Published ahead of print on August 18, 2005, doi:10.1164/rccm.200503-417OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200503-417OC
Acquired Rifamycin Resistance with Twice-Weekly Treatment of HIV-related TuberculosisDenver Public Health; University of Colorado Health Sciences Center, Denver, Colorado; Veterans Affairs Medical Center of Washington, DC; George Washington University Medical Center, Washington, DC; Centers for Disease Control and Prevention, Atlanta, Georgia; Los Angeles County–University of Southern California Medical Center, Los Angeles, California; Baylor College of Medicine, Houston; Tarrant County Public Health Department; University of North Texas Health Sciences Center, Fort Worth; University of Texas Health Science Center; South Texas Veterans Health Care System, San Antonio, Texas; University of Medicine and Dentistry of New Jersey–New Jersey Medical School National Tuberculosis Center, Newark, New Jersey; Harlem Hospital Center; and Columbia University College of Physicians and Surgeons, New York, New York Correspondence and requests for reprints should be addressed to William Burman, M.D., Denver Public Health, 605 Bannock Street, Denver, CO 80204. E-mail: bburman{at}dhha.org
Rationale: Rifabutin was recommended in place of rifampin during treatment of HIV-related tuberculosis (TB) to facilitate concomitant potent antiretroviral therapy, but this approach has not been evaluated in a prospective study. Objective: To evaluate the activity of intermittent rifabutin-based therapy. Methods: Patients with culture-confirmed TB were treated under direct supervision with 2 mo of rifabutin, isoniazid, pyrazinamide, and ethambutol (given daily, thrice-weekly, or twice-weekly per the local tuberculosis control program), followed by 4 mo of twice-weekly rifabutin plus isoniazid. Measurements: Culture-positive treatment failure or relapse. Main Results: A total of 169 eligible patients were enrolled. Most had advanced HIV disease; the median CD4 cell count and HIV-RNA level were 90 cells/mm3 (interquartile range, 35–175) and 5.3 log10 copies/ml (interquartile range, 4.8–5.7), respectively. Nine (5.3%) patients had culture-positive treatment failure (n = 3) or relapse (n = 6). Eight of these nine (89%) cases had isolates with acquired rifamycin resistance. Treatment failure or relapse was associated with baseline CD4 lymphocyte count, being 12.3% (9/73; 95% confidence interval, 6.5–22.0%) among patients with CD4 < 100 cells/mm3 versus 0% (0/65; 95% confidence interval, 0.0–4.5%) among those with higher CD4 lymphocyte counts (p < 0.01). One hundred thirty-seven (81%) patients received antiretroviral therapy during TB treatment. Adverse events were common, but only two patients (1%) permanently discontinued study drugs. Conclusions: Intermittent rifabutin-based therapy for HIV-related TB was well tolerated, but there was a high risk of treatment failure or relapse with acquired rifamycin resistance among patients with low CD4 lymphocyte counts.
Key Words: antiretroviral therapy • HIV • rifabutin • rifamycin resistance • tuberculosis In many parts of the world, tuberculosis (TB) is the most common opportunistic infection among persons with HIV disease (1–3). TB complicating HIV infection is associated with a 20 to 50% risk of death within the first year after diagnosis (4–6). Death soon after the diagnosis may be due to TB, but most deaths among patients with HIV-related TB are due to other complications of HIV (4, 5). Antiretroviral therapy markedly improves survival among patients with advanced HIV disease (7, 8) and may improve survival with HIV-related TB (9). However, the use of combination antiretroviral therapy during treatment of TB is complicated by overlapping side-effect profiles of anti-TB and antiretroviral drugs, complex drug–drug interactions between the rifamycins and many antiretroviral drugs, and immune reconstitution reactions after initiation of antiretroviral therapy (10). Rifampin markedly decreases serum concentrations of the HIV-1 protease inhibitors (11), and has substantial effects on concentrations of two of the available nonnucleoside reverse-transcriptase inhibitors (delavirdine and nevirapine) (12), thus limiting the choice of antiretroviral drugs. An alternate rifamycin antibiotic, rifabutin, has much less effect on serum concentrations of these antiretroviral drugs (12) and appears to be as potent as rifampin for the treatment of TB (13–15). Therefore, rifabutin-based therapy was recommended by the Centers for Disease Control and Prevention (CDC) as a way of avoiding serious drug–drug interactions and allowing use of potent antiretroviral therapy during TB treatment (16). However, there is relatively little published clinical experience with rifabutin-based therapy for HIV-related TB (15). The Tuberculosis Trials Consortium evaluated the efficacy and tolerability of a standard rifabutin-based regimen for treatment of HIV-related TB (TBTC Study 23). Preliminary results of this study were reported in the form of an abstract (17) and a special "Note to Readers" of the Morbidity and Mortality Weekly Report (18).
