Amphotericin B Colloidal Dispersion Is Effective and Safe for the Treatment of Invasive Aspergillosis in Chinese Patients with Hematologic Disease: A Multicenter, Prospective, Observational, Real-World Study.

Introduction

Invasive aspergillosis (IA) represents a major complication in patients with hematologic disease, particularly among that undergoing hematopoietic stem cell transplantation (HSCT). 1 , 2 It is estimated that approximately 27,000 annual cases of IA occur globally among 173,629 patients with acute myeloid leukemia (AML) and allogeneic HSCT (allo-HSCT). 3 , 4 Despite advances in diagnostic and antifungal treatments, IA remains associated with significant mortality. Globally, post-treatment IA remains associated with a high mortality rate of approximately 45%, corresponding to an estimated 11,000 attributable deaths annually. 3 , 4 Given the significant morbidity and mortality of IA in patients with hematologic disease, prompt and effective antifungal therapy is essential for patient survival. 5

Current clinical guidelines recommend triazoles, particularly voriconazole, as first-line therapy for IA, whereas lipid formulations of amphotericin B (AmB) remain important alternatives for patients with azole intolerance, drug interactions, or suspected azole-resistant Aspergillus infections. 6 , 7 AmB is a broad-spectrum polyene antifungal agent with fungicidal activity against most Aspergillus species, yet the use of conventional amphotericin B deoxycholate (AmB-D) is severely limited by its nephrotoxicity. 8 To address this limitation, novel lipid formulations of AmB, including amphotericin B colloidal dispersion (ABCD), liposomal amphotericin B (L-AmB), and amphotericin B lipid complex (ABLC), exhibit improved safety profiles, particularly nephrotoxicity. 9 Notably, ABLC is not commercially available in China, while L-AmB remains far less accessible than ABCD in clinical practice. The availability rate of L-AmB in tertiary hospitals was only 5.37%, significantly lower than that of ABCD (27.5%) and AmB-D (54.9%). 10 Furthermore, the high cost of L-AmB presents a substantial barrier to treatment. Thus, despite the higher clinical acceptance of L-AmB, 11 ABCD may represent a more pragmatic choice given the accessibility constraints in the current Chinese clinical setting.

ABCD comprises a stable 1:1 complex of AmB and cholesteryl sulphate in colloidal dispersion, which facilitates targeted delivery to macrophages and reduces renal accumulation. A study in HIV-infected patients with talaromycosis showed that ABCD exhibited comparable clinical efficacy to AmB-D with less nephrotoxicity. 12 In a Phase I/II trial, ABCD doses of up to 6 mg/kg/day resulted in no differences in serum creatinine levels, even in patients with pre-existing renal failure. 13 However, clinical evidence supporting ABCD for IA remains limited, with few data available regarding its efficacy and safety in immunocompromised patients with hematologic diseases, particularly in real-world Chinese clinical practice. This study aimed to provide the first real-world evaluation in China of the clinical outcomes and safety profile of ABCD in this high-risk population, offering preliminary insights into its potential clinical use.

Materials and Methods

Study Design

This exploratory, multicenter, prospective, observational, real-world study enrolled Chinese patients with hematologic disease and proven/probable/possible IA from 7 participating hospitals between January 2022 and May 2024. Eligible patients received ABCD for up to 28 days or until investigator-assessed clinical response or intolerance. Clinical efficacy was assessed at the end of the ABCD treatment, whereas all-cause mortality was evaluated within 28 days following the last dose of ABCD. This study was carried out in accordance with the ethical principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Initial approval was granted by the Ethics Committee of the Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences (No. IIT2021035-EC-3), with subsequent ethical approvals were obtained from the institutional review boards of all participating hospitals. This trial was registered on the www.Chictr.org.cn under the identification number ChiCTR2100052369. All patients provided written informed consent.

Patient Eligibility

The study population consisted of adults (18 to 70 years) who with hematologic disease confirmed by bone marrow cytology, histochemistry, and immunology, who were diagnosed with proven/probable/possible IA per the European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/MSG) 2020 criteria. 14 Specifically, patients classified as possible IA satisfied the criteria for host risk factors and IA-consistent clinical features, with no supporting microbiological evidence of aspergillosis. Host factors included hematologic malignancies, hematopoietic stem cell transplantation, prolonged neutropenia, or receipt of immunosuppressive therapy. Key exclusion criteria included prior intravenous polyene therapy, known hypersensitivity to AmB or its cholesterol sulfate complex, significant hepatic or renal dysfunction, uncontrolled hypokalemia prior to therapy initiation, and pregnant, lactating, or unwillingness to use effective contraception when applicable.

