Itraconazole and fluconazole offer advantages over amphotericin B or ketoconazole as maintenance therapy for fungal disease in patients with AIDS [3]. Both are well absorbed after oral administration and cause few side effects. Although both drugs have been effective in the treatment of murine [4, 5] and human histoplasmosis [6], itraconazole was selected for further study because there is more experience with it in the treatment of histoplasmosis in humans [6]. We carried out an open-label, nonrandomized trial to assess the efficacy and tolerance of itraconazole in the prevention of histoplasmosis relapse in patients with AIDS who had been treated with amphotericin B.

Methods

Top Methods Results Discussion Author & Article Info References

Patients

Participants were at least 13 years old, had human immunodeficiency virus (HIV) infection, and had received amphotericin B induction therapy at a total dose of at least 15 mg/kg body weight administered for 3 to 12 weeks for proven disseminated histoplasmosis within 6 weeks before enrollment. Successful induction treatment was documented by the absence of clinical findings associated with histoplasmosis and negative fungal blood cultures at the time of enrollment. Patients were excluded from the study if they had a history of allergy to azole antifungal agents; a bilirubin level more than 2.5 times the upper limit of normal; an alanine aminotransferase level more than five times the upper limit of normal; an alkaline phosphatase level more than five times the upper limit of normal; a creatinine level more than three times the upper limit of normal; neutropenia (neutrophil count, <0.75 x 10⁹/L) or thrombocytopenia (platelet count, <75 x 10⁹); a Karnofsky score of less than 60; and concurrent treatment with phenytoin, barbiturates, rifampin, corticosteroids in doses exceeding those used in maintenance therapy for adrenal insufficiency, cytotoxic chemotherapy, systemic antifungal agents, or investigational drugs.

Patients requiring histamine blockers or antacids were not allowed to receive such therapy within 4 hours of itraconazole administration.

Intervention and Evaluation

Eligible patients received itraconazole, 200 mg twice daily orally with meals.

Patients were evaluated for efficacy and toxicity until the last enrolled patient completed at least 52 weeks of treatment. Patients were evaluated biweekly for 4 weeks, monthly for 12 months, and then bimonthly until the completion of the study. Each evaluation included an assessment of clinical findings to identify evidence of histoplasmosis relapse or side effects of itraconazole therapy, complete blood counts, serum chemistries, and determinations of Histoplasma capsulatum variety capsulatum antigen levels in urine and serum. Plasma itraconazole levels were measured at weeks 2, 4, 12, 24, and 36, and blood was collected for fungal culture at enrollment and at weeks 12, 24, 36, and 52, or as indicated based on clinical findings.

Adverse events were graded using the standard AIDS Clinical Trial Group severity scoring system. Selected laboratory criteria are listed in the results section, but complete criteria are available on request.

Plasma itraconazole concentrations were measured by bioassay at the Fungus Testing Laboratory, San Antonio, Texas [7]. Histoplasma capsulatum var. capsulatum antigen levels in urine and serum were measured at Indiana University [8, 9]. Antigen levels were measured as specimens were received and in batch after all patients had completed 1 year of treatment or had withdrawn from the study for other reasons. All specimens were stored at –70°C.

Statistical Analysis

The distribution of the times to events was estimated using the method of Kaplan and Meier [10]. In the determinations of duration of follow-up, patients who died were not considered to be "lost to follow-up." Mean antigen levels at initiation and completion of 52 weeks of study therapy were compared using a paired t-test. Differences in proportions of patients with positive antigen results at initiation and completion of 52 weeks of study therapy were compared using the McNemar test.

Results

Top Methods Results Discussion Author & Article Info References

Forty-two patients were enrolled in the study from 10 AIDS Clinical Trials units and 2 Mycoses Study Group sites. Forty-one of the 42 patients were diagnosed with disseminated histoplasmosis based on demonstration of organisms consistent with H. capsulatum var. capsulatum in extrapulmonary tissues by special stain or culture; the diagnosis in the remaining patient was based on detection of H. capsulatum var. capsulatum antigen in urine and demonstration of granulomas in the bone marrow.

