Abstract
Background:
Invasive Fungal Sinusitis (IFS) is a deadly and morbid disease of the paranasal sinuses. Immunocompromised patients are at risk of developing IFS, and survival at 6 months is approximately 50%. Previously studied treatments such as granulocyte stimulating factor have not improved survival. Thus, the immune cells responsible for survival in this vulnerable population are not well defined.
Objective:
The goal of this study was to immunophenotype cells associated with IFS survival.
Methods:
Blood samples from consenting patients were collected at the time of surgery for IFS between April 2022 and December 2023. Peripheral blood mononuclear cells (PBMCs) were evaluated for major lymphoid and myeloid subsets using mass Cytometry by Time of Flight (CyTOF) between IFS survivors and non-survivors, defined by vital status 6 months after the time of diagnosis. Data were analyzed using the dimensionality reduction algorithm optimized t-distributed stochastic neighbor embedding (opt-SNE).
Results:
Seven patients had PBMCs viable for analysis (3 survivors and 4 non-survivors). Total white blood cell counts were lower in non-survivors. Total T cell and B cell counts were decreased in non-survivors as compared to survivors, but there were no differences in other immune cell populations, such as NK cells or monocytes. Percentage differences were identified as deficiencies in CD4 T cells, particularly memory cells, in the IFS non-survivors.
Conclusion:
Future work includes the investigation of CD4 T cells as a biomarker for IFS survival and the study of T cell manipulation or stimulation in IFS patients.
Keywords: Invasive fungal sinusitis, Aspergillus, Mucor, Sinusitis, Sinonasal, Adaptive immunity
1. Introduction
Invasive fungal infections cause 1.5 million deaths annually [1]. Invasive fungal sinusitis (IFS) is an aggressive disease of the paranasal sinuses and surrounding structures. Mortality is reported as high as 80 % [2]. Risk factors include immunocompromised state, and prognosis is often associated with reversal of the underlying disease process. However immune reconstitution is often not possible due to the need for ongoing chemotherapy or iatrogenic immunosuppression, such as in bone marrow transplantation (BMT). As the number of immunocompromised patients rises due to medical advances, invasive fungal infections continue to increase [3].
Most prior IFS work has focused on clinical outcomes and prognostic factors [4,5]. This work has identified that prompt treatment and reversal of the immune compromised state are associated with improved survival. However, a deeper understanding of the systemic immune system and the timing and specifics of immune cell recovery are unknown.
Several small studies investigated whether immune stimulators or replacement therapies improve the immune response to fungal invasion. Peripheral recovery of neutrophil counts is associated with control of IFS. However, granulocyte infusions and other immune stimulators have failed to improve overall survival [2,6]. While the systemic immune system is clearly critical for control of IFS and ultimate survival, the specific cell phenotypes responsible for survival are unclear. Understanding how specific immune cell populations control IFS is integral for developing future targeted therapies and improving IFS survival in immune-deficient patients.
In this study, we investigate immune cell phenotypes that are associated with IFS survival using CyTOF to differentiate PBMCs present at the time of diagnosis between IFS patients who survived 6 months or more, and those who did not. Through this work, we attempt to identify immune phenotypes critical for IFS survival.
2. Methods
2.1. Patient recruitment and sample collection
Patients undergoing endoscopic sinus surgery were approached for potential inclusion in our sinonasal tissue bank (IRB# 201908204). This tissue bank is an ongoing collection of sinonasal tissue, nasal swabs, nasal aspirates and blood from consenting patients undergoing surgery within the Department of Otolaryngology and has been active since 2022. Analysis of blood samples from patients with confirmed IFS on surgical pathology collected between April 2022 and December 2023 was conducted under a secondary IRB protocol (IRB# 202202016). Patients with overall survival greater than 6 months after the time of initial IFS diagnosis were considered survivors in this study.
Additional data collected for the study through the biobank included basic demographic information, past medical history including immunodeficiencies; other bacterial, viral and fungal infections; and medications including antibiotics, antifungals, immunosuppressive medications, steroids, chemotherapy and immune modulators. Laboratory results were also collected including total white blood cell count, absolute neutrophil count and absolute lymphocyte count.
