Alternating and Intermittent Regimens of Zidovudine and Dideoxycytidine in Patients with AIDS or AIDS-Related Complex

  1. Gail Skowron, MD;
  2. Samuel A. Bozzette, MD;
  3. Lynette Lim, PhD;
  4. Carla B. Pettinelli, MD, PhD;
  5. Herbert H. Schaumburg, MD;
  6. Joseph Arezzo, PhD;
  7. Margaret A. Fischl, MD;
  8. William G. Powderly, MD;
  9. David J. Gocke, MD;
  10. Douglas D. Richman, MD;
  11. John C. Pottage, MD;
  12. Diana Antoniskis, MD;
  13. George F. McKinley, MD;
  14. Newton E. Hyslop, MD;
  15. Graham Ray, BSN;
  16. Gary Simon, MD, PhD;
  17. Nancy Reed, MSN, CANP;
  18. Marsha L. LoFaro, MA;
  19. Raj B. Uttamchandani, MD;
  20. Lawrence D. Gelb, MD;
  21. Steven J. Sperber, MD;
  22. Robert L. Murphy, MD;
  23. John M. Leedom, MD;
  24. Michael H. Grieco, MD;
  25. James Zachary, MD;
  26. Martin S. Hirsch, MD;
  27. Stephen A. Spector, MD;
  28. Joseph Bigley, MS;
  29. Whaijen Soo, MD, PhD; and
  30. Thomas C. Merigan, MD
  1. Requests for Reprints: Gail Skowron, MD, Division of Infectious Diseases, Roger Williams Hospital, 825 Chalkstone Avenue, Providence, RI 02908. Acknowledgments: The authors thank the patients who participated in this study; Dennis Dixon, PhD, and Susan Ellenberg, PhD, for their review of the manuscript; Lori Brashears, RN, Janet Austen Herman, MBA, Elizabeth Hawkins, AA, Mary Culnane, MS, Bernard Landry, MPH, and Pamela Clax, DPM, in the AIDS Clinical Trials Group Operations Office for administrative support; and Susan Auclair, Bill Jesdale, and Lucile Lopez for assistance with manuscript preparation. Grant Support: In part by the AIDS Clinical Trials Group of the National Institute of Allergy and Infectious Diseases.
     
    Next Section

    Abstract

    Objective: To determine whether alternating regimens consisting of zidovudine and 2′,3′-dideoxycytidine (ddC) reduce the toxicity and maintain or increase the antiretroviral effect associated with each drug alone.

    Design: An unblinded, randomized (phase II) clinical trial in which seven treatment regimens were compared.

    Setting: Outpatient clinics of 12 AIDS Clinical Trials Units.

    Patients: One hundred thirty-one patients with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex and serum p24 antigenemia (≥ 70 pg/mL).

    Intervention: Treatments included weekly or monthly alternating zidovudine (200 mg every 4 hours) and ddC (0.01 or 0.03 mg/kg body weight every 4 hours); weekly intermittent zidovudine, 200 mg every 4 hours, or ddC, 0.03 mg/kg every 4 hours; and continuous zidovudine.

    Measurements: Toxicity, CD4 cell counts, serum p24 antigen levels, and clinical end points. Data were analyzed for the first 48 weeks of therapy (median follow-up, 40 weeks).

    Results: Hematologic toxicity was significantly less frequent in patients who received zidovudine therapy every other week (11% to 15%) or every other month (11% to 14%) than in those who received continuous zidovudine therapy (33%) (P < 0.02). Weekly alternating therapy with zidovudine and ddC, 0.03 mg/kg, or intermittent therapy with ddC, 0.03 mg/kg, produced high rates of peripheral neuropathy (41% and 50%, respectively). Neuropathy occurred in 10% to 21% of patients in the other three alternating-therapy limbs and in 17% of patients receiving zidovudine alone (intermittently or continuously). Initial increases in CD4 cell counts were sustained in three alternating-therapy limbs, but counts returned to baseline by week 28 in the remaining limbs. The median weight gain at week 48 was significantly greater in patients treated with alternating regimens (0.9 to 3.8 kg) compared with those treated with continuous zidovudine therapy ( −0.7 kg) (P = 0.008). Patients treated with alternating regimens and those treated with continuous zidovudine had similarly sustained decreases in p24 antigen levels.

    Conclusions: These findings suggest that alternating therapy with zidovudine and ddC reduces the toxicity associated with each drug alone while maintaining strong antiretroviral activity.

    Zidovudine is a nucleoside analog that reduces progression of human immunodeficiency virus type 1 (HIV)-associated disease and prolongs survival [1-3]. The use of this drug at high doses in patients with advanced disease has been limited by hematologic toxicity [4-6]. Initial studies of another nucleoside analog, 2′,3′-dideoxycytidine (ddC), have shown that it is a potent inhibitor of HIV replication; therapy with ddC results in both the rapid suppression of serum p24 antigen levels and increases in CD4 cell counts [7, 8]. The principal dose-limiting toxicity of ddC is sensory peripheral neuropathy, which is related to both dose and duration of therapy [7-10]. Both zidovudine and ddC have been studied at lower doses to construct a tolerable monotherapy regimen [2, 6, 9, 11].

