the adrenal glands. The cancer cases with adrenal hyperplasia col- lected were in the majority not cachectic, however, and weight loss had not been great in two-thirds of the group. Whereas the presence of cancer alone was not thought sufficient to explain the adrenal cortical en- largements observed. inanition did not ap- pear to account for its frequency.
Adrenal cortical hyperplasia could be in- duced secondarily through the metabolic and endocrinologic activities of established can- cers. Evidence for such secondary hyper- functions includes the uniform finding of increased hypertrophic amphophile cells in the anterior pituitary lobes of patients dying of a variety of types of cancer .* Mellgren 1: first recognized hypertrophic amphophile cells as connoting adrenal cortical hyper- function, either nonspecific or of clinically specific types, like Cushing’s syndrome. While more data need to be accumulated. at present it would appear that a pituitary- adrenocortical hyperfunction, involving pi- tuitary and adrenal hormones not yet iso- lated. is the rule in persons dying with malignant tumors. Corticotropin (ACTH) and cortisone do not seem to participate. Adrenal stimulation by pituitary amphophile and hypertrophic amphophile cells results in an adrenal cortical hyperplasia in animals and probably in some patients. Secondary stimulation following cancer growth may likely explain the hyperplastic adrenals ac- companying the cancers of skin and cervix, as well as some breast cancers.
Whether adrenal cortical hyperplasia pre- cedes or follows the establishment of a can- cer. its structure and the abundant lipid suggest functional activities. The relative frequency with which adrenal enlargement accompanied respiratory tract, prostatic, and gastrointestinal cancers, particularly those of the stomach, would favor the employment of surgical or “medical” adrenalectomy pro- cedures on late inoperable cases of these types. The rather heterogeneous urinary tract cancers might provide another favor-
* References 9 and 12.
able group for adrenalectomy. Persons with multiple primary cancers of these sites would also be theoretically more likely to benefit from ablation or inhibition of the adrenal fasciculate zone functions. A present-day regimen developed to manage prostatic car- cinoma involves the successive use of (1) surgical ablation of the primary site and orchidectomy, (2) cortisone, (3) adrenal- ectomy, and (4) hypophysectomy.1+
It is reiterated that a retrospective study of this type has recognized limitations of scope and significance. Differences in the frequency of adrenal cortical hyperplasia accompanying certain cancers or combina- tions of cancers, however, warrant attention because of the distinguishable host reactions to different forms of cancer.
SUMMARY
Adrenal cortical hyperplasia was found in 72 of 1405 autopsied cases reviewed patho- logically, including 60 cases among 1012 cancer autopsies. Adrenal cortical hyper- plasia accompanied 7.5% of the instances of multiple primary cancer, 5.2% of single- cancer cases, and 3.1% of 393 noncancer controls. Relatively high incidences of hyper- plasia were observed with cancers of the respiratory tract, prostate, and gastroin- testinal tract and in urinary tract neoplasms. Whether adrenal cortical hyperplasia de- veloped before or after the establishment of cancer growth is discussed in reference to the mechanism of adrenal stimulation. Thera- peutic adrenalectomy appears worthy of trial, particularly with inoperable cancers of the respiratory and gastrointestinal tracts.
The following pathologists provided material for this study: Dr. A. T. Hertig, Free Hospital for Women; Drs. S. P. Hicks and G. L. Dammin, Peter Bent Brigham Hospital ; Dr. P. M. LeCompte, Faulkner Hospital ; Dr. W. A. Meissner, New Eng- land Deaconess Hospital, and Dr. L. Reiner, Beth Israel Hospital, Boston. Drs. A. T. Hertig, Hugo Muench, and Shields Warren made suggestions and reviewed the manuscript.
REFERENCES
1. Huggins, C., and Bergenstal, D. M .: In- hibition of Human Mammary and Prostatic Can- cers by Adrenalectomy, Cancer Res. 12:134, 1952.
2. Lemon, H. M .; Walker, B. S .; Reynolds, M. M. D., and Wotiz, H. H .: Biochemistry of Human Cancer (Concluded), New England J. Med. 251:1011, 1954.
3. Dobriner, K .; Lieberman, S .; Wilson, H .; Ekman, B .; Pearson, O., and Eliel, L .: Adrenal Function and Steroid Excretion in Neoplastic Dis- ease, in Proceedings of the First Clinical ACTH Conference, edited by J. R. Mote, Philadelphia, The Blakiston Company, 1950, pp. 158-166.
4. Warren, S., and Ehrenreich, T .: Multiple Primary Malignant Tumors and Susceptibility to Cancer, Cancer Res. 4:554, 1944.
5. Sommers, S. C .; Murphy, S. A., and Warren, S .: Pancreatic Duct Hyperplasia and Cancer, Gastroenterology 27:629, 1954.
6. LeCompte, P. M .: Width of Adrenal Cortex in Lymphatic Leukemia, Lymphosarcoma and Hyperthyroidism, J. Clin. Endocrinol. 9:158, 1949.
7. Frantz, M .; Kirschbaum, A., and Casas, C .: Endocrine Interrelationship and Spontaneous Tumors of Adrenal Cortex in NH Mice, Proc. Soc. Exper. Biol. & Med. 66:645, 1947.
8. Lemon, H. M., and Sommers, S. C .: Clinico- pathologic Conference, Am. J. Clin. Path. 24:1402, 1954.
9. Burt, A. S., and Castleman, B .: Some Histo- logical Effects of Estrogens and Castration on the Anterior Pituitary in Women with Carcinoma of the Breast, Cancer 6:236, 1953.
10. Tepperman, J .; Engel, F. L., and Long, C. H. : A Review of Adrenal Cortical Hypertrophy, Endocrinology 32:373, 1943.
11. Sommers, S. C., and Lombard, O. M .: Cancer Associated with Ovarian Stromal Hyper- plasia, A. M. A. Arch. Path. 56:462, 1953.
12. Russfield, A. B .: The Endocrine Glands After Bilateral Adrenalectomy Compared with Those in Spontaneous Adrenal Insufficiency, Can- cer 8:523, 1955.
13. Mellgren, J. A .: The Anterior Pituitary in Hyperfunction of the Adrenal Cortex, Acta path. et microbiol. scandinav., Supp. 60, 1945.
14. Harrison, J. H .; Leman, C. ; Munson, P. L., and Laidlaw, J. C .: Hormone Excretion Before and After Castration and Adrenalectomy, New England J. Med. 252:425, 1955.