Pregnancy outcome in cancer patients. Experience in a large cooperative group.

To evaluate the potential teratogenicity and mutagenicity of modern cancer treatment, the authors enumerated from a cooperative clinical trial group 133 pregnancies in 66 women with malignant neoplasms (53% with Hodgkin's disease, 26% with other lymphomas and leukemia, and 21% with solid tumors). The gestations were divided into the following groups: Group 1, 43 pregnancies ending before therapy; Group 2, therapy given at conception or during 32 pregnancies; and Group 3, 58 pregnancies after therapy. Although the total frequencies of abnormalities were similar in Groups 1 and 2 (23% of 35 pregnancies not electively aborted and 28% of 25, respectively), there were slightly more elective abortions and birth defects related to radiation exposure at a susceptible time of gestation in Group 2. Still, there were eight normal infants among the ten fetuses who were liveborn and had first trimester exposure to chemotherapy alone; so, drug therapy early in pregnancy is not inevitably teratogenic. The apparent and surprising excess of abnormal outcomes in Group 3, 40% of 50 pregnancies, was due to low birth weight and premature terminations of pregnancy, rather than an excess of congenital anomalies. The type of unfavorable outcomes in Group 3 and their concentration in the first year posttherapy suggested they could represent defects in factors (e.g., uterine or hormonal) that normally maintain gestations, and not genetic damage to oocytes. Limitations of the data, collected by mail from physicians and their patients, included biases of self-reporting and low statistical power. Prospective study, probably through interinstitutional collaboration, seems necessary, if accurate estimates are to be made of the frequency of certain outcomes, such as spontaneous abortion and minor anomalies.