International Association of Physicians in AIDS Care, February 2000 Journal
Renslow Sherer, MD
Director, Coordinated HIV Services, The CORE Center, Cook County Hospital, Associate Professor of Medicine, Rush Medical College
Chicago, Illinois, USA
Retrospective studies done in the past few years show that genotypic and phenotypic resistance testing of HIV can predict virologic outcome.1-9 More recently, two prospective studies documented short-term improvement in virologic outcome when genotyping was used to help select a new regimen for patients in whom highly active antiretroviral therapy (HAART) had failed.10,11 Similarly, a recent prospective study demonstrated superior short-term virologic outcomes for patients whose clinicians saw phenotypic resistance results at the time of first HAART failure.12 In November 1999 the US Food and Drug Administration (FDA) began requiring resistance testing in clinical trials of new antiretroviral agents.13 Clinical use of resistance testing is widespread; in a recent symposium in the United States, 54 percent of 900 clinicians indicated occasional use of these assays in clinical practice.14 The increasing prevalence of single drug mutations in 5 to 15 percent of newly infected persons and of multiple resistance mutations in 1 to 3 percent underscores the epidemiologic imperative to track the spread of drug-resistant virus accurately and to assess newly infected individual carefully.14
|
This article addresses the question of whether genotypic and phenotypic resistance testing should be considered the standard of care for individuals taking antiretroviral therapy. Table 1 notes the factors to be considered in attempting to answer this question; each will be considered here.
Almost two years ago antiretroviral guidelines noted the predictive value of resistance testing.15,16 These reviews shared a cautious optimism about the potential value of resistance testing, while falling short of unequivocally recommending its use in clinical practice. An updated version of the IAS-USA Resistance Testing Guidelines, which were first published in July 1998,15 is currently in review. Table 2 outlines potential indications for resistance testing in the most recent update of the Department of Health and Human Services (HHS) guidelines.17
The most common shortcomings of genotypic and phenotypic resistance tests include lack of long-term clinical follow-up in reported studies, unreliable standardization, variable reproducibility of testing methodology, disorganized or confusing reporting formats for test results, lack of physician expertise in interpretation, excessive cost, and limited access.
| Clinical setting/recommendation | Rationale |
|---|---|
| Recommended | |
| Virologic failure during HAART | Determine the role of resistance in drug failure and maximize the number of active drugs in the new regimen, if indicated |
| Suboptimal suppression of viral load after initiation of antiretroviral therapy | Determine the role of resistance and maximize the number of active drugs in the new regimen, if indicated |
| Consider | |
| Acute HIV infection | Determine if drug-resistant virus was transmitted and change regimen accordingly |
| Not generally recommended | |
| Chronic HIV infection before initiation of antiretroviral therapy | Uncertain prevalence of resistant virus; current assays may not detect minor drug-resistant species |
| After discontinuation of a drug | Drug resistance mutations may become minor species in the absence of selective pressure; current assays may not detect minor drug-resistant species |
| Plasma viral load less than 1000 HIV RNA copies/mL | Resistance assays cannot be performed reliably because of low HIV RNA copy number |
| *From the Department of Health and Human Services guidelines, January 28, 2000.17 | |
There is ample evidence that certain genotypic changes and phenotypic evidence of decreased susceptibility to a drug predict poor virologic outcome, although there are some inconsistencies in these data. Table 3 lists several retrospective studies showing the predictive value of genotypic and phenotypic testing.
