Isolated Report of Kaletra Resistance

13th International HIV Drug Resistance Workshop (Tenerife, Canary Islands, Spain, June 8-12, 2004)

Written by David Margolis, M.D. (Dallas VA Medical Center & University of Texas-Southwestern)

Additional contributions by Jules Levin

Primary resistance to lopinavir/ ritonavir (Kaletra) has not been previously reported. Theoretical considerations argue that resistance to this drug is difficult to generate in vivo (in patients).

Steve Deeks (abstr. 70) presented an interesting anecodote showing that under very unusual circumstances genotypic resistance to Kaletra could evolve in vivo. Deeks reported at Tenerife Kaletra resistance developed by one patient showing perhaps that resistance can develop. Initially treated with a standard regimen, resulting in suppression to < 75 copies, this subject was then switched to therapy containing Kaletra. The patient then stopped taking his backgrounds medicines and took Kaletra monotherapy for a year before reappearing in clinic. It appeared that the patient may not have been adherent but this was not substantiated. Viral rebound had occurred (>14,000 copies/ml) with genotype showing an unusual mutation at I47A, with V32I and M46M/I mutations in protease. Phenotypic resistance to lopinavir (FC 38) and low RC (2.9%), with susceptibility to all other PIs was seen (the highest FC for other PIs was 2.7 for indinavir) and hypersusceptibility to saquinavir (Fold Change 0.16).

The I47A mutation has been selected in lab passage experiments in the past, but is seen in only 0.08% of samples in the ViroLogic database after the approval of Kaletra. It appears to be a favored pathway of solo lopinavir resistance, but may be highly disfavored in the setting of combination therapy.

The C-terminal gag sequence contained R429G and S451N near the p7/p1 and p1/p6 cleavage sites, as well as a 3 amino acid insertion before the PTAPP domain. In a database of over 70000 clinical samples with phenotype data, no samples were identified that had isolated lopinavir resistance (defined as lopinavir FC >10 and FC <5 for amprenavir, indinavir, nelfinavir, ritonavir and saquinavir). The genotype database contained 34 unique samples exhibiting the I47A mutation, 26 of which also had V32I; 25 of these had phenotypes available. These samples displayed high-level resistance to lopinavir (median FC 154) and other PIs (median FC 6 to 40), with the exception of saquinavir (median FC 0.7). The prevalence of PI-resistant samples with I47A since 2002 has remained steady at approximately 0.4%.

Deeks concluded: Although primary resistance to lopinavir is difficult to generate in vivo, it is not impossible. The preferred pathway for lopinavir resistance may involve V32I and I47A. The lack of isolated lopinavir resistance in the other I47A-containing samples described here likely reflects the emergence of this mutation when lopinavir was used in PI-experienced patients. Continued phenotypic/genotypic analysis in patients receiving lopinavir as their first-line PI will be needed to confirm these results.