Amphotericin

Amphotericin deoxycholate (DAMB)

The adverse effects of amphotericin deoxycholate have been reviewed in a retrospective analysis of 102 adult patients (median age 61 years) with a variety of underlying conditions who were admitted to a small community hospital in Honolulu and who received the drug for treatment of presumed or proven fungal infections that were mostly due to Candida species [ ]. The average total dose of amphotericin deoxycholate was comparatively low at 162 (range 10–840) mg. The initial dose averaged 16 (range 1–50) mg and the total duration of therapy was 8.3 (range 1–46) days. Chills, fever, and/or nausea were noted in 25% of the patients. Hypokalemia (a serum potassium concentration below 3.5 mmol/l) occurred in 19%, and nephrotoxicity (defined as a serum creatinine concentration of at least 141 μmol/l (1.6 mg/dl) with an increase of at least 44 μmol/l (0.5 mg/dl) during amphotericin deoxycholate therapy) in 15% of the patients. Nephrotoxicity increased with increasing total dose of amphotericin, while infusion-associated toxicity decreased with advancing age. The overall response rate to therapy with amphotericin deoxycholate was 83%.

Amphotericin colloidal dispersion (ABCD)

The safety and efficacy of amphotericin colloidal dispersion have been evaluated in 148 immunocompromised patients with candidemia [ ]. ABCD was given intravenously in a median daily dose of 3.9 (range 0.1–9.1) mg/kg for a median of 12 (range 1–72) days. In the safety analysis (n = 148 patients), nephrotoxicity occurred in 16% of the patients, with either doubling of the baseline serum creatinine concentration or an increase of 88 μmol/l (1.0 mg/dl) or a 50% fall in calculated creatinine clearance. Severe adverse events were believed to be probably or possibly related to ABCD in 36 patients (24%), including chills and fever (9.5%), hypotension and abnormal kidney function (4%), tachycardia, asthma, hypotension (3%), and dyspnea (2%). ABCD was withdrawn in 12% because of toxicity. The overall response rate in 89 evaluable patients was 66% with candidemia alone and 14% with disseminated candidiasis.

The safety and efficacy of ABCD have been studied in 133 patients with invasive fungal infections and renal impairment due to either amphotericin deoxycholate or pre-existing renal disease [ ]. The mean daily dose of ABCD was 3.4 (range 0.1–5.5) mg/kg, and the mean duration of therapy was 21 (range 1–207) days. Although individual patients had increases in serum creatinine concentrations, ABCD did not have an adverse effect on renal function: the mean serum creatinine concentration tended to fall slightly with days on therapy, and increases were not dose-related. Six patients discontinued ABCD therapy because of nephrotoxicity. Infusion-related adverse events occurred at least once in 74 patients (56%); however, while 43% of patients had infusion-related toxic effects on day 1, only 18% reported these events by day 7. There were complete or partial responses in 50% of the intention-to-treat population and in 67% of the 58 evaluable patients.

The safety of ABCD has been reviewed using data from 572 immunocompromised patients refractory to or intolerant of standard therapies enrolled in five phase I/II clinical trials [ ]. The mean daily dose of ABCD was 3.85 (median 3.8, range 0.1–9.1) mg/kg and the mean duration of treatment was 25 (median 16, range 1–409) days. Overall, the principal adverse events associated with ABCD therapy were chills (52%), fever (39%), and hypotension (19%). These infusion-related reactions were dose-related and were the dose-limiting adverse events, defining the maximum tolerated dosage at 7.5 mg/kg. ABCD did not adversely affect renal function, as measured by overall changes in serum creatinine from baseline to the end of therapy, even in patients with pre-existing renal impairment; only 3.3% of patients discontinued therapy because of nephrotoxicity. Complete or partial responses to treatment were reported in 149 of 260 evaluable patients (57%).

The safety and efficacy of ABCD have been studied in 220 bone marrow transplant recipients enrolled in the same five phase I or phase II studies [ ]. The median dose in this population was 4 (range 0.4–8.0) mg/kg, and the median duration of treatment was 16 (range 1–409) days. Overall, 37 (19%) of the patients had nephrotoxicity, defined as a doubling of serum creatinine from baseline, an increase of 88 μmol/l from baseline, or at least a 50% fall in calculated creatinine clearance. There were no significant changes in hepatic transaminases, alkaline phosphatase, or total bilirubin. Fever and chills were reported by 12% and 11% of patients respectively. Other acute, severe, infusion-related adverse events were hypoxia (4.1%), hypertension (2.7%), and hypotension (2.7%).

Mucormycosis has an exceedingly high mortality rate in immunocompromised patients. In five phase I and phase II studies of ABCD, 21 patients were given ABCD (mean dose 4.8 mg/kg per infusion for a mean duration of 37 days) on the basis of pre-existing renal insufficiency, nephrotoxicity during amphotericin B therapy, or refractory infections [ ]. Of 20 evaluable patients, 12 responded to ABCD, and there was no renal or hepatic toxicity. However, a previous randomized, comparative trial showed an at least similar if not increased frequency and severity of infusion-related reactions compared with conventional amphotericin B [ ].

