Adrenocortical Function in Bronchogenic Carcinoma: Study of 100 Patients
HURST B. HATCH, JR., M.D., ALBERT SEGALOFF, M.D., ALTON OCHSNER, M.D.
From the Section on Chest Diseases, and the Departments of Internal Medicine and Surgery, Ochsner Clinic, and the Division of Endocrine Research, Alton Ochsner Medical Foundation, New Orleans, Louisiana
CERTAIN observations2, 4-6, 12 suggest a relationship between adrenocortical func- tion and carcinoma of the lung: 1) bron- chogenic carcinoma has a predilection for metastasizing to the adrenal glands and oc- casionally causing adrenal insufficiency from adrenal destruction; 2) postmortem studies indicate a significant incidence of adrenal hyperplasia in patients with car- cinoma of the lung and 3) adrenal hyper- function has been demonstrated in some patients with bronchogenic carcinoma.
We became interested in the effect of bronchogenic carcinoma on adrenocortical function when three patients died after operation for bronchogenic carcinoma in an unrecognized but, in retrospect, classi- cal state of adrenal insufficiency. Review of necropsies performed at Ochsner Foun- dation Hospital from 1954 through 1959 on 54 patients who died from bronchogenic carcinoma revealed a striking incidence of metastasis to the adrenal gland (54% ).
We therefore studied 100 cases of carci- noma of the lung encountered at the Ochsner Clinic from September, 1959 to July, 1963 to ascertain whether or not there was a relationship between bronchogenic cancer and function of the adrenal glands.
Presented before the Southern Surgical As- sociation, Dec. 8-10, 1964, Boca Raton, Fla.
This study was supported by the Daisy Deen Herring Memorial Grant for Cancer Research T-103E from the American Cancer Society.
Ninety-six of the patients were men and four women. The age distribution accord- ing to decade of life is shown in Table 1.
Method
In each patient the diagnosis was con- firmed by examination of sputum, bron- choscopy, scalene node or other biopsy procedure, or thoracotomy. In each patient serum electrolyte concentrations, including sodium, potassium, chloride and CO2, were determined and routine laboratory studies were performed.
As ancillary tests, each patient had tests of hepatic function, including bromsul- phalein excretion and alkaline phosphatase, glutamic oxalacetic transaminase, glutamic pyruvic transaminase and bilirubin deter- minations in serum, as well as assessment of renal function. Most patients also had the usual studies to exclude metastases out- side the thorax.
The first 25 patients in the series were studied during ACTH infusion. Urine and blood were collected before, during and after infusion for corticoid and ketosteroid assays. These data added little, if any, help- ful information since response to ACTH infusion was considered normal. Therefore these studies were not carried out in the re- maining patients. Routine 24-hour urine collections were not done on all patients because, with the exception of one who had Cushing’s syndrome, there was no evi-
dence of increased urinary excretion of corticoids. In addition to the foregoing adrenal function tests, a Kepler water load- ing test was performed on 61 patients.
To study adrenal function 50 cc. of heparinized blood was drawn from the antecubital vein and analyzed in duplicate for Porter-Silber chromogens according to the method of Rongone.8 In 40 patients the transcortin or protein-bound cortin deter- minations were performed according to the equilibrium dialysis method of Slaunwhite and Sandberg.10
Results
Clinical Features. Symptoms produced by bronchogenic carcinoma were essentially those due to local irritation, erosion of the tumor or invasion and metastasis. Meta- static disease and extrathoracic manifesta- tions of carcinoma were insignificant with respect to this study. Only one patient had Cushing’s syndrome at the time of admis- sion and none had clinical evidence of adrenal insufficiency.
Chemical Features. Seventy-one patients had elevated plasma corticoid levels. In 25, the level was above 20 µg .; the highest value was 54 µg. (Table 2).
Six of the 40 patients on whom the trans- cortin or protein-bound cortin in plasma was determined had significantly elevated transcortin levels.
