Immunological monitoring and immunotherapy in carcinoma of the lung

Abstract

One hundred and seven patients with carcinoma of the lung underwent immunologic testing, and 62 of these patients were randomized to an immunotherapy protocol comparing the effects of Pasteur strain BCG, either alone or combined with allogeneic tumor cells, to the effects of no immunotherapy. Patients with residual disease left at the time of surgery or with metastatic disease at the time of diagnosis showed no increase in survival as a result of this form of immunotherapy. An insufficient number of patients with less advanced disease, in whom we would expect the most beneficial effect, have been entered in this study. In general, we were unable to document substantial effects of immunotherapy on the immunologic parameters tested. Only in recall antigen skin testing was there a statistically significant increase in reactivity in the immunotherapy groups. Tests of general immune status appeared to have a predictive value in monitoring lung cancer patients. Anergic patients had a poorer prognosis than did patients who demonstrated skin test reactivity. Patients with normal percentages of lymphocytes (T cells) forming rosettes with sheep erythrocytes at 29 degrees C were generally normal in other tests of immune competence. In serial studies of rosette formation, all patients who developed recurrent disease had a pattern of depressed or falling rosette values, and these abnormalities occurred an average of 3.1 months prior to clinical detection of recurrence. Patients with large-cell anaplastic carcinoma were found to have a significantly higher incidence of depressed rosette levels than the other histologic types. Both large and small-cell anaplastic patients had significantly depressed lymphocyte proliferation by mitogens and allogeneic cells. Although lung cancer patients have been described as immunologically depressed, they are capable of recognizing tumor-associated antigens. When tested in leukocyte migration inhibition assays with tumor-associated antigens, the majority of the patients in our study were found to be reactive. The use of a 3 M KCl extract of pleural effusion cells from a patient with pulmonary adenocarcinoma has given good reactivity and specificity in lung cancer patients of all histologic types. In addition, these patients have been shown to respond in a mixed lymphocyte/tumor interaction to tumor-associated antigens (Dean, 1976b).