    Alternating courses of zidovudine and ddC have been studied in an attempt to reduce dose-limiting toxicity. By limiting the continuous administration of each drug, the use of alternating therapy may allow any short-term toxicity from one drug to resolve during the administration of the other, and cumulative toxicity may be avoided. In addition to a reduction in toxicity, several other benefits of alternating therapy have been suggested. First, such combination therapy may have a synergistic or additive effect on virologic and immunologic markers; preliminary studies in vitro and in vivo suggest that alternating regimens have an enhanced antiretroviral effect [7, 12]. Concurrent zidovudine and ddC therapy has shown similar in vitro and in vivo effects and has recently been approved for use by the Food and Drug Administration [13, 14]. Second, it may be possible to maintain continuous high-dose therapy, thereby maximizing central nervous system concentrations. Third, in limiting the time of exposure to each drug, the sequential use of two agents may reduce or retard the emergence of drug-resistant isolates.

    We evaluated monthly and weekly alternating schedules of zidovudine and ddC to determine whether a reduction in toxicity or an enhanced antiretroviral effect, or both, could be achieved by this method of administration of the two agents. Intermittent zidovudine and ddC limbs were included to assess the relative contribution of each of the components to the overall effect. Continuous zidovudine, 1200 mg/d, served as the control treatment regimen.

    Previous SectionNext Section

    Methods

    Patients

    The study sample consisted of patients with AIDS-related complex or AIDS. We defined AIDS-related complex as the documented presence of one of the following: weight loss exceeding 15 pounds or 10% of body weight within 120 days before entry; a temperature greater than 38.5 °C that persisted for more than 14 consecutive days or was documented on more than 15 days in a 30-day interval; diarrhea (three or more liquid stools per day) for at least 30 days without a definable cause; recurrent oral candidiasis; hairy leukoplakia; or a history of herpes zoster. We defined AIDS according to Centers for Disease Control criteria [15]. All study patients had circulating serum p24 antigen levels of 70 pg/mL or more on two occasions before enrollment. No restriction was placed on entry CD4 cell count.

    Patients were excluded from the study if they met any of the following criteria: a hemoglobin concentration of less than 95 g/L; transfusion dependence; an absolute granulocyte count less than 1.2 × 109/L; a platelet count less than 100 × 109/L; a calculated creatinine clearance rate of 50 mL/min per 1.73 m2 body surface area or less; transaminase levels 5 or more times the upper limit of normal; a Karnofsky performance score of less than 60; pregnancy or lactation; significant malabsorption; cardiac or liver disease; or AIDS-defining conditions requiring systemic maintenance chemotherapy. Patients with evidence of preexisting peripheral neuropathy and those who had had previous ddC treatment were also excluded. We encouraged all patients with less than 200 CD4 cells/mm3 to undergo aerosolized pentamidine prophylaxis (300 mg every month) for Pneumocystis carinii pneumonia.

    Previous SectionNext Section

    Study Design

    An unblinded, randomized trial (Protocol 047) was conducted at 12 AIDS Clinical Trials Units. Seven regimens were tested (Figure 1). At the time the study began, the standard zidovudine dose was 200 mg every 4 hours.

    Figure 1. ddC = 2′,3′-dideoxycytidine.
    View larger version:
    Figure 1. ddC = 2′,3′-dideoxycytidine. The seven treatment regimens.

    Randomization was stratified by center and according to CD4 cell count (> 200 cells/mm3 or ≤ 200 cells/mm3). At least 16 patients were assigned to each limb. Patients who did not complete the first 8 weeks of therapy for any reason other than toxicity were replaced. All study patients gave informed consent and institutional review board approval was obtained at each institution.

    Study medication was to be administered for up to 49 weeks, with an additional follow-up period of 3 weeks. Study patients being treated with an intermittent regimen who had less than a 50% reduction in the serum p24 antigen level at 6 months had the option to change to an alternating regimen. Patients who completed 49 weeks of therapy and showed evidence of a beneficial response (defined as a 50% reduction in p24 antigen level or an improvement in CD4 cell count) had the option to continue therapy past 52 weeks. For the present analyses, only the first 48 weeks of data were used.

    Previous SectionNext Section

    Patient Evaluation

    Pretreatment evaluations included a medical history, physical examination, weight measurement, Karnofsky score, complete blood count, hepatic and renal function studies, T-cell subset analysis, and serum p24 antigen determination (Abbott Laboratories, North Chicago, Illinois). Follow-up evaluation was done weekly for the first 8 weeks and every 4 weeks thereafter. Serum specimens for p24 antigen determination were collected weekly for the first 8 weeks and then 3 of every 4 weeks thereafter. An evaluation for peripheral neuropathy was done on enrollment and was repeated as follows: neuropathy symptom questionnaire every 2 weeks, peripheral neurologic examination every 4 weeks, and quantitative sensory testing of vibration every 8 weeks [10]. Signs or symptoms suggestive of opportunistic infection or malignancy were evaluated according to standard AIDS Clinical Trials Group guidelines [15].

    Hematologic toxicity for patients being treated with a zidovudine-containing regimen was defined and managed as follows: Transfusion was allowed for symptomatic grade 2 anemia (hemoglobin, 80 to 94 g/L) and for grade 3 (hemoglobin, 65 to 79 g/L) or grade 4 anemia (hemoglobin < 65 g/L). Study drug dosage was reduced or held in cases of grade 3 or 4 anemia if the patient declined transfusion or required more than 4 units of packed red blood cells in 6 weeks; or in cases of grade 3 (granulocyte count, 500 to 750; leukocyte count < 1.5 × 109/L) or grade 4 (granulocyte count >500, leukocyte count < 1.0 × 109/L) granulocyte-leukocyte toxicity. Patients were removed from the study if they developed grade 3 or 4 anemia or granulocyte-leukocyte toxicity that either failed to respond to dose adjustment or recurred. Patients treated with intermittent ddC were removed from the study for the following reasons: a first episode of grade 3 anemia if the patient declined transfusion or required more than 4 units of packed red blood cells in 6 weeks; grade 3 granulocyte-leukocyte toxicity that either failed to respond to dose adjustment or recurred; or a first episode of grade 4 anemia or granulocyte-leukocyte toxicity.

