Abstract

Background

   Anxiety disorders (AD), ADHD, and head and neck cancer (HNC) are
   complex conditions with potential genetic interconnections that remain
   to be fully elucidated. The purpose of this study is to investigate
   gene-level connections among ADHD, AD, and HNC.

Method

   A comprehensive literature mining approach identified potential
   gene-disease relationships from PubMed and bioinformatics databases,
   analyzing 19,924 genes. An AI-driven computational process constructed
   a gene-disease relationship table using the Adjusted Binomial Method
   Algorithm (ABMA) to evaluate association reliability. Overlapping genes
   were analyzed through protein-protein interaction (PPI) networks,
   functional annotations, and literature-based pathway analyses to
   elucidate shared and unique genetic mechanisms linking these diseases.

Results

   The analysis identified 141 significant genes associated with AD, 153
   with ADHD, and 1,065 with HNC (q-value < 0.05). These genes
   demonstrated significant overlap (odds ratio ≥ 1.8; p ≤ 2.58E-2) and
   high interconnectivity (PPI network density ≥ 0.39, clustering
   coefficient ≥ 0.76, and diameter ≤ 3). Centrality analysis revealed
   core genes such as IL-6, MYC, NLRP3, and CXCR4 as critical mediators.
   Functional enrichment analysis identified key pathways, including
   serotonergic synapse, inflammatory response, and Toll-like receptor
   signaling, highlighting the involvement of neuronal and immune
   mechanisms. Functional pathway analysis demonstrated reciprocal genetic
   influences among AD, ADHD, and HNC, emphasizing shared and distinct
   gene-level connections that may underlie their co-occurrence and mutual
   risk factors.

Conclusion

   This study reveals a complex and interconnected genetic network among
   AD, ADHD, and HNC, highlighting shared pathways, unique mechanisms, and
   critical genes, providing valuable insights into the genetic
   underpinnings of these conditions and potential avenues for therapeutic
   exploration.

   Keywords: gene-level connections, anxiety disorders, ADHD, head and
   neck cancer, therapeutic exploration

Introduction

   Anxiety disorders (AD) are a group of mental health conditions
   characterized by excessive fear, worry, and related behavioral
   disturbances. Epidemiologically, ADs are highly prevalent, affecting
   approximately 19.1% of adults in the United States in a given year,
   with a lifetime prevalence of around 29% ([31]1). These disorders can
   arise from a combination of genetic, environmental, and psychological
   factors, making understanding their epidemiology crucial for effective
   prevention and treatment strategies ([32]2).

   Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental
   disorder characterized by persistent patterns of inattention,
   hyperactivity, and impulsivity that can interfere with functioning or
   development. ADHD affects approximately 6-7% of children worldwide
   ([33]3), with varying prevalence rates across different regions and
   populations.

   Head and Neck Cancer (HNC) encompasses a diverse group of malignancies
   that arise in the oral cavity, pharynx, larynx, and other related
   structures. Epidemiologically, HNC accounts for approximately 4% of all
   cancers in the United States, with an estimated 54,540 new cases and
   10,780 deaths projected for 2023 ([34]4).

   To clarify terminology, AD in this study refer to the broad spectrum of
   anxiety-related conditions as defined by the Diagnostic and Statistical
   Manual of Mental Disorders (DSM-5), including but not limited to
   generalized anxiety disorder (GAD), panic disorder, separation anxiety
   disorder, and social anxiety disorder. While these subtypes differ in
   their specific symptomatology, they share core features of excessive
   fear, worry, and behavioral disturbances. We use the term AD to
   encompass this spectrum, acknowledging that this inclusive approach may
   obscure subtype-specific nuances but allows for a broader exploration
   of shared genetic mechanisms. Importantly, AD in this context refers
   specifically to Anxiety Disorders and not Alzheimer’s disease, which is
   another psychiatric condition with distinct pathophysiology and
   clinical presentation.

   Similarly, HNC encompasses a diverse group of malignancies arising in
   the oral cavity, pharynx, larynx, sinuses, and salivary glands. This
   includes squamous cell carcinomas, adenocarcinomas, and other
   histological subtypes. By using the term HNC, we aim to capture the
   shared genetic and clinical features across these malignancies,
   recognizing that this broad categorization may overlook
   subtype-specific differences. In this study, gene-disease associations
   were analyzed at the level of full clinical diagnoses rather than
   individual symptoms to ensure consistency across ADHD, AD, and HNC.
   While ADHD is generally treated as a single diagnostic entity, AD and
   HNC encompass multiple subtypes with distinct but overlapping genetic
   underpinnings. While this approach facilitates a comprehensive analysis
   of shared genetic pathways, it also necessitates caution in
   interpreting findings, as the heterogeneity within HNC subtypes may
   influence results.

