Cervical cancer is the fourth most common cancer among women worldwide, affecting nearly 600,000 women annually (1). The application of the human papillomavirus (HPV) vaccine and screening programs have significantly reduced the incidence of cervical cancer; however, it is still highly prevalent in developing countries as the second most common cause of cancer-related deaths in women (2). Even though the disease at its early stages can be amenable to surgery or radiotherapy, recurrent or metastatic cervical cancer is still incurable and calls for novel therapeutic approaches (3). In the past decade, the ICI, pembrolizumab, was the only novel treatment approved by the FDA for treating PD-L1-positive, recurrent or metastatic cervical cancer patients with disease progressing on or after chemotherapy. Though it offers new hope for advanced disease, it is notable that its efficacy was still poorly limited, with an objective response rate of 14.6% in patients with PD-L1–positive tumors (4). Thus, a better understanding of the genomic feature of cervical cancer is a fundamental part for the identification of biomarkers for the development of novel therapeutic approaches and improvement of the efficacy of ICIs.

With the great advances of next-generation sequencing, it enables the researchers to find a comprehensive genomic feature and identify the treatment-related biomarkers in cervical cancer patients. The genomic profiles of Western patients with cervical cancer have been revealed by the TCGA project in 2017 (5). A high prevalence of genes, namely, PIK3CA, EP300, FBXW7, and PTEN was identified as the genomic feature of Western cervical cancer patients, and inferred as novel potential therapeutic targets for drug development in future. Meanwhile, in a pan-cancer study, researchers from the Memorial Sloan-Kettering Cancer Center (MSKCC) uncovered that over one third of metastatic cervical cancer patients harbored at least one actionable alteration (6). However, comparing with other tumor types, the number of studies in genetic profiling on cervical cancer is relatively limited. Furthermore, previous studies were predominantly on the Caucasian patients, leaving an unsolved question on whether there were genetic differences between Chinese and Western cervical patients.

Recently, it has been found that DDR alteration(s) could influence the inflammatory signaling pathways which have the ability of reshaping tumor microenvironment (7), and are emerging as an effective biomarker for predicting the response of ICI, for example, (1) DDR alterations were significantly correlated with clinical benefit in urothelial carcinoma patients who received the therapeutic treatment of anti-PD1/PDL1 (8), (2) ICIs therapy could improve the survival of non-small cell lung cancer patients having co-mutations of DNA damage response and repair pathways (9), and (3) DDR mutations were correlated with improved overall survival of patients with colorectal cancer (10). Meanwhile, it has been comprehensively studied in other gynecological tumors, especially in ovarian and endometrial carcinoma as the hallmark event for precision medicine or prognosis classification. However, the role of DDR in cervical cancer has not been specifically clarified yet. In cervical cancer, HPV could manipulate DDR genes to improve its viral life and prevent the viral apoptosis (11). In addition, the progression of cervical cancer is significantly associated with the increased genetic instability, which is primarily caused by the abnormal regulation of DDR genes (12). Thus, the latest Clinical Trials Planning Meeting from the National Cancer Institute (NCI) in 2020 have stressed the development of clinical trials to explore the potential role of DDR in the treatment of cervical cancer (13).

To our knowledge, there existed several studies describing genomic features of Chinese cervical cancer patients, namely, 13 cervical cancer cases of Chinese Hong Kong women (14), 20 endocervical adenocarcinoma cases (15), 32 cervical cancer cases (16), 32 advanced cervical cancer (17), and 74 cervical cancer cases (18) of Chinese mainland women. However, the latter two studies focused on the molecular profiles of integrated gynecologic cancers containing ovarian cancer, endometrial cancer, and cervical cancer. In the present study, we performed the NGS to determine the genomic profiling of 129 Chinese cervical cancer patients, especially the actionable alterations to explore some potential therapeutic strategies. Furthermore, by comparison with the data from Western cohort, it was the first time to figure out the genetic difference(s) between Chinese and Western patients with cervical cancer. Subsequently, we further explored the DDR alteration and tumor microenvironment based on the public dataset.