Results of the Kepler water loading test were positive (indication of adrenocortical hypofunction ) in 46% of 61 patients (Table 3).
| Decade | Cases |
|---|---|
| 30-39 | 3 |
| 40-49 | 12 |
| 50-59 | 42 |
| 60-69 | 32 |
| 70-79 | 11 |
| Total | 100 |
| Porter-Silber Chromogens (ug.) | Cases |
|---|---|
| 4-11 (Normal) | 29 |
| 12-19 | 46 |
| 20-27 | 20 |
| >27 ** | 5 |
| Total | 100 |
* Normal mean = 10.5 µg. S.D. = 2.3.
** Highest value = 54 µg.
Discussion
Several clinical features are evidently the result of a relationship between carcinoma of the lung and adrenocortical function: 1) elevated plasma corticoid levels and functional hypo-adrenalism, as evidenced by absence of diuresis after water loading tests; 2) normal adrenal function; 3) typical Cushing’s syndrome-the most commonly reported abnormality (but actually among the rarest ) in this relationship 2, 4-6, 12; 4) rapidly progressive weakness associated with severe hypokalemic alkalosis, muscle wasting and death-with electrolyte loss suggestive of excessive mineral-corticoid effect11; and 5) Addison’s disease due to adrenal destruction by metastatic disease.
In 70 of our patients only laboratory evidence of abnormal adrenocortical func- tion was demonstrable; there were no clini- cal features suggestive of a relationship between carcinoma and adrenocortical function. The only patient with a clinical abnormality had typical Cushing’s syn-
| Plasma Corticoids | % |
|---|---|
| Elevated | |
| Positive Kepler | 35 |
| Negative Kepler | 29 |
| Normal | |
| Positive Kepler | 11 |
| Negative Kepler | 25 |
drome. “Chemical hypercorticalism” is ap- parently more common than is generally known.1
Explanations for the frequently reported hyperfunction of adrenal glands of pa- tients with carcinoma of the lung, as well as of those with cancer of other organs, have been purely speculative. One theory is that the stress produced by the carci- noma during a prolonged period serves as a stimulus to the pituitary-adrenal axis. Studies conducted at Ochsner Foundation Hospital, however, have shown that pa- tients exposed to long periods of stress as a result of chronic pulmonary disease dem- onstrate no evidence of adrenocortical hy- peractivity.9 Another hypothesis is that these patients have a predisposition to hyperadrenal function which may be re- lated to the carcinoma itself, specifically in those in whom the tumor has secreted an ACTH-like substance. This explanation is unsatisfactory because only rarely do pa- tients with carcinoma of the lung and elevated plasma corticoids have Cushing’s syndrome with the usual clinical and uri- nary hormonal changes.
We believe that some patients with car- cinoma of the lung have increased protein- bound cortin in plasma. This might explain the plasma elevation and the normal uri- nary excretion of corticoids, since protein- bound corticoid is not excreted in the urine. The transcortin level was deter- mined in a minority of patients, and the oc- currence of moderately elevated transcortin levels agrees with recent observations of Doe and associates.3 These studies indicate that estrogen, like bronchogenic carcinoma, elevates the plasma corticoid and trans- cortin concentrations, but in bronchogenic carcinoma the elevation of transcortin is proportionately less. Further investigation is needed to explain the discrepancy be- tween clinical and chemical features of pa- tients with carcinoma of the lung and adrenocortical hyperfunction.
The Kepler water test ” is helpful in as-
certaining the status of adrenal function in patients with bronchogenic carcinoma. We believe that it should be included in the routine preoperative study of all patients with carcinoma of the lung. If the result is positive the patient should be prepared with cortisone as if he had adrenal insuf- ficiency. The possibility of relative, or even absolute, adrenal insufficiency cannot be excluded in these patients on clinical evi- dence alone. Since utilizing the Kepler test, we have had no postoperative deaths at- tributable to adrenal insufficiency. There is little correlation between the plasma corti- coid level and the result of the Kepler test. This emphasizes the importance of per- forming more than one test of adrenal func- tion on patients with bronchogenic carci- noma.
