|
|
| |||||||||||||||||||||||||||||||
Background Both didanosine and zalcitabine are commonly used to treat patients with human immunodeficiency virus (HIV) infection who cannot tolerate zidovudine treatment or who have had disease progression despite it. The relative efficacy and safety of these second-line therapies are not well defined.
Methods In this multicenter, open-label trial we randomly assigned 467 patients who previously received zidovudine and had 300 or fewer CD4 cells per cubic millimeter or a diagnosis of the acquired immunodeficiency syndrome (AIDS) to treatment with either didanosine (500 mg per day) or zalcitabine (2.25 mg per day).
Results After a median follow-up of 16 months, disease progression or death occurred in 157 of 230 patients assigned to didanosine and 152 of 237 patients assigned to zalcitabine, for a relative risk of 0.93 for the zalcitabine group as compared with the didanosine group (P = 0.56), which decreased to 0.84 (P = 0.15) after adjustment for the CD4 count, Karnofsky score, and presence of AIDS at base line. There were 100 deaths in the didanosine group and 88 in the zalcitabine group, for a relative risk of 0.78 (P = 0.09) and an adjusted relative risk of 0.63 (P = 0.003).
A majority of patients in each group (66 percent) had at least one adverse event during treatment (153 patients taking didanosine and 157 taking zalcitabine). Peripheral neuropathy and stomatitis occurred more often with zalcitabine and diarrhea and abdominal pain more frequently with didanosine.
Conclusions For patients with HIV infection who have not responded to treatment with zidovudine, zalcitabine is at least as efficacious as didanosine in delaying disease progression and death.
Patients in whom zidovudine treatment had failed or intolerance of zidovudine had developed were enrolled in a multicenter, randomized clinical trial comparing the efficacy and safety of didanosine and zalcitabine, initiated in December 1990 by the Terry Beirn Community Programs for Clinical Research on AIDS (the acquired immunodeficiency syndrome). This report compares survival, clinical disease progression, changes in the CD4 lymphocyte count, and adverse effects during the two treatments.
Methods
Study Population
Patients were enrolled at 133 clinical sites in 16 units participating in the Community Programs for Clinical Research on AIDS (CPCRA). The CPCRA is sponsored by the National Institute of Allergy and Infectious Diseases and is a consortium of administrative units in 13 American cities, established to conduct clinical trials in the community where patients with HIV infection receive their primary care.
Patients with HIV infection who were at least 13 years of age (except women who were pregnant or breast-feeding) were eligible if they met the following criteria: they had an AIDS-defining condition or two CD4 lymphocyte counts of 300 cells or less per cubic millimeter, with either a positive serologic test for HIV or a clinician's working diagnosis of HIV infection; and they had undergone zidovudine therapy that led to intolerance of the drug or progression of disease during therapy.
Zidovudine intolerance was defined as hematologic if a patient treated with 600 mg or less daily for more than eight weeks had two neutrophil counts of 750 cells or less per cubic millimeter or two hemoglobin values of 9.0 g per deciliter or less. Nonhematologic intolerance was indicated by any of the following features: a syndrome of headache and agitation, nausea and vomiting causing severe discomfort despite appropriate therapy with antiemetic drugs, symptoms of zidovudine myopathy, or other, persistent severe symptoms directly attributable to zidovudine therapy and recurring on rechallenge.
Zidovudine treatment was considered to have failed if patients who had received a daily dose of 500 mg or more for a cumulative duration of at least six months had disease progression, defined as the development of a new or recurrent opportunistic infection, the development of a new or recurrent opportunistic neoplasm (excluding Kaposi's sarcoma), the development of stage 1 (mild) AIDS dementia complex, or a failure to thrive despite adequate therapy. Failure to thrive was indicated by an involuntary weight loss of more than 10 percent of the maximal weight during zidovudine therapy, accompanied by either diarrhea due to an undiagnosed cause or documented fever and chronic weakness lasting more than 30 days in the absence of a concurrent illness.
