Maternal diabetes appears to raise the risk for certain types of cancers in children, according to an analysis of two large population-based registries.
Most notably, researchers found an elevated risk of glioma among children who had been prenatally exposed to type 1 diabetes in one registry and to gestational diabetes in the other.
“Our study supports the potential role for maternal diabetes in cancer risk in offspring,” the authors write.
But the researchers and an outside expert urged “caution” in interpreting the findings, especially those involving small numbers of cancer cases.
The study was published online in the British Journal of Cancer.
Diabetes during pregnancy accelerates fetal growth and increases expression of proinflammatory cytokines in the placenta — two possible biologic mechanisms linking maternal diabetes and childhood cancers.
However, the impact of maternal diabetes on childhood cancer risk has not been extensively studied.
Using two population-based registries in Denmark and Taiwan, Xiwen Huang, PhD candidate, UCLA Fielding School of Public Health, Los Angeles, and colleagues examined the contributions of different types of maternal diabetes — gestational, type 1, and type 2 — to the risks of childhood cancers.
The cohort from Denmark included 6420 cancer patients and 160,484 control persons, and the cohort from Taiwan included 2160 cancer patients and 2,076,877 control persons.
In Denmark, type 1 diabetes was associated with an almost 2.5-fold increased risk for central nervous system (CNS) tumors (odds ratio [OR], 2.44), but no such association was found for gestational diabetes (OR, 0.95).
Huang and colleagues also found that prenatal exposure to type 1 diabetes was associated with an increased risk for gliomas (OR, 2.33; 95% CI,1.04 – 5.22), especially for astrocytoma (OR, 3.61 for six cases that involved exposure).
Owing to the limited sample size, the researchers could not evaluate associations between type 2 diabetes and specific cancer types in the Denmark population.
Turning to Taiwan, the authors found an increased risk of gliomas in offspring who had been exposed prenatally to gestational diabetes (hazard ratio [HR], 1.59; 95% CI, 1.01–2.50), but there was no association between type 2 diabetes and gliomas.
However, offspring prenatally exposed to maternal type 2 diabetes did have a twofold increased risk of hepatoblastoma (HR, 2.02; 95% CI, 1.02 – 4.00). Taiwan, like other regions in Asia, has a high rate of hepatoblastoma compared to the US and Europe. (A link between diabetes and hepatoblastoma could not be assessed in the Danish analysis because of insufficient statistical power, the researchers note.)
Although the findings suggest that “maternal diabetes increased the risk of certain childhood cancer in Denmark and Taiwan,” the authors acknowledge several limitations to the study.
The prevalence of diabetes types across the populations differed — type 1 diabetes was much more common in Denmark, and type 2 diabetes and gestational diabetes were more common in Taiwan. This difference “limits comparability” between the populations and affects “our statistical power to estimate effects,” the authors note.
In addition, the authors note that “cancer types involving <10 exposed cases, as well as those with wide 95% confidence intervals, should be interpreted cautiously.”
Reached for comment, Jian-Min Yuan, MD, PhD, a cancer epidemiologist, said the findings for maternal diabetes and risk for CNS tumors, glioma, or hepatoblastoma are “interesting” but that the results require “some caution” in interpretation.
First, the findings were not consistent between the two studies. “In the Denmark study, the association was for diabetes, mainly type 1 diabetes, with risk of glioma or CNS tumors, whereas in the Taiwan study, it was type 2 diabetes with risk of hepatoma,” Yuan, of UPMC Hillman Cancer Center, Pittsburgh, Pennsylvania, told Medscape Medical News.
Second, adjustments were not made for maternal obesity or body mass index in the statistical analysis, since obesity and diabetes are highly correlated; this could complicate the interpretation of biological mechanism, Yuan said.
Another caveat: The effective sample size — the number of cases in the high-risk groups — was “extremely small” (fewer than 10 cases), which could mean there is a high likelihood of chance findings, Yuan noted.
The Danish study was supported by a grant from the National Institutes of Health. The Taiwanese study was supported by a grant from Alex’s Lemonade Stand Foundation. The authors and Yuan have disclosed no relevant financial relationships.
Br J Cancer. Published September 10, 2022. Full text
For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.
Source: Read Full Article