† Patick AK, Zhang M, Hertogs K, et al. Correlations of virological response with genotype and phenotype of plasma HIV-1 variants in patients treated with nelfinavir in the US expanded access program. Antiviral Ther 1998;3(suppl 1);1998:39. ‡ Lorenzi P, Opravil M, Hirschel B, et al. Impact of drug resistance mutations on virologic response to salvage therapy. Swiss HIV Cohort Study. AIDS 1999;13:F17-F21. ¶ Lanier R, Ait-Khaled M, Madison S, et al. Analysis of possible predictors of response to abacavir (ABC) in antiretroviral-experienced adults: comparison of viral genotype, viral phenotype and patient treatment history. Presented at: 6th Conference on Retroviruses and Opportunistic Infections; January 31-February 4, 1999; Chicago. Abstract 134. ¥Deeks SG, Hellman NS, Grant RM, et al. Novel four drug salvage treatment .regimens after failure of a human immunodeficiency virus type 1 protease inhibitor containing regimen: antiviral activity and correlation of baseline phenotypic drug susceptibility with virologic outcome. J Infect Dis 1999;179:1375-1381. |
Clinical history alone was sufficient to predict the failure of subsequent therapy in some studies, for example, with abacavir in heavily nucleoside-experienced patients3 or with nelfinavir in patients with prolonged protease inhibitor experience.4 Zolopa and colleagues found that clinical history alone predicted 40 percent of the treatment response in a univariate analysis;5 however, genotype alone predicted 67 percent of the treatment response, and adding clinical history into the model in the multivariate analysis raised that value only to 71 percent. This evidence indicates that resistance testing improves a clinician's ability to select second-line regimens compared with treatment history alone.
Several studies looked at either genotypic or phenotypic resistance tests and found no association between test results and virologic outcome.18,19 It is important to learn from these studies as well. There are numerous possible reasons for lack of association between the presence of resistance mutations and virologic failure, because there are many reasons besides resistance for virologic failure. These include poor adherence to therapy and pharmacologic factors such as incomplete drug absorption, drug-drug interactions, or rapid metabolism or excretion of drugs. An important feature of future clinical trial design will be the simultaneous prospective study of all of these factors in addition to the common surrogate markers, CD4+-cell count and viral load.
There is insufficient retrospective evidence to judge the superiority of one type of resistance test over the other; both offer important evidence and are likely to be complementary in clinical practice. Nonetheless, the relative predictive value of each test in relation to the other must be addressed prospectively in the present generation of studies and by analyzing large data sets of genotypic and phenotypic information linked to patient outcomes.
The cost efficacy of each test and the relative cost efficacy of genotypic assays compared with phenotypic assays are also uncertain. Preliminary data from the VIRADAPT study10 (see below) suggest that genotypic testing improves short-term virologic outcome without increasing the overall cost of care. These data need to be bolstered by additional studies with larger cohorts and longer follow-up.
Two prospective studies--the European VIRADAPT study and the GART study in the US--showed the positive predictive value of genotypic resistance testing.10,11 Both studies examined patients in whom at least one HAART regimen had failed virologically.
The VIRADAPT investigators randomized participating clinicians to receive genotypic information immediately or six months after they switched patients to a new regimen.10 Virologic endpoints were noted at six months. Patients in the genotype arm had a mean viral load reduction 0.6 log greater than patients in the control arm (P = 0.015). After six months patients were allowed to cross over to the genotype arm for an additional six months. This crossover resulted in improved virologic outcomes after 12 months among patients originally randomized to the standard-of-care arm, while virologic responders in the genotype arm maintained those responses through a year.
The GART study randomized 120 patients to switch their regimen on the basis of history alone or with the help of genotypic testing, expert interpretation of test results, and treatment history.11 Virologic outcomes were measured at 8 and 12 weeks. Short-term results were similar to those in the VIRADAPT study: Patients randomized to the GART arm had a mean 0.5 log greater reduction in viral load than did patients in the control arm (P = 0.003 at week 12).
The GART investigators also analyzed physician behavior regarding the expert interpretation. Roughly 33 percent followed the recommendations completely, 33 percent disregarded the recommendations, and 33 percent followed some but not all of the recommendations. Patients of the physicians who followed the recommendations completely did best, with a 1.6-log viral load reduction. Importantly, patients in the GART arm had significantly fewer drug changes and fewer inactive drugs in their new regimen than control patients, a finding suggesting that genotypic testing may reduce the inappropriate use of potentially toxic and costly medications.
Taken together, these studies suggest that genotypic testing leads to improved short-term virologic outcomes. Whether genotyping will improve long-term clinical outcome has yet to be proved. Nonetheless, the virologic benefit and the reduced number of unnecessary drugs are important steps towards establishing the utility of real-time genotypic testing in clinical practice. Note the importance of standardization of laboratory testing and expert interpretation in these studies. Directions for future genotyping studies include using genotypic assays without any additional interpretation, and using real-world clinical labs to perform the assays.