Blood samples obtained at time of surgery were immediately processed and cryopreserved in the Tissue Procurement Core. Cell Preparation Tubes (CPT; Fisher Scientific) were remixed by gentle inversion and centrifuged at 1500 xg for 20 min at room temperature. The mononuclear cell/platelet layers were collected and centrifuged at 400 xg for 10 min at room temperature. Cells were cryopreserved at 5 × 106 cells/aliquot in 1 ml of freezing media (90 % FBS and 10 % DMSO).
2.2. Mass cytometry by time of flight
Cryopreserved samples were thawed and washed twice with CyFACS buffer. After determining viability, cells were stained with an antibody panel designed to evaluate immune cell phenotypes (Table 1) at room temperature and then washed twice with Cell Staining Buffer (Standard Biotools). Samples were resuspended in Fixation Buffer (BD Biosciences) and placed on ice for 30 min. Samples were permeabilized and barcoded (Standard Biotools) for 30 min on ice. Cells were then resuspended in 1:3000 DNA intercalator (191lr and 193Ir) with 2 % PFA in isotonic CyPBS and placed at 4C overnight. Prior to sample acquisition, samples were washed and then suspended in a 10 % EQ Bead solution at a concentration of 1 × 106 cells/mL. All samples were run on a CyTOF 2. 1-II (Helios) machine. Following data acquisition, files were normalized.
Table 1.
CyTOF markers.
| Antibody | Mass |
|---|---|
| CD45 | 89Y |
| CD196 | 141Pr |
| CD123 | 143Nd |
| CD19 | 144Nd |
| CD4 | 145Nd |
| CD8a | 146Nd |
| CD11c | 147Nd |
| CD16 | 148Nd |
| CD45Ro | 149Sm |
| CD45RA | 150Nd |
| CD161 | 151E |
| CD194 | 152Sm |
| CD25 | 153Eu |
| CD27 | 154Sm |
| CD57 | 155Gd |
| CD183 | 156Gd |
| CD185 | 158Gd |
| CD28 | 160Gd |
| CD38 | 161Dy |
| CD56 | 163Dy |
| TCRgd | 164Dy |
| CD294 | 166Er |
| CD197 | 167Er |
| CD14 | 168Er |
| CD3 | 170Er |
| CD20 | 171Yb |
| CD66b | 172Yb |
| HLA-DR | 173Yb |
| IgD | 174Yb |
| CD127 | 176Yb |
| CD68 | 175Lu |
| CD94 | 169Tm |
| CD303 | 165Ho |
FCS files were analyzed using Cytobank software using previously published methods [7,8]. OptSNE plots were generated using proportional sampling of all live PBMCs and included 1,900,000 events from all FCS files. Dimensionality reduction included all markers except CD45 and the optSNE advanced setting values included perplexity of 50. Population statistics were exported from Cytobank and analyzed with GraphPad Prism 10.0.3 for macOS software (GraphPad, La Jolla, CA).
3. Results
Seven patients had PBMCs available for analysis (Table 2). On the day of surgery, survivors had higher total white blood cell counts on average, as compared to non-survivors (11.4 ± 5.9 K/cumm vs 1.2 ± 1.1 K/cumm, t = 3.5, p = 0.02). Absolute neutrophil count were also higher in survivors compared to non-survivors (8.1 ± 5.2 K/cumm vs 0.4 ± 0.6 K/cumm, t = 3.0, p = 0.03).
Table 2.
Patient characteristics.
| Patient cohort | Age (y) | Reason for immunosuppression/comorbidities | Fungal species |
|---|---|---|---|
| IFS survivors | 69 | Lung/rectal squamous cell carcinoma, chemotherapy, pulmonary aspergillosis | Aspergillus |
| 60 | Hyperglycemia | Mucor | |
| 33 | Hyperglycemia | Mucor, Aspergillus, candida | |
| IFS non-survivors | 53 | AML, BMT | Alternaria |
| 54 | AML, BMT | Aspergillus | |
| 47 | Thymoma, chemotherapy, aplastic anemia | Presumed Alternaria | |
| 68 | Myelodysplastic syndrome, BMT | Aspergillus, Alternaria |
After gating for live cells using established biaxial gating schemes [9,10] we performed optimized t-distributed stochastic neighbor embedding (opt-SNE) analysis using 32 markers on one healthy control, three survivors and four non-survivors of IFS (Fig. 1A–B). Opt-SNE is an unsupervised dimension reduction algorithm that allows for visualization of multidimensional data on a 2-dimensional map. We identified myeloid, T cell, and B cell populations based on visualization of surface markers on the optSNE plots (Fig. 1C).
