Oral squamous cell carcinoma (OSCC) is one of the most predominant malignant neoplasms of the oral cavity (Ferlay et al., 2021), constituting over 90% of oral cancer cases and ranking as the sixth most prevalent cancers globally. Arising from the epithelial lining of regions such as the buccal mucosa, tongue, mouth floor, and palate, OSCC is associated with substantial morbidity, mortality, and impaired quality of life (Rezazadeh et al., 2023). Based on the GLOBOCAN database (2022), oral cancer accounts for approximately ∼389,846 to 600,000 new cases and ∼188,438 deaths annually, with a significantly higher disease burden observed in South and Southeast Asia (Bray et al., 2024). In Pakistan, OSCC is the most frequently diagnosed cancer in men (12.3%) and the second most common cancer in women (5.3%), contributing to 8.6% of all cancer cases in 2022 (Ikram et al., 2023). In Western regions, including Europe and the United States, OSCC remains a persistent challenge, with over 130,000 and 34,000 new cases annually, respectively (Ghanem et al., 2024).

The etiological factors for OSCC are heterogeneous, with established risk factors such as tobacco use in both smoked and smokeless forms, alcohol consumption, snuff dipping, and betel quid chewing practices that are particularly common in South Asia (Gupta et al., 2016). Alarmingly, in 2022, an estimated 120,000 new oral cancer cases were linked to smokeless tobacco and areca nut use, particularly in South Central Asian regions and low- and middle-income countries (LMICs) (Warnakulasuriya et al., 2021). OSCC significantly influences the patients’ quality of life, leading to functional impairments, emotional distress, and social isolation. Despite developments in treatment techniques, the five-year survival rate for OSCC continues to fall below 50%, with women generally experiencing slightly better outcomes than men (Montero and Patel, 2015). This rising trend underscores the critical importance of developing effective methods for early diagnosis and improved therapeutic interventions. Surgery continues to be the leading treatment for oral cancer but often results in functional and esthetic impairments and requires long-term recovery and support (Dwivedi et al., 2020). Meanwhile, chemotherapy, radiotherapy, and emerging immunotherapeutic approaches are constrained by toxicity, intolerance, and limited efficacy (Raghavi and Anbarasu, 2024). Consequently, the identification of robust molecular biomarkers is critical for improving diagnostic accuracy and therapeutic efficacy.

Recent advances in molecular oncology have underscored the role of kinase gene mutations in the progression of OSCC (Su et al., 2017; Capra et al., 2006). Among these, the NIMA-related kinase (NEK) family has emerged as a key player in regulating a wide array of cellular processes, including apoptosis, DNA damage response, and cell-cycle regulation, particularly active in the S and G₂/M phases, contributing to centrosome separation and mitosis¹³. Dysregulation and overexpression of NEK genes have been linked to poor prognosis, increased tumor aggressiveness, and therapeutic resistance in several cancers, suggesting that NEK genes may function as potential biomarkers and targets for cancer therapy (Cao et al., 2018).

Despite the growing recognition of NEKs in cancer biology, there remains a significant gap in understanding their mutational landscape in OSCC, especially within the Pakistani population. Building on our previous studies on the genetic landscape in oral cancer (Naeem et al., 2025; Nawab et al., 2025), this study aims to fill that gap by profiling mutations in selected NEK family genes using next-generation whole-exome sequencing (NG-WES) of OSCC samples of Khyber Pakhtunkhwa. Furthermore, by applying in silico analyses such as protein modeling and functional annotation, we seek to evaluate the biological impact of these mutations. This research could contribute to the identification of novel prognostic markers and therapeutic targets, ultimately supporting the advancement of precision oncology in OSCC.

