Deciphering the stromal molecular landscape: the correlation between p16 and α-SMA in epithelial ovarian cancer.

This retrospective study was conducted on 64 patients with high-grade serous ovarian cancer, clear-cell ovarian cancer or endometrioid ovarian cancer who underwent primary debulking surgery (PDS) at the Departments of Gynecology and Obstetrics of the University of Pisa between January 2019 and June 2021. The hospital records, including surgical notes, pathological reports and follow-up data, were collected using a common form with standardized items and a common database. The choice for a PDS approach was individually established on the basis of an accurate evaluation of both the spread of disease at clinical, radiological [chest-abdomen-pelvic computed tomography (CT) scan] and, sometimes, surgical examination and the patient general conditions after an exhaustive discussion with the patient by a multidisciplinary team, as reported in a previous paper (Gadducci et al. 2017 ). The tumor stage and histological diagnosis of each patient were determined according to the FIGO criteria and the histological typing system of the World Health Organization (WHO), respectively. The baseline characteristics (age, FIGO stage, histological type, tumor grade, peritoneal spread, lymph node metastases, peritoneal cytology, and residual disease after PDS) were reported for each patient. Before starting this work, the entire study was approved by the Regional Ethical Committee for the Clinical Trials in Tuscany (Protocol no 19735). All patients enrolled in the study provided written informed consent. The Formalin-fixed paraffin-embedded (FFPE) tissue samples of primary EOCs were retrieved from the archives of the Division of Pathology, Department of Translational Research and New Technologies in Medicine and Surgery, University of Pisa. All pathological specimens collected during PDS were analyzed by two expert gynecological pathologists (S.P. and A.G.N.) following the guidelines of the WHO International Classification of Ovarian Tumors (McCluggage et al. 2020 ). Stage was classified according to FIGO recommendations and was updated and revised during data collection. The key inclusion criteria were EOC diagnosed in chemo-naive patients, who were > 18 years old, who were candidates for PDS, who had no history of prior chemotherapy for other diseases or prior radiation therapy to the pelvis. All immunostainings were performed on FFPE samples from all patients using the Ventana Medical System and specific antibodies. After initial deparaffinization, 0.3% hydrogen peroxide was used to block endogenous peroxidase activity. The sections were microwaved in 10 mM citrate buffer (pH 6.0) or Tris–EDTA to unmask the epitopes. Then, the slides were incubated with the following primary antibodies according to optimized protocols: monoclonal mouse α-smooth muscle actin (α-SMA) protein (clone 1A4, α-SMA, Ventana Medical System) and monoclonal mouse anti-p16 (CINtec ® histology V-Kit, clone E6H4, Roche Tissue Diagnostics). The peroxidase- labeled polymer ADVANCE ™ HRP Detection System (Dako) was applied for 30 min at room temperature. DAB chromogen substrate (3-diaminobenzidine, Sigma-Aldrich, St. Louis, MO, USA) was added, at a ratio of 0.06 g to 100 mL of PBS, 500 μL of 3% hydrogen peroxide and 1 mL of dimethyl sulfoxide at 37 °C for 5 min. The slides were washed in water, counterstained in haematoxylin, dehydrated and mounted. The positive controls for each antibody were tissue samples with adequate immunoreactivity. Negative controls were produced by omitting of the primary antibodies. The same two gynecological pathologists (S.P. and A.G.N.) with expertise in EOC, who were blinded to the clinical and pathological data, scored the samples. Protein expression of p16 (CINtec ® histology V-Kit, clone E6H4; Roche Tissue Diagnostics) and α-SMA (clone 1A4; α-SMA; Ventana Medical System) was assessed via immunohistochemistry using standard techniques. The expression of p16 and α-SMA was divided into a positive and negative staining (defined by complete absence of signal). Somatic BRCA1/2 mutations were detected in FFPE carcinoma samples, and the most appropriate material, containing almost 50% of the tumor cells, was chosen with high accuracy. DNA extraction from tumour tissue was performed using a DNA Sample Preparation Kit according to the manufacturer's protocol. The mutational analysis was performed using the next-generation sequencing (NGS) technique and the ILLUMINA HPS- panel (San Diego, California). The expression of p16 and α-SMA was considered a dichotomous variable. The statistical analysis was performed by the Spearman's Rank-Order Correlation (ρ), and p-values of 0.05 or less were considered to indicate statistical significance. Statistical analyses were conducted using ‘R’ Software version 4.1.1.

