Background: There is conflicting evidence on specific reverse transcriptase inhibitors to which the N348I mutation in the connection domain of HIV type-1 reverse transcriptase confers resistance. Here, we examined associations between the emergence of N348I and antiretroviral history in a large clinical database.
Methods: We analysed 5,353 resistance tests (that were sequenced beyond codon 348) among 2,266 antiretroviral-experienced patients. Associations between N348I and individual antiretroviral drug exposure were estimated using a matched case-control approach. Cases were defined as the first resistance test where N348I was detected; for each case, the 10 closest (in calendar time) N348N tests were selected as controls. Odds ratios (ORs) adjusted for effects of all other drugs were estimated by conditional logistic regression.
Results: N348I was detected in 198 (8.7%) cases. Drugs that were statistically significantly positively associated with N348I were efavirenz (OR 1.55, 95% confidence interval [CI] 1.08-2.23; P=0.017) and nevirapine (OR 2.06, 95% CI 1.49-2.85; P<0.001). Tenofovir disoproxil fumarate (TDF) was significantly negatively associated (OR 0.27, 95% CI 0.15-0.48; P<0.001) with N348I. Similar findings were observed when the analysis was repeated to include only those tests within 2 years of the resistance test. Effects for zidovudine and stavudine were evident only in an additional analysis, which considered exposure to both drugs jointly within 2 years prior to the resistance test: exposure to zidovudine alone (OR 4.61, 95% CI 1.83-11.61; P<0.001) and exposure to stavudine alone (OR 3.39, 95% CI 1.32-8.71; P=0.011).
Conclusions: This is the first clinical evidence to suggest that efavirenz might select for N348I in addition to nevirapine, that stavudine might select for N348I in addition to zidovudine and that TDF might protect against the mutation.