Amphotericin B lipid complex (ABLC)

The safety and efficacy of ABLC have been evaluated in 556 cases of proven or presumptive invasive fungal infection treated in an open, single-patient, US emergency-use study of patients who were refractory to or intolerant of conventional antifungal therapy [ ]. The daily dosage was either 5 mg/kg (87%) or 3 mg/kg. The investigators had the option of reducing the daily dosage as clinically warranted. Treatment was for 7 days in 540 patients (97%). During the course of ABLC therapy, serum creatinine concentrations in all patients fell significantly from baseline. In 162 patients with serum creatinine concentrations of at least 221 μmol/l (2.5 mg/dl) at baseline, the mean serum creatinine concentration fell significantly from the first week to the sixth week. The serum creatinine concentration increased from baseline to the end of therapy in 132 patients (24%). Hypokalemia (serum potassium concentration of less than 3 mmol/l) developed in 4.6%, and hypomagnesemia (serum magnesium concentration of less than 0.75 mmol/l) in 18%. There was a rise in serum bilirubin in 142/284 patients (33%); the overall increase was from 79 to 112 μmol/l (4.66–6.59 mg/dl) at the end of therapy. The mean alkaline phosphatase activity rose from 273 to 320 IU/l. There was no significant change overall in alanine transaminase activity, but the activity increased by the end of treatment in 16% of patients with initially normal values. There were complete or partial responses to therapy with ABLC in 167 of 291 mycologically confirmed cases evaluable for therapeutic response (57%).

The safety and efficacy of ABLC 5 mg/kg/day in patients with neutropenia and intolerance or refractoriness to amphotericin deoxycholate have been reported in two smaller series of 25 treatment courses from the UK. In one, the mean serum creatinine at the start of therapy was 139 μmol/l and at the end of therapy 132 μmol/l; there were no infusion-related adverse events [ ]. There was an increase in alanine transaminase activity in 12 of the 22 analysed treatment courses. In the other, there was an increase in serum creatinine in 5 of 18 courses (28%), and hypokalemia (less than 2.5 mmol/l) in two courses (11%); premedication for infusion-associated reactions was required in three courses (17%) [ ]. There were modest increases in serum alanine transaminase activities in five patients (30%).

In contrast to these reports, there was a high prevalence of adverse events with ABLC in the treatment of suspected or documented invasive fungal infections in 19 Scandinavian patients with mostly hematological malignancies [ ]. The mean starting dose of ABLC was 4.1 mg/kg/day, given for a median of 3 (range 1–19) days. ABLC was withdrawn because of adverse events in 14/19 patients (74%). These included rising creatinine concentrations (n = 12), increased serum bilirubin (n = 7), erythema (n = 6), increased alanine transaminase (n = 6), fever and chills (n = 5), hypoxemia (n = 3), hemolysis (n = 2), and back pain and increased serum alkaline phosphatase activity (n = 1 each). In patients with renal adverse effects, there were significantly increased serum creatinine concentration (from 85 to 199 mmol/l) and increased bilirubin concentration (from 17 to 77 μmol/l) in seven patients. The authors stated that while all the patients were very ill at the time of the start of ABLC therapy, in all cases the adverse effects had a direct and obvious correlation with the administration of ABLC. However, the reason for this unusual high rate of adverse events remains unclear.

Safety data have been published in a retrospective analysis of 551 patients with invasive fungal infections intolerant of or refractory to conventional antifungal therapy, 73 of whom received ABLC initially at 3 mg/kg/day instead of 5 mg/kg/day, as recommended in the protocol [ ]. There were no notable differences in adverse events (increased serum creatinine, infusion-related chills) between the two groups. Serum creatinine values were improved or stable at the end of therapy in 78% and 70% of patients respectively.

Two smaller series have addressed the safety of ABLC in immunocompromised patients [ , ]. Each included about 30 patients who were treated with median dosages of 4.8 and 5.0 mg/kg for a median of 8 and 14 days. In contrast to a previous retrospective analysis that showed a 74% withdrawal rate, mostly due to infusion-related reactions [ ], ABLC was well tolerated, with withdrawal rates of 6% and 0% and an overall trend for stable or improved serum creatinine values at the end of therapy. Similarly, 13 infusion-related reactions have been reported among 308 infusions in four of ten patients with hematological malignancies receiving ABLC 3 mg/kg/day [ ]. These reactions (fever, rigors, myalgias), occurred during the first infusions, were judged to be mild, and resolved during later infusions. ABLC was well tolerated by 30 persistently febrile neutropenic patients with hematological malignancies who received it in a low dosage of 1 mg/kg/day for a median of 7.5 (range 2–19) days [ ]. Seven patients (23%) had mild to moderate infusion-related reactions, and no patient had nephrotoxicity. In one patient, ABLC was discontinued owing to intolerable infusion-related fever and chills.

The safety and efficacy of low dose ABLC (1 mg/kg/day) for empirical treatment of fever and neutropenia have been studied in 69 episodes in 61 patients with hematological malignancies [ ]. The median duration of therapy was 8 (range 2–19) days and 13 patients had mild to moderate infusion-related adverse events. Creatinine concentrations remained stable in 42 cases, improved in 9, and deteriorated in 18. There were no other toxic effects. The response rate (resolution of fever during neutropenia and absence of invasive fungal infection) was 67%.

Fungal infection remains an important cause of morbidity and mortality in lung transplant patients. In a prospective non-comparative evaluation in lung or heart–lung transplant recipients, ventilated patients received undiluted aerosolized ABLC 100 mg, and extubated patients received 50 mg; in all, 381 treatments were given (98 in ventilated patients and 283 in extubated patients) [ ]. The treatment was administered by face mask jet nebulizer with compressed oxygen at a flow rate of 7–8 l/minute and inhaled over 15–30 minutes. Treatments were delivered once every day for four consecutive days, then once a week for 2 months. In all, 381 treatments were given to 51 patients, and ABLC was subjectively well tolerated in 98%. Pulmonary function worsened by 20% or more in under 5% of all treatments. There were no significant adverse events.

