4.1.1. Molecular Divergence Between SCC and ADC

The main histological subtypes of CC are SCC and ADC. Despite their distinct histological features, detailed comparisons of tumor heterogeneity between SCC and ADC of cervix remain relatively underexplored in the literature [27]. Single-cell analyses have unveiled distinct molecular profiles between SCC and ADC of the cervix. For instance, Li et al. compared tumor cells from patients with SCC and ADC of the cervix. They revealed that SCC of the cervix is characterized by upregulation of immune-related pathways such as epithelial–mesenchymal transition (EMT), hypoxia, and inflammation, whereas ADC of the cervix shows enrichment in cell cycle pathways, including MYC targets and the G2M checkpoint [27]. Further studies by Qiu et al. confirmed these findings, demonstrating that SCC-specific markers (e.g., S100A8 and KRT14) and ADC-specific markers (e.g., LGALS4 and CLDN3) could effectively stratify subtypes and guide therapeutic targeting [22]. Additionally, distinct epithelial subclusters (Epi1–Epi6) were identified, each associated with unique pathway activation profiles: Epi1 (ADC-like) exhibited activation of hypoxia and oncogenic signaling pathways; Epi2 and Epi4 showed metabolic reprogramming; and Epi5 (SCC-like) was characterized by EMT pathways [30]. These findings underscore the subtype-specific vulnerabilities and identify potential biomarkers for precision therapy.

4.1.2. HPV-Driven Tumor Cell Heterogeneity

To study the intratumor heterogeneity between HPV-associated and HPV-independent cancer, Li et al. used scRNA-seq to investigate the differentiation-related heterogeneity of viral genes and host cells in HPV-associated CC tissue. The clusters containing the highest proportion of HPV-associated cells in the epithelium exhibited deviations from normal epithelial markers, revealing functional heterogeneity and alterations in epithelial structure. This finding indicates substantial molecular heterogeneity within the cancer tissue, with these HPV-associated cells displaying different gene expression profiles compared to normal epithelial cells. Specifically, compared to HPV-independent cells, the HPV-associated cells exhibited unique gene expression patterns related to the extracellular matrix, cell adhesion, proliferation, and apoptosis [34].

4.2.1. Functional Dynamics of CD4⁺ and CD8⁺ T Lymphocytes in Antitumor Immunity

scRNA-seq has been used to delineate multiple functional subpopulations of classical T cells through gene expression analysis. To date, the majority of relevant studies have primarily focused on CD4⁺ and CD8⁺ T cells.

CD4⁺ T cells play complex and multifaceted roles in tumors, primarily by promoting tumor control through cytokine release and related helper cell functions [42]. Using scRNA-seq and high expression of relevant genes, Li et al. identified three subgroups of CD4⁺ T cells in samples from different stages of CC development and progression: naïve CD4⁺ T cells, regulatory T cells (Tregs), and central memory T cells. Among these, the Tregs in the SCC group showed high immunosuppressive activity and were associated with shorter survival. CD4⁺ T cells are one of the main components involved in the early immune response in metastatic lymph nodes (LNs) [28]. Additionally, Qiu et al. reported that Type 1 helper T cells (Th1-like CD4⁺ T cells) expressing CCL4, GZMA, and GZMB are involved in viral protein interactions, natural killer cell–mediated cytotoxicity, and T cell receptor signaling pathways, revealing their positive role in tumor immune defense [22].

CD8⁺ T cells are a key component of the TME and a fundamental element in cancer immunotherapy. Li et al. primarily analyzed the expression of relevant genes in CD8⁺ T cells, identifying three subgroups: naïve CD8⁺ T cells, cytotoxic CD8⁺ T cells, and exhausted CD8⁺ T cells. Naïve CD8⁺ T cells express markers such as CCR7, KLF2, SELL, and LEF1. Cytotoxic CD8⁺ T cells upregulate the expression of cytotoxic genes, such as PRF1, GNLY, GZMA, GZMB, and GZMK. In contrast, exhausted CD8⁺ T cells display increased expression levels of exhaustion-related molecules, such as PDCD1 (PD-1), CTLA4, LAG3, and HAVCR2 (TIM-3). Additionally, the authors used Monocle2 for pseudotime trajectory analysis, which indicated that CD8⁺ T cells in tumors undergo a process of exhaustion. This finding suggests that tumors exhibit an immunosuppressive state characterized by the infiltration of exhausted CD8⁺ T cells [28].

