To the Editor:
Genotypic resistance testing (GRT) is recommended and is increasingly used in the routine clinical care of HIV-infected patients for whom highly active antiretroviral therapy (HAART) is failing. 1 GRT has proven efficacy in managing patients for whom HAART fails 2,3; however, GRT is expensive (approximately $200 per test). In addition, because of the replicative advantages of wild-type virus (WTV) over drug-resistant virus, 4,5 GRT, if performed when a patient has stopped taking antiretroviral medications, may yield results that do not accurately reflect acquired drug mutations. In previous series of patients for whom HAART failed, the average rate of WTV detection by GRT was 28% (range, 10%–41%). 6–9 Because of limited resources at our large public clinic, we sought to maximize the cost-effectiveness of GRT by implementing guidelines for ordering resistance tests.
We performed a retrospective review of results of all GRT performed at the CORE Center (Chicago, IL) between September 2000 and December 2002 as a part of routine care. The CORE Center is the largest HIV clinic in the midwestern United States, serving >3500 HIV-infected patients in 2002. Sixty percent of patients have no insurance coverage for laboratory tests. Before the availability of resistance testing in 2000, guidelines were developed to improve the cost-effectiveness of testing. A quality assurance plan was established to review the rate of WTV detection results under these guidelines.
For GRT, patients had to be currently receiving antiretroviral therapy, have an HIV RNA level of ≥5,000 copies/mL (later changed to ≥1,000 copies/mL [in 2001] as sensitivity of the assay improved), and be considered adequately adherent at the time of testing as determined by a clinical pharmacist or attending physician. To demonstrate adherence, patients had to be compliant with clinic visits and have adherence assessed by the primary medical care provider. Each request for GRT was reviewed and approved by a supervising infectious disease attending physician. Health care providers were also encouraged to have the patients evaluated by the on-site clinical pharmacist, who assessed adherence with screening questions including pill count, dosing frequency, pill recognition, and number of missed doses. Patients who were thought to be completely nonadherent (taking no medications) were instructed to restart ordered medications or work with their health care providers to obtain a more tolerable regimen and then undergo repeated testing for viral load after 4–6 weeks of therapy. Patients were told that the test was expensive and that the test results would indicate whether therapy was failing due to the development of drug resistance or only apparently failing due to complete nonadherence.
Sequencing of the reverse transcriptase and protease genes was performed on plasma viral RNA by Quest Diagnostics (Nichols Institute, San Juan Capistrano, CA) and later in the study period by LabCorp (Research Triangle Park, NC) in accordance with the instructions of the manufacturer.
There were 424 GRT procedures performed for 406 patients (18 patients had ≥2 tests). Fifty patients had the test ordered and approved but failed to get blood samples drawn for GRT. The clinical pharmacist evaluated 233 (55%) of the 406 patients who underwent GRT. Most patients were highly treatment experienced, and their second, third, or fourth regimen was failing. WTV was sequenced in 16 patients (3.8%); in 12 patients (2.8%), M184V was the only mutation present. The other 394 tests demonstrated resistance mutations consistent with current or previous antiretroviral therapy.
Nine (56%) of 16 patients who had WTV sequenced were not evaluated by the clinical pharmacist. Further evaluation of the 7 patients who were evaluated by the clinical pharmacist showed that 3 were adherent and approved, 3 were nonadherent but GRT was performed against the recommendation of the clinical pharmacist, and 1 was treatment naive.
Detection of WTV in patients with apparent failure of HAART due to nonadherence is not uncommon, with rates ranging from 10% to 41% in clinic and trial cohorts. 6–9 In this study, we found a 3.8% rate of WTV detection among our cohort, which to our knowledge is the lowest rate reported by any center to date. The largest series, performed by reference laboratories, reported a 29%–37% prevalence of WTV detection among >35,000 samples submitted for genotyping by physicians between 1999 and 2002. 6
For a patient with virologic failure who has WTV sequenced, the conclusion drawn is usually complete nonadherence. If these patients with WTV are subsequently questioned about tolerability and counseled on the importance of adherence to the original regimen, an intensified regimen or a more tolerable regimen is often successful in suppressing viral replication. We do not believe that we had fewer patients with apparent failure of HAART due to nonadherence in our clinic compared with other cohorts. However, by instituting a system of reviewing for adherence and informing patients of the ability of GRT to detect nonadherence, both health care providers and patients were more likely to order the tests in the setting where the results would be useful. This may also have played a part for the 50 patients who had GRT approved but did not undergo testing. It could be argued that using GRT to detect nonadherence may be more cost-effective than continuing to prescribe and dispense medications that the patient is not taking. However, it is unlikely that a health care provider would discontinue medications completely for a patient judged to be nonadherent.