Baseline demographic, clinical, and laboratory characteristics, as well as data on induction antifungal therapy, are summarized in Table 1. The median CD4 lymphocyte count at baseline for these patients was 0.047 x 10⁹/L (range, 0.005 to 0.272). The CD4 count was less than 0.050 x 10 (⁹/L) in 19 of 37 patients (51%), less than 0.100 x 10⁹/L in 30 patients (81%), and less than 0.200 x 10⁹/L in 36 patients (97%); CD4 counts were not available in 5 patients. Disseminated histoplasmosis was the primary AIDS-defining illness in 25 patients (60%). Disseminated histoplasmosis occurred a median of 3 months (range, 0.3 to 12.8 months) after the AIDS-defining illness in the 17 patients in whom histoplasmosis was a second or later AIDS-defining event. Thirty-eight patients had received amphotericin B at a total dose of at least 15 mg/kg given over 4 to 12 weeks, and the remaining 4 patients had received a total dose of at least 10 mg/kg. The median total dosage was 20.9 mg/kg (range, 14.0 to 59.2 mg/kg) given over a median of 36 days (range, 9 to 132 days).

Of the 42 patients who enrolled in the study, 15 failed to meet all eligibility criteria but were included in the data analysis. Reasons for failure to meet eligibility criteria included a neutrophil count of less than 0.75 x 10⁹/L or a platelet count of less than 75 x 10⁹/L (five patients), induction treatment with less than 15 mg/kg of amphotericin B (four patients), completion of induction therapy more than 6 weeks before enrollment (three patients), and negative cultures at the time of diagnosis of histoplasmosis (three patients). In two of these latter three patients, silver staining of lymph node specimens showed organisms resembling H. capsulatum var. capsulatum; in the third, antigen was detected in the urine.

Compliance

Compliance with itraconazole treatment was assessed by counts of returned medication and measurement of plasma itraconazole concentrations at specified visits. Less than 25% of medication was returned at 692 of the 720 visits (96%). The 28 visits at which more than 25% of medication was returned were made by 16 patients (38%). Plasma itraconazole concentrations of at least 0.75 µg/mL were documented for 233 of 255 determinations (91%) in the 40 patients for whom results were available. Overall, 25 of the 40 patients (63%) had concentrations of at least 0.75 µg/mL for every determination. In the remaining 15 patients (37%), concentrations were less than 0.75 µg/mL for 31 of 95 determinations (33%). Itraconazole concentrations were determined in specimens obtained at (four occasions) or within 4 weeks (nine occasions) of the visits for which more than 25% of unused medication was returned. Concentrations were at least 0.75 µg/mL in 10 of the 13 specimens (77%).

Follow-Up

The median duration of follow-up was 109 weeks (range, 4 to 134 weeks), and the median survival time was 98 weeks (range, 4 to 134+ weeks) (Figure 1). As of 4 February 1992, when the data were analyzed, 17 patients continued to receive study treatment. Among the 25 patients who were no longer receiving therapy, reasons for discontinuation included death (15 patients), voluntary withdrawal (3 patients), inability to keep scheduled follow-up (3 patients), preterminal HIV-associated disease (2 patients who died within 7 days of discontinuing treatment), toxicity (1 patient), and relapse (1 patient) (Table 2). Of the 10 patients who withdrew or were withdrawn from study, 6 subsequently died.

View larger version (12K): [in this window] [in a new window]   Figure 1. Kaplan-Meier estimates of survival and relapse. The numbers in parentheses along the x-axis refer to number of patients still taking the study drug at that week of follow-up. Twenty-one patients were still living at the time of the analysis, and 17 were still receiving study medication.

Overall, 21 patients had died as of 4 February 1992. Causes of death included progressive HIV infection with wasting (10 patients), opportunistic infections (9 patients), and HIV-related malignancy (2 patients). Of the 9 patients with opportunistic infections, 2 had disseminated Mycobacterium avium-intracellulare complex infection, 2 had pneumonia of uncertain cause, and 1 each had toxoplasmosis, sepsis, disseminated Pneumocystis carinii infection, an undiagnosed central nervous system syndrome with paralysis and coma, and disseminated histoplasmosis. Only 1 patient had an autopsy, and disseminated Pneumocystis carinii infection was found to be the cause of death; cultures for H. capsulatum var. capsulatum were negative.

Risk for Relapse

One proven and one possible relapse occurred during the study. The proven relapse occurred in a patient who withdrew from the study at week 8 and died of disseminated histoplasmosis 18 weeks later while receiving no maintenance therapy. The possible relapse occurred in a patient who did not comply with therapy; itraconazole was not detected in blood samples at two consecutive clinic visits before the presumed relapse at week 68 of the study. Clinical findings included fever, splenomegaly, and the presence of diffuse pulmonary infiltrates. Amphotericin B therapy was started for a presumed relapse of histoplasmosis. The patient also received treatment with broad-spectrum antibiotics and pentamidine for possible bacterial and Pneumocystis carinii infections. Relapse was not confirmed by a positive culture or by an increase in antigen levels. After re-induction therapy with amphotericin B, 500 mg, the patient resumed itraconazole maintenance therapy from weeks 72 to 92 of the study. This patient was subsequently withdrawn from therapy because of incarceration. At 108 weeks, the patient reported during telephone follow-up that he had not relapsed while receiving ketoconazole. It is noteworthy that no evidence of relapse was observed during the initial 56 weeks of therapy when plasma itraconazole levels ranged from 4.3 and 20 µg/mL.

