U.S. treatment guidelines advise against the use of genotypic resistance tests when a patient’s viral load is below 1,000. But if a patient has persistent, low-level viremia and emerging resistance is a concern, can a genotype still be useful?

A small study presented at this conference suggests that the answer is yes: In the case of the Trugene genotype, at least, a low viral load does not compromise the test’s accuracy.

Genotypic resistance testing is an important tool in the management of patients who experience virologic failure while on antiretroviral therapy. The presence of documented resistance mutations suggests which drugs are no longer likely to be effective, and the absence of other mutations suggests which drugs can be used in subsequent regimens. Resistance testing allows the patient and the clinician to make an informed decision on how best to change antiretroviral therapy.

Unfortunately, in clinical practice, the use of resistance testing is not always this simple. One particularly frustrating obstacle is the issue of how to deal with low-level replication of HIV in the bloodstream. Low-level viremia (50 to 500 copies/mL) can be documented by sensitive viral load testing. Sometimes this low, but detectable, viral load is isolated, becomes undetectable on repeat testing, and is of no clinical significance (i.e., a viral load “blip”). But at other times the low-level viral load is persistent, which raises concerns about the development of resistance.

In the setting of persistent, low-level viremia, the patient and clinician may choose to change antiretroviral therapy, in order to prevent a further increase in viral load and an accompanying decline in CD4+ cell count. But how does a clinician choose a new regimen in this situation? Should he or she change one, two or all of the drugs in the regimen? Ideally, a genotypic resistance test would help answer these questions. However, according to U.S. Department of Health and Human Services guidelines, genotypes should be performed only when a patient’s viral load is 1,000 copies/mL or higher, due to the unreliability of virus amplification.

H.B. Gale and V.L. Kan, of the Veterans Affairs Medical Center in Washington, D.C., sought to address this clinical conundrum by evaluating a commonly used genotype kit, Trugene, at lower-than-recommended viral loads. They performed a Trugene genotype on a total of 20 blood samples from patients with extensive HIV drug resistance and a viral load greater than 1,000 copies/mL (1,032 to 345,560). Then they took a second round of samples, diluted the blood so that the viral load was approximately 100 copies/mL (<75 to 254), and repeated the genotype. They were not able to perform a genotype on eight of the 20 diluted samples at the first attempt, but they were successful on a second attempt after using ultracentrifugation (i.e., spinning the samples at a high rate of speed) to concentrate the HIV.

Overall, the original and diluted genotypes were remarkably concordant. The interpretation for individual drugs (stratified by whether there was no evidence of resistance, possible resistance, resistance or insufficient evidence) was the same in 97.1% of the paired comparisons (96% for nucleoside/tide reverse transcriptase inhibitors, 95% for non-nucleoside reverse transcriptase inhibitors, and 99% for protease inhibitors). This was not significantly different than the concordance of a repeat genotype on the original, undiluted sample; in those cases, the researchers found that 97.6% of interpretations were the same.

Although the study sample was small, these results suggest that genotypes can be attempted at low viral loads (100 to 500 copies/mL). Given that genotypes are usually performed at a different laboratory than viral load tests, this study also suggests that clinicians can use the results of a genotype for clinical management, even if a concurrent viral load is low.

It is important to note, however, that eight of the 20 diluted blood samples required centrifugation prior to obtaining a genotype result; many clinical laboratories may not perform this step. In addition, many laboratories may be using a genotype other than Trugene, in which case this study would not apply.

Obtaining a genotype at a low viral load may help many patients choose a new drug regimen when resistance is still limited and a switch can help preserve future treatment options. It would be nice to see larger studies to confirm these results, and to test whether this improved laboratory tool translates into better clinical results for patients experiencing low-level virologic failure.