For molecular dynamic simulations, five mutations (NEK1 ^p.D409Y , NEK1 ^p.N643K , NEK9 ^p.H174Y , NEK10 ^p.R275C , and NEK10 ^p.E596K ) were prioritized based on their predicted pathogenicity, as identified as deleterious by 5–4 different bioinformatics tools. All these mutations were located at critical interaction sites within the protein structure, suggesting a likely impact on protein stability and function. All the mutant types of the proteins maintained consistently higher Rg values than their WT counterparts throughout the 100-ns simulation, indicating potential structural destabilization (Supplementary Figure S1C–S5C). The average radius of gyration for all the mutations is shown in Supplementary Figure S6. Among the analyzed mutations, NEK1 ^p.D409Y exhibited an average radius of gyration (Rg) of 5.62 nm for the mutant protein and 4.45 nm for the WT. Similarly, for NEK9 (NEK9 ^p.H174Y ) and NEK10 (NEK10 ^p.R275C and NEK10 ^p.E596K ) IS mutations, the average Rg values for the mutant protein were higher than those of the WT (Supplementary Figure S6). RMSD analyses revealed substantial structural deviations in mutant proteins compared to those in WT. All the NEK gene mutants exhibit consistently higher backbone deviations compared to the WT throughout the 100-ns simulation. NEK1 mutations, i.e., NEK1 ^p.D409Y and NEK1 ^p.N643K , showed a substantial deviation, particularly after 20 and 40 ns, as indicated in Supplementary Figures S4A, S5A. NEK9 ^p.H174Y exhibited major deviations after 10 ns and minor deviations after 50 ns (Supplementary Figure S3A). Similarly, NEK10 mutants (NEK10 ^p.R275C and NEK10 ^p.E596K ) exhibited greater structural deviations than WT, indicating reduced conformational stability (Supplementary Figures S1A, S2A).

RMSF profiles further supported these findings. All the mutants exhibited increased flexibility around the mutated residues compared to WT proteins, suggesting increased local flexibility and potential functional disruption (Supplementary Figures S1B–S5B). In addition, other energy parameters (i.e., density, temperature, and pressure) also showed noticeable fluctuations in mutant proteins, supporting the hypothesis of compromised structural stability due to IS mutations. Ramachandran plots were also generated, indicating minor changes in residue conformational distributions (Supplementary Figures S1–S5H,I). All these interacting-site mutations were superimposed and visualized using PyMOL. HOPE analysis further predicted that these substitutions may induce structural destabilization and functional alterations in the proteins, as illustrated in Figure 9.

Identifying reliable genetic biomarkers in oral squamous cell carcinoma remains a central challenge. Despite advances in cancer genomics, the molecular architecture of OSCC has not been fully elucidated, particularly regarding the contribution of underexplored kinases such as the NIMA-related kinase (NEK) family. NEKs are a family of 11 serine/threonine kinases (NEK1–NEK11) emerging as critical regulators of various mitotic processes (cell-cycle regulation, apoptosis, centrosome function, spindle assembly, and the G₂/M transition processes) that are frequently dysregulated in cancer (Nguyen et al., 2023). By maintaining genomic stability and cellular homeostasis, these kinases play an essential role in regulating normal cellular functions. In this study, we present the first comprehensive mutational profiling of all 11 NEKs in OSCC patients from Khyber Pakhtunkhwa (KPK), revealing novel insights into their demographic associations, mutation spectra, and prognostic relevance/significance. Furthermore, we carried out in silico analysis to determine the pathogenic significance and potential effects of the variants on protein function using multiple bioinformatics tools.