Treatment

All patients received ABCD for up to 28 days or until investigator-assessed clinical response or intolerance, whichever occurred first, as clinically appropriate. Prior to ABCD treatment, premedication with dexamethasone ± promethazine was administered to prevent infusion-related reactions. ABCD treatment started with a test dose (1.6–8.3 mg in 10 mL) over 15–30 minutes, patients who tolerated the test dose well received the protocol-specified treatment dose. The initial dose of ABCD was 0.5–1.0 mg/kg/day was individualized based on baseline risk factors (eg, 0.5 mg/kg/day for mild renal impairment, hypokalemia, prior hepatic injury, or history of severe infusion reactions; 1.0 mg/kg/day for normal organ function with severe infection progression), increased daily according to the patient’s condition to reach a maintenance dose of 2.0–4.0 mg/kg/day by the third day, which was continued until the end of treatment. The dosage and regimen of patients were adjusted by the investigators according to the patient’s condition. Treatment could be stopped at the investigator’s discretion if it was determined that continuing would not be in the patient’s best interest, including due to adverse events (AEs) or IA complete remission, irreversible progression of underlying malignancy, persistent severe neutropenia with clinical deterioration despite supportive care, or patient refusal of continued therapy.

Efficacy and Safety

Efficacy assessment was performed on the end of ABCD therapy. The primary efficacy outcome was the clinical response rate, assessed according to the Mycoses Study Group and European Organization for Research and Treatment of Cancer (MSG-EORTC) consensus criteria. 15 Clinical response comprised complete remission (CR) or partial remission (PR). CR was defined as the absence of all IFD-related clinical signs and symptoms, along with complete or near-complete resolution of imaging abnormalities. PR was defined as clinically meaningful improvement in IFD-related signs and symptoms, accompanied by significant improvement of imaging abnormalities (≥ 50%). Clinical failure encompassed stable disease (SD) and treatment failure. SD was defined as survival without disease progression but with minimal or no improvement according to clinical, radiological, and mycological criteria (< 50%). Treatment failure was defined as the occurrence of disease progression, IFD-related death, or discontinuation of ABCD treatment due to AEs. Thus, SD indicated insufficient improvement without clinical deterioration, whereas treatment failure represented worsening disease or inability to continue therapy. The clinical efficacy of ABCD was independently assessed by at least two investigators. Moreover, the secondary efficacy outcome was all-cause mortality within 28 days following the last dose of ABCD.

Safety was evaluated based on the incidence, treatment-relatedness, and severity of AEs that occurred throughout the course of ABCD treatment. AEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 5.0. All AEs were coded according to the Medical Dictionary for Regulatory Activities (MedDRA, version 27.0).

Statistical Analysis

The sample size was calculated based on the primary efficacy outcome (clinical response rate). Previous studies reported that the clinical response rates of AmB in patients with IFD were 23.4% 16 and 43.2%. 17 Hence, the expected response rate for AmB in hematologic patients with IA was estimated to be 33.3%. The anticipated treatment effect of ABCD was based on the historical cohort study by White et al 16 which showed a 25.4% absolute improvement in response rates over conventional AmB-D (48.8% vs 23.4%). Considering potential differences between historical and current study populations, a conservative treatment effect of 17.7% was assumed, corresponding to an increase from 33.3% to 51%. With a two-sided alpha of 0.05 and a power (1-β) of 0.8, the minimum required sample size was calculated to be 58 patients. Accounting for an anticipated dropout rate of approximately 10%, a final sample size of 65 patients was planned.

Data analysis was conducted using SPSS 25 (SPSS Inc., Chicago, IL). For data description, continuous variables are presented as mean ± standard deviation (SD) or as median (range), while categorical variables are summarized as numbers and percentages. Two analysis sets were defined: the full analysis set (FAS) included all patients, and the per-protocol set (PPS) required protocol adherence with at least 7 days of treatment. Exploratory subgroup analysis of the clinical response rate was performed according to IFD diagnosis, Eastern Cooperative Oncology Group performance status score (ECOG) score, baseline body temperature, and hematologic diseases. Continuous variables were assessed for normality before analysis. Normally distributed variables were compared using the independent-samples Student’s t -test, whereas non-normally distributed variables were analyzed using the Mann–Whitney U -test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. No imputation was performed for missing data; a complete-case analysis was applied. Univariate and multivariate logistic regression analyses were conducted as exploratory analyses to identify potential predictors associated with clinical response, with results expressed as odds ratio (OR) with 95% confidence interval (95% CI) and P values. Confidence intervals were calculated using the Clopper–Pearson exact binomial method. A P value < 0.05 was considered statistically significant.

Results

Clinical Characteristics

A total of 65 hematologic patients with IA were included in this study between January 2022 and May 2024. At baseline, the median (range) age of patients was 43 (18–68) years, and the majority were male (63.08%, 41/65) ( Table 1 ). Underlying hematologic disease included AML (45/65, 69.23%), acute lymphoblastic leukemia (ALL) (12/65, 18.46%), myelodysplastic syndrome (3/65, 4.61%), multiple myeloma (2/65, 3.08%), lymphoma (2/65, 3.08%), and extremely severe aplastic anemia (1/65, 1.54%). Among the 65 patients, 6 (9.23%) had probable IA, and 59 (90.77%) remained as possible IA. Across all patients, the most common infection site was the lung (92.31%, 60/65). Baseline galactomannan (GM) testing was performed in 43 patients (66.15%): serum GM was positive in 4/36 (11.11%), plasma GM in 3/6 (50.00%), and bronchoalveolar lavage fluid (BALF) GM positive in 0/1. The remaining patients were not tested, mainly due to prior empirical antifungal therapy reducing test sensitivity or critical illness precluding baseline collection. Of the 78.46% of patients who received prior antifungal prophylaxis or treatment, voriconazole was the most commonly used agent (44.62%, 29/65), followed by posaconazole (24.62%, 16/65) and caspofungin (18.46%, 12/65). Table 1 Patients’ Demographics and Main Baseline Characteristics Characteristics Total (n=65) Age, year, median (range) 43 (18–68) Male, n (%) 41 (63.08) Underlying hematologic diseases, n (%) Acute myelogenous leukemia 45 (69.23) Acute lymphoblastic leukemia 12 (18.46) Myelodysplastic syndrome 3 (4.61) Multiple myeloma 2 (3.08) Lymphoma 2 (3.08) Very severe aplastic anemia 1 (1.54) IFD diagnosis, n (%) Possible IA 59 (90.77) Probable IA 6 (9.23) Infection site*, n (%) Lung 60 (92.31) Paranasal sinus 2 (3.08) Unspecified 4 (6.15) Patients receiving GM testing 43 (66.15) Serum GM positive/tested 4/36 (11.11) Plasma GM positive/tested 3/6 (50.00) BALF GM positive/tested 0/1 (0.00) ECOG score, n (%) 0–2 61 (93.85) 3–4 4 (6.15) Previous antifungal prevention or treatment, n (%) 51 (78.46) Voriconazole 29 (44.62) Posaconazole 16 (24.62) Caspofungin 12 (18.46) Micafungin 3 (4.62) ABCD pretreatment, n (%) Dexamethasone 41 (63.08) Dexamethasone ± promethazine 22 (33.85) Hydrocortisone 2 (3.08) ABCD administration Duration, day, median (range) 19 (1–28) Mean dose, mg/kg/d, mean ± SD 2.17 ± 0.77 Cumulative dose, mg, mean ± SD 2608.92 ± 1768.80 Notes : *One patient had concurrent pulmonary and sinus infection. Four patients had unspecified infection sites due to incomplete imaging evaluation at enrollment caused by critical illness. Abbreviations : IFD, invasive fungal disease; IA, invasive aspergillosis; GM, galactomannan; BALF, bronchoalveolar lavage fluid; ECOG, Eastern Cooperative Oncology Group; ABCD, amphotericin B colloidal dispersion.

ABCD Therapy

All patients received premedication with ABCD to reduce infusion-related reactions. Among them, 41 patients (63.08%) received dexamethasone alone, 22 patients (33.85%) received dexamethasone ± promethazine, and 2 patients (3.08%) received hydrocortisone. Further, the median duration of ABCD treatment was 19 days (range, 1–28 days), the mean (SD) daily dose was 2.17 ± 0.77 mg/kg/d, and the mean (SD) cumulative dose was 2608.92 ± 1768.80 mg. Fifty-one of the 65 patients (78.46%) received ABCD treatment for ≥ 7 days and thus formed the PPS.

Efficacy Analysis

Across all patients in the FAS, 64.62% of patients (42/65) achieved clinical response with ABCD treatment, which included 6 (9.23%) CR and 36 (55.38%) PR ( Table 2 ). Notably, in the PPS, the clinical response rate increased to 80.39% (41/51), with CR and PR rates of 11.76% (6/51) and 68.63% (35/51), respectively. Additionally, only two deaths were recorded in the FAS within 28 days following the last dose of ABCD, corresponding to a low all-cause mortality rate of 3.08%. Two patients developed progressive underlying hematological disease accompanied by severe immunosuppression, refractory infection, and multiple organ dysfunction syndromes, and subsequently died after their families withdrew treatment. Table 2 The Clinical Outcomes of ABCD for Hematologic Patients with IA Clinical Outcomes, n (%) [95% CI] FAS (n=65) PPS (n=51) Clinical response 42 (64.62) [51.8–76.1] 41 (80.39) [66.9–90.2] Complete response 6 (9.23) [3.5–19.0] 6 (11.76) [4.4–23.9] Partial response 36 (55.38) [42.5–66.7] 35 (68.63) [54.1–80.9] Clinical failure 23 (35.38) [23.9–48.2] 10 (19.61) [9.8–33.1] Stable disease 15 (23.08) [13.5–35.2] 6 (11.76) [4.4–23.9] Treatment failure 8 (12.31) [5.5–22.8] 4 (7.84) [2.2–18.9] All-cause mortality Survival 63 (96.92) [89.3–99.6] 50 (98.04) [89.6–100.0] Death 2 (3.08) [0.4–10.7] 1 (1.96) [0–10.4] Abbreviations : FAS, full analysis set; PPS, per-protocol set.

Exploratory subgroup analyses demonstrated generally consistent trends in clinical response rates across subgroups between the FAS and PPS populations, with more pronounced differences observed in the PPS ( Table 3 and Figure 1 ). For IFD diagnosis, patients with possible IA had significantly higher response rate than those with probable IA in both analysis sets (FAS: 69.49% vs 16.67%, P = 0.010; PPS: 88.89% vs 16.67%, P < 0.001), with an even greater difference observed in the PPS. By baseline performance status, patients with ECOG of 0–2 had numerically higher response rates than those with ECOG of 3–4 in both analysis sets (FAS: 65.57% vs 50%; PPS: 81.25% vs 66.67%), though not statistically significant. Similarly, patients with a baseline body temperature < 38°C had a significantly higher response rate than those with a temperature ≥ 38°C (FAS: 69.49% vs 16.67%, P = 0.010; PPS: 85.11% vs 25.00%, P = 0.004). Furthermore, regarding underlying diseases, the response rate was numerically higher in patients with ALL than in those with AML (FAS: 65.57% vs 50%; PPS: 81.25% vs 66.67%), without statistical significance in either analysis set. However, given the exploratory nature of these subgroup analyses and the limited sample sizes, the findings should be interpreted cautiously and warrant validation in larger populations. Table 3 The Subgroup Analysis of Clinical Response Rate for Hematologic Patients with IA Clinical Response rate, n/N* (%) [95% CI] FAS (n=65) P Value PPS (n=51) P Value IFD diagnosis 0.010 < 0.001 Possible IA 41/59 (69.49) [56.1–80.8] 40/45 (88.89) [75.9–96.3] Probable IA 1/6 (16.67) [0.4–64.1] 1/6 (16.67) [0.4–64.1] ECOG score 0.528 0.531 0–2 40/61 (65.57) [52.3–77.3] 39/48 (81.25) [67.4–91.1] 3–4 2/4 (50) [6.8–93.2] 2/3 (66.67) [9.4–99.2] Body temperature 0.010 0.004 <38°C 41/59 (69.49) [56.1–80.8] 40/47 (85.11) [71.7–93.8] ≥38°C 1/6 (16.67) [0.4–64.1] 1/4 (25.00) [0.6–80.6] Underlying hematologic disease 0.168 0.386 Acute myelogenous leukemia 28/45 (62.22) [46.5–76.2] 27/34 (79.41) [62.1–91.3] Acute lymphoblastic leukemia 10/12 (83.33) [51.6–97.9] 10/11 (90.91) [58.7–99.8] Notes : *n/N represents the number of patients with effective treatment out of the total patients. Statistically significant differences (P < 0.05) are highlighted in bold font. Abbreviations : FAS, full analysis set; PPS, per-protocol set; IFD, invasive fungal disease; IA, invasive aspergillosis; ECOG, Eastern Cooperative Oncology Group. Figure 1 Clinical response rates in different subgroups of full analysis set ( A ) and per-protocol set ( B ). *indicates P < 0.05; ** indicates P < 0.01; *** indicates P < 0.001. Orange bars indicate a higher treatment success rate than the overall population, while blue bars indicate a lower one. Image A displays a bar graph of clinical response rates (%) by subgroup. Categories on the horizontal axis include: Possible IA, Probable IA, ECOG 0-2, ECOG 3-4, Body temperature <38 degrees C, Body temperature greater than or equal to 38 degrees C, Acute myelogenous leukemia and Acute lymphoblastic leukemia. The vertical axis shows clinical response rates (%) with tick marks at 20, 40, 60, 80 and 100. A reference line is at 64.62%. Bar values: Possible IA 69.49%*, Probable IA 16.67%, ECOG 0-2 65.57%, ECOG 3-4 50%, Body temperature <38 degrees C 69.49%*, Body temperature greater than or equal to 38 degrees C 16.67%, Acute myelogenous leukemia 62.22%, Acute lymphoblastic leukemia 83.33%. Image B shows a similar bar graph with a reference line at 80.39%. Bar values: Possible IA 88.89%***, Probable IA 16.67%, ECOG 0-2 81.25%, ECOG 3-4 66.67%, Body temperature <38 degrees C 85.11%**, Body temperature greater than or equal to 38 degrees C 25%, Acute myelogenous leukemia 79.41%, Acute lymphoblastic leukemia 90.91%. A grouped bar graph showing clinical response rate across subgroups in two analysis sets. Abbreviations : IA, invasive aspergillosis; ECOG, Eastern Cooperative Oncology Group performance status score.

Safety

AEs related to ABCD treatment are presented in Table 4 . Most treatment-related AEs (TRAEs) were mild to moderate (grade 1–2), and no life-threatening serious AEs were reported. The most common TRAEs were hypokalemia (21/65, 32.31%), hypoalbuminemia (19/65, 29.23%) and elevated alanine aminotransferase (ALT) (16/65, 24.62%). Infusion-related reactions mainly manifested as post-infusion fever (14/65, 21.54%) and chills (13/65, 20.00%). Notably, no severe nephrotoxicity was observed; elevated creatinine occurred in 8 patients (12.31%), all of which were grade 1–2. Hepatic abnormalities included elevated ALT, aspartate aminotransferase (AST), and total bilirubin levels in 16 (24.62%), 7 (10.77%), and 4 (6.15%) patients, respectively, with grade 3 elevations occurring in 3 patients with ALT elevation and 1 patient with AST elevation. Table 4 Main AEs Related to ABCD Treatment for Hematologic Patients with IA AEs, n (%) Grade 1 Grade 2 Grade 3 Grade 4 Total Fever 6 (9.23) 7 (10.77) 1 (1.54) 0 14 (21.54) Chill 10 (15.38) 3 (4.62) 0 0 13 (20.00) Headache 2 (3.08) 2 (3.08) 0 0 4 (6.15) Hypokalemia 9 (13.85) 9 (13.85) 3 (4.62) 0 21 (32.31) Anoxia 0 3 (4.62) 0 0 3 (4.62) Elevated alanine aminotransferase 11 (16.92) 2 (3.08) 3 (4.62) 0 16 (24.62) Elevated aspartate aminotransferase 4 (6.15) 2 (3.08) 1 (1.54) 0 7 (10.77) Elevated creatinine 7 (10.77) 1 (1.54) 0 0 8 (12.31) Elevated bilirubin 3 (4.62) 1 (1.54) 0 0 4 (6.15) Hypoproteinemia 3 (4.62) 16 (24.61) 0 0 19 (29.23) Abbreviation : AEs, adverse events.

Predictors of Clinical Response

Univariate analysis revealed that IFD diagnosis (P = 0.032), prior posaconazole treatment (P = 0.04), duration of ABCD therapy (P < 0.001), mean daily dose (P = 0.001), and cumulative dose (P = 0.001) were all significantly associated with clinical response ( Table 5 ). In the multivariable logistic regression analysis, longer duration of ABCD therapy remained statistically associated with clinical response after adjustment for other variables (P = 0.028; OR: 1.335, 95% CI: 1.032–1.729). However, this association should be interpreted cautiously, as longer treatment duration may partly reflect better baseline condition, treatment tolerance, or survival status, and the possibility of immortal time bias and residual confounding cannot be excluded. Table 5 Univariate and Multivariate Analyses of Clinical Response in Hematologic Patients with IA Variables Clinical Response (n=42) Non-Response (n=23) Univariate Analysis Multivariate Analysis OR (95% CI) P Value OR (95% CI) P Value Age, years, mean ± SD 42.48±11.66 45.30±12.99 0.980(0.939–1.023) 0.367 Gender, no. (%) 0.185 Male 24 (57.14) 17 (73.91) 0.471(0.155–1.433) Female 18 (42.86) 6 (26.09) 2.125(0.698–6.470) Underlying hematologic disease, n (%) AML 28 (66.67) 17 (73.91) 0.706(0.228–2.186) 0.546 ALL 10 (23.81) 2 (8.70) 3.281(0.653–16.496) 0.149 IFD Diagnosis 0.032 0.033(0.001–1.429) 0.076 Possible IA 41 (97.62) 18 (78.26) 0.088(0.010–0.806) Probable IA 1 (2.38) 5 (21.74) 11.389(1.240–104.589) Infection site, n (%) Pulmonary 38 (90.48) 22 (95.65) 2.316(0.243–22.045) 0.465 Paranasal sinus 1 (2.38) 1 (4.35) 1.864(0.111–31.259) 0.665 ECOG score, n (%) 0–2 36(85.71) 18(78.26) 0.600(0.161–2.234) 0.446 3–4 2(4.76) 2(8.70) 1.905(0.250–14.901) 0.534 Previous antifungal prevention or treatment, n (%) Voriconazole 19(45.24) 10(43.48) 0.931(0.334–2.593) 0.891 Posaconazole 14(33.33) 2(8.70) 0.190(0.039–0.930) 0.040 0.473(0.066–3.045) 0.457 Caspofungin 6(14.29) 6(26.09) 2.118(0.595–7.541) 0.247 Micafungin 1(2.38) 2(8.70) 3.905(0.334–45.586) 0.277 ABCD administration Duration, day, median (range) 21(3~40) 5(1~32) 1.136(1.063–1.215) <0.001 1.335(1.032–1.729) 0.028 Mean daily dose, mg/kg/d, mean ± SD 2.45±0.62 1.68±0.78 4.594(1.824–11.571) 0.001 2.103(0.424–10.422) 0.363 Cumulative dose, mg, mean ± SD 3158.98±1355.35 1491.91±1801.48 1.001(1.000–1.001) 0.001 0.999(0.997–1.001) 0.278 Notes : Statistically significant differences (P < 0.05) are highlighted in bold font. Abbreviations : AML, Acute myelogenous leukemia; ALL, Acute lymphoblastic leukemia; IFD, invasive fungal disease; IA, Invasive aspergillosis; SD, standard deviation; ECOG, Eastern Cooperative Oncology Group; ABCD, amphotericin B colloidal dispersion.

Discussion

This multicenter, prospective, real-world study represents the first evaluation of ABCD in Chinese hematologic patients with IA, aiming to address this critical evidence gap within this high-risk population. Our findings demonstrated that following ABCD treatment, the clinical response rate was 64.62% in the FAS and increased to 80.39% in PPS, both of which were higher than the predefined response rate of 51%. Meanwhile, the safety profile was favorable, with TRAEs being predominantly mild to moderate in severity. These results indicate that ABCD may provide clinical benefits in this high-risk population, thereby generating valuable real-world evidence to support clinical decision-making for such patients.

Cumulative evidence supports the favorable efficacy profile of ABCD for IFD, 18 , 19 and targeted investigations in hematologic patients with IA have been limited in recent years. The clinical response rate observed in this study (64.62%) was similar to the recently reported efficacy of ABCD in the proven/probable/possible IFD (67.76%), 18 further corroborating the therapeutic efficacy of ABCD in this high-risk population. Notably, the clinical response rate reached 80.39% in PPS patients who completed ≥ 7 days of ABCD treatment, which was not only higher than that of the FAS population, but also far exceeded the historical response rates (30–50%) of ABCD or AmB in immunocompromised populations. 16 , 20 This observed higher response rate may be attributed to several key factors. On one hand, this study included a high proportion of patients with possible IA, who achieved significantly superior response rate compared with those with probable IA due to the lower disease burden. These data highlight the critical role of prompt empirical treatment in improving outcomes for patients with suspected IA. On the other hand, the higher response rate observed in the PPS compared with the FAS highlights the positive impact of longer ABCD treatment duration on therapeutic outcomes. Similarly, the mortality rate reported in this study (3.08%) was significantly lower than previously reported rates in immunocompromised populations. 16 , 20 This may be related to the relatively large number of possible IA cases enrolled, and concurrently reflects the favorable efficacy and safety profile of ABCD.

In particular, exploratory subgroup and regression analyses identified several clinical factors associated with variations in clinical response to ABCD. Subgroup analysis revealed that the response rate was significantly higher in patients with possible IA than in those with probable IA. Additionally, patients with better performance status (ECOG score 0–2) and lower baseline body temperature (< 38°C) showed a trend toward higher efficacy. These findings may suggest that patients with lower disease burden or milder systemic inflammatory responses are more likely to achieve favorable treatment outcomes. Consistent with current guidelines emphasizing the importance of early empiric therapy in high-risk patients with suspected IA, 21 , 22 our findings suggest that early recognition and timely treatment may have clinical implications. Similarly, univariate analysis results showed that the IFD diagnosis was significantly correlated with clinical response, suggesting that differences in disease characteristics and diagnostic stratification may influence treatment outcomes. Furthermore, longer ABCD treatment duration, higher daily dose, and higher cumulative dose were all significantly correlated with clinical response. Exploratory multivariable analysis suggested that longer duration of ABCD treatment remained statistically associated with clinical response. However, this association should be interpreted with caution, as longer treatment duration may be a marker of better baseline status, treatment tolerability, or survival rather than a causal factor; immortal time bias and residual confounding remain possible. In contrast, a study conducted in immunocompromised patients demonstrated that shorter antifungal treatment duration was independently associated with IA recurrence, 23 together underscoring the importance of completing a full course of antifungal therapy. Moreover, the response rate was significantly higher in the ALL subgroup than in AML patients, a finding that may be related to differences in disease characteristics, chemotherapy intensity or degree of immunosuppression, but should be interpreted cautiously given the limited sample size and requires validation in larger studies. Taken together, these findings suggest that timely initiation and appropriate duration and dosing of ABCD therapy may be important considerations for optimizing clinical outcomes in patients with IA and hematological malignancies. However, given the exploratory nature of these subgroup and regression analyses, the observed associations should be interpreted cautiously and warrant further validation in larger cohorts.

The safety profile of ABCD was favorable in this study, particularly with respect to renal safety. No grade 3/4 nephrotoxicity events were observed during the treatment period, suggesting a low risk of severe renal toxicity in this population. The incidence of elevated serum creatinine (12.31%) was consistent with prior meta-analysis data for lipid-based formulations of AmB, 24 while remaining lower than that reported for conventional AmB (26.1%–33.8%). 24 , 25 Despite the low incidence of severe renal toxicity observed in our cohort, patients with hematologic diseases often have multiple risk factors for kidney injury, including underlying disease conditions, concomitant nephrotoxic chemotherapies, and sepsis. Therefore, careful monitoring of renal function remains essential during ABCD therapy. Nevertheless, although a low incidence of grade 3/4 hypokalemia (4.62%), the overall rate was 32.31%, indicating the need for continued monitoring of electrolyte abnormalities during therapy. All elevations in total bilirubin observed in this study were grade 1–2, consistent with previous reports characterizing ABCD-associated hepatic abnormalities as mild and reversible. 11 , 26 , 27 Overall, ABCD demonstrated a favorable tolerability profile in patients with hematologic diseases, supporting its clinical use in this vulnerable population.

Despite prospectively establishing the clinical benefits of ABCD in patients with hematologic disease complicated with IA, the present study has certain limitations. First, the single-arm design and lack of a comparator group in this study makes it susceptible to time-dependent bias. Second, although the overall sample size met the statistical requirements, the relatively small size of certain subgroups may have compromised the power of the subgroup analyses. Therefore, the corresponding findings of the exploratory subgroup analysis and regression analysis require verification in larger-scale cohort. Third, the predominance of patients with possible IA (90.77%) in our cohort may limit the robustness and generalizability of the efficacy findings. Given the rapid progression and high mortality of IA in patients with hematological malignancies, initiating treatment based on clinical and radiological evidence before microbiological confirmation is common practice in high-risk populations and reflects real-world clinical management. Nevertheless, the relatively high response rate and low mortality observed in our cohort should be interpreted cautiously and may not be directly extrapolated to patients with microbiologically confirmed IA. Finally, the relatively short follow-up period precluded a comprehensive assessment of long-term outcomes, including relapse, recurrence, and survival. However, this exploratory observational study primarily focused on characterizing the short-term clinical response and safety profile of ABCD in high-risk patients with IA. Future studies should include a larger number of patients with proven/probable IA patients, in order to validate potential efficacy differences across diagnostic levels. Longer follow-up periods are also warranted to provide more comprehensive evidence regarding long-term outcomes.

In conclusion, this multicenter real-world observational study suggests that ABCD may be a potential treatment option for IA in Chinese patients with hematologic diseases, with a favorable safety profile. However, given the lack of a control group and the predominance of patients with possible IA, larger prospective studies with appropriate comparator groups are needed to further validate these findings.

Acknowledgments

We would like to acknowledge all the participants and their families and caregivers as well as the investigators and site staff.

Data Sharing Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics Approval and Informed Consent

This study was carried out in accordance with the ethical principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Initial approval was granted by the Ethics Committee of the Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences (No. IIT2021035-EC-3), with subsequent approval obtained from all participating sites. This trial was registered on the www.Chictr.org.cn under the identification number ChiCTR2100052369. All patients provided written informed consent.

Consent for Publication

The subjects gave written informed consent for the publication of any associated data.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors have no competing interests to declare.

References

1. Vergidis P , Sendi P , Alkhateeb HB , Nguyen MH . How do I manage refractory invasive pulmonary aspergillosis . Clin Microbiol Infect . 2024 ; 30 ( 6 ): 755 – 11 . doi: 10.1016/j.cmi.2024.01.015 38286175

2. Iyadorai T , Tay ST , Liong CC , et al. A review of the epidemiology of invasive fungal infections in Asian patients with hematological malignancies (2011–2021) . Epidemiol Rev . 2024 ; 46 ( 1 ): 1 – 12 . doi: 10.1093/epirev/mxae003 38778757

3. Gras E , Azoyan L , Monzo-Gallo P , et al. Risk factors for invasive mould infections in adult patients with hematological malignancies and/or stem cell transplant: a systematic literature review and meta-analysis . J Infect . 2025 ; 91 ( 3 ): 106574 . doi: 10.1016/j.jinf.2025.106574 40803450

4. Denning DW . Global incidence and mortality of severe fungal disease . Lancet Infect Dis . 2024 ; 24 ( 7 ): e428 – e438 . doi: 10.1016/S1473-3099(23)00692-8 38224705

5. Patterson TF , Thompson III GR , Denning DW , et al. Practice Guidelines for the Diagnosis and Management of Aspergillosis: 2016 Update by the Infectious Diseases Society of America . Clin Infect Dis . 2016 ; 63 ( 4 ): e1 – e60 . doi: 10.1093/cid/ciw326 PMC4967602 27365388

6. Lamoth F , Calandra T . Pulmonary aspergillosis: diagnosis and treatment . Eur Respir Rev . 2022 ; 31 ( 166 ): 220114 . doi: 10.1183/16000617.0114-2022 PMC9724826 36450372

7. Wichmann D , Hoenigl M , Koehler P , et al. Diagnosis and treatment of invasive pulmonary aspergillosis in critically ill intensive care patients: executive summary of the German national guideline (AWMF 113-005) . Infection . 2025 ; 53 ( 4 ): 1299 – 1310 . doi: 10.1007/s15010-025-02572-2 PMC12316785 40465080

8. Lee JSF , Cohen RM , Khan RA , et al. Paving the way for affordable and equitable liposomal amphotericin B access worldwide . Lancet Glob Health . 2024 ; 12 ( 9 ): e1552 – e1559 . doi: 10.1016/S2214-109X(24)00225-0 PMC11345448 39151989

9. Hoenigl M , Lewis R , van de Veerdonk FL , Verweij PE , Cornely OA . Liposomal amphotericin B-the future . J Antimicrob Chemother . 2022 ; 77 ( Suppl_2 ): ii21 – ii34 . doi: 10.1093/jac/dkac353 PMC9693803 36426674

10. Chen L , Zhao C , Xu Z , Yu Y , Xuan J . Comprehensive clinical evaluation of amphotericin B in the treatment of invasive fungal disease . Chin J Mod Appl Pharm . 2025 ; 42 ( 15 ): 2698 – 2705 .