Fungal blood cultures done at baseline and at weeks 12, 24, 36, and 52 were negative in all cases.

Development of Other Fungal Infections

Six patients developed oral thrush. At the time of diagnosis, three patients had plasma itraconazole levels of 3.8, 6.2, and 19.3 µg/mL; the drug was not detectable in the other three patients. Antifungal susceptibility testing to identify in-vitro resistance as a cause for treatment failure was not done. Treatment with nystatin or clotrimazole eradicated thrush in three of four patients. No patient was reported to have esophageal or disseminated candidiasis, cryptococcosis, or any other systemic fungal infection during treatment.

Antigen Clearance from Serum and Urine

Antigen levels in serum and urine decreased progressively during treatment (Figure 2). The rate of decline, based on the entire study group, was 0.03 units/week in serum and 0.16 units/week in urine. Twenty of 40 patients (50%) had antigen in serum at enrollment, but only 5 of 40 patients (13%) had serum antigen at week 52 or at completion of study if sooner (P = 0.0003). The five patients who had detectable antigen at the completion of therapy were among the 20 with positive test results at enrollment. Antigen levels in serum decreased by at least 2 units in 5 of 7 patients (71%) who had initial values of at least 2.6 units, and assay results became negative in 15 of 20 patients (75%) who initially had a positive result. The mean antigen level in serum (±SD) decreased by 0.89 ± 1.25 units, from 1.55 ± 1.39 units at enrollment to 0.66 ± 0.37 units at week 52 or at completion of study if sooner (P = 0.0001).

View larger version (31K): [in this window] [in a new window]   Figure 2. Histoplasma capsulatum var. capsulatum in serum and urine. Initial antigen levels and final levels after 1 year of treatment (or at withdrawal from study if sooner) are shown for individual patients. The number of patients with negative results at enrollment and at follow-up is represented by the thick bar at the bottom of the figure.

Antigen was detected in the urine of 35 of 39 patients (90%) at enrollment and in 22 of 39 patients (56%) at week 52 or at completion of study if sooner (P = 0.004). Antigen levels in urine decreased by at least 2 units in 27 of 30 patients (90%) with initial levels of at least 2.6 units, and assay results became negative in 15 of 35 patients (43%) who initially had a positive result. The mean antigen level in urine decreased by 3.52 ± 5.28 units, from 6.91 ± 5.62 units at enrollment to 3.39 ± 4.53 units at the final analysis (P = 0.0002).

None of 40 patients showed an increase of 2 units or more in the serum antigen level, and only 1 of 39 patients (3%) showed such an increase in the urinary antigen level. The single patient with the increase in urinary antigen showed no clinical evidence of relapse and subsequently became antigen negative. The antigen level in the patient with the presumed relapse at week 68 of treatment was 0.6 units in both urine and serum compared with levels of 3.8 and 0.9 units, respectively, at enrollment. Specimens were not available at relapse for the patient who withdrew from the study at week 8 and who died of disseminated histoplasmosis 18 weeks later while not receiving antifungal maintenance therapy.

Plasma Itraconazole Concentrations

Itraconazole concentrations were measured in blood obtained a median of 4 hours (range, 0 to 20 hours) after oral administration of a 200-mg dose. A total of 255 determinations were done in 40 patients. The median itraconazole concentration was 6.8 µg/mL (range, 0 to 21.3 µg/mL). Concentrations were more than 0.75 µg/mL in 233 of 255 specimens (91%), including 51 of 54 specimens (94%) obtained during the first month and 182 of 201 specimens (91%) obtained during the remainder of treatment. Twenty-five patients had concentrations greater than 0.75 µg/mL at every determination. Itraconazole was undetectable in 14 specimens (5%) that came from 10 patients (24%). However, multiple specimens were tested for 39 of 40 patients, and no patient had a concentration of less than 0.75 µg/mL on each determination. Concentrations were at least 0.75 µg/mL on 64 of 95 determinations (67%) for the 15 patients with concentrations of less than 0.75 µg/mL on at least one determination.