In a retrospective comparison of outcomes in liver transplant recipients with invasive aspergillosis who received amphotericin B lipid complex (ABLC) or conventional amphotericin B, the 60-day mortality rate was lower in the ABLC cohort: four of 12 patients versus 24 of 29 patients. Only one of four ABLC recipients with definite invasive aspergillosis died, compared with all 11 in the amphotericin B group. The 60-day survival probability curves was significantly lower in the amphotericin B group. ABLC therapy was the only independent mortality-protective variable (OR = 0.31; 95% CI = 0.07, 0.44) [ ]. In a prospective historical control study 131 patients with acute myelogenous leukemia or high-risk myelodysplastic syndrome undergoing induction chemotherapy were given ABLC 2.5 mg/kg intravenously 3 times weekly as antifungal prophylaxis and were compared with 70 who had previously received LAMB 3 mg/kg 3 times weekly [ ]. Grade 3 and 4 adverse events (hyperbilirubinemia, 3% versus 6%; infusion-related adverse events, 3% versus 7%) were statistically similar between the groups, as were rates of withdrawal because of adverse events (18% versus 15%) and survival rates (92% versus 86%).

ABLC has shown promise in lung transplant recipients as a convenient means of delivering protective drug to the upper airways, avoiding systemic adverse effects [ ]. Aerosolized ABLC in 40 subjects undergoing allogeneic hemopoietic stem cell transplantation has been prospectively investigated in an open, non-comparative study. ABLC was given once daily for 4 days then once weekly for 13 weeks in addition to systemic fluconazole. Cough, nausea, taste disturbances, or vomiting occurred in 2.2% of 458 administrations of inhaled ABLC; 5.2% of administrations were associated with at least a 20% reduction in pulmonary function (FEV ₁ or FVC), but none required treatment with bronchodilators or withdrawal from the study. Four mild adverse events were considered possibly or probably related to treatment; no deaths or withdrawals were attributed to treatment. Of three proven invasive fungal infections that occurred during the study, only one, a catheter-related case of disseminated fusariosis, occurred while the subject was taking the study medication.

The risk of hematological, renal, and hepatic toxicity associated with ABLC has been assessed in a multicenter, open, non-comparative study in 93 patients from 17 different hospitals who received ABLC because of proven or suspected systemic fungal infection or leishmaniasis [ ]. Optimum treatment with ABLC comprised a 2-hour infusion of 5 mg/kg/day for a minimum of 14 days; the mean dose was 235 (range 9–500) mg/day and the total cumulative dose was 2894 (range 30–10 200) mg. In the whole group, the mean serum creatinine concentration was similar before and after ABLC (1.00 versus 1.20 mg/dl). There were no significant changes in concentrations of hemoglobin, potassium, and bilirubin, or in hepatic transaminase activities. There was no significant correlation between the dose and the serum creatinine concentrations. There was no greater nephrotoxicity in the patients with previous renal insufficiency, or in those who had previously received amphotericin B. There were serious adverse events in five patients, but other alternative causes were found in three of them. There were fevers or chills in 23% of the patients during the infusion of ABLC, but only in one case did this necessitate withdrawal. The overall rate of withdrawals due to adverse events was 9.7%.

Liposomal amphotericin (LAmB)

The safety, tolerance, and pharmacokinetics of liposomal amphotericin have been evaluated in an open, sequential dose escalation, multiple-dose phase I/II study in 36 patients with neutropenia and persistent fever requiring empirical antifungal therapy [ ]. The patients received doses of 1, 2.5, 5.0, or 7.5 mg/kg/day of liposomal amphotericin for a mean of 9.2 days. Liposomal amphotericin was well tolerated: infusion-related adverse effects (fever, chills, rigor) occurred in 15 (5%) of all 331 infusions, and only two patients (5%) required premedication (dyspnea and generalized flushing; facial urticaria). Hypotension (one infusion) and hypertension (three infusions) were infrequent. One patient each had sharp flank pain and dyspnea during one infusion; these symptoms did not recur during subsequent infusions. Serum creatinine, potassium, and magnesium concentrations were not significantly changed from baseline, and there were no net increases in serum transaminases. There was, however, a significant increase in serum alkaline phosphatase activity and increase in bilirubin concentration in the overall population as well as in individual dosage groups. One patient who received concomitant l -asparaginase had increases in serum lipase and amylase activities in association with symptoms of pancreatitis while receiving liposomal amphotericin; however, as he continued to receive the drug, the serum lipase and amylase returned to baseline. Liposomal amphotericin had non-linear pharmacokinetics consistent with reticuloendothelial uptake and redistribution. There were no breakthrough fungal infections during therapy.

The efficacy of two dosages of liposomal amphotericin in the treatment of proven or probable invasive aspergillosis in neutropenic patients with cancer or those undergoing bone marrow transplantation has been studied in a prospective, randomized, open, multicenter trial in 120 patients randomized to receive either 1 mg/kg/day or 4 mg/kg/day of liposomal amphotericin; 87 patients were available for evaluation [ ]. There was at least one toxic event during treatment in 15 of 41 patients given 1 mg/kg/day and 25 of 46 given 4 mg/kg/day, but the numbers of events per patient were similar. These events included headache, nausea, diarrhea, rash, liver toxicity, myalgia, dyspnea, fever, chills, and back pain. Renal toxicity definitely related to liposomal amphotericin occurred in 1/41 patients treated with 1 mg/kg/day and 5/46 patients treated with 4 mg/kg/day. Only in one case was treatment permanently discontinued because of toxicity related to liposomal amphotericin (4 mg/kg/day). No patient died from liposomal amphotericin toxicity. Overall, liposomal amphotericin was effective in 50–60% of patients; however, the number of cases with proven invasive aspergillosis was too small to allow a meaningful comparison of the two dosages regarding efficacy in this life-threatening disease.

