Genotypic resistance test has been recommended to evaluate HIV drug resistance and guide the effective
regimens of antiretroviral therapy (ART) in HIV-infected patients with treatment failure. In patients with multiple
treatment failures, drug resistance-associated mutations may disappear due to the loss of selective drug pressure after
switching regimens. A cohort study was conducted among HIV-infected patients who had ≥2 genotypic resistance tests
during 2003-2011. HIV-1 pol nucleotide sequencing of reverse transcriptase and protease region was carried out using
TRUGENE HIV-1 Genotypic Assay. Sequencing data was analyzed using Stanford rule-based interpretation algorithms.
Of 54 patients with mean age of 30.1 years, 46.3% were males. HIV-1 subtype A/E was observed in 88.9% of patients. At
the latest failure, 55.3% were receiving protease inhibitor-based regimens. Median CD4 and HIV RNA were 167
cells/mm³ and 22,359 copies/mL. During a median duration of ART of 38.6 months, 72.2%, 22.2%, and 5.6% had 5, 3,
and 2 genotype tests, respectively. When compared between using cumulative (CG) and last genotypes (LG), CG
interpreted resistance to any drug 59.3% higher than LG did. For NRTI, NNRTI, and PI drug classes, CG interpreted as
resistance 42.6%, 27.8%, and 7.4% higher than LG, respectively. The most common drugs that CG interpreted resistance
with the higher rate than LG were lamivudine/emtricitabine, nevirapine, efavirenz, etravirine and abacavir. In conclusion,
CG interprets HIV drug resistance at a higher rate than LG and may be more accurate to use for selecting the next
effective regimen of ART among HIV-infected patients with multiple treatment failures.