Plasma itraconazole concentrations did not vary with time after oral administration of a 200-mg dose (Figure 3). The protocol specified that plasma samples should be obtained 2 to 4 hours after an oral dose, but the timing varied. The time of sampling after oral administration of a 200-mg dose was known for 254 determinations; 150 samples (59%) were obtained within 4 hours, 76 (30%) between 6 and 11 hours, and 28 (11%) between 12 and 22 hours after administration. Linear regression analysis showed no relation between plasma concentration and time of sampling after drug administration (P > 0.2).

View larger version (26K): [in this window] [in a new window]   Figure 3. Plasma itraconazole concentrations as a function of time after drug administration. The analysis is based on 254 observations in patients receiving itraconazole, 200 mg twice daily. The horizontal lines through the bars represent the median concentrations; the upper and lower ends of the bars represent the 5th and 95th percentile, respectively; and the vertical lines extend to 1.5 times the interquartile range of concentrations. The number at the bottom of the bar represents the number of observations at that time after administration.

Adverse Events

Thirty-seven patients (88%) experienced grade 3 or 4 (severe or life-threatening) adverse events, but only 1 patient was removed from the trial because of itraconazole toxicity. Most of the adverse events were attributed to underlying HIV infection, opportunistic infections, or side effects of other medications. The patient who withdrew from the trial because of an adverse event experienced hypokalemia (potassium level, 2.6 mmol/L) despite potassium replacement. Grades 3 and 4 adverse events included the following: rash (2 patients); nausea (1 patient); vomiting (2 patients); hypertension (1 patient); hemoglobin concentration of less than 80 g/L (5 patients); a neutrophil count of less than 0.75 x 10⁹/L (15 patients); a platelet count of less than 50 x 10⁹/L (5 patients); an aspartate aminotransferase level of more than five times the upper limit of normal (2 patients); an alkaline phosphatase level more than five times the upper limit of normal (6 patients); a bilirubin level more than five times the upper limit of normal (4 patients); a creatinine level more than three times the upper limit of normal (6 patients); and hypokalemia (potassium level < 2.5 mmol/L) (2 patients). One patient developed cefuroxime-sodium-induced Stevens-Johnson syndrome. This patient had received itraconazole for 6 months without difficulty and had previously had a rash during ceftazidime therapy.

Discussion

Top Methods Results Discussion Author & Article Info References

In our study, itraconazole was highly effective as maintenance therapy for patients with AIDS who had had histoplasmosis. Ninety-five percent of the patients remained free of histoplasmosis during 97 patient-years of follow-up. Itraconazole was well tolerated; only one patient was withdrawn from the study because of toxicity.

Itraconazole appears to have similar and possibly superior efficacy and to be better tolerated than amphotericin B when used to prevent histoplasmosis relapse in patients with AIDS. In one study, relapse occurred in 19% of 21 patients who received amphotericin B for a median duration of 8 months [1]. In another study, 5% of 36 patients experienced relapse (median follow-up, >12 months) [2]. Amphotericin B doses in these two studies ranged from 50 to 100 mg administered weekly or every 2 weeks. These two studies were retrospective and were conducted before our study; the patients were similar to ours in most respects, making comparisons of outcomes of patients receiving maintenance therapies possible. In the two retrospective studies, all patients had completed amphotericin B induction therapy for histoplasmosis before beginning maintenance treatment, and histoplasmosis was the initial AIDS-defining infection in more than two thirds of cases. The CD4 counts were not reported for these historical control groups, but subsequent studies have shown that the median CD4 count was 0.033 to 0.047 x 10⁹/L, with counts less than 0.10 x 10⁹/L in more than 80% of patients (13; Wheat J. Unpublished observations; McKinsey DS. Unpublished observations). Although amphotericin B was highly effective as maintenance therapy, the inconvenience of administration and toxicity limit the popularity of amphotericin B among patients and physicians. Bacteremia, catheter infection, or thrombophlebitis occurred in nearly two thirds of patients receiving such therapy [11].

Itraconazole appears to be more effective than ketoconazole as maintenance therapy for histoplasmosis. Although relapse was observed in 10 of 20 patients (50%) who received ketoconazole as maintenance therapy in Indianapolis [1], these patients were not evaluated as carefully as the subsequent cohorts of patients who received amphotericin B [1, 2] or the patients who received itraconazole in our study. However, this experience is consistent with the results of a study by Johnson and colleagues [12], who observed histoplasmosis relapse in 4 of 11 patients (36%) receiving ketoconazole maintenance therapy. The limited effectiveness of ketoconazole for histoplasmosis in patients with AIDS is also supported by other studies: In one study, histoplasmosis occurred during ketoconazole treatment for other fungal infections in 5 of 36 patients (14%) [13], and in another study, 10 of 11 patients (91%) who received ketoconazole as induction therapy experienced treatment failure [1].

