No Inherent Association between Minor Mutations in HIV Protease at Baseline and Selection of the L90M Mutation at the Time of the First Virological Failure

To the Editor—Using data from the Italian Cohort Naive Antiretrovirals (I.C.O.N.A.), Perno et al. provide evidence for an association between the presence of minor mutations in the HIV protease at baseline and the presence of the primary resistance mutation L90M at the time of the first virological failure of a protease-inhibitor (PI)-based antiretroviral regimen [1]. In this analysis of 93 cohort subjects for whom baseline and posttherapy resistance genotype data were available, the presence of L90M at the time of virological failure was significantly associated with the presence of the minor mutations L101/V and/or M361 at baseline. Using data from the Highly Active Antiretroviral Therapy [HAART] Observational Medical Evaluation and Research (HOMER) cohort of 1191 HIV infected, antiretroviral-naive individuals initiating their first triple-drug regimen in British Columbia, Canada [2, 3], we investigated this association. We identified 500 subjects who initiated PI-based HAART and for whom baseline resistance genotype data were available. Of these 500 subjects, 111 experienced virological failure (defined as a plasma virus load [pVL] ≥500 copies/ mL after an initial suppression of pVL to <500 copies/mL) at some point during the first 30 months of follow-up. For 98 (88.3%) of these 111 subjects, resistance genotype data were available for the samples collected at the time of the first virological failure. At baseline, 9 (9.2%) of these 98 subjects had the L101/V mutation, 22 (22.4%) had the M361 mutation, and 2 (2.0%) already had the L90M mutation—results that are comparable to those in the I.C.O.N.A. [1]. At the time of the first virological failure (a median of 74 weeks after initiation of HAART), no additional L90M mutations were observed in this group. One explanation for the lack of L90M mutations observed in the HOMER cohort may be that only 7.1% of these 98 subjects were prescribed hard-gel saquinavir (Invirase) [4, 5] as their initial regimen, compared with 50.5% of the subjects in the I.C.O.N.A. [1]. Resistance genotype data from a minimum of 30 months of follow-up were available for all subjects in the HOMER cohort. We therefore investigated the appearance of the L90M mutation in the larger group of 500 subjects who initiated PI-based HAART, to determine whether the appearance of the L90M mutation at any point during the 30 months following the initiation of PI-based HAART was associated with the presence of minor protease mutations at baseline. At baseline, 41 (8.2%) of the 500 subjects had the L101/ V mutation, 107 (21.4%) had the M36I mutation, and 3 (0.6%) had the L90M mutation. At some point during the course of follow-up, new L90M mutations were selected in 9 subjects. A Cox proportional hazards regression model was used to identify factors associated with the appearance of the L90M mutation in this population. In a univariate analysis, the presence of the M361 mutation (hazard ratio [HR], 4.6; P = .02), a higher pVL (HR, 5.6/log10, increment; P = .03), and an AIDS-defining illness at baseline (HR, 4.7; P = .02) were associated with earlier selection of the L90M mutation in this population. Initiation of HAART with an indinavir-based regimen (vs. HAART containing other PIs) was associated with a decreased risk of selection of the L90M mutation (HR, 0.07; P = .0008). However, in a multivariate analysis adjusting for protease mutations at baseline, pVL, AIDS-defining illness, and type of PI at time of initiation of HAART as first therapy, the presence of the L101/V and M361 mutations at baseline was not significantly associated with the time to development of the L90M mutation in this population. However, there was a trend toward this association that may not have reached significance because of the small number of new L90M resistance mutations observed (HR, 3.4 [95% confidence interval, 0.91–13.2]; P = .07). Indinavir-containing therapy remained associated with later development of the L90M mutation (HR, 0.09; P = .002). In summary, data from the HOMER cohort do not support an association between the appearance of the L90M mutation at the time of the first virological failure and the presence of either the L 101/V or the M36I mutation. However, analysis of additional follow-up data from the HOMER cohort suggests that the presence of the M36I mutation may be associated, in some individuals, with the eventual appearance of the L90M mutation. The potential association between minorprotease mutations at baseline and the eventual selection of the L90M mutation is intriguing and merits further investigation in additional cohorts of HIV-infected individuals initiating antiretroviral therapy.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.