Tamiwe Tomoka

SUMMARY: Introduction: Cervical cancer (CC) represents a neoplastic transformation of the cervical epithelium. CC is the third most common cancer type in Malawi. In 2018, the estimated cervical cancer mortality rate was 51.5 deaths per 100,000 women per year among Malawian women. Although recent burden association and meta-analysis studies have identified candidate cervical cancer susceptibility variants in Malawian women, they did not establish the functional significance of individual variants. Aim: This study aimed to identify germline genetic variants most likely to contribute to cervical cancer susceptibility among Malawian women. Methodology: Whole blood samples were collected from 20 newly diagnosed cervical cancer patients at the Malawi National Cancer Treatment Centre between October and December 2023, and whole genome sequencing was performed. Variants were annotated via VEP and prioritized by ClinVar pathogenic status, Condel (SIFT and PolyPhen2) and minor allele frequency (MAF ≤ 0.1) in the 1000 Genomes project and GenomAD databases. Genes containing the prioritized variants were annotated using the DAVID tool. Results: We identified five known pathogenic variants in the ROS1 gene among Malawian cervical cancer patients. We also identified 51 moderate impact variants, 64.7% of which were located in known genes associated with CC, including ADRB1, APOBEC3B, FAT1, IL12RB1, DST, EXO1 and FGFRL1 genes. Discussion and Conclusion: The presence of known pathogenic variants in the ROS1 gene, which is associated with a high recurring rate of CC, as well as the identified moderate impact variants in other genes associated with CC, provide novel insights into the genetic landscape of cervical cancer in Malawi and highlight potentially functional variants that warrant further investigation.

The tumor immune microenvironment (TIME) and tumor heterogeneity shape prognosis and therapeutic response in diffuse large B-cell lymphoma (DLBCL), but human immunodeficiency virus (HIV)–associated (HIV+) DLBCL remains poorly understood. To characterize the TIME and tumor-intrinsic features of DLBCL across a spectrum of HIV-associated immune dysregulation, we applied imaging mass cytometry (IMC), spatial transcriptomics (ST), and whole-exome sequencing to 61 pretreatment DLBCL biopsies, stratified by HIV status and antiretroviral therapy (ART) exposure. In IMC, a tumor–immune–hot neighborhood was associated with inferior survival in HIV+ DLBCL (HR = 3.16; P = 0.013). Using ST, HIV+/ART-experienced tumors demonstrated greater macrophage infiltration (P = 0.02) and enrichment for M2 macrophage markers (CD163, MRC1, and MARCO) and PDCD1:CD274 coexpression. Unsupervised clustering identified 17 shared subclonal niches, with PRC1/BMI1 dysregulation in HIV+ tumors and PRC2/EZH2 enrichment in HIV-negative (HIV−) tumors. Pseudotime positioned HIV+/ART-naïve, HIV+/ART-experienced, and HIV− tumors at progressively later transcriptional states. CD47 increased over pseudotime in HIV+/ART-experienced tumors but decreased in HIV− tumors, whereas CD58 loss occurred in HIV− tumors only. In an exploratory paired exome subset, high−mutational burden tumors transitioned from proliferative to immune-engaged states, whereas low-burden tumors accumulated extracellular matrix and myeloid signatures suggesting fibrotic microenvironments. The data define distinct immune architectures in DLBCL across a spectrum of HIV-associated immune dysregulation. HIV and ART exposure seem to shape the DLBCL TIME in a manner that may reflect the degree of immune reconstitution during lymphomagenesis. The immunosuppressive, checkpoint-enriched microenvironment of ART-experienced tumors supports further investigation of immune checkpoint blockade and macrophage-targeting strategies in a population historically excluded from such trials. Significance:This study demonstrates that the systemic host immunity in which DLBCL arises influences the tumor microenvironment, tumor evolution, and immune evasion strategies that develop, with HIV+ DLBCL exhibiting earlier developmental states and selection for immunosuppressive environments. The findings provide biological rationale for further investigation of checkpoint therapy, and macrophage inhibitors, in HIV-associated lymphomas, despite historic exclusion of this population from immunotherapeutic clinical trials.

Spatial immune characterization of DLBCL by IMC and ST. A, Representative regions of interest of IMC data of three representative tumors. HIV+/ART-naïve (T15, red), HIV+/ART-exp (T03, green), and HIV− (T33, blue). In the top three panels, each spot represents a single cell and is colored by cell type. In the bottom three panels, each spot represents a single cell and is colored by cell-type neighborhood (k = 20 and center = 6). B, Kaplan–Meier survival curve of tumor–immune–hot neighborhood presence by median cutoff [above median = “Hot” (red); below median = “Cold” (blue)] in HIV+ DLBCL (log-rank P = 0.018; Cox HR = 3.16; P = 0.013). C, Gene expression of select markers used for spot-labeling of three representative tumors. HIV+/ART-naïve (T21, red), HIV+/ART-exp (T01, green), and HIV− (T06, blue). Spots colored by dominant cell-type assignment in the final panel. D, Comparison of cell types present in tumor by spot-labeling, normalized to total number of spots per tumor, by HIV/ART status (n = 31, Wilcoxon rank-sum test). E, PCA plot of HIV+ tumors by ST identified three clusters. Each tumor is colored by tumor–immune–hot neighborhood presence assigned by IMC using median cutoff, as in B [above median = “high” (red); below median = “low” (blue)], and shaped by ART status (circle = ART-naïve; triangle = ART-exp.). Clusters are denoted by colored ellipses (cluster one = red, cluster 2 = green, and cluster 3 is one tumor, T19). cDC, conventional Dendritic Cells.

Supplemental Figure S10. Cell-of-origin classification and its relationship to HIV/ART status, cell-type composition, and spatial immune architecture.

Supplemental Figure S11. Oncogenic pathway enrichment across shared subclonal niches stratified by HIV/ART status.

Supplemental Figure S12. Pseudotime complexity by HIV/ART status.

Supplemental Figure S1. Quality control metrics of Visium spatial transcriptomic data.

Supplemental Figure S13. Exploratory analysis of tumor-intrinsic gene expression, subclonal architecture, and pseudotime trajectories by tumor mutational burden.

Supplemental Figure S14. Heatmap of the top 50 pseudotime-associated genes in TMB-H tumors.

Supplemental Figure S15. Heatmap of the top 50 pseudotime-associated genes in TMB-L tumors.