More recently, a randomized, prospective 16-week study tested the Virco phenotypic assay in a cohort experiencing their first failure of HAART.12 Investigators assigned 117 patients to receive phenotypic testing during a four-week lead-in period between randomization at HAART failure and regimen change; 111 patients randomized to the standard-of-care received second-line therapy on the basis of treatment history. In both groups second-line regimens were limited to three or four drugs, and 96 percent of patients were NNRTI naive. After 16 weeks the viral load in the phenotype group decreased by 1.27 logs compared with 0.56 log in the comparison arm, a significant difference (P = 0.011).
Though much discussion in the past year has focused on the relative strengths and weaknesses of genotypic and phenotypic resistance testing, studies completed to date have not compared these two assays in a rigorous manner. Fortunately, several prospective studies now under way will do so: The CPCRA FIRST study, the European INITIO trial, and ACTG 384, among others, will compare genotyping and phenotyping in nested studies to assess the predictive value and relative merits of both tests. It is tempting to speculate that phenotypic resistance testing alone may be sufficient, given its simpler and more familiar reporting format. However, because genotypic changes usually precede phenotypic changes by weeks to months, and because results may be technically and economically more accessible, genotyping may offer more timely and accessible results than phenotyping, at least in the short term.
Several recent studies have assessed large international databases of genotypic mutations and corresponding phenotypic changes.20-24 Analyses of these databases have made it possible to identify primary resistance mutations associated with current antiretroviral agents. Though these databases still lack direct clinical correlation, particularly with the resistance imprinting known to occur with various sequences of antiretroviral regimens, they are invaluable in providing cross-sectional glimpses at the prevalence of genotypic mutations and the correspondence of genotypic changes with in vitro phenotypic resistance. One such compilation is available to researchers and clinicians via the Internet at http://hiv-web.lanl.gov.
As with any diagnostic test, resistance assays must be held to the familiar standards of sensitivity and specificity, positive and negative predictive value, and cost efficacy in real time. These determinations can only be made with larger prospective studies over longer periods of follow-up.
Whenever a new clinical technology performed in experienced laboratories is found to improve patient care, there is a period of assimilation in less experienced laboratories during which quality and standardization of the new assay are uncertain. It appears that genotypic resistance testing is entering this era. Excellent congruity of genotypic results from experienced reference labs has been established. However, there are large disparities in performance between local laboratories.
One recent study of regional laboratories in the United States and Europe found substantial improvement over 18 months in correctly identifying genetically engineered viruses with up to five primary protease gene mutations and five reverse transcriptase gene mutations, yet substantial inaccuracies remained.25 The labs did best in identifying either wild-type virus or virus with all 10 mutations, and worst with samples that contained varying prevalences of mutations; up to 50 percent of participating laboratories failed to identify these mixed viruses accurately.
Phenotypic resistance testing is available from two international reference laboratories, Virco in Belgium and ViroLogic in the United States, though other assays are in development. Phenotypic resistance testing is technically difficult because of the sophisticated molecular genetic technology required for high throughput and rapid turnaround, so it is unlikely to be available in many laboratories in the near future.
The GART study was instructive in establishing the value of expert interpretation of HIV clinical treatment histories and genotypic resistance testing.11 In contrast, in VIRA 3001 no instructions were given to clinicians who received phenotypic test results before choosing the second-line regimen.12 Physician expertise in interpreting both genotypic and phenotypic results will clearly be an important dimension in establishing resistance testing as the standard of care.
Interpreting genotypic testing is more complex because it requires an ongoing knowledge of primary and secondary resistance mutations. Phenotypic resistance testing is more immediately familiar to clinicians because of its similarities to resistance testing of antimicrobials and antimycobacterial agents. However, there are many important differences between such tests and HIV resistance testing. For example, antibiotic resistance assays measure killing of bacteria in vitro; antiretroviral phenotypic resistance assays measure the sensitivity of HIV strains to an agent that may be used in the future against an HIV swarm containing unidentifiable minority species that may be fully or partly resistant to the drug in question.
