A genotypic test is just as accurate as a more expensive tropism test in determining which treatment-experienced patients are suitable for treatment with the CCR5 inhibitor maraviroc, researchers from the University of British Columbia reported at the Fifth International AIDS Society conference in Cape Town last week.
The finding opens the prospect of much cheaper testing to determine eligibility for CCR5 inhibitor treatment, and coincides with a re-analysis of 96-week results from the MERIT study showing that maraviroc is just as effective as efavirenz in treatment-naive patients, albeit better tolerated.
CCR5 inhibitors are a new class of antiretroviral drug that block the CCR5 co-receptor used by HIV to gain entry to CD4 cells. CCR5 inhibitors only have an antiviral effect in people who are predominantly infected with HIV that is adapted to using the CCR5 receptor, typically patients with higher CD4 counts and less advanced HIV disease.
In order to determine whether a person has HIV that is susceptible to a CCR5 inhibitor, a blood sample is tested to determine whether HIV uses the CCR5 receptor or not – its so-called 'tropism'. An assay developed by Monogram Biosciences, called the Trofile assay, is the only current means of determining which patients will benefit from a CCR5 inhibitor.
The drawbacks of the Trofile assay are its cost (launched in 2007 at around US $1960 per test) and the turnaround time for results – up to three weeks. In patients who are switching treatment it can only be used where viral load has risen above 1000 copies/ml.
The alternative is to look at a region of HIV’s envelope protein called the V3 loop to determine tropism. This can be done during a standard genotyping test for drug resistance, at no extra cost. (Anyone starting or changing treatment should undergo resistance testing, US and European treatment guidelines state.) The genotype is then used to predict the phenotype, or tropism, an approach called 'geno2pheno'.
Until now, comparisons of the accuracy of genotyping and tropism testing have been carried out in fairly small patient groups, and have left researchers unconvinced about the equivalence of the two testing methods.
The study presented last week by Professor Richard Harrigan and colleagues from the University of British Columbia Centre for Excellence in HIV/AIDS, Vancouver, analysed the performance of the two tests in predicting virological response to maraviroc in 1216 patients who participated in MOTIVATE 1 and 2 (999 patients) and in the A4001029 study (165 patients). Each study evaluated response to maraviroc in treatment-experienced patients.
Using stored blood samples, the researchers compared the ability of the baseline testing methods to predict virologic response to treatment at weeks 8 and 24.
The study found that Trofile and genotyping produced comparable results in terms of detecting viral phenotype and in predicting virologic response, regardless of patient characteristics.
| R5 tropism | Sensitivity for virologic suppression at w8 | Specificity for virologic suppression at w8 | Predicted proportion with viral load below 50 copies/ml, w24 | |
| Trofile | 72% | 72% | 92% | 46.4% |
| Geno2pheno | 72% | 72% | 89% | 46.1% |
Based on viral tropism and the degree of sensitivity to drugs in the background regimen (as assessed by genotype), the researchers were also able to predict the degree of virologic suppression and the proportion of patients with viral load below 50 copies, and found very high levels of agreement between the two assays. Patients identified as CCR5-tropic had almost identical degrees of viral load reduction (-2.5log at week 8).
“HIV V3 genotyping shows promise as a significantly faster and more cost-effective way to correctly identify patients who would benefit from CCR5 antagonists like maraviroc,” said Professor Harrigan.
“Since the genotypic test is based on methods that are already widely used through the same labs that provide HIV drug resistance testing, this approach could become broadly available and conducted at the same time as resistance testing to determine susceptibility to all drugs, including maraviroc.”
The study, which was supported by maraviroc’s manufacturer Pfizer, comes at an important time for the company in its attempts to breathe new life into sales of maraviroc.
According to Pharmacogenomics Reporter at www.genomeweb.com, Monogram Biosciences is currently testing less than 2500 samples a quarter for CCR5 tropism, indicating the small size of the market for a drug that Pfizer hoped would be a blockbuster HIV product.
Although the cost of the assay is one obstacle to wider use, the other barrier to adoption has been a previous failure to demonstrate that maraviroc is as effective as the current standard of care, efavirenz (Sustiva or Stocrin), in first-line treatment.
However, at IAS 2009 Pfizer presented results of a new analysis of 96-week results from the MERIT study, using a more sensitive version of the Trofile assay, showing equivalence between maraviroc and efavirenz in virologic outcomes, but superior immunologic and lipid outcomes for maraviroc-treated patients.
MERIT was a randomised comparison of maraviroc or efavirenz, combined with AZT/3TC, in CCR5-tropic patients. The 48-week results of the study, presented at the 2005 IAS conference in Sydney, showed an inferior virologic response among maraviroc-treated patients.
Pfizer has always contended that the reason for the inferior response to maraviroc was failure of the assay to detect a handful of patients with dual-tropic or CXCR4-tropic virus.
Since the study was carried out Monogram has developed a more sensitive version of the Trofile assay, improving its ability to accurately identify CCR5-tropic virus and dual-tropic virus.
When this enhanced assay was used retrospectively to analyse response according to baseline tropism, virologic responses at week 96 were similar (58.5% < 50 copies/ml in the maraviroc group vs 62.4% in the efavirenz group) in CCR5-tropic patients.
An almost identical proportion of patients in each arm remained on treatment after two years, but there were differences in the reasons for drug discontinuation. In the maraviroc group, 12.5% discontinued due to insufficient clinical response, compared to 5.9% in the efavirenz group. On the other hand, discontinuations due to adverse events were more frequent in the efavirenz arm (15.5% vs 6.1%), and occurred overwhelmingly in the first 48 weeks of the study. The proportion of patients experiencing serious adverse events was similar at weeks 48 and 96.
CD4 counts rose higher in maraviroc-treated patients (a 41-cell difference at week 96), and maraviroc-treated patients were significantly less likely to require lifestyle changes or lipid-lowering drugs as a result of cholesterol elevations (38% of efavirenz-treated patients had total cholesterol above 5.2mmol/l at week 96, compared to 10% of maraviroc-treated patients) (p<0.0001).
The results of the MERIT re-analysis are likely to provide enough evidence to justify licensing maraviroc for first-line use in the United States and Europe, while the results of the British Columbia study could make it much easier and cheaper to identify which patients might benefit from maraviroc. However, whether the twice-daily dosing of maraviroc will find favour in first-line treatment remains to be seen.