Patients were required to have the following values at entry: a hemoglobin level above 9.0 g per deciliter (with or without support from transfusions or epoetin), an absolute neutrophil count above 750 per cubic millimeter, a platelet count above 50,000 per cubic millimeter, a serum creatinine concentration below 2.0 mg per deciliter (180 µmol per liter), a total bilirubin concentration below 2.0 mg per deciliter (34 µmol per liter), a uric acid concentration below 10 mg per deciliter (590 µmol per liter), hepatic aminotransferase levels less than five times the upper limit of normal, and a total serum amylase level equal to or less than twice the upper limit of normal.
Patients were excluded if they had any of the following features: previous treatment with didanosine or zalcitabine, a history of pancreatitis, recent excessive alcohol use, a history or symptoms of peripheral neuropathy, an uncontrolled seizure disorder, AIDS dementia complex of stage 2 (moderate) or worse, contraindications to the study drugs, or current treatment for any active AIDS-defining opportunistic infection or neoplasm.
All patients gave written informed consent for participation in the study.
Study Design
This study was designed as a multicenter, randomized, open-label, community-based clinical trial. A total of 400 patients were to be recruited in 1 year and follow-up was planned to last at least 12 months, to ensure that the study had a power of 80 percent to detect a difference in the hazard rates for events indicating clinical disease progression (including death) of 92 and 64 events per 100 patient-years in each treatment group, with an alpha level of 0.05 (two-tailed). After their eligibility was confirmed, the patients were randomly assigned to the treatment groups by a central statistical office using a permuted-block design; they were then stratified according to their study unit and whether zidovudine treatment had failed or led to intolerance of this drug.
The primary objective of the study was to determine which dideoxynucleoside monotherapy was better in patients in whom zidovudine treatment had failed or intolerance of the drug had developed. Both delay of disease progression and drug tolerance were taken into account in the determination of which treatment was superior. Clinical disease progression was defined as the first occurrence of an AIDS-defining condition according to a description adapted from the 1987 classification of the Centers for Disease Control and Prevention7 or as a recurrence of any of the following: Pneumocystis carinii pneumonia, esophageal candidiasis, herpes simplex infection, disseminated herpes zoster, and septicemia due to nontyphoidal salmonella. Events that might indicate disease progression were reviewed by a clinical-events committee and, if judged to be confirmed or probable, were considered end points for the study.
Adverse events were graded on a five-point scale (I through V) by study clinicians while the patients were taking the study medication and during the three months after they had discontinued it. An adverse event was reportable if it was at least possibly related to the study drug and its grade was III or IV. If the grade of the event was II and the event was at least possibly related to the study drug, it was also reportable if the dose of drug was decreased or its administration was interrupted by the investigator.
Treatment Regimens, Evaluation of Patients, and Follow-up
Didanosine (Videx, Bristol-Myers Squibb, Wallingford, Conn.) was provided in sachets containing a buffer-powder blend to protect the drug from degradation by gastric acid. Patients weighing more than 45 kg were initially given 250 mg of didanosine orally twice a day. Zalcitabine (Hivid, Hoffmann-LaRoche, Nutley, N.J.) was provided in tablets that each contained 0.375 mg of drug. Patients weighing 40 kg or more were initially given 0.75 mg of zalcitabine orally three times a day.
Prophylaxis against P. carinii pneumonia was recommended for all patients with absolute CD4 lymphocyte counts of less than 200 cells per cubic millimeter, the decision being left to the discretion of the clinician8.
After base-line evaluation, routine follow-up visits were required twice during the first month, once during the second month, and every other month thereafter. At each visit a symptom-severity checklist was completed, a blood sample was obtained to monitor hematologic and biochemical values, and compliance was assessed as good, fair, or poor by a clinician interviewing the patient. Absolute CD4 cell counts were determined at 2, 6, 12, and 18 months.
Each patient was permitted to switch from one study drug to the other once during the course of the trial. If drug intolerance developed at any time or disease progressed anew after 12 weeks of treatment, the patient could switch to the alternate study drug after a washout period of at least 3 days or could continue to take the drug originally assigned unless drug intolerance was severe.
Data Management and Statistical Analysis
All 16 study units were monitored by external investigators at least every three months. The monitors determined whether findings in clinical source documents corresponded to those in case-report forms, especially in regard to conformity with the criteria for eligibility, reporting of adverse events, and clinical end points.
