In the spring of 1993 Michael T. presented to a local emergency room with oral thrush and marked lymphadenopathy -- classical markers of advanced HIV disease in those days. He was assigned a dual-nucleoside regimen of ZDV and ddC, and he was started on TMP-SMX as prophylaxis against pneumocystis pneumonia. At the time, his CD4 count was 432 cells/mm3. Over the next three years it dropped to 121 cells/mm3, and in March of 1996, on his first visit to our institution, Michael T. was given a diagnosis of frank AIDS. We switched him to the three-drug combination of d4T, 3TC, and the original formulation of saquinavir (Table). At the time, saquinavir was the only protease inhibitor that had received F.D.A. approval, and the effectiveness of 3TC in reversing resistance to ZDV and d4T had just been discovered.
At the time, this combination of agents was the most potent available to us, and it did reduce Michael T.s viral load to 504 copies/mL. However, the patient promptly developed chronic and debilitating diarrhea on this combination of antiretroviral agents. This adverse reaction to therapy was attributed to the saquinavir, and so after a mere three months we substituted a newly available and far more potent protease inhibitor, indinavir, for the saquinavir. At the time of this substitution the patients viral load was 10,028 copies/mL and his CD4 count was 209 cells/mm3.
During the 15 months that Michael T. took d4T, 3TC, and indinavir, his viral load never fell below 3,000 copies/mL. By late November of 1997 the patients viral load, which had risen steadily over the past year, reached almost 12,000 copies mL, and his CD4 count, which had dropped just as steadily, fell to 151 cells/mm3. This evidence of increasing resistance led us to switch our patient to a combination of two familiar drugs and two new ones: d4T, ddI, the soft-gel capsule formulation of saquinavir, and low-dose ritonavir, given to potentiate the bioavailability of the other protease inhibitor.
Over the next year, on the combination of ddI, 3TC, nelfinavir, and hydroxyurea, our patients viral burden dropped precipitously, but it never fell below the level of detection of the ultrasensitive HIV RNA assays that had become available by that time. Indeed, his lowest HIV RNA measurement, during the year he was on this, his fourth antiretroviral regimen, was 6,746 copies/mL. And when his viral load began yet another inexorable climb, we switched him to yet another combination of two familiar drugs and two new ones: the coformulation of ZDV and 3TC plus abacavir and amprenavir, the latter drug at a dose of 1200 mg b.i.d. Although Michael T.s viral load held steady at around 20,000 copies/mL for eight months on this new regimen, it never fell below that level, and his CD4 count dropped to 108 cells/mm3.
In April of 2000 we evaluated Michael T.s case in our salvage clinic. In preparation for that evaluation we assessed his adherence (which remained high), obtained a plasma trough-level of amprenavir, and ran genotype and phenotype assays on our patients viral isolate. The results of the latter assay are reproduced on pages 26 and 27, just as we received them from the reporting laboratory (Figures 1 and 2). They show that Michael T. has exhausted his susceptibility to 3TC and has developed high-level resistance to virtually all of the other agents to which he has been exposed over the past decade.
Interestingly enough, Michael T. remains sensitive to non-nucleoside reverse-transcriptase inhibitors as a class. His only prior exposure to NNRTIs came in 1995, when nevirapine was briefly added to his two-nucleoside regimen -- and he promptly developed a severe rash. Although life-threatening, Stevens-Johnson-like hypersensitivity reactions to nevirapine are seen in less than 1% of patients, rash occurs in up to 33% of individuals who are started on this NNRTI. To reduce the likelihood that patients will develop rash on nevirapine, therapy is initiated at 200 mg daily for two weeks and then increased to the therapeutic dose of 200 mg twice daily. In patients who experience only mild erythema and maculopapular eruptions, nevirapine is maintained at 200 mg daily until symptoms resolve, after which the dose is increased. In patients who experience more severe reactions, as Michael T. did, nevirapine is discontinued permanently.
Given what you know of this patient's treatment history -- and given the results of the phenotyping of his much-mutated viral isolate -- what combination of antiretroviral agents would you recommend at this juncture?
The results of Michael T.s phenotype assay indicate that he has developed some degree of resistance to all five of the protease inhibitors that have been prescribed to him as components of the half-dozen antiretroviral regimens he has been assigned over the course of more than a decade. In such a situation, differences in degree of resistance become all-important. Phenotypic assays derive their findings by determining the concentration of a given drug that is needed to inhibit 50% of the HIV replication in a given patients viral isolate -- and then comparing that value, the IC50, with the concentration of the drug that is needed to inhibit 50% of the HIV replication in a wild-type viral isolate that has never been exposed to the drug. This comparison yields the fold-change bar graph seen at the right side of Figure 1.
The assay we used (PhenoSense®, ViroLogic) also gives IC50 values in µM (seen in the third column from the left side in Figure 1), and these values can be used to calculate how much of a particular drug will be needed to overcome a particular patients resistance to that drug. In Michael T.s case, what this comparison of IC50 values reveals is that it will now take concentrations of indinavir, nelfinavir, or saquinavir that are a hundred times greater than the standard clinical doses to achieve a 50% reduction in viral replication in this patient (Figure 2, panels 9-12).
The news is markedly better where amprenavir is concerned, however. Here, the fold change between the patients IC50 and the IC50 of drug-sensitive virus is only 22 (Figure 1, center column, and Figure 2, panel 8). We recognized that we could boost our patients plasma levels of amprenavir well above the IC50 fold-shift of 22. Our conclusion was based on our calculation that it would require 0.127 µg/mL of drug to overcome the resistance in Michael T.s viral isolate. The assay that we used, which is run without protein, reports how much unbound drug is needed to inhibit the patients viral strain. It is therefore necessary to calculate how much protein-bound drug will be needed to inhibit viral reproduction. Population-based studies have established that amprenavir is 10% protein bound, so it will take 1.27 µg/mL of the drug to overcome the resistance exhibited by Michael T.s viral isolate.
Our conclusion is supported by data from a recent clinical trial conducted by Piscitelli et al. They found that combining amprenavir at a dose of 1200 mg b.i.d. with ritonavir at a dose of 200 mg b.i.d. resulted in a mean serum concentration of amprenavir of 2.18 µg/mL, with a range of 0.94 to 3.54 µg/mL.
We therefore switched Michael T. to the four-drug combination of these two protease inhibitors at these doses, plus d4T and ddI at the standard doses. After only two weeks on this regimen, his viral load had fallen from more than 20,000 copies/mL, to 577 copies/mL. We also obtained a plasma trough-level of amprenavir, which was 1.48 µg/mL -- or 1.2-fold above the IC50 for this protease inhibitor. Although Michael T.s HIV RNA levels have risen over the six months that he has been on his latest salvage regimen, they have tended to hover in the range of 4,000 to 8,000 copies/mL.
Andrew D. Luber, Pharm.D., is Executive Director of Pacific Oaks Research Beverly Hills, CA.