Infiltration of CD8+ cytotoxic T-cells and expression of PD-1 and PD-L1 in ovarian clear cell carcinoma.

This study included 18 patients with OCCC diagnosed at Peking Union Medical College Hospital (PUMCH) between January 2015 and April 2018, all of whom had qualified paired tumor and adjacent non-tumor samples. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The ethics committee of PUMCH granted approval for this study (No. JS-1626) . All women agreed to participate by providing informed consent. Patients with OCCC were included if they met the following criteria: (i) had lesions originating in the ovary; (ii) underwent comprehensive staging surgery, cytoreductive surgery or laparoscopic biopsy surgery; (iii) FIGO stage I to stage IV disease; (iv) had a clear pathological diagnosis in which pathological specimens were preserved intact; and (v) had complete follow-up data. Oncological outcomes were assessed by gynaecologic oncologists. The clinicopathological characteristics of these patients, including surgical scope, FIGO stage, treatment plan, and recurrence, were summarized in Table 1 . Progression-free survival (PFS) and overall survival (OS) were defined as from the date of pathological diagnosis of OCCC to the date of first progression or recurrence confirmed by imaging or histopathology, the date of death from OCCC or the date of death from any cause, respectively. The researchers conducted the follow-up procedure until January 2024. All pathological diagnoses were verified by pathologists. Table 1 Surgical outcomes of OCCC. Characteristics N = 18 Age (years) Mean ± SD 52.06 ± 7.07 Median (Range) 65 (35–70) Total time of operation (min) Median 200 Range 150–270 Blood loss (ml) Median 300 Range 80–1800 Residual Disease R0 15 (83.33%) R1 2 (11.11%) R2 1 (5.56%) FIGO stage I 9 (50%) II 3 (16.67%) III 4 (22.22%) IV 2 (11.11%) Postoperative Complication Yes 4 (22.22%) No 14 (77.78%) Chemotherapy Treatment Course  < 6 6 (33.33%)  ≥ 6 12 (66.67%) Recurrence Yes 9 (50%) No 9 (50%) Surgical outcomes of OCCC. Tissue blocks preserved in formalin and embedded in paraffin were acquired from the PUMCH Pathology Laboratory, and 4-mm-thick consecutive sections were prepared for later use in immunohistochemistry (IHC) and mIF. We used tyramine signal amplification (TSA) multiple fluorescence immunohistochemical staining for multicolour labelling. The immunofluorescence markers used were PDL1 (1:500, Servicebio, GB11339), PD1 (1:200, Servicebio, GB14131-50), and CD8A (1:400, Servicebio, GB15068). Two gynaecologic pathologists independently reviewed the prepared mIF slides without any prior information about the patients' clinical history. Multicolour fluorescently labelled samples could be photographed and recorded using a spectral fluorescence microscope. A professional quantitative pathological analysis software (inform V2.4) was used to identify and analyse the collected spectral images. The assessment of the combined positive score (CPS) involved analysing the presence of PD-L1 and PD1 expression. To calculate CPS, count the cells expressing PD-L1 (such as tumor cells, lymphocytes, and macrophages), divide by the total tumor cell count, and then multiply by 100. To determine the cutoff value for PD-L1 positivity, we integrated the prior experience of the Department of Pathology at our institution and consulted the FDA-approved CPS evaluation criteria for immune checkpoint inhibitors, which designates CPS ≥ 1 as the cutoff for immunotherapy in recurrent/metastatic cervical cancer. Furthermore, in the KEYNOTE-100 clinical trial on PD-L1 in ovarian cancer, CPS ≥ 1 was also used as one of the cutoff values. Consequently, CPS ≥ 1 was established as the threshold for defining