    Dose-limiting peripheral neuropathy was defined by the occurrence of bilateral burning or shooting pains in the lower extremities that were of moderate (grade 2) intensity and persisted for 72 hours or more; a symptom of severe (grade 3) intensity of any duration; or a symptom of moderate intensity of any duration plus either a supporting abnormality in the affected limb on standardized peripheral nerve examination or a supporting abnormality in the affected limb on quantitative sensory testing.

    Previous SectionNext Section

    Statistical Analysis

    Data from all enrolled patients were used in the analyses; however, data collected beyond the point at which patients switched from their initially assigned therapy were not included in the analysis. The Fisher exact test and chi-square tests were used to compare subgroups when the data were discrete [16] and Wilcoxon-Mann-Whitney tests were used for continuous data [17]. The Kaplan-Meier method and log-rank tests were used to analyze time-to-event data [18]. All P values were two-sided.

    The area under the curve (AUC) of absolute CD4 cells/mm3 was calculated, and the resulting total area was divided by time on therapy (T) to derive AUC/T [19]. The baseline CD4 count was subtracted from AUC/T to obtain the average change in CD4 cell count from baseline. A linear regression model was then calculated, with the average change in CD4 cell count as the dependent variable and both the baseline CD4 cell count and the assigned treatment limb as the independent variables.

    Previous SectionNext Section

    Results

    Patients

    One hundred thirty-one patients were enrolled in the study between 24 June 1988 and 17 October 1989 (Table 1). Most patients were male (96%) and white (76%) and had not received zidovudine within 90 days of study entry (96%). Seventy-seven percent had AIDS-related complex and 23% had AIDS. The median CD4 cell count at entry was 142 cells/mm3, and the median serum p24 antigen level was 257 pg/mL.

    View this table:
    Table 1. Patient Characteristics at Study Entry*

    The median follow-up for the entire study group was 40 weeks. Fifty-nine patients (45%) completed 48 weeks of therapy (Table 2). Twenty patients were withdrawn from the study because of toxicity related to ddC or zidovudine. Twenty-eight patients were removed because of AIDS-related opportunistic infection, malignancy, or dementia; 13 of these patients were removed from the study before completing 8 weeks of therapy. These 13 patients—along with 4 patients who voluntarily withdrew from the study, were lost to follow-up, or were removed from the study for protocol violations in the first 8 weeks—were replaced.

    View this table:
    Table 2. Patient Status*

    Adverse Events

    The principal dose-limiting adverse event was peripheral neuropathy (Table 3). Because the protocol required that patients who developed neuropathic symptoms of grade 2 or greater severity (as described above) be withdrawn from the study, no dose reduction, dose interruption, or rechallenge occurred. The incidence of peripheral neuropathy was significantly higher in patients receiving the 0.03 mg/kg dose of ddC (34% for limbs B, D, and F combined) than in those receiving the 0.01 mg/kg dose (15% for limbs A and C combined, P = 0.04). Patients being treated with the monthly alternating regimen of ddC, 0.03 mg/kg, and zidovudine (limb D), however, had a significantly lower incidence of neuropathy (14%) than did patients treated with the weekly alternating regimen of ddC, 0.03 mg/kg, and zidovudine (46% for limbs B and F combined, P = 0.03). Overall, the incidence of peripheral neuropathy in three of the four alternating-therapy limbs (A, C, and D) was similar to that in the two zidovudine-alone treatment limbs (E and G). Twenty-one of the 31 patients with neuropathy (68%) developed it within the first 24 weeks of therapy.

    View this table:
    Table 3. Toxicity*

    Hematologic toxicity was most common in the continuous-zidovudine limb (G), occurring in 6 of 18 patients (33%) (see Table 3). Multiple dose reductions, dose interruptions, and transfusions were allowed by the protocol. The one patient who required discontinuation of therapy because of serious hematologic toxicity [recurrent grade 4 anemia] had been assigned to the continuous-zidovudine limb. Compared with patients treated with continuous zidovudine, those receiving zidovudine either intermittently or in alternation with ddC (limbs A, B, C, D, and E) had a significantly lower risk for hematologic toxicity (P < 0.02).

    Six additional patients were removed from the study for treatment-related adverse events. Three adverse events were attributed to ddC: anaphylactoid reaction (one patient in limb B), fever (one patient in limb D), and reversible hearing loss (one patient in limb D). Two adverse events were attributed to zidovudine: malaise (one patient in limb A) and elevated transaminase levels (one patient in limb E). A rash occurring in a patient assigned to limb B was of unknown cause.

    CD4 Cell Responses

    No significant differences were observed in baseline CD4 cell counts among the seven limbs; median CD4 cell counts were lowest in patients receiving continuous zidovudine. All regimens produced an initial increase in the median CD4 cell count; gains were more modest in patients receiving intermittent therapy with zidovudine or ddC (limbs E and F) than in those receiving alternating regimens or continuous zidovudine (limbs A, B, C, D, and G; [Figure 2]). The median peak CD4 cell responses for each group were greater than 100 cells/mm3 in limbs A, B, and D [149, 121, and 136 cells/mm3, respectively] and less than 100 cells/mm3 in limbs C, E, F, and G (90, 40, 50, and 76 cells/mm3, respectively).

