Amphotericin B colloidal dispersion: an improved antifungal therapy

Advanced Drug Delivery Reviews 47 (2001) 149–163 www.elsevier.com / locate / drugdeliv

Amphotericin B colloidal dispersion: an improved antifungal therapy Luke S.S. Guo* ALZA Corporation, 1050 Hamilton Court, Menlo Park, CA 94025, USA

Abstract Amphotericin B colloidal dispersion (ABCD) is a near 1:1 discoidal complex of amphotericin B (AMB) and sodium cholesteryl sulfate (SCS) arranged as a bilayer of SCS interspersed with AMB via noncovalent interactions. The complex is stable in blood and plasma with minimal dissociation. In vitro and in vivo studies show that ABCD is as effective and four to five times safer than conventional AMB (CAB) for fungal infection. Compared with CAB treatment, ABCD demonstrates reduced peak plasma levels, prolonged residence time, and lowered AMB levels in most tissues including kidney, the major target of toxicity for CAB. In 572 patients with systemic fungal infections secondary to severe underlying disease, ABCD doses # 6 mg / kg / day were well tolerated, even in those who failed to tolerate or respond to CAB. Mild-to-moderate, dose-dependent, infusion-related adverse events typically seen with CAB were also observed with ABCD, with no sign of renal or hepatic toxicity. Complete or partial recovery was seen in 57.3%. Therefore, ABCD should be considered as an alternative treatment of systemic fungal infections.  2001 Elsevier Science B.V. All rights reserved. Keywords: Lipid complex of amphotericin B; Sodium cholesteryl sulfate; Sodium deoxycholate; Structure of ABCD; Clinical safety and efficacy of ABCD; Pharmacokinetics of ABCD

Contents 1. Introduction ............................................................................................................................................................................ 2. Formulation of ABCD ............................................................................................................................................................. 3. Structure of ABCD.................................................................................................................................................................. 4. Size-exclusion chromatography of ABCD ................................................................................................................................. 5. Stability of ABCD in blood or plasma ...................................................................................................................................... 6. In vitro and in vivo activity ...................................................................................................................................................... 7. Pharmacokinetics .................................................................................................................................................................... 8. Preclinical toxicology .............................................................................................................................................................. 9. Clinical safety of ABCD .......................................................................................................................................................... 10. Clinical efficacy of ABCD ..................................................................................................................................................... 10.1. Bone marrow transplant patients with invasive fungal infections........................................................................................ 10.2. Immunocompromised patients with candidemia................................................................................................................ 10.3. Patients with invasive aspergillosis .................................................................................................................................. 10.4. Renal impaired patients with invasive fungal infections.....................................................................................................

*Tel.: 1 1-650-564-4596; fax: 1 1-650-617-3080. E-mail address: [email protected] (L.S.S. Guo). 0169-409X / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0169-409X( 01 )00104-1

150 151 151 153 154 155 157 158 158 158 158 159 159 159

150

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

10.5. Empirical treatment of fever and neutropenia ................................................................................................................... 10.6. Other systemic mycosis .................................................................................................................................................. 10.7. Overview trials .............................................................................................................................................................. 11. Summary .............................................................................................................................................................................. Acknowledgements ...................................................................................................................................................................... References ..................................................................................................................................................................................

1. Introduction Amphotericin B (AMB) is a broad-spectrum fungicidal antibiotic used primarily in the treatment of life-threatening systemic fungal infections [1,2]. It is a polyene macrolide containing seven double bonds along the hydrophobic side of the ring, multiple hydroxyl groups along the hydrophilic side, and a mycosamine side chain (Fig. 1). This molecule exhibits strong binding affinity for membrane sterols. The most widely accepted model for the anticellular activity of this agent involves the formation in cell membranes of 1:1 AMB / sterol aggregates, which subsequently associate into a transmembrane channel leading to ion and solute leakage and cell death [3]. Binding of AMB to cholesterol in mammalian cell membranes results in various adverse effects. The stronger association of AMB with ergosterol in fungal cells versus cholesterol explains the higher toxicity toward fungi [4]. AMB may also induce oxidative damage to the cell membrane and stimulate host immune response [5,6]. Due to its poor aqueous solubility, the conventional AMB (CAB) used in the clinic is suspended in a bile salt, sodium deoxycholate, which forms a micel-

159 160 160 161 161 161

lar suspension. The clinical utility of this formulation is complicated by frequent and severe side effects, including fever, chills, nausea, vomiting, anemia, and nephrotoxicity [1,2]. Abnormal renal function occurs in most patients after multiple dosing, and therapy must be interrupted to prevent further tissue damage. The dose-limiting side effects often lead to therapeutic failures. Beginning in the early 1980s, a number of lipid or liposome formulations of AMB were investigated in an attempt to increase the specificity and reduce the toxicity of this potent antifungal agent [7–15]. Three lipid formulations, liposomal amphotericin B (AmBisome  ), amphotericin B lipid complex (ABLC, ABELCET  ), and amphotericin B colloidal dispersion (ABCD, AMPHOTECH  , AMPHOCIL  ) have been successfully tested in the clinic and all have shown an increased therapeutic index compared with CAB. ABCD, developed by ALZA Corporation (formerly SEQUUS Pharmaceuticals, Inc.), has received regulatory approval for the treatment of invasive aspergillosis in patients where renal impairment or unacceptable toxicity precludes the use of CAB. It is a colloidal complex of AMB and sodium cholesteryl

Fig. 1. Molecular structure of amphotericin B.

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

sulfate (SCS) in near 1:1 molar ratio. This manuscript reviews the formulation, structure, in vitro and in vivo properties, and preclinical pharmacology and toxicology of this drug dosage form. Clinical safety and efficacy of ABCD are also summarized.