11. Hamill RJ . Amphotericin B formulations: a comparative review of efficacy and toxicity . Drugs . 2013 ; 73 ( 9 ): 919 – 934 . doi: 10.1007/s40265-013-0069-4 23729001

12. Liu M , Yang H , Liu Q , He K , Yuan J , Chen Y . Amphotericin B Colloidal Dispersion is Efficacious and Safe for the Management of Talaromycosis in HIV-Infected Patients: results of a Retrospective Cohort Study in China . Infect Drug Resist . 2024 ; 17 : 5581 – 5593 . doi: 10.2147/IDR.S481856 PMC11653854 39697558

13. Herbrecht R . Safety of amphotericin B colloidal dispersion . Eur J Clin Microbiol Infect Dis . 1997 ; 16 ( 1 ): 74 – 80 . doi: 10.1007/BF01575124 9063677

14. Donnelly JP , Chen SC , Kauffman CA , et al. Revision and Update of the Consensus Definitions of Invasive Fungal Disease From the European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium . Clin Infect Dis . 2020 ; 71 ( 6 ): 1367 – 1376 . doi: 10.1093/cid/ciz1008 PMC7486838 31802125

15. Segal BH , Herbrecht R , Stevens DA , et al. Defining responses to therapy and study outcomes in clinical trials of invasive fungal diseases: mycoses Study Group and European Organization for Research and Treatment of Cancer consensus criteria . Clin Infect Dis . 2008 ; 47 ( 5 ): 674 – 683 . doi: 10.1086/590566 PMC2671230 18637757

16. White MH , Anaissie EJ , Kusne S , et al. Amphotericin B colloidal dispersion vs amphotericin B as therapy for invasive aspergillosis . Clin Infect Dis . 1997 ; 24 ( 4 ): 635 – 642 . 9145737

17. White MH , Bowden RA , Sandler ES , et al. Randomized, double-blind clinical trial of amphotericin B colloidal dispersion vs amphotericin B in the empirical treatment of fever and neutropenia . Clin Infect Dis . 1998 ; 27 ( 2 ): 296 – 302 . doi: 10.1086/514672 9709879

18. Zhang L , Zhang Q , Ma W , et al. Efficacy and Safety of Amphotericin B Colloidal Dispersion for Patients with Invasive Fungal Disease and Febrile Neutropenia: a Registry-Based, Multicenter, Retrospective, Real-World Study . Infect Drug Resist . 2025 ; 18 : 2475 – 2487 . doi: 10.2147/IDR.S494985 PMC12085143 40384797

19. Li Z , Li F , Chen T , Yang R . Efficacy and Safety of Amphotericin B Colloidal Dispersion via Nebulized Inhalation Combined with Intravenous Therapy for Invasive Pulmonary Fungal Disease: a Single-Center, Retrospective Cohort Study . Infect Drug Resist . 2025 ; 18 : 3415 – 3425 . doi: 10.2147/IDR.S519105 PMC12258249 40662166

20. Bowden R , Chandrasekar P , White MH , et al. A double-blind, randomized, controlled trial of amphotericin B colloidal dispersion versus amphotericin B for treatment of invasive aspergillosis in immunocompromised patients . Clin Infect Dis . 2002 ; 35 ( 4 ): 359 – 366 . doi: 10.1086/341401 12145716

21. Ruhnke M , Cornely OA , Schmidt-Hieber M , et al. Treatment of invasive fungal diseases in cancer patients-Revised 2019 Recommendations of the Infectious Diseases Working Party (AGIHO) of the German Society of Hematology and Oncology (DGHO) . Mycoses . 2020 ; 63 ( 7 ): 653 – 682 . doi: 10.1111/myc.13082 32236989

22. Taplitz RA , Kennedy EB , Bow EJ , et al. Outpatient Management of Fever and Neutropenia in Adults Treated for Malignancy: American Society of Clinical Oncology and Infectious Diseases Society of America Clinical Practice Guideline Update . J Clin Oncol . 2018 ; 36 ( 14 ): 1443 – 1453 . doi: 10.1200/JCO.2017.77.6211 29461916

23. Shin DH , Yoo S-J , Jun KI , et al. Short course of voriconazole therapy as a risk factor for relapse of invasive pulmonary aspergillosis . Sci Rep . 2020 ; 10 ( 1 ): 16078 . doi: 10.1038/s41598-020-73098-w PMC7527978 32999399

24. Steimbach LM , Tonin FS , Virtuoso S , et al. Efficacy and safety of amphotericin B lipid-based formulations-A systematic review and meta-analysis . Mycoses . 2017 ; 60 ( 3 ): 146 – 154 . doi: 10.1111/myc.12585 27878878

25. Botero Aguirre JP , Restrepo Hamid AM . Amphotericin B deoxycholate versus liposomal amphotericin B: effects on kidney function . Cochrane Database Syst Rev . 2015 ; 2015 ( 11 ): Cd010481 . doi: 10.1002/14651858.CD010481.pub2 PMC10542271 26595825

26. Guo LS . Amphotericin B colloidal dispersion: an improved antifungal therapy . Adv Drug Deliv Rev . 2001 ; 47 ( 2–3 ): 149 – 163 . doi: 10.1016/S0169-409X(01)00104-1 11311990

27. Miller MA . Reversible hepatotoxicity related to amphotericin B . Can Med Assoc J . 1984 ; 131 ( 10 ): 1245 – 1247 . PMC1483710 6594184