The algorithm used for the interpretation of doravirine transmitted drug resistance strongly influences clinical practice and guideline recommendations

Abstract Objectives We report the results of the reverse transcriptase (RT)/protease (PR) transmitted drug resistance (TDR) prevalence study in 2018, focusing on doravirine resistance-associated mutations and the differences observed when Stanford or French National Agency for AIDS Research (ANRS)/Spanish Network of AIDS Research (RIS)/IAS-USA resistance interpretation algorithms are used to describe clinically relevant resistance. Methods We used the WHO 2009 list to investigate the prevalence of NNRTI, NRTI and PI TDR, in treatment-naive HIV-1-infected patients, adding mutations E138A/G/K/Q/R, V106I, V108I, V179L, G190Q, H221Y, F227C/L/V, M230IDR, L234I, P236L and Y318F in RT. The prevalence of doravirine resistance-associated mutations, as described by Soulie et al. in 2019, was evaluated. Clinically relevant TDR was investigated using the latest versions of ANRS, RIS, IAS-USA and Stanford algorithms. Results NNRTI mutations were detected in 82 of 606 (13.5%) patients. We found 18 patients (3.0%) with NRTI mutations and 5 patients (0.8%) with PI mutations. We detected 11 patients harbouring doravirine resistance-associated mutations (prevalence of 1.8%). Furthermore, we observed important differences in clinically relevant resistance to doravirine when ANRS/RIS (0.7%), IAS-USA (0.5%) or Stanford algorithms (5.0%) were used. V106I, which was detected in 3.8% of the patients, was the main mutation driving these differences. V106I detection was not associated with any of the clinical, demographic or virological characteristics of the patients. Conclusions The prevalence of NRTI and PI TDR remains constant in Spain. Doravirine TDR is very infrequent by RIS/ANRS/IAS-USA algorithms, in contrast with results using the Stanford algorithm. Further genotype–phenotype studies are necessary to elucidate the role of V106I in doravirine resistance.