Using WES of 31 tumor tissues, we identified multiple somatic and germline variants (36/46; 78.26% and 10/46; 21.74%) in NEK genes, several of which are predicted to be pathogenic (NEK1, NEK2, NEK5, NEK9, NEK10, and NEK11 mutations are deleterious) based on consensus outputs from SIFT, PolyPhen-2, and other pathogenicity tools, as reported above. Among these genes, NEK1 (14/46; 30.43%), NEK5 (6/46; 13.04%), NEK10 (7/46; 15.21%), and NEK11 (5/11; 10.86%) showed the highest mutation burden. These findings highlight the potential oncogenic relevance of NEKs in OSCC. Our results are in line with their previously reported oncogenic roles in other cancer types, including ovarian, lung, brain, breast, and colorectal cancers (Fry et al., 2012; Chen et al., 2025; Tsunoda et al., 2009). Previous literature reports that NEK10 has been identified as one of the important kinases likely harboring driver mutations, based on cancer whole-genome sequencing datasets, with 13 cataloged missense variants across multiple tumor types (Greenman et al., 2007). Our study also supports this finding, revealing a high frequency of NEK10 mutations in oral squamous cell carcinoma patients from Pakistan. The non-synonymous SNVs (NEK10 ^p.R275C and NEK10 ^p.E596K ) were reported as the most deleterious mutations in NEK10, which occur at the interacting site and have a destabilizing impact on protein structure. Similarly, NEK1 mutations have been documented in diverse cancers such as ovarian, colorectal, lung, and skin tumors, where NEK1 dysfunction is linked to chromosomal instability and defective DNA damage checkpoint phenotypes that promote malignant transformation in both cellular and animal models (Chen et al., 2011). NEK11 has also been frequently implicated in somatic alterations across various cancers. NEK11 exhibits both oncogenic and tumor-suppressive functions across different cancer types (Li H. et al., 2025). Exome sequencing studies in melanoma and colorectal cancers have identified several mutations in NEK11. Functional analyses further reveal that loss of NEK11 disrupts the G₂/M cell-cycle checkpoint in response to DNA damage, leading to heightened genomic instability and reduced apoptosis, thereby potentially promoting tumor progression (Sabir et al., 2015). However, NEK11 ^p.E347V was one of the most common mutations in our population, which was frequently reported in all 31 OSCC patients.

NEKs play key roles in regulating mitotic progression, centrosome dynamics, spindle formation, and genomic stability. Dysregulation of these processes is a hallmark of cancer, and alterations in NEK genes have been linked to abnormal proliferation, chromosomal instability, and enhanced metastatic behavior in multiple malignancies. Mutations identified in this study, particularly those predicted as damaging, may impair catalytic activity or substrate binding, disrupting cell-cycle checkpoints and mitotic control. These changes can also interact with key OSCC pathways such as PI3K/AKT, MAPK, and DNA damage response, underscoring the role of NEKs in tumorigenesis and as potential biomarkers (Fry et al., 2012; Moniz et al., 2011; Xie et al., 2020). The inclusion of TCGA-HNSCC data places our findings within a broader genomic context and supports the relevance of the identified genes in head and neck cancers. Although mutation frequencies were comparatively higher in our cohort, TCGA analysis confirms that these genes are not unique to our dataset but are recurrently altered in larger populations. Given the exploratory nature of this study and the modest sample size, these observations should be interpreted cautiously and warrant validation in larger, independent cohorts.

Furthermore, the convergence of computational predictions strengthens the conclusion that these mutations are likely impactful in OSCC pathogenesis. All these genes have destabilized proteins. Nearly 41% (19/46) coincide with protein–protein interaction sites, and key mutations in NEK1 (NEK1 ^p.D409Y and NEK1 ^p.N643K ), NEK9 (NEK9 ^p.H174Y ), and NEK10 (NEK10 ^p.R275C ) occur in highly conserved, functionally critical regions, with scores of 8 and 9. Using the STRING server and GeneMANIA, we analyzed the interaction networks for these genes (Figure 13; Supplementary Figure S7). The resulting interactome pathways reveal that these proteins participate in multiple biological processes associated with genome integrity, and any mutation in them may disrupt these interactions, ultimately contributing to tumor progression. These in silico findings extend observations of NEKs influencing mitotic regulation, DNA repair, and signal transduction pathways (Zhu et al., 2024).

Recent advances in computational and deep learning methods have enhanced mutation detection, functional impact prediction, and biomarker prioritization in cancer genomics. For instance, heterogeneous information network learning models with neighborhood-level structural representation predict biologically relevant interactions, such as lncRNA–miRNA networks (Zhao et al., 2024), whereas geometric deep learning frameworks enable drug repositioning over heterogeneous networks (Zhao et al., 2022). Similarly, sequence-based deep learning models accurately predict functional sites, including HIV-1 protease cleavage sites (Li D. et al., 2025), and structure-based and molecular modeling advances, such as fine-tuned MM/PBSA (GBSA)-based methods and multi-perspective 3D molecular representations, improve variant interpretation and clinically relevant biomarker identification (Zhang et al., 2025; Zhao et al., 2025).