4.2.2. Regulatory Roles of B Cells in Immune Modulation

Compared to T cells, B cells have received relatively less research attention. However, B cell signatures are often associated with favorable clinical outcomes in patients with CC [43]. Recently, there has been increasing recognition of the antitumor characteristics of B cells [44]. To investigate the role of B cells in CC, Cao et al. classified B cells into six subpopulations: activated B cells (marked by TNFRSF13B and CD83), memory B cells (marked by BACH2 and KLF4), germinal center B cells (marked by MME, AICDA, and BCL6), plasma cells (further divided into PC_IGHA1 and PC_IGHG4 and marked by MZB1 and XBP1), and transitional B cells (also marked by MZB1 and XBP1). Functional analysis revealed that genes specific to activated B cells were enriched in antigen processing and presentation pathways, while genes related to complement activation and immunoglobulin production pathways were enriched in plasma cells. This indicates distinct antitumor functions among different B cell subgroups [21]. Furthermore, B cells mainly originate from tumor samples rather than adjacent normal tissues, suggesting that a B cell response is elicited after stimulation within the TME [21].

4.2.3. Polarization Dynamics of M1 and M2 Macrophages

In the TME, accumulating macrophages are referred to as tumor-associated macrophages (TAMs). Clinical data and extensive research suggest that TAMs generally exert procancer effects [45]. TAMs often exhibit different functional phenotypes, broadly categorized into M1 and M2 macrophages. M1 macrophages are typically associated with tumor-killing activities, primarily involved in antitumor responses and immune promotion. In contrast, M2 macrophages possess immunosuppressive properties and may facilitate tissue repair and tumor development [45–47]. TAMs exhibit functional plasticity, transitioning between proinflammatory M1 and immunosuppressive M2 phenotypes. Li et al. observed a shift from M1-dominant macrophages in precancerous lesions (high-grade squamous intraepithelial lesions, HSILs) to M2-dominant macrophages in invasive tumors. This shift correlates with immune evasion and disease progression. Paradoxically, macrophages associated with HSIL exhibited mixed pro- and anti-inflammatory signatures, reflecting early immune activation followed by suppression. Guo et al. further demonstrated that M2 polarization promotes malignant transformation, suggesting that targeting macrophage reprogramming (e.g., blocking M2-inducing cytokines like IL-10) could restore antitumor immunity [24]. These findings underscore the dual role of macrophages across different disease stages and their therapeutic relevance.

4.2.4. CAFs in Stromal Remodeling

CAFs are key components of the tumor matrix and play a crucial role in facilitating interactions between cancer cells and the TME [48]. Numerous scRNA-seq studies have investigated the role of CAFs in tumor metastasis and their impact on prognosis, suggesting that CAFs may serve as potential therapeutic targets [12, 49].In their study of the TME in CC, Qiu et al. conducted a detailed investigation of CAFs, reclassifying them into two main types: inflammatory CAFs (iCAFs, characterized by high expression of CKB, CFD, and DPT) and myofibroblastic CAFs (myCAFs, characterized by high expression of SULF1, INHBA, and COL8A1). Further functional analysis revealed that iCAFs actively participate in the NFKB signaling pathway, complement and coagulation cascades, and cytokine–cytokine receptor interactions, highlighting their importance in promoting inflammation-mediated carcinogenesis in the development of CC [22]. Additionally, myCAFs not only play a crucial role in extracellular matrix remodeling but also significantly contribute to antigen processing and presentation [22]. In summary, both iCAFs and myCAFs promote CC progression through various pathways [22].

