Risk factors for Kaposi’s sarcoma, non-Hodgkin’s lymphoma, and non-AIDS-defining cancers in the era of highly active anti-retroviral therapy

503 The fact that ovulation induction can raise estradiol and progesterone levels has raised concern about effects on a variety of hormone-sensitive tumors. We recently reported that clomiphene citrate may increase the risk of uterine cancers, and that fertility drugs may also be associated with slight increases in breast and ovarian tumors after long latency periods. This report examines risk of fertility drug use related to melanoma and thyroid, cervical and colon cancers, all of which have been suggested as having possible hormonal etiologies. We conducted a retrospective cohort study of 8422 women (155,527 women-years) evaluated for infertility during 1965-88. Through 1999, 581 cancers were ascertained by questionnaire or cancer and death registries. Standardized incidence ratios (SIR) compared cancer risk to the general female population. Analyses within the cohort estimated adjusted rate ratios (RR) and 95% confidence intervals (CI) associated with fertility drug use. Relative to the general population, infertile woman were at higher risk of melanoma (SIR=1.57; 95% CI, 1.1-2.1) and colon (SIR=1.76; 1.2-2.6) cancers, but not thyroid (SIR=0.99; 0.6-1.6) or cervical (SIR=0.61; 0.3-1.0) cancers. Within the cohort, 39% and 10% of women used clomiphene or gonadotropins, respectively. Clomiphene use was not significantly associated with the risk of melanoma (RR=1.66; 95% CI, 0.9-3.1), or thyroid (1.42; 0.5-3.7), cervical (1.61; 0.5-4.7), or colon (0.83; 0.4-1.9) cancers. Additionally, we found no relationship between clomiphene dose or cycles of use and any of these cancers. Clomiphene use appeared to impart stronger effects on risks of melanoma (RR=2.08; 95% CI, 0.9-4.9), thyroid (1.54; 0.3-8.0), and cervical (2.67; 0.6-11.9) cancers among women who were followed for 15+ years and on risks of melanoma and thyroid cancer among women who remained nulliparous through follow-up (RRs when compared to unexposed parous women of 2.00 (0.9-4.6) and 4.23 (1.0-17.1), respectively). Although used by very few women, we found no evidence that gonadotropins were associated with risk of these cancer sites. Although our data do not implicate fertility drugs as having strong effects on these possibly hormonally-related cancers, further follow-up should be pursued to assess long-term risks and to monitor effects among women who remain nulliparous.