PD-L1 positivity. The study utilized Cut-off Finder to analyse the receiver operating characteristic curve and determine the best threshold for CD8 + cytotoxic T-cells. Professional quantitative pathological analysis software (inform V2.4) was used to count cells in the tumor areas and the adjacent nonneoplastic areas in biopsy specimens. CD8 + cytotoxic T-cells were quantified in a high-power field at 200 × magnification, with 10 fields counted per section. Cut-off values of 100/mm2 for CD8 + cytotoxic T-cells and differences in CD8 + NTILs and CD8 + TILs of 70 cells/mm 2 were used to classify the patients into low- and high-expression groups. Women attended follow-up visits every 3 months for the initial 2 years, every 6 months for the next 3 years, and annually thereafter. Physical and gynaecological examinations, abdominal (including both kidneys) and pelvic ultrasound examinations and serum tumor marker data were collected during each follow-up visit. If tumor recurrence was suspected due to clinical observations or imaging results, a positron emission tomography CT (PET-CT) scan was performed to investigate the extent of disease. Gynecologic oncologists confirm disease recurrence by finding lesions on a biopsy or by a positive PET-CT scan. Categorical variables were assessed utilizing Pearson's chi-square test or Fisher's exact test based on the expected values. Continuous variables were compared and presented as medians and interquartile ranges (IQRs) and the mean ± standard deviation (SD). The Kaplan–Meier method and log-rank test were performed to analyse survival and evaluate differences in PFS, respectively. Hazard ratios (HRs) with 95% confidence intervals (CIs) were also calculated. P < 0.05 was considered statistical significance. R software version 3.5.3 (The R Foundation for Statistical Computing, Vienna, Austria) and SPSS 26.0 (IBM SPSS Statistics, Armonk, NY: IBM Corp) were used in this study.

The clinicopathologic characteristics and surgical outcomes of the patients are shown in Table 1 . Eighteen OCCC patients were included, with a median age of 65 years (range 35 to 70 years). According to the FIGO staging, 3 patients (16.67%) had stage I/II disease, and 6 patients (33.33%) had stage III/IV disease. Postoperative residual disease was found in 3 patients (16.67%). Median surgical blood loss was 300 ml (range: 80 to 1800 ml). The median duration of surgery was 200 min (range: 150 to 270 min). Four patients (22.22%) experienced postoperative complications (including poor incision healing, incisional infection, and lower extremity deep vein thrombosis). Biopsies were subjected to mIF and analysed via professional quantitative pathological analysis as described above (Fig.  1 ). As shown in Table 2 , the percentages of PDL1-positive tumor cells and tumor stromal cells were 44.44% and 61.1%, respectively. PD1 expression in lymphocytes was positive in 12 patients (66.67%). There was no significant difference in the expression of PD-L1 or PD1 in OCCC tissues according to tumor size, tumor location, disease stage, CA-125 level or recurrence. Fig. 1 Morphological observation and immunohistochemistry and multiplex immunofluorescence staining for PD-L1, PD-1, and CD8 in OCCC specimens (200 × magnification). ( A ): OCCC morphological observation. ( B ): CD8A expression in OCCC tumor cells and immune cells. ( C ): PD-L1 expression in OCCC tumor cells and immune cells. ( D ): PD-1 expression in OCCC tumor cells and immune cells. ( E ): Multiplex immunofluorescence staining for the indicated immune cell markers in OCCC tumor tissue. PD-L1: purple; PD-1: orange; CD8A: green; and DAPI: blue. (OCCC: ovarian clear cell carcinoma; PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1). Table 