The baseline characteristics of the patients are shown in Table  1 . The median age of the patients was 60 years (range, 37–87 years). Most patients (n.45, 70.31%) were in FIGO stage IIb-IV at presentation EOCs. Table 1 Baseline characteristics of patients N0 pts % Age    50 years 51 79,69 Histotype  HGSC 54 84,38  HGEC 5 7,81  CCC 5 7,81 Stage  I 18 28,13  IIb 6 9,38  III 33 51,56  IV 6 9,38  Unknown 1 1,56 BRCA  Wild-type 40 62,50  m-BRCA 1 13 20,31  m-BRCA 2 10 15,63  Unknown 1 1,56 HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer Baseline characteristics of patients HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer The mutational BRCA status was reported for 63 women. The wild-type BRCA1 gene was found in 40 patients (62,50%), whereas the mutated BRCA-1 gene and mutated BRCA-2 gene were detected in 13 (20,31%) and 10 (15,63%) patients, respectively. With respect to histological type, the BRCA1 and BRCA2 genes were mutated in 12 (22,22%) and 7 (12,96%) of the 54 women with high-grade serous ovarian cancer, respectively. Of the 5 patients with endometrioid ovarian cancer, 2 (40.00%) had a mutated BRCA2 gene, and none had a mutated BRCA1 gene. Instead, 5 patients were diagnosed with clear cell ovarian cancer: one woman had a mutated BRCA1 gene (20.00%), and the other had a mutated BRCA2 gene (20.00%). The pathologists described the histologic and architectural patterns of all the all samples: SET features (‘Solid, pseudo-Endometrioid and/or Transitional cell carcinoma-like’) were detected in 8 samples (14,80%) of high grade serous ovarian cancer, and 7 of those had BRCA germline and somatic mutations (87,50%). However, in the group of endometrioid ovarian cancers harboring BRCA germline or somatic mutations, 63,16% of the patients had a classic architectural pattern, and 36,84% had SET features. Table 2 reports the stromal expression of p16 and α-SMA in EOCs. Positive p16 immunostaining was found in 7 out of 64 patients (10,94%), and according to histological type, it was found in 5 of 54 (9,25%) patients with high-grade serous ovarian cancer and 2 of 10 patients (20,0%) with endometrioid or clear cell ovarian cancer. p16 stromal expression was positive in 6 out of 38 (15,78%) patients with wild-type BRCA and in only 1 of the 22 (4,5%) patients with mutated BRCA. Table 2 The stromal expression of p16 and α-SMA in EOC and its correlation with BRCA mutational status Histotype p16 stromal expression α-SMA stromal expression Negative,n(%) Positive,n(%) Unknown,n(%) Negative,n(%) Positive,n(%) Unknown,n(%) HGSC 45 (83,30%) 5 (9,25%) 4 (7,40%) 4(7,40%) 46 (85,20%) 4 (7,40%) wt-BRCA 28 (87,50%) 4 (12,50%) – 4 (12,50%) 28 (87,50%) – m-BRCA 17 (94,44%) 1 (5,56%) – 0 18 (100,00%) – HGEC&CCC 8 (80,00%) 2 (20,00%) – 1 (10,00) 9 (90,00) – wt-BRCA 4 (66.66%) 2 (33,34%) – 0 6 (100,00%) – m-BRCA 4 (100,00%) 0 – 1 (25,00%) 3 (75,00%) HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer, wt-BRCA wild-type BRCA, m-BRCA mutated BRCA The stromal expression of p16 and α-SMA in EOC and its correlation with BRCA mutational status HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer, wt-BRCA wild-type BRCA, m-BRCA mutated BRCA Immunostaining for α-SMA was positive in 55 out of 64 (85,90%) patients and by histological type in 46 of 54 (85,20%) (Table  3 ) patients with high-grade serous ovarian cancer and 9 of 10 patients (90,00%) with endometrioid or clear cell ovarian cancer (Fig.  1 ). According to the BRCA mutational status, positive α-SMA expression was detected in 34 of 38 patients with wild-type BRCA (89,47%) and in 21 of 22 patients (95,45%) with mutated BRCA. Table 3 The stromal expression of p16 and α-SMA in EOC and its correlation with BRCA mutational status and stage HGSC, HGEC, CCC p16 stromal expression α-SMA stromal expression Negative, n (%) Positive, n (%) Negative, n (%) Positive, n (%) Stage I-II 20 (83,30%) 4 (16,70%) 1 (4,17%) 23 (95,83%) wt-BRCA 9 (75,00%) 3 (25,00%) 1 (8,33%) 11 (91,67%) m-BRCA 11 (91,70%) 1 (8,30%) 0 12 (100%) Stage III-IV 33 (97,10%) 3 (2,90%) 4 (10,81%) 33 (89,19%) wt-BRCA 23 (88,40%) 3 (11,60%) 3 (11,54%) 23 (88,46%) m-BRCA 10 (100,00%) 0 1 (10,00%) 9 (90,00%) HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer, wt-BRCA wild-type BRCA, m-BRCA mutated BRCA Fig. 1 Immunohistochemistry was performed using standard techniques to evaluate the expression of α-SMA. A A 10 × representation of positive stromal tumor α-SMA. B A 20 × image of stromal α-SMA expression The stromal expression of p16 and α-SMA in EOC and its correlation with BRCA mutational status and stage HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer, wt-BRCA wild-type BRCA, m-BRCA mutated BRCA Immunohistochemistry was performed using standard techniques to evaluate the expression of α-SMA. A A 10 × representation of positive stromal tumor α-SMA. B A 20 × image of stromal α-SMA expression As shown in Table  3 , immunostaining for α-SMA was positive in 95.83% of patients with stage I-II disease and 89.19% of those with stage III-IV disease. Positive p16 stromal