Fig. 1.
Survivors and non-survivors of IFS show differences in lymphoid populations. A Representative opt-SNE of PBMCs from a healthy control with assigned immune cell populations. B Density contour map of concatenated survivors and non-survivors. C Heatmap expression of surface markers used to assign immune cell population identities.
No cell populations were different by percentage between non-survivors and survivors. However, in non-survivors, the absolute number of T cells (0.3 ± 0.2 K/cumm vs 5.9 ± 3.0 K/cumm, t = 3.7, p = 0.01), and B cells (0.004 ± 0.004 K/cumm vs 1.0 ±0.8 K/cumm, t = 2.8, p = 0.04) were statistically significantly decreased compared to survivors (Fig. 2A, B).
Fig. 2.
Lymphoid populations are associated with survival in IFS. A Percentage and B absolute numbers (K/cumm) of major immune populations in survivors and non-survivors with IFS. *p < 0.05.
Next, we examined subsets of T cells and B cells to determine the relative contribution of CD4 T cells, CD8 T cells, γδ T cells, memory B cells, and naïve B cells to the overall decrease in these two lymphoid populations. We found that the percentage of CD4 T cells (12.67 ± 6.8 % vs 36.9 ± 8.7 %, t = 4.1, p = 0.009) and specifically CD4 memory T cells (10.8 ± 3.3 % vs 22.9 ± 1.6 %, t = 5.8, p = 0.002) but not CD8 T cells, memory B cells, or naïve B cells was statistically significantly decreased in non-survivors compared to survivors of IFS (Fig. 3A). γδ T cell percentages trended towards a decrease in non-survivors compared to survivors (1.1 ± 1.1 % vs 3.7 ± 2.2 %, t = 2.1, p = 0.09). The absolute numbers of all T cell and B cell subsets were decreased in non-survivors compared to survivors (Figs. 3B, 4, Table 3, units for cells are K/cumm).
Fig. 3.
CD4 T cells are associated with survival in IFS. A Percentage and B absolute numbers (K/cumm) of CD4, CD8, and γδ T cell populations in survivors and non-survivors with IFS. *p < 0.05.
Fig. 4.
B cell absolute numbers are higher in IFS survivors. A Percentage and B absolute numbers (K/cumm) of naïve and memory B cells in survivors and non-survivors with IFS. *p < 0.05.
Table 3.
T and B cell absolute numbers in survivors and non-survivors of IFS.
| Population | Survivors | Non-survivors | t-Statistic | p-Value |
|---|---|---|---|---|
| CD8 T cells | ||||
| Naive | 1.0 ± 0.5 | 0.08 ± 0.1 | 3.4 | 0.02 |
| Memory | 0.6 ± 0.2 | 0.05 ± 0.04 | 4.5 | 0.007 |
| CD4 T cells | ||||
| Naive | 1.4 ± 0.9 | 0.03 ± 0.05 | 3.1 | 0.03 |
| Memory | 2.7 ± 1.5 | 0.1 ± 0.1 | 3.4 | 0.02 |
| γδ T cells | 0.4 ± 0.2 | 0.01 ± 0.02 | 3.8 | 0.01 |
| B cells | ||||
| Naive | 0.6 ± 0.4 | 0.003 ± 0.004 | 2.8 | 0.04 |
| Memory | 0.4 ± 0.3 | 0.001 ± 0.001 | 2.8 | 0.04 |
4. Discussion
In this study, we identified specific immune cell phenotypes associated with IFS survival. While we did not identify differences in percentages of all cell types, absolute numbers of all T and B cell subtypes were lower in non-survivors as compared to the survivor cohort. Interestingly, the percentage and number of monocytes was not associated with IFS survival in this study, despite evidence supporting the importance of monocytes in fungal infection response and the role of tissue macrophages in IFS [1]. Similarly, the percentage and number of NK cells, another component of innate immunity, were not different between groups.