4.3.1. Developmental Origins of Cervical SCC

To gain a deeper understanding of the occurrence and progression of CESC, Liu et al. employed scRNA-seq to analyze samples from 3 normal cervix (NC) tissue samples, 2 cervical intraepithelial neoplasia (CIN) samples, 3 early-stage CESC samples (International Federation of Gynecology and Obstetrics (FIGO) 2018 IB2 stage), and 5 locally advanced-stage CESC samples (FIGO2018 IIB to IIIC). As a proven marker highly expressed in normal cervical squamous cells, KRT14 showed high expression in NC samples but low expression in early and advanced CESC. The study revealed that its expression was high in NC samples but significantly reduced in both early- and locally advanced-stage CESC samples. In contrast, the expression of malignant squamous cell markers (KRT17) and typical proliferation markers (CDKN2A, mki67, and TOP2A) progressively increased from CIN to advanced-stage CESC, aligning with the tumorigenic process and progression of CESC [20]. In addition, the authors used Monocle analysis for trajectory inference and identified two distinct trajectories: one representing the tumorigenic trajectory of squamous cells and the other representing the developmental trajectory of nonmalignant squamous cells. The squamous cell subcluster Epi1, located at the beginning of the trajectory, was composed of TP63KRT5++ basal stem cells. On the one hand, the Epi1 cluster may give rise to the malignant squamous cell subclusters Epi7 and Epi8, with increased expression of the malignant markers MKI67 and TOP2A. On the other hand, Epi1 can also differentiate into the squamous cell subcluster Epi5, which shows more differentiated characteristics, including high expression of SPRR3 and SPRR2a. Furthermore, cell cycle analysis of epithelial cells revealed that these epithelial cells were gradually activated along the tumorigenic trajectory, thereby further confirming the tumorigenesis of CESC [20].

4.3.2. Molecular Transitions From Premalignant Lesions to Invasive Cancer

Guo et al. comprehensively mapped the transcriptional changes that occurred during the progression from normal epithelium to cancer induced by HPV. They discovered that three genes (SPRR3, CEACAM7, and APOBEC3A) were upregulated following HPV infection. These genes promoted viral persistence and genomic instability through their antiviral responses, effectively acting as “second hits” that accelerated the process of malignancy. In contrast, the downregulation of protective genes (TCN1, TFF3, and BPIFB1) in HSIL and cancer weakened the defenses against transformation. Additionally, six hub genes (CALML5, CXCL10, KRT6C, KRTDAP, S100A7, SBSN) were uniquely elevated during the transition from HSIL to cancer, serving as potential biomarkers for early detection and intervention [24]. This study provides a detailed molecular roadmap of cervical carcinogenesis and highlights actionable targets for intercepting disease progression.

Unraveling the intricate processes of tumorigenesis and progression is crucial, as it not only helps identify potential therapeutic targets but also provides valuable biomarkers for diagnosis and prognosis. In our study, we meticulously mapped the mechanisms underlying the progression from a normal cervix to CC (Figure 2). Zhang et al. performed scRNA-seq on 122,400 cells derived from 20 cervical biopsy samples. These samples included 5 healthy controls, 4 patients with HSIL, 5 patients with microinvasive cervical cancer (MIC), and 6 patients with invasive cervical squamous cell carcinoma (ICC). By employing single-sample gene set enrichment analysis (ssGSEA) of tumor-associated features to compare the malignant cell clusters in HSIL, MIC, and ICC samples, they uncovered that key tumor-associated features, including EMT, metastasis, angiogenesis, dormancy, inflammation, hypoxia, and differentiation of malignant cells, progressively intensified with disease progression. Further investigation into the metabolic profiles of these cells revealed distinct metabolic signatures in the malignant cells of the HSIL, MIC, and ICC groups. Overall, these findings underscore the escalating heterogeneity and malignancy of tumor cells as the disease progresses [17]. Li et al. observed that HPV-infected epithelial cells no longer maintain the orderly arrangement typical of normal epithelial cells. This observation implies that the orderly cell arrangement is disrupted during the processes of cell differentiation, transformation, and cancer development. This study highlights the significant heterogeneity among cells in cancer tissues, which may exhibit unique genetic traits that distinguish them from normal epithelial cells [34]. Moreover, in an effort to investigate the heterogeneity of epithelial cells during the progression of CC, Liu et al. conducted scRNA-seq on 76,911 cells derived from 13 human cervical malignant tumor tissue samples at various stages. They identified eight distinct epithelial cell subclusters. Notably, Epi7 and Epi8—two epithelial cell subsets—showed specific enrichment in early- and late-stage CESCs and exhibited gene expression profiles associated with active tumor-related pathways, such as the cell cycle, DNA repair, and tumor invasion. Epi8, in particular, expressed genes related to potential immune suppression (indoleamine 2,3-dioxygenase 1, (IDO1)) and inflammation (HLA-DRA, HLA-DPA1, IFIT, and CXCL9-11), suggesting direct communication with T cells through ligand–receptor interactions [20].