expression was detected in 3 out of 12 (25,00%) patients with wild-type BRCA in stage I-II disease and in 11,60% of wild-type BRCA patients affected by advanced disease. The statistical analysis carried out by the nonparametric test Spearman's rank-order correlation showed a monotonic relationship between the stromal expression of α-SMA and that of p16 (r = −0,32, p = 0,023) (Table 4 ) (Supplementary Fig. 1). Table 4 The correlation between the stromal expression of α-SMA and p16 in EOCs HGSC, HGEC, CCC α-SMA stromal expression negative α-SMA stromal expression positive No % No % p16 stromal expression negative 4 6,66 53 88,33 p16 stromal expression positive 1 1,67 3 5,00 HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer The correlation between the stromal expression of α-SMA and p16 in EOCs HGSC high-grade serous cancer, HGEC high-grade endometrioid cancer, CCC clear cell cancer

The tumor stroma consists of cellular components such as CAFs, immune cells, endothelial cells, pericytes, adipocytes, and so forth, as well as acellular components such as extracellular matrix proteins (ECMs) (Serrano et al. 1993 ). All the components of the tumor microenvironment have been shown to play important roles in the initiation and progression of several tumors, including EOC. Therefore, the molecular spectrum of this aggressive tumor landscape warrants global efforts to further understand its biology to identify novel clinical biomarkers and develop new therapeutic strategies. Recent advances in oncological research have led to several investigations focusing on the diagnosis, prognostic stratification and therapeutic strategy for gynecological cancer (Perelli et al. 2023 ). However, relatively little is known about the relationships among morphology, biology and genotype. Although most BRCA1 germline mutation-associated EOCs have been reported to be of high-grade serous histology, relatively few BRCA1-mutated patients have other histological types. In our series, the BRCA1 and BRCA2 genes were mutated in 22.22% and 12.96%, respectively, of high-grade serous ovarian cancers. Conversely, the most frequent BRCA mutations in endometrioid and clear cell ovarian cancer were found to involve the BRCA 2 gene. Our findings confirmed the characteristic morphologic appearance of HGSCs with BRCA1 abnormalities, as previously described in the literature. Compared with patients lacking BRCA abnormalities, patients with BRCA mutations are significantly more likely to have undergone SET (Soslow et al. 2012 ). Nonetheless, SET tumors are considered biologically and clinically distinctive, such that it is reasonable to call attention to their existence in diagnostic reports (high-grade serous carcinoma, transitional cell carcinoma-like variant). The correlation between SET morphology and BRCA mutation seems to be unilateral. SET features were reported in 8 samples (14,8%) of high-grade serous ovarian cancers, 7 of which had BRCA germline and somatic mutations (87,5%). Conversely, SET features were found in only 36.84% of patients with BRCA-mutated endometrioid ovarian cancer. These results suggest the possibility of evaluating the use of SET morphology as an indicator of BRCA mutation status in centers without the opportunity to perform genetic tests routinely; further studies are needed to validate this hypothesis. Tumor cells orchestrate the recruitment and activation of stromal cells within the TME; instead, CAFs are involved in the remodeling of the TME to facilitate cancer progression and metastasis development in EOC (Dasari et al. 2018 ; Sun and Fu 2019 ; Leung et al. 2018 ; Mhaidly and Mechta-Grigoriou 2021 ): the molecular mechanism of CAF-mediated tumor enhancement is poorly defined. EOC–associated CAFs often express α-SMA, while normal fibroblasts do not. However, the expression of α-SMA is only one of the several findings detected in activated fibroblasts. The expression of α-SMA may also change between CAFs (Dasari et al. 2018 ), and CAFs in ovarian tumors are predominantly α-SMA positive (Mhawech-Fauceglia et al. 2014 ) and are often correlated with a poor prognosis, as reported in other cancers. In addition to their role in controlling cancer cell behavior, CAFs are emerging as the main players in shaping the TME toward an immunosuppressive and growth-promoting phenotype, via increased production of immunosuppressive cytokines and enhanced expression of immune checkpoints (Rambau et al. 2018 ). In our study, significant stromal expression of α-SMA was detected in 85,19% of tumor samples from high-grade serous ovarian cancer patients. Notably, stromal α-SMA expression was positive in 87,50% of patients with wild-type BRCA and in 100,00% of those with mutated BRCA. Regardless of the mutational status, ⍺-SMA seems to be expressed by activated fibroblasts in the early stages of carcinogenesis. Among the cellular and noncellular heterogeneities of the TME, cellular senescence is well established as one of the most important mechanisms for suppressing tumorigenesis; moreover, p16 is known