Successful use of genotypic resistance testing in HIV-1-infected individuals with detectable viraemia between 50 and 1000 copies/ml

Abstract

Traditionally an HIV viral load of over 1000 copies/ml was presumed necessary for genotypic resistance testing. We performed a retrospective analysis to assess rates of viral amplification on viral loads between 50 and 1000 copies/ml. Amplification was significantly more likely for 200–1000 copies/ml than 50–200 copies/ml; most samples amplified successfully, supporting the use of genotyping for low-level viraemia.

In the era of HAART viral load monitoring is an integral part of therapeutic monitoring in developed countries. Virological suppression below 20–50 copies/ml is associated with a more durable virological response than suppression to 50–400 copies/ml ^[1–3].

The early diagnosis of virological failure is essential, because an important cause is the development of genotypic mutations in the reverse transcriptase or protease genes that confer resistance to antiretroviral agents. Genotypic testing is used to detect which mutations have developed and select the subsequent antiviral regimen accordingly. Historically, a lower cut-off of 1000 copies/ml had been used to determine when a genotype can be performed; this was based on physician experience and anecdotal evidence.

At our centre resistance tests are routinely reviewed in a weekly ‘virtual clinic’, and it was noticed that a number of tests had been requested on samples with a viral load below 1000 copies/ml, but the sample had been successfully amplified for genotyping. We therefore reviewed all resistance assays performed in samples with an HIV viral load of less than 1000 copies/ml.

We have a large, prospectively collected database including demographic data, treatment history, clinical events and the results of laboratory investigations, including viral load assays and resistance tests. Using this database, we identified all individuals between July 2001 and July 2003 with at least two consecutive viral loads above 50 but less than 1000 copies/ml who underwent genotyping in the same period. For the resistance assays, HIV-1 protease–reverse transcriptase sequences were generated on an ABI 3730XL sequencer using Virco BVBA (Mechelen, Belgium) proprietary primer mixers. Viral loads at the time of resistance testing were stratified into ranges and chi-squared testing was used to analyse the differences in rates of successful genotyping within those ranges.

A total of 112 genotype tests were attempted on individuals with two consecutive viral loads in the 50–1000 copies/ml range. The likelihood of the successful amplification of genetic material and sequencing is demonstrated in Table 1. Overall, 73% of samples were clade B and 27% were non-B.

Table 1: Likelihood of the successful amplification of genetic material and sequencing.

The likelihood of successful genotyping was significantly greater for a viral load between 200 and 1000 copies/ml compared with 50 and 200 copies/ml (P = 0.009).

In summary, we demonstrate a significant chance of successful genotypic analysis on samples with an HIV viral load less than 1000 copies/ml. In particular, for samples with a viral load between 200 and 1000 copies/ml there was a greater than 90% chance of successful amplification. Even for samples in the 50–200 copies/ml range there was an almost 70% chance of successful genotyping. We recommend genotypic testing on all individuals with a viral load in excess of 200 copies/ml and also for individuals with persistent viraemia below 200 copies/ml.