The prevalence of integrase strand transfer inhibitor (INSTI)–transmitted drug resistance (TDR) may increase with the increasing use of INSTIs.

We analyzed the prevalence of INSTI TDR in the Swiss HIV Cohort Study (2008–2014). 1316 patients had ≥1 genotypic drug resistance test performed for the integrase gene before the first exposure to an INSTI (earliest GRT per patient chosen). Samples retrieved before 2008 were summarized together as a group. We considered drug resistance mutations listed by the IAS–USA in 2015 and differentiated between minor mutations (T66AK, L74 M, E92G, T97A, E138AK, G140AS, and R263K) and major mutations (T66I, E92Q, F121Y, Y143CHR, S147G, Q148HKR, and N155H). To account for potential reversion of TDR mutations in the absence of drug pressure, we performed a subanalysis that included only GRTs from recently infected, treatment-naive patients.

We compared different aspects of the period after the introduction of INSTI (2008–2014) to the periods after introduction of NNRTIs (1998–2004), unboosted PIs (1996–2002), and PI/r (1999–2005).

Treatment failure was defined as ≥1 viral load of ≥ 500 HIV-1 RNA copies/mL (after 180 days of continuous treatment or previous viral suppression) followed by a treatment change or stop.

In 1 of 1316 drug-naive samples (0.1%), a major INSTI TDR mutation was detected (T66I, found in a sample retrieved in 2001). The most common minor mutations, most likely polymorphic, were L74M (17 of 1316 [1.3%]) and T97A (16 of 1316 [1.2%]). Minor mutations were more common in subtype non-B infections as compared to subtype B infections (24 of 466 [5.2%] vs 14 of 850 [1.6%]; P < .001, by the Fisher exact test). Prevalence of minor mutations was stable, although INSTIs were increasingly used. The results were similar when we restricted the analysis to recently infected patients (no major mutation in 303 samples). The low prevalence may be explained by the low number of patients who were potential transmitters of INSTI resistance (between 2008 and 2014, 85 patients experienced a treatment failure on ART including INSTIs in the entire SHCS database).

We showed that this is in contrast to the introduction of previous drug classes, in which more treatment failures with resistant strains occurred and TDR was observed more rapidly. In the 7 years after the introduction of unboosted PIs, PI/r, and NNRTIs, 18.2 times, 5.7 times, and 7.2 times more patients did not respond to the respective ART.

Conclusion : We demonstrated on a population-level that it is possible to avoid TDR to a new drug class for years.