to be a tumor suppressor, and its expression status in the peritumoral stroma has rarely been investigated. In our analysis, stromal p16 expression was positive in 6 out of 38 (15,78%) patients with wild-type BRCA and in only 1 of the 22 (4,5%) patients with mutated BRCA. A higher prevalence of p16-positive tumors was detected in patients with stage I-II disease (16,70% vs 2,90%). In the advanced stage of disease, a mutation in a BRCA gene is correlated with a complete absence of p16 stromal expression. Research is actively trying to unravel the possible correlation between stromal p16 and alpha-SMA in EOC may lie in the role of the tumor stroma in cancer progression. The specific patterns of expression could correlate with increased aggressiveness and senescence-associated secretory phenotype. In our pilot study, we searched for a possible link between the role of CAFs in promoting tumor development and the stromal expression of p16. In previous studies, Rambau et al. investigated the associations of the p16 staining pattern in tumor cells with overall survival and reported differences across EOC histotypes. The absence of p16 expression was most common in HGSC patients (56%) and was significantly associated with shorter survival in patients with endometriosis-related ovarian carcinomas (Mhaidly and Mechta-Grigoriou 2020 ). Nevertheless, this previous large-scale collaborative study did not validate p16 expression as a prognostic marker in HGSC. More recently, the tumor suppressors p16 and p53 have been identified as factors that contribute to cellular senescence. Several conflicting findings about the stromal expression of p16 and its role in stromal senescence have been reported in the literature. Whereas senescence within the tumor itself suppresses tumor growth, senescence in the microenvironment has been shown to increase tumor growth (Harper et al. 2018 ). A significant discrepancy was observed between the stromal expression of p16 and that of ⍺-SMA in all the samples, and a monotonic relationship was shown between these two biomarkers. Our results suggested different biological implications for the stromal expression of p16 and ⍺-SMA. To our knowledge, this is the first study describing a significant discrepancy in the stromal expression of α-SMA and p16 in EOCs. Due to the complexity of the TME, p16 is known to be a tumor suppressor, and a decrease in stromal p16 is inversely correlated with the stromal expression of α-SMA, a marker of fibroblast activation. Our results suggest that the pro-tumorigenic functions of CAFs could be attributed either to their ability to produce pro-survival factors, which directly and positively impact tumor progression by enhancing cancer cell proliferation, survival and metastasis, or to their role in regulating the antitumor immune response by inducing an immunosuppressive microenvironment (Rambau et al. 2018 ). The weakness of this investigation is represented by the relatively small number of samples. This is a pilot study: additional studies on a larger number of patients and evaluating several biomarkers are strongly warranted to elucidate the role of stroma in the development and progression of EOC, especially high-grade serous ovarian cancer, and to correlate stromal biomarkers with BRCA mutational status. Currently, according to international guidelines, the BRCA test should be requested for all women with non-mucinous and nonborderline tumors, regardless of age and family history. BRCA testing could be preferentially performed on neoplastic tissues. In the presence of a positive tumor test, a genetic test should always be performed on a blood sample to differentiate between germline mutations, which require counseling and genetic testing of family members, and somatic mutations. However, because of the limited population sample, the differences among the subgroups were not statistically significant. The limited sample in our investigation could also be an accidental reason why the percentage of patients with a BRCA mutation was higher than that in other recent population-based studies. To our knowledge, no previous studies have investigated the correlation between BRCA mutational status and the stromal landscape in patients with ovarian cancer. The development of new targeted therapies and patient stratification methods has been hampered by the lack of a detailed understanding of the specific tumor microenvironment in EOC. In conclusion, CAFs are leading players in high-grade ovarian cancer, and their protumorigenic functions are attributed to their ability to produce prosurvival factors and to their role in regulating the antitumor immune response by inducing an immunosuppressive microenvironment. Continuous reciprocal interactions between CAFs, cancer cells and other elements of the TME shape their fate, marker expression, and function in the tumor. Further research toward a better understanding of the plasticity, regulation, function, and heterogeneity of these cells could greatly increase the understanding of tumors.

