Background
Histoplasmosis, caused by Histoplasma capsulatum, poses a diagnostic challenge due to its worldwide distribution, its heterogeneous clinical presentation, and the limitations of current diagnostic modalities. The clinical spectrum of Histoplasma infections is broad and includes pulmonary and extrapulmonary disease, including cardiac, gastrointestinal, and central nervous system disease. Additionally, individuals may present acutely after infection or with chronic undiagnosed disease. Diagnostic testing for this infection has traditionally relied upon culture and histopathology, both of which can frequently miss acute pulmonary infection with this organism due to limited analytical sensitivity. Antigen and serology diagnostics are more sensitive but miss a non-trivial proportion of cases.Reference Azar and Hage1 Because of these factors, the diagnosis of histoplasmosis is frequently elusive and appropriate antifungal coverage is often delayed while de-escalation of unnecessary antimicrobials is prolonged.
In the United States, next generation sequencing of microbial cell-free DNA (mcfDNA NGS) is available as an open-ended testing assay from patient sera and can detect hundreds of bacteria, fungi, DNA-based viruses, and parasites. At Baylor-St. Luke’s Medical Center (BSLMC), a quaternary medical center (Houston, Texas), mcfDNA NGS is often used once traditional diagnostics (eg, antigen assays, etc.) were unrevealing or in cases where other tests or specimen sources (eg, tissue biopsy, etc.) were infeasible.
Methods
The medical records at BSLMC were queried for mcfDNA NGS results with Histoplasma capsulatum between 2018 and 2025. Cases were included if they were found to be positive for Histoplasma capsulatum mcfDNA NGS. A test was considered positive if the mcfDNA NGS concentration threshold in molecules per microliter (MPM) was above that in a previous validation study of healthy controls.Reference Blauwkamp, Thair and Rosen2 McfDNA NGS was performed on patient sera using the Karius test (Redwood City, CA) while cultures, cytology, and histopathology were performed in-house. MiraVista Diagnostics (Indianapolis, IN) performed antigen testing.
For cases with positive mcfDNA NGS for H. capsulatum, the electronic medical record was reviewed for presenting symptoms, available testing, the first test to establish the diagnosis of histoplasmosis, treatment courses, and outcomes. Clinical turnaround times (TAT) were calculated as the time between collection of serum samples for mcfDNA NGS testing, fungal cultures, tissue biopsy, or bronchoalveolar lavage and the reporting of results. This study was approved by the Baylor Institutional Review Board.
Results
Eight patients with positive mcfDNA NGS for H. capsulatum were included. Their presentation and test results are summarized in table 1. The median age was 57.5 and all but one were male. Three presented with isolated pulmonary disease, two had pulmonary disease in combination with extrapulmonary disease, and three were diagnosed with exclusively extrapulmonary disease. Five had known immunocompromising conditions including HIV and inflammatory bowel disease requiring monoclonal antibody therapy. Two had prosthetic material, including a ventriculoperitoneal shunt and a mechanical mitral valve replacement.
Clinical characteristics, laboratory testing, and outcomes for eight patients with histoplasmosis
Table 1. Long description
The table presents data on eight patients with histoplasmosis, detailing their clinical characteristics, laboratory testing results, and outcomes. It includes columns for patient descriptions, Histoplasma antigen tests, fungal cultures, fungal stains, tissue biopsies, organisms detected in mcfDNA NGS and concentration, changes in management, and disposition and outcome. The table has eight rows, each corresponding to a different patient, and multiple columns providing specific test results and clinical details. Notable trends include the detection of H. capsulatum in various tests and the subsequent changes in patient management, such as the start of itraconazole or liposomal amphotericin B treatments.
mcfDNA, microbial cell-free DNA; NGS, Next-generation sequencing; HIV, Human immunodeficiency virus; AIDS, Acquired immunodeficiency syndrome; HLH, Hemophagocytic lymphohistiocytosis; EIA, Enzyme immunoassay; MPM, Molecules per microliter; GMS, Grocott’s methylamine silver.
All patients had urine Histoplasma antigen tests; only four had antigen levels above the assay detection thresholds. One had an immunodiffusion assay which had a positive M band, and one had a negative cerebrospinal fluid antigen test. Seven had fungal cultures of various tissues including blood, bone marrow, skin, cerebrospinal fluid, bronchoalveolar lavage, and valve tissue, three of these cultures were positive. The same seven patients also had a biopsy of corresponding tissue (bone marrow, skin, aortic valve tissue, bronchoalveolar lavage, and lung tissue), of which six yielded histopathologic evidence of Histoplasma. Five patients had quantified mcfDNA NGS concentrations, while three underwent an earlier version of the test that did not report quantified results. Among those with quantified results, the results ranged between 65 and 2,729 MPM.