All lymphocyte counts were severely reduced in the non-survivor population, highlighting the role of the adaptive immune system in IFS survival. Lymphocytes are well documented as critical for survival in the setting of fungal infections [11,12]. B cells produce antibodies with anti-fungal effects and also aid T cells [13]. CD8 T cells appear to play a protective role in fungal immunity which is independent of CD4 T cell function [13]. Upon further evaluation of T cell subsets, CD4 T cells and specifically, CD4 memory T cells, showed differences in proportions and total counts between survivors and non-survivors in our study. CD4 T cells are known to be critical in controlling fungal infections by influencing the innate immune system, facilitating B-cell responses, and cytotoxic CD8 T cell activation [3].
In our practice caring for IFS patients, current treatment options such as GM-CSF and granulocyte and leukocyte transfusions have shown limited and variable improvements in survival [2]. However animal work has supported the use of in vitro expanded fungal-specific CD4 T cell adoptive transfer for protection against Aspergillus infections [3,14]. This study and others, including one small human study, highlight the possibility of an established bank of HLA-matched donor antifungal T-cells, for use in the setting of invasive fungal diagnoses in vulnerable patients, such as those undergoing BMT [14–18]. Furthermore, our recently published work evaluating adaptive immunity in sinonasal mucosa supports dysregulated tissue lymphocytes in IFS patients [19].
There are limitations of this study. Several recruited and consenting patients did not have sufficient viable cells for analysis, and thus, our total analyzable sample size is small. Survival was determined by overall survival as disease-specific survival can be difficult to determine in these complicated patients with multiple ongoing medical conditions. Furthermore, patients with IFS in the literature are notoriously heterogenous with a wide range of comorbidities and immune modulating medications, making generalizable conclusions challenging [5]. In our study, two-thirds of our survivors were affected by IFS due to a temporary hyperglycemic state, whereas non-survivors were severely immunocompromised, often suffering from AML. Therefore, non-survivors were arguably more immunocompromised than their surviving counterparts. Additionally, different fungal species infected survivors and non-survivors, likely due to their underlying diagnoses [2].
Regardless of underlying patient differences between cohorts, we believe that the immune cell phenotype differences identified in this study are important. This is the first investigation of differences in immune cell populations at the time of diagnosis between surviving and non-surviving IFS patients. Our observed relative lymphocyte deficits in non-survivors support the possibility of future therapeutics targeting lymphocyte deficits in patients diagnosed with IFS. Our next steps include validation of these findings in a larger cohort from patients diagnosed with IFS from other institutions around the country.
5. Conclusion
Our work supports the key role of lymphocytes in IFS survival. Additional work in larger cohorts to determine whether lymphocyte counts may be used as a biomarker to predict survival is warranted. This work is critical to the treatment of IFS in vulnerable populations.
Acknowledgements
Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number KL2TR002346 (LTR). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This work was supported by the American Rhinologic Society Friends in Research Award (LTR).
This work was supported, in part, by the Bursky Center for Human Immunology and Immunotherapy Programs at Washington University, Immunomonitoring Laboratory.
Footnotes
CRediT authorship contribution statement
Iris Lee: Data curation, Formal analysis, Methodology, Writing – review & editing. Sonam Verma: Formal analysis, Investigation, Methodology, Writing – review & editing. John S. Schneider: Data curation, Writing – review & editing. Nyssa F. Farrell: Data curation, Writing – review & editing. Peggy L. Kendall: Conceptualization, Data curation, Formal analysis, Supervision, Writing – review & editing. Lauren T. Roland: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing.
Ethical approval
IRB approval was obtained for this study: WashU IRB# 201908204.
