Understanding the risk-benefit trade-off for pregnancy and infant outcomes in clinical trials of pregnant women is complex due to multiple outcomes of interest. Clinical trials often summarize risks and benefits in separate analyses, which can be misleading. Alternatively, risk and benefit can be compared by arms using a desirability of outcome ranking (DOOR) with weights to account for severity. We employed this strategy using data from the IMPAACT 2010 (VESTED) trial.
643 pregnant women living with HIV in 9 countries were randomized in 2018-2019 to one of three antiretroviral treatment arms: dolutegravir (DTG)+emtricitabine (FTC)/tenofovir alafenamide (TAF); DTG+FTC/tenofovir disoproxil fumarate (TDF); or efavirenz (EFV)/FTC/TDF. Mother-infant (MI) pair adverse outcomes were grouped according to the most severe outcome experienced: 1) infant death, 2) spontaneous abortion or stillbirth, 3) infant HIV infection (benefit via reduction), 4) very preterm delivery (<32 weeks), 5) major congenital anomaly, 6) preterm delivery (<37 weeks), 7) small for gestational age (<10th percentile, SGA), 8) infant hospitalization, and 9) infant grade 3 or 4 adverse event. Ordinal logistic regression was used to compare the odds of a more severe outcome across arms. Supplementary analyses weighted the ranked outcome according to the study team's belief of their relative severity using a tipping point method.
79/216 (37%), 93/213 (44%), and 101/211 (48%) MI pairs experienced at least one of the ranked outcomes in the DTG+FTC/TAF, DTG+FTC/TDF, and EFV/FTC/TDF arms, respectively. Ordinal logistic regression resulted in a better risk-benefit trade-off for DTG+FTC/TAF compared to EFV/FTC/TDF (OR=0.60, 95% confidence interval(CI):0.42, 0.88). In the severity-weighted analysis, DTG+FTC/TAF had a better risk-benefit trade-off relative to DTG+FTC/TDF (OR=0.64, 95%CI:0.49, 0.84) and EFV/FTC/TDF (OR=0.28, 95%CI:0.21, 0.36); DTG+FTC/TDF had a better risk-benefit trade-off relative to EFV/FTC/TDF (OR=0.41, 95%CI:0.32, 0.53).
The risk-benefit trade-off was clearer with these ranked outcome analyses, compared to the many separate previously reported analyses which favored different arms for outcomes of different severity in IMPAACT 2010. Overall, DTG+FTC/TAF provided the best and clearest risk-benefit trade-off when more severe outcomes were given more weight. Similarly, DTG+FTC/TDF had a better risk-benefit profile than EFV/FTC/TDF.