According to estimates from GLOBOCAN 2018, cervical cancer was the fourth most common cancer among women, with approximately 570,000 new cases and 311,000 deaths (1). Cure rate for earliest stages is more than 90%, whereas locally advanced lesions treated with multimodality therapy can only achieve 65% cure rate in stage II lesions and 55% in stage III lesions (2). Therapies to improve survivorship are in desperate need for cervical cancer especially locally advanced disease.

With rapid development of precision medicines, novel therapeutic strategies, especially immunotherapies, have been proposed to significantly improve clinical outcomes of cervical cancer (3, 4). Harnessing an antitumor immune response has been a fundamental strategy in cancer immunotherapy. A paradigm shift has appeared in cancer immunotherapy: from traditional immune enhancement with low-objective responses and frequent adverse events to more effective and less toxic reactions immune normalization (5, 6).

Cancer immunotherapies, such as administration of the cytokine Interleukin-2 (IL-2), adoptive cell transfer, and the checkpoint modulators CTLA-4 and PD-1, have proved effective in clinical practice (6). Blockade of the checkpoint modulators cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and programmed cell death protein-1 (PD-1) starts the field of immune normalization in immunotherapy. Upregulated PD-1 in tumor microenvironment inhibits an effector T cell antitumor immune response, and therapies blocking this pathway have proven effective against multiple tumor types (5). Anti-PD therapy perform antitumor immunity mainly through the following three principles:1) targeting a tumor-induced immune escape mechanism, 2) selectively modulating immunity in the tumor microenvironment, and 3) resetting immunity in the tumor microenvironment (5). It has been proved that patients with high mutation burden and burden of potential neoepitopes benefit more from immunological checkpoint blockade (7–11). Immunotherapy plays a dispensable role in management of cervical cancer. In KEYNOTE-158, pembrolizumab has been approved by the US Food and Drug Administration for use in advanced cervical cancer with progressive disease either during or after chemotherapy (12). The objective response rate was 26.3% with a disease control rate of 68.4% for immunotherapy in cervical cancer (4). However, only small proportion patients benefited from immunotherapy, and this proportion will hopefully increase with better patient selection and combination therapy. Hence, it is urgent to find potential biomarkers for prediction of response to checkpoint immunotherapy and the rationale for the use of checkpoint immunotherapy.

In the present study, we qualify immune cells infiltration in cervical cancer and analyze the correlation between immune cells and cancer prognosis. Hub genes regulating prognosis through immune infiltration in cervical cancer were identified by weighted gene co-expression network analysis (WGCNA) and least absolute shrinkage and selection operator (LASSO). It was suggested by infiltrated immune cells and pathway enrichment analysis that our immune-related signature was closely related to tumor prognosis and could predict response of immunotherapy. A robust immune-related prognostic signature based on transcriptomics in cervical cancer was constructed and validated.

Cervical cancer represents a major public health problem even in developed countries such as United States, with 14,000 new cases and 4,285 deaths estimated in 2022 (20). Increasing incidence of cervical adenocarcinoma and attenuation of earlier declines for cervical squamous cell carcinoma emphasize the importance of the need for improved therapeutic options to reduce the burden of cervical cancer (21). The most important risk factors affecting prognosis of cervical cancer are stage, status of the lymph nodes, tumor volume, depth of tumor invasion into the cervical stroma, and Lymphovascular space invasion (LVSI) (22). The 5-year OS decreased significantly with rising stages (23). A substantial percentage of patients with advanced cervical cancer will undergo recurrence and poor prognosis (2, 24), and the recurrence rate fluctuates from 10% to 74% (25). Tumor stage was one of the most pivotal factors related to recurrence: 10% for stage IB, 17% for stage IIA, 23% for stage IIB, 42% for stage III, and 74% for stage IVA (26). Prognosis of metastatic and recurrent cervical cancer was extremely poor with a median survival time of 12 months (27). Therefore, improved therapeutic options in management of cervical cancer, especially locally advanced cervical cancer is in urgent need.

High-risk subtypes of the human papilloma virus (HPV) are the cause of cervical cancer (28, 29). Viral oncoproteins E6 and E7 leads to dysregulation of p53 and Hypoxia-inducible factor-1 (HIF-1) alpha, thus affecting cell cycle proteins and VEGF expression (28–32). HPV is highly immunogenic and elicits immune responses in humans; thus, immune response might play important roles in carcinogenesis of cervical cancer. Incidence of cervical cancer was substantial declined in countries with high HPV vaccine coverage (33–35). Immunotherapy fights against tumor cells through activating endogenous immune response, which seems to switch on the new frontier of the anticancer treatment (36). Immunotherapy includes different approaches, such as active immunotherapy (vaccine), passive immunotherapy (adoptive cellular transfer, antibodies, and cytokines), and immunomodulation (cyclooxygenase 2 inhibitor).