Declaration of competing interest
LTR: medical advisory board - Sanofi. NFF: medical advisory board - Sanofi. The other authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
- [1].Heung LJ. Monocytes and the host response to fungal pathogens. Front Cell Infect Microbiol 2020;10:34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Roland L, Humphreys I, Le C, et al. Diagnosis, prognosticators, and management of acute invasive fungal rhinosinusitis: an evidence-based review with recommendations. Int Forum Allergy Rhinol 2023;13(9):1615–714. [DOI] [PubMed] [Google Scholar]
- [3].Sharma J, Mudalagiriyappa S, Nanjappa SG. T cell responses to control fungal infection in an immunological memory lens. Front Immunol 2022;13:905867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Wandell G, Miller C, Rathor A, et al. A multi-institutional review of outcomes in biopsy-proven acute invasive fungal sinusitis. Int Forum Allergy Rhinol 2018;8 (12):1459–68. [DOI] [PubMed] [Google Scholar]
- [5].Turner JH, Soudry E, Nayak JV, Hwang PH. Survival outcomes in acute invasive fungal sinusitis: a systematic review and quantitative synthesis of published evidence. Laryngoscope 2013;123(5):1112–8. [DOI] [PubMed] [Google Scholar]
- [6].Monroe MM, McLean M, Sautter N, et al. Invasive fungal rhinosinusitis: a 15-year experience with 29 patients. Laryngoscope 2013;123(7):1583–7. [DOI] [PubMed] [Google Scholar]
- [7].Diggins KE, Ferrell PB Jr, Irish JM. Methods for discovery and characterization of cell subsets in high dimensional mass cytometry data. Methods 2015;82:55–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Nicholas KJ, Greenplate AR, Flaherty DK, et al. Multiparameter analysis of stimulated human peripheral blood mononuclear cells: a comparison of mass and fluorescence cytometry. Cytometry A 2016;89(3):271–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Wilfong EM, Bartkowiak T, Vowell KN, et al. High-dimensional analysis reveals distinct endotypes in patients with idiopathic inflammatory myopathies. Front Immunol 2022;13:756018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Wilfong EM, Vowell KN, Bunn KE, et al. CD19 +CD21lo/neg cells are increased in systemic sclerosis-associated interstitial lung disease. Clin Exp Med 2021;22(2):209–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Drummond RA, Desai JV, Ricotta EE, et al. Long-term antibiotic exposure promotes mortality after systemic fungal infection by driving lymphocyte dysfunction and systemic escape of commensal bacteria. Cell Host Microbe 2022;30(7):1020–1033.e1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Velasco-de-Andrés M, Català C, Casadó-Llombart S, et al. The lymphocyte scavenger receptor CD5 plays a nonredundant role in fungal infection. Cell Mol Immunol 2021;18(2):498–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Dunne MR, Wagener J, Loeffler J, Doherty DG, Rogers TR. Unconventional T cells – new players in antifungal immunity. Clin Immunol 2021;227:108734. [DOI] [PubMed] [Google Scholar]
- [14].Castellano-González G, McGuire HM, Luciani F, et al. Rapidly expanded partially HLA DRB1-matched fungus-specific T cells mediate in vitro and in vivo antifungal activity. Blood Adv 2020;4(14):3443–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Tramsen L, Beck O, Schuster FR, et al. Generation and characterization of ANTI-Candida T cells as potential immunotherapy in patients with Candida infection after allogeneic hematopoietic stem-cell transplant. J Infect Dis 2007;196(3):485–92. [DOI] [PubMed] [Google Scholar]
- [16].Cenci E, Mencacci A, Bacci A, Bistoni F, Kurup VP, Romani L. T cell vaccination in mice with invasive pulmonary aspergillosis. J Immunol 2000;165(1):381–8. [DOI] [PubMed] [Google Scholar]
- [17].Kumaresan PR, Manuri PR, Albert ND, et al. Bioengineering T cells to target carbohydrate to treat opportunistic fungal infection. Proc Natl Acad Sci U S A 2014;111(29):10660–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Perruccio K, Tosti A, Burchielli E, et al. Transferring functional immune responses to pathogens after haploidentical hematopoietic transplantation. Blood 2005;106(13):4397–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Nguyen KM, Schneider JS, Farrell NF, Zenga J, Kendall PL, Roland LT. Immune receptor repertoire analyses reveal dynamics of adaptive immunity in acute invasive fungal sinusitis. Int Forum Allergy Rhinol 2025. [DOI] [PubMed] [Google Scholar]