Results from the French Perinatal Cohort suggest that HIV-negative children exposed to antiretrovirals in the womb are no more likely to develop cancer than other children in France. However, the study, published in the October 18th issue of AIDS, noted an increase in rare nervous system tumours. The authors urge further study into the potential genotoxic effects of antiretrovirals, especially NRTIs.
Effective antiretroviral therapy during pregnancy dramatically reduces the risk of an HIV-positive mother passing the virus to her child. AZT monotherapy was the first treatment discovered to reduce transmission, and later, combination therapy lowered rates even further. The triple strategy of the two nucleoside reverse transcriptase inhibitors (NRTIs) AZT (zidovudine, Retrovir) and 3TC (lamivudine, Epivir) plus a protease inhibitor is now widely used, although other combinations are also available.
The long-term effects of prenatal HIV therapy on children are largely unknown. NRTIs by their nature damage DNA structure, and so are called genotoxic. Given that foetal and neonatal life sees prolific genetic activity as cells divide and differentiate, it may be particularly susceptible to genotoxic agents such as NRTIs. AZT exposure in utero has led to biological abnormalities after birth, specifically mild anaemia and other blood disorders, however its impact on cancer is unknown. During 20 years of use, neither AZT nor other NRTIs have been associated with cancer in adults or children who take them.
The French Perinatal Cohort, established in 1984, is a massive database of over 11,000 mother and child pairs and is one of the premier tools for perinatal HIV research in the field. In the current report, cohort researchers Benhammou and colleagues share findings from a study determining the incidence and types of cancer among cohort participants.
Investigators identified 10,979 children born HIV-negative to HIV-positive mothers from 1984 to mid-2007. Of the total, 9127 children (83%) were exposed to at least one NRTI during pregnancy or after birth. Length of prenatal therapy increased, from a median of eleven weeks with AZT monotherapy early in the study to 20 weeks of combination therapy by the end of the study. Only 3% of infants received therapy during and after delivery only. Following birth, 98% of infants received prophylactic therapy, either AZT monotherapy (82% of cases) or AZT plus 3TC for four to six weeks. Median age of the children at the time of analysis was a little under five and a half years.
There were ten cases of cancer detected in the cohort. Five cases were cancers of the central nervous system, three were leukaemias and two were a form of cancer of the eye called retinoblastomas. Mean age of diagnosis was 54 months, and six boys and four girls were affected. No cases of cancer were detected among the 1852 HIV-negative children not exposed to antiretroviral therapy before birth.
All children with cancer had been exposed to NRTIs: four to AZT monotherapy, three to AZT and 3TC, two to AZT with 3TC plus ddI and one to 3TC and ddI. Two children also received two doses each of nevirapine.
The investigators calculated the expected number of cases from two different French registries that tracked the general population: the 1990 to 1999 regional rates and the 2000 to 2004 national rates. Both gave numbers very similar to the observed number: 8.9 for the 1990 to 1999 data and 9.6 for the 2000 to 2004 rates. “The overall risk of cancer was not greater than that for the general population,” the investigators write. A similar finding had been seen previously in two smaller cohort studies.
When investigators looked at the rates of the different kinds of cancer, they noticed that central nervous system tumours tended to be more common than expected: five reported cases compared with 1.6 or 2.1 expected cases. This difference tended to statistical significance depending on comparison (p = 0.05 to 0.14). These cases are “worrying” say the investigators, because the tumours are generally quite rare and are related in nature, perhaps suggesting a specific effect of pre-natal therapy.
Looking at the impact of drug choice during pregnancy and comparing with AZT monotherapy, AZT plus 3TC use did not increase the risk of cancer, but 3TC and ddI was associated with a 12.5-fold increase in risk (95% CI: 2.4 to 66.1). Treatment with ddI and any other drug except 3TC was not linked to increased risk. Neither the use of AZT during delivery, nor administering NRTIs to newborns after birth was linked to increased cancer risk.
“Didanosine–lamivudine [ddI and 3TC] was given to less than 4% of treated women of the cohort and is associated with one third of tumors,” the investigators note. However, “given the small number of cancer events seen, interpretation of this difference must be cautious.”
Antiretroviral therapy to reduce transmission, which is incredibly effective, should also be made as safe as possible conclude the investigators: “Our findings certainly highlight the need for further data concerning the genotoxicity of the various NRTI combinations.”
Benhammou V et al. Incidence of cancer in children perinatally exposed to nucleoside reverse transcriptase inhibitors. AIDS 22:2165 – 2177, 2008.