Many different treatment strategies are being explored in patients with a history of resistance to multiple HIV drugs or drug classes. Because most drugs have lost their effectiveness at improving CD4+ cell counts or reducing viral loads in these clinical situations, it would be desirable to recycle drugs that have lost their activity due to resistance. Strategies have included complete or partial treatment interruptions, followed by reintroduction of standard or megaHAART regimens.

It has been previously shown that patients with evidence of multidrug-resistant virus, who undergo a treatment interruption for at least 8 weeks, can show complete or at least partial disappearance of resistant virus in their blood plasma. The resistant virus has not disappeared, however, but has faded into the background at very low or undetectable levels. Controversy exists as to the most beneficial and safest strategy to pursue in this regard, and whether patients can benefit clinically from such strategies.

Stocker et al presented data from a pilot study looking at intermittent switching of HIV medications, based on viral genotype, in patients with multidrug-resistant virus. The 15 patients enrolled in this study were started on a 3-drug regimen consisting entirely of nucleoside reverse transcriptase inhibitors (NRTIs). If a patient's viral load increased by 0.5 log10/mL, a genotypic resistance test was given, the results of which determined whether the patient would be switched to a dual protease inhibitor (PI)-only regimen. If the patient's viral load again increased by 0.5 log10/mL after the switch, the genotype was again tested, the PI regimen was discontinued, and the patient was switched back to a triple-NRTI regimen.

Twelve patients have completed 48 weeks of observation. The baseline median viral load was quite high at 5.8 log10/mL, and the median CD4+ cell count was 130 cells/mm³. After 48 weeks, there was a mean change of -0.22 log10/mL in viral load and -32 cells/mm³ in CD4+ cell count.

Of note, when patients were on the NRTI regimen, previously noted PI mutations (such as V82A) reverted to wild type; when patients were on the PI regimen, some of the NRTI mutations (such as M184V, which is associated with lamivudine [3TC, Epivir] resistance) reverted to wild type. This was not uniform across all patients. A single patient was presented in further detail to show changes at each switching period. This patient experienced significant reductions in viral load (1-2 log10/mL) for a few weeks, followed by rebound. This also correlated with the return of susceptibility to PIs or lamivudine while this patient was on a NRTI or PI regimen, respectively.

The study demonstrates that, at least in this small number of patients, the strategy of switching between single-class regimens is safe, in that no significant reduction in CD4+ cell count was seen. On the other hand, no sustained significant reduction in viral load was seen either. Thus, patients with evidence of multidrug and multiclass resistance can be safely maintained on a single-class regimen for significant periods of time. Some patients apparently experienced greater virologic benefit than others, but whether there were particular mutation patterns or drug-switching combinations that were more beneficial than others was not presented. Non-nucleoside reverse transcriptase inhibitors (NNRTIs), such as nevirapine (NVP, Viramune) or efavirenz (EFV, Sustiva, Stocrin), were not used in this study because it is known that resistance mutations in patients taking these drugs do not disappear from virus in blood plasma. Thus, starting and stopping these medicines would probably not offer any advantage. This study is significant in that patients experiencing pill fatigue while maintaining a complex multidrug, multiclass regimen to which their virus is completely resistant may be offered some relief by safely maintaining a single-class regimen. It still remains to be determined whether this switching strategy (including drugs used and time interval between switching) is clinically beneficial or not.