This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Integrating Human Proteomes with Genome-Wide Association Data Reveals Prioritized Therapeutic Candidates for Lung Squamous Cell Carcinoma
1
School of Life Sciences and Medical Engineering, Anhui University, Hefei 230601, China
2
Institutes of Physical Science and Information Technology, Anhui University, Hefei 230601, China
*
Author to whom correspondence should be addressed.
Biology 2025, 14(12), 1640; https://doi.org/10.3390/biology14121640
Submission received: 18 October 2025
/
Revised: 19 November 2025
/
Accepted: 20 November 2025
/
Published: 21 November 2025
(This article belongs to the Section Cancer Biology)
Simple Summary
Lung squamous cell carcinoma (LUSC) represents a clinically challenging subtype of non-small-cell lung cancer, where targeted treatment options remain limited. To address this unmet need, we systematically investigated the causal roles of blood plasma proteins in LUSC development by integrating genetic data from large-scale studies. Using a Mendelian randomization (MR) approach that leverages natural genetic variation as instrumental variables, we identified twelve circulating proteins whose levels are causally linked to LUSC risk. Among these, five proteins (including DOK2, FKBPL, NCF2, PDIA3, and TCL1A) exhibited strong evidence of shared genetic architecture with LUSC. We further classified these candidates into prioritized tiers based on colocalization, differential gene expression, interaction networks, and druggability profiles. In addition, mediation analysis revealed that modifiable risk factors, such as smoking and physical activity, partly influence the protein–LUSC relationship. Our study provides a genetically validated roadmap for targeting novel therapeutic candidates and supports the integration of lifestyle interventions in LUSC prevention and treatment strategies.
Abstract
Lung squamous cell carcinoma (LUSC) is the second most prevalent type of lung cancer worldwide. Despite its global health burden, the molecular mechanisms driving LUSC remain poorly characterized, posing considerable challenges for the development of targeted preventive therapies. Here, by integrating human plasma proteomes (n = 54,219) with GWAS summary data for LUSC (7426 cases and 55,627 controls), we performed genome-wide Mendelian randomization (MR) and colocalization analyses to identify potential druggable targets associated with LUSC risk. After applying Bonferroni correction, sensitivity analyses, and reverse causation detection, we identified 12 potential druggable proteins significantly associated with LUSC risk. Five of these proteins (DOK2, FKBPL, NCF2, PDIA3, and TCL1A) showed strong evidence of colocalization. Furthermore, protein–protein interaction (PPI) networks and druggability assessments were used to refine therapeutic target selection. Additionally, mediation analyses were performed to elucidate the mediating effects of modifiable risk factors on the relationship between plasma proteins and LUSC risk, and we identified 14 modifiable risk factors that could mitigate LUSC risk through targeted interventions. More importantly, we stratified the 12 proteins into four tiers based on colocalization, differential expression, PPI networks, and druggability potential. Notably, DOK2 emerged as a Tier 1 target, while FKBPL, NCF2, AXL, and PDIA3 were classified as Tier 2 targets, representing promising candidates for further drug development. Overall, we identified 12 proteins with druggable potential associated with LUSC risk and demonstrated how modifiable risk factors mediate these associations. These findings advance our understanding of LUSC etiology and provide a foundation for developing targeted therapeutic strategies while emphasizing the importance of addressing modifiable risk factors in both prevention and treatment efforts.
Keywords:
lung squamous cell carcinoma; drug target; Mendelian randomization; pQTL; GWAS
