Prognostic Impact of Spatial Distribution of B-cells in High-grade Serous Ovarian Cancer (HGSOC)

Abstract

High grade serous ovarian cancer (HGSOC) is accompanied by an immune-rich but heterogenous tumor microenvironment. Although tumor associated tertiary lymphoid structures (TLS) are linked to favourable antitumor immunity, B cells may also adopt spatial distribution patterns favourable for immune suppression. This study evaluated whether the abundance and spatial distribution patterns of CD20+ B cells were associated with overall survival in HGSOC. Eight tissue microarray (TMA) slides were analyzed by chromogenic immunohistochemistry (IHC) for CD20 and CD3. Cells were segmented in QuPath using a optimised Cellpose-SAM model and further classified based on their marker intensities. Binary Moran’s Index quantified the CD20+ cell autocorrelation, while a random labelling adjusted Ripley cross-L metric summarised CD20-CD3 spatial interactions. Image derived variables were collected and aggregated on a patient basis and evaluated using Cox proportional hazard regression models. CD20 abundance was not associated with overall survival after clinical adjustment. In contrast, higher CD20+ cell clustering was associated with poorer overall survival. Each 0.1-unit increase in patient-level mean binary Moran's I corresponded to an adjusted hazard ratio of 1.26 (95% CI, 1.00-1.57; p = 0.046). Interaction of CD20-CD3 cells was not independently prognostic. These findings indicate that the spatial distribution of B cells might be more informative than their abundance alone and support future phenotypic analysis of regulatory B cells in spatial studies of ovarian cancer.

Popular Abstract

High-grade serous ovarian cancer (HGSOC) is the most common and one of the most aggressive types of ovarian cancer. It is often discovered late, after the cancer has already spread within the abdomen or to other parts of the body.
Many patients respond well at first to surgery and chemotherapy, but the cancer often comes back. Long-term survival is therefore still difficult for many patients. One reason is that HGSOC can vary a lot between patients, and even between different areas of the same tumor. These differences may affect how the cancer behaves and how the immune system reacts to it.
This study investigated whether the location and organisation of B cells in ovarian cancer tissue are linked to patient survival. B cells are immune cells, and the study asked whether it matters if they are spread out, grouped together, or found close to other immune cells.
Tumor tissue samples from patients with HGSOC were placed on special microscope slides. The samples were stained so that B cells could be identified using the marker CD20, and T cells using the marker CD3. The stained slides were then analysed with digital image software called QuPath. Several cell detection methods were tested, and the final method was chosen because it gave a good balance between accuracy and analysis time.
The study measured three main things:
how many B cells were present,
whether the B cells were grouped together or spread out,
whether B cells and T cells were located close to each other.
The results showed that metastatic tumor tissue had slightly more B cells than primary tumor tissue. However, the total number of B cells was not clearly linked to patient survival after clinical factors were taken into account. This means that simply having more B cells did not explain why some patients lived longer than others.
The most important finding was that patients whose tumors had stronger clustering of B cells had poorer overall survival. However, this does not mean that all B cell clusters are harmful. Some immune cell clusters may be helpful, but this study could not determine exactly what type of clusters were present.
The relationship between B cells and T cells varied between samples and was not clearly linked to survival after adjustment for clinical factors.
Overall, the study shows that it is not enough to simply count immune cells in tumor tissue. Where the cells are located and how they are organised may also be important. Future studies should use more detailed markers to better understand which B cell patterns are linked to helpful immune responses and which may be linked to poorer outcomes.

• author: Choudhary, Arnab
• supervisor: Anna Sandström Gerdtsson; Asger Meldgaard Frank
• organization: Department of Immunotechnology; Educational programmes, LTH
• course: KIMM01 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Keywords: High-grade serous ovarian cancer, B cells, CD20, CD3, Tertiary lymphoid structures, Moran's Index, Ripley cross-L, Tissue microarray, Digital pathology, Overall survival
• language: English
• id: 9240516
• date added to LUP: 2026-06-17 16:36:17
• date last changed: 2026-06-17 16:36:17