observations of this complication were in multiple myeloma, lymphoma, and ovarian cancer patients is undoubtedly that these were the first patients to receive prolonged therapy with nitrogen mustards who survived long enough to develop this complication.

As experience with cancer chemotherapy has increased, it has become obvious that this complication occurs in approximately 3 to 5 percent of all patients treated with therapeutic courses of nitrogen mustards and other alkylating agents (Tucker et al., 1988), and in some groups of patients treated with prolonged, intensive courses, this rate has been as high as 30 percent (Einhorn, 1978). The acute leukemia that is seen after alkylating agent therapy is very malignant and responds poorly to conventional therapy. The peak time of onset of this leukemia has been reported to be between three and nine years after the original treatment (Blayney et al., 1987). As larger numbers of cancer patients have been treated with alkylating agents, the evidence has become very strong that there is also an increase in the rate of solid tumors in these patients (Tucker et al., 1988). In one study of patients who had been treated for Hodgkin's disease, the patients' 10-year actuarial risk of ANL was estimated to be 5.9 percent, for lymphoma 3.5 percent, and for solid tumors 5.8 percent (Koletsky et al., 1986). Similar risks have been described in other studies, including children treated for cancer.

Therapeutic nitrogen mustards are administered systemically (except for a small experience with topical application to the skin, see Chapter 9) and are given repeatedly for periods of weeks to months. They are less reactive and have a different systemic pharmacology than sulfur mustard (Colvin and Chabner, 1990). Therefore, it is difficult to make quantitative extrapolations to the carcinogenicity of sulfur mustard and to which tumors sulfur mustard would be expected to produce. For example, acute leukemia has not been reported as a late consequence of sulfur mustard exposure. This might well be because those patients with sufficient skin or inhalation exposure to deliver a leukemic dose to the bone marrow would have succumbed to pulmonary and other complications. However, because sulfur mustards and nitrogen mustards have similar effects on DNA, the clinical experience with nitrogen mustards supports the evidence that sulfur mustard is carcinogenic in man.

There are three major gaps in the epidemiologic literature on occupational exposure. The first gap concerns the limited cohorts of workers that have been studied, Japanese and British. Other countries (including the United States) manufactured war gases; hence other cohorts of