The safety and efficacy of liposomal amphotericin have been compared with that of ABLC in a retrospective analysis of 59 adult patients with hematological malignancies who received 68 courses of either liposomal amphotericin (n = 32) or ABLC (n = 36) for a variety of presumed or confirmed invasive fungal infections [ ]. The median daily dosages were 1.9 (range 0.7–4.0) mg/kg for liposomal amphotericin and 4.8 (range 1.9–5.8) mg/kg for ABLC. There was no statistically significant difference in the overall outcome; febrile reactions were significantly more common with ABLC (36% versus 6%), but there were no significant differences in the median creatinine concentrations at baseline and at the end of therapy or in the number of patients with urinary loss of potassium or magnesium.

In an open, sequential phase II clinical study of three different regimens of liposomal amphotericin for visceral leishmaniasis (2 mg/kg on days 1–6 and on day 10; 2 mg/kg on days 1–4 and on day 10; 2 mg/kg on days 1, 5, and 10) in Indian and Kenyan patients in three developing countries, there were few infusion-associated adverse effects [ ]. Of 32 Brazilian patients (15 of whom received 2 mg/kg on days 1–10 because of poor responses to the first regimen, 37% had a fever with one or more infusions, 9% had chills, and 6% had back pain; in addition, three patients had respiratory distress and/or cardiac dysrhythmias. There were different response rates to the three regimens in the different countries, leading to the recommendation of 2 mg/kg on days 1–4 and day 10 in India and Kenya, and 2 mg/kg on days 1–10 in Brazil.

In order to determine the maximum tolerated dosage of liposomal amphotericin B, a phase I/II study was conducted in 44 adult patients with proven (n = 21) or probable (n = 23) infections due to Aspergillus species and other filamentous fungi [ ]. The dosages were 7.5, 10, 12.5, and 15 mg/kg/day. The number of infusions was 1–83 with a median duration of 11 days. The maximum tolerated dosage was at least 15 mg/kg. Infusion-related reactions included fever in 8 and chills and rigors in 5 of 43 patients. Three patients developed a syndrome of substernal chest tightness, dyspnea, and flank pain, which was relieved by diphenhydramine. Serum creatinine increased two times above baseline in 32% of patients, but this increase was not dose-related. Hepatotoxicity developed in one patient. Altogether, the most common adverse events included fever (48%), increased creatinine concentration (46%), hypokalemia (39%), chills (32%), and abdominal pain (25%), with no obvious dose-dependency. Nine patients (20%) stopped taking the drug because of an adverse event. The reasons included raised serum creatinine, renal insufficiency, pancreatitis, hyperbilirubinemia, hypotension associated with the infusion, cardiorespiratory failure, multiorgan failure, and relapse of the primary malignancy. The last three events were attributed to the underlying disease process. Discontinuation was unrelated to dosage. Pharmacokinetic analysis showed dose-related non-linear kinetics at dosages of 7.5 mg/kg/day and over.

A dermatosis commonly known as post-kala-azar dermal leishmaniasis can develop after treatment of human visceral leishmaniasis. In about 15% of cases the disfiguring lesions persist, sometimes for many years. The usefulness of LAMB 2.5 mg/kg/day for 20 days in the treatment of persistent post-kala-azar dermal leishmaniasis has been evaluated in 12 Sudanese subjects, who were regularly screened for adverse effects; LAMB completely cleared the rash in 10 (83%) of the patients and caused no detectable adverse effects [ ].

Liposomal amphotericin B (L-AmB) has been studied in an open study with a combination of fluconazole + itraconazole as prophylaxis in patients undergoing induction chemotherapy for acute myelogenous leukemia and myelodysplastic syndrome [ ]. Patients were randomized to receive either fluconazole 200 mg orally every 12 hours + itraconazole 200 mg orally every 12 hours (n = 67) or L-AmB 3 mg/kg intravenously 3 times a week (n = 72). Altogether, 47% of the patients completed antifungal prophylaxis without a change in therapy for proven or suspected fungal infection. Three patients in each arm developed a proven fungal infection. Because of persistent fever 23% of those treated with L-AmB and 24% of those treated with fluconazole + itraconazole were changed to alternative antifungal therapy. Increases in serum creatinine concentrations to over 177 μmol/l (2 mg/dl) (20% versus 6%) and increases in serum bilirubin concentrations to over 2 mg/dl (43% versus 22%) were more common with L-AmB. There were infusion-related reactions in five patients who received L-AmB. Responses to chemotherapy and induction mortality rates were similar in the two arms. Thus, while L-AmB and fluconazole + itraconazole appeared to have similar efficacy, L-AmB was associated with higher rates of increased serum bilirubin and creatinine concentrations.

Response rates and the adverse effects of treatment with L-AmB have been assessed in a phase IV cohort study in 406 patients aged 1 day to 77 years, of whom 83% had malignancies and 66 % had fever of unknown origin [ ]. The mean duration of treatment was 20 days and the mean daily dose 2.3 mg/kg. There was either a complete or partial response in 314 patients (77%). There were drug-related adverse events in 94 patients (23%). Among these, hypokalemia (6.2%) and abnormal liver function tests (5.2%) were the most common; there was nephrotoxicity in 17 patients (4.2%).