    Figure 2. Alternating regimens (limbs A, B, C, and D) compared with continuous zidovudine (limb G). Intermittent regimens (limbs E and F) compared with continuous zidovudine (limb G). Imputed combined alternating regimens (indicated by X) compared with continuous zidovudine (limb G); missing values for patients not completing 48 weeks of study were imputed using the average of the last two CD4 counts on therapy. The curve labeled X represents combined values for alternating-therapy limbs A, B, C, and D. The baseline CD4 cell count was calculated as the mean of two pretreatment values. The number of participants on therapy in each limb at each time point is indicated along the abscissa. The standard errors of the mean change in CD4 cell count from baseline to weeks 8 and 24 were as follows: limb A, 16.7 and 33.7; limb B, 25.4 and 37.8; limb C, 19.1 and 33.9; limb D, 18.6 and 27.8; limb E, 21.7 and 10.9; limb F, 22.3 and 43.8, and limb G, 228.3 and 20.3. For a detailed description of the regimens, see .
    View larger version:
    Figure 2. Alternating regimens (limbs A, B, C, and D) compared with continuous zidovudine (limb G). Intermittent regimens (limbs E and F) compared with continuous zidovudine (limb G). Imputed combined alternating regimens (indicated by X) compared with continuous zidovudine (limb G); missing values for patients not completing 48 weeks of study were imputed using the average of the last two CD4 counts on therapy. The curve labeled X represents combined values for alternating-therapy limbs A, B, C, and D. The baseline CD4 cell count was calculated as the mean of two pretreatment values. The number of participants on therapy in each limb at each time point is indicated along the abscissa. The standard errors of the mean change in CD4 cell count from baseline to weeks 8 and 24 were as follows: limb A, 16.7 and 33.7; limb B, 25.4 and 37.8; limb C, 19.1 and 33.9; limb D, 18.6 and 27.8; limb E, 21.7 and 10.9; limb F, 22.3 and 43.8, and limb G, 228.3 and 20.3. For a detailed description of the regimens, see . Median change in CD4 cell count from baseline.Top.Middle.Bottom.Figure 1

    Sustained increases in the CD4 cell count past 24 weeks were observed in three of the alternating-therapy limbs (A, B, and D; see Figure 2, top). These limbs also had the greatest proportion of patients with a 50-cell or greater increase from baseline after 48 weeks of therapy (27%, 40%, and 60% for limbs A, B, and D, respectively). In contrast, the median change in CD4 cell count decreased to less than the baseline level by week 16 in patients receiving intermittent zidovudine or ddC (limbs E and F) and by week 28 in those receiving continuous zidovudine (limb G; see Figure 2, middle). By week 48, no patients receiving either of the zidovudine-alone regimens (limbs E and G) showed a 50-cell or greater increase in CD4 count.

    The greater CD4 cell increases seen in patients receiving the alternating regimens did not appear to be attributable to the selective withdrawal of patients with low CD4 cell counts. Missing values were replaced by each of the following: the lowest CD4 cell count on therapy, the last CD4 cell count on therapy, the average of the last two CD4 cell counts on therapy, or the average of all CD4 cell counts on therapy. Using any of these substitution methods, we found that regimens A, B, and D still appeared to be associated with sustained CD4 increases. As shown in Figure 2, bottom, patients receiving the four alternating regimens were combined and then compared with patients receiving continuous zidovudine; missing values were replaced by the average of the last two CD4 cell counts on therapy. At each time point, CD4 gains appeared to be greater and more sustained in patients receiving the alternating regimens than in those receiving continuous zidovudine.

    The regression model used to analyze the average change in CD4 cell count during the entire 48-week study period (see Methods) showed that patients with higher baseline CD4 cell counts had greater increases in CD4 cell counts while on therapy (P = 0.002). However, even after adjustments were made for baseline CD4 counts, patients assigned to treatment limbs A and D showed the greatest increases in average CD4 cell counts (P = 0.07 and 0.02, respectively), whereas those assigned to limb E showed the greatest decrease (P = 0.02). This model predicts that, given a baseline measure of 200 cells/mm3, the average on-therapy CD4 count would be 247, 231, 220, 250, 186, 210, and 197 for limbs A through G, respectively. Patients receiving alternating regimens (limbs A, B, C, and D) differed significantly from those receiving continuous zidovudine regarding average on-therapy CD4 cell count (P = 0.05). Although these results are based on small numbers of patients, they remained the same when the average CD4 change was imputed using the last two values on therapy.

    Serum p24 Antigen Response

    Each of the four alternating regimens, as well as continuous zidovudine, produced rapid decreases in serum p24 antigen levels (Figure 3). Suppression of serum p24 antigen was sustained in many patients throughout the study, with more than 70% of patients receiving alternating regimens showing a 50% or greater decrease in p24 antigen level from the baseline value by week 24 of therapy. Among alternating regimens, the monthly alternating regimen of ddC, 0.01 mg/kg, and zidovudine (limb C) had the weakest effect on the p24 antigen level. Intermittent zidovudine therapy (limb E) appeared to be the least effective of the seven regimens; about 25% of the patients assigned to this regimen showed a 50% reduction in the p24 antigen level throughout the study. Two patients who received intermittent zidovudine (limb E) had increasing serum p24 antigen levels and were removed from the study to begin continuous zidovudine by prescription. Four additional patients in this limb were crossed over to a weekly alternating regimen of zidovudine and ddC, 0.03 mg/kg, after failing to show at least a 50% decrease in the serum p24 antigen level from the baseline level by month 6 of therapy (see Table 2).

