This pilot intervention study demonstrates that six weeks of HCQ treatment improves insulin sensitivity in obese non-diabetic subjects without a known systemic inflammatory condition. Animal models, randomized controlled clinical trials, and two epidemiologic studies have all shown that HCQ positively affects insulin and glucose metabolism [15
]. We found a statistically significant increase in ISI after 6 weeks of HCQ and a decrease in ISI toward baseline after stopping HCQ. The area under the curve analysis agrees with the ISI analysis, in that less insulin maintained similar serum glucose levels. Reduction in HOMA-IR during HCQ also suggests improved insulin sensitivity, in that less insulin was required to control glucose at week 6. Statistical significance was not established for these results, but the small sample size could have produced a false negative error. There were no important changes in secondary outcomes, such as CRP or IL-6, during this short-term study, and BMI was consistent.
While this pilot study does not allow us to determine the clinical relevance of HCQ's effect on insulin sensitivity, this degree of improvement in insulin sensitivity may translate into a reduced risk of DM, as suggested by two large epidemiologic studies among persons with rheumatoid arthritis [18
]. While we did not study subjects with rheumatic disease, we did look at obese individuals with increased baseline insulin resistance and an elevated baseline CRP, similar to persons with rheumatic disease. With HCQ's benefit as a disease-modifying antirheumatic drug, if it also improves insulin sensitivity in persons with rheumatic disease, it may be beneficial to maintain HCQ use in the rheumatic disease population.
The small sample size and short duration of drug administration limits this study, which should be viewed as hypothesis generating. Also, in Figures and , two different study subjects who were potential outliers were included in the statistical analysis, and the results remained statistically significant. It is important to take note of these skewed data points, even though it cannot be determined whether they are due to individualized responses to HCQ or the possibility that the subject may not have fasted before testing despite affirming to the study staff that they had.
The use of the Matsuda ISI as a surrogate measure for a clinical outcome, such as DM, is another limitation of this study. However, a paper on the current approaches to measuring insulin sensitivity suggests that the Matsuda ISI is highly correlated with results from the gold standard of metabolic testing, the euglycemic insulin clamp [24
]. Additionally, the Matsuda ISI provides a dynamic measure for analyzing both glucose uptake and insulin secretion in response to a challenge, and has proved to be an accurate predictor of DM in epidemiologic studies [24
]. Clinical studies among a variety of patient populations including those with DM, obesity, and stroke have successfully used ISI to evaluate a change in insulin sensitivity after undergoing drug treatment [25
]. Studies using metformin, moxonidine, glyburide, glargine insulin, rosiglitazone and pioglitazone have reported changes in ISI ranging from 0.14 to 1.18 after treatment [25
]. In this study, even larger changes in ISI after HCQ treatment were reported.
In this small pilot study we found that during HCQ use, obese non-diabetic subjects experienced a significant benefit in insulin sensitivity. No concurrent evidence of improvement in inflammatory markers was observed (for example, CRP and IL-6 did not change). This argues for a direct effect of HCQ on insulin metabolism-reduced degradation or enhanced activity at the receptor level, rather than an indirect effect through reduced inflammation. An important next step in this line of investigation would be a larger and longer study examining the effect of HCQ on insulin sensitivity, focusing on subjects with systemic inflammatory conditions, a population at an increased risk for insulin resistance and DM [30
]. If HCQ improves insulin sensitivity in people with systematic rheumatic disease, a diabetes prevention trial should be considered for high risk patients with RA or SLE.