Kate Johnson

October 31, 2014

UPDATED November 5, 2014 // Editor's note: This article has been updated with corrected figures.

HONOLULU — Children conceived with the help of assisted reproductive technologies are not at increased risk for common childhood cancers, but the risk for rarer cancers remains unclear, new research shows.

The population-based cohort study is ongoing, however, and the numbers in the analysis to date are small.

"Our preliminary analysis has excluded moderate to strong associations of assisted reproduction with overall childhood cancer and with acute lymphoblastic leukemia, which is the most common childhood cancer," said investigator Logan Spector, PhD, from the University of Minnesota in Minneapolis. "Because of our wide confidence intervals, moderate to strong associations cannot be excluded for the less common cancers, and estimates will stabilize as data are added."

The study, which linked state cancer registries with live birth records from 2004 to 2009 in the Society for Assisted Reproduction Clinic Outcome Reporting System, involves 16 states, but not all have submitted complete data yet, Dr Spector explained. He presented the findings here at the American Society for Reproductive Medicine 2014 Annual Meeting.

"We're awaiting imminent data from California, Florida, and Illinois, which will increase our sample size appreciably," he reported. "But since we have achieved a sample size equal to that of the two other major prospective reports, we thought it was time to report on what we've found to date."

The preliminary findings differ from those of two other reports, both of which found an increased relative risk (RR) for cancer associated with assisted reproductive technology, he noted.

A British study of more than 100,000 children conceived with assisted reproductive technology demonstrated associations with hepatoblastoma (RR, 3.3) and rhabdomyosarcoma (RR, 2.6) (N Engl J Med. 2013;369:1819-1827). And a Nordic study of almost 92,000 such children demonstrated associations with central nervous system tumors (RR, 1.4) and carcinomas (RR, 2.0) (Hum Reprod. 2014;29:2050-2057).

To date, the American study involves 113,892 children conceived with assisted reproductive technology; these children are linked in a 1:10 ratio to naturally conceived children.

The mean time from birth to cancer diagnosis, infant death, or end of follow-up was 3.8 years. That period "covers the time-span of most risk for childhood leukemia and embryonal tumors [also known as neuroblastoma, retinoblastoma, nephroblastoma, and hepatoblastoma], which are two of the types we had the strongest suspicion about," Dr Spector explained.

After adjustment for sex, plurality, birth weight z-score, and maternal age, race, and education, there was no increased risk for cancer overall (RR, 1.07) or for the combination of leukemia and acute lymphoblastic leukemia.

"For other cancers, we see some elevated relative risk but very wide confidence intervals, especially ependymomas, which had the highest relative risk we observed but an extremely wide confidence interval," he noted.

"The results in this area are inconsistent and it is important to have further studies in this field," said Marie Hargreave, MSc, from the Danish Cancer Society Research Center in Copenhagen.

Hargreave has been involved in several studies showing an increased risk for cancer in children conceived with assisted reproductive technology. Her group recently reported on 123,322 such children born in Denmark from 1964 and 2006 (Int J Cancer. Published online September 25, 2014).

Although the association between fertility drugs and risk for overall childhood cancer was not significant, progesterone use by the mother before childbirth markedly increased the child's risk for acute lymphocytic leukemia (hazard ratio [HR] for any use, 4.95; HR for 3 or more cycles, 9.96) and for sympathetic nervous system tumors (HR for any use, 5.79; HR for 3 or more cycles, 8.51).

In a meta-analysis conducted by Hargreave and colleagues of 25 cohort and case–control studies involving children born after fertility treatment, the researchers found an increased risk for all cancers (RR, 1.33), for hematologic cancers (RR, 1.59), central nervous system/neural cancers (RR, 1.88), and other solid cancers (RR, 2.19) (Fertil Steril. 2013;100:150-156).

Although many of the studies in that meta-analysis did not individually show a significant increase in risk, for the pooled data, "the number was large enough to show that assisted reproductive technology was associated with a statistically significant increased risk of cancer in children, both overall and for many specific types of cancer," she explained.

For specific cancer types, there were increased risks for leukemias (RR, 1.65), neuroblastomas (RR, 4.04), and retinoblastomas (RR, 1.62).

"It is difficult to examine outcomes such as cancer in children conceived with assisted reproductive technology, because both the procedure and childhood cancer are rare events. Based on the findings of our studies and the two other studies mentioned previously, continued monitoring of the risk of long-term health effects such as cancer in these children is extremely important, as several well-conducted large studies have indicated increased risks."

This study was supported by the National Institutes of Health. Dr Spector and Ms Hargreave have disclosed no relevant financial relationships.

American Society for Reproductive Medicine (ASRM) 2014 Annual Meeting: Abstract 0-148. Presented October 21, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: