Gastric-type cervical adenocarcinoma (GCA) is a rare and aggressive subtype of cervical adenocarcinoma. Despite its clinical significance, its molecular carcinogenesis and therapeutic targets remain poorly understood. This study aimed to compare the clinicopathological, immunohistochemical, and molecular profiles of GCA and usual-type cervical adenocarcinoma (UCA), exploring prognostic and therapeutic biomarkers in a Japanese population. A total of 110 cervical adenocarcinoma cases, including 16 GCA and 94 UCA cases, were retrospectively analyzed for clinicopathological features, and a panel of immunohistochemical markers was assessed. Sanger sequences were performed for the KRAS, PIK3CA, and BRAF genes, and survival and clinicopathological correlations were assessed using Kaplan–Meier and Cox regression analyses. GCA was significantly associated with more aggressive features than UCA, including lymph node involvement, advanced FIGO stages, increasing recurrence rate, and poor survival status. High ARID1B expression was observed in a subset of GCA cases and correlated with worse progression-free and overall survival. Additionally, PD-L1 expression was more frequent in GCA than UCA and was associated with unfavorable prognostic factors. Conversely, UCA cases showed strong p16 expression, supporting their HPV-driven pathogenesis. Molecular profiling revealed KRAS and PIK3CA mutations in both subtypes, while BRAF mutations were identified exclusively in GCA. These findings reveal distinct clinical and molecular profiles for both tumor types and underscore ARID1B and PD-L1 as predictive prognostic and therapeutic biomarkers in GCA, implicating the use of subtype-specific treatment strategies.
The molecular mechanisms through which endometriosis-related ovarian neoplasms (ERONs) develop from benign endometrioma remain unclear. It is especially a long-standing mystery why ovarian endometrioma has the potential to develop into two representative histological subtypes: endometrioid ovarian carcinoma or clear cell ovarian carcinoma. This study aimed to investigate the molecular carcinogenesis of ERONs using newly developed in vitro and in vivo carcinogenesis models. Epithelial cells were isolated and purified from surgically removed benign endometrioma samples, followed by immortalization by overexpressing cyclinD1/CDK4 in combination with the human TERT gene. Immortalized cells were subjected to various molecular manipulations by combining knockout or overexpression of several candidate drivers, including ARID1A, KRAS, PIK3CA, AKT, and MYC, based on previous comprehensive genome-wide studies of ERONs. These cells were then inoculated into immunocompromised mice and evaluated for malignant transformation. Inoculated cells harboring a combination of three genetic alterations successfully developed tumors with malignant features in mice, whereas those with two genetic manipulations failed to do so. Especially, ARID1A gene knockout, combined with overexpressing the KRAS oncogenic mutant allele (or overexpressing AKT) and c-Myc overexpression led to efficient tumor formation. Of note, these three combinations of genetic alterations produced tumors that histologically represented typical clear cell carcinoma in SCID mice, while the same combination led to tumors with endometrioid histology in nude mice. A combination of ARID1A mutation, KRAS mutation or AKT activation, and c-Myc overexpression were confirmed to be the main candidate drivers for the development of ERONs, as suggested by comprehensive genetic analyses of ERONs. A tumor immune microenvironment involving B-cell signaling may contribute to the diverse histological phenotypes. The present model may help to clarify the molecular mechanisms of ERON carcinogenesis and understand their histological diversity and novel molecular targets.
Ovarian cancer is classified as type 1 or 2, representing low- and high-grade serous carcinoma (LGSC and HGSC), respectively. LGSC arises from serous borderline tumor (SBT) in a stepwise manner, while HGSC develops from serous tubal intraepithelial carcinoma (STIC). Rarely, HGSC develops from SBT and LGSC. Herein, we describe the case of a patient with HGSC who presented with SBT and LGSC, and in whom we analyzed the molecular mechanisms of carcinogenesis. We performed primary debulking surgery, resulting in a suboptimal simple total hysterectomy and bilateral salpingo-oophorectomy due to strong adhesions. The diagnosis was stage IIIC HGSC, pT3bcN0cM0, but the tumor contained SBT and LGSC lesions. After surgery, TC (Paclitaxel + Carbopratin) + bevacizumab therapy was administered as adjuvant chemotherapy followed by bevacizumab as maintenance therapy. The tumor was chemo-resistant and caused ileus, and bevacizumab therapy was conducted only twice. Next-Generation Sequencing revealed KRAS (p.G12V) and NF2 (p.W184*) mutations in all lesions. Interestingly, the TP53 mutation was not detected in every lesion, and immunohistochemistry showed those lesions with wild-type p53. MDM2 was amplified in the HGSC lesions. DNA methylation analysis did not show differentially methylated regions. This case suggests that SBT and LGSC may transform into HGSC via p53 dysfunction due to MDM2 amplification.
