Application of coamplification at lower denaturation temperature-PCR sequencing for early detection of antiviral drug resistance mutations of hepatitis B virus

Abstract

Nucleoside/nucleotide analogue for the treatment of chronic hepatitis B virus (HBV) infection is hampered by the emergence of drug resistance mutations. Conventional PCR-sequencing cannot detect minor variants of <20%. We developed a modified CO-amplification at Lower Denaturation temperature-PCR (COLD-PCR) method for the detection of HBV minority drug resistance mutations. The critical denaturation temperature for COLD-PCR was determined to be 78°C. Sensitivity of COLD-PCR sequencing was determined using serially-diluted plasmids containing mixed proportions of HBV reverse transcriptase (rt) wild-type and mutant sequences. Conventional PCR-sequencing detected mutations only if they existed in ≥25%, whereas COLD-PCR sequencing detected mutations when they existed in 5-10% of the viral population. The performance of COLD-PCR was compared to conventional PCR-sequencing and a line probe (LiPA) assay, using 215 samples obtained from 136 lamivudine- or telbivudine-treated patients with virological breakthrough. Among these 215 samples, drug resistance mutations were detected in 155 (72 %), 148 (69 %) and 113 samples (53 %) by LiPA, COLD-PCR, and conventional PCR-sequencing, respectively. Nineteen (9 %) samples had mutations detectable by COLD-PCR but not LiPA, while 26 (12 %) samples had mutations detectable by LiPA but not COLD-PCR, indicating both methods were comparable (P = 0.371). COLD-PCR was more sensitive than conventional PCR-sequencing: 35 (16 %) samples had mutations detectable by COLD-PCR but not conventional PCR-sequencing, while none had mutations detected by conventional PCR-sequencing but not COLD-PCR (P < 0.0001). COLD-PCR sequencing is a simple method which is comparable to LiPA and superior to conventional PCR-sequencing in detecting minor lamivudine/telbivudine resistance mutations.