2 Relationship between tumor outcomes and PD‐L1 and PD‐1 expression in OCCC. Characteristics N PD‐L1 expression PD‐1 expression Cancerous cell ( +) P -value Stromal cells ( +) P -value Lymphocyte ( +) P -value Total 18 8 (44.44%) 11 (61.1%) 12 (66.67%) Tumor size (cm)  < 10 9 3 (33.33%) 0.319 5 (55.56%) 0.500 6 (66.67%) 0.500  ≥ 10 9 5 (55.56%) 6 (66.67%) 5 (55.56%) Ascites Yes 6 3 (50%) 0.563 3 (50%) 0.428 4 (66.67%) 0.694 No 12 5 (41.67%) 8 (66.67%) 8 (66.67%) Tumor site Unilateral 16 6 (37.5%) 0.183 9 (56.25%) 0.359 10 (62.5%) 0.431 Bilateral 2 2 (100%) 2 (100%) 2 (100%) CA-125 level Negative 8 4 (50%) 0.520 5 (62.5%) 0.648 5 (62.5%) 0.563 Positive 10 4 (40%) 6 (60%) 7 (70%) FIGO stage I, II 12 5 (41.67%) 0.563 7 (58.33%) 0.572 8 (66.67%) 0.706 III, IV 6 3 (50%) 4 (66.67%) 4 (66.67%) Recurrence Yes 9 3 (33.33%) 0.319 5 (55.56%) 0.500 5 (55.56%) 0.310 No 9 5 (55.56%) 6 (66.67%) 7 (77.78%) Morphological observation and immunohistochemistry and multiplex immunofluorescence staining for PD-L1, PD-1, and CD8 in OCCC specimens (200 × magnification). ( A ): OCCC morphological observation. ( B ): CD8A expression in OCCC tumor cells and immune cells. ( C ): PD-L1 expression in OCCC tumor cells and immune cells. ( D ): PD-1 expression in OCCC tumor cells and immune cells. ( E ): Multiplex immunofluorescence staining for the indicated immune cell markers in OCCC tumor tissue. PD-L1: purple; PD-1: orange; CD8A: green; and DAPI: blue. (OCCC: ovarian clear cell carcinoma; PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1). Relationship between tumor outcomes and PD‐L1 and PD‐1 expression in OCCC. The mean follow-up time was 60.41 months (range, 8.83–85.1 months). Nine women experienced recurrence, for a recurrence rate of 50%. One (5.56%) person died due to widespread disease. Our study did not reach the endpoint of OS, so PFS was used for the survival analysis. The relationship between tumor outcomes and PD-L1 and PD-1 expression in OCCC patients was shown in Fig.  2 . The expressions of PDL1 and PD1 in tumor cells and tumor stromal cells were not significantly correlated with PFS. Fig. 2 The relationship between PFS and the expression of PD-L1 and PD-1 in tumours ( A ): The relationship between PFS and the expression of PD-1 in tumor tissues. ( B ): The relationship between PFS and the expression of PD-L1 in tumor tissues. ( C ): The relationship between PFS and the expression of PD-1 in stromal cells. ( D ): The relationship between PFS and the expression of PD-L1 in stromal cells. ( E ): The relationship between PFS and the expression of PD-1 in tumor cells. ( F ): The relationship between PFS and the expression of PD-L1 in tumor cells. (PFS: progression-free survival; PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1). The relationship between PFS and the expression of PD-L1 and PD-1 in tumours ( A ): The relationship between PFS and the expression of PD-1 in tumor tissues. ( B ): The relationship between PFS and the expression of PD-L1 in tumor tissues. ( C ): The relationship between PFS and the expression of PD-1 in stromal cells. ( D ): The relationship between PFS and the expression of PD-L1 in stromal cells. ( E ): The relationship between PFS and the expression of PD-1 in tumor cells. ( F ): The relationship between PFS and the expression of PD-L1 in tumor cells. (PFS: progression-free survival; PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1). We quantified the density of CD8 + cytotoxic T-cells in tumor tissues and adjacent nonneoplastic regions (summarized in Table 3 ). CD8 + cytotoxic T-cells density (including tumor-infiltrating lymphocytes (TILs) and non-tumor-infiltrating lymphocytes (NTILs)) in the tumor tissue