2. Formulation of ABCD ABCD is a near 1:1 molar ratio complex of AMB and SCS [16] that is formulated based on the unique affinity of AMB for sterol. The lipid carrier, SCS, is a naturally occurring metabolite of cholesterol that is distributed widely in human tissue and fluids [17]. The commercial form of ABCD is available as a lyophilized product in two dosage sizes of 50 or 100 mg AMB per vial. Following reconstitution with sterile water for injection, ABCD occurs as an opalescent or clear colloidal dispersion at an AMB concentration of 5 mg / ml. The mean hydrodynamic diameter of ABCD determined by dynamic light scattering is 115630 nm (specification range of the commercial product).

3. Structure of ABCD The structure of ABCD was characterized by several physicochemical methods, including electron microscopy (EM), X-ray diffraction, and differential scanning calorimetry (DSC). The images of ABCD were examined by three electron microscopic techniques, including negative staining, platinum shadowing, and thin sectioning. Negative staining of

151

ABCD by EM revealed the presence of thin discs (Fig. 2, top). In more concentrated preparations, many of the discs were folded and seen as an electron-lucent strip or edge. Occasionally, a small portion of a disc edge was folded. In platinumshadowed preparations, ABCD were round and appear to lie flat with no evidence of folding (Fig. 2, middle). Notches were frequently seen at the edges of the discs. In thin sectioning preparations, embedded ABCD revealed parallel layers delineated by dark bands (Fig. 2, bottom). The bands appeared to be in pairs with as many as four adjacent band pairs observed. Two band pairs seemed to occur most frequently. The diameter and thickness of ABCD measured directly from EM images from a typical development lot of ABCD are presented in Table 1. The thickness of ABCD was also determined by small-angle X-ray diffraction. Diffraction patterns were recorded using the photographic film method. A 24-h X-ray exposure of ABCD aligned on mica produced two central diffraction lines oriented vertically. The stronger spacing was calculated to be 4.3 nm and the weaker spacing was 3.1 nm. The stronger spacing of 4.3 nm confirmed the thickness of ABCD determined by EM techniques. DSC analyses were performed on SCS and ABCD over the temperature range of 10–958C. SCS was studied at several concentrations of water as a function of temperature from 18.8 to 74.4% water. On heating SCS at 18.8% water showed a single sharp endotherm at 54.78C with an enthalpy of 5.8 kcal / g (Fig. 3). On cooling, a recrystallization exotherm was seen at about 44.68C with an enthalpy of 5.0 kcal / g. Repeated heating showed essentially

Table 1 Thickness and diameter of ABCD measured directly from electron microscopic images a Image characteristic

Technique

Band spacing Height / thickness Thickness of folded discs (2 layers) Thickness of disc edge (1 layer) Diameter (nm)b

Thin sectioning Shadowing Negative staining

88 82 96

Negative staining

6

Negative staining

335

a

Particle measured

Mean6S.D. (nm)

Range (nm)

4.560.3 3.360.9 8.663.4

3.9–5.4 1.7–4.6 4.1–20.5

4.5

3.9–4.6

122648

–

Data were presented at the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy, 1991. b Average of two perpendicular measurements.

152

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

the same endotherm at 53.38C. No other transitions were seen up to a temperature of 958C. At higher water concentrations (54 and 75%), similar endotherms and exotherms were observed, except that two slightly broader endothermic peaks at 78.8 and 87.58C were also seen on reheating. After further cooling and heating there were no endothermic events, indicating a somewhat metastable structure for the SCS over this range of temperature at these high water concentrations. DSC of the complex of AMB and SCS on the other hand, showed no melting behavior between 10 and 958C, indicating the disruption of the SCS lattice and the formation of a stable complex. Therefore, ABCD is a disc-shaped bilayer of SCS interspersed with AMB via noncovalent interactions, including van der Waal’s, electrostatic, and hydrogen bond forces. The thickness of the disc is about 4.3 nm, which is roughly equivalent to a bilayer thickness of SCS. The interaction of SCS with AMB is intimate in that it abolishes all of the transitions and structures that SCS forms by itself in water. Because of the presence of a polar hydroxyl region, AMB probably forms a shield at the disc edges and may be oriented in such a way that the seven hydroxyl groups along its side are exposed to the polar aqueous environment [16]. Fig. 4 shows a proposed model structure of ABCD.

Fig. 2. Electron micrographs of ABCD prepared by different techniques. (Top) Negative staining: round and flat disc-shaped particles were seen. Many of the discs were folded. The portion of the disc that folded under was superimposed on the rest of the disc (long arrows). Occasionally, a small portion of a disc edge was folded up (short arrow). (Middle) Platinum-shadowing: flat, round discs with frequent notches at the edges were seen. The discs appeared to lie flat and have very narrow shadow margins. In contrast, a shadowed polystyrene bead (seen as a white spherical particle) of 109 nm diameter had a sharp and elongated shadow margin. Note a stack of two discs as delineated by the arrow. (Bottom) Thin-sectioning: parallel layers were revealed, delineated by dark bands. The bands appeared to be in pairs with as many as four adjacent band pairs observed. Two band pairs seemed to occur most frequently. Length estimates of several sets of band pairs suggested similarity of the size range of particle diameters, which were determined by negative staining. The mean band spacing was 4.5 nm. (Data were presented at the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy, 1991.)

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

153

Fig. 3. Differential scanning calorimetry of sodium cholesteryl sulfate in various amounts of water. On heating, a single sharp endotherm between 54 and 558C with an enthalpy of about 5.5 kcal / g was observed. On cooling, there was a recrystallization exotherm between 44 and 478C. (Data were presented at the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy, 1991.)

4. Size-exclusion chromatography of ABCD To define further the nature of the AMB / SCS complexes, ABCD was separated into fractions by size using size-exclusion chromatography (SEC) with a calibrated Sephacryl S-1000 column [16]. Chromatography of ABCD yielded a single AMB peak with mean hydrodynamic diameter fractions ranging from 40 to 170 nm (Fig. 5). Analysis of AMB and SCS ratios in each fraction showed that

the drug-to-lipid molar ratio was approximately 1:1 and was consistent throughout the identified range of particle sizes. Recovery of AMB from the column was almost 100% and was associated exclusively with SCS in the complex form. These results confirmed that ABCD is a near 1:1 complex of AMB and SCS. In sharp contrast, chromatography of CAB, a micellar suspension of amphotericin B and sodium deoxycholate, resulted in micelle dissociation and

154

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

intact during passage through columns in which the sample is diluted approximately 50-fold.

5. Stability of ABCD in blood or plasma

Fig. 4. Proposed model structure of ABCD. Top: Cartoon molecules of SCS and AMB. Middle: cross-section view of ABCD shows a bilayer of SCS with AMB interspersed within it. Bottom: side view of a discoidal complex. AMB forms a shield at the disc edges and is oriented so that the seven hydroxyl groups (black dots) along its side are exposed to the polar aqueous environment. (Data were presented at the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy, 1991.)

drug precipitation within the column bed with no detectable drug eluted [16]. These results demonstrate one of the key differences between ABCD and CAB. The micellar system is at a thermodynamic equilibrium that depends largely on the critical micellar concentrations of the constituents. Rapid dilution leads to instant instability. The data also suggest that ABCD is a stable complex that remains

After intravenous administration, CAB is believed to dissociate rapidly, resulting in binding of AMB to the surrounding plasma lipoproteins and cell membranes [18,19]. Plasma lipoprotein-bound AMB is more toxic than AMB alone in vivo [20]. Moreover, exposure of erythrocytes to CAB can lead to hemolysis [21]. The hemolytic effect of ABCD was compared with that of CAB micelles in vitro. Fresh human blood containing 0.1% EDTA was mixed with ABCD or CAB to give a final AMB concentration of 28 mg / ml. Samples were incubated at 378C for 30 min. When whole human blood was incubated with CAB, hemoglobin was rapidly released, indicating erythrocyte lysis, but this was not observed in human blood incubated with ABCD [16]. The hemolytic effect of the CAB was probably due to the combined effects of AMB and the deoxycholate carrier. In vitro studies in human plasma have shown that the ABCD complex is significantly more stable than CAB [16]. ABCD or CAB was incubated with freshly isolated human plasma at 378C for 2 h. At the end of the incubation, three plasma fractions (plasma lipoproteins, proteins, and a lipoprotein / protein-poor middle fraction) were isolated by sequential preparative ultracentrifugation, and drug concentrations were measured. Incubation of CAB with human plasma, at an AMB concentration of 50 mg / ml, resulted in extensive complex dissociation. More than 90% of the AMB was dissociated and bound to plasma lipoproteins and proteins (Table 2). At higher incubation concentrations (100 and 300 mg / ml), plasma appeared to saturate as greater proportions of the drug remained in the protein-poor middle fraction. In contrast, when ABCD was incubated with plasma under the same experimental conditions, greater than 80% of the ABCD remained nondissociated regardless of incubation concentrations. In fact, ABCD-like particles were observed in the protein / lipoprotein-poor fraction by EM (Fig. 6). Together, these studies indicate that AMB remains bound to its lipid carrier after ABCD administration,

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

155

Fig. 5. Elution profile of ABCD from a Sephacryl S-1000 column (1.5 3 65 cm) [16]. Closed and open circles are AMB and SCS concentrations in each fraction, respectively, determined by HPLC. Numbers along the elution profile are mean hydrodynamic diameters of ABCD determined by dynamic light scattering. Arrows show the calibrated void volume (V0 ) and included volume (Vi ) of the column.

Table 2 Dissociation of AMB from CAB or ABCD in plasma Plasma fraction a (Density)

Distribution of AMB in plasma fractions (%) CAB (mg AMB / ml plasma)

Plasma lipoproteins (d , 1.21 g / ml) Lipoprotein / protein-poor fraction (d 5 1.21–1.30 g / ml) Plasma proteins (d . 1.30 g / ml)

ABCD (mg AMB / ml plasma)

50

100

300

76.6

51.2

23.8

8.9

36.6

14.5

13.4

50

100

300

4.8

3.8

2.7

65.4

80.2

85.2

84.8

10.8

14.5

11.1

12.5

a

Freshly isolated human plasma was incubated with CAB or ABCD at AMB concentrations of 50–300 mg / ml at 378C for 2 h. The mixtures were then separated into three density fractions by sequential preparative ultracentrifugation. AMB concentration in each fraction was determined by HPLC.

thus reducing the possibility of rapid distribution of the drug to red blood cells and plasma lipoproteins in the blood circulation.