Epithelial ovarian cancer (EOC) is the leading cause of death from gynecologic malignancies in Western countries (Jemal et al. 2009 ). Usually, it is diagnosed in International Federation of Gynecology and Obstetrics (FIGO) stage IIb-IV disease through the spread of the disease into the peritoneal cavity and other organs (Jelovac and Armstrong 2011 ). EOCs are divided into five main distinct diseases according to their different genetic alterations, pattern of spread, clinicopathological features and prognosis. These histologic types include high-grade serous, endometrioid, clear cell, mucinous and low-grade serous carcinoma, accounting for 70%, 10%, 10%, 3% and < 5%, respectively, of the cases respectively (Jelovac and Armstrong 2011 ; Prat et al. 2018 ). However, the mechanism of carcinogenesis in EOC has not been fully elucidated. Mutations in the genes BRCA1 and BRCA2 lead to increased cancer predisposition, approximately 14% of EOCs present these gene mutations, according to recent population-based studies (Alsop et al. 2012 ). EOC is characterized by considerable biological, histological and molecular heterogeneity. Recent studies have already revealed that the tumor microenvironment (TME) plays a central role in promoting cancer initiation, progression and metastasis, as well as in the response to chemotherapy (Pietras and Östman 2010 ; Goubran et al. 2014 ). Cancer-associated fibroblasts (CAFs) are the major constituents of the TME and play critical roles in promoting many aspects of tumor function and in modulating tumor cell proliferation and spread (Dasari et al. 2018 ). α-smooth muscle actin (α-SMA) seems to be a good marker for CAFs: CAFs expressing α-SMA could enhance stromal reaction and contribute to a pro-tumorigenic environment. p16 INK4A (hereafter referred to as p16) is a cyclin-dependent kinase inhibitor that has several biological functions, including the inhibition of cell cycle progression (Serrano et al. 1993 ), and the induction of Senescence-Associated Secretory Phenotype (SASP), secreting factors that modify the TME in ways that promote cancer cell invasion and metastasis (Ohtani et al. 2004 ). The tumor suppressor functions of p16 have been extensively studied in tumor cells, but little attention has been given to its potential effects on the surrounding stromal cells, including CAFs (Al-Ansari et al. 2013 ). The possible correlation between stromal expression of p16 and α-SMA in EOC may lie in the role of the tumor stroma in cancer. The stromal expression of p16 could reflect an attempt of the stroma to counteract tumor progression (tumor-suppressive effects), instead CAFs expressing α-SMA could increase tumor aggressiveness. Research is actively trying to unravel these complex interactions and differential outcomes: the specific patterns of expression could correlate with different clinical outcomes, where certain combinations of p16 and alpha-SMA expression are indicative of either better or worse prognosis, depending on how they interact within the tumor ecosystem to understand better how they can be targeted therapeutically to inhibit ovarian cancer progression. The aim of this preliminary study was to investigate the correlation between the stromal expression of a tumor suppressor, p16, and the expression of α-SMA (a marker of CAF activation) in EOC.

Below is the link to the electronic supplementary material. Supplementary file1 (PDF 85 KB) Supplementary file1 (PDF 85 KB)