Successful salvage treatment of Mucor nidicola and Aspergillus calidoustus co-infection with oral encochleated amphotericin B (MAT2203) in a pediatric patient with acute myeloid leukemia Authors: Christopher P. Ouellette, Monica I. Ardura, Connie Gibas, Nathan Wiederhold, Eunkyung Song Published: 2026 ABSTRACT Antifungal treatment of invasive fungal infections (IFI) in immunocompromised children is challenging. We present a case of rhinocerebral and pulmonary IFI caused by Mucor nidicola and Aspergillus calidoustus with limited azole susceptibility successfully treated with MAT2203 after developing toxicity from intravenous liposomal amphotericin B. Case summary A 15-year-old, 97 kg female (BMI 38 kg/m2) with acute myeloid leukemia (AML) presented with fever, right-sided facial pain and swelling three weeks after completion of intensification II chemotherapy (day 0 of symptom onset). Her absolute neutrophil count (ANC) had remained <100 cells/μL for the preceding 2 weeks, during which she received levofloxacin and voriconazole prophylaxis (voriconazole trough concentration 2.38 μg/mL on admission; Fungus Testing Laboratory, University of Texas Health Sciences Center, San Antonio, TX). Nasopharyngoscopy revealed necrosis and edema of the right middle turbinate. Magnetic resonance imaging (MRI) demonstrated right paranasal sinus disease without contrast enhancement, bony destruction, orbital or central nervous system (CNS) involvement. Ophthalmology examination was unremarkable. Chest computed tomography (CT) revealed multiple pulmonary macronodules (Fig. 1). Empiric IV liposomal amphotericin B (LAMB, 5 mg/kg/day) and IV posaconazole (300 mg q12h x 2 followed by 300 mg daily) were initiated concurrently with an emergent sinus debridement and bronchoalveolar lavage (BAL), performed approximately 24 hours after symptom onset.Fig. 1Timeline of Diagnosis, Pathogen Identification, Imaging, and Management course in a Pediatric Patient with Invasive Mold Co-InfectionA. Sinus computed tomography (CT) showing mucosal thickening and opacification involving the right sphenoid, ethmoid, and maxillary sinuses, with obstruction of the right ostiomeatal unit infundibulum.B. Periodic acid–Schiff (PAS) staining of sinus tissue from the right middle turbinate reveals necrotic debris and the presence of fungal elements: B-1. Septate hyphae; B-2. Non-septate hyphaeC. Chest CT demonstrating new right-sided patchy ground-glass and nodular opacities, suggestive of pulmonary fungal involvement.Abbreviation: CNS central nervous system, PCR polymerase chain reaction, BAL bronchoalveolar lavage, Asp GM aspergillus galactomannan, TIW three times a week PAS stain of sinus tissue demonstrates persistent fungal elements along the surgical margin. PAS stain of sinus tissue reveals absence of fungal elements at the tissue margin. Timeline of Diagnosis, Pathogen Identification, Imaging, and Management course in a Pediatric Patient with Invasive Mold Co-Infection A. Sinus computed tomography (CT) showing mucosal thickening and opacification involving the right sphenoid, ethmoid, and maxillary sinuses, with obstruction of the right ostiomeatal unit infundibulum. B. Periodic acid–Schiff (PAS) staining of sinus tissue from the right middle turbinate reveals necrotic debris and the presence of fungal elements: B-1. Septate hyphae; B-2. Non-septate hyphae C. Chest CT demonstrating new right-sided patchy ground-glass and nodular opacities, suggestive of pulmonary fungal involvement. Abbreviation: CNS central nervous system, PCR polymerase chain reaction, BAL bronchoalveolar lavage, Asp GM aspergillus galactomannan, TIW three times a week PAS stain of sinus tissue demonstrates persistent fungal elements along the surgical margin. PAS stain of sinus tissue reveals absence of fungal elements at the tissue margin. Sinus tissues demonstrated necrosis with both non-septate and septate hyphae on Periodic Acid-Schiff (PAS) stain, concerning for concurrent Mucorales and Aspergillus spp., respectively (Fig. 1). Sinus tissue fungal