The relationship of chemotherapeutic and endocrine intervention on natural killer cell activity in human breast cancer.

Peripheral blood natural killer (NK) activity against K562 target tumor cells was monitored in patients with breast cancer receiving no treatment, combination chemotherapy, and/or endocrine therapy. NK activity in untreated Stage I patients with no evidence of disease (ned) was significantly higher than in healthy controls. NK activity was shown to decline in individuals with cytotoxic drug therapy (P equals 0.036). There also were reduction in lymphocyte recoveries concomitant with chemotherapeutic intervention (P less than 0.001). Lymphocyte counts were incorporated in a calculation of absolute NK activity that more accurately reflected the significant reduction in NK activity that occurred in patients with localized and systemic disease on chemotherapy. Different chemotherapeutic agents were found to selectively affect NK activity. Stage II patients on phenylalanine mustard (P)/5-fluorouracil (F) (PF) and cyclophosphamide (C)/methotrexate (M)/5-fluorouracil (F) (CMF) protocols showed significant reductions in overall NK activity relative to healthy controls and Stage I patients with ned. Patients on P/doxorubicin (A)/F/tamoxifen (Tx) (PAFT) protocols showed reduced NK activity relative to Stage I patients. Patients on the short-dose C/A (CA) protocol showed normal levels of overall NK activity. High-risk Stage I patients on methotrexate (M)/F (MF)with sequential leucovorin rescue and patients with metastatic disease on endocrine therapy, i.e., Tx or megestrol acetate (Meg) showed overall NK activities in the range of healthy controls. Patients with systemic disease on CMF, CMF/vincristine/prednisone (CMFVP), vinblastine/A/thiotepa/fluoxymesterone (VATH), mitomycin/mitoxantrone (MtMx), and A regimens showed overall levels of absolute NK that were significantly less than either healthy controls or metastatic patients undergoing endocrine therapy. NK cytolytic data, monitored at multiple effector to target ratios, were subjected to exponential regression analysis. The elevation of NK cell responses in Stage I patients with ned and the decline of NK cell responses with cytotoxic chemotherapy were due to alterations in the maximal plateau levels of NK cell cytotoxicity represented by the A (asymptote) values. The k values obtained on regression analysis and indices of the relative killing capacities of individual NK cells remained unaltered in all populations. These results suggest that the cytolytic lymphocyte NK pool, elevated in Stage I patients with cancer, selectively declines as a result of cytotoxic therapy.