Background: Enfuvirtide (ENF) is the first fusion inhibitor to undergo Phase III clinical trials. Novel technologies have been developed to assess entry inhibitor susceptibility, co-receptor tropism, and HIV-1 envelope genotype.

Objectives: To describe baseline (BL) ENF susceptibility, co-receptor tropism and genetic variation of entry-inhibitor naïve HIV-1 in patients from TORO 1 & 2.

Methods: Designation of viruses as R5, X4, or dual tropic was based on the ability to generate reportable phenotypic susceptibility results on CD4/CXCR4 and/or CD4/CCR5 cells. Susceptibility was expressed as the fold change in IC₅₀ (FC) compared to an R5 (JRCSF) or X4 (HXB2) reference virus. BL susceptibility to ENF was log normally distributed on both CXCR4 and CCR5 cells. To adjust for differences in susceptibility of the X4 and R5 reference viruses, FC values were normalized (nFC) such that the geometric mean of the results determined on either X4 or R5 cells was set to1. The env gene was sequenced, amino acid sequences of gp41 were compiled, and substitutions relative to HXB2 were reported.

Results: Assay reproducibility (< 3-fold) was defined by trending the susceptibility of reference viruses over time, and repeat testing of primary isolates. Assay accuracy was verified using viruses containing R5 and X4 tropic envelope sequences bearing one or more ENF resistance mutations. 612 BL viruses were evaluated. The percentages of R5, X4, and dual tropic viruses at BL were 62% (n = 378), 4% (n = 23) and 34% (n = 211), respectively, 2% (n = 12) were non-B clade. The distribution of nFC values was broader for viruses that replicated on CXCR4 cells (IQR_X4 = 0.4–2.7) compared to CCR5 cells (IQR_R5 = 0.5–1.8). Pure X4 tropic viruses exhibited slightly higher nFC (GM = 2.7, IQR = 1.7–4.6) compared to pure R5 (GM = 1.2, IQR = 0.6–2.1) and dual tropic viruses (GM_X4 = 0.9, IQR_X4 = 0.3–2.3, GM_R5 = 0.8, IQR_R5 = 0.4–1.5). BL gp41 substitutions in aa 36–45 relative to HXB2 were uncommon (≤ 2%) except at aa 42 (N42S = 15%), which was more prevalent in viruses that were most susceptible to ENF (p < 0.001).

Conclusions: At TORO 1&2 study baseline, pure R5 tropic viruses were more prevalent than dual tropic viruses, and pure X4 tropic viruses were rare. BL ENF susceptibility distributions were generally similar for all virus populations, however pure X4 tropic viruses had slightly higher nFC compared to R5 and dual viruses. While polymorphisms in the ENF binding site were uncommon, N42S was associated with lower nFC to ENF.