We believe that an abnormal plasma corticoid level in patients with a pulmonary lesion provides evidence that the lesion may be malignant in addition to that evi- dence obtained clinically and by tissue diagnosis. Plasma corticoid determination is of value in follow up of patients who have had surgical, chemical or radiation therapy for pulmonary carcinoma. We have noted that 4 to 6 weeks after surgical re- moval of bronchogenic carcinoma the plasma corticoid level has returned to nor- mal. In patients in whom plasma corticoid concentrations have been determined at frequent intervals, we have noted eleva- tion concomitant with recurrence of car- cinoma. In several instances the plasma corticoid level has risen before symptomatic or laboratory evidence of recurrence.
Summary
Certain observations suggest a relation- ship between adrenocortical function and carcinoma of the lung, and in order to de- termine their validity the adrenocortical function of 100 patients with bronchogenic carcinoma was studied. Seventy-one per cent had elevated plasma corticoid levels, the explanation for which remains specula-
tive. The Kepler water loading test is a valuable screening procedure for patients undergoing thoracotomy for bronchogenic carcinoma. Positive results indicate func- tional adrenal insufficiency. On the basis of results obtained in this study, plasma corticoid level has diagnostic and prog- nostic significance in patients with bron- chogenic carcinoma.
References
1. Allott, E. N. and M. O. Skelton: Increased Adrenocortical Activity Associated with Ma- lignant Disease. Lancet, 2:278, 1960.
2. Bornstein, P., J. P. Nolan and D. Bernanke: Adrenocortical Hyperfunction in Association with Anaplastic Carcinoma of the Respira- tory Tract. New Engl. J. Med., 264:363, 1961.
3. Doe, R. P., R. Fernandez and U. S. Seal: Measurement of Corticosteroid-Binding Glob- ulin in Man. J. Clin. Endocr., 24:1029, 1964.
4. Kovach, R. D. and L. H. Kyle: Cushing’s Syndrome and Bronchogenic Carcinoma. Amer. J. Med., 24:981, 1958.
5. McDaniel, H. G., J. A. Pittman, S. R. Hill and W. R. Starnes: Adrenal Cortical Hyper- function Associated with Bronchogenic Car- cinoma. Amer. J. Med., 35:427, 1963.
6. Parker, T. G. and S. C. Sommers: Adrenal Cortical Hyperplasia Accompanying Cancer. Arch. Surg., 72:485, 1956.
7. Robinson, F. J., M. H. Power and E. J. Kepler: Two New Procedures to Assist in the Recognition and Exclusion of Addison’s Disease; a Preliminary Report. Proc. Staff Meet. Mayo Clin., 16:577, 1941.
8. Rongone, E. L .: Method for Determination of 17, 21-Dihydroxy-20-Ketosteroids in Urine and Plasma. Proc. Soc. Exp. Biol. Med., 101:709, 1959.
9. Segaloff, A., H. B. Hatch and E. L. Rongone: Elevated Plasma Corticoids Associated with Lung Cancer. Cancer Chemother. Rep., 16: 343, 1962.
10. Slaunwhite, W. R., Jr. and A. A. Sandberg: Transcortin; a Corticosteroid-Binding Pro- tein of Plasma. J. Clin. Invest., 38:384, 1959.
11. Spaulding, W. B., W. A. Ollie and A. G. Gornall: Mineralo-Corticoid-like Disturbance Associated with Adrenal Metastases from Bronchogenic Carcinoma. Ann. Intern. Med., 42:444, 1955.
12. Werk, E. E., Jr. and L. J. Sholiton: Adreno- cortical Function in Carcinoma of the Lung. Cancer, 13:469, 1960.