Clearly the increasing use of HIV resistance testing will require additional education and training for both experienced and less-experienced clinicians. There is ample evidence that care from physicians who lack HIV experience is associated with poorer clinical outcomes than care from more experienced physicians. Resistance testing will likely require additional time for consultation with HIV researchers knowledgeable in interpreting both types of resistance assays. As a corollary requirement, clear, unambiguous reporting of the results of resistance testing is needed to avoid confusion.
There is sufficient evidence of short-term virologic benefit with both genotypic and phenotypic assay results at the time of HAART failure to warrant their use in specific clinical settings, as noted in the recent update in the HHS guidelines (Table 2). In general, these are settings in which clinical uncertainty exists regarding the optimal next HAART regimen and in which multiple treatment options are available to the patient. Several reviews and guidelines over the past year have noted the utility of resistance testing and recommended its use in selected clinical settings. Thus resistance testing has entered an era in which it has become an element of the standard of care in HIV disease management.
Numerous unresolved questions and important issues related to the use of resistance testing require urgent attention. The relative merits of genotypic and phenotypic resistance testing need further study and clarification. Standardization of testing methodology and quality controls for laboratories that perform these assays must be established and tested before widespread testing is done with erroneous or incomplete data. In addition, international standards for reporting formats are needed to simplify the complex task of interpreting data. As with any clinical test, determining the sensitivity, specificity, and positive and negative predictive values of resistance tests should be a high priority. An additional compelling question being addressed by long-term prospective clinical trials is the long-term impact of resistance assays.
The important public health question of the cost efficacy of resistance assays must be addressed. Specifically, the potential benefit and cost efficacy of resistance testing will have to be compared with other measures associated with improvements in morbidity, mortality, and quality of life. Finally, the critical need for aggressive physician training and education in using and interpreting resistance testing will be an essential component of the ascent of another difficult and important learning curve for HIV clinicians.
2. Condra JH. Virologic and clinical implications of resistance to HIV-1 protease inhibitors. Drug Resistance Updates 1998;1:292-299.
3. Lanier R, Danehower S, Daluge S, et al. Genotypic and phenotypic correlates of response to abacavir (ABC, 1592). Antiviral Ther 1998;3(suppl 1):36.
4. Hammer S, Squires K, DeGruttola V, et al. Randomized trial of abacavir (ABC) & nelfinavir (NFV) in combination with efavirenz (EFV) & adefovir dipivoxil (ADV) as salvage therapy in patients with virologic failure receiving indinavir (IDV). Presented at: 6th Conference on Retroviruses and Opportunistic Infections; January 31-February 4, 1999; Chicago. Abstract 490.
5. Zolopa AR, Shafer RW, Warford A, et al. HIV-1 genotypic resistance patterns predict response to saquinavir-ritonavir therapy in patients in whom previous protease inhibitor therapy had failed. Ann Intern Med 1999;131:813-821.
6. Harrigan PR, Montaner JS, Hogg RS, et al. Baseline resistance profile predicts response to ritonavir/saquinavir therapy in a community setting. Antiviral Ther 1998;3(suppl 1);1998:38.
7. Patick AK, Zhang M, Hertogs K, et al. Correlations of virological response with genotype and phenotype of plasma HIV-1 variants in patients treated with nelfinavir in the US expanded access program. Antiviral Ther 1998;3(suppl 1);1998:39.
8. Deeks SG, Hellmann HS, Grant RM, et al. Novel four-drug salvage treatment regimens after failure of a human immunodeficiency virus type 1 protease inhibitor-containing regimen: antiviral activity and correlation of baseline phenotypic drug susceptibility with virologic outcome. J Infect Dis 1999;179:1375-1381.
9. Call S, Westfall A, Cloud G, et al. Predictive value of HIV phenotypic susceptibility testing in a clinical cohort. Presented at: 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 26-29, 1999; San Francisco. Abstract LB17.
10. Durant J, Clevenbergh P, Halfon S, et al. Drug-resistance genotyping in HIV-1 therapy: the VIRADAPT randomised controlled trial. Lancet 353;1999:2195-2199.