6. In vitro and in vivo activity Hanson and Stevens [22] compared the antifungal activity of ABCD and CAB against 41 isolates of 15 pathogenic fungi known to be sensitive to AMB. ABCD showed some species-dependent differences

from CAB, but the range of antifungal activities observed was similar for the two formulations. The in vivo activity of ABCD has been studied in murine models of Aspergillus fumigatus, Candida albicans, Coccidioides immitis, and Cryptococcus neoformans infections [23–26]; a rabbit model of Aspergillus fumigatus [27]; and a hamster model of Leishmania donovani [28] infections. In these studies, at least one equivalent dose of CAB was included for comparison. Efficacy was assessed based on survival and / or tissue colony count. Be-

156

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

Fig. 6. Electron microscopic image of ABCD (top) and ABCD-like particles isolated from human plasma after 2 h of incubation at 378C (bottom). Samples were prepared by negative staining technique. White spherical particles are plasma lipoproteins.

cause systemic fungal infections often occur in immunocompromised patients, some studies were performed in animals that were genetically immunodeficient or pharmacologically immunosuppressed. These studies showed either that ABCD and CAB were equally effective at the same dose or that ABCD was as effective as CAB at three- to five-fold

higher doses. Because ABCD is well tolerated at doses up to five to eight times the maximum tolerated dose of CAB [24], greater efficacy with ABCD was achieved through dose. ABCD was better at eradicating fungal infections in several of the animal models tested because of its improved therapeutic index compared with that of CAB [25–

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

27]. In the hamster model of Leishmania donovani infection, ABCD was approximately four times as active as CAB [28].

Table 4 AMB tissue distribution in rats and dogs: ratio of tissue concentration of ABCD to CAB after 14 daily repeat dosings [24] Tissue

Ratio of tissue concentration (ABCD/ CAB)

7. Pharmacokinetics Pharmacokinetics and tissue distribution of ABCD or CAB were studied in rats [24,29,30] and dogs [24,31] that received either a single dose or up to 13 weeks of daily doses of each formulation. In these studies, plasma and tissue levels of AMB were determined by high-performance liquid chromatography (HPLC). Maximal plasma concentrations of AMB in both rats and dogs that received equivalent doses of ABCD and CAB were lower in ABCDtreated animals, because of its rapid uptake by the liver. However, the area under the plasma concentration–time curve was greater in the ABCDtreated animals owing to the prolonged plasma terminal half-life of AMB (Table 3) [29]. When compared with CAB treatment, ABCD treatment was associated with lower concentrations of AMB in most tissues (Table 4), including the kidney, which is the primary site of toxicity after CAB treatment. ABCD was distributed primarily to the liver after intravenous injection and also to the other organs of the mononuclear phagocytic system (MPS), including the spleen and bone marrow. A separate study, in which liver cell types were separated using centrifugal elutriation, showed that after treatment with ABCD the majority of AMB in the liver was in the Kupffer cells or in the macrophages of the MPS, with a small proportion in the hepatocytes [24]. The rate of washout of AMB from the tissues after

157

Liver Kidney Spleen Lung Heart Bone marrow Brain Skeletal muscle a

Rats (daily dose 1 mg / kg)

Dogs (daily dose 0.6 mg / kg)

5.2 0.3 0.4 0.3 0.3 nd a 0.4 0.6

1.69 0.15 1.09 0.02 0.01 2.78 0.07 nd

nd, not detected.

the end of treatment [30,31] was very similar to its plasma elimination half-life in the animal studies. The dose proportionality of the plasma terminal elimination half-life of AMB in ABCD-treated animals was due to the dose-proportionate accumulation of AMB in the tissues. Thus, the plasma pharmacokinetics of AMB after ABCD administration not only represent the actual rate of AMB elimination from the plasma, but also reflect the amount of AMB accumulated and its rate of washout from sites of accumulation. The organs of the MPS may serve as a reservoir from which AMB is slowly released. A similar dose-dependence was seen in the plasma terminal elimination half-life of AMB in healthy human subjects who received doses of ABCD ranging from 0.25 to 1.5 mg / kg [32]. These studies demonstrate that the pharmacokinetics and tissue distribution of AMB differ in several respects after administration of ABCD vs.

Table 3 Pharmacokinetic parameters a of CAB and ABCD in a single dose study in rats [29] Drug and dose

Cmax at l h (ng / ml)

AUC 0 – ` (ng / h / ml)

CL (l / h / kg)

MRT (h)

t1 / 2 (h)

CAB 1 mg / kg

275646

3986

0.26

16.59

10.4

ABCD 1 mg / kg 5 mg / kg

102611 170615

4421 11 620

0.23 0.43

40.38 46.57

27.1 41.12

a

Cmax , maximal concentration; AUC 0 – ` , area under the first moment of the plasma concentration–time curve to infinity; CL, clearance; MRT, mean residence time; t 1 / 2 , terminal half-life.

158

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

CAB, due to the increased stability of ABCD in plasma and the very rapid uptake of ABCD by the macrophages of the MPS, primarily in the liver. Compared with animals receiving CAB, animal receiving ABCD had reduced peak levels in plasma, prolonged residence time, and lowered levels of AMB in most tissues, including the kidney.

caused only minimal and readily reversible inflammation at the injection site. Reproductive toxicity studies of ABCD in rats and rabbits demonstrated that ABCD was neither embryotoxic nor teratogenic [24].

9. Clinical safety of ABCD 8. Preclinical toxicology Studies in three species (mice, rats, and dogs) that employed daily dosing up to 13 weeks have demonstrated that ABCD induces a spectrum of adverse effects similar to that of CAB, but at dose levels that are four- to five-fold higher [23,24]. No toxicities unique to ABCD were observed, despite the markedly different disposition of AMB after administration of each product. The most common finding in ABCD-treated animals was dose-related renal tubular nephrosis similar to the renal damage produced by CAB [24]. Renal toxicity, as assessed by clinical chemistry and histopathological changes, was observed after as few as 14 daily doses. In general, it was partially or fully reversible, particularly at a dose level of ABCD below 5 mg / kg / day in rats and below 1 mg / kg / day in dogs. The histology of the renal changes was very similar in ABCD- and CAB-treated animals, suggesting that both were caused by the action of AMB. Renal toxicity of comparable severity was only seen after doses of ABCD four to five times higher than those of CAB. These studies have shown that treatment with ABCD affected neither the glomerular filtration rate nor blood flow to the kidney tubules, whereas the same dose of CAB substantially decreased each. The hepatic toxicity commonly seen after CAB treatment was observed only at high doses of ABCD. Overall, despite significant accumulation of AMB in the liver (up to 20-fold the levels measured after the same dose of CAB), hepatic toxicity was generally less notable after ABCD treatment than after treatment with CAB [24]. When administered intravenously or intra-arterially, CAB also exhibited moderately severe irritation. In contrast, ABCD was less irritating and caused only slight to moderate, reversible perivascular irritation by either route [24]. Extravasation of ABCD