Over the past decades, the overall survival of advanced cervical carcinoma has not been strikingly improved, mainly attributing to slow drug development. Unlike ovarian carcinoma, the most prevalent genes in cervical carcinoma patients have poor relationship with any target therapy with high efficacy (5). Furthermore, the genetic feature of Chinese cervical carcinoma patients has not been clarified yet.

Initially, we found significant differences in the genetic features between our cohort and the TCGA database, namely, a different prevalence of KMT2C, RET, and TP53. The frequency of TP53 in our cohort was nearly equivalent with previous result in a 32-patient cohort (15.50% vs 15.60%, p >0.05) (17), both of which were more prevalent among Chinese cervical patients in comparison with those (7.90%) in the Western cohort (p <0.05). Meanwhile, in a 32-sample cohort the frequency of KMT2C was also significantly lower than the result in the TCGA cohort (<9.00% vs 18.56%, p <0.05), but the Western groups had similar frequency of KMT2C (TCGA cohort vs 182-patient cohort: 18.56 vs 16.00%, p >0.05) (26). Of note, the prevalence of RET in Chinses cohort was first found in the present study. Our study revealed the different genetic profiles of cervical cancer patients with different genetic backgrounds. However, the prevalence of the most recurrent genes and actionable genes were similar, and notably, over 40% of investigated Chinese and Western cervical cancers patients harbored at least one actionable genomic alteration, which was also close to the previous findings of Zehir et al. (6). The most prevalent actionable alterations were in PI3K signaling pathway, especially for PIK3CA and PTEN. A recent study also demonstrated the high prevalence of PIK3CA alterations in cervical carcinoma patients with 31.30% altered samples, suggesting the promising targeted therapy with related PI3K or mTOR inhibitor (27). However, only limited evidences supported the correlation of PIK3CA alterations with the response to the mTOR inhibitors in cervical carcinoma (28). Besides, ERBB2 is widely altered in solid tumors, especially breast and gastric cancers. Previous studies found that nearly 5.5 and 3.15% of Western and Chinese invasive cervical carcinoma patients had ERBB2 alterations, which were associated with a worse prognosis (29, 30). In our study, we found 9.30% of the patients had ERBB2 alterations, but 6.97% had oncogenic alterations, including 6 gains of function missense variants and 3 of amplification. Interestingly, all the oncogenic ERBB2 alterations were only identified in the tissue samples. Early research on patient-derived xenograft derived from the cervical carcinoma patients found anti-HER2 therapy, the combination of trastuzumab and lapatinib inhibited tumor growth. Neratinib, an ERBB2 inhibitor, showed a confirmed objective response rate of 25% and progression-free survival of 7.0 months in 10 cervical carcinoma patients from the phase 2 SUMMIT basket trial (31). In addition, 10.88% of Chinese cervical carcinoma patients in our cohort were identified to harbor functional DNA damage repair alterations, similar to the prevalence in Western patients (16.15% in the TCGA cohort and 13.2% in another cohort with 824 Western cervical patients) (32). In the past decades, PARP inhibitors have been the promising targeted therapies for pan-cancers, especially for those with homology recombination deficiency. Though they have made remarkable progress in multiple solid tumors, namely, ovarian, breast, pancreatic, and prostate carcinoma, results of the efficacy of PARP inhibitors in cervical cancer are still quite poor. One study analyzed the combination of chemotherapy (paclitaxel and cisplatin) with PARP inhibitor (Veliparib) in 34 biomarker-unselected persistent or recurrent cervical carcinoma patients, showing a promising ORR of 34%, and the median PFS and OS were 6.2 and 14.5 months, respectively (33). Enlightened by the results of trails on biomarker-guided match-therapy (34), it would be recommended that these cervical cancer patients with actionable alterations in our cohort could try the matched therapy when they progressed following prior treatment or without satisfactory alternative standard treatment options.