Survival after histoplasmosis in our study was relatively good compared with survival after other opportunistic infections in patients with AIDS. The median survival was 98 weeks; 81% of patients survived at least 1 year (CI, 69% to 93%), and 50% survived for more than 2 years (CI, 34% to 60%). For purposes of comparison, median survival was 34 weeks in patients receiving ketoconazole or amphotericin B maintenance therapy for cryptococcal meningitis [14], 37 to 55 weeks in patients receiving ganciclovir or foscarnet maintenance treatment for cytomegalovirus retinitis [15], and 83 weeks in patients receiving inhaled pentamidine maintenance treatment for Pneumocystis carinii pneumonia [16]. We believe that this favorable response was the result of early diagnosis, prompt induction treatment, and use of a convenient, well-tolerated, and effective maintenance treatment to prevent relapse.

Clearance of antigen provided a useful laboratory marker to evaluate efficacy. Although differences in study design complicate comparison of our results with those in historical controls treated with amphotericin B, antigen levels decreased at rates similar to those previously reported [17]. The rate of antigen clearance in serum for itraconazole recipients was 0.03 units/week and 0.05 units/week for amphotericin B recipients; corresponding rates of antigen clearance in urine were 0.16 units/week and 0.05 units/week. In our study, 1 of 42 patients (2%) showed an increasing urinary antigen level suggestive of relapse [18]. Specimens were not available for the patient who relapsed after withdrawal from the study, and the antigen level did not increase in the noncompliant patient with possible relapse.

Itraconazole therapy may have prevented the development of other serious fungal infections. No patient developed cryptococcal meningitis. The expected incidence in the United States is about 7% [14]. Six patients developed oropharyngeal candidiasis; in three of these patients, itraconazole could not be detected in plasma samples. The efficacy of itraconazole as prophylaxis for fungal infection can only be determined in a prospective, controlled trial.

Itraconazole was variably absorbed, but plasma levels greater than 0.75 µg/mL were achieved in more than 90% of patients. Although we could not determine the cause for variation in plasma concentrations, the timing of sampling after oral administration was not the cause. Poor compliance may have accounted for low plasma concentrations in some patients, but compliance was assessed as good at more than 95% of visits. Because the two relapses occurred in patients who were not taking their medication, we could not determine the blood level required to prevent relapse. We suggest that blood levels be measured if noncompliance or malabsorption is suspected during treatment with medications that accelerate the metabolism of or reduced absorption itraconazole, such as carbamazepine, barbiturates, or_histamine antagonists, and at the time of suspected relapse. Rifampin or phenytoin and itraconazole should not be used concurrently because rifampin and phenytoin profoundly lowers itraconazole concentrations [7]. Demonstration that itraconazole was undetectable in blood would provide useful information for modifications of therapy: education about the importance of compliance, treatment of the underlying cause for malabsorption, initiation of treatment to increase gastric acidity, withdrawal of interacting medications, elevation of itraconazole dose, or replacement of itraconazole with another antifungal agent.

Itraconazole was well tolerated. Although adverse events occurred in most patients, they were generally mild; only one patient withdrew from the study because of toxicity (hypokalemia). Severe hypokalemia (potassium levels < 2.5 mmol/L) occurred in two patients (5%) in our study. In other studies in which similar doses of itraconazole were administered for more than 6 months, fewer than 5% of patients withdrew because of toxicity [19].

In conclusion, itraconazole appears to be a safe and effective alternative to amphotericin B for prevention of histoplasmosis relapse in patients with AIDS. Although the efficacy of itraconazole in patients with central nervous system involvement is unknown, studies of patients with cryptococcal meningitis suggest that it may prevent recurrence of Histoplasma meningitis as well. Further studies of itraconazole are needed: In addition to its use as preventive therapy in patients with AIDS who have had histoplasmosis, itraconazole is an excellent candidate for induction therapy in such patients.

Appendix

Other contributors to the study were Mary Culnane, BS, National Institutes of Health, Bethesda, Maryland; Judith Deutsch, MD and Julie Relue, RN, Indiana University School of Medicine, Indianapolis, Indiana; Judy Johnson, MS, Research Triangle Institute, Raleigh, North Carolina; Robert Legendre, Janssen Research Foundation, Titusville, New Jersey; Robert Larsen, MD sup, University of Southern California, Los Angeles, California; Rebecca Clark, MD, Tulane University School of Medicine, New Orleans, Louisiana; Peter Frame, MD, University of Cincinnati Medical Center, Cincinnati, Ohio; Margaret Ragni, MD,^sup University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; and Sarah Cheeseman, MD, University of Massachusetts Medical Center, Worcester, Massachusetts.