Herbrecht and colleagues [33,34] reviewed the safety of ABCD in five open-label Phase I / II clinical trials in 572 selected patients who had a fungal infection secondary to a severe underlying disease. In these studies, the populations were selected to include patients particularly at risk of nephrotoxicity either because of their illness or because of preexisting renal impairment mainly due to prior therapy with CAB. ABCD doses of up to 6 mg / kg / day resulted in no change from baseline in serum creatinine levels, even in patients with preexisting renal failure. ABCD therapy resulted in no change from baseline in liver function as measured by liver enzymes and total bilirubin levels in serum. The dose-limiting toxicity was infusion-related rigors, chills and hypotension, also seen with CAB therapy. Adverse events related to ABCD requiring discontinuation of therapy occurred in 70 patients (12.2%). The most frequent of these were infusion-related adverse events, which occurred in 5.4% of patients. The maximum tolerated dose was set at 7.5 mg / kg / day. These studies show that ABCD can be administered safely to patient with reduced risk of renal toxicity, even when renal impairment has already developed following therapy with CAB.

10. Clinical efficacy of ABCD

10.1. Bone marrow transplant patients with invasive fungal infections Bowden et al. [35] conducted a Phase 1 doseescalation study of ABCD in 75 bone marrow transplant patients with invasive fungal infections (primarily Aspergillus or Candida species). The objective of the study was to determine the toxicity profile, maximum tolerated dose, and clinical response of ABCD in this patient population. Escalat-

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

ing doses of 0.5–8.0 mg / kg in 0.5-mg / kg / patient increments were given for up to 6 weeks. This study showed that ABCD could be delivered safely up to a daily dose of 7.5 mg / kg. The complete or partial response rate across dose levels and infection types was 52%. For specific types of infections, 53% of patients with fungemia had complete response, and 52% of patients with pneumonia had complete or partial response. Noskin et al. [36] reviewed the efficacy of ABCD in a total of 220 bone marrow transplant recipients from five open-label clinical trials of ABCD therapy (include Bowden’s 75 patients mentioned above). All patients had suspected or documented life-threatening fungal infections. ABCD was administered intravenously once daily with a median dose of 4 mg / kg for up to 409 days. Successful therapeutic response to treatment (complete or partial) was reported in 52% of the 99 evaluable patients with proven infection and in 40% of all 220 patients. In the evaluable patient population, the response and mortality rates were 51 and 73%, respectively.

10.2. Immunocompromised patients with candidemia The efficacy of ABCD was evaluated in 148 immunocompromised patients who were diagnosed with candidemia [37]. The patients were recruited from five open-label clinical trials of ABCD therapy for fungal infections subsequent to bone marrow transplantation, hematologic malignancies, solid tumors, solid-organ transplantation, or other severe underlying disorders. ABCD was given intravenously (median daily dose 3.9 mg / kg) for up to 72 days. Response rates were 53% overall, 66% for patients with candidemia alone, and 14% for patients with disseminated candidemia. Patients infected with C. albicans or C. parapsilosis responded better than patients infected with other Candida species. In a separate report by Dupont [38], efficacy was evaluated in a total of 346 patients with systemic candidal infections (107 evaluable and 239 in the intent-to-treat population). The response rates (complete or partial) for the evaluable and intent-to-treat populations were 70 and 50%, respectively. The efficacy of ABCD at doses of 3–4 mg / kg / day in

159

disseminated candidiasis appears to be similar to that seen in studies using CAB.

10.3. Patients with invasive aspergillosis To assess the efficacy and renal safety of ABCD in the treatment of invasive aspergillosis, 82 patients with proven or probable aspergillosis who were treated with ABCD were compared retrospectively with 261 patients with aspergillosis who were treated with CAB at six cancer and transplant centers [39]. The groups were balanced in terms of underlying disease. Response rates (48.8%) and survival rates (50%) among ABCD-treated patients were both significantly higher than those (23.4 and 28.4%, respectively) among the historical control CABtreated patients (P , 0.001). Renal dysfunction developed less frequently in ABCD recipients than in the retrospective comparison to CAB recipients (8.2 vs. 43.1%, respectively; P , 0.001). This retrospective study suggests that in the treatment of aspergillosis, ABCD causes less nephrotoxicity than CAB and the efficacy of ABCD is at least comparable with that of CAB.

10.4. Renal impaired patients with invasive fungal infections Because ABCD is designed to minimize drug distribution in the kidney and reduce nephrotoxicity, Anaissie et al. [40] studied the efficacy and safety of ABCD in 133 renally compromised patients with invasive fungal infections. These patients had either nephrotoxicity from CAB or preexisting renal disease. ABCD was given intravenously at 4 mg / kg for up to 6 weeks. ABCD did not appear to have an adverse effect on renal function. Serum creatinine levels tended to decrease slightly over the course of the therapy, regardless of the total dose of ABCD. Complete or partial response to treatment was reported for 50% of the 133 intent-to-treat patients.

