Geotrichum Capitatum Infections in the Hematologic Setting

Introduction: Geotrichum capitatum is a ubiquitous fungus in nature and can be responsible for invasive fungal infections (IFI) in immunocompromised patients (pts) in the hematologic setting, most often within the context of a diagnosis of acute leukemia.

Approximately 85% of cases of G. capitatum IFI have been reported in Europe (with approximately 85% of which in Italy and Spain), and it has been suggested that the mediterranean climate could be a contributing factor to its epidemiology; considering the limited geographic distribution of reported cases, the largest published multicentric clinical series comprises just 35 pts. The narrow clinical experience with this IFI thus hinders the development of treatment guidelines, and therefore larger series are needed to address these limitations.

Methods: We reviewed all fungal infections in our mediterranean Hematology Department in adult pts (18 years and over) over a 10.5 year period from October 2011 to April 2022, selecting cases with positive fungal identifications of G. capitatum by mycologic methods. The date of "Diagnosis” (Dx) was considered to be the moment of collection of the positive sample.

Results: We identified 39 patients with G. capitatum isolates. The average number of cases per year was 3.7 (range: 0-6), with a peak of infections in 2014-2015 temporally coinciding with construction work in the main hospital building, in proximity to the Hematology ward. The mean age at IFI-Dx was 59.0 years (range: 23.2-83.2; median: 64.5) and 61.5% of pts were male.

The majority of pts had a Dx of acute leukemia - 61.5% (n=24) myeloid (AML) and 7.7% (n=3) lymphoblastic (ALL) -, with a further 5.1% (n=2) with myelodysplastic syndromes (MDS) with excess blasts; 5.1% (n=2) had Burkitt's lymphoma, 12.8% (n=5) had other chronic lymphoproliferative diseases, 5.1% (n=2) had multiple myeloma (MM), and one pt (2.5%) had aplastic anemia.

Two-thirds of pts (66.7%, n=26) presented with fungemia in cultures of peripheral blood (PB) or central venous access (CVA) blood (with 38.5%, n=15, in both); 28.2% (n=11) had positive cultures in either sputum or bronchial aspirate (n=1 in both); 7.7% had urinary isolates and one pt each had positive cultures in a wound exudate and in abscess material.

The median OS after IFI-Dx was 17 days (range: 0 days-8.5 years); 31% of pts (n=12) died within a week of Dx, half of whom within the first 3 days, often before fungal identification was established; these pts (none of whom were autopsied) were considered a posteriori to have a G. capitatum IFI as their cause of death. A further 33% of pts (n=13) died from a week to a month after Dx; the direct cause of death in this group of pts could not be established in the absence of autopsy, and both the infection and the underlying hematologic disease were considered equally likely causes. A final 21% of pts (n=8) died between one month and one year after IFI-Dx, and the likeliest cause of death was considered to be progressive disease. Five pts (13%), treated with voriconazole+liposomal amphotericin B, are still alive, with no evidence of recurring IFI, and in complete remission (CR) of the underlying hemato-oncology disease: one with ALL at 8.3 months post-IFI-Dx, one with diffuse large B-cell lymphoma at 11.2 months post-IFI-Dx, one with MM at 4.1 years, and two cases of AML, at 7.2 and 8.5 years post-IFI-Dx, respectively. While the ALL pt had fungemia (in both PB and CVA cultures), the other four surviving pts had no evidence of fungemia and had positive cultures in sputum (n=3) and abscess fluid (n=1).

Discussion: We present a very large unicentric clinical series, overtaking the largest published multicentric series reported in the literature. In our cohort, the acute leukemias together with high risk MDS account for three quarters of cases.

Approximately one third of patients died of active infection in the acute phase of IFI, while one fifth died of progressive hemato-oncologic disease; in one third of patients, who were not autopsied, the cause of death (IFI vs progressive disease) was not established. Only one-eighth of the cohort currently survives in CR; this subset of pts is relevant for the relative absence of fungemia - present in only 20% of pts, vs 66.7% of the full cohort.

Our retrospective series of a rare infection, although underpowered to allow statistical analysis (as has been the case with previously published data) relevantly expands the current knowledge base on G. capitatum IFI in the hematologic patient.

No relevant conflicts of interest to declare.