Background: Male survivors of childhood cancer treated with alkylating agents (AA) are at risk for azoospermia; however, the long-term dose-risk relationship is unknown. Methods: Of the 287 SJLIFE participants treated with AA but no radiation therapy, 214 provided an evaluable semen sample (mean age at diagnosis: 7.9 years, mean age at evaluation: 29.8 years, and mean years from diagnosis to evaluation: 21.6 years). Survivors were categorized as azoospermia, oligospermia (sperm concentration > 0 and < 15 million/ml), or normospermia (sperm concentration ≥ 15 million/ml). AA exposure was estimated using the cyclophosphamide equivalent dose (CED). Risks were estimated using the odds ratio (OR) and 95% confidence intervals (CI) from multinomial logistic regression analyses. Results: Among survivors 24.8% had azoospermia, 27.6% oligospermia, and 47.6% normospermia (Table). CED was negatively correlated with sperm concentration (r = - 0.25, p< 0.001). Neither age at diagnosis nor age at evaluation was significantly associated with increased risk for azoospermia. Treatment with cis-platinum or carboplatinum did not increase the risk of azoospermia or oligospermia. Adjusting for age at diagnosis and at follow-up, CED was statistically significantly associated with azoospermia (OR= 1.22, 95% CI 1.11, 1.34) and oligospermia (OR=1.14, 95% CI, 1.04, 1.25) for each 1000 mg/m²increase of CED. Conclusions: We identified neither a threshold dose, below which impaired spermatogenesis was unlikely, nor a dose above which azoospermia was uniformly present. This result suggests that other factors, including genetic variation in drug metabolizing pathways, may modulate the impact of AA exposure on spermatogenesis.