of the recurrence group was greater than that in the non-recurrence group ( p  =  0.047 ). There was no significant difference in the expression of CD8 + T-cells in NTILs or TILs between the recurrence and non-recurrence groups ( p  =  0.415 vs p  =  0.347 ). The density of CD8 + T-cells among the total cells and stromal cells in the recurrent patients was higher than in the non-recurrent patients in the adjacent nonneoplastic regions ( p  =  0.015 vs p  =  0.008 ). In the epithelial cells of the adjacent nonneoplastic regions, there was no difference in the expression of CD8 + cytotoxic T-cell between the recurrence patients and the non-recurrence patients ( p  =  0.104 ). Table 3 Summary of CD8 + T cell density measurements. CD8 + T Cell density Total (N = 18) Mean ± STD, cells/mm2 No Recurrence (N = 9) Mean ± STD, cells/mm2 Recurrence (N = 9) Mean ± STD, cells/mm2 p-value total cells of tumor tissue 96.61 ± 101.48 62.67 ± 78.06 130.33 ± 115.30 0.047* non-tumor-infiltrating lymphocytes 119.71 ± 118.99 79.51 ± 94.79 132.17 ± 44.06 0.415 tumor-infiltrating lymphocytes 64.86 ± 87.61 38.88 ± 12.96 115.86 ± 38.62 0.347 total cells of adjacent non-neoplastic regions 269.78 ± 368.27 54.98 ± 18.33 471.95 ± 157.32 0.015* stromal cells of adjacent non-neoplastic regions 217.26 ± 350.85 33.53 ± 11.18 460.87 ± 153.62 0.008* epithelial cells of adjacent non-neoplastic regions 627.67 ± 551.96 275.47 ± 91.82 665.15 ± 221.72 0.104 * P < 0.05. Summary of CD8 + T cell density measurements. * P < 0.05. The cutoff value for CD8 + T cell density was set at 100 cells/mm 2 , dividing patients into low expression and high expression groups. According to the Kaplan–Meier survival analysis, Fig.  3 A illustrated the PFS difference between the high-expression and low-expression groups of all CD8 + T cells (including TILs and NTILs) in tumor tissues (p = 0.042). The expression of CD8 + TILs and CD8 + NTILs did not correlate with the PFS (p = 0. 32; p = 0.15; respectively) (shown in Fig.  3 B and C). We analysed the relationship between the difference in the number and percentage of patients with CD8 + TILs and CD8 + NTILs in tumor tissues and patient prognosis. The ratio of CD8 + TILs to CD8 + NTIL cell density was not correlate with PFS (p = 0.61) (shown in Fig.  3 D). In Fig.  3 E, the difference in cell density between CD8 + NTILs and CD8 + TILs exceeding 70 cells/mm 2 was associated with poorer PFS (p = 0.042). Fig. 3 The relationship between PFS and the number of CD8 + T-cells. ( A ): Relationships between PFS and the expression of CD8 + NTILs and CD8 + TILs. ( B ): The relationship between PFS and the number of CD8 + NTILs. ( C ): The relationship between PFS and the number of CD8 + TILs. ( D ): The relationship between PFS and the ratio of CD8 + TILs to CD8 + NTIL density. ( E ): The relationship between PFS and the difference in CD8 + NTIL to CD8 + TIL density. ( F ): The relationship between PFS and the expression of CD8 + T-cells in adjacent nonneoplastic regions. ( G ): The relationship between PFS and the expression of CD8 + T-cells in epithelial cells of adjacent nonneoplastic regions. ( H ): The relationship between PFS and the expression of CD8 + T-cells in stromal cells of adjacent nonneoplastic regions. (PFS: progression-free survival; non-tumor-infiltrating lymphocytes: NTILs; tumor-infiltrating lymphocytes: TILs). The relationship between PFS and the number of CD8 + T-cells. ( A ): Relationships between PFS and the expression of CD8 + NTILs and CD8 + TILs. ( B ): The relationship between PFS and the number of CD8 + NTILs. ( C ): The relationship between PFS and the number of CD8 + TILs. ( D ): The relationship between PFS and the ratio of CD8 + TILs to CD8 + NTIL density. ( E ): The relationship between PFS and the difference in CD8 + NTIL to CD8 + TIL density. ( F ): The relationship between PFS and the expression of CD8 + T-cells in adjacent nonneoplastic regions. ( G ): The relationship between PFS and the expression of CD8 + T-cells in epithelial cells of adjacent nonneoplastic regions. ( H ): The relationship between PFS and the expression of CD8 + T-cells in stromal cells of adjacent nonneoplastic regions. (PFS: progression-free survival; non-tumor-infiltrating lymphocytes: NTILs; tumor-infiltrating lymphocytes: TILs). In the adjacent nonneoplastic regions, PFS was significantly lower in patients with high expression of CD8 + cytotoxic T-cells in both total and the stromal cells than in patients with low CD8 + cytotoxic T-cells (p = 0.012; p = 0.007; respectively) (shown in Fig.  3 F and H). The PFS was not correlate with the expression of CD8 + T-cells in the epithelial cells of the adjacent nonneoplastic region (p = 0.37). (shown in Fig.  3 G). According to the above survival analysis, we further investigated the differences in the immune-related microenvironment between tumor tissue and adjacent non- tumor tissue. Significant difference was not shown in the expression of PD1 or PD-L1 between adjacent nonneoplastic regions and tumor samples (p = 0.122 and p = 0.297) (Fig.  4 A and B). There were more CD8 + cytotoxic T-cells in the adjacent nonneoplastic tissues than in the tumour tissue samples (p < 0.001) (Fig.  4 C). The difference of CD8 + cytotoxic T cells (adjacent non-neoplastic minus tumor area) in the recurrent group was slightly higher than that in the non-recurrent group (p = 0.078) (Fig.  4 D). Fig. 4 Differential immune infiltration between tumours and adjacent nonneoplastic tissues. ( A ): Differences in PD-1 expression between tumours and adjacent nonneoplastic tissues ( p  =  0.122 ). ( B ): The difference in PD-L1 expression between tumours and adjacent nonneoplastic tissues ( p  =  0.297 ). ( C ): The difference in the percentage of CD8 + T-cells between the tumor and adjacent nonneoplastic tissues ( p  <  0.001 ). ( D ): The relationship between the difference in the expression of CD8 + T-cells (adjacent nonneoplastic area minus the tumour area) and recurrence ( p  =  0.078 ). (PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1). Differential immune infiltration between tumours and adjacent nonneoplastic tissues. ( A ): Differences in PD-1 expression between tumours and adjacent nonneoplastic tissues ( p  =  0.122 ). ( B ): The difference in PD-L1 expression between tumours and adjacent nonneoplastic tissues ( p  =  0.297 ). ( C ): The difference in the percentage of CD8 + T-cells between the tumor and adjacent nonneoplastic tissues ( p  <  0.001 ). ( D ): The relationship between the difference in the expression of CD8 + T-cells (adjacent nonneoplastic area minus the tumour area) and recurrence ( p  =  0.078 ). (PD-L1: programmed cell death ligand 1; PD-1: programmed cell death 1).

In conclusion, we found that CD8 + cytotoxic T-cells have significant potential in the prediction of the prognosis of OCCC patients, which lays a foundation for the development of biomarkers for immune checkpoint blockade treatment response in OCCC patients. Considering the prolonged disease progression and complex immune microenvironment of ovarian clear cell carcinoma, simply evaluating CD8 + cytotoxic s in the entire tumor tissue may not be sufficient to accurately assess the risk of tumor recurrence. This suggests that we need to expand the sample size and use RNA-seq analysis to further study the differences in immune cells and pathway enrichment analyses between tumor tissues and adjacent nonneoplastic regions in the future.

