icon-folder.gif   Conference Reports for NATAP  
  9th Conference on Retroviruses and Opportunistic Infections
Seattle, Washington, February, 2002
Back grey_arrow_rt.gif
Genotypic Mutations to Predict Kaletra Viral Response
  Reported by Jules Levin, including commentary excerpted from "HIV Resistance" report from Retrovirus 2002 written for NATAP by Andrew Zolopa, MD, Stanford University. Link to full report written by Dr. Zolopa:
Three posters reported at Retrovirus Conference 2002 on using genotypic mutations to predict response to Kaletra.
Virologic response to therapy with Kaletra (lopinavir/ritonavir), with reverse transcriptase (RT) inhibitors, has previously been shown to be associated with baseline phenotypic and genotypic susceptibility to lopinavir (LPV). Although genotypic resistance to Kaletra has not been completely defined, 11 mutations in HIV protease have been found to be associated with significant changes in the in vitro susceptibility to LPV. The total number of these 11 mutations constitutes the LPV mutation score. Kempf from Abbott previously reported in Study M98-957, virologic response to Kaletra, efavirenz, and nucleoside RT inhibitors in multiple PI-experienced, non-nucleoside RT inhibitor-naive patients was highest in those patients with a baseline LPV mutation score of 5 or less. Intermediate response was observed in patients with a baseline mutation score of 6-7, and the lowest response was observed in those patients with a baseline mutation score of 8 or more.
In poster 559-T at Retrovirus 2002, Kempf and the ANRS HIV French Resistance Group reported new genotypic mutation information in predicting response to Kaletra from a large patient group of 800. The old Kaletra mutation score was based on these 11 mutations: 10, 20, 24, 46, 53, 63, 71, 82, 84, and 90. Kempf reports below that a new group of 10 mutations not containing the L90M better predicts the response to Kaletra. They found that mutations at codons at 33, 36, 47 and 48 contributed to a diminished response to LPV/r. A Canadian research group reported at Retrovirus 2002 that the L90M was predictive in a smaller group of 167 patients in their Expanded Access Program.
Kempf reported that although the LPV mutation score is a useful predictor of virologic response to Kaletra, the initial Kempf analysis of response with respect to genotype has not been adequately characterized. Furthermore, the LPV mutation score was based on an analysis of a limited number (112) of viral isolates from PI-experienced patients. Mutations that were underrepresented in that panel of isolates may not be included in the LPV mutation score, despite the possibility that some may contribute to significantly reduced susceptibility to LPV. In order to further characterize genotypic resistance to Kaletra, and to quantitatively assess the effect of individual mutations within the mutation score on virologic response, we analyzed data from the Kaletra ATU ("Authorisation Temporaire dčUtilisation", Provisional Authorization of Use) program conducted in France. The Kaletra ATU database is an observational database of 792 antiretroviral-experienced patients who initiated combination therapy including Kaletra, and who had genotype available prior to (median: 105 days) the change to Kaletra therapy. The availability of a large number of patients in this database allowed us to directly assess the influence of baseline genotype on virologic response. Furthermore, the influence of individual mutations included in (or excluded from) the LPV mutation score was studied.
Virologic response was defined as achieving a viral load <400 copies/mL within 12 months after initiation of Kaletra. Mutations at positions 10, 20, 54, and 82 were found to statistically significantly influence virologic response in the context of the remainder of the LPV mutation score. Although not statistically significant, mutations at positions 24 and 84 had an odds ratio of 0.8 or less in the logistic regression analysis. These mutations also appear to influence virologic response in the context of LPV mutation score. Mutations at positions 46, 63 and 90 have a smaller effect on the rate of virologic response.
In this study the ANRS Group and Abbott found PI mutation patterns that predict failure to enhanced PI regimens such as Kaletra are complex, and complicated algorithms and large study populations are necessary to quantify the effect of baseline genotype on virologic response. They reported a new set of 10 mutations (positions 10, 20, 24, 33, 36, 47, 48, 54, 82 and 84) that predicted virologic response better than the current LPV mutation score.
Previous in vitro studies have suggested that the I50V mutation, in the context of other mutations, confers significantly reduced susceptibility to LPV. The prevalence of the I50V mutation in this study population was very low (<1%). However, none of the 5 patients whose genotype displayed the I50V mutation prior to initiation of LPV/ritonavir therapy experienced virologic response. One of the 5 patients had only 3 additional mutations along with the I50V (82, 36, 10). In this study, the prevalence of some mutations, particularly the I50V mutation, was too low for assessment of its effect on virologic response.
In poster 560, Loufty and a Canadian research group found the L90M mutation associated with viral failure: 26% (16/60) with the L90M mutation had a viral response while 52% (22/42) of patients who did not have the L90M had a viral response. This was a look at 167 treatment-experienced patients receiving Kaletra in the Canadian Expanded Access Program at 4 Toronto sites. Patients were often NNRTI-experienced. They used the old list of 11 genotypic mutations. Before receiving Kaletra the average LPV mutation score was 4.8 mutations. 85% had a mutation score of 0-5, 13% had a score of 6-7, and 1.2% (2 patients) had a score of >7 mutations. By month 4-6, 40% with a score of 0-5 had <50 copies/ml, 25% with a score of 6-7 had <50 copies/ml, and 0% with a score >7 achieved <50 copies/ml. Loufty reported 37% had <50 copies/ml at month 4-6, and a mutational score >3 was predictive of a lesser response.
Neil Parkin from Virologic and colleagues showed in poster 581 that additional protease mutations (including those listed above) were significantly associated with LPV/r phenotype. One striking observation is that 25/26 isolates with 50V mutation had a LPV/r fold-change >10. Again supporting the potential of cross resistance between amprenavir and LPV/r.
Clearly, interpreting resistance patterns is becoming increasingly complex and requires an expertise among HIV treaters.