Microwave-Synthesized Multicomponent Nanoparticles for Enhanced Proton Therapy Efficacy in Non-Small Cell Lung Cancer: Integration of NGS-Based Next-Generation Sequencing and Single-Cell Transcriptomic Database Analysis for Target Identification and Acute

Abstract

Background and Rationale: Non-small cell lung cancer (NSCLC) remains the most prevalent and lethal form of thoracic malignancy worldwide, accounting for approximately 85% of all lung cancer diagnoses. Despite advances in proton therapy, targeted therapies, and immunotherapy, clinical outcomes remain suboptimal due to intratumoral heterogeneity, therapy resistance, and off-target systemic toxicity. The rational design of novel radiosensitizing nanoparticle-drug combinations demands a deep understanding of NSCLC's molecular landscape — a gap that large-scale genomic and single-cell transcriptomic datasets are uniquely positioned to bridge.

Objectives: This study pursues three integrated objectives: (1) to perform comprehensive bioinformatics analysis of publicly available NGS and single-cell RNA sequencing (scRNA-seq) datasets to identify key molecular targets, cell-type-specific vulnerabilities, and tumor microenvironment (TME) signatures in NSCLC; (2) to synthesize a library of multicomponent nanoparticles — including CuO, ZnO, superparamagnetic Fe₃O₄-decorated hexagonal boron nitride (h-BN) nanosheets, Ni-Cu alloys, and Ag-doped lanthanum manganite (Ag:LaMnO₃) — using microwave synthesis; and (3) to evaluate the anticancer selectivity and acute systemic toxicity of over 100 novel combinations of these nanoparticles with FDA-approved chemotherapeutic (Gemcitabine, Cisplatin, Carboplatin, Paclitaxel) and targeted agents (Tepotinib, Osimertinib, Rybrevant) alongside alkali metal salts (RbCl, CsCl, Rb₂CO₃, Cs₂CO₃) in the context of proton therapy enhancement.

Bioinformatics and Database Analysis: Large-scale transcriptomic profiling was performed by mining publicly available repositories including TCGA-LUAD/LUSC (The Cancer Genome Atlas), GEO (Gene Expression Omnibus, datasets: GSE131907, GSE117570, GSE189357), and CELLxGENE single-cell atlas. Raw NGS data (bulk RNA-seq and whole-exome sequencing) were processed through standardized pipelines (STAR aligner, DESeq2, GATK) for differential gene expression (DEG) analysis, somatic mutation profiling, and copy number variation (CNV) assessment. Single-cell RNA-seq datasets comprising >200,000 cells from NSCLC patient tumors and adjacent healthy tissues were re-analyzed using Seurat and Scanpy frameworks. Cell clustering, trajectory inference (Monocle3), and regulon analysis (pySCENIC) revealed transcriptionally distinct malignant subpopulations, cancer-associated fibroblast (CAF) niches, and immunosuppressive macrophage states within the TME. Pathway enrichment (GSEA, KEGG, Reactome) identified upregulated oxidative stress, metal-ion homeostasis, and DNA damage response (DDR) pathways as primary vulnerabilities amenable to nanoparticle-mediated radiosensitization. Molecular docking and network pharmacology analysis (STRING, Cytoscape) were used to map the polypharmacological interactions of drug-nanoparticle combinations against NSCLC driver targets (EGFR, MET, KRAS, TP53, ALK).

Materials and Methods: Nanoparticles were synthesized via microwave-assisted hydrothermal routes and characterized by XRD, Raman spectroscopy, and FTIR. In vitro anticancer activity was assessed by MTT assay and Annexin V-FITC/PI flow cytometry on A549 (NSCLC) vs. NHDF (healthy) cell lines to compute selectivity index (SI). In vivo acute toxicity was evaluated using a dual-model platform: (i) chicken embryo chorioallantoic membrane (CAM) assays with non-invasive photoplethysmography (PPG) cardiac monitoring, and (ii) a Wistar rat behavioral maze protocol incorporating blood pressure, reactive oxygen species (ROS), oxygen saturation, and body temperature measurements — quantified via a composite Acute Toxicity Index (ATI).

Results: scRNA-seq re-analysis identified a transcriptionally aberrant NSCLC subpopulation characterized by elevated MT1G/MT2A (metallothionein), HMOX1 (heme oxygenase-1), and SLC31A1 (copper transporter) expression, providing mechanistic rationale for the preferential cytotoxicity of CuO and ZnO nanoparticles in malignant vs. healthy cells. DEG analysis from TCGA confirmed overexpression of boron transporter genes (SLC6A12, SLC4A11) in LUAD, supporting the use of ¹⁰B-enriched h-BN nanosheets for proton/neutron capture therapy. Experimentally, the multicomponent formulations achieved a 3–7-fold increase in selectivity index compared to standard clinical drugs. ZnO-containing combinations demonstrated ~20% superior selectivity versus CuO-based equivalents. All 50 primary combinations exhibited lower acute toxicity in embryo models than their individual clinical counterparts (ATI reduction: 1.5–1.8×). Formulations remained physicochemically stable (< 3% sedimentation) for 9 months at +4°C. Magnetic nanoparticles (Ni-Cu, Ag:LaMnO₃) synthesized with Curie temperatures of 41–45°C demonstrated suitability for self-regulated magnetic hyperthermia, synergizing with the proton beam at reduced energies (70 MeV vs. standard 140–240 MeV).

Conclusions: This study establishes a novel NGS- and scRNA-seq-guided nanoparticle design framework for NSCLC-targeted proton therapy enhancement. Single-cell and bulk transcriptomic analysis from large public databases provided mechanistic, target-level justification for the superior performance of CuO, ZnO, and h-BN nanoparticle formulations. The integrative approach — linking genomic vulnerability mapping with multicomponent synthesis, in vitro/in vivo toxicity testing, and proton physics — represents a significant methodological advance toward precision nano-radiotherapy in NSCLC. The proposed combinations offer a cost-effective, repurposing-based pathway with a Therapeutic Value (TV) 8.7–15.2× higher than standard-of-care monotherapy.