Table 3.
Comparison of Women With Virologic Failure Versus Antiretroviral Therapy (ART) Suppression During Efavirenz-Based ART Within Prior PROMISE Antepartum Treatment Arms
| Characteristic | Antepartum ZDV Monotherapy | Antepartum ZDV/3TC + LPV/r | Antepartum TDF/FTC + LPV/r | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Virologic Failure (n = 76) | ART-Suppressed (n = 453) |
Unadjusted OR (95% CI) | Virologic Failure (n = 89) |
ART-Suppressed (n = 437) |
Unadjusted OR (95% CI) | Virologic Failure (n = 24) |
ART-Suppressed (n = 153) |
Unadjusted OR (95% CI) | |
| Clinical data, median (IQR) | |||||||||
| Age, y (analyzed per 5 y) | 24 (21–27) | 26 (23–30) | 0.65 (0.49–0.84)** | 24 (21–28) | 27 (23–31) | 0.67 (0.53–0.85)*** | 22.5 (20.8–28) | 26 (23–30) | 0.67 (0.41–1.05) |
| HIV RNA load at PROMISE entry, log10 copies/mL | 3.95 (3.56–4.56) | 3.88 (3.47–4.35) | 1.30 (0.90–1.87) | 4.12 (3.70–4.60) | 3.91 (3.42–4.45) | 1.44 (1.03–2.00)* | 4.24 (3.54–4.44) | 3.99 (3.57–4.42) | 1.09 (0.58–2.00) |
| HIV RNA load at EFV initiation, log10 copies/mL | 4.02 (3.64–4.74) | 3.97 (3.30–4.51) | 1.43 (1.06–1.94)* | 3.82 (3.46–4.57) | 3.74 (3.28–4.37) | 1.23 (0.90–1.69) | 3.96 (3.51–4.54) | 3.91 (3.36–4.45) | 1.39 (0.77–2.53) |
| Time between delivery and EFV initiation, wk (analyzed per 26 wk) | 4.22 (2.73–5.66) | 4.73 (2.85–6.15) | 0.94 (0.84–1.05) | 4.77 (2.54–5.92) | 4.69 (2.96–6.23) | 0.97 (0.87–1.08) | 2.88 (2.41–4.01) | 2.42 (1.46–3.35) | 1.25 (0.96–1.63) |
| Time between last viremic visit and EFV initiation, wk (analyzed per 4 wk) | 0.04 (0–0.94) | 0.29 (0–6.21) | 0.96 (0.92–0.99)** | 0.21 (0–3.11) | 0.29 (0–7.29) | 0.97 (0.95–1.00)* | 0.25 (0–3.54) | 0.71 (0–6.18) | 0.96 (0.88–1.02) |
| HIV RNA load at delivery | |||||||||
| ≤400 copies/mL | 12 (15.8) | 107 (23.6) | Reference | 49 (55.1) | 364 (83.3) | Reference | 11 (45.8) | 115 (75.2) | Reference |
| >400 copies/mL | 62 (81.6) | 324 (71.5) | 1.66 (.90–3.29) | 39 (43.8) | 68 (15.6) | 4.25 (2.59–6.95)*** | 12 (50) | 36 (23.5) | 3.44 (1.42–8.44)** |
| Hepatitis B virus test result | |||||||||
| HBsAg negative | 76 (100) | 439 (96.9) | Reference | 89 (100) | 425 (97.3) | Reference | 23 (95.8) | 138 (90.2) | Reference |
| HBsAg positive | 0 (0) | 14 (3.1) | 0.2 (.00–1.51) | 0 (0) | 12 (2.8) | 0.19 (.00–1.47) | 1 (4.17) | 15 (9.80) | 0.57 (.06–2.48) |
| HIV genotype at EFV initiationa,b | |||||||||
| Wild-type | 68 (89.5) | 387 (86.4) | Reference | 76 (85.4) | 373 (85.8) | Reference | 15 (65.2) | 134 (88.7) | Reference |
| Any resistance mutations | 8 (10.5) | 61 (13.6) | 0.78 (.34–1.60) | 13 (14.6) | 62 (14.3) | 1.05 (.54–1.94) | 8 (34.8) | 17 (11.3) | 4.21 (1.55–11.1)** |
| NRTI mutation(s) only | 0 | 7 | 0.38 (.00–3.16) | 0 | 5 | 0.44 (.00–3.98) | 0 | 0 | … |
| K103N only | 5 | 38 | 0.81 (.29–1.90) | 6 | 34 | 0.92 (.35–2.07) | 5 | 9 | 5.02 (1.47–16.0)* |
| Y181C only | 0 | 3 | 0.81 (.01–8.47) | 0 | 5 | 0.44 (.00–3.98) | 2 | 2 | 8.68 (1.26–60.1)* |
| G190A only | 0 | 2 | 1.13 (.01–14.1) | 0 | 1 | 1.63 (.01–30.8) | 0 | 2 | 1.74 (.01–22.6) |
| Other single NNRTI mutation | 0 | 7 | 0.38 (.00–3.16) | 2 | 11 | 1.06 (.20–3.69) | 1 | 4 | 2.89 (.28–17.0) |
| ≥2 NNRTI mutations | 1 | 3 | 2.42 (.23–15.0) | 1 | 5 | 1.33 (.14–6.80) | 0 | 0 | … |
| ≥1 NRTI and ≥1 NNRTI mutation | 2 | 1 | 9.43 (1.24–104)* | 4 | 1 | 14.6 (2.66–148)** | 0 | 0 | … |
Abbreviations: 3TC, lamivudine; ART, antiretroviral therapy; CI, confidence interval; EFV, efavirenz; FTC, emtricitabine; HBsAg, hepatitis B virus surface antigen; HIV, human immunodeficiency virus; IQR, interquartile range; LPV/r, ritonavir-boosted lopinavir; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleos(t)ide reverse transcriptase inhibitor; OR, odds ratio; TDF, tenofovir disoproxil fumarate; ZDV, zidovudine.
aTwo separate models were generated to examine the effects of genotype; one to compare any drug resistance detected vs wild-type and the other to compare each mutational category (n = 7) to wild-type.
bSamples without genotype (n = 10) did not amplify.
* P < .05, **P < .01, ***P < .001 by unadjusted logistic regression with Firth bias reduction.