It is known that hormone therapy leads to changes in insulin resistance with an increase in insulin resistance with feminizing hormone therapy, whereas reverse is true for masculinizing hormone therapy. Van Velzen and colleagues aimed to clarify whether this change in insulin resistance leads to an effect on the incidence of type 2 diabetes in patients on hormone therapy. Their study, published in the Journal of Clinical Endocrinology and Metabolism, enrolled 4099 transgender people on hormone therapy to determine incidence rates of diabetes from 2007 to 2018 compared with that in the general population.
The study used a national database register for dispensing medications (started in 2006) to identify individuals with type 2 diabetes as those on glucose lowering agents, with age cutoffs used to account for type 1 diabetes and gestational diabetes. The authors then used this data to calculate standardized incidence ratios.
The study found no difference in the incidence of diabetes in transgender individuals compared to the general population during a median follow-up since the start of hormone therapy to censoring of 11.3 years for transwomen and 5.2 years for transmen. Repeat analysis in different age categories at the start of hormone therapy as well as years in which therapy was commenced did not reveal any difference between sub-groups.
When looking at risk factors, smoking and use of alcohol did not confer any increased risk. However, body mass index at initiation of hormone therapy was associated with a hazard ratio of 1.12 in transwomen and 1.09 in transmen. This was also found to be the case for a 5-year increase in age, which also resulted in an increase in hazard ratio.
This is the first study with a relatively large sample size and longer follow-up duration to show that hormone therapy does not alter the incidence of diabetes. The authors postulate that the increase in insulin resistance in transwomen may be due to cyproterone acetate rather than hormone therapy, as it did not lead to an increase in incidence of diabetes on long follow-up.
Interestingly, another study published in the same journal recently by Islam and colleagues from the STRONG cohort did show an increased incidence and prevalence of type 2 diabetes in transgender women compared to cisgender female referrals. The STRONG cohort study used electronic health records to determine diabetes diagnosis and had a comparable number of patients as well as follow-up duration. In addition, the STRONG cohort is more ethnically and racially diverse compared to participants of the present study who were >90% white and residents of the Netherlands. The STRONG cohort, however, did not have data on risk factors for type 2 diabetes.
Taken together, these studies do not suggest any change in clinical practice for now. Current endocrine society guidelines on hormone therapy suggest diabetes screening as part of cardiovascular risk factor evaluation in individuals on hormone therapy and the van Velzen study would advocate using risk factors for diabetes as a way of individualizing this.
Limitations of this study included the lack of individual patient data, a definition of type 2 diabetes used in study, lack of racial and geographic diversity, as well as different results from another electronic health records-based cohort study. This means that prospective studies are needed before any change in current management can be suggested.
Nauman Jadoon, MRCP, FHEA, FRSPH is an Endocrinology Registrar at NHS Greater Glasgow & Clyde, Scotland.
Read the study here and a Q&A with the lead author here.