June 07, 2019
Vitamin D3 supplementation in people at high risk of developing diabetes but who did not have vitamin D insufficiency does not reduce the chances of developing the disease compared with placebo, the new results of a randomized, placebo-controlled trial show.
The findings from the Vitamin D and Type 2 Diabetes (D2d) trial were presented here at the American Diabetes Association 2019 Scientific Sessions by Anastassios Pittas, MD, from Tufts Medical Center, Boston, Massachusetts.
However, he added that "in a post-hoc analysis we did see that vitamin D supplementation potentially had a benefit in those with very low vitamin D levels (20% of the study population was deemed at least vitamin D 'insufficient' as opposed to the remainder who were vitamin D sufficient)."
After 2.5 years of follow-up of over 99% of the participants, no significant difference was found in the development of diabetes between those taking vitamin D supplementation and those on placebo (P = .12).
The study was simultaneously published in the New England Journal of Medicine, along with an editorial by Deborah J. Wexler, MD, from Massachusetts General Hospital Diabetes Center and Harvard Medical School, Boston.
Wexler notes that D2d "is the largest" of a number of randomized controlled trials looking at vitamin D supplementation to prevent progression to type 2 diabetes and, as such, these results are "long-awaited."
But she points out that the findings show that "Any benefit of vitamin D for diabetes prevention, if present, is modest and clearly does not pertain to a vitamin D-sufficient population."
"Whether targeting populations with vitamin D levels below 12 ng/mL, many of whom have additional risk factors for diabetes, would have an effect on beta-cell function and progression to type 2 diabetes remains unresolved."
Trial Participants Had Prediabetes but Most Were Not Vitamin D Insufficient
The D2d multicenter trial aimed to test whether vitamin D supplementation reduces the risk of the development of type 2 diabetes among adults at high risk for the disease, ie, those with prediabetes.
The authors note the biological plausibility of vitamin D status influencing risk for type 2 diabetes, highlighting that "both impaired pancreatic beta-cell function and insulin resistance have been reported with low blood [levels] of 25-hydroxyvitamin D."
They add that observational studies suggest a low blood vitamin D level is associated with diabetes risk, and this is further supported by mechanistic studies showing that vitamin D supplementation improved pancreatic beta-cell function by 40%.
In the study, a total of 2423 participants across 22 US cities were randomized (1211 to the vitamin D group and 1212 to the placebo group) to receive 4000 IU/day of vitamin D or placebo, regardless of baseline serum 25-hydroxyvitamin D level.
The 4000 IU/day dose was selected to balance safety and efficacy and resulted in a large difference in the serum 25-hydroxyvitamin D level between the trial groups in the first 2 years of follow-up.
Participants were required to meet at least two of three glycemic criteria for prediabetes (fasting plasma glucose level, 100-125 mg/dL; plasma glucose level 2 hours after a 75-g oral glucose load, 140-199 mg/dL; and HbA1c, 5.7%-6.4%) with no diagnostic criteria for diabetes.
Researchers specifically designed D2d to include participants at high risk for type 2 diabetes regardless of their vitamin D level.
When the study began, just under 80% of participants had vitamin D levels considered sufficient by US nutritional standards (≥ 20 ng/mL).
A further 17.4% had levels between 12 and 19 ng/mL, so they were not deficient but did not have sufficient levels, while 4.3% of the overall trial participants were vitamin D "deficient" (< 12 ng/mL).
Women comprised 44.8% of participants, mean age was 60 years, and mean BMI was 32.1 kg/m2; 33.3% were nonwhite and 9.3% were Hispanic. Participants had a mean HbA1c of 5.9% (48 mmol/mol).
Participants were followed for new-onset diabetes, with blood tests performed every 6 months for a median of 2.5 years.
The primary outcome was a diagnosis of new-onset diabetes, based on annual glycemic testing of fasting plasma glucose, HbA1c, and 2-hour postload plasma glucose, and semi-annual testing of fasting plasma glucose and HbA1c.
After 2.5 years of follow-up, new-onset diabetes had developed in 293 participants in the vitamin D group and 323 patients in the placebo group (9.4 events and 10.7 events/100 person-years, respectively), Pittas reported.
So no significant difference was found in the development of diabetes between those taking vitamin D supplementation and those taking placebo (hazard ratio [HR], 0.88; 95% CI, 0.75 - 1.04; P = .12).
By month 24, the mean serum 25-hydroxyvitamin D level in the vitamin D group was 54.3 ng/mL (from 27.7 ng/mL at baseline) compared with 28.8 ng/mL in the placebo group (from 28.2 ng/mL at baseline).
The incidence of adverse events did not differ significantly between the two groups, including hypercalcemia, a fasting urine calcium–creatinine ratio of more than 0.375, a low estimated glomerular filtration (eGFR) rate, and nephrolithiasis.
In the vitamin D supplementation group, 135 participants discontinued their pills, 15 started a diabetes medication, and seven started a weight-loss medication. compared with 107, 19 and 10 participants, respectively, in the placebo group.
"We did a per-protocol exploratory analysis to see if, by excluding patients who started medications or took out-of-trial vitamin D supplements above the trial limit of 1000 IU/day, made a difference."
"We found that the primary outcome occurred in 22.0% in the vitamin D group and 25.1% in the placebo group with an HR of 0.84 [95% CI, 0.71 - 1.00]," Pittas explained.
Finally, Pittas noted that in the US population "the proportion of people with vitamin D insufficiency has improved over recent years; however, the proportion with vitamin D deficiency has remained stable, paving the way for further research with existing data."
Effect of Vitamin D Supplementation in Those With Deficiency Unknown
In her editorial, Wexler remarks that the observed HR of 0.88 for the primary endpoint of the trial does not rule out a modest benefit of vitamin D. A larger and longer trial might be necessary to show a significant benefit in a vitamin D-sufficient population, she explains.
Wexler goes on to say that the effect might be greater if vitamin D supplementation was given to people who truly have a vitamin D deficiency, in comparison to participants in the D2d trial who largely did not.
In the US, "correlates of vitamin D insufficiency include older age, black, Asian or Hispanic race, and obesity," among other factors.
"Indeed, in the D2d trial, a post hoc analysis of data from the 103 participants with vitamin D deficiency (< 12 ng/mL) showed an HR for [new-onset] diabetes with vitamin D supplementation of 0.38 (95% CI, 0.18 - 0.80)," compared with placebo, she stresses.
But Wexler adds that, overall, in a prespecified subgroup analysis of those who did not have sufficient levels, "the HR in participants with a level [of vitamin D] < 20 ng/mL was essentially the same as that in participants with a 'sufficient' level of ≥ 20 ng/mL or higher (0.87 and 0.89, respectively)."
Pittas noted that other large trials from Japan, the United States, and Norway have shown similar HRs to the D2d trial with a range of 0.87 to 0.9.
Pittas has disclosed no relevant financial relationships. Wexler has sat on the data monitoring committee for oral semaglutide, Novo Nordisk.