Hormone Therapy Not Linked to Sex-Based Difference in Glioma Incidence

Postmenopausal hormone therapy (HT) did not significantly increase the risk of glioma, leaving unresolved the reason for the sex disparity in incidence, a large retrospective analysis showed.

A review of data for 75,335 women produced a nonsignificant 16% increase in the hazard ratio for glioma among women with a history of HT. Analyses of current versus former use of HT and duration of HT showed no significant associations with glioma.

An extensive subgroup analysis revealed a positive association only for college-educated women, but a test for the interaction effect of education did not achieve statistical significance, reported Hui Tang, MD, of North Sichuan Medical College in Nanchong, China, and co-authors in Menopause.

“Fourteen studies have evaluated the effect of HT on women’s glioma risk,” the authors noted in the discussion of their findings. “In retrospective case-control studies, an inverse association with HT use and glioma was reported across all studies, although some findings were not statistically significant. Conversely, studies that collected HT exposure data prospectively found no significant association.”

“Our current study aligns with most previous prospective research, showing a non-significant increase in glioma risk among women who used HT.”

At least two factors might explain the lack of consistency across studies: potential recall bias and lack of differentiation between various components HT in the retrospective studies.

“[Our findings] indicate that HT use is not significantly associated with glioma risk,” Tang and colleagues concluded. “To confirm this relationship, future studies with larger sample sizes, prospective design, and extended follow-up periods are necessary, with particular focus on the specific components of HT and cumulative duration of use.”

The association between HT and glioma has been extensively studied without producing any clear associations. Perhaps it’s time to move on, suggested Stephanie Faubion, MD, of the Mayo Clinic in Jacksonville, Florida, and medical director of the Menopause Society.

“I don’t know that we’re going to get a lot more out of it [research into HT and glioma] because these brain tumors are rare,” Faubion told MedPage Today. “The ones that are more common are the ones that we really need to look at. The Mayo Clinic is actually looking at this in meningiomas. About 90% of meningiomas occur in women, and they’ve never been looked at with regard to hormone therapies.”

“Overall, I think we need to understand the disease in general,” she added. “When it’s more common in women, what makes it more common in women? Is it simplistic to think of these things as just being involved with estrogen? While it could be that simple, I think we have to look at other things that could be playing a role.”

Tang and colleagues continued the investigation of HT and glioma by analyzing data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO). Information collected from trial participants included women’s use of HT and an extensive list of sociodemographic factors. The analysis involved women ages 50-78 enrolled from 1993-2001, with a median follow-up of 11.82 years.

Among the participants in the study population, 66,777 were white, 4,270 were Black, 2,508 were Asian, 1,195 were Hispanic, 358 were Pacific Islander, 201 were Native American, and 26 had unknown racial classification. For statistical analyses, participants were categorized as white/non-Hispanic and others.

The participants had a median age of 62, and 69% were married or cohabitating. HT use was reported by 50,019 participants.

Investigators identified 101 female PLCO participants who had glioma diagnoses during the follow-up period. An unadjusted analysis comparing HT use versus none showed no association (HR 1.04, 95% CI 0.69-1.58). After adjustment for marital status, education level, and other social factors, the hazard for glioma among HT users increased slightly but failed to achieve statistical significance (HR 1.16, 95% CI 0.75-1.81).

Further adjustment for body mass index, oral contraceptive use, hysterectomy, oophorectomy, pregnancy history, age at menarche, and age at natural menopause yielded a single positive association among college-educated participants (HR 3.00, 95% CI 1.02-8.84). A test for interaction yielded a non-significant association. Analyses of HT duration yielded nonsignificant results for ≤5 years use (HR 1.33, 95% CI 0.82-2.17), 6-9 years (HR 0.47, 95% CI 0.18-1.23), and ≥10 years (HR 1.23, 95% CI 0.70-2.16).

“I think [this study] highlights how we don’t understand some of the sex-based differences that exist,” said Faubion. “This is a classic example. Migraine is another neurological condition that is way more common in women. We don’t know why. There are some hormonal variations and hormonal associations with migraine, but there are many diseases that have sex-based differences, whether they’re more common in men or women. The more we learn about these, the more we learn about them in the whole population.”

“Sex-based differences are so incredibly important, and any factors we can understand better about them is going to benefit everybody,” she added.