OCCC exhibits resistance to chemotherapy and lacks effective treatment options for advanced and recurrent disease 26 . With the in-depth study of the molecular characteristics and tumor microenvironment of OCCC, a variety of possible effective molecular targeted inhibitors have been identified, and related clinical trials have been carried out 27 . Immune checkpoint inhibitors such as PD-1/PD-L1 pathway blockers have been widely used in the treatment of various type of cancers 28 , 29 . Among them, the PD-1/PD-L1 pathway blockers has evolved swiftly in the therapy of non-small cell lung cancer (NSCLC) 30 . However, only a portion of patients develop an immune response to antiPD-1/PD-L1 therapy, and identifying patients who are eligible for PD1/PDL1 therapeutic intervention continues to pose a significant challenge 31 . Recent released results of a phase II clinical trial of pembrolizumab for recurrent OCCC (known as the PEACOCC study; NCT03425565 ) have brought hope for immunotherapy in OCCC. The study enrolled 48 patients with advanced clear cell gynaecological cancer in the UK, 85% of whom had OCCC. The 12-week PFS rate was 44%, and the objective response rate (ORR) was only 25% 32 . Additionally, MOCCA/APGOT-OV2/GCGS-OV3 study compared the efficacy of durvalumab and chemotherapy in 47 patients with recurrent OCCC, without significant difference in clinical benefit rate (37.5% vs 32,1%), ORR (10.7% vs. 18.8%) and PFS (7.4 vs. 14.0 weeks) between the two regimens 33 . Hamanishi et al. found that PD-L1 present on tumor cells directly inhibited anti-tumor CD8 + T cells, so the prognosis of patients with high PD-L1 expression was significantly worse than those with low expression 34 . Numerous studies have demonstrated that PD-L1 is expressed on various immune cells within the TME, including macrophages, dendritic cells, and regulatory T cells. This expression plays a critical role in regulating tumor immunity and influencing tumor progression 35 . In contrast to the findings of previous studies, the percentages of PDL1- and PD1-positive tumor cells and stromal cells were not significantly associated with PFS in our study. It is necessary to evaluate the relationship between the expression of PD1 or PD-L1 and patient prognosis further. The discrepancies observed in the expression of PD-1/PD-L1-positive tumor cells and PFS across different studies may be attributed to variations in the type of biological material analysed, as well as differences in the antibodies and platforms employed for the detection of PD-1 and PD-L1 36 , 37 . We tested whether the density of CD8 + cytotoxic T-cells was a reliable prognostic biomarker for the survival of OCCC patients. Niu D et al. demonstrated that the abnormal expression profile of CD8 + cytotoxic T-cells in cancer tissues was closely related to the clinical characteristics and disease phenotype of patients 38 . Deng M et al. reported that higher counts of CD8 + TILs and CD8 + NILs are correlated with poorer OS and PFS in intrahepatic cholangiocarcinoma patients 39 . A previous study reported that higher levels of CD8 + TILs were correlated with shorter PFS and OS in OCCC patients 40 . In our study, the PFS of the high CD8 + expression group in tumor tissue was significantly lower than that of the low CD8 + expression group. Interestingly, a difference in cell density between CD8 + NILs and CD8 + TILs greater than 70 cells/mm2 was associated with poor PFS. All OCCC patients had a certain degree of peritumoral inflammation, with a large number of inflammatory cells infiltrating the interstitium, mainly including lymphocytes and macrophages located in the matrix and luminal space adjacent to tumor cells 19 . Although there might be a large number of immune cells around tumor cells, they were unable to penetrate into the core of tumor cells to exert immune killing effects but only gathered in the surrounding matrix of tumor cells. It is also known as an immune-exclude tumor. The immunosuppressive environment of OCCC enables tumours to achieve immune escape, subsequently leading to the proliferation, invasion, and metastasis of OCCC. Patients with OCCC are often reported to have a history of endometriosis, which has prompted speculation that OCCC may originate from this disease 41 . It can be hypothesized that OCCC carcinogenesis after endometriosis may initiate in an immune-infiltrating environment or that these immune cells may exist as immune stalemates 42 . A integrated analysis of the transcriptomes of tumor tissues and tumor-adjacent normal tissues revealed that tumor-adjacent tissues were a unique intermediate state between healthy tissues and tumors, which could propagate proinflammatory signals to their surroundings, which may be critical for tumorigenesis and/or tumor progression 43 , 44 . The increased density of CD8 + cytotoxic