Weekly prophylactic high-dose L-AmB 7.5 mg/kg has been investigated in 21 adults receiving high-dose prednisone 2 mg/kg/day for acute graft-versus-host disease after allogeneic hemopoietic stem cell transplantation [ ]. Patients received a median of 4 (range 1–8) infusions of L-AmB. Seven withdrew because of drug-related adverse events, including raised serum creatinine (> 1.5 times from baseline; n = 5), hypotension and pain (n = 1), and severe chest pain and dysrhythmias (n = 1). The overall frequency of infusion-related reactions was 29% (n = 6), but these reactions were always transient and relieved by stopping the infusion.

In an open, prospective pilot study in adults receiving chemotherapy for acute leukemia (n = 21) or allogeneic hemopoietic stem cell transplantation (n = 8), the former received weekly infusions of L-AmB 10 mg/kg for 4 weeks for and the latter 10 mg/kg for 8 weeks [ ]. The most frequent drug-related adverse events were infusion-related reactions, 12 of which (of a total of 76 infusions) led to increased infusion duration for better tolerance. No adverse events led to withdrawal of prophylactic treatment in the patients with acute leukemia. In those with hemopoietic stem cell transplants, eight adverse events (in six patients) were reported to be related to the study treatment and led to withdrawal.

In 164 HIV-negative children (median age 1.6 years; range 4 months to 14 years) with Mediterranean visceral leishmaniasis L-AmB 3 mg/kg given on days 1–5 and 10 was not associated with adverse events [ ].

Aerosolized amphotericin

In a prospective, randomized, multicenter trial, inhalation of aerosolized amphotericin (10 mg bd) has been investigated as prophylaxis against invasive aspergillosis in 382 cancer patients with an anticipated duration of neutropenia of at least 10 days [ ]. While there was no difference in the incidence of invasive aspergillosis, infection-related mortality, and overall mortality, 31% of the patients discontinued amphotericin prophylaxis prematurely owing to adverse effects (55%; most commonly cough, bad taste, nausea), inability to cooperate further (30%), violation of the study protocol (11%), and non-adherence (4%).

Comparisons of different formulations of amphotericin

Amphotericin deoxycholate in glucose versus amphotericin deoxycholate in Intralipid : The safety of two formulations of intravenous amphotericin deoxycholate has been investigated in a randomized, open comparison in neutropenic patients with refractory fever of unknown origin or pulmonary infiltrates [ ]. Amphotericin deoxycholate was given in a dose of 0.75 mg/kg/day either in 250 ml of a 5% glucose solution or mixed with 250 ml of a 20% lipid emulsion (Intralipid 20%) on eight consecutive days and then on alternate days as a 1–4 hour infusion. The mean number of days of treatment was 11.3 versus 9.9 days. There were no statistically significant differences between the two cohorts with respect to the incidence of infusion-related adverse events, such as fever and chills, renal impairment, or treatment failure. However, grade 3–4 acute dyspnea occurred slightly more often with the lipid emulsion formulation, and there were significantly more other severe respiratory events in patients receiving lipid emulsion, raising the possibility of a causal relation via fat overload or incompatibility between amphotericin deoxycholate and the lipid emulsion.

The efficacy and tolerability of amphotericin prepared in Intralipid 20% have been evaluated in 16 patients with HIV infection and esophageal candidiasis or cryptococcosis and compared with standard amphotericin in a matched group of 24 patients [ ]. While both formulations had apparently similar clinical and microbiological efficacy, fewer patients receiving the lipid emulsion formulation required premedication or symptomatic therapy for infusion-associated adverse events, and fewer patients were withdrawn because of adverse effects. Renal adverse effects (a rise in serum creatinine and/or electrolyte loss) were more common in patients who received the conventional formulation.

The efficacy and safety of amphotericin in Intralipid 20% or 5% glucose has been evaluated in a retrospective case analysis in 30 patients with AIDS and cryptococcal meningitis who received either formulation 1 mg/kg/day for 20 days with or without flucytosine (n = 20) or fluconazole (n = 4), followed by maintenance therapy with fluconazole 400 mg/day [ ]. Twenty patients received amphotericin deoxycholate in 500 ml 5% glucose over 5 hours, and 10 received amphotericin deoxycholate in 100 ml of 20% Intralipid given over 2 hours. Complete clinical resolution was obtained in 55% and 60% of the patients respectively. There were no differences regarding infusion-related adverse effects, nephrotoxicity, or anemia.

Amphotericin deoxycholate versus amphotericin B colloidal dispersion : Amphotericin colloidal dispersion has been compared with amphotericin deoxycholate in a prospective, randomized, double-blind study in the empirical treatment of fever and neutropenia in 213 patients [ ]. Patients were stratified by age and concomitant use of ciclosporin or tacrolimus and then randomized to receive ABCD (4 mg/kg/day) or amphotericin deoxycholate (0.8 mg/kg/day) for 14 days. Renal dysfunction was less likely to develop and occurred later with ABCD than with amphotericin deoxycholate. Likewise, the absolute and percentage fall in the serum potassium concentration from baseline to the end of therapy was greater with amphotericin deoxycholate than ABCD. However, probable or possible infusion-related hypoxia and chills were more common with ABCD than amphotericin deoxycholate. There was a therapeutic response in 50% of the patients who received ABCD and 43% of those who received amphotericin deoxycholate. Thus, ABCD was of comparable efficacy and less nephrotoxic than amphotericin deoxycholate, but infusion-related events were more common with ABCD.

