Cancer in women after assisted reproductive technology

Women initiating treatment with assisted reproduction technology (ART) have no greater short-term risk for developing cancer compared with the general population or other women treated with ART.

Barbara Luke, Sc.D., M.P.H., Morton B. Brown, Ph.D., Logan G. Spector, Ph.D., Stacey A. Missmer, Sc.D., Richard E. Leach, M.D., Melanie Williams, Ph.D., Lori Koch, B.A., Yolanda Smith, M.D., M.S., Judy E. Stern, Ph.D., G. David Ball, Ph.D., Maria J. Schymura, Ph.D.

Volume 104, Issue 5, Pages 1218-1226


To evaluate the risk of cancer after assisted reproductive technology (ART) therapy.

Longitudinal cohort study.

Not applicable.

New York, Texas, and Illinois residents between 2004 and 2009, treated with ART, comprising cycles of 113,226 women, including 53,859 women without prior ART treatment, who were linked to their respective state cancer registries and whose cycles were reported to the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System (SART CORS).


Main Outcome Measure(s):
Diagnosis of cancer, as reported to the state cancer registry; standardized incidence ratios (SIR) and their 95% confidence intervals, comparing the observed to expected cancer cases based on age-specific cancer rates in the general population of each state.

Among the cohort of women without prior ART therapy, hazard ratios (HR) and 95% confidence intervals (CI) were calculated for treatment parameters and reproductive history factors. The mean follow-up period was 4.87 years; among women without prior ART, 450 women developed 460 cancers. Women treated with ART had a statistically significantly lower risk for all cancers (for all women: SIR 0.78; CI, 0.73–0.83; women without prior ART: SIR 0.75; CI, 0.68–0.82), breast cancer, and all female genital cancers; a non-statistically-significant lower risk for endocrine and uterine cancer; and a non-statistically-significant higher risk for melanoma and ovarian cancer. Among women without prior ART, we found no statistically significant increased HR by parity, number of cycles, cumulative follicle-stimulating hormone dosage, or cycle outcome.

Women initiating ART treatment have no greater risk for developing cancer after nearly 5 years of follow-up compared with the general population and with other women treated with ART.

  • Barbara Luke

    Perhaps. There might be differences in urban versus rural populations, and across racial and ethnic groups. This was a pilot for a larger, more geographically diverse study.

  • Shvetha Zarek

    Reassuring evidence to answer a question that patients ask about daily. Excellent news that the study will be expanded to more states and will be able to evaluate long term risk. Do the authors anticipate seeing geographical variations in cancer risk outcomes?

  • Barbara Luke

    This is an excellent question. Number of cycles had no effect–which is quite reassuring. We are planning on expanding this study to more States and for a longer period.

  • msamplaski

    I personally get asked this question quite frequently: Have the hormones used in IVF been shown to have any adverse effects in the mother? This study addresses that question. Were the authors able to sub-stratify women into the number of IVF cycles that they underwent? Obviously, more hormones would seemingly be more dangerous.

Translate »