Although the present study employed traditional in silico tools to assess mutation impact and protein–protein interactions, recent advancements in machine learning and deep learning are revolutionizing the interpretation of cancer-associated mutations and network-level analysis. Ensemble approaches, such as rotation forest models, enhance accuracy in disease–gene association predictions by integrating multi-feature similarity profiles. Applying analogous ensemble or deep learning strategies to NEK gene mutations could provide enhanced insights into mutation pathogenicity, protein–protein interaction disruptions, and network-level effects that are not fully captured by conventional methods (Guo et al., 2019).

Furthermore, current trends in computational PPI prediction, propelled by modern machine learning and deep learning frameworks, reflect a transition from traditional sequence-based tools such as ISPRED-SEQ. Modern approaches integrate structural and evolutionary protein features to more accurately assess mutation-induced interaction changes. Methods including AlphaFold Multimer, graph neural network (GNN)-based PPI predictors, and transformer-based protein language models have shown improved performance in identifying interaction interfaces and evaluating the functional consequences of variants (Wang Y. et al., 2020; Wang et al., 2019). These transformer-based models learn contextual relationships across protein sequences, enabling detection of long-range dependencies and subtle sequence–structure effects that are often missed by conventional tools. Incorporating such approaches into future analyses of NEK gene mutations in oral cancer may enhance the identification of critical PPI disruptions, support prioritization of variants for experimental validation, and strengthen the interpretability of in silico findings.

Additionally, recent deep learning approaches have significantly enhanced drug–target interaction (DTI) prediction and hold relevance for future NEK druggability assessment. LSTM-based neural networks and transformer models capture complex protein–ligand interactions, predicting potential small-molecule binders more accurately than traditional docking or similarity-based tools. These methods could prioritize selective NEK inhibitors and guide therapeutic development, as supported by recent studies on LSTM-based DTI frameworks that underscore the benefits of integrating such models into NEK-focused drug discovery (Wang Y.-B. et al., 2020).

The somatic mutational profile of cancers is strongly influenced by inherited genetic predispositions and the distribution of allele frequencies within specific populations. In regions characterized by rich ethnic diversity, such as South Asia, these factors become particularly significant. Our cohort from KPK reflects a genetically distinct ethnic community with unique hereditary backgrounds and characteristic lifestyle exposures, including the widespread use of naswar, i.e., smokeless tobacco, smoking, and betel nut consumption (IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, World Health Organization and International Agency for Research on Cancer, 2004; Ferlay et al., 2020; India Project Team of the International Cancer Genome Consortium, 2013). These carcinogenic factors are known to cause DNA damage and may potentially alter the pattern and frequency of mutations in NEK genes (Sharan et al., 2012). We observed a higher incidence of OSCC in men (71%) and individuals older than 56 years (58%), which is consistent with global trends of OSCC prevalence in men and older age groups. Notably, naswar use emerged as the primary risk factor (58%), while a significant subset of patients also presented with dental issues (32%). These findings align with prior studies linking smokeless tobacco and poor oral hygiene to OSCC risk, reinforcing the role of these exposures in regional populations (Chen et al., 2025). Furthermore, NEK4 and NEK9 mutations demonstrated a statistically significant association with tumor site (p = 0.02) and tobacco intake (p = 0.01), suggesting a potential site-specific oncogenic role of NEK4 and a strong link between NEK9 mutation and tobacco exposure in OSCC. These associations may aid in understanding the molecular mechanisms of NEK gene dysregulation and their relationship with clinicopathological features in oral cancer.