10.5. Empirical treatment of fever and neutropenia White et al. [41] conducted a prospective, randomized, double-blind study comparing ABCD with CAB in the empirical treatment of fever and neutro-

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

160

penia. Patients with neutropenia and unresolved fever after at least 3 days of empirical antibiotic therapy were stratified by age. They were then randomized to receive therapy with ABCD (4 mg / kg / day) or CAB (0.8 mg / kg / day) for up to 14 days. A total of 213 patients were enrolled, of whom 196 were evaluable for efficacy. ABCD appeared comparable in efficacy with CAB (50 vs. 43.2%, respectively, P 5 0.31). Renal toxicity associated with ABCD was significantly (P , 0.001) less frequent than that associated with CAB for both adults and children, but infusion-related hypoxia and chills were more common with ABCD treatment than with CAB treatment (P 5 0.013 and P 5 0.018, respectively).

10.6. Other systemic mycosis Limited experience presented in several case reports suggested that ABCD has been efficacious in the treatment of life-threatening rhinocerebral mucormycosis [42], cryptococcal meningitis [43], and zygomycosis [44], but failed to cure the infection in paracoccidiodomycosis [45]. ABCD was evaluated in these patients because most of the

patients had renal dysfunction due to prior treatment with CAB or other complications.

10.7. Overview trials Herbrecht et al. [33] conducted an overview of efficacy of ABCD in 572 patients who has documented or suspected systemic mycoses in five openlabel Phase I / II clinical trials. Most patients in these studies had failed to adequately respond to treatment with CAB, had developed AMB-induced nephrotoxicity, or had preexisting renal impairment. The mean daily dose of ABCD was 3.85 mg / kg, and most of them received doses of 3–6 mg / kg / day. As shown in Table 5, positive responses (complete or partial recovery) were achieved in 149 of the 260 evaluable patients with an overall response rate of 57.3%. When therapeutic responses were classified by underlying disease and type of infection, patients with Candida infection were found to respond better than those with systemic aspergillosis (70.1% recovery vs. 48.8%). Patients with other infections had a positive response rate of 47.9%. Bone marrow transplant recipients appeared to have the lowest therapeutic response rates regardless of the type of

Table 5 Therapeutic response of ABCD [34] Disease

Response rate in % (numbers) of treated patients a Total

Aspergillosis

Candidiasis

Other / multiple

ITT

Eval

ITT

Eval

ITT

Eval

ITT

Eval

Bone marrow transplant

38.8 (224)

52.5 (99)

27.9 (68)

35.3 (34)

50.5 (101)

67.4 (46)

30.9 (55)

47.4 (19)

Hematological malignancy

42.2 (135)

67.9 (53)

40.4 (57)

66.7 (21)

47.4 (38)

76.9 (13)

40.0 (40)

63.2 (19)

Solid tumor

40.7 (27)

71.4 (14)

33.3 (3)

33.3 (3)

42.9 (21)

80.0 (10)

33.3 (3)

100 (1)

Transplant

50.7 (69)

57.5 (40)

39.3 (28)

50.0 (14)

63.0 (27)

64.7 (17)

50.0 (14)

55.6 (9)

Other

37.6 (117)

51.9 (54)

36.4 (22)

62.5 (8)

46.2 (52)

71.4 (21)

27.9 (43)

32.0 (25)

Total

40.9 (572)

57.3 (260)

34.8 (178)

48.8 (80)

49.8 (239)

70.1 (107)

34.2 (155)

47.9 (73)

a

ITT, intent to treat group; Eval, evaluable group.

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163

161

infection. The data showed no difference in efficacy in the higher-dose levels, but longer courses are clearly required for the treatment of infection with Aspergillus than for Candida infection.

under the direction of Dr Donald Small. The author is grateful to Drs Bob Albra and Harry Wong for reviewing this manuscript.

11. Summary

References

The data presented here demonstrate that ABCD, an almost equimolar complex of AMB and SCS, can modulate the toxic effects of the drug. Interaction of AMB and SCS results in disc-shaped particles with thickness of about 4.3 nm and a mean hydrodynamic diameter of 115 nm. In vitro and in vivo studies have shown that ABCD retains the spectrum and activity of AMB but is less toxic to the kidneys than CAB. The reduced renal toxicity of ABCD is probably related to the stability of the ABCD complex in plasma, the rapid uptake of ABCD by the macrophages of the MPS (primarily in the liver), and the reduction of delivery of AMB to the major target organs for toxicity, including the kidneys. Clinical studies in 572 patients with systemic fungal infections have shown that ABCD up to 6 mg / kg / day is well tolerated and results in no change in serum creatinine levels even in patients with preexisting renal failure. ABCD therapy results in no change in liver function as measured by liver enzymes and total bilirubin levels in serum. Infusionrelated adverse events, also commonly seen with CAB, are the most frequently reported side effects of ABCD. The overall therapeutic response rate with complete or partial recovery is 57.3%. These results show that ABCD should be considered in the treatment of select systemic fungal infections with reduced risks of renal toxicity, even when renal dysfunction had already developed following therapy with CAB. The major advantage of ABCD therapy appears to be the ability to deliver AMB with reduced nephrotoxicity.