The TAT for mcfDNA NGS was compared with Histoplasma antigen testing, fungal cultures, and tissue biopsy or cytology. Median TAT for mcfDNA NGS was 2.5 days (range 2–5 d), compared with 4.5 days (range 2–8 d) for Histoplasma urine antigen, 6.5 days (range 6–13 d) for positive fungal culture, and 2 days (range 1–4 d) for tissue pathology or cytology. Antifungal therapy targeting Histoplasma was initiated based on mcfDNA NGS results in five of eight patients (62.5%), following biopsy in two patients, and empirically in one patient with a prior history of histoplasmosis. Six were initially treated with liposomal amphotericin B and transitioned to itraconazole following clinical improvement (median of 14 d of amphotericin B therapy, range 10–21 d), while two were started on itraconazole without a prior course of liposomal amphotericin B. Following transition to itraconazole, six were discharged, one was transferred to a long-term acute care hospital, and one died in the hospital from septic shock.
Discussion
This case series describes eight patients with Histoplasma infections and challenging diagnostic courses. Most underwent invasive tissue sampling procedures to obtain cultures and histologic samples. Serum and blood-based diagnostics were frequently negative, and several patients diagnosed by mcfDNA NGS or Histoplasma serology still underwent invasive sampling, either while these tests were pending or after results to confirm the presence of fungi in the suspected infection site and establish the clinical syndrome along with the causative organism.
McfDNA NGS has been recognized along with other molecular assays as a valuable modality for diagnosing histoplasmosisReference Barros, Wheat and Hage3 and previous case studies in children and adults have documented its utility when antigen assays and cultures are unrevealing or histopathology is inconclusive.Reference Patel, Doss and Anche4–Reference Chang, Saleh and Wales5 In this case-series, mcfDNA NGS offered a faster diagnosis and facilitated prompt pathogen-directed antifungals in five of eight patients, particularly in those without easily accessible tissue.
Although cytology results had a slightly shorter TAT than mcfDNA NGS, this metric does not account for delays in tissue acquisition, which can be prolonged by clinical instability or logistical barriers to performing invasive procedures. Such delays can substantially extend the time to diagnosis. The mcfDNA NGS assay is therefore most likely to be useful when sent in tandem with the decision to pursue invasive sampling rather than after the sampling has occurred, though it may still be useful when tissue cultures or biopsies are unrevealing or inconclusive.
This study is limited by its small sample, and the inability to ascertain the test’s sensitivity and specificity for this mycosis. Though every person in this series had a compatible syndrome, a positive test result may not necessarily correlate with an imaging finding and thus invasive testing may still be required after mcfDNA NGS results. Larger cohorts are necessary to determine if mcfDNA matches for this fungus can be attributed to an infectious syndrome and to characterize the false-positive rate. Furthermore, the prohibitive cost of the test compared to more widely available commercial assays should also prompt careful consideration of the need for time-sensitive diagnosis before sending patient sera for analysis, particularly in patients with mild or moderate disease for whom rapid treatment may not be immediately necessary.
Histoplasma infections are common throughout the United States and not just in the Mississippi and Ohio River Valleys where it has been traditionally thought to be endemic, with its distribution changing over recent decades.Reference Benedict, Toda and Jackson6 It is therefore not surprising that the manufacturers of the Karius test have found Histoplasma mcfDNA to be the most common among dimorphic fungi in samples received by the largest company performing NGS in the United States.Reference Park, Chang and Ledeboer7 Larger studies are necessary to determine the optimal timing and role for this assay in identifying the large and growing number of patients for whom prompt diagnosis may translate into better outcomes.
Acknowledgments
None.
Financial support
None reported.
Competing interests
All authors report no conflicts of interest relevant to this article.
This research did not receive any grant from the public, commercial, or not-for-profit sectors.
This study was approved by the institutional review board of Baylor College of Medicine and complies with journal ethics policies.
Ethical standard
This study was reviewed and approved by the Baylor College of Medicine Institutional Review Board (protocol H-45143).
Consent for publication
The Institutional Review Board waived the requirement for written informed consent for this retrospective analysis of fully de-identified data.
Open access
