Consensus guidelines recommend routine use of genotypic tropism testing: new focus on maraviroc as a switching option (2010)
Simon Collins, HIV i-Base
New consensus guidelines on tropism testing, produced by a panel of European virologists with expertise in HIV drug resistance produced from a literature review of 57 journal papers and 42 conference abstracts were presented by Annemarie Wensing.
Currently the use of maraviroc is technically restricted to people who are shown to have R5-tropic HIV determined by phenotypic tropism testing (Trofile ES). However, genotypic population sequencing of the V3 loop of env has shown quantifiable differences between X4 and R5 tropic virus and using a genotypic predictive algorithm has similar accuracy at predicting R5 tropism. While phenotypic tropism testing requires viral load >1000 copies/mL, genotype testing can be used for patients who are currently virologically suppressed on treatment, increasing the confidence in identifying people who might benefit from maraviroc as a switch option.
Key recommendations include tropism testing for ARV-naive patients (who have limited drug choices, perhaps due to toxicity of their first treatment, drug resistance or drug interactions) and treatment experienced patients prior to a change of treatment (noting the importance of repeat testing over time due to shifting tropism).
The panel recommended both the phenotypic Enhanced Sensitivity Trofile Assay (ESTA) and genotypic population sequencing, but recognise that for patients not based in the US, genotypic testing will be preferred due to greater accessibility, lower cost and shorter turnaround time. The list price of ESTA is close to $2000 (although ViiV can subsidise this cost) compared to 150-250 for genotype tests. 
The panel also provides guidance on technical aspects and interpretation issues.
For genotype testing triplicate PCR amplification and sequencing is advised using the online geno2pheno interpretation (coreceptor) tool  with a false positive rate (FPR) of 10%. If the viral load is below the level of amplification, proviral DNA can be used, and the panel recommends performing triplicate testing and use of an FPR of 10%. If genotypic DNA testing is not performed in triplicate the FPR should be increased to 20%.
Although prospective data on the use of genotypic tropism testing are generally from small observational studies, the ability to set the FPR depending on patient circumstances is important. For patients with very limited treatment options or with a strong background regimen, using a lower cut-off may be acceptable to improve the chance of using maraviroc when the presence of low-level X4 may be acceptable. Increasing the precision for excluding X4 by setting a higher FPR may be more important in less experienced patients to avoid the loss of other drugs in the combination. 
The geno2pheno includes adjustment for CD4 nadir, relevant given that HIV progression correlates positively with the risk of shifting from R5 to X4 virus.
BHIVA guidelines recommend routine use of triplicate testing but not routine tropism testing for all naive patients (unless maraviroc is being considered as a treatment option. 
- Wensing AMJ et al. Consensus statement of the European guidelines on clinical management of HIV-1 tropism testing. 10th International Congress on Drug Therapy in HIV Infection. 711 November 2010. Glasgow. Abstract O121. Published in Journal of the International AIDS Society 2010, 13(Suppl 4):O50doi:10.1186/1758-2652-13-S4-O121. http://www.jiasociety.org/content/13/S4/O7
- Harrigan PR, Geretti AM. Genotypic tropism testing: evidence-based or leap of faith? AIDS 2010 (Published ahead of print, 19 November 2010). doi: 10.1097/QAD.0b013e32834113f9. http://journals.lww.com/aidsonline/Citation/publishahead/Genotypic_tropism_testing__evidence_based_or_leap.99335.aspx
- Online Geno2pheno coreceptor calculator. http://coreceptor.bioinf.mpi-inf.mpg.de
- Geretti AM, Mackie N. Determining HIV-1 tropism in routine clinical practice (2009). BHIVA web publication. http://www.bhiva.org/Tropism.aspx