The molecular mechanisms through which endometriosis-related ovarian neoplasms (ERONs) develop from benign endometrioma remain unclear. It is especially a long-standing mystery why ovarian endometrioma has the potential to develop into two representative histological subtypes: endometrioid ovarian carcinoma or clear cell ovarian carcinoma. The present study aimed to investigate the molecular carcinogenesis of ERONs using newly developed in vitro and in vivo carcinogenesis models. Epithelial cells were isolated and purified from surgically removed benign endometrioma samples, followed by immortalization by overexpressing cyclinD1/CDK4 in combination with the human TERT gene. Immortalized cells were subjected to various molecular manipulations by combining knockout or overexpression of several candidate drivers, including ARID1A, KRAS, PIK3CA, AKT, and MYC, based on previous comprehensive genome-wide studies of ERONs. These cells were then inoculated into immunocompromised mice and evaluated for malignant transformation. Inoculated cells harboring a combination of three genetic alterations successfully de-veloped tumors with malignant features in mice, whereas those with two genetic mutations failed to do so. Especially, ARID1A gene knockout combined with overexpressing the KRAS oncogenic mutant allele (or overexpressing AKT) and c-Myc overexpression led to efficient tumor formation. Of note, these three combinations of genetic alterations produced tumors that histologically represented typical clear cell carcinoma in SCID mice, while the same combination led to tumors with endometrioid histology in nude mice. A combination of ARID1A mutation, KRAS mutation or AKT activation, and c-Myc overexpression were confirmed to be the main candidate drivers for the development of ERONs, as suggested by comprehensive genetic analyses of ERONs. A tumor immune microenvironment involving B cell signaling may contribute to the diverse histological phenotypes. The present model may help to clarify the molecular mechanisms of ERON carcinogenesis and understand their histological diversity and novel molecular targets.
Background/objective: Mucinous borderline tumors of the ovary (MBOTs) are characterized by their unique histological features and intermediate malignant potential; however, the factors underlying their molecular carcinogenesis and tumor biology remain largely unknown. Developing cell lines from these tumors presents an ongoing challenge. The purpose of this study is to establish MBOT cell lines and characterize their biological features. Methods: Epithelial cells were collected and purified from surgically removed MBOT samples and then stably maintained with an extended life span by overexpressing CyclinD1/CDK4 in combination with human telomerase reverse transcriptase. The characterization of resulting cell lines was defined by morphology, growth kinetics, functional analysis, whole-exome sequencing, and tumorigenicity in mice. Results: Two independent cell lines, HMucBOT-1 and HMucBOT-2, were successfully established from the tissues of a patient with an MBOT, with the latter showing more aggressive growth capacity. In the patient-derived xenograft model, HMucBOT-1 cells retained the original morphological characteristics of the MBOT, whereas HMucBOT-2 cells displayed a transition to mucinous carcinoma accompanying undifferentiated carcinoma, suggestive of dedifferentiated carcinoma. Genetic analysis of the original tumor sample and HMucBOT-2 cells revealed shared oncogenic mutations. However, KRAS amplification and certain copy number alterations were uniquely observed in the HMucBOT-2 cells. Conclusions: The above results indicate that HMucBOT-1 can serve as a preclinical model for investigating the biological behavior of and potential targeted therapies for human MBOTs, with HMucBOT-2 serving as a valuable tool for studying the heterogeneity and genetic diversity of this tumor and explaining the potential causes of treatment failure or relapse.
Although endometriosis is a benign disease, it is associated with cancer-related gene mutations, such as KRAS or PIK3CA. Endometriosis is associated with elevated levels of inflammatory factors that cause severe pain. In a previous study, we demonstrated that KRAS or PIK3CA mutations are associated with the activation of cell proliferation, migration, and invasion in a patient-derived immortalized endometriotic cell line, HMOsisEC10. In this study, we investigated the effects of these mutations on progesterone resistance. Since the HMOsisEC10 had suppressed progesterone receptor (PR) expression, we transduced PR-B to HMOsisEc10 cell lines including KRAS mutant and PIK3CA mutant cell lines. We conducted a migration assay, invasion assay, and MTT assay using dienogest and medroxyprogestrone acetate. All cell lines showed progesterone sensitivity with or without mutations. Regarding inflammatory factors, real-time quantitative RT-PCR revealed that the KRAS mutation cell line exhibited no suppression of Cox-2 and mPGES-1 on progesterone treatment, whereas IL-6, MCP-1, VEGF, and CYP19A1 were significantly suppressed by progesterone in both mutated cell lines. Our results suggest that KRAS mutation and PIK3CA mutation in endometriotic cells may not be associated with progesterone resistance in terms of aggressiveness. However, KRAS mutations may be associated with progesterone resistance in the context of pain.