cultures grew two Mucor spp. (UTHSCSA DI25-108, GenBank PV528836 and UTHSCSA DI25-109, GenBank PV528837), later identified as Mucor nidicola by BLASTn searches using DNA sequences of the ITS region (99.78% similarity to Mucor nidicola NRRL 54520). Universal fungal PCR (uPCR, University of Washington, Seattle, WA) on sinus tissue detected M. nidicola and M. irregularis. Fungal culture and uPCR from the BAL specimen were negative. However, the Aspergillus galactomannan antigen (AGM; BioRad Platelia Aspergillus Galactomannan Ag EIA, France) was positive in both serum (optical density index (ODI) 1.17) and BAL fluid (ODI 2.11). Plasma cell-free metagenomic next generation sequencing (mNGS, KARIUS®, CA) detected Aspergillus calidoustus (56 molecules per microliter, reference range <10) only. Fever resolved after the second sinus surgery, however, during week 2 of antifungal therapy, she developed worsening facial pain, swelling, and headache, coinciding with ANC recovery (700 cells/μL). Repeat facial and brain MRI revealed new bilateral dural enhancement, prompting an increase in LAMB to 7.5 mg/kg/day to balance concerns for CNS involvement with anticipated nephrotoxicity during prolonged therapy. Lumbar puncture revealed no cerebral spinal fluid (CSF) pleocytosis. Fungal CSF cultures were sterile, however, Mucorales nested uPCR on CSF detected M. ardhlaengiktus. Posaconazole was transitioned to isavuconazole (372 mg daily following loading). She underwent sinus debridement 4 times during the first 2 weeks of therapy, each continuously showing fungal elements at tissue margins. After the fifth debridement, 12 days after the initial surgery and antifungal therapy, PAS stains became negative, corresponding with improvement in facial pain and swelling. Antifungal susceptibility testing against M. nidicola demonstrated high minimum inhibitory concentrations (MIC) to posaconazole and isavuconazole (>16 μg/mL) as well as the investigational agent, manogepix (MIC > 4μg/mL), whereas the amphotericin B MIC was 0.06 μg/mL (Fig. 1). Reported MICs for M. irregularis and M. ardhlaengiktus to isavuconazole in literature were relatively high (>8μg/mL), with amphotericin B demonstrating the lowest MIC [1]. Because A. calidoustus similarly exhibits poor azole activity and no breakpoints exist to guide therapy [2], and therapeutic azole levels would not exceed the M. nidicola MIC, the regimen was transitioned to LAMB monotherapy. By week 7, the dose was reduced to 5 mg/kg/day as MRI showed resolution of dural enhancement and chest CT demonstrated smaller macronodules. After 10 weeks, serum AGM ODI remained <0.5. By week 12, LAMB was decreased to thrice-weekly dosing as a consolidative therapy to mitigate cumulative nephrotoxicity while maintaining antifungal exposure. However, the patient developed acute kidney injury with Fanconi syndrome attributed to LAMB, requiring several hospitalizations for hydration and electrolyte correction. Due to limited therapeutic alternatives and toxicity concerns, an emergent investigational new drug application for MAT2203 was obtained, and therapy (Dose 1200 mg/day divided q6h, oral suspension) began at week 14. Renal function began to improve within 10 days, and serum creatinine normalized by 4 weeks of MAT2203 without further toxicities. Chest CT at week 16 showed reduction of the largest macronodule, and imaging at week 24 (week 10 of MAT2203) showed near-complete resolution with only a few sub-5 mm micronodules. MAT2203 was discontinued at that time. At one-year follow-up, the patient remains well with complete resolution of infection and ongoing AML remission. Discussion IFI among immunocompromised children cause substantial morbidity and mortality and require early diagnosis and timely medical and surgical management to improve outcomes. Treatment may be complicated by limited antifungal agents, drug-mediated toxicities or interactions with chemotherapy, and the potential mixed-mold infections as demonstrated in this patient. M. nidicola, previously identified only in an invasive paper wasp nest [3], was detected by fungal culture and uPCR on sinus tissues. Our clinical isolate exhibited in vitro resistance to all tested azoles and manogepix, suggesting that M. nidicola may represent a newly recognized, highly drug-resistant human pathogen that can lead to rhino-cerebral disease. This case highlights both the importance and limitations of current diagnostic approaches for IFI. Tissue biopsy remains essential, providing histopathological confirmation and material for culture and susceptibility testing. However, culture alone may have reduced diagnostic sensitivity in immunocompromised hosts, and may fail to recover all pathogens in polymicrobial infections when more rapidly growing organisms predominate [[4], [5], [6]]. Diagnosis was further complicated by mixed-mold infections, with A. calidoustus detected exclusively by plasma mNGS despite negative cultures and uPCRs on multiple specimens. While adjunctive non-culture methods including galactomannan, tissue PCR, and plasma mNGS improved overall diagnostic yield, results were discordant across modalities; M. irregularis, M. ardhlaengiktus and A. calidoustus were identified only through molecular approaches, whereas culture recovered a single organism, M. nidicola. Although BAL galactomannan and PCR individually demonstrate variable sensitivity, their combined use may improve diagnostic performance, with some studies reporting sensitivities up to 100% and specificities of 95–98% [7,8]. Nonetheless, even with multimodal testing, no single diagnostic modality reliably captures the full spectrum of pathogens in complex IFI. Instead, an integrated approach incorporating histopathology, culture, antigen assays, and molecular diagnostic testing should be obtained to guide optimal antifungal therapy. Toxicity and resistance limited conventional antifungal options. MAT2203, an orally bioavailable encochleated amphotericin B formulation, encases AMB within a nanocrystal lipid bilayer that protects it from gastric degradation. After phagocytosis by monocytes and macrophages, the lipid bilayer unravels to release AMB, enabling high drug concentrations of AMB at sites of infection while minimizing systemic toxicities [9]. Early clinical experience, including a randomized clinical trial comparing MAT2203 with IV LAMB for HIV-associated cryptococcal meningitis in adults, has provided proof-of-concept that MAT2203 can achieve comparable antifungal activity with IV LAMB with a reassuring safety profile [10]. Case series data suggest a potential utility of MAT2203 in other IFIs due to Rhodotorula sp., Candida krusei, Fusarium spp, and Histoplasma capsulatum. [11] In our patient, MAT2203 provided a well-tolerated alternative to prolonged IV LAMB therapy and enabled complete resolution of rhinocerebral and pulmonary infections [2,[12], [13], [14]]. To our knowledge, this represents the first reported pediatric use of MAT2203 as a salvage therapy resulting in successful treatment of this mixed-mold co-infection. Additional studies are needed to define the role of MAT2203 beyond cryptococcal meningitis and as a therapeutic option in immunocompromised children. Ethics statement Written informed consent for publication of clinical details and images was obtained from the patient in accordance with institutional policy. The Institutional Review Board at Nationwide Children's Hospital determined that this single-patient case report did not constitute human subjects research and did not require IRB review. Financial support Not applicable. CRediT authorship contribution statement Christopher P. Ouellette: Writing – review & editing. Monica I. Ardura: Writing – review & editing. Connie Gibas: Methodology. Nathan Wiederhold: Methodology, Writing – review & editing. Eunkyung Song: Conceptualization, Methodology, Writing – original draft, Writing – review & editing. Potential conflicts of interest NPW has received research support (paid to UTHSCSA) from Basilea, bioMérieux, Bruker, F2G, Mycovia, Scynexis, and Sfunga.