If we presume that the actual rate of nonadherence leading to apparent HAART failure is the same as in other studies that did not use restrictions (average WTV detection rate, 28%), compared with our rate of 3.8%, we prevented ∼90 tests at a cost savings of approximately $18,000. The upper limit of WTV detection in reported series, ∼40%, if true in our clinic would have resulted in approximately $37,000 in additional testing costs. In the absence of these guidelines, we believe that our rate of WTV detection would have been as high as or higher than those for other clinics.
Perhaps, more importantly, by identifying nonadherence without waiting for the ∼2- to 3-week turnaround time of GRT, we created the opportunity to intervene in improving adherence or changing an intolerable regimen sooner. It can be reasonably speculated that earlier adherence intervention may, by preventing the emergence of resistance in a partially adherent patient, prolong the effectiveness of a given set of medications.
The limitations of this study include its retrospective, noncontrolled nature, the diversity of regimens involved, and the lack of information on how many requested tests were not approved by the attending physician or clinical pharmacist.
In conclusion, we found that clinicians and clinical pharmacists may be able to detect complete nonadherence in patients for whom HAART is failing before ordering GRT and thus optimize the use of these tests. The use of guidelines for ordering GRT was effective in keeping detection of WTV to a minimum, thus increasing the cost-effectiveness of GRT. This may be particularly important in resource-limited settings.
Toyin M. Adeyemi, MD*
Blake Max, PharmD†
Sheila M. Badri, MD*
David E. Baker, MD*
1. US Department of Health and Human Services. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. Updated February 2002. Available at http://www.hivatis.org
2. Tural C, Ruiz L, Holtzer C, et al. Clinical utility of HIV-1 genotyping and expert advice: the HAVANA trial. AIDS. 2002; 16:209–218.
3. Baxter JD, Mayers DL, Wentworth DN, et al. A randomized study of antiretroviral management based on plasma genotypic antiretroviral resistance testing in patients failing therapy. CPCRA 046 Study Team for the Terry Beirn Community Programs for Clinical Research on AIDS. J Infect Dis. 2001; 183:401–408.
4. Deeks SG, Grant RM, Wrin T, et al. Persistence of drug-resistant HIV-1 after a structured treatment interruption and its impact on treatment response. AIDS. 2003; 17:361–370.
5. Izopet J, Massip P, Souyris C, et al. Shift in HIV resistance genotype after treatment interruption and short-term antiviral effect following a new salvage regimen. AIDS. 2000; 14:2247–2255.
6. Lanier ER, Ait-Khaled M, Craig C, et al. Prevalence of mutations associated with resistance to antiretroviral therapy from 1999–2002 [abstract 635]. 10th Conference on Retroviruses and Opportunistic Infections, February 10–14, 2003.
7. Badri SM, Smith KY, Huang D, et al. Utility of a genotypic assay in HIV-1 infected patients failing salvage therapy [abstract 364]. 38th annual meeting of the Infectious Diseases Society of America, September 7–10, 2000.
8. Erice A, Ribaudo H, Demeter LM, et al. Antiretroviral resistance in ACTG 388 participants with virologic failure [abstract 564-T]. 9th Conference on Retroviruses and Opportunistic Infections, January 2002.
9. Miller MD, Margot NA, McColl DJ, et al. Genotypic and phenotypic characterization of virologic failure through 48 weeks among treatment-naive patients taking tenofovir DF or stavudine in combination with lamivudine and efavirenz [abstract]. 6th International Congress on Drug Therapy in HIV Infection, Glasgow, UK, November 2002.