Patients 18 yr or older at participating sites in the United States and Canada with confirmed HIV infection and suspected or confirmed TB were eligible for enrollment. Patients were excluded if they had received treatment for active or latent TB within the previous 6 mo, were pregnant or breast-feeding, or if the initial mycobacterial cultures were negative or grew rifamycin-resistant Mycobacterium tuberculosis. The study was reviewed and approved by the institutional review boards of the CDC and each clinical site. All patients gave written, informed consent. Patients could be enrolled at any time during the first 2 mo of TB treatment. Patients who enrolled before the completion of the intensive phase of therapy (the first 2 mo) received rifabutin, isoniazid, pyrazinamide, and ethambutol, given daily for the first 2 wk and then either daily (5 d/wk), thrice-weekly, or twice-weekly based on the preference of the local TB control program. After 2 mo, all patients were treated with rifabutin (300 mg) plus isoniazid (15 mg/kg; maximum, 900 mg) given twice weekly. All TB treatment was given as directly observed therapy, defined as the ingestion of study medication under the observation of a member of the study team. Completion of study therapy was defined as observed ingestion of 36 twice-weekly continuation-phase doses, not by calendar time. The type and timing of initiation of antiretroviral therapy were not determined by the study protocol. Rifabutin doses were adjusted according to current guidelines if given with the HIV-1 protease inhibitors or efavirenz (19). TB treatment duration was 6 mo, with a mandatory extension to 9 mo for patients with a positive culture or lack of a clinical response at 2 mo of treatment (defined as persistent fever or progressive weight loss, and increase in size of lymph nodes, abscesses, or other tuberculous lesions) (16). Patients having an isoniazid-resistant isolate or who were intolerant of isoniazid were treated with 9 mo of rifabutin, pyrazinamide, and ethambutol; those having a pyrazinamide-resistant isolate or who were intolerant of pyrazinamide were treated with 9 mo of rifabutin plus isoniazid. CD4 lymphocyte count and HIV-RNA level (bDNA Bayer Versant 3.0; Bayer Corporation, Tarrytown, NY) were performed at enrollment, after 2 mo of TB therapy, at end of treatment, and every 6 mo thereafter. We attempted to obtain sputum specimens from all patients at 2 mo of treatment, regardless of the presence or absence of clinically evident pulmonary involvement. Patients with pulmonary involvement had sputum cultures obtained monthly during therapy and then at 1.5, 3, 6, 12, 18, and 24 mo after completing therapy. At all of these time points, patients who could not produce an expectorated sputum specimen underwent sputum induction. Mycobacterial cultures were performed using liquid plus solid media at certified local laboratories. Isolates obtained at diagnosis and at the time of treatment failure or relapse underwent confirmatory drug-susceptibility testing (using solid media and the proportion method), DNA fingerprinting (20, 21), and sequencing of the rpoB gene (22).
