Assessment of minority frequency pretreatment HIV drug-resistant variants in pregnant women and associations with virologic non-suppression at term.

OBJECTIVE: To assess in ART-naïve pregnant women randomized to efavirenz- versus raltegravir-based ART (IMPAACT P1081) whether pretreatment drug resistance (PDR) with minority frequency variants (<20% of individual's viral quasispecies) affects antiretroviral treatment (ART)-suppression at term. DESIGN: A case-control study design compared PDR minority variants in cases with virologic non-suppression (plasma HIV RNA >200 copies/mL) at delivery to randomly selected ART-suppressed controls. METHODS: HIV pol genotypes were derived from pretreatment plasma specimens by Illumina sequencing. Resistance mutations were assessed using the HIV Stanford Database, and the proportion of cases versus controls with PDR to their ART regimens was compared. RESULTS: PDR was observed in 7 participants (11.3%; 95% CI 4.7, 21.9) and did not differ between 21 cases and 41 controls (4.8% vs 14.6%, p = 0.4061). PDR detected only as minority variants was less common (3.2%; 95% CI 0.2, 11.7) and also did not differ between groups (0% vs. 4.9%; p = 0.5447). Cases' median plasma HIV RNA at delivery was 347c/mL, with most (n = 19/22) showing progressive diminution of viral load but not ≤200c/mL. Among cases with viral rebound (n = 3/22), none had PDR detected. Virologic non-suppression at term was associated with higher plasma HIV RNA at study entry (p<0.0001), a shorter duration of ART prior to delivery (p<0.0001), and randomization to efavirenz- (versus raltegravir-) based ART (p = 0.0085). CONCLUSIONS: We observed a moderate frequency of PDR that did not significantly contribute to virologic non-suppression at term. Rather, higher pretreatment plasma HIV RNA, randomization to efavirenz-based ART, and shorter duration of ART were associated with non-suppression. These findings support early prenatal care engagement of pregnant women and initiation of integrase inhibitor-based ART due to its association with more rapid suppression of plasma RNA levels. Furthermore, because minority variants appeared infrequent in ART-naïve pregnant women and inconsequential to ART-suppression, testing for minority variants may be unwarranted.

Accuracy of a rapid real-time polymerase chain reaction assay for diagnosis of group B Streptococcus colonization in a cohort of HIV-infected pregnant women.

OBJECTIVE: There are limited data regarding Xpert performance to detect Group B Streptococcus (GBS) in HIV-infected pregnant women. We evaluated the accuracy of a rapid real-time polymerase chain reaction (PCR) test in a cohort of HIV-infected women. METHODS: At 35-37 weeks of pregnancy, a pair of combined rectovaginal swabs were collected for two GBS assays in a cohort of sequentially included HIV-infected women in Rio de Janeiro: (1) culture; and (2) real-time PCR assay [GeneXpert GBS (Cepheid, Sunnyvale, CA)]. Using culture as the reference, sensitivity, specificity, positive and negative-likelihood ratios were estimated. RESULTS: From June 2012 to February 2015, 337 pregnant women met inclusion criteria. One woman was later excluded, due to failure to obtain a result in the index test; 336 were included in the analyses. The GBS colonization rate was 19.04%. Sensitivity and specificity of the GeneXpert GBS assay were 85.94% (95% CI: 75.38-92.42) and 94.85% (95% CI: 91.55-96.91), respectively. Positive and negative predictive values were 79.71% (95% CI: 68.78-87.51) and 96.63% (95% CI: 93.72-98.22), respectively. CONCLUSIONS: GeneXpert GBS is an acceptable test for the identification of GBS colonization in HIV-infected pregnant women and represents a reasonable option to detect GBS colonization in settings where culture is not feasible.