Print this page    •   Back to Web version of article

Resistance Assay Update

By Dave Gilden

October 1998

Treatment Issues last June published a description of two "phenotypic" drug resistance assays that were slated to enter the market soon. Such phenotypic resistance assays have certain advantages over assays that confine themselves to analyzing HIV genetic sequences ("genotypic" assays). They culture a form of HIV in the presence of different individual drugs to get a direct measure of these drugs' effectiveness against a person's HIV. Their disadvantage is that they are much slower and more complicated to carry out. Also, as pointed out in the June article, the new phenotypic assays are not really testing patients' HIV but rather the effect of genes extracted from a person's HIV and inserted into a standard carrier virus for testing purposes. These assays are really a hybrid that is one step removed from the traditional culture methods for testing drug resistance, and the new technology they embody requires extensive validation.

One of the new assays, developed by ViroLogic in California, is still undergoing quality control testing. The other, conducted by the Belgian company Virco, entered the U.S. market last summer via LabCorp, the national medical lab chain. LabCorp collects the blood samples and does some of the initial processing before sending them to Belgium. The Virco phenotypic test (the "Antivirogram") costs a hefty $880 to check on 12 anti-HIV drugs, including abacavir and efavirenz, but not delavirdine. Right now, a four-week turnaround time is the norm. Virco and Labcorp also offer a speedier genotypic analysis (the "Vircogen") that lists specific mutations in one's HIV protease and reverse transcriptase genes and attempts to interpret how they affect drug resistance. This complementary genetic sequencing costs an extra $450.

Validating the Assays

In the past few months, a number of reported studies have given some confidence that phenotypic assay results actually reliably predict response to therapy in many cases. The largest body of such reports occurred at the Second International Workshop on HIV Drug Resistance and Treatment Strategies, which took place in Italy at the end of June, as well as the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), held in San Diego on September 24-27.

In one presentation on salvage therapy (Resistance Workshop abstract 53), Nick Hellman of ViroLogic described using his company's drug resistance assay to explain the response of 18 persons to a four-drug combination during a salvage therapy study. Eight volunteers who had failed to respond to an indinavir-containing combination were treated with abacavir/nevirapine/nelfinavir/saquinavir (Fortovase) while another ten received abacavir/a second nucleoside analog/nelfinavir/Fortovase. (The volunteers had never received nevirapine or any other NNRTI.)

Notably, four of the volunteers turned out to have indinavir-sensitive HIV. No one had HIV sensitive to all four drugs in their regimen, but those with HIV still sensitive to two or three of the drugs in the combination experienced a profound reduction in viral load that reached a median of 2.5 logs (99.7%). The volunteers with HIV sensitive to zero or one drug had an immediate viral load drop that reached a median of 1.3 logs (95%) by the second week. This was as good as the first group had achieved at that point, but the second group's viral load then rapidly returned to the baseline value.

In another example of studying the relationship between resistance assay results and treatment outcome, Richard Harrigan, of the British Columbia Centre for Excellence in HIV/AIDS, presented results from a study of the Centre's 84 patients who received combined ritonavir/saquinavir therapy between September 1996 and June 1997 (ICAAC abstract I-78). Inability to respond to combination regimens containing these two protease inhibitors was highly correlated with phenotypic resistance to saquinavir, especially, but also to ritonavir to some extent. No person with resistance to both attained viral loads below 500, the limit of detection with the viral load assay in use. (Resistance determinations were made by Virco.) But the patients whose tests showed sensitivity to both drugs were not always successfully treated. Only about half the patients with prior protease inhibitor therapy and Virco test results indicating sensitivity to both PIs achieved viral loads below 500.

Virco itself has had a similar experience. The company has amassed a database containing over 8,000 phenotypic assay results and 1,000 genotypic ones. It is using these data to create algorithms for predicting phenotypic resistance from genotypic mutational analyses. So far, the genotypic results can predict resistance to multiple protease inhibitors with about 98% accuracy, but about 15% of the HIV found to be without known relevant mutations is nevertheless PI-resistant (Resistance Workshop abstract 51).

Virco is now embarking on a trial with Glaxo Wellcome to evaluate the value of its phenotypic assay in clinical practice. The 268 study participants must have a limited treatment history consisting of at least two nucleoside analogs, just one protease inhibitor and no nonnucleoside reverse transcriptase inhibitors. All will get prior phenotypic testing and will then be randomized into two groups. One group will receive treatment on the basis of the Virco results while the other will not be informed of the Virco findings and will be prescribed antiviral medication on the basis of their physicians' standard practice. At 16 weeks, the two groups will be compared in regards to changes in viral load, CD4 counts and drug resistance as determined by genotypic and phenotypic test results. The California Collaborative Trials Group is embarking on a similar trial that will last one year instead of 16 weeks.

Failure Despite Lack of Evident Resistance

Our knowledge of how genetic mutations affect HIV drug resistance is still rather sketchy, and for that reason the correlation between negative genotype findings and lack of phenotypic resistance is not as tight as one would hope. One report at ICAAC, from the Visible Genetics company in Toronto, described how specific seemingly innocuous mutations consistently appear before given major resistance mutations and serve as an early warning of coming drug resistance (ICAAC abstract I-76).