The primary endpoint of the study was culture-confirmed treatment failure (positive culture during treatment, at or after Month 4) or relapse (positive culture after completing treatment). The planned sample size was 215 patients, allowing for 80% power to detect whether the rate of treatment failure plus relapse was higher than 10%, twice that expected for treatment of drug-susceptible TB (
Data Analysis
A total of 180 patients were enrolled between December 1998 and March 2002 (Figure 1), 11 of whom were later excluded from the study (eight patients who did not have a positive mycobacterial culture and three patients who enrolled at sites that were closed soon after study initiation). The 169 remaining patients were used for the analyses that follow. Most patients were male and of Black or Hispanic race/ethnicity; 40% were born outside of the United States and Canada (Table 1). Nearly 45% of patients had extrapulmonary manifestations of TB; 22 (13%) had only extrapulmonary involvement. Most patients had advanced HIV disease at the time of TB diagnosis; the median CD4 cell count and HIV-RNA level were 90 cells/mm3 (interquartile range, 35–175) and 5.3 log10 copies/ml (interquartile range, 4.8–5.7), respectively.
A data and safety monitoring board reviewed outcome data three times. In March 2002, the board recommended that enrollment be stopped because five cases of acquired rifamycin resistance had occurred among patients with treatment failure or relapse. At the time enrollment was stopped, investigators were encouraged to use more frequent therapy (daily or thrice-weekly) and/or add medications such as ethambutol for patients having CD4 lymphocyte counts of less than 100 cells/mm3 at study entry. Patients were monitored for 2 yr after completing therapy.
TB Treatment
Treatment Failure and Relapse Eight of the nine isolates (89%) obtained at the time of treatment failure or relapse had acquired rifamycin resistance (three cases of treatment failure and five cases of relapse). The median time from enrollment to acquired rifamycin resistance was 8.3 mo (range, 4.7–24.5 mo). Multidrug resistance developed in one of these cases. A patient with an isolate having baseline pyrazinamide resistance developed recurrent suppurative cervical adenopathy at 5 mo of directly observed therapy. A percutaneous aspirate yielded an isolate with acquired rifamycin and isoniazid resistance. For all nine cases, the DNA fingerprint patterns of paired isolates at initial diagnosis and the time of treatment failure or relapse were identical. The rpoB gene sequences of the eight isolates having acquired rifamycin resistance had mutations previously associated with rifampin resistance (Ser531Leu [three cases], Gln513Leu [two cases], His526Tyr [two cases], His526Asp [one case]) (22). All patients with treatment failure or relapse responded to retreatment with a regimen active against rifamycin-resistant TB.
Factors Associated with Rifamycin-resistant Treatment Failure or Relapse
Use of Antiretroviral Therapy Thirty-eight (22%) patients were on antiretroviral therapy at the time of TB diagnosis, and 99 patients (59%) started antiretroviral therapy during TB treatment. Therefore, 81% of patients (137 of 169) received antiretroviral therapy during TB treatment. Patients not started on antiretroviral therapy during TB treatment had higher CD4 lymphocyte counts than those who received antiretroviral therapy during TB treatment (median CD4 lymphocyte count among those who did not start antiretroviral therapy, 149 cells/mm3 [interquartile range, 73–261] vs. 79 [interquartile range, 32–150]; p = 0.02). All patients who were treated with antiretroviral treatment received combination therapy; the most common regimens were two nucleoside analogs plus nelfinavir (47 patients) and two nucleoside analogs plus efavirenz (45 patients). CD4 lymphocyte counts rose substantially and HIV-RNA levels decreased during TB treatment (Figure 2). Twenty-six of the 169 patients (15%) were diagnosed with immune reconstitution reactions, all of which occurred among patients receiving antiretroviral therapy (26 of 137, 19%).
