On the Pathological Relations Between Cancer and Tuberculosis.

Abstract

For some time past we have been engaged in a biometrical and experimental investigation of the relation between cancer and tuberculosis.∗ In the biometrical study there was first set up, by a critically adequate procedure, a case-for-case control series of autopsied patients, each one of whom was of the same sex, age and color (race) as one of the 816 persons autopsied in the Johns Hopkins Hospital and found to have carcinoma, sarcoma, or some other form of malignant tumor. The individuals in the control series did not have cancer or any other malignant tumor. With these malignant and non-malignant series of cases it was then shown that:

1. Active tuberculous lesions were found at autopsy in only 6.6% of 816 persons having malignant growths. On the other hand such tuberculous lesions were found in 16.3% of 816 persons without malignant tumors, but of the same race, sex, and age as the former group. Active tuberculous lesions therefore occurred over twice as frequently in the controls as they did in the malignant group in the present material taken as a whole.

2. Active tuberculous lesions occur 2.2 times more frequently among the controls of the carcinomatous than among those having carcinoma; whereas active tuberculous lesions occur 3.3 times more frequently among the controls of the sarcomatous than among those having sarcoma.

3. Healed tuberculous lesions (apical scars, calcified tubercles in glands, etc.) occur with equal frequency in the malignancy and control groups, whether they are taken as wholes, or when carcinoma and sarcoma (and other) are separately treated.

4. In each decade of age, and in each sex and race division of the material, the percentage of persons showing active tuberculous lesions at autopsy is markedly higher in the control group than in the malignancy group.

5. Looking at the matter from the reverse aspect, it was found that, in this autopsy experience, persons who had tuberculous lesions of any degree of activity whatever, showed a relative incidence of cancer and other malignant tumors, only a little more than half as great as that exhibited either by persons with no tuberculosis discoverable at autopsy, or by those who have only old, healed tuberculous lesions.