Mutations are not the whole story in regards to drug resistance, though, and the totality of factors that contribute to a treatment's failure to suppress HIV in a given individual remains elusive. Several recent drug trials have made some surprising observations concerning the relationship between phenotypic resistance and treatment outcome.

One of these was ACTG 343, a maintenance therapy study in which volunteers who had reached viral loads below 200 after six months on AZT/3TC/indinavir were switched to either AZT/3TC or indinavir monotherapy and compared to a cohort that stayed on the triple combination. The reduced maintenance therapies were clearly inferior almost from the beginning. After two months on the new regimens, 23% in both the indinavir and AZT/3TC arms had experienced viral rebound whereas this happened to only 4% of those on AZT/3TC/indinavir.

These results were not surprising, and in fact they were anticipated by the numerous people who criticized this trial at its onset. What was surprising was the outcome of the resistance assays, which employed the ViroLogic test as well as a gene-sequencing technique. Ten volunteers did not achieve stable viral load suppression while on triple therapy. For nine of these, the only HIV resistance detected was to 3TC (accompanied by the standard mutation at codon 184 of the reverse transcriptase gene). Loss of susceptibility to indinavir was never observed. As for the nine volunteers who experienced viral rebound on the indinavir-only maintenance therapy, none showed phenotypic or genotypic resistance to indinavir or the other two drugs in the induction regimen. (Those rebounding on AZT/3TC maintenance were largely the ones whose HIV had AZT-resistance mutations prior to the trial's commencement.)

These findings were described during an ICAAC lecture (session 63-I, see also Resistance Workshop abstract 74) by Douglas Richman, of the University of California San Diego. Dr. Richman explained the rebounds without broad resistance to the therapeutic combination by invoking a "more food for the virus theory" that has increased in popularity since the Resistance Workshop at Lake Maggiore, Italy. At this workshop, Angela McLean, of the Institute for Animal Health in Berkshire, UK, described the latest version of the mathematical models that support this theory. The models indicate that increases in CD4 cell populations arising during successful antiviral therapy are not completely good. They also perversely increase the likelihood of viral rebound even in the absence of obvious drug resistance (abstract 1).

In ACTG 343, the chance of HIV rebound was in fact correlated with the increase in CD4 cells a person experienced during the initial induction phase (the risk of rebound increased 40% for every 100 cell/mm3 rise in CD4 count). As Dr. Richman interpreted the ACTG 343 observations of rebound while on triple therapy, CD4 counts first increased in the study participants, but the emergence of 3TC resistance opens up more targets for escape HIV. The resistance to 3TC may have arisen accidentally prior to therapy in a small HIV subpopulation or might easily arise during the first few weeks of therapy, when residual HIV replication is always substantial. HIV gains resistance to 3TC randomly as well as rapidly after exposure to the drug because it only depends on the single point mutation at codon 184 of the reverse transcriptase gene. In either case, the result could be that the escape HIV has the opportunity to proliferate and eventually develop resistance to the other antiviral agents in a combination regimen if new drugs are not added in time to resuppress the virus.

(Note that a just-published French maintenance therapy trial, the Trilege study, did not find an association between CD4 count increase and viral rebound, although other results were similar to ACTG 343.)

The inability of resistance testing to predict such evolving resistance could be taken as a weakness. Conversely, it is definitely a strength that testing at the right time point, immediately after viral rebound begins, can show that only some drugs in a combination are problematic. The tests can provide an early warning, allowing only the problem drugs to be replaced so that the others' activity can be preserved as HIV replication once again is stanched.

Without a precise understanding of what is happening, a doctor would have to grope to find a successful salvage treatment strategy. One way would be to replace all the drugs with a completely new combination, but this might be unnecessary and runs the risk of further promoting multidrug resistant HIV. Another tactic would be to blindly add extra drugs to make a more complicated, intensified regimen. But patients might find this intolerable over the long run. Finally, a person could stay on a combination regimen until it fails completely, at which point that individual would have needlessly lost the ability to benefit from any of the antiviral agents he or she was taking.

HIV experts nonetheless remain unsure whether these assays really help to provide longer survival and better quality of life than current practice. At a special ICAAC "interactive" session the audience at one point had the opportunity to vote on which assay they would use to help find a salvage regimen for a patient with rebounding viral load. Thirty-five percent chose genotypic testing, 30% chose phenotypic testing and 29% favored the third proffered option -- a Ouija board.

Back to GMHC Treatment Issues October 1998 contents page.

This article was provided by Gay Men's Health Crisis. It is a part of the publication GMHC Treatment Issues. You can find this article online by typing this address into your Web browser:

General Disclaimer: is designed for educational purposes only and is not engaged in rendering medical advice or professional services. The information provided through should not be used for diagnosing or treating a health problem or a disease. It is not a substitute for professional care. If you have or suspect you may have a health problem, consult your health care provider.