Adverse Events Of 169 patients in the study, 25 (14.8%) died and 25 (14.8%) developed a new AIDS-defining condition (24). None of the deaths were due to TB; 17 were due to complications of HIV disease (AIDS wasting, five patients; opportunistic infection, nine patients; HIV-related malignancy, three patients), three to illicit drug overdose, and five to other causes (suicide, cardiovascular disease, liver failure, non–HIV-related cancer). Although adverse events were common during TB treatment, few were believed to be due to the TB treatment regimen, and only two patients permanently stopped study drugs due to an adverse event (Table 4).
This study is the largest prospective evaluation of a rifabutin-based TB treatment regimen in conjunction with use of potent antiretroviral therapy. The rate of treatment failure or relapse observed in the study was approximately 5%, similar to that seen with standard rifampin-based TB therapy (5, 25), demonstrating the potency of rifabutin-based treatment for the treatment of HIV-related TB. The treatment regimen was also well tolerated, with only 1% of patients permanently discontinuing rifabutin due to an adverse event. However, eight of the nine patients who had treatment failure or relapse had isolates with acquired rifamycin resistance. Because rifamycins are the key to short-course TB treatment, the occurrence of acquired rifamycin resistance led to the early discontinuation of the study. One of the most important benefits of directly observed therapy for TB is that it prevents selective drug taking, and thereby almost completely prevents acquired drug resistance (14, 26, 27). However, this study demonstrates that this axiom of TB treatment is not true for patients with advanced HIV disease. It is also notable that the drug to which resistance was acquired in this and other studies (5, 25, 28, 29) was nearly always the rifamycin component of the regimen (rifampin, rifabutin, or rifapentine). Among patients with HIV-related TB treated with directly observed therapy, there are two consistent associations in this and previous studies with the occurrence of acquired rifamycin resistance: the presence of severe immunodeficiency (CD4 lymphocyte count < 100 cells/mm3) and the use of once- or twice-weekly therapy. Acquired rifamycin resistance has been reported with twice-weekly rifabutin plus isoniazid (4.7% in the present study) and with twice-weekly rifampin plus isoniazid (2.0, 3.7, and 1.7%) (5, 29, 30). In contrast, the rate of acquired rifamycin resistance among HIV-infected patients treated with daily (5–7 d/wk) rifampin-based therapy (29, 31, 32), or rifabutin-based therapy (29, 31) has been very low. The interpretation of these data from different clinical trials and observational cohorts is difficult. We conclude that there is a risk of acquired rifamycin resistance with rifabutin or rifampin when given twice weekly, but whether the risk is higher with one of these agents is not clear. In addition to the associations with severity of immunosuppression and dosing frequency, abnormal pharmacokinetics of anti-TB drugs may also be a part of the pathophysiology of acquired rifamycin resistance. Among patients in this trial who enrolled in a pharmacokinetic substudy, patients with low concentrations of rifabutin were at increased risk of acquired rifamycin resistance, even after adjusting for baseline CD4 cell count and dosing frequency (33). Thus, it appears that three factors are closely associated with acquired rifamycin resistance: severe immunodeficiency, highly intermittent dosing, and low concentrations of the rifamycin component of the regimen. These associations suggest that, in the relative absence of immune control, low concentrations of the rifamycin near the end of the prolonged dosing interval allow mycobacterial replication and selection for rifamycin resistance. The risk of acquired rifamycin resistance in our study was lower among patients who received antiretroviral therapy during the first 2 mo of TB treatment. The interpretation of this finding is difficult: the study could not capture the many factors that led to differences in the timing of antiretroviral therapy, and this association has not been evaluated in previous studies. Although our sample size is relatively small, these data support the hypothesis that immune reconstitution is beneficial in the treatment of HIV-associated TB. The question of the optimal timing of antiretroviral therapy during TB treatment is complex and is currently being evaluated in randomized clinical trials. The TB treatment regimen in this study was chosen to determine if TB treatment could be used safely with potent antiretroviral therapy. In that respect, the study regimen succeeded; most patients received and benefited from potent antiretroviral therapy during TB therapy. Baseline characteristics of patients in this trial were quite similar to those of patients enrolled in a trial that was performed in many of the same study sites, before the availability of potent antiretroviral therapy (Table 5). The markedly lower rates of death (4.9 vs. 20.0%) and new opportunistic illness during the first year after TB diagnosis in the present study suggest that the benefits of antiretroviral therapy can be extended to patients being treated for TB. This is an important finding because TB is the most common opportunistic infection in areas of the world with high rates of HIV infection, areas in which combination antiretroviral therapy is now being introduced. Rifabutin has only been approved in a relatively small number of resource-rich countries (e.g., United States and western Europe). This limits the direct application of our results to many countries with high rates of HIV-related TB in which rifampin is the only available rifamycin antibiotic. However, some potent antiretroviral regimens (e.g., efavirenz or ritonavir with two nucleoside analogs) can be used with rifampin-based TB therapy (11, 34), and should allow patients to obtain the benefits of potent antiretroviral therapy given during TB treatment.