Amphotericin deoxycholate versus liposomal amphotericin : Liposomal amphotericin 5 mg/kg/day and amphotericin deoxycholate 1 mg/kg/day have been compared in the treatment of proven or suspected invasive fungal infections in neutropenic patients in a randomized, multicenter study [ ]. Significantly more patients given amphotericin deoxycholate had a greater than 100% increase in baseline serum creatinine. Treatment was temporarily discontinued or the dosage reduced because of an increase in serum creatinine in 18/54 (33%) patients treated with amphotericin deoxycholate versus 2/51 (4%) treated with liposomal amphotericin. There was no statistically significant difference in the number of patients with infusion-related toxicity (fever/chills), hypokalemia, or increases in serum transaminases, alkaline phosphatase, or serum bilirubin. In 66 patients eligible for analysis of efficacy, there was a trend to an improved overall response rate and a significant difference in the rate of complete responses in favor of liposomal amphotericin; death rates were also lower in patients treated with liposomal amphotericin.

The results of a randomized, double-blind, multicenter comparison of liposomal amphotericin (3.0 mg/kg/day) with conventional amphotericin deoxycholate (0.6 mg/kg/day) for empirical antifungal therapy in patients with persistent fever and neutropenia have been reported [ ]. The mean duration of therapy was 10.8 days for liposomal amphotericin (343 patients) and 10.3 days for amphotericin deoxycholate (344 patients). While the composite rates of successful treatment were similar (50% for liposomal amphotericin and 49% for amphotericin deoxycholate), significantly fewer of the patients who received the liposomal preparation had infusion-related fever (17% versus 44%), chills or rigors (18% versus 54%), or other reactions, including hypotension, hypertension, and hypoxia. Nephrotoxicity (defined by a serum creatinine concentration twice the upper limit of normal) was significantly less frequent among patients treated with liposomal amphotericin (19%) than among those treated with conventional amphotericin deoxycholate (34%).

Aerosolized DAMB 50 mg and ABLC 25 mg have been compared in lung transplant recipients in a prospective, randomized, double-blind trial in 100 subjects [ ]. The study drug was withdrawn because of intolerance in six of 49 patients treated with DAMB and three of 51 treated with ABLC. Those who received DAMB were more likely to have had an adverse event (OR = 2.16, 95% CI = 1.10, 4.24).

Amphotericin lipid complex versus liposomal amphotericin : Liposomal amphotericin and ABLC have been compared in an open randomized study in 75 adults with leukemia and 82 episodes of suspected or documented mycosis [ ]. The median durations of treatment and dosages were 15 days at 4 mg/kg/day for liposomal amphotericin and 10 days at 3 mg/kg/day for ABLC. Acute but not dose-limiting infusion-related adverse events occurred in 36% versus 70%. Bilirubin increased to over 1.5 times baseline in 59% versus 38%. There was no difference in the effects of either agent on renal function and drug-related withdrawals. The overall response rate to therapy in documented fungal infections (29% and 30% respectively) was not different between the two drugs.

Amphotericin deoxycholate in glucose versus amphotericin in nutritional fat emulsion : The safety of DAMB prepared in nutritional fat emulsion (a non-approved mode of amphotericin administration) has been reviewed [ , ]. It is not clear whether it has a better therapeutic index than other formulations, and methods of preparing it have not been standardized. The adverse effects of amphotericin prepared in nutritional fat emulsion have been compared with those of amphotericin prepared in 5% dextrose in two studies. While one of the studies showed a significantly lower frequency of infusion-related reactions and hypokalemia in patients receiving the fat emulsion [ ], there were no differences in safety and tolerance between the two formulations in the other study [ ]. The safety of amphotericin prepared in nutritional fat emulsions has been reviewed [ , ]. Because of stability concerns and lack of systematic safety data, this form of amphotericin cannot be recommended.

Aerosolized liposomal amphotericin versus deoxycholate amphotericin : Aerosolized liposomal amphotericin and deoxycholate amphotericin B have been retrospectively compared in 38 consecutive lung transplant recipients [ ]. In all, 1206 doses of DAMB and 1149 doses of L-AmB were administered; 18 patients received DAMB only, 11 received L-AmB only, and 9 received the two medications sequentially. The total numbers of complaints were 1.0% of doses of DAMB and 1.2% of doses of L-AmB. There were no differences between the groups on lung biopsy specimens. Plasma amphotericin concentrations were 0.2–0.9 mg/l with DAMB and under 0.2 mg/l with L-AmB.

Antifungal azoles

Fluconazole : There has been an open, randomized comparison of amphotericin deoxycholate 0.5 mg/kg/day intravenously versus fluconazole 400 mg/day orally for empirical antifungal therapy in neutropenic patients with cancer and fever refractory to broad-spectrum antibiotics [ ]. Patients with abnormal hepatic or renal function were excluded, as were those with proven or suspected invasive fungal infection. The mean duration of therapy was 8.3 days with amphotericin deoxycholate and 7.9 days with fluconazole. Altogether, 32/48 patients randomized to amphotericin deoxycholate and 19/52 randomized to fluconazole had adverse effects (67% versus 36%). Two patients developed immediate hypersensitivity reactions (flushing, hypotension, bronchospasm) to amphotericin deoxycholate and had to be withdrawn. Hypokalemia was noted in 25 patients (52%), and nephrotoxicity, defined as a rise in serum creatinine of 44 μmol/l (0.5 mg/dl) or more compared with the baseline value, in nine patients (19%). The corresponding frequencies with fluconazole were 23% and 6% respectively. Treatment success rates and mortality were similar (46% versus 56% and 33% versus 27% respectively).

