Joseph Woodward Carlson

High-grade serous ovarian cancer (HGSOC) is characterized by remarkable intratumoral heterogeneity and variable tissue architectures, dictating differential therapeutic responses. Although BRCA1/2 mutations correlate with improved responses to DNA-damaging agents, how BRCA genotype shapes the tumor immune microenvironment (TIME) remains incompletely understood. Here, we integrated whole exome sequencing, total RNA sequencing and spatial transcriptomics to profile a cohort of BRCA-mutated (BRCA-MT) and non-BRCA-mutated (BRCA-WT) HGSOC tumors. While bulk transcriptomics highlighted global phenotypic shifts, suggesting a more pro-inflammatory milieu in BRCA-MT tumors, spatially resolved analysis revealed how these differences manifest within specific tissue architectures. Integrated clustering revealed recurrent spatial transcriptional states across genotypes. Specifically, we identified a transcriptionally distinct immune-enriched region (IC7) characterized by low tumor purity, interferon signaling, and antigen-presentation programs. Reference-based RCTD deconvolution provided orthogonal computational support that IC7 was enriched for T/NK and dendritic cells relative to non-IC7 regions. At the patient level, the functional state of IC7 was highly BRCA-dependent: BRCA-MT tumors exhibited stronger enrichment of interferon-α/γ responses and inflammatory signaling than BRCA-WT tumors. Exploratory CellChat analysis further suggested genotype-associated differences in predicted immune-related signaling. We also identify a conserved structural interface between CAF-like region (IC3) and macrophage-rich region (IC4), accompanied by a predicted CXCL12–CXCR4 chemokine recruitment axis. Taken together, these findings support a model in which BRCA mutation status is associated with functional differences within broadly conserved spatial regions, rather than extensive reorganization of the overall spatial architecture. This spatially resolved framework identifies BRCA-associated immune transcriptional patterns in HGSOC and provides a basis for future mechanistic validation.

Supplementary Figure S3. IFN measurement and organ BLI of the short-term in vivo study. A. IFN quantification by ELISA in post-mortem mouse blood plasma. B. Representative BLI images of isolated organs from mice treated with PBS, NT-T, and CAR-T cells at the indicated T cell doses after 32 days. C. Organ BLI quantification. N=3 for control. N=5 for all groups, except N=2 for NT-T 1M, N=3 for CAR-T 1M, and N=4 NT-T 2M. Mean ± SD. One-way ANOVA with Tukey’s multiple comparison tests. (A) ***p=0.0005, ****p=0.0001. (C) *p=0.04, **p≤0.001 to PBS group. Not significant, n.s., differences were attributed to p≥0.05. (M=1x106).

Supplementary Figure S2. Short-term in vivo study, dose-response, and weight monitoring. A. Treatment groups and used T cell doses. B. Weight monitoring in the indicated mice groups. CAR-T 3M treated mice showed a trend of mild weight loss. (M=1x106).

Supplementary Figure S6. Expression of SSEA-4 in normal mouse tissues. Immunofluorescence detection of SSEA-4 (green) in the indicated tissues of healthy 10 weeks old (C57BL/6) mouse using anti-SSEA-4 antibody.

SSEA-4 expression in an ovarian cancer cohort.

Supplementary Figure S5. Weight monitoring during the long-term in vivo study. Mice treated with 2M CAR-T dose showed pronounced weight loss and health deterioration at earlier time points than mice treated with 1M CAR-T dose. PBS, tumor-free CAR/NT-T 1M, and NT-T groups did not show weight loss and remained healthy during the duration of the experiment. Unless otherwise indicated, mice groups are tumor-bearing. (M=1x106).

Supplementary Figure S7. T cell detection in mouse tissues after the long-term survival study. Indicated tissues were stained with HOECHST 33342 for nuclei (blue), anti-EGFR to detect CAR-T cells (middle panel; green) and anti-CD3 for T cells (right panel, white). Left panel shows a merged image of all three channels. Arrows highlight CAR-T and CD3 positive cells, while arrowheads highlight erythrocytes with an intense background staining.

Supplementary Figure S1. SSEA-4 staining specificity and LDH cytotoxicity assay. A. SSEA-4 immunohistochemistry in paraffin-embedded gastric cancer tumor microarrays. B. Upper panel sections were treated with standard xylene/ethanol series used in paraffin embedded section treatment, whereas lower panel samples were untreated. C. Consecutive sections of fresh frozen pancreatic cancer patient samples were stained with anti-SSEA-4 antibody. The staining was blocked with SSEA-4 hexaose (middle panel) or SSEA-3 (right panel). D. Representative immunofluorescence micrographs of SSEA-4 (red) in OVCAR-4 cells (high expression) and Caov-3 cells (no expression). Nucleus (blue; Hoechst 33342). N=3. E. LDH assay to measure toxicity of anti-SSEA-4 CAR-T cells on SSEA-4 positive OVCAR-4 cells and SSEA-4 negative Caov-3 cells as a control (N=2). LDH assay was performed after 72 hours of CAR/NT-T cell treatment. ****p=0.0001, ***p=0.0002 and **p=0.0025.

Supplementary Data S1. Pathology Report of the long-term survival study. Post-mortem lung, intestine and kidney tissue samples collected and analyzed by Histowiz.