During the study all investigators were blinded to end-point events. Interim findings were monitored on five occasions by an independent data and safety monitoring board following prespecified guidelines for stopping the study. At the last review, August 1992, the board recommended that the study end as originally planned, on September 20, 1992, one year after the last patient had been assigned to treatment.
The treatment groups were compared for survival and clinical disease progression according to the patients' original treatment assignment (intention-to-treat analysis). All analyses, unless otherwise stated, were stratified according to study unit and status for treatment failure of zidovudine or intolerance of zidovudine, in keeping with the method of randomization.
Comparisons of the treatment groups for survival, clinical disease progression, discontinuation of study treatment, and adverse events were based on the use of time-to-event methods, including Kaplan-Meier estimation, stratified log-rank tests, and proportional-hazards regression models stratified as for randomization9,10. Data on adverse events in patients who switched to an alternate drug were censored with respect to the time that the original drug was permanently discontinued. Longitudinal measurements of the CD4 lymphocyte count and other hematologic variables, biochemical values, body weight, and Karnofsky performance scores were summarized with use of a random-effects model described by Laird and Ware11.
Proportional-hazards analyses of disease progression and death included both unadjusted and adjusted tests for significance and estimates of relative risk. These adjustments were stratified according to study unit and controlled for imbalance of the treatment groups in the following prognostic covariates at base line: previous AIDS-defining illness, CD4 lymphocyte count, and Karnofsky performance score. Relative risks and their 95 percent confidence intervals were computed from adjusted and unadjusted proportional-hazards regression models (the risk with zalcitabine relative to the risk with didanosine). Event rates are reported as the numbers of events per 100 patient-years at risk.
Results
Study Population
The 16 study units enrolled 467 patients from December 1990 through September 1991. Of the total enrolled, 6 percent did not meet the criteria for entry; most of these patients had minor violations of the limits for laboratory values, but eight patients (1.7 percent) were later found not to have met the exact criteria for either failure of zidovudine treatment or intolerance to this drug. Analyses were based on all 467 patients enrolled; the conclusions of these analyses did not differ from those of analyses that excluded the patients determined to be ineligible after randomization.
The base-line characteristics of the patients are shown in Table 1. Almost all patients initially received the full dose of 250 mg of didanosine twice a day or 0.75 mg of zalcitabine three times a day. Three patients assigned to didanosine whose weight was low at base line received a lower dose (167 mg), and one patient assigned to zalcitabine initially received half the full dose.
|
Of the patients alive and being followed at the end of the trial, 41 (35 percent) assigned to didanosine and 55 (42 percent) assigned to zalcitabine were still receiving their initially assigned drug. Six months after randomization, 33 percent of patients assigned to didanosine and 33 percent of those assigned to zalcitabine had permanently discontinued taking their assigned drug; 12 months after randomization, these figures were 62 percent and 55 percent, respectively. The reasons for discontinuation included death during treatment with the assigned drug (14 patients assigned to didanosine and 14 assigned to zalcitabine), toxic effects (94 assigned to didanosine and 88 assigned to zalcitabine), disease progression (18 assigned to didanosine and 13 assigned to zalcitabine), and a decision by the patient or clinician to withdraw (28 assigned to didanosine and 34 assigned to zalcitabine).
One hundred twenty-one patients assigned to didanosine and 102 assigned to zalcitabine became eligible to switch to the other study drug because of drug intolerance. Of these, 43 patients actually switched from didanosine to zalcitabine and 26 from zalcitabine to didanosine. Twelve months after randomization, 12.4 percent of patients assigned to didanosine and 18.8 percent of those assigned to zalcitabine had reinitiated zidovudine treatment.