11. Baxter JL, Mayers DL, Wentworth DN, et al. A pilot study of the short-term effects of antiretroviral management based on plasma genotypic antiretroviral resistance testing (GART) in patients failing antiretroviral therapy. Presented at: 6th Conference on Retroviruses and Opportunistic Infections; January 31-February 4, 1999; Chicago. Abstract LB8.
12. Cohen C, Hunt S, Sension M, et al. Phenotypic resistance testing significantly improves response to therapy: a randomized trial (VIRA3001). Presented at: 7th Conference on Retroviruses and Opportunistic Infections; January 30-February 2, 2000; San Francisco. Abstract 237.
13. New York Times. November 4, 1999:8.
14. Boucher C. Role of resistance testing in clinical practice: the value of resistance assays when changing therapy. Presented at: 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 26-29, 1999; San Francisco. Abstract 1368.
15. Hirsch MS, Conway B, D'Aquila R, et al. Antiretroviral drug resistance testing in adults with HIV infection: implications for clinical management. JAMA 1998;279:1984-1991.
16. Report of the NIH Panel to Define Principles of Therapy of HIV Infection and Guidelines for the Use of Antiretroviral Therapies in HIV Infected Adults and Adolescents. Ann Intern Med 1998;128:1057-1100. Available at: ftp://ftp.cdc.gov/pub/Publications/mmwr/rr/rr4705.pdf.
17. Department of Health and Human Services (DHHS) and the Henry J. Kaiser Family Foundation Panel on Clinical Practices for Treatment of HIV Infection. Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents. January 28, 2000. Available at: http://www.hivatis.org/trtgdlns.html#AdultAdolescent.
18. Schmuck A, Chanzy B, Brengel Pesce B, et al. Lack of usefulness of genotypic resistance assay in naive HIV positive patients with low levels of viremia after 6 months of treatment. Presented at: 7th European Conference on Clinical Aspects and Treatment of HIV Infection; October 23-27, 1999; Lisbon. Abstract 368.
19. Zolopa A, Tebas P, Gallant J, et al. The efficacy of ritonavir/saquinavir antiretroviral therapy in patients who failed nelfinavir: a multicenter cohort study. Presented at: 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 26-29, 1999; San Francisco. Abstract 2065.
20. Hertogs K, deBethune MP, Miller V, et al. A rapid method for simultaneous detection of phenotypic resistance to inhibitors of protease and reverse transcriptase in recombinant human immunodeficiency virus type 1 from patients treated with antiviral drugs. Antimicrobial Agents Chemother 1998;42:269-276.
21. Young B, Johnson S, Bahktiari M, et al. Resistance mutations in protease and reverse transcriptase genes of HIV-1 isolates from patients failing combination antiretroviral therapy. J Infect Dis 1998;178:1497-1501.
22. Pauwels R, Hertogs K, Kemp S, et al. Comprehensive HIV drug resistance monitoring using rapid, high throughput phenotypic and genotypic assays with correlative data analysis. Antiviral Ther 1998;3(suppl 1):35-36.
23. Hertogs K, Deroey V, Vandeneynde C, et al. Common, rare, and new genotypic and/or phenotypic HIV-1 resistance profiles observed in over 5,000 isolates. Presented at: 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 26-29, 1999; San Francisco. Abstract 425.
24. Kuritzkes D. Managing drug resistance. Medscape HIV/AIDS. 1999. Available at: http://www.medscape.com/medscape/HIV/journal/ 1999/v05.n03s/mha0622.05.kuri/mha0622.05.kuri-01.html.
25. Schuurman R, Brambila D, de Groot T, Boucher C. Second worldwide evaluation of HIV-1 drug resistance genotyping quality using the ENVA 2 panel. Antiviral Ther 1999;4(suppl 1):41.
000210
IA000207
ÆGIS is made possible through unrestricted grants from Roxane Laboratories, the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeard in 2000. This material is designed to support, not replace, the relationship that exists between you and your doctor.
Copyright © 2000 - International Association of Physicians in AIDS Care. All rights reserved. http://www.iapac.org
This information is designed to support, not replace, the relationship that exists between you and your doctor.
55 percent <400 HIV RNA copies/mL