In the present study, it was found that DDR alteration was positively correlated with the hypoxia score, especially for the deleterious DDR alteration indicating the higher hypoxia score. In addition, both the buffa and ragnum hypoxia scores, described in previous studies (22, 23), were negatively implicated with the level of CD8+ T cells which play a pivotal role in cancer immunity and are associated with a better response in patients receiving ICIs (35). The immune checkpoint inhibitors are promising treatments for various advanced cancers. FDA had approved pembrolizumab for treating patients with recurrent or metastatic cervical cancer based on the phase II KEYNOTE-158 study, though its objective response rate (ORR) was only 12.2% (4). Given the limited response rate of anti-PD-1 therapy, it is vital to identify robust biomarkers for distinguishing cervical patients who may benefit from ICIs treatment. DDR alteration was widely suggested as an effective biomarker for predicting the potential responder in multiple types of cancer, including lung, bladder, and renal cell carcinoma (36–38). Furthermore, DDR alteration may lead to genomic instability, namely, mismatch instability and chromosomal rearrangements, and further affects the tumor immune microenvironment by activating of T cells and adaptative immune system (39). However, there was no study revealing the relationship among DDR alteration, tumor microenvironment, and ICIs efficacy in cervical cancer. Our study is the first one suggesting that although DDR alteration was associated with a higher TMB value, and it was also positively related to increasing hypoxia feature, which may reshape the immune suppressive tumor microenvironment. DDR alterations, regardless of their specific function, were positively associated with both the higher hypoxia score and hypoxia feature in cervical cancer patients. Previous studies have suggested a complex relationship between hypoxia and DDR function, revealing a multifaceted regulatory role of hypoxia for DDR (40). For chronic tumor hypoxia, it downregulated most DDR pathways to silence their function in maintain genomic stability. Tumor hypoxia is not only associated with the development of malignancy and therapeutic resistance as an indicator for poor outcomes but also serves a vital determinant of tumor microenvironment (41). Previous studies also demonstrated that hypoxia could suppress the NK cell function, affect the contents of effective and regulatory T cells, and promote the polarization of macrophages to M2, a immunosuppressive phenotype (42). Thus, it could be suggested that DDR alteration could not function as a robust determinant for predicting the efficacy of ICIs in cervical cancer patients as other types of cancers, which need to be further verified.

Furthermore, we identified CDHR5 as the significant hub gene solely related to both DDR alteration and hypoxia score. This gene belongs to the superfamily of cadherin, and participates in multiple physical processes including cell adhesion and branching morphogenesis of organs (43). Previous studies have suggested controversial roles of CDHR5 in the cancer progression in different cancer types, but according to the decreased expression level in tumor tissues than the adjacent non-tumor tissues, it’s more likely to function as tumor suppressor (43–46). Its decreased expression in the tumor was associated with hypermethylation and transcriptional regulation. Though Beck and his colleagues found the negative correlation between CDHR5 and DNA replication and repair (44), the explicit relationship between CDHR5 and DDR or hypoxia has not been established yet. This is the first study that suggested the negative correlation between CDHR5 and DDR or hypoxia in the cervical cancer, which merited further study.

The work presented here has several limitations. Firstly, it is limited by the sample size to comprehensively understand the genetic profiling of Chinese cervical cancer patients, and further study with a larger sample size is required to fully evaluate the findings. Secondly, we just investigated the potential correlation between DDR alteration, hypoxia feature, and tumor microenvironment, but whether the efficacy of ICIs in cervical cancer patients with or without DDR alteration is different merits further study.

In summary, this study provided a comprehensive analysis of genomic alterations in Chinese patients with cervical cancer. Genomic profiling of Chinese patients uncovered a unique genomic feature and widely prevalent actionable variants, especially in PI3K and DDR pathways, which could guide clinical management in future. Moreover, we found the association between DDR alteration, hypoxia feature, and tumor microenvironment in cervical cancer, namely, the negatively regulated hub gene CDHR5, suggesting that DDR alteration(s) could not function as a robust predictor of ICIs in cervical cancer patients.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by the Fudan University Cancer Center Ethics Committee. The patients/participants provided their written informed consent to participate in this study.

JL, HW and X-LZ proposed the design of this study. HW and Q-HG collected samples and conducted data analysis. JL, HW, Q-HG and X-LZ wrote the manuscript draft. X-HW and JL revised the manuscript. All authors contributed to the article and approved the submitted version.

This study was supported by the Female Tumor Project of Shanghai Key Clinical Specialty from Shanghai Municipal Health Commission (No. SHSLCZDZK06301).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.