Disease Progression
A total of 459 nonfatal events indicating disease progression were reported during follow-up. Of these, 251 were considered confirmed events (114 events among patients taking didanosine and 137 among those taking zalcitabine) and 112 were considered probable events (58 events among patients taking didanosine and 54 among those taking zalcitabine) by the clinical-events committee. Clinical disease progression or death occurred in 157 patients taking didanosine (rate, 93.3 per 100 patient-years) and 152 taking zalcitabine (rate, 87.7 per 100 patient-years) (relative risk, 0.93; 95 percent confidence interval, 0.74 to 1.18; P = 0.56). The rates of disease progression are shown as Kaplan-Meier curves in Figure 1.
|
|
|
Table 3 summarizes both the types of first events indicating disease progression, including death, and the total numbers of events in the treatment groups. Of the 230 patients assigned to didanosine, 157 had 272 events; of the 237 patients assigned to zalcitabine, 152 had 279 events. The development of disseminated Mycobacterium avium complex infection was the most common nonfatal first event in both treatment groups. P. carinii pneumonia, esophageal candidiasis, and cytomegalovirus disease were the next most frequent opportunistic infections among events indicating disease progression.
|
Adverse Events during Treatment
At least one adverse event occurred in 153 patients assigned to didanosine and 157 assigned to zalcitabine, before they discontinued their initial study drug (relative risk, 0.92; P = 0.47). Half these patients had two or more adverse events.
Although the numbers of patients with at least one adverse event and the rates of events in the didanosine and zalcitabine groups were similar, the types of events were different (Table 4). Peripheral neuropathy was significantly more frequent in the zalcitabine group (69 patients, vs. 33 patients in the didanosine group; P = 0.002). Stomatitis occurred only in the zalcitabine group (eight patients), and pancreatitis only in the didanosine group (four patients). Diarrhea and abdominal pain occurred significantly more often in the didanosine group (P<0.001 and P = 0.03, respectively).
|
Our randomized, open-label trial was a direct comparison of didanosine with zalcitabine as monotherapy in patients who could no longer tolerate zidovudine or whose disease progressed during zidovudine treatment. As more patients start to receive antiretroviral therapy earlier in the course of HIV infection, clinicians will more often face the problem of how best to treat such patients. Didanosine was initially approved for the treatment of this population12. Zalcitabine is currently approved for use only in combination with zidovudine in patients with CD4 lymphocyte counts below 300 per cubic millimeter who have evidence of disease progression despite zidovudine therapy13. As yet, there is no officially approved indication for zalcitabine monotherapy. Combining zalcitabine or didanosine with zidovudine is not an option in patients with a marked intolerance of zidovudine.
In this study, zalcitabine was found to be as efficacious as didanosine in delaying disease progression, including death; in addition, zalcitabine may have provided a survival advantage. This finding suggests that zalcitabine can be used as an alternative second-line therapy. However, during the median follow-up period of 16 months, the majority of patients in both treatment groups had disease progression (66 percent) and at least one adverse event during treatment (66 percent), and more than one third of the patients in each group died. Thus, neither treatment offered substantial long-term benefit.
Recent studies of didanosine given in various doses as salvage monotherapy in a similar study population have produced results similar overall to the findings of this trial. The European/Australian Alpha Trial compared survival in 1775 patients with zidovudine intolerance who were assigned to one of two doses of didanosine (750 mg and 200 mg per day)14. Half the patients enrolled had AIDS. The median survival of patients in the two dosage groups was virtually the same, with a much lower rate of pancreatitis and peripheral neuropathy in the group taking the lower dose. The AIDS Clinical Trials Group of the National Institute of Allergy and Infectious Diseases recently concluded a study that compared doses of 200, 500, and 750 mg of didanosine given daily to patients with zidovudine intolerance15. Of the 650 patients enrolled, 50 percent had an AIDS-defining condition, and the median CD4 cell count was 33 cells per cubic millimeter. Overall, the death rate in that study was high and was similar to the rate in our study (Allan JD, et al.: personal communication).
In the preliminary results of a program giving expanded access to zalcitabine, patients with zidovudine intolerance or disease progression during zidovudine treatment were eligible to receive zalcitabine in a dose of 0.75 mg three times a day or half this dose. Of 3979 patients enrolled, 1965 took the standard dose and 2014 the low dose; there were no significant differences between these dosage groups in the time to the first opportunistic infection or in survival16.
