Although studies show fairly consistent associations between the home use of insecticides or herbicides and other childhood cancer types, the evidence for retinoblastoma is less clear. This retrospective study found a 2.8-fold higher risk between use of these chemicals and the development of nonhereditary unilateral retinoblastoma in children.
While this study’s findings are tempered by recall bias and a small sample size of available cases with appropriately-matched controls, it highlights the need for increased awareness of the potential link between pesticide exposure before or during pregnancy and the development of retinoblastoma.
Unilateral retinoblastoma is not heritable, and arises when somatic mutations occur to both RB1 genes in a cell during early fetal development. The authors of this study focused on maternal exposure in these cases.
Detailed interviews about the use of pesticides within the home during the perinatal period were given to families of children (aged 0-15 years) with retinoblastoma. Subjects included 282 cases (186 unilateral and 96 bilateral) from the Children’s Oncology Group, a large, multi-center case-control study in which telephone interviews were conducted with the mothers. Then the researchers analyzed risk using healthy, age-matched controls.
Use of products to kill insects or lawn disease (such as Raid) 1 month prior to pregnancy or during was found to be associated with 2.8-fold higher risk of unilateral retinoblastoma (95% CI, 1.1-6.7). The use of professional lawn or landscaping services was also associated with an odds ratio of 2.8 (95% CI, 1.0-8.2) for development of unilateral disease. Home weed killer products also showed an association with higher risk, but the results were not statistically significant.
Though analysis of the matched pairs suggests that exposure also increased the odds of bilateral disease, the confidence intervals were too wide due to the small sample size.
Analysis of combined parental exposures for both types of retinoblastoma also suggests that the increased risk conveyed by pesticide use was similar among different types of pesticides, regardless of indoor or outdoor use, the time at which they were used during pregnancy, and whether it was the mother or father who applied them. However, these confidence intervals were also inconclusive.
There are several limitations to this study. Chiefly, the small sample size of available cases with appropriately-matched controls inhibited the statistical analysis. Only 99 pairs of unilateral/matched control cases and 56 bilateral/controls pairs completed the interview. Although efforts were made to recruit age-matched children who were not biologically related to the mother for unilateral cases or father for bilateral cases, in some instances the study accepted controls who were either related (7.8% of controls) or not the same age. Another issue is that the interviews were conducted with the patients’ mother regardless of whether maternal or paternal exposure was of interest, which may have excluded instances where the father was exposed unbeknownst to the mother. Finally, the data analyzed is strongly affected by recall bias, as the authors relied on the memories of parents about their use of products months to years – depending on the age of the child at enrollment – after the fact.
Awareness should be raised on the risk of unilateral retinoblastoma from parental exposure. Future research is needed to determine if certain products are more harmful than others, or if exposure during certain phases of pregnancy is more dangerous than others. In addition, further investigation into the effects parental exposure may have on bilateral disease should be undertaken, as currently there is less data on this type. The authors assert that all future studies should evaluate the causes of heritable vs. non-heritable retinoblastoma separately to avoid confounding results.