Cerebrospinal fluid mononuclear cell profiling among patients with HIV associated cryptococcal meningitis: Immunological predictors of early mortality and 1 year survival

Abstract

Introduction: Despite the introduction of new treatment strategies for the induction phase of HIV-associated cryptococcal meningitis (CM), the 2-week mortality continues to be unacceptably high and at 1 year more than half are reported dead. The strategic framework for ending CM deaths set a goal of reducing deaths from CM by 90% by 2030. Examining host-pathogen interactions to identify cellular biomarkers in patients at high risk of death after diagnosis of CM could lay the foundation for interventions to reduce mortality from HIV-associated CM and lay the groundwork for future therapeutic targets and treatment strategies.

Objectives: To compare the cerebrospinal fluid (CSF) cellular phenotype and activation among CM patients who died within two weeks, and to describe the changes in CSF characteristics and immune profile from baseline to 1 year after CM diagnosis and treatment among survivors. Methods: This was a prospective cohort study nested within a parent study with a nested case-control study. In the case-control study, the CSF cellular immune profile associated with early mortality was determined. Thawed CSF cell pellets and supernatant were processed and analyzed to determine the immunophenotype and activation of CSF T lymphocytes (CD4+, CD8+), monocytes (CD+14), and natural killer cells (CD56+) using polychromatic flow cytometry and quantitative CSF cryptococcal antigen (CrAg) titer using CrAg lateral flow assay respectively. Statistical analysis was performed using Mann Whitney U test, logistical regression and Wilcoxon signed rank sum test where appropriate with a P value <0.05 being considered as significant. Results: Sixty CSF samples from 39 participants were used in this study. At CM diagnosis, participants who died within 2 weeks significantly had low absolute CSF CD8+ T cell and CD14+ monocyte counts, increased expression of PD-1 on all CSF mononuclear cells i.e., CD8+ and CD4+ T cells, CD14+ monocytes, and CD56+ natural killer cells at the time of CM diagnosis compared to survivors. Likewise, the CSF CD14+ monocytes among those who died expressed less HLA-DR+ and more CD163+ compared to those who survived beyond 2 weeks. There was a significant improvement in the CSF opening pressure, protein, lactate, and glucose among the CM survivors at 1 year compared to CM diagnosis. We also noted a significant reduction in the median CSF CrAg titer from 1:2560 at baseline to 1:5 at 1 year (p <0.0001). The median reduction in CSF CrAg dilutions was 8.0 (IQR= 5.0-9.0) with 8 of 21 participants (38.1%) having a negative CSF CrAg test at 1 year. Similarly, there was a significant reduction in the HLA-DR+ on CD4+ and CD8+ T lymphocytes, CD163+ expression on CD14+ monocytes, and CD57+ expression on NK cells at 1 year compared to baseline among CM survivors. Conclusion: In addition to the known clinical risk factors, a low CD8+T cell count, abundance of alternatively activated macrophages and immune exhaustion of CSF mononuclear cells may contribute to the early mortality seen among patients with CM. There is a need for adjunctive immune therapy in addition to the current standard of care among CM patients with poor CSF immunological profile. After 1 year of CM treatment, 62% of participants still had a positive CSF CrAg and the CSF mononuclear cells that had previously been activated returned to a deactivated state. More studies are needed to explore the mechanisms as to why some CM patients have their CSF CrAg convert at 1 year and others do not.