Furthermore, survival analysis revealed distinct trends across the NEK gene family. The observed differences in overall survival between NEK-mutant and WT OSCC patients likely reflect the context-dependent dual roles of NEKs in cancer. Previous studies indicate that NEKs can function as either oncogenes or tumor suppressors depending on cellular context and mutation type (Nguyen et al., 2023; Chen et al., 2023; Anuraga et al., 2021). Our findings that NEK10-mutant patients exhibited a significantly better overall survival (p < 0.05), whereas mutations in NEK1, NEK2, NEK5, and NEK9 showed non-significant trends toward improved survival, and mutations in NEK4 and NEK8 showed a trend toward worse survival are consistent with a context‐dependent dual role of NEKs in tumorigenesis. NEK4 generally acts as a tumor suppressor involved in DNA damage response and cell-cycle checkpoint control; thus, its mutation may impair genomic stability and contribute to poorer survival (Park et al., 2016). Similarly, NEK8 is reported to be overexpressed in\ gastric, colorectal, and breast cancers, where it promotes proliferation and is linked with poor prognosis (Cao et al., 2023; Gao et al., 2022). In contrast, NEK1, NEK2, NEK5, NEK9, and NEK10 often exhibit oncogenic functions, promoting centrosome amplification, mitotic progression, and proliferation; loss-of-function mutations in these genes may weaken their pro-tumorigenic potential, which could explain the better survival of patients with mutant forms (Zhu et al., 2024; Yang et al., 2022). In particular, NEK10 regulates G₂/M checkpoint and MAPK signaling, and its impaired function may reduce tumor growth potential, aligning with the improved survival seen in our cohort (Moniz et al., 2011). Collectively, these findings suggest that NEK mutations may contribute to heterogeneous clinical outcomes in oral cancer and warrant further validation in larger patient cohorts to clarify their prognostic relevance.

Based on our in silico findings, the druggability of NEK family mutations remains largely unexplored due to the limited availability of selective NEK inhibitors in clinical practice. These mutations are located within or near conserved kinase domains, suggesting a potential impact on catalytic activity or PPIs. However, their presence may modulate cellular signaling pathways, influencing tumor cell sensitivity or resistance to broader classes of kinase inhibitors. Therefore, understanding the functional consequences of these mutations is critical, as they may either serve as future drug targets or act as biomarkers for treatment stratification and therapeutic response. NEK2 is consistently overexpressed across all types of gastrointestinal cancers, including gastric and pancreatic tumors, and contributes to tumor progression and drug resistance. Currently, T-1101 tosylate, emerging as the most advanced NEK2 inhibitor, has shown promising preclinical results (Xia et al., 2025). Similarly, NEK10, as a potential tumor-promoting factor in lung adenocarcinoma, could serve as a new therapeutic target and warrant further investigation in OSCC (Dutt et al., 2024).

In conclusion, this study provides the first comprehensive mutational profiling of all 11 NEKs in OSCC patients from KPK, revealing novel insights into their mutation spectrum, demographic associations, and potential prognostic implications. Recurrent alterations were most frequent in NEK1, NEK9, NEK10, and NEK11, with some localizing to functionally relevant regions. Computational analyses suggest possible effects on protein stability and interactions, though these require experimental confirmation. Observed associations with clinicodemographic factors (naswar use and tumor site) and survival (NEK10 with improved outcomes) are preliminary and limited by cohort size. Overall, our findings expand the knowledge of NEK alterations in OSCC and underscore population-specific patterns, but validation in larger cohorts and functional studies is essential to elucidate mechanistic roles and clinical utility.

Furthermore, this study has certain limitations. The relatively limited sample size posed challenges in establishing robust correlations between molecular findings and clinicopathological parameters. In addition, the study primarily relies on WES and in silico analyses without experimental validation. Although in vitro or in vivo functional assays would provide mechanistic confirmation of the biological impact of NEK gene mutations, such analyses were beyond the scope of the present work. Future studies involving larger cohorts from the same population, along with functional validation experiments, are needed to elucidate the mechanistic contributions, prevalence, and clinical relevance of NEK mutations in OSCC.