    Figure 3. Alternating regimens (limbs A, B, C, and D) compared with continuous zidovudine (limb G). Intermittent regimens (limbs E and F) compared with continuous zidovudine (limb G). Combined alternating regimens (indicated by X) compared with continuous zidovudine (limb G). The curve labeled X represents combined values for alternating-therapy limbs A, B, C, and D. Baseline serum p24 antigen values were calculated as the mean of two pretreatment values. The number of participants on therapy in each limb at each time point is indicated along the abscissa. Antigen testing was done in batch at the conclusion of the study. The standard errors of the mean percent changes in p24 antigen level from baseline to weeks 8 and 24 were as follows: limb A, 32.3 and 5.3; limb B, 12.8 and 22.4; limb C, 4.6 and 11.7; limb D, 21.4 and 12.3; limb E, 21.2 and 20.6; limb F, 41.0 and 11.9; and limb G, 5.4 and 8.3. One patient in limb D was excluded from this calculation. For a detailed description of the regimens, see .
    View larger version:
    Figure 3. Alternating regimens (limbs A, B, C, and D) compared with continuous zidovudine (limb G). Intermittent regimens (limbs E and F) compared with continuous zidovudine (limb G). Combined alternating regimens (indicated by X) compared with continuous zidovudine (limb G). The curve labeled X represents combined values for alternating-therapy limbs A, B, C, and D. Baseline serum p24 antigen values were calculated as the mean of two pretreatment values. The number of participants on therapy in each limb at each time point is indicated along the abscissa. Antigen testing was done in batch at the conclusion of the study. The standard errors of the mean percent changes in p24 antigen level from baseline to weeks 8 and 24 were as follows: limb A, 32.3 and 5.3; limb B, 12.8 and 22.4; limb C, 4.6 and 11.7; limb D, 21.4 and 12.3; limb E, 21.2 and 20.6; limb F, 41.0 and 11.9; and limb G, 5.4 and 8.3. One patient in limb D was excluded from this calculation. For a detailed description of the regimens, see . The median percent change in serum p24 antigen from baseline.Top.Middle.Bottom.Figure 1

    Weight Gain

    Patients receiving the four alternating regimens showed greater weight gains than did those who received continuous zidovudine (P = 0.008, at week 48). Sustained weight gain was greatest in limbs A, B, and C (median weight gain at week 48, 3.4 kg, 3.8 kg, and 2.4 kg, respectively). More modest weight gains were seen in limb D (0.9 kg) and limb F (1.3 kg); net losses in median weight were seen in patients taking zidovudine alone (limb E, −1.0 kg; limb G, −0.7 kg).

    Progression of HIV-associated Disease

    Opportunistic infections occurred in five patients after 8 weeks of therapy: Two patients developed Pneumocystis carinii pneumonia and three acquired Mycobacterium avium complex infection. Both patients who developed Pneumocystis carinii pneumonia had been receiving prophylaxis with aerosolized pentamidine. In addition, one patient developed HIV-associated dementia and two patients developed lymphoma. The number of opportunistic infections and malignancies was similar among the treatment limbs (see Table 2).

    Three patients died while receiving therapy or within 4 weeks of stopping therapy. These deaths were due to HIV-associated dementia (limb C, week 41), cardiomyopathy (limb E, week 3), and lymphoma with renal failure (limb F, week 20). Six additional patients died 16 to 30 weeks after stopping study medication. No death was thought to be related to study drug administration.

    Previous SectionNext Section

    Discussion

    Early clinical trials showed that zidovudine therapy reduces risks for morbidity and death in patients with AIDS or AIDS-related complex [1, 20]. However, continuous zidovudine therapy causes toxicity, chiefly neutropenia and anemia, in 40% to 55% of patients who receive the drug in a dose of 1200 to 1500 mg/d [5, 6]. The administration of zidovudine, 600 mg/d, in patients with AIDS who had had Pneumocystis carinii pneumonia reduced the rates of severe neutropenia and anemia to 37% and 29%, respectively [6]. However, even at this dose, hematologic toxicity remains a significant reason for dose interruption and withdrawal of drug in patients with advanced HIV-associated disease [6].

    Continuous therapy with ddC is associated with a dose-limiting peripheral sensory neuropathy [7-11, 21]. When given as monotherapy in previous studies, ddC at doses of 0.03 and 0.01 mg/kg resulted in neuropathy in 100% and 80% of patients, respectively [8]. Acceptable rates of neurotoxicity were achieved only with reduced doses of 0.01 mg/kg every 8 hours or less [11].

    Our study showed a reduction in expected toxicities when zidovudine and ddC were given as alternating therapy. No participant who received alternating therapy had hematologic toxicity requiring permanent drug withdrawal, and patients treated with three of the four alternating regimens (limbs A, C, and D) showed rates of neuropathy that were similar to those seen in patients who did not receive ddC. Thus, despite the high doses of both agents, dose-limiting drug effects in these three alternating-therapy limbs were acceptable.

