Background:  In prior data, the prevalence of transmitted drug resistance in recently HIV-infected persons in the Options Project cohort in San Francisco declined from an annual peak of 19 to 27% of patients in 2000 to 2002 to 10 to 11% in 2003 to 2004. We sought to assess changes in the prevalence of transmitted drug resistance in San Francisco since 2003.

Methods:  All patients enrolled in the Options Project, a longitudinal cohort study in San Francisco of acute/early HIV (<12 months), underwent genotype analysis at or soon after entry into the cohort. Patients were assessed for the presence on an initial genotype of at least one drug-resistance mutation, using published guidelines, which exclude common polymorphisms. Genotypes were excluded if performed >10 days after initiation of ART. The prevalence of transmitted drug-resistance mutations, as well as the prevalence of these mutations by class (NRTI, NNRTI, PI) was assessed in each year from 2003 to 2007. A c² test for linear trend was used to assess whether prevalence changed significantly over time.

Results:  Of a total of 224 patients who entered the cohort and underwent initial genotyping, 36 (16%) had evidence of transmitted drug resistance mutations. There was a statistically significant increase in the prevalence of transmitted drug resistance mutations over the study period (p = 0.01). There was a statistically significant increase in the prevalence of NNRTI mutations (p = 0.04). Changes over time in the prevalence in NRTI mutations were not significant (p = 0.2), nor were changes in PI mutations (p = 0.9).

Conclusions:  We observed a fairly steady and statistically significant increase in the prevalence of transmitted drug resistance in San Francisco from a nadir in 2003. This was driven most strongly by increases in NNRTI resistance, while PI resistance was steady or declined. This increase may be related to the increasing use of NNRTI and subsequent development of resistance in persons receiving treatment.