Fluconazole and amphotericin as empirical antifungal drugs in febrile neutropenic patients have been investigated in a prospective, randomized, multicenter study in 317 patients randomized to either fluconazole (400 mg qds) or amphotericin deoxycholate (0.5 mg/kg qds) [ ]. Adverse events (fever, chills, renal insufficiency, electrolyte disturbances, and respiratory distress) occurred significantly more often in patients who were given amphotericin (128/151 patients, 81%) than in those given fluconazole (20/158 patients, 13%). Eleven patients treated with amphotericin, but only one treated with fluconazole, were withdrawn because of an adverse event. Overall mortality and mortality from fungal infections were similar in both groups. There was a satisfactory response in 68% of the patients treated with fluconazole and 67% of those treated with amphotericin. Thus, fluconazole may be a safe and effective alternative to amphotericin for empirical therapy of febrile neutropenic patients; however, since fluconazole is ineffective against opportunistic molds, the possibility of an invasive infection by a filamentous fungus should be excluded before starting empirical therapy. Similarly, patients who take azoles for prophylaxis are not candidates for empirical therapy with fluconazole.

Conventional amphotericin deoxycholate (0.2 mg/kg qds) and fluconazole (400 mg qds) have been compared in a prospective randomized study in 355 patients with allogeneic and autologous bone marrow transplantation [ ]. The drugs were given prophylactically from day 1 until engraftment. There was no difference in the occurrence of invasive fungal infections, but amphotericin was significantly more toxic than fluconazole, especially in related allogeneic transplantation, after which 19% of patients developed toxicity compared with none of those who received fluconazole.

Itraconazole : Amphotericin and itraconazole have been compared in a multicenter, open, randomized study in 277 adults with cancer and neutropenia [ ]. Itraconazole oral solution (100 mg bd, n = 144) was compared with a combination of amphotericin capsules and nystatin oral suspension (n = 133). Adverse events were reported in about 45% of patients in each group. The most frequent were vomiting (14 versus 12 patients), diarrhea (12 versus 9 patients), nausea (5 versus 12 patients), and rash (2 versus 13 patients). There were no differences in liver function test abnormalities. Treatment had to be withdrawn because of adverse events (including death) in 34 patients who took itraconazole and 33 of those who took amphotericin plus nystatin; there were 17 deaths in each group and death was recorded as adverse event in 13 and nine patients respectively.

Intravenous amphotericin deoxycholate (0.7–1.0 mg/kg) and itraconazole (400 mg intravenously for 2 days, 200 mg intravenously for up to 12 days, then 400 mg/day orally) have been compared in 384 granulocytopenic patients with persistent fever in a randomized, multicenter trial [ ]. The median duration of therapy was 8.5 days. The incidence of drug-related adverse events (54% versus 5%) and the rate of withdrawal due to toxicity (38% versus 19%) were significantly higher with amphotericin. The most frequent reasons for withdrawal in patients taking itraconazole were nausea and vomiting (5%), rash (3%), and abnormal liver function tests (3%). Significantly more of the patients who received amphotericin had nephrotoxicity (24% versus 5%); however, fewer had hyperbilirubinemia (5% versus 10%). There was no difference in gastrointestinal adverse events between the two groups.

Amphotericin in capsules 500 mg qds has been compared with itraconazole elixir 2.5 mg/kg bd for the prophylaxis of systemic and superficial fungal infections in a double-blind, randomized, placebo-controlled, multicenter trial for 1–59 days [ ]. While itraconazole significantly reduced the frequency of superficial fungal infections, it was not superior in reducing invasive fungal infections or in improving mortality. Adverse events were reported in 222 patients taking itraconazole (79%) and in 205 patients taking amphotericin (74%). The commonest adverse events were gastrointestinal, followed by rash and hypokalemia, with no differences between the two regimens. In both groups, 5% of the adverse events were considered to be definitely drug-related. Comparable numbers of patients in the two groups permanently stopped treatment because of adverse events (including death), 78 (28%) in the amphotericin group and 75 (27%) in the itraconazole group. Nausea (11% and 9%) and vomiting (7% and 8%) were the most frequently reported adverse events that led to withdrawal. Biochemical changes were comparable in the two groups.

Amphotericin and itraconazole have been compared in empirical antifungal drug treatment of febrile neutropenia in an open, randomized study in 162 patients who received either intravenous itraconazole followed by oral itraconazole suspension or intravenous amphotericin for a maximum of 28 days [ ]. Itraconazole was associated with significantly fewer withdrawals because of any adverse event (22% versus 57%). The main reason was a rise in serum creatinine (1.2% versus 24%). Renal toxicity was significantly worse and there were more drug-related adverse events with amphotericin.

Echinocandins : In a randomized, double-blind comparison of amphotericin (0.5 mg/kg intravenously) and caspofungin acetate (35, 50, or 70 mg) once daily for 7–14 days in 140 patients with oropharyngeal and/or esophageal candidiasis, 63% had esophageal involvement and 98% were infected with HIV [ ]. Response rates were 63% with amphotericin and 74–91% with caspofungin. More patients receiving amphotericin had drug-related adverse effects (fever, chills, nausea, vomiting) than those receiving any dose of caspofungin. Two patients who took caspofungin 35 mg and one who was given amphotericin withdrew because of adverse effects. Drug-related laboratory abnormalities were also more common in patients who received amphotericin. The most common drug-related laboratory abnormalities in patients who received caspofungin were raised alanine transaminase, aspartate transaminase, and alkaline phosphatase, which were typically less than five times the upper limit of normal and resolved despite continued treatment. None of the patients receiving caspofungin and nine of those who received amphotericin developed drug-related increases in serum creatinine concentrations. No patient withdrew because of drug-related laboratory adverse effects.

