Adverse Drug Reactions to Integrase Strand Transfer Inhibitors

Katherine J. Lepik; Benita Yip; Ana C. Ulloa; Lu Wang; Junine Toy; Linda Akagi; Viviane Dias Lima; Silvia Guillemi; Julio S.G. Montaner; Rolando Barrios


AIDS. 2018;32(7):903-912. 

In This Article


The large cohort of 1344 persons included 1464 unique person-INSTI exposures to raltegravir, elvitegravir-cobicistat, and dolutegravir, with a total of 2377 person-years follow-up. After adjustment for potential confounders, the ADR rate for elvitegravir-cobicistat was approximately twice that of dolutegravir (hazard ratio 2.24, 95% CI 1.13, 4.44); however, there were no statistically significant differences between elvitegravir-cobicistat or dolutegravir versus raltegravir.

Other observational studies have reported INSTI ADR rates (proportions) as 5.7–7.6/100 person-years (3.6–5.8%) for raltegravir,[7–11] 4.4–10.3/100 person-years (3.4–12.3%) for elvitegravir-cobicistat,[6–11] and 3.8–13.9/100 person-years (2.5–13.7%) for dolutegravir.[3,4,6–11] Our study's findings are consistent with other cohorts, but at the lower end of the range for raltegravir 2.91/100 person-years (4.4%), and dolutegravir 2.96/100 person-years (5.2%), and at the higher end of the range for elvitegravir-cobicistat 5.94/100 person-years (9.6%).

In our cohort, the higher incidence of ADR-related discontinuation for elvitegravir-cobicistat was largely due to gastrointestinal symptoms. To date, one other study with a comparably high incidence of elvitegravir-cobicistat ADRs documented a high proportion of renal events, and noted these were likely attributable to tenofovir.[8] In the present study, renal ADRs which the reporter attributed to tenofovir did not qualify as study events unless elvitegravir-cobicistat was also specifically implicated, thus there were few elvitegravir-cobicistat renal ADRs in our cohort (2/394, 0.5%). The observation of a higher proportion of ADR-related discontinuations in the second year of elvitegravir-cobicistat treatment could not be explained by the available data, and was neither due to the occurrence of unexpected symptoms, nor influenced by the availability of newer drugs. Although we observed a higher occurrence of CNS ADRs with dolutegravir relative to other INSTIs, this difference was not statistically significant, and the 3.5% incidence of dolutegravir-related CNS effects was lower than the more than 5% incidence reported by others.[3,8] There have been recent reports of weight gain associated with INSTIs.[28,29] In this cohort, there were only two INSTI discontinuations due to weight gain (one elvitegravir-cobicistat, one dolutegravir; Supplemental Table S6a,

Taken together with the work of others, this study affirms that all INSTIs are generally well tolerated, although side-effect profiles differ between drugs. Persons who experience ADRs with one INSTI may tolerate an alternative drug in this class; however, the converse is also true, and switching from one INSTI to another may result in new side effects. In this cohort, approximately one-third of persons with an INSTI ADR had switched to the INSTI to simplify or modernize the ART regimen. This included regimen changes from twice daily raltegravir to once-daily elvitegravir-cobicistat or dolutegravir, and switches from older drugs such as atazanavir or efavirenz to an INSTI. A high proportion of these persons subsequently switched back to their previous ART regimen following the INSTI ADR. Although ADR rates were not significantly higher in persons who switched to the INSTI for the purpose of ART simplification versus other reasons for regimen change (Supplemental Table S7,, these findings serve as a reminder that ART changes for the purpose of modernization or simplification may not necessarily improve quality of life,[30] and may result in ADRs to the new regimen. Approximately one-quarter of persons experiencing an INSTI ADR had a period of ART interruption associated with the event, which underscores the clinically important consequences of ART tolerability on adherence.[31]

Comparisons of older and newer drugs can be confounded by shorter follow-up times for newer agents and changing prescribing patterns over time. Strengths of our study included providing equal, 2-year follow-up opportunity for all INSTIs, adjusting for time periods defined by INSTI availability, and excluding ADRs identified as unlikely associated with the INSTI in an independent, blinded causality review. In British Columbia, all antiretrovirals are centrally dispensed at no charge to patients; therefore ADR-related ART changes were not influenced by drug cost constraints. This population-based INSTI cohort included prescriptions from 345 physicians, reducing the influence of individual prescribing preferences. The availability of longitudinal pharmacovigilance data permitted adjustment for prior ADR history, which was found to have a significant association with subsequent ADR-related INSTI discontinuation.

The study's results must be interpreted in the context of its limitations. The application of a causality assessment tool permits consistent classification, but cannot prove the relationship between drug exposure and suspected ADR.[22] Observed ADR rates might differ in other healthcare settings or patient populations. These results cannot be generalized to children and adolescents. This study only analyzed ADRs that resulted in INSTI discontinuation, which might underestimate overall ADR rates. Persons with clinically important ADRs might be unable to switch to a different regimen if their therapeutic choices are limited by drug resistance or other contraindications. Although this analysis attempted to adjust for patient demographics and the differences in prior ART exposure and co-prescribed antiretrovirals, the inability to completely adjust for both known and unmeasured confounders may have influenced the observed ADR-related INSTI discontinuations.

In conclusion, the present study affirms that raltegravir, elvitegravir-cobicistat, and dolutegravir are generally well tolerated when prescribed in routine clinical practice. Consideration of potential differences in INSTI side-effect profiles can inform individualization of ART.