[1] G. Medoff, J. Brajtburg, G.S. Kobayashi, J. Bolard, Antifungal agents useful in therapy of systemic fungal infections, Annu. Rev. Pharmacol. Toxicol. 23 (1983) 303– 330. [2] T.J. Walsh, A. Pezzo, Treatment of systemic fungal infection: recent progress and current problems, Eur. J. Clin. Microbiol. Infect. Dis. 7 (1988) 460–475. [3] T. Andreoli, On the anatomy of AMB–cholesterol pores in lipid bilayer membranes, Kidney Int. 4 (1973) 337–345. [4] J.D. Readio, R. Bittman, Equilibrium binding of amphotericin B and its methyl ester and borate complex to sterols, Biochim. Biophys. Acta 685 (1982) 219–224. [5] M. Stokol-Anderson, J. Brajtburg, G. Medoff, AMB-induced oxidative damage and killing of Candida albicans, J. Infect. Dis. 154 (1986) 76–83. [6] M. Stokol-Anderson, J.E. Sligh Jr., S. Elberg, J. Brajtburg, G.S. Kobayashi, G. Medoff, Role of cell defense against oxidative damage in the resistance of Candida albicans to the killing effects of AMB, Antimicrob. Agents Chemother. 32 (1988) 702–705. [7] R.R. New, M.L. Chance, S. Health, Antileishmanial activity of amphotericin and other antifungal agents entrapped in liposomes, Antimicrob. Agents Chemother. 8 (1981) 371– 381. [8] J.R. Graybill, P.C. Craven, R.L. Taylor, D.M. Williams, W.E. Magee, Treatment of murine cryptococcosis with liposomeassociated amphotericin B, J. Infect. Dis. 145 (1982) 748– 752. [9] R.L. Taylor, D.M. Williams, P.C. Craven, J.R. Graybill, D.J. Drutz, W.E. Magee, Amphotericin B in liposomes: A novel therapy for histoplasmosis, Am. Rev. Respir. Dis. 125 (1982) 610–611. [10] G. Lopez-Berestein, V. Fainstein, R. Hopfer, K. Mehta, M.P. Sullivan, M. Keating, M.G. Rosenblum, R. Mehta, M. Luna, E.M. Hersh, J. Reuben, R.L. Juliano, G.P. Bodey, Liposomal amphotericin B for the treatment of systemic fungal infections in patients with cancer: A preliminary study, J. Infect. Dis. 151 (1985) 704–710. [11] G. Lopez-Berestein, G.P. Bodey, L.S. Frankel, K. Mehta, Treatment of hepatosplenic candidiasis with liposomal amphotericin B, J. Clin. Oncol. 5 (1987) 310–317. [12] G. Lopez-Berestein, G.P. Bodey, V. Faintein, M. Keating, L.S. Frankel, B. Zeluff, L. Gentry, K. Mehta, Treatment of systemic fungal infections with liposomal amphotericin B, Arch. Intern. Med. 149 (1989) 2533–2536. [13] R.L. Juliano, C.W.M. Grant, K.R. Barber, M.A. Kalp,

Acknowledgements The physicochemical characterization studies of ABCD were performed at Biophysics Institute, Boston University School of Medicine, Boston, MA,

162

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23] [24]

[25]

[26]

[27]

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163 Mechanism of the selective toxicity f amphotericin B incorporated into liposomes, Mol. Pharmacol. 31 (1987) 1–11. F.C. Szoka Jr., D. Milholland, M. Barza, Effect of lipid composition and liposome size on toxicity and in vitro fungicidal activity of liposome-intercalated amphotericin B, Antimicrob. Agents Chemother. 31 (1987) 421–429. T.F. Patterson, P. Miniter, J. Dijkstra, F.C. Szoka Jr., J.L. Ryan, V.T. Andriole, Treatment of experimental invasive aspergillosis with novel amphotericin B / cholesterol–sulfate complexes, J. Infect. Dis. 159 (1989) 717–724. L.S.S. Guo, R.M. Fielding, D.D. Lasic, R.L. Hamilton, Novel antifungal drug delivery: stable amphotericin B–cholesteryl sulfate discs, Int. J. Pharm. 75 (1991) 45–54. M. Iwamori, H.W. Moser, Y. Kishimoto, Cholesterol sulfate in rat tissues: Tissue distribution, developmental change and brain subcellular localization, Biochim. Biophys. Acta 441 (1976) 268–279. L.C. Edmonds, L. Davidson, J.S. Bertino Jr., Solubility and stability of amphotericin B in human serum, Ther. Drug Monit. 11 (1989) 323–326. J. Brajtburg, S. Elberg, J. Bolard, G.S. Kobayashi, R.A. Levy, R.E. Ostlund Jr., D. Schlessinger, G. Medoff, Interaction of plasma proteins and lipoproteins with amphotericin B, J. Infect. Dis. 149 (1984) 986–997. M.H. Koldin, G.S. Kobayashi, J. Brajtburg, G. Medoff, Effects of elevation of serum cholesterol and administration of amphotericin B complexed to lipoproteins on amphotericin B-induced toxicity in rabbits, Antimicrob. Agents Chemother. 28 (1985) 144–145. S.C. Kinsky, J. Avruch, M. Permutt, H.B. Rogers, A.A. Schonder, The lytic effect of polyene antifungal antibiotics on mammalian erythrocytes, Biochem. Biophys. Res. Commun. 9 (1962) 503–507. L.H. Hanson, D.A. Stevens, Comparison of antifungal activity of amphotericin B deoxycholate suspension with that of amphotericin B cholesteryl sulfate colloidal dispersion, Antimicrob. Agents Chemother. 36 (1992) 486–488. L.S.S. Guo, P.K. Working, Complexes of Amphotericin B and cholesteryl sulfate, J. Liposome Res. 3 (1993) 473–490. P.K. Working, Amphotericin B colloidal dispersion: preclinical review, Antimicrob. Agents Chemother. 45 (Suppl. 1) (1999) 15–26. K.V. Clemons, D.A. Stevens, Comparative efficacy of amphotericin B colloidal dispersion and amphotericin B deoxycholate suspension in treatment of murine coccidiomycosis, Antimicrob. Agents Chemother. 35 (1991) 1829–1883. J.S. Hostetler, K.V. Clemons, L.H. Hanson, D.A. Stevens, Efficacy and safety of amphotericin B colloidal dispersion compared with those of amphotericin B deoxycholate suspension for the treatment of disseminated murine crytococcosis, Antimicrob. Agents Chemother. 36 (1992) 2656– 2660. M.C. Allende, J.W. Lee, P. Francis, K. Garrett, H. Dollenberg, J. Berenguer, C.A. Lyman, P.A. Pizzo, T.J. Walsh, Dose dependent antifungal activity and nephrotoxicity of amphotericin B colloidal dispersion in experimental pulmon-