Discovery of immune checkpoint such as CTLA-4 and PD-1 plays an indispensable role in the development of cancer immunotherapy. It was surprising that immune checkpoint inhibitors anti–CTLA-4 and anti–PD-1 displayed enormous success in solid tumors (37). Similar to other solid tumors, novel immunotherapeutic approaches, such as immune checkpoint inhibitors, have shown encouraging results in cervical cancer (12). Implementing immunotherapeutic approaches earlier in advanced cervical cancer would seem to be most appropriate. However, the objective response rate with anti–PD-1/PD-L1 monotherapy is hovering at 20%. Moreover, immune-related toxicities and severe adverse effects can occur during PD-1/PD-L1 blockade therapy (38). PD-1/PD-L1 blockade therapy does not demonstrate efficacy in almost 80% of patients with cervical cancer, suggesting that the potential mechanism of PD-1/PD-L1 in immunotherapy remains to be further clarified. Thus, new immune checkpoint inhibitors or comprehensive understanding of specific mechanism underlying PD-1/PD-L1 regulation in carcinogenesis is in urgent need.

Immune infiltration of cervical cancer determines the immune activation of tumor microenvironment and is related with clinical outcome of patients. In this study, immunophenoscore of 28 infiltrated immune cells in TCGA cervical cancer was calculated. Univariate Cox regression analysis showed that activated B cell, activated CD8 T cell, eosinophil, monocyte, activated CD4 T cell, effector memory CD8 T cell, immature B cell, and plasmacytoid dendritic cell were strongly associated with OS of cervical cancer. WGCNA revealed that magenta module was the most relevant module to immune infiltration. In total, we identified 10 kinds of infiltrated immune cells with strong correlation with magenta module. These were as follows: activated B cell (R = 0.88), activated CD8 T cell (R = 0.82), activated dendritic cell (R = 0.79), central memory CD4 T cell (R = 0.78), effector memory CD8 T cell (R = 0.85), immature B cell (R = 0.93), macrophage (R = 0.72), Myeloid-derived suppressor cells (MDSC) (R = 0.85), regulatory T cell (R = 0.71), T follicular helper cell (R = 0.72), and type 1 T helper cell (R = 0.83).

By the use of WGCNA, magenta module showed the highest correlation with the immune infiltration of cervical cancer. The magenta module contained 609 immune-related genes, and 153 genes were picked up for further analysis after taking intersection of prognosis and immune infiltration. Through LASSO analysis, 15 immune-related genes (LAG3, CD74, CCL22, CH25H, OLR1, MIAT, BATF, IKZF3, TRARG, ACSL6, C11orf21, GTSF1, APOL1, CD1C, and LINC00158) were included in prognostic classifier. In addition to CTLA-4– and PD-1/PD-L1–targeted cancer immunotherapy, LAG3 (lymphocyte activation gene-3, CD223) is the third clinically targeted inhibitory receptor (39). CD74 (invariant chain) plays a dispensable role in process of immune systems that it participates in antigen presentation, B-cell differentiation, and inflammatory signaling. CD74 has the potential to be a therapeutic target in cancer and autoimmune disease (40). The chemokine CCL22 promoted regulatory T cell communication with dendritic cells to control immunity and was associated with poor prognosis (41, 42). CH25H produces 25-hydroxycholesterol, which inhibited tumor-derived extracellular vesicles uptake and correlated with prognosis in patients with melanoma (43). OLR1 (oxidized low-density lipoprotein (LDL) receptor 1) was a possible link between obesity, dyslipidemia, and cancer. OLR1 played carcinogenic role by activating Nuclear factor kappa B (NF-κB) pathway to promote proliferation and migration and to inhibit apoptosis and de novo lipogenesis (44). MIAT Long non-coding RNAs (lncRNA) was overexpressed in a number of malignancies and caused poor prognosis (45–47). Basic leucine zipper transcription factor, ATF-like (BATF) was an important transcription factor regulating differentiation of early effector CD8⁺ T cells (48) and was a prognostic indicator for patients with colon cancer (49). IKZF3 promoted growth of multiple tumors to cause poor prognosis (50).

Immune infiltration played important roles in the survival of cervical cancer. In this study, we identified 15 immune infiltration associated genes in cervical cancer and built a prognostic signature. Immune scores depended on the expression of these 15 genes and were associated with the survival of cervical cancer. High immune scores meant good prognosis. GSEA analysis showed that the 15-gene prognostic signature was obviously associated with PD-L1 expression and PD-1 checkpoint pathway in cancer. The prognostic signature could provide basis for potential immunotherapy in the future. Similarly, other researchers also constructed prognostic signatures for cervical cancer based on immune-related genes (51, 52). Compared with our study, although these studies use different types and different quantity of genes, all the signatures can well predict the prognosis of cervical cancer. However, the study has several limitations. First, no in vitro or in vivo molecular experiment was performed to verify our analysis. Second, our study was a retrospective study. Thus, prospective study is in need to validate the findings of our study in the future.

In conclusion, we successfully constructed a 15-gene prognostic signature with powerful predictive function. Differences in the OS of high- and low-risk groups are implicated in immune infiltration, tumor microenvironment, PD-L1 expression, and PD-1 checkpoint pathway. These findings revealed the underlying mechanism of immunotherapy and provided basis for cervical cancer pathogenesis and clinical treatment.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

RS is responsible for experimental design. CWJ and HB are responsible for instrument operation. JX is responsible for data analysis. LZ and CJJ are for providing overall ideas. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (grant no. 82002736 to CJJ) and the clinical characteristic medical technology cultivation plan of Shanghai General Hospital (no. 02.DY12.06.22.07 to LZ and no. 02.DY12.06.22.12 to JX).

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.