How Relevant is the Interaction Between Dolutegravir and Metformin in Real Life?

    loading  Checking for direct PDF access through Ovid


To the Editors:
Song et al1 have reported recently the results of a phase I parallel-group study aimed to evaluate the effect of dolutegravir on the steady-state pharmacokinetics (PK) of metformin in healthy volunteers. They found that co-administration of dolutegravir dose-dependently (dolutegravir was in fact given at 50 mg q24h or 50 mg q12h) increased metformin area under the curve (AUC) and Cmax by 79%–145% and 66%–111%, respectively. Overall, combined treatment was well-tolerated, with 1 of 29 nondiabetic subjects experiencing clinical signs of hypoglycemia. According to these findings, the authors recommended that dose adjustments of metformin should be considered to maintain optimal glycemic control when HIV-infected patients are starting/stopping dolutegravir while taking metformin. Here, we intend to verify the clinical relevance of this drug-to-drug (DDI) pharmacokinetic interaction in a real-life setting.
In this retrospective analysis, we considered all HIV-infected patients from our local database fulfilling the following mandatory criteria: (1) a diagnosis of type II diabetes from at least 1 year; (2) metformin therapy for at least 12 months; and (3) switch from any antiretroviral regimen to dolutegravir-based therapies from at least 6 months. Glycemic control was expressed both as fasting blood glucose concentrations (the last 3 measurements before the switch and the first 3 measurements after the switch to dolutegravir) and glycosylated hemoglobin (HbA1c) (the 2 most recent measurements for each observational period, with the third measurement taken at least 3 months after the switch to dolutegravir). Hypoglycemia was defined as fasting blood glucose concentrations less than 70 mg/dL, in agreement with the definition adopted by Song et al.1 Dolutegravir plasma trough concentrations were assessed by high performance liquid chromatography method with ultraviolet detection.2 Comparative analysis on glycemic control before versus after the switch to dolutegravir were performed by paired t test (significance level was verified for P < 0.05). The association between glycemic control parameters and dolutegravir plasma trough concentrations was investigated by linear regression analyses.
All HIV-infected adult patients concomitantly given metformin and dolutegravir identified from our database were included in this study (n = 15). The patients were mainly Caucasians (92%), men (92%), and had mean age of 59 ± 10 years. After the switch to dolutegravir, all patients had a good viro-immunological status (CD4 cell count: 776 ± 423 cells/mL; HIV-RNA <37 copies/mL). None of them had clinical signs of liver (serum alanine aminotransferase: 29 ± 12 IU/mL) or kidney (serum creatinine: 1.1 ± 0.3 mg/dL) dysfunction. Mean body weight and body mass index were 76 ± 13 kg and 26.2 ± 4.0 kg/m2, respectively.
As shown in Figure 1, no significant differences were observed while comparing predolutegravir versus postdolutegravir switch on mean fasting blood glucose concentrations (left panel: 140 ± 31 versus 134 ± 43 mg/dL, mean difference −4.0%; P = 0.538) or on HbA1c (right panel: 48 ± 10 versus 47 ± 13 mmol/mol, mean difference +3.0%; P = 0.589).
All patients were given dolutegravir at 50 mg/d in combination with abacavir + lamivudine (n = 4), tenofovir + emtricitabine (n = 1), atazanavir/ritonavir (n = 4), darunavir/ritonavir (n = 2), nevirapine (n = 1), or lamivudine (n = 1). A wide distribution of dolutegravir trough concentrations was observed, with values ranging from 197 to 4288 ng/mL. No significant associations were found between dolutegravir trough concentrations and the percentage differences in blood glucose levels (coefficient of correlation = −0.027) or the percentage differences in HbA1c (coefficient of correlation = −0.030) after the switch to dolutegravir. No changes in the hypoglycemic therapy (drugs and/or drug doses) were done after the switch to dolutegravir. Metformin was well-tolerated when either administered without or when administered with dolutegravir.
    loading  Loading Related Articles