The clinical and economic impact of genotypic resistance testing for people diagnosed with persistent virological non-suppression on tenofovir-lamivudine-dolutegravir in South Africa: a modelling study

Abstract

Background: Persistent virological non-suppression among people with HIV receiving tenofovir-lamivudine-dolutegravir (TLD) can result from poor adherence with or without resistance; however, genotypic resistance testing (GRT) is not recommended routinely in South Africa. We examined the clinical and economic effect of GRT for all South African adults diagnosed with persistent virological non-suppression on TLD.

Methods: In this modelling study, we used the previously validated Cost-Effectiveness of Preventing AIDS Complications-International microsimulation model to compare three strategies: (1) continued TLD (baseline); (2) immediate switch to tenofovir-lamivudine plus ritonavir-boosted darunavir; and (3) GRT prompting switch to tenofovir-lamivudine plus ritonavir-boosted darunavir for people with dolutegravir resistance or TLD continuation for people without dolutegravir resistance. We estimated that 2·3% and 28·5% of the baseline population have dolutegravir resistance and nucleoside reverse transcriptase inhibitor (NRTI) resistance, respectively. We also examined the effect of a low-cost, point-of-care urine tenofovir test in development to detect recent antiretroviral therapy use (84% sensitivity and 50% specificity), with GRT only when positive. Costs included GRT (US$157 per test), TLD ($45 per year), tenofovir-lamivudine plus ritonavir-boosted darunavir ($247 per year), and urine tenofovir testing ($2 per test). Outcomes included life-years, costs (provider perspective), and incremental cost-effectiveness ratios (ICERs; $ per disability-adjusted life-year [DALY]). We considered cost-effectiveness thresholds of less than $3310 per DALY (base case) and less than $1100 to $4250 per DALY.

Findings: Based on our model, we estimated that continued TLD results in 14·11 undiscounted life-years and costs $5380 discounted at 3%; GRT results in 14·36 life-years and costs $5860 (0·14 discounted DALYs averted; ICER $3500 per DALY). Immediate switch results in fewer DALYs averted and higher costs. GRT has an ICER of $3310 per DALY or less when baseline dolutegravir resistance prevalence is ≥2·5% or genotypic resistance test costs ≤$147 per test. Urine tenofovir testing to identify GRT eligibility results in an ICER of $2300 per DALY; the ICER would be less than $1100 per DALY if urine test specificity is 0·87 or greater and costs $2 per test or test specificity is higher than 0·98 and costs $10 per test or less.

Interpretation: GRT could increase life expectancy for people with HIV and persistent virological non-suppression on TLD in South Africa and could be cost-effective, especially at lower test costs. At current effectiveness and costs of tenofovir-lamivudine plus ritonavir-boosted darunavir, an immediate switch would not be preferred.

Funding: National Institute of Allergy and Infectious Diseases, the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and the MGH Jerome and Celia Reich Endowed Scholar in HIV/AIDS Research Award.