Different doses of either didanosine or zalcitabine seem to differ little in their effect on the progression of disease or survival in patients with advanced HIV infection, but they may have different toxic effects. Thus, the toxicity profiles of these two drugs should be an important consideration in choosing antiretroviral therapy for an individual patient. Patients with advanced disease appear to have a clinical course that may not be markedly altered by any of the currently available nucleoside analogues used as monotherapy, although this cannot be definitively stated in the absence of a group receiving no treatment. Other studies evaluating concurrent combination treatments or alternating sequential treatments in a population with advanced HIV disease are ongoing and may yield better results.
We are indebted to Charles Kaplan, M.D., for his efforts in the design of the study protocol, to Rosemary Bell for her assistance in the preparation of the manuscript, and to our collaborators at the following participating study units: Research and Education Group -- Doug Beers, M.D., Karen Loveless, R.N., and Norma Martinez, R.N.; Denver Community Program for Clinical Research on AIDS -- Carol A. Mesard, R.N., Randall R. Reves, M.D., and Jack A. Rouff, M.B.A.; Chicago Community Programs for Clinical Research on AIDS -- Thomas Klein, M.D., Richard McDonald, M.D., and Ross Slotten, M.D.; Community Consortium of San Francisco -- Carol Brosgart, M.D., Lawrence Drew, M.D., and Jeffrey Fessel, M.D.; AIDS Research Consortium of Atlanta -- Richard Hudson, D.O., Carlos Lopez, M.D., and Sumner Thompson, M.D.; Henry Ford Hospital -- Jones O. Kumi, M.D., Diane Mastro-Polak, R.N., and Louis D. Saravolatz, M.D.; Richmond AIDS Consortium -- Thomas M. Kerkering, M.D., Robert A. Higginson, P.A.-C., and Lawrence Gernon, M.D.; Bronx Lebanon Hospital Center -- Mordechi Bar, M.D., Douglas Mendez, M.D., and Jerome Ernst, M.D.; Washington Regional AIDS Program -- John Scott, R.N.C., Betsy Finley, R.N., and Douglas Ward, M.D.; Louisiana Community AIDS Research Program -- Ted Wisniewski, M.D., William Brandon, M.D., and Janice Walker, R.N.; Comprehensive AIDS Alliance of Detroit -- Colleen Calhoun, R.N., Jack R. Ebright, M.D., and Paula Schuman, M.D.; Delaware Community Program for Clinical Research on AIDS -- Arlene Bincsik, R.N., Karen Swanson, R.N., and Dean Winslow, M.D.; North Jersey Community Research Initiative -- Ann Sheridan, R.N., Victoria Taylor, R.N., and George Perez, M.D.; Harlem AIDS Treatment Group -- Wafaa El-Sadr, M.D., Cheryl Guity, R.N., and Josephine Juleau, P.A.; Clinical Director's Network of Region II/Community Program for Clinical Research on AIDS -- Steven Salvati, M.D., Colleen Dowling, R.N., and Graciela I. Salvador, M.D.; and Hill Health Corporation -- A.J. Burns, P.A.-C., and Si-hoi Lam, M.D.
Source Information
From the San Francisco Community Consortium on AIDS, University of California, San Francisco (D.I.A., A.H.); the Division of Biostatistics, University of Minnesota, Minneapolis (A.I.G., C.L., J.D.N.); the Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Md. (J.A.K., L.D.); the Comprehensive AIDS Alliance of Detroit, Harper Hospital, Detroit (L.R.C.); Denver Community Programs for Clinical Research on AIDS, Denver Public Health, Denver (M.G., D.L.C.); Chicago Community Programs for Clinical Research on AIDS, Chicago (S.W.); R.O.W. Sciences, Rockville, Md. (K.M.); Pharmaceutical Product Development, Wilmington, N.C. (S.K.); the Department of Medicine, St. Joseph Hospital, Chicago (R.L.-H.); the Division of Infectious Diseases, Henry Ford Hospital, Detroit (N.M.); the Research and Education Group, Portland, Oreg. (J.H.S.); and AIDS Research Consortium of Atlanta, Atlanta (M.T.).
Address reprint requests to Dr. Deyton at the National Institute of Allergy and Infectious Diseases, National Institutes of Health, 6003 Executive Blvd., Bethesda, MD 20892.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