    Our ability to detect statistically significant differences in clinical or surrogate markers of HIV infection between patients receiving alternating regimens and those receiving continuous zidovudine was limited by study size and design. Assuming that at least 16 patients per limb remained on therapy for at least 8 weeks, the statistical power to detect differences of 25 cells in CD4 cell count change between alternating-therapy limbs and the continuous-zidovudine limb is only 14% at week 8. We could have detected a difference in the CD4 cell count change from baseline of 82 cells with 80% power. A difference in change from baseline of 1.26 kg for weight or of 30% for p24 antigen level between alternating-therapy limbs and the continuous-zidovudine limb could have been detected with 80% power. The power to detect differences of these magnitudes between individual limbs would be lower (approximately 21%). Higher rates of discontinuation (particularly in limbs B, E, and F) would further reduce power. Although patients were stratified by entry CD4 count and were randomly assigned to a treatment limb, neither patients nor investigators were blinded to the assigned therapy. The results of the study, however, do not suggest that this aspect of the study design was a significant source of bias in the described outcomes.

    Despite the limitations of the study, the descriptive analyses of surrogate markers of antiretroviral activity, such as CD4 cell count and p24 antigen level, suggest some comparisons between subgroupings. Although, overall, no sustained increase in CD4 cell count was seen in the study patients (P > 0.20, global test of differences), patients receiving three of the four alternating regimens appeared to have sustained increases in CD4 cell counts, whereas patients taking zidovudine alone showed only a temporary increase. The regression analysis further suggests that perceived differences between patients assigned to alternating therapy and those assigned to continuous zidovudine were not attributable to the lower median CD4 count at entry in the latter patients.

    Sustained suppression of serum p24 antigen levels was observed in the alternating-therapy limbs and in the continuous-zidovudine limbs; intermittent therapy with zidovudine or ddC was less effective in this regard. It has been noted previously that p24 antigen suppression by zidovudine occurs rapidly after institution of therapy and reverses rapidly after discontinuation of therapy [22, 23]. Our results suggest that neither zidovudine nor ddC exerts a sufficient antiretroviral carry-over effect into the washout phase to be effective as intermittent therapy.

    Weight change was used in our study to indicate the clinical effect of the various treatment regimens. Both the magnitude and the duration of weight gain was greater in the four alternating-therapy limbs than in the continuous- or intermittent-monotherapy limbs.

    Patients receiving intermittent regimens did not have encouraging toxicity-efficacy profiles. Therapy with intermittent zidovudine (limb E) resulted in poor CD4 enhancement and p24 suppression, although such therapy was relatively well tolerated. Intermittent ddC therapy at a dose of 0.03 mg/kg (limb F) resulted in suppression of p24 antigen levels but was associated with poor CD4 enhancement and an unacceptably high incidence of peripheral neuropathy. Based on our data, continuous use of nucleoside-analog therapies appears to be warranted and the intermittent use of any one drug should be avoided.

    Results in patients who received weekly alternating therapy with ddC, 0.03 mg/kg, and zidovudine (limb B) show the importance of a sufficient period of washout in reducing the incidence of peripheral neuropathy. Although this regimen produced marked CD4 enhancement and p24 antigen suppression, the rate of peripheral neuropathy was unacceptably high. A much lower incidence of neuropathy, in association with similar antiviral and immune-system-enhancing effects, was seen when the same total daily dose of ddC was given in a monthly alternating fashion.

    The ddC dose of 0.01 mg/kg did not perform as well in a monthly alternating regimen (limb C) as the 0.03 mg/kg dose (limb D). Because of the greater antiretroviral activity, the higher dose may be able to maintain the CD4 enhancement and p24 antigen suppression initiated during the month of zidovudine therapy.

    Acceptable toxicity and prolonged antiretroviral and immunologic effects were seen in the weekly alternating regimen of ddC, 0.01 mg/kg, and zidovudine (limb A) and in the monthly alternating regimen of ddC, 0.03 mg/kg, and zidovudine (limb D). These regimens allowed the use of twice the daily dose of zidovudine and two to six times the daily dose of ddC as that used for monotherapy. High doses of drug might be desirable because antiretroviral activity in vitro and penetration into the central nervous system are concentration dependent [7, 24]. In one clinical study, trends were seen toward a reduction in the incidence of AIDS-dementia complex and total AIDS-defining events in patients receiving 1200 mg/d compared with those receiving 800 or 400 mg/d [25]. In another study of zidovudine in HIV-related dementia, a trend was noted toward the greater effect of a high dose (2000 mg/d) compared with a “lower” dose (1000 mg/d) [26]. Our study showed that high doses of zidovudine and ddC can be well tolerated in an alternating regimen for more than 1 year and may be particularly suited to patients in whom maximal serum levels of an antiretroviral agent may be preferred.

    In patients with advanced HIV-associated disease, the reduced hematologic toxicity of lower daily doses of zidovudine (500 to 600 mg/d) allowed for the continuous use of an antiretroviral agent [6, 25], which may be a factor in the improved survival seen in patients receiving this dose [6]. Other regimens that reduce long-term toxicity, such as the alternating regimens described, may also show the benefits of continuous antiretroviral therapy.

    Monotherapy with ddC, 0.75 mg every 8 hours, has been shown to be inferior to zidovudine monotherapy in patients with AIDS or advanced AIDS-related complex who have not previously received therapy with an antiretroviral nucleoside analog (Hoffmann-La Roche, Inc. Personal communication). The zidovudine group (100 mg every 4 hours) had both longer survival and greater CD4 cell increases than did the ddC group. These results may raise the concern that alternating therapy with zidovudine and ddC would be inferior to therapy with zidovudine alone in prolonging survival. The results of our study, however, suggest that the alternating regimens are superior to zidovudine alone in prolonging increases in the CD4 cell count. Experience with zidovudine and didanosine monotherapy shows that improved clinical course and prolonged survival accompany a favorable change in CD4 cell count, suggesting that alternating regimens should be superior to zidovudine alone [14]. Clearly, this must be proved in a large, comparative, controlled trial.

