Treatment failure and tropism changes in maraviroc trial related to previously undetected CXCR4, rather than a mutational shift from CCR5

3 July 2007. Related: Conference reports, Drug resistance, Intl Drug Resistance Workshop 16 Barbados 2007.

Simon Collins, HIV i-Base

Understanding the mechanism behind treatment failure with CCR5 inhibitors is as important as the new data provided on integrase: maraviroc is also already available in an expanded access programme and both maraviroc and vicriviroc are in Phase III studies.

Understanding CCR5 inhibitor resistance is still unclear though, as are the implications for receptor tropism changes from CCR5 to CXCR4. The tropism question is complicated by limitations of current assays to detect minority X4 virus. The clinical significance for potentially changing the natural course of infection – X4 is associated with later stage infection – has led to arguably tougher regulation than for developers of drugs in existing classes – and rightly so according to many of the debates at the meeting.

Marilyn Lewis from Pfizer presented results from a phylogenic sub-study using env clones sequenced from 20 patients from the Phase 3 MOTIVATE 1&2 studies, in whom X4 virus was detected on-study (using the Monogram Trofile test).

Four of these patients were in the placebo arm and 16 were receiving maraviroc.

In 14 patients (11 maraviroc, 3 placebo), the X4 virus that emerged during the study was identical to X4 detected in baseline sample: 4 patients with >10% X4 populations were identified as dual/mixed tropism at baseline and ten patients were found to have had X4 at 1-6%, below the 10% threshold for detecting X4 at baseline.

X4 samples from the remaining six patients (5 maraviroc, 1 palcebo) were too phylogenetically different from both baseline and on-treatment R5 tropic virus (with 7-15 differences in the 35-amino acid V3 loop alone) to have evolved from the R5 virus.

The authors concluded that this showed that maraviroc did not cause viral evolution from R5 to X4 virus and that X4 virus emerging as a response to CCR5 inhibition was entirely from pre-existing X4 populations, generally below the limit of detection for the Trofile tropism assay.

The limited clinical data from people who subsequently stop maraviroc is that majority (detectable) tropism returns to R5 virus. This increases the importance of understanding the sensitivity of the current tropism test and natural history of tropism changes, especially given that 4/20 patients with detectable X4 virus were in the placebo arm and that 5 patients classified as R5 when screened for these studies were later found to have dual/mixed tropism at study entry (day 1).

50% of patients screened for the MOTIVATE studies were not eligible for entry due to dual/mixed or X4 virus at screening.

Testing for viral tropism

Currently the Monogram Trofile tropism assay has been used in CCR5 registrational trials and is recognised as the gold standard, but this is expensive and requires sample to be shipped to the US. Several other labs are looking at whether genotype-based systems are able to provide similar results for use in clinical practice.

In the MOTIVATE sub-study of tropism changes reported above, there was a 90% concordance between tropism for individual clones assignment using the Trofile test and Position Specific Mutation Scoring (PSMS) algorithm (see Table 1).

However, Marylin Lewis and colleagues from Pfizer reported that the sensitivity of both the 11/25 rule and PSMS to detect X4 usage was lower than for detecting R5 tropism. (see Table 1) [1]

Although several other tropism assays are in development (see reports from the European resistance workshop in the May issue of HTB [2, 3]), none of these tests are commercially available yet.

Table 1: Concordance between tropism assays in MOTIVATE study

References:

1. Lewis M, Simpson P, Fransen S et al. CXCR5-using virus detected in patients receiving maraviroc in the Phase III studies MOTIVATE 1 and 2 originates from a pre-existing minority of CXCR4-using virus. XVI International HIV Drug Resistance Workshop, 12-16 June 2007, Barbados. Abstract 56. Published as part of Antiviral Therapy (Volume 12 Issue 5). Antiviral Therapy. 2007;12:S65.
2. Mascolini M. Several genotyping systems offer hope as tropism predictors. HTB May 2007.
   http://www.i-base.info/htb/2498
3. Mascolini M. Accuracy of Monogram’s Trofile assay in determining tropism and new tropism assays in development. HTB May 2007.
   http://www.i-base.info/htb/2501