T-cells in adjacent tissues may contribute to the regulation of the TME, leading to tumor initiation 45 . Consistent with previous published studies, we found that high CD8 + cytotoxic T-cell density in adjacent tissues was associated with poor prognosis. Therefore, we speculated that immune cell infiltration in adjacent tumour tissues might contribute to the progression of OCCC. Unfortunately, our results showed that there was only a slight tendency in CD8 + differential infiltration between the recurrent group and the nonrecurrent group. In the future, we will expand the sample size to further study the relationship between the differential infiltration of immune cells between paired tumours and adjacent nontumor areas and patient prognosis. The concurrent use of RNA sequencing, flow cytometry, and other detection methods alongside IHC can provide valuable insights for future research and clinical practice. Zheng et al. characterized the TME in intrahepatic cholangiocarcinoma using RNA sequencing and multiple immunofluorescence analyses, further validating the role of specific cell subsets through flow cytometry 46 . Similarly, Peng et al. combined multiple immunofluorescence analysis with single-cell RNA sequencing (scRNA-seq) in NSCLC to investigate the spatial architecture and complex interactions between immune cells and tumor cells 47 . These approaches not only enhance the identification and validation of distinct cell types but also enable the exploration of specific targets and the elucidation of potential mechanisms from a multi-omics perspective, including transcriptional data. However, relevant studies in ovarian cancer, particularly OCCC, remain limited. In previous studies, the distinctive germline alterations and immune infiltration characteristics of OCCC have been elucidated through next-generation sequencing and transcriptomic profiling 48 , 49 . Specific transcriptional features, such as APOBEC3 and SOX6, have been further investigated in relation to the immune microenvironment, including TILs, as well as the prognosis of ovarian cancer 50 , 51 . Additionally, single-cell mRNA sequencing and T cell receptor sequencing have revealed that ARID1A mutations are associated with the enrichment of neoantigen-reactive CXCL13 + CTLA4 + CD8 + T cells. Inhibition of VEGF was found to remodel the OCCC stromal environment, restore T cell function, and enhance the effectiveness of immunotherapy 52 . However, these studies have been limited by small sample sizes. Therefore, expanding the cohort and employing advanced methodologies are critical for further investigating the differences in immune cell dynamics and pathway enrichment between tumor and adjacent non-tumor tissues. Previous studies have provided valuable insights into the immune landscape and prognostic implications of OCCC 53 , establishing a foundation for the development of immunotherapies for OCCC and identifying immune subtypes that influence prognosis. Concurrently, the NRG Oncology study GY016 evaluated the combination of pembrolizumab and the IDO-1 inhibitor epalrestat in patients with recurrent OCCC, yielding promising results 54 . Despite these advances, a critical clinical challenge persists: the development of effective methods for assessing the prognosis of OCCC patients and formulating tailored immunotherapy strategies. This challenge remains a central focus of ongoing research. This study innovatively revealed that the density of CD8 + cells were worthy of attention in the prognosis of women with OCCC. But our study also had several limitations. Initially, our investigation was directed towards elucidating the correlation between CD8 + cytotoxic T lymphocytes and the prognostic outcome in patients suffering from OCCC, excluding any evaluation of the functional significance of alternative immune cell populations. Secondly, we only evaluated the expression of CD8 + T cells, PD-1, and PD-L1 via IHC, and the reason why increased CD8 + T cell density leads to poor prognosis was unclear. Thirdly, our study was confined to a limited number of cases, and the endpoint for OS was not reached. Finally, the relationship between the dynamic changes of CD8 + T cells during immunotherapy and prognosis was unclear, and future research will require the use of single-cell sequencing and dynamic monitoring for further investigation. OCCC is a rare subtype of EOC that is related to endometriosis. Due to inherent chemotherapy resistance, the prognosis of patients with advanced and recurrent OCCC is poor. Developing more efficient treatment approaches for OCCC is imperative and urgent. Because OCCC is relatively rare and the number of cases is limited, international and domestic multicentre cooperation is needed to carry out clinical research and basic research to develop effective strategies for the treatment of OCCC.

