HIV-1 Coreceptor Tropism, Ultradeep Sequencing

Tropism refers to the type of cytokine coreceptor used by HIV-1 when infecting the host cell. The viruses in most (>80%) treatment-naïve patients use the CCR5 (R5) coreceptor.¹ Conversely, the viruses in up to 50% of treatment-experienced patients use either the CXCR4 (X4) coreceptor or both coreceptors (ie, R5 and X4).² Viruses that use both coreceptors are called dual-mixed (D/M) viruses.

A coreceptor tropism test determines whether a patient exclusively harbors R5-tropic virus or has X4-tropic or D/M virus. Patients who exclusively harbor R5-tropic virus can be treated with CCR5 antagonists such as maraviroc (Selzentry^®). CCR5 antagonists block R5 viruses from binding to the CCR5 coreceptor and infecting cells. CCR5 antagonists are ineffective in patients with X4 or D/M virus. Thus, coreceptor tropism testing can help determine patient eligibility for CCR5 antagonist therapy.

A coreceptor tropism test should be performed when the use of a CCR5 antagonist is being considered.^3,4 Coreceptor tropism testing might also be considered for patients who exhibit virologic failure while taking a CCR5 inhibitor.^3,4

Two methods are available in a few clinical laboratories: 1) phenotypic analysis of recombinant viruses and 2) genotypic analysis of the V3 loop of the HIV-1 envelope gene. Although phenotypic tropism testing is generally preferred because of a greater weight of supporting evidence, genotypic tropism testing should be considered as an alternative assay, according to the Department of Health and Human Services 2013 Guidelines.³ Genotypic testing costs less than phenotypic testing and has faster analytical times.³

Phenotypic tropism tests such as Trofile^® (Monogram Biosciences) use a cloned envelope gene from a patient’s HIV to infect R5 and X4 indicator cells that emit light when infected. The relative amount of light emitted by each cell type can then be measured to determine whether the patient’s virus is R5-tropic, X4-tropic, or D/M-tropic.

Genotypic tropism tests determine the DNA sequence of the third variable loop (V3) region of the HIV envelope gene, the primary determinant of viral tropism. The patient’s viral RNA is reverse-transcribed to DNA and amplified using PCR. After DNA sequencing, a number of different bioinformatic algorithms can be used to infer viral coreceptor usage from the V3 loop nucleic acid sequence.

At Quest Diagnostics, we reverse transcribe and amplify the HIV V3 loop sequences in triplicate to increase the chances of detecting minority X4 variants.^5,6 We then utilize ultradeep sequencing (UDS) and bioinformatics analysis to detect X4 variants.⁷ If X4-tropic virus is detected at a level of 2% or higher, the tropism test result is reported as X4 “detected” because CCR5 antagonists are unlikely to be effective in these cases.⁸

Yes, testing has been clinically validated in both treatment-experienced and treatment-naïve patient populations. Predictions of maraviroc response determined by genotypic tropism methods are similar to predictions generated by phenotypic methods.^6-9

Harrigan and Geretti summarized the results of several retrospective clinical studies and concluded HIV genotypic tropism testing is suitable for clinical use.¹⁰ This review article is available at Journals.lww.com/aidsonline/Fulltext/2011/01140/Genotypic_tropism_testing__evidence_based_or_leap.17.aspx#.

Treatment-experienced patient population

• In a retrospective clinical study, virologic responses to maraviroc predicted by population sequencing were consistently similar to the responses predicted by the original (less sensitive) Trofile assay.⁶
• When UDS was used to reanalyze more than 1,800 samples from treatment-experienced patients in the MOTIVATE and A4001029 clinical trials, 49% of maraviroc recipients with an R5 genotype maintained a viral load of <50 copies/mL after 48 weeks, compared to 46% with an R5 phenotype determined by the original Trofile method.⁸ A viral load of <50 copies/mL was maintained by 26% and 23% (genotype and phenotype, respectively) of patients predicted to harbor D/M or X4 virus.⁸
• Using a subset of 327 patients from the MOTIVATE and A4001029 studies, maraviroc response predictions made by Quest Diagnostics genotypic tropism assay (population sequencing with reflex to UDS) were compared to that of the enhanced sensitivity Trofile assay (ESTA). Both assays predicted the same short-term virologic response to maraviroc treatment. At week 8, the proportion of patients with R5 virus who responded to treatment (positive predictive value) was 66% for the ESTA and 65% for the Quest Diagnostics assay. The proportion of patients with D/M or X4 virus who did not respond to treatment (negative predictive value) was 59% and 58% for ESTA and the Quest Diagnostics assay, respectively.⁷
• Both assays also predicted the same CD4+ immunologic response at week 24. In patients with R5 virus, the CD4+ T-cell count increased by 88.0 cells/mL and 88.5 cells/mL for the ESTA and Quest Diagnostics assays, respectively. In patients with X4 virus, the increase in T-cell count was much smaller: 48.3 cells/mL and 35.5 cells/mL for the ESTA and Quest Diagnostics assays, respectively.⁷

Treatment-naïve patient population

• The clinical performance of UDS was also evaluated in a reanalysis of the MERIT study, which used maraviroc in a treatment-naïve patient population.⁹ Among patients identified as having R5 virus by UDS or by ESTA, 67% and 68%, respectively, achieved a viral load of <50 copies/mL at week 48. A lower proportion of patients found to harbor D/M or X4 virus achieved a virologic response at week 48: 46% for UDS and 45% for ESTA.

The original Trofile phenotypic assay was able to reliably detect 10% minority X4 variants, based on mixtures of R5 and X4 DNA clones.¹¹ When the same clonal experiments were conducted with the enhanced sensitivity assay

(ESTA), 0.3% minority X4 clones could be reliably detected.¹² Similarly, the UDS instrument used for Quest Diagnostics genotypic tropism testing can detect 0.5% X4 in mixed clones.⁷

Biological sensitivity is defined as the ability to reliably detect X4 virus in mixed R5/X4 plasma samples similar to those taken from HIV-1 patients. Unlike clonal analysis, biological sensitivity measures the sensitivity of the entire assay, which may be influenced by the efficiency of nucleic acid extraction and PCR amplification of minor variants, as well as the viral load in the sample.

The biological sensitivity of the Quest Diagnostics UDS assay to detect X4 virus in 90% of dual-mixed samples is 18% X4 at a viral load of 25,000 copies/mL and 6% X4 at a viral load of 100,000 copies/mL.⁷ The biological sensitivity of the Trofile assay, however, is unpublished. Because the Trofile assay also uses extraction and amplification, its biological sensitivity is likely higher than the published technical sensitivity.

Regardless of these potential differences in sensitivity, the Quest Diagnostics genotype assay and the current Trofile assay appear equal when predicting maraviroc response.⁷

If the plasma viral load is <1,000 copies/mL, we might not be able to amplify sufficient viral RNA to perform a standard tropism test. A proviral DNA tropism test (HIV-1 Coreceptor Tropism, Proviral DNA, test code 91299(X)) is available for tropism testing in these patients. For more information about the proviral DNA test, go to Education.QuestDiagnostics.com/faq/FAQ87.