4.3.3. Metastatic Niche Adaptation and Immune Evasion Mechanisms

Tumor metastasis, as a crucial biological behavior, is one of the defining hallmarks of cancer [53]. Advanced CC is predominantly characterized by its invasive and metastatic capabilities, which are underpinned by alterations in a complex array of gene expression patterns. Unraveling the metastatic mechanism of CC is of paramount importance for gaining a comprehensive understanding of its development and for devising effective treatment strategies [54]. To investigate the cellular diversity and molecular characteristics underlying CC metastasis, Li et al. performed scRNA-seq analysis on a total of 5 samples from 4 CC patients, including 3 tumor tissues, 1 positive LN, and 1 negative LN. Their findings revealed distinct signaling pathway profiles between nonmetastatic and metastatic tumors. Specifically, nonmetastatic tumors exhibited significant enrichment in cell cycle signaling pathways, such as the G2M checkpoint, E2F target, MYC target V2, and mitotic signaling pathways. In contrast, metastatic tumors displayed upregulation of inflammation-related signaling pathways, including the IL6-JAK-STAT3 signaling pathway and pathways involved in coagulation and inflammatory responses. The authors found that, compared to nonmetastatic tumors, metastatic tumors exhibited a higher infiltration of immune cells, such as macrophages, neutrophils, and NK/T cells, while having fewer epithelial cells [29]. These findings suggest that once tumor metastasis occurs, tumor cells undergo a series of functional changes to adapt to the TME, facilitating immune escape. Moreover, tumor cells from primary tumors and metastatic LNs showed increased copy number variation (CNV) levels in different regions. Metastatic LNs exhibited an early activated TME, characterized by an increase in proliferative T cells, NK cells, and Tregs, as well as a decrease in cytotoxic CD8⁺ T cells and naïve T cells. The authors found that MRC1 expression in metastatic LNs was higher than that in metastatic tumors, leading them to conclude that a high percentage of C1QA + MRC1 macrophages might be associated with tumor metastasis. Additionally, they demonstrated that CAFs in metastatic LNs played prominent roles in antigen presentation, cell adhesion, and immune regulation. These results provide valuable insights into the mechanisms of CC metastasis and pave the way for new therapeutic approaches for CC [29].

4.4.1. Molecular Basis of Chemotherapy Resistance

Gu et al. conducted a detailed analysis of T cells, B cells, and myeloid cells using scRNA-seq and found that several signaling pathways closely associated with chemotherapy resistance, such as the PI3K/AKT and mitogen-activated protein kinase (MAPK) signaling pathways, were enriched across all three cell subpopulations. The PI3K/AKT signaling pathway plays a pivotal role in tumorigenesis, tumor progression, cell survival, and apoptosis. Meanwhile, the MAPK signal pathway, which regulates cell migration, survival, proliferation, and progression, was found to be significantly enriched and upregulated in patients resistant to chemotherapy. In addition, integrins and soluble factors secreted within the TME activate survival pathways involving PI3K/AKT and MAPK, resulting in the elevated expression of antiapoptotic proteins, thereby enhancing cell viability and contributing to drug resistance. The enrichment of the PI3K/AKT pathway in most differentially expressed genes (DEGs) suggests that the activation status of this pathway, which is commonly involved in tumor development, progression, and apoptosis, may be a key driver of chemoresistance in CC patients. Targeting specific components of the PI3K/AKT and MAPK pathways not only has the potential to enhance the efficacy of chemotherapy in CC but may also help overcome drug resistance [32].

4.4.2. Cellular Adaptations Underlying Radiation Resistance

Radiotherapy is the primary radical treatment modality for patients with Stage IB3 to IVA CC and also represents an optional radical therapy for those with Stage IB1 to IB2 CC [3]. However, a significant challenge associated with radiotherapy is the potential development of radioresistance in cancer cells, rendering them less susceptible to ionizing radiation, which ultimately leads to treatment failure in cancer management [57].

Bi et al. used scRNA-seq to analyze three radioresistant CC samples. They found that radiation-resistant conditional reprogrammed (CR) cells, which exhibit radioresistance, predominantly clustered in the G1 phase of the cell cycle. In contrast, most radiation-sensitive cells, which are more susceptible to radiation, were found in the G2 phase. The study revealed that endoplasmic reticulum protein 29 (ERp29) plays a role in promoting endoplasmic reticulum stress–mediated radioresistance. Additionally, oxidative stress induces the monoubiquitination and transcriptional activity of FoxO4 in radioresistant cells, while simultaneously reducing the expression of the deubiquitinase USP7, thereby causing G0/S arrest and preventing radiation-induced cell death. Notably, the level of USP7 was decreased in radioresistant cells, whereas there was no significant difference in the expression of ERp29 [18].