Background: BF is crucial to reducing infant morbidity and mortality but may result in HIV transmission if the mother is HIV-infected. Both mART and iNVP are effective in preventing postnatal HIV transmssion. PROMISE is the first randomized trial designed to directly compare the efficacy/safety of these two strategies during extended BF.
Methods: PROMISE was conducted in sub-Saharan Africa (13 sites) and India (1 site). HIV-infected women with CD4+ counts >350 cells/mm3 (or > country-specific threshold for therapy if higher) and their HIV-uninfected newborns were randomized at 6 - 14 days postpartum to mART or iNVP. These regimens were continued until 18 months post-delivery unless there was cessation of BF, infant HIV infection, or toxicity. Kaplan-Meier (K-M) probabilities and incidence rates per 100 person-years were used in primary analyses of efficacy/safety.
Results: 2431 mother-infant pairs were enrolled between June 2011 and October 2014. Women were asymptomatic (median CD4 count - 686; 97% WHO Clinical Stage I) with median age of 26 years. Infant''s median gestational age and birthweight were 39 weeks and 2.9 kg, respectively. Baseline characteristics were comparable by study arm. Median duration of BF was 15 months and not significantly different by study arm (p=0.85). K-M estimates of MTCT of HIV at ages 6, 9 and 12 months were 0.3% (95% CI 0.1-0.6), 0.5% (95% CI 0.2-0.8) and 0.6% (95% CI 0.4-1.1), respectively, and not significantly different between the two arms. Infant 12-month survival rate was extremely high (98.9%) and not significantly different by regimen. Incidence rates of maternal/infant safety outcomes did not differ significantly by regimen (Table).

 mART N=1220iNVP N=1211P-value, K-M log-rank test
Composite maternal Grade 3/4 signs/symptoms; Grade 2-4 lab events; or maternal death (2 deaths in mART arm and 1 in iNVP arm)14.8 (95% CI 12.7-17.3)14.6 (95% CI 12.5-16.9)0.99
Composite severe maternal safety outcomes (same as row 1 but excludes Grade 2 lab events)5.1 (95% CI 4.3-6.1)5.6 (95% CI 4.8-6.6)0.61
Composite infant Grade 3/4 signs/symptoms; Grade 3/4 lab event; or infant death44.1 (95% CI 39.2-49.5)43.5 (95% CI 38.7-48.8)0.95
Infant death (16 in mART arm and 14 in iNVP arm)1.2 (95% CI 1.0-1.5)1.1 (95% CI 0.9-1.3)0.72
[Safety outcome incidence rates (per 100 person years) by study arm]

Conclusions: Both mART and iNVP were safe and were associated with very low MTCT rates during extended BF and high infant survival rates. For mothers who do not adhere to or tolerate ART, iNVP throughout BF offers a safe efficacious alternative.