    The addition of ddC to zidovudine in either a weekly or monthly alternating regimen showed both the antiretroviral activity of ddC and the potential importance of combination therapy in enhancing the efficacy of both these drugs. Similar prolonged increases in CD4 cell counts were also noted with the concomitant administration of zidovudine and ddC [13, 14]. The regimen of concomitant ddC and zidovudine (0.75 mg and 200 mg, respectively, every 8 hours) has been approved by the Food and Drug Administration for patients with a CD4 cell count of 300/mm3 or less who have demonstrated significant clinical or immunologic deterioration (HIVID package insert, Hoffman-La Roche, Inc., Nutley, New Jersey). However, neither alternating nor concomitant administration of the two antiretroviral agents has yet demonstrated the ability to prevent the emergence of in vitro resistance to zidovudine (Richman D. Personal communication). Whether this finding suggests that the enhanced effects of the combined administration of zidovudine and ddC on surrogate markers is independent of the development of resistance awaits the further evaluation of both modes of combination therapy.

    The ideal regimen of alternating zidovudine and ddC may be one in which optimally effective doses of each drug are administered with sufficient washout periods to minimize toxicity. Two regimens studied, the alternating weekly regimen of zidovudine and ddC, 0.01 mg/kg (limb A), and the alternating monthly regimen of zidovudine and ddC, 0.03 mg/kg (limb D), had favorable toxicity profiles and were associated with sustained CD4 cell increases, p24 antigen suppression, and weight gain. These regimens deserve further evaluation for clinical efficacy in a larger study population.

    This study was presented in part at the Sixth International Conference on AIDS, San Francisco, California, 20-24 June 1990.

    Previous SectionNext Section

    Appendix

    Other persons contributing to the ACTG 047 study included Lisa Tieger, Albert Einstein College of Medicine, Bronx, New York; David Amato, PhD, Harvard School of Public Health, Boston, Massachusetts; Jan Reddick, FNP, and Frank Baumgartner, FNP, Stanford University, Stanford, California; David Clifford, MD, Janet Voorhees, RN, and Michael Klebert, RN, Washington University, St. Louis, Missouri; Alana Fraser, RN, FNP, and Joanne Santangelo, RN, MSN, FNP, University of California, San Diego, California; Pamela Urbanski, RN, MSN, Rush-Presbyterian Hospital, Chicago, Illinois; William Richardson, RN, Donna Sutherland, BA, MLT (ASCP), and Bisher Akil, MD, University of Southern California, Los Angeles, California; Brenda Kolatch, MS, Julie Linksman-Rivera, RN, and James O'Connor, MD, St. Luke's-Roosevelt Hospital Center, Columbia University, New York, New York; Debra Greenspan, BS, MN, Tulane University, New Orleans, Louisiana; Ellen Godfrey, MA, Theresa Flynn, RN, and Robert T. Schooley, MD, Harvard University and Massachusetts General Hospital, Boston, Massachusetts; Susan LeLacheur, PA, David M. Parenti, MD, and Jane Courtless, RN, The George Washington University, Washington, DC; Nancy Priller, RN, Northwestern University, Chicago, Illinois; Henry H. Balfour, MD, University of Minnesota, Minneapolis, Minnesota; Miklos P. Salgo, MD, PhD, and Cheryl N. Karol, PhD, Hoffmann-La Roche, Inc., Nutley, New Jersey.