Amphotericin has been compared with caspofungin in a multicenter, double-blind, randomized trial in 128 adults with endoscopically documented symptomatic Candida esophagitis [ ]. There was endoscopically verified clinical success in 63% of patients given amphotericin deoxycholate 0.5 mg/kg/day and in 74% and 89% of the patients who received caspofungin 50 and 70 mg/day respectively. Therapy was withdrawn because of drug-related adverse events in 24% of the patients who were given amphotericin and in 4% and 7% of those who were given caspofungin 50 and 70 mg/day respectively. More patients who received amphotericin had drug-related fever, chills, or nausea than those who received caspofungin. More patients who received amphotericin (91%) than caspofungin (61% and 32%) developed drug-related laboratory abnormalities. There were drug-related increases in blood urea–nitrogen concentrations in 15% of the patients who received amphotericin but none of those who received caspofungin. Likewise, serum creatinine concentrations increased in 16 patients who received amphotericin but in only one who received caspofungin. In summary, caspofungin was as effective as amphotericin but better tolerated in the treatment of esophageal candidiasis.

In a double-blind, randomized trial, amphotericin deoxycholate was compared with caspofungin for the primary treatment of invasive candidiasis [ ]. Patients who had clinical evidence of infection and a positive culture for Candida species from blood or another site were enrolled. They were stratified according to the severity of disease, as indicated by the presence or absence of neutropenia and the Acute Physiology and Chronic Health Evaluation (APACHE II) score, and were randomly assigned to receive either amphotericin (0.6–0.7 mg/kg/day or 0.7–1.0 mg/kg/day for patients with neutropenia) or caspofungin (50 mg/day with a loading dose of 70 mg on day 1). Of the 239 patients enrolled, 224 were included in the modified intention-to-treat analysis. Baseline characteristics, including the percentage of patients with neutropenia and the mean APACHE II score, were similar in the two treatment groups. The efficacy of amphotericin was similar to that of caspofungin, with successful outcomes in 62% of the patients treated with amphotericin and in 73% of those treated with caspofungin. There were significantly more drug-related adverse events (fever, chills, and infusion-related events) associated with amphotericin. Amphotericin caused more nephrotoxicity, as defined by an increase in serum creatinine of at least twice the baseline value or an increase of at least 88 μmol/l) (8.4% versus 25%). Only 2.6% of those who were given caspofungin were withdrawn because of adverse events, compared with 23% of those who were given amphotericin. Thus, caspofungin was at least as effective as amphotericin for the treatment of mostly non-neutropenic patients with invasive candidiasis but significantly better tolerated.

Effects on blood pressure

Changes in blood pressure (hypotension as well as hypertension) have been reported [ , ].

• A 67-year-old man with multiple intraperitoneal and urinary fungal pathogens and a history of well-controlled chronic hypertension developed severe hypertension associated with an infusion of ABLC [ ]. He received a 5 mg test dose, which was tolerated without incident. About 60 minutes into the infusion (5 mg/kg), his blood pressure rapidly increased to 262/110 mmHg from a baseline of 150/80 mmHg. His temperature increased to 39.8 °C, and tachycardia developed (up to 121/minute). The infusion was stopped, and he was given morphine, propranolol, and paracetamol. His blood pressure returned to baseline over the next 2 hours. Rechallenge with ABLC on the next day resulted in an identical reaction despite premedication with pethidine, diphenhydramine, and morphine. ABLC was permanently withdrawn, and the infection was managed with high dosages of fluconazole.

The etiology of amphotericin-associated hypertension has not been elucidated, but it may be related to vasoconstriction. Of note, the traditional test dose appears not to identify individuals predisposed to hypertensive reactions; four of six cases of amphotericin-associated hypertension received test doses without incident.

There have been eight cases of hypertension in patients receiving amphotericin; six occurred within 1 hour [ ]. All except one had received a non-lipid-containing formulation.

• A 19-year-old girl with acute lymphoblastic leukemia developed sustained severe arterial hypertension shortly after being given amphotericin and continuing for several hours after the infusion [ ].

Cardiac dysrhythmias

In 6 of 90 children given intravenous amphotericin there was a significant fall in heart rate, and monitoring of heart rate was recommended in children with underlying heart disease [ ]. These immediate reactions follow intravenous administration and occur particularly with excessively rapid infusion of DAMB.

Ventricular dysrhythmias have been reported after rapid infusion of large doses of DAMB [ ] in patients with hyperkalemia and renal insufficiency, but not in patients with normal serum creatinine and potassium concentrations, even if they have received the drug over a period of 1 hour. Slower infusion rates and infusion during hemodialysis have been advocated in patients with terminal kidney insufficiency, in order to avoid hyperkalemia.

• A 41-year-old woman with cryptococcal meningitis and no previous cardiac disease developed a fatal cardiac dysrhythmia, acute renal failure, and anemia after an acute overdose of amphotericin B deoxycholate [ ]. The intention had been to give liposomal amphotericin 5 mg/kg/day; however, amphotericin B deoxycholate 5 mg/kg was inadvertently given instead, the usual dose of the deoxycholate formulation being 0.5–0.8 mg/kg/day.