Ovarian clear cell carcinoma (OCCC) represents a relatively uncommon histological subtype of epithelial ovarian cancer (EOC) 1 , 2 , exhibiting variable incidence rates across different ethnic populations 3 . OCCC represents 3–12% of all EOC cases in European and North American 4 , 5 and 25% of EOC cases in East Asia 6 . These differences between countries are partly related to the incidence of endometriosis, a recognized precursor of OCCC and endometrioid ovarian cancer 7 , 8 . The prognosis of early OCCC is good, but the sensitivity of advanced OCCC to first-line chemotherapy based on platinum is low. Compared with other type of EOC in the same period, the prognosis is worse. Melphalan, a bifunctional alkylating agent, has demonstrated notable efficacy in patients with platinum-resistant EOC harbouring BRCA mutations. However, research on the use of alkylating agents in OCCC remains limited 9 , 10 . Despite the unclear mechanisms of carcinogenesis and chemoresistance in OCCC, its various genetic alterations have been extensively studied. The mutation frequencies of AT-rich interaction domain 1A ( ARID1A ) and phosphatidylinositol-4,5-diphosphate 3-kinase catalytic subunit alpha ( PIK3CA ) are increased in OCCC 11 , resulting in an abnormal cell cycle and loss of proliferation control 12 . PIK3CA is recognized as the second most frequently mutated gene in human cancers. Literature reports indicate that PIK3CA mutations are present in 33% to 55% of OCCC patients 13 – 15 . Recent studies have further established an association between PIK3CA mutations and OCCC survival outcomes 16 , with emerging evidence suggesting that these mutations may also be linked to distinct immune characteristics 17 . Therefore, further research should focus on uncovering new chemoresistance tumor markers and developing new therapeutic targets for OCCC. Tumor-induced immunosuppression is a crucial issue because it not only promotes tumor progression but also impacts the efficiency of anticancer treatments. PD-L1 is a transmembrane protein and the major ligand for programmed cell death, which may be an effective mechanism to prevent tumors from escaping the immune system 18 . PD1 plays an essential role in regulating the immune system response and inhibiting T cell inflammatory activity to regulate the immune system and promote self-tolerance 19 . The PD-1/PD-L1 pathway inhibits the function of T lymphocytes within the tumor microenvironment (TME), reducing the immune killing capacity of tumors and facilitating immune evasion 20 – 23 . A pan-cancer study showed that high expression of CD8 + T-cells and PD-L1 are satisfactory indicators for predicting the efficacy of anti-PD-1/PD-L1 therapy 24 . However, data regarding the role of PD-L1, PD1 and CD8 + T-cells expression in the prognosis of OCCC are relatively limited. Although preliminary studies have investigated the immunohistochemical expression of PD-L1 in OCCC and its correlation with clinicopathological characteristics 18 , 25 , no significant prognostic predictive value has been established. Therefore, further research that integrates PD-1, CD8 + , and other markers, alongside more precise partitioning of tumor tissues and accurate identification of cell subtypes, may enhance predictive accuracy and offer greater clinical utility. The purpose of this study was to analyse the expression of PD-L1, PD1, and CD8 + T cells in OCCC patients by multiple immunohistochemical quantitative methods and to explore the relationship between their expression patterns and the clinical characteristics and prognosis of OCCC patients.