   Associations have been suggested between ADHD and AD, with ADHD often
   co-occurring with some form of anxiety ([35]5). Genome-wide association
   studies have demonstrated significant genetic correlations between ADHD
   and AD (rg = 0.34), with both conditions sharing genetic risks linked
   to neuroticism (rg = 0.81) and major depressive disorder, suggesting a
   common polygenic architecture that may explain their frequent
   comorbidity ([36]6, [37]7). Mendelian randomization analyses further
   reveal that socioeconomic factors, such as higher educational
   attainment and income, serve as protective influences for both ADHD and
   anxiety disorders, underscoring the role of gene–environment
   interactions in their co-occurrence ([38]8). Together, these findings
   highlight the importance of exploring the shared and unique genetic
   factors underlying these conditions.

   Extending this framework to HNC involves investigating shared
   biological pathways such as neuroinflammation and immune dysregulation.
   In HNC patients, higher pretreatment anxiety levels are significantly
   associated with poorer 2-year overall survival, with tumor response
   mediating this relationship, suggesting that AD may negatively impact
   cancer outcomes ([39]9). Conversely, HNC patients, particularly those
   who have undergone radiotherapy, may develop anxiety and depressive
   disorders ([40]10). Moreover, shared genes have been identified as
   playing roles in all three disorders, including CYP2D6 ([41]11–[42]13).
   Emerging hypotheses propose that systemic inflammation—implicated in
   ADHD and anxiety through genetic variants in pathways like IL6 and
   TNF-α—may also contribute to oncogenic processes in HNC ([43]14,
   [44]15). For instance, chronic inflammation and oxidative stress,
   common in neuropsychiatric conditions, are established drivers of
   carcinogenesis. Although direct genetic links between HNC and
   psychiatric disorders remain underexplored, the overlap in inflammatory
   pathways provides a plausible mechanistic bridge. Additionally, while
   ADHD-associated behaviors (e.g., tobacco use) may elevate HNC risk,
   genetic predispositions to immune dysregulation could further compound
   susceptibility ([45]16). By elucidating these shared mechanisms,
   research may uncover transdiagnostic therapeutic targets and inform
   preventive strategies across neurodevelopmental, psychiatric, and
   oncological conditions.

   This study aims to address the gap in knowledge by exploring gene-level
   connections among these conditions using a computational biology
   approach. We hypothesize that there are significant overlapping genetic
   pathways and core genes that contribute to the co-occurrence and mutual
   risk factors of AD, ADHD, and head and neck cancer. The findings of
   this study could provide valuable insights into the genetic
   underpinnings of these conditions, potentially informing future
   research and therapeutic strategies.

Method

Study workflow

   This study followed a structured multi-step workflow to explore the
   genetic relationships among ADHD, AD, and HNC. First, relevant
   gene-disease associations were retrieved from multiple bioinformatics
   databases and literature sources, including PubMed and the AIC
   Bioinformatics Database (ABD). Next, an AI-driven data processing
   pipeline was applied to filter and refine gene-disease associations,
   ensuring high-quality data for subsequent analysis. Overlapping and
   unique gene sets across the three diseases were identified, and
   statistical enrichment analyses were conducted to assess their
   biological significance. Finally, functional and pathway analyses were
   performed to investigate potential mechanisms linking ADHD, AD, and
   HNC. Details of each step, including data sources, computational
   approaches, and statistical methods, are provided in the subsequent
   sections. The workflow of the current study is depicted in [46]Figure 1
   . To note, as described in the Introduction, AD refers to anxiety
   disorders as defined by DSM-5, and HNC encompasses multiple
   malignancies arising in the head and neck region. This study considers
   ADHD as a single diagnostic entity, while AD and HNC include multiple
   subtypes with shared genetic features.

Figure 1.

   Figure 1
   [47]Open in a new tab

   Workflow diagram of this study. Generated using
   [48]https://bioinfogp.cnb.csic.es/tools/venny/.

Disease gene identification using literature-based mining

   To systematically identify gene-disease associations for ADHD, AD, and
   HNC, we conducted a comprehensive literature mining approach utilizing
   multiple bioinformatics resources. The analysis encompassed
   whole-genome genes (19,924 genes) and aimed to extract relevant
   disease-gene relationships from curated scientific literature.

   First, we employed the Entrez API
   ([49]http://www.ncbi.nlm.nih.gov/Entrez/) to programmatically query
   PubMed ([50]https://pubmed.ncbi.nlm.nih.gov) for relevant publications.
   This automated process retrieved references containing associations