Leukemia in breast cancer patients following adjuvant chemotherapy or postoperative radiation: the NSABP experience.

Since 1971, 8,483 women with primary breast cancer participated in seven trials evaluating adjuvant chemotherapy. Leukemia occurred in only three of 2,068 patients treated by operation alone. The cumulative risk was 0.06% after 10 years in those free of metastases or a second primary tumor, and 0.27% in those with tumor. Thus, leukemia is not an important factor in the natural history of breast cancer. Five of 646 women receiving postoperative regional radiation developed leukemia, an overall risk of 1.39 +/- .49% at 10 years. Twenty-seven cases of leukemia (0.5%) and seven of myeloproliferative syndrome (0.1%) were recorded in 5,299 patients who received L-phenylalanine mustard (L-PAM)-containing regimens. The maximum cumulative risk of leukemia in chemotherapy recipients (leukemia of any type and myeloproliferative syndrome) was 1.68 +/- .33% at 10 years following operation. The risk excluding those with myeloproliferative syndrome was 1.29 +/- .28%. The risk of leukemia in patients free of metastases or a second primary was 1.11 +/- .30% at 10 years, and when combined with myeloproliferative syndrome, it was 1.54 +/- .36%; risks not significantly greater than observed following radiation (P = .58 and .29). No cases of leukemia were observed during the 2 years of chemotherapy and none have occurred after the seventh postoperative year. Comparisons with the surveillance, epidemiology, and end results tumor registries (SEER) data indicate an increased relative risk of acute myelogenous leukemia following postoperative regional radiation (P less than .01) and adjuvant chemotherapy (P less than .001). The findings indicate that hematologic disorders are side effects of both radiation and alkylating agents used in the adjuvant treatment of primary breast cancer. The risk of such events is lower than that reported following treatment of other solid tumors and hematologic malignancies by chemotherapy. The benefit from adjuvant chemotherapy for breast cancer exceeds the risk of leukemia. Since chemotherapy is not uniformly beneficial, efforts should be directed toward identifying responders so that only those who will benefit are exposed to the risk.