    Previous SectionNext Section

    Abbreviation

    ddC: 2′,3′-dideoxycytidine

    Previous Section
     

    References

    1. 1.↵
      Fischl MA, Richman DD, Grieco MH, Gottlieb MS, Volberding PA, Laskin OL, et al. The efficacy of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex. A double-blind, placebo-controlled trial. N Engl J Med. 1987; 317:185-91.
    2. 2.↵
      Volberding PA, Lagakos SW, Koch MA, Pettinelli C, Myers MW, Booth DK, et al. Zidovudine in asymptomatic human immunodeficiency virus infection. A controlled trial in persons with fewer than 500 CD4-positive cells per cubic millimeter. The AIDS Clinical Trials Group of the National Institute of Allergy and Infectious Diseases. N Engl J Med. 1990; 322:941-9.
    3. 3.↵
      Fischl MA, Richman DD, Hansen N, Collier AC, Carey JT, Para MF, et al. The safety and efficacy of zidovudine (AZT) in the treatment of subjects with mildly symptomatic human immunodeficiency virus type I (HIV) infection. A double-blind, placebo-controlled trial. The AIDS Clinical Trials Group. Ann Intern Med. 1990; 112:727-37.
    4. 4.↵
      Richman DD, Fischl MA, Grieco MH, Gottlieb MS, Volberding PA, Laskin OL, et al. The toxicity of azidothymidine (AZT) in the treatment of patients with AIDS-related complex. A double-blind, placebo-controlled trial. N Engl J Med. 1987; 317:192-7.
    5. 5.↵
      Dournon E, Matheron S, Rozenbaum W, Gharakhanian S, Michon C, Girard PM, et al. Effects of zidovudine in 365 consecutive patients with AIDS or AIDS-related complex. Lancet. 1988; 2:1297-302.
    6. 6.↵
      Fischl MA, Parker CB, Pettinelli C, Wulfsohn M, Hirsch MS, Collier AC, et al. A randomized, controlled trial of a reduced daily dose of zidovudine in patients with the acquired immunodeficiency syndrome. N Engl J Med. 1990; 323:1009-14.
    7. 7.↵
      Yarchoan R, Perno CR, Thomas RV, Klecker RW, Allain JP, Wills RJ, et al. Phase I studies of 2′,3′-dideoxycytidine in severe human immunodeficiency virus infection as a single agent and alternating with zidovudine (AZT). Lancet. 1988; 1:76-81.
    8. 8.↵
      Merigan TC, Skowron G, Bozzette SA, Richman D, Uttamchandani R, Fischl M, et al. Circulating p24 antigen levels and responses to dideoxycytidine in human immunodeficiency virus (HIV) infections: a phase I and II study. Ann Intern Med. 1989; 110:189-94.
    9. 9.↵
      Merigan TC, Skowron G. Safety and tolerance of dideoxycytidine as a single agent. Results of early-phase studies in patients with acquired immunodeficiency syndrome (AIDS) or advanced AIDS-related complex. Am J Med. 1990; 88(Suppl 5B):11S-5S.
    10. 10.↵
      Berger AR, Arezzo JC, Schaumburg HH, Skowron G, Bozzette S, Soo W. 2′,3′-dideoxycytidine (ddC) toxic neuropathy: a study of 52 patients. Neurology. (In press.)
    11. 11.↵
      Gottlieb M, Galpin J, Thompkins J, Wilson D, Donatacci L, Soo W. 2′,3′-dideoxycytidine (ddC) in the treatment of patients with AIDS and ARC (Abstract). In: Proceedings of the Fifth International Conference on AIDS, Montreal, Quebec, June 4-9, 1989. Ottawa, Ontario: International Development Research Centre; 1989.
    12. 12.↵
      Spector SA, Ripley D, Hsia K. Human immunodeficiency virus inhibition is prolonged by 3′-azido-3′-deoxythymidine alternating with 2′,3′-dideoxycytidine compared with 3′-azido-3′-deoxythymidine alone. Antimicrob Agents Chemother. 1989; 33:9203.
    13. 13.↵
      Meng TC, Fischl MA, Boota AM, Spector SA, Bennett D, Bassiakos Y, et al. Combination therapy with zidovudine and dideoxycytidine in patients with advanced human immunodeficiency virus infection. A phase I/II study. Ann Intern Med. 1992; 116:13-20.
    14. 14.↵
      Food and Drug Administration. Antiviral Drug Products Advisory Committee, April 20, 1992.
    15. 15.↵
      Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR. 1987; 36 (Supp 1S):1S-15S.
    16. 16.↵
      Cox DR. The Analysis of Binary Data. London: Metheun; 1970.
    17. 17.↵
      Lehmann EL. Nonparametrics: Statistical Methods Based on Ranks. San Francisco: Holden Day; 1975.
    18. 18.↵
      Kalfleish JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York: Wiley; 1980.
    19. 19.↵
      Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ. 1990; 300:230-6.
    20. 20.↵
      Yarchoan R, Klecker RW, Weinhold KJ, Markham PD, Lyerly HK, Durack DT, et al. Administration of 3′-azido-3′-deoxythymidine, an inhibitor of HTLV-III/LAV replication, to patients with AIDS or AIDS-related complex. Lancet. 1986; 1:575-80.
    21. 21.↵
      Dubinsky RM, Dalakas M, Yarchoan R, Broder S. Follow-up of neuropathy from 2′,3′-dideoxycytidine (Letter). Lancet. 1988; 1:832.
    22. 22.↵
      de Wolf F, Lange JM, Goudsmit J, Cload P, de Gans J, Schellekens PT, et al. Effect of zidovudine on serum human immunodeficiency virus antigen levels in symptom-free subjects. Lancet. 1988; 1:373-6.
    23. 23.↵
      Jackson GG, Paul DA, Falk LA, Rubenis M, Despotes JC, Mack D, et al. Human immunodeficiency virus antigenemia in AIDS and the effect of treatment with zidovudine. Ann Intern Med. 1988; 108; 75-180.
    24. 24.↵
      Klecker RW, Collins JM, Yarchoan R, Thomas R, McAtee N, Broder S, et al. Plasma and cerebrospinal fluid pharmocokinetics of 3′-azido-3′-deoxythymidine: a novel pyrimidine analog with potential application for the treatment of patients with AIDS and related diseases. Clin Pharmacol Ther. 1987; 41:407-12.
    25. 25.↵
      Nordic Medical Research Councils' HIV Therapy Group. Double-blind dose-response study of zidovudine in AIDS and advanced HIV infection. BMJ. 1992; 304:13-7.
    26. 26.↵
      Sidtis JJ, Gatsonis C, Price RW, Singer EJ, Collier AC, Richman DD, et al. Zidovudine treatment of the AIDS Dementia Complex: Results of a placebo-controlled trial. Ann Neurol. (In press.)
    « Previous | Next Article »Table of Contents

    This Article

    1. Ann Intern Med March 1, 1993 vol. 118 no. 5 321-330
    1. AbstractFREE
    2. Figures
    3. » Full Text
    4. Full Text (PDF)

    Rapid Responses

    1. Submit a response
    2. No responses published

    Navigate This Article

    Current Issue

    1. November 17, 2009, 151 (10)
    1. Alert me to current issues