The major limitation of this study is the lack of a randomized comparison group. Although use of twice-weekly therapy was associated with acquired rifamycin resistance among patients with low baseline CD4 lymphocyte counts, our study did not include a group randomized to more frequent dosing. Evaluating the relationship between dosing frequency and the success of treatment of HIV-related TB should be a high priority for future randomized studies. Although this is one of the largest prospective studies of treatment of HIV-related TB, the sample size limits the power of subgroup comparisons. Finally, the substantial proportion of patients lost after the first year of follow-up, but before completing 2 yr of follow-up, is a limitation of our study. A largely twice-weekly rifabutin-based regimen for HIV-related TB was associated with an unacceptable risk of acquired rifamycin resistance. The results of this and other studies were the basis for national recommendations that persons with CD4 lymphocyte counts of less than 100 cells/mm3 should not be treated with twice-weekly therapy, at least during the first 2 mo of therapy (18, 35). Patients with baseline CD4 lymphocyte counts more than 100 cells/mm3 did well in this study, but additional experience is needed to evaluate intermittent therapy in this important subgroup. Intermittent therapy greatly facilitates directly observed therapy, so randomized trials of dosing frequency should be a high priority among patients with HIV-related TB. Finally, our study supports the recommendation to use antiretroviral therapy among patients with advanced HIV disease and TB (16), as such patients can obtain dramatic benefits from potent antiretroviral therapy.
The authors thank the many patients who contributed to the success of this trial. They also thank Drs. Kenneth Castro, Rick O'Brien, and M. Elsa Villarino for continued support within CDC, and local TB program staff, who assisted in the clinical management of the participants. Bayer Corporation provided test kits (Versant HIV-1 RNA 3.0 assay [bDNA]) for the HIV viral load assessments. Participating clinical sites (principal investigators and study coordinators, with numbers of patients enrolled in brackets) were as follows: Los Angeles County/University of Southern California Medical Center (Brenda Jones, M.D., Claudia Silva, R.N. [29]); University of North Texas Health Science Center (Stephen Weis, D.O., Barbara King, R.N., Norma Shaffer, L.V.N. [15]); Houston VA Medical Center (Christopher Lahart, M.D., Ruby Nickson, R.N., Terry Scott, R.N., Robert Awe, M.D. [14]); Audie L. Murphy VA Hospital, San Antonio (Marc Weiner, M.D., Melissa Engle, R.N. [13]); Johns Hopkins University School of Medicine (Richard Chaisson, M.D., Tim Sterling, M.D., Kristina Moore, R.N., Judith Hackman, R.N. [13]); Carolinas Medical Center (James Horton, M.D., Beth Quinn, R.N. [12]); Nashville VA Medical Center (Douglas Kernodle, M.D., Anthony Chapdelaine, M.D., Guat-Siew McKee, M.D., Linda Reeves-Hammock, R.N. [11]); New Jersey Medical School National TB Center–University of Medicine and Dentistry of New Jersey (Lee B. Reichman, M.D., M.P.H., Bonita T. Mangura, M.D., Marilyn Owens, R.N., Cora Leus, R.N. [8]); Bellevue Hospital Center, New York University (Neil Schluger, M.D., William Rom, M.D., Rany Condos, M.D., Laurie Sandman, R.N. [7]); University of California at San Francisco Medical Center (Charles Daley, M.D., Llewellyn Stanton, R.N. [7]); University of British Columbia Health Center (Mark Fitzgerald, M.B., Eduardo Hernandez, M.D., Banafsheh Peyvandi, M.D. [7]); Harlem Hospital Center (Wafaa El-Sadr, M.D., Frantz Medard, M.D., Mary Klein, R.N. [6]); Denver Health and Hospitals (Randall Reves, M.D., William Burman, M.D., Jan Tapy, R.N. [5]); Chicago VA Medical Center–Lakeside (Mondira Bhattacharya, M.D., Julie Fabre, R.N. [5]); University of Manitoba (Earl Hershfield, M.D., Gerry Izon, R.N. [4]); Columbia University College of Physicians and Surgeons and the New York City Department of Health (Neil Schluger, M.D., Joseph Burzinski, M.D., Sonal Munsiff, M.D., Vilma Lozano, R.N. [4]); Hines VA Medical Center and Suburban Cook County TB District, Illinois (Constance Pachuki, M.D., Mary Samuel, R.N., James Gallai, M.D. [3]); Seattle King County Health Department (Charles Nolan, M.D., Stefan Goldberg, M.D., Marcia Stone, R.N., Debra Schwartz, R.N. [2]); VA Medical Center–Washington DC (Fred Gordin, M.D., Debra Benator, M.D., Donna Sepulveda Conwell, R.N., Thomas Walsh, M.D., Margaret Lankford, R.N., Charlotte Quinlan-Mauzy, R.N. [2]); and Duke University Medical Center and Durham VA Medical Center (Carol Dukes Hamilton, M.D., Ann Mosher, R.N. [2]). Trial administration and data management were performed at CDC by Sharlene Broadnax, Crystal Carter, Kenneth Dansbury, Melissa Fagley, Lon Gross, Constance Henderson, Margaret Jackson, Tashawnya Rainey. Microbiologic data collection was supervised by Lorna Bozeman, M.S. CDC laboratory testing was performed or supervised by Robert Cooksey, Ph.D., Jack Crawford, Ph.D., Lois Diem, Beverly Metchock, Dr.P.H., Glenn Morlock, David Sikes, Lauren Steinlein, David Temporado, and Sean Toney. Data and safety monitoring were performed by Harold Jaffe, M.D., James Neaton, Ph.D., and John Bass, M.D.
Supported by the U.S. Centers for Disease Control and Prevention. Originally Published in Press as DOI: 10.1164/rccm.200503-417OC on August 18, 2005 Conflict of Interest Statement: W.B. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. D.B. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. A.V. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. A.K. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. B.J. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. C.S. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. C.L. is employed by Glead Sciences, Inc., since August 2004, and owns 200 shares of Glead Common Stock in an employee stock purchase plan. S.W. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. B.K. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. B.M. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. M.W. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. W.E.-S. does not have a financial relationship with a commercial entity that has an interest in the subject matter of this manuscript. Received in original form March 16, 2005; accepted in final form August 16, 2005
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
5%). 
2, and continuous variables with the Wilcoxon two-sample test. To identify factors associated with acquired rifamycin resistance, we compared patients having acquired rifamycin resistance to all other patients in the study. All patients in the study completed 8 wk of four-drug therapy, but there were differences in the amount of that therapy that was administered twice-weekly. For the analysis of risk factors for acquired rifamycin resistance, patients who received four or more doses of their treatment as twice-weekly dosing during the first 2 mo were defined as having received twice-weekly therapy during induction phase. Because of the small number of cases of treatment failure or relapse it was not appropriate to perform multivariate analysis (