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

ary aspergillosis, Antimicrob. Agents Chemother. 38 (1994) 518–522. J.D. Berman, G. Ksionski, W.L. Chapman, V.B. Waits, W.L. Hanson, Activity of amphotericin B cholesterol dispersion (Amphocil) in experimental visceral leishmaniasis, Antimicrob. Agents Chemother. 36 (1992) 1978–1980. R.M. Fielding, P.C. Smith, L.H. Wang, J. Porter, L.S.S. Guo, Comparative pharmacokinetics of amphotericin B after administration of a novel colloidal delivery system, ABCD, and a conventional formulation to rats, Antimicrob. Agents Chemother. 35 (1991) 1208–1213. L.H. Wang, R.M. Fielding, P.C. Smith, L.S.S. Guo, Comparative tissue distribution and elimination of amphotericin B colloidal dispersion (Amphocil) and Fungizone after repeated dosing in rats, Pharm. Res. 12 (1995) 275–283. R.M. Fielding, A.W. Singer, L.H. Wang, S. Babbar, L.S.S. Guo, Relationship of pharmacokinetics and drug distribution in tissue to increased safety of amphotericin B colloidal dispersion in dogs, Antimicrob. Agents Chemother. 36 (1992) 299–307. S.W. Sanders, K.N. Buchi, M.S. Goddard, J.K. Lang, K.G. Tolman, Single-dose pharmacokinetics and tolerance of a cholesteryl sulfate complex of amphotericin B administered to health volunteers, Antimicrob. Agents Chemother. 35 (1991) 1029–1034. R. Herbrecht, Safety of amphotericin B colloidal dispersion, Eur. J. Clin. Microbiol. Infect. Dis. 16 (1997) 74–80. R. Herbrecht, V. Letscher, E. Andres, A. Cavalier, Safety and efficacy of amphotericin B colloidal dispersion, Chemotherapy 45 (Suppl. 1) (1999) 67–76. R.A. Bowden, M. Cays, T. Gooley, R.D. Mamelok, J. van Burik, Phase I study of amphotericin B colloidal dispersion for the treatment of invasive fungal infection after bone marrow transplant, J. Infect. Dis. 173 (1996) 1208–1215. G. Noskin, L. Pietrelli, M. Gurwith, R. Bowden, Treatment of invasive fungal infections with amphotericin B colloidal dispersion in bone marrow transplant recipients, Bone Marrow Transplant. 23 (1999) 697–703. G.A. Noskin, L. Pietrelli, G. Coffey, M. Gurwith, L. Liang, Amphotericin B colloidal dispersion for treatment of candidemia in immunocompromised patients, Clin. Infect. Dis. 26 (1998) 461–467. B. Dupont, Clinical efficacy of amphotericin B colloidal dispersion against infectious caused by Candida spp, Chemotherapy 45 (Suppl. 1) (1999) 27–33. M.H. White, E.J. Anaissie, S. Kusne, J.R. Wingard, J.W. Hiemenz, A. Cantor, M. Gurwith, C. Du Mond, R.D. Mamclok, R.A. Bowden, Amphotericin B colloidal dispersion vs. amphotericin B as therapy for invasive aspergillosis, Clin. Infect. Dis. 24 (1997) 635–642. E.J. Anaissie, G.N. Mattiuzzi, C.B. Miller, G.A. Noskin, M.J. Gurwith, R.D. Mamelok, L.A. Pietrelli, Treatment of invasive fungal infections in renal impaired patients with amphotericin B colloidal dispersion, Antimicrob. Agents Chemother. 42 (1998) 606–611. M.H. White, R.A. Bowden, E.S. Sandler, M.L. Graham, G.A. Noskin, J.R. Wingard, M. Goldman, J. van Burik, A.

L.S.S. Guo / Advanced Drug Delivery Reviews 47 (2001) 149 – 163 McCabe, J. Lin, M. Gurwith, C.B. Miller, Randomized, double-blind clinical trial of amphotericin B colloidal dispersion vs. amphotericin B in the empirical treatment of fever and neutropenia, Clin. Infect. Dis. 27 (1998) 296–302. [42] A.E. Moses, G. Rahav, Y. Barenholz, J. Eliden, B. Azaz, S. Gillis, M. Brickman, I. Polacheck, M. Shapiro, Rhinocerebral mucormycosis treated with amphotericin B colloidal dispersion in three patients, Clin. Infect. Dis. 26 (1998) 1430–1433. [43] G. Valero, J.R. Graybill, Successful treatment of cryptococcal meningitis with amphotericin B colloidal dispersion: Report

163

of four cases, Antimicrob. Agents Chemother. 39 (1995) 2588–2590. [44] L.S. Tkatch, S. Kusne, D. Eibling, Successful treatment of zygomycosis of the paranasal sinuses with surgical debridement and amphotericin B colloidal dispersion, Am. J. Otolaryngol. 14 (1993) 249–253. [45] R. Dietze, V.G. Fowler Jr., T.S. Steiner, P.M. Pecanha, G.R. Corey, Failure of amphotericin B colloidal dispersion in the treatment of paracoccidioidomycosis, Am. J. Trop. Med. Hyg. 60 (1999) 837–839.