Background:  Increasing protease inhibitor (PI) exposure in patients with PI-resistant HIV using normalized inhibitor quotient (NIQ) -based therapeutic drug monitoring (TDM) may improve outcome. This is the final primary analysis of a randomized trial evaluating this strategy.

Methods:  Adults with ≥1 failed PI and viral load ≥1000 copies/mL on ARV began a new FDA-approved PI regimen at entry. A PI NIQ was generated at week 2 (patients IQ = week 2 PI trough/screening IC₅₀ fold-change (vircoTYPE, a virtual phenotype); NIQ = patients IQ/IQ of a reference population with viral load response to that PI). Patients with NIQ ≤1 were hypothesized to benefit from PI dose escalation, and randomized 1:1 at week 4 to receive standard of care or PI dose escalation (TDM arm). All patients had PI troughs drawn 2 and 4 weeks later; only TDM patients received NIQ and PI dose increase for persistent NIQ ≤1. The primary comparison was the difference between TDM and standard of care in the log₁₀ viral load change between randomization and 20 weeks later. Of the total, 180 patients gave 80% power to detect a 0.6 log₁₀ difference in the primary endpoint, if SD = 1.3.

Results:  We randomized 92 patients to TDM, and 91 to standard of care of whom 90% were men, 49% white non-Hispanic, median age 45 years, and median entry viral load, CD4, and number of active PI (based on vircoTYPE susceptibility score) were 4.36 log₁₀ copies/mL, 194/µL, and 0.7, respectively. Median trough concentration increased more in TDM vs standard of care for all PI (p = 0.01 to 0.05); except fos-amprenavir (fos-APV) (p = 0.13), taken by 32% of patients. NIQ increased more in TDM than standard of care (median +69% vs +25%, p = 0.01; +87% vs +25%, p = 0.006, excluding patients on fos-APV). Overall, TDM and standard of care did not differ in the primary endpoint (intent to treat, +0.09 vs +0.02 log₁₀ copies/mL; Gehan-Wilcoxon p = 0.17), other outcome measures, or toxicities. In subgroup analyses, patients on ≥0.7 active PI benefited from TDM (p = 0.002); those on <0.7 did not (p = 0.35, test for interaction p = 0.003). Hispanic and black patients (49% of subjects) benefited more from TDM (p = 0.035, 0.05, and 0.40 for Hispanic, black, and white patients, respectively; test for interaction p = 0.04). The 3 racial/ethnic groups had similar body mass index; number of active PI; PI trough concentration; change in NIQ; fos-APV use; and baseline adherence, viral load and CD4.

Conclusions:  There was no overall benefit of TDM in this study. The inability to increase APV concentration with dose escalation in patients on fos-APV is not understood, but may have contributed to the lack of a detectable TDM effect. TDM may confer more benefit in black and Hispanic patients than white patients, for unclear reasons. Subgroup analyses suggest a TDM effect was obscured by the inclusion of patients with highly PI-resistant HIV.