Highly active multi-drug antiretroviral regimens have reversed some of the morbidity and mortality rates from human immunodeficiency virus (HIV) infection. However, the inevitable result of this greater drug use is increased rates of drug-resistant virus. Little and associates put specific numbers on the rising resistance levels by comparing blood samples from patients infected with HIV between 1995 and 1998 with blood from patients more recently infected, in 1999 and 2000.

The patients were enrolled from seven Acute Infection and Early Disease Research Program sites covering 10 different North American cities. New HIV infection was defined as documented HIV seroconversion within the past 12 months or laboratory evidence of acute HIV infection. No patient was eligible if he or she had received more than seven days of antiretroviral treatment. A total of 377 eligible patients were analyzed between May 1995 and June 2000. The majority of patients were white men whose risk factor for HIV infection was having sex with men.

Blood samples were screened for drug resistance in two ways. The phenotypic assay quantified the concentration of various anti-retroviral agents required to inhibit growth of a patient's virus in the laboratory. High-level resistance was defined as an inhibitory drug concentration for the patient's virus that was at least 10 times higher than a drug-sensitive reference standard. The trial also included a genotypic analysis of resistance that checked for mutations in the HIV genome known to confer drug resistance.

The rate of phenotypic high-level drug resistance was 3.4 percent in the older cohort of patients (1995 to 1998) and 12.4 percent in the newer cohort (1999 and 2000). The prevalence of multi–drug-resistant HIV increased from 1.1 percent to 6.2 percent, respectively.

Unlike those in the older group, the majority of multi–drug-resistant patients in the newer group had high-level resistance to one or more agents in all three antiretroviral drug classes (nucleoside reverse-transcriptase inhibitors, non-nucleoside reverse-transcriptase inhibitors, and protease inhibitors). The genotypic assay showed that drug resistance mutations increased from 8.0 percent in the older cohort to 22.7 percent among newer patients. Genotypic evidence of resistance to multiple classes of drugs rose from 3.8 percent to 10.2 percent.

The authors also provided some information on the clinical toll associated with this increased drug resistance. Their data showed that, among patients with high-level resistance, the time to initial virus suppression with drug therapy was prolonged and the length of time to eventual failure of the anti-retroviral regimen was shorter.

The authors advocated for incorporation of drug resistance testing, even into the initial treatment of HIV infection, because the prevalence of highly resistant HIV among newly infected patients has increased.

editor's note: Hirsch writes in an accompanying editorial that the cost of drug-resistance testing is in line with other commonly accepted HIV treatments (e.g., prophylaxis of opportunistic infections). The case for employing drug resistance testing at the time of initial treatment may be even more compelling if a physician knows that the prevalence of drug-resistant HIV in the community is relatively high.—B.Z.