AbstractNon-small cell lung cancer (NSCLC) represents a heterogeneous group of malignancies characterised by diverse histological and molecular features. Some NSCLCs, particularly adenocarcinomas, harbour genomic alterations in receptor tyrosine kinases or downstream RAS/RAF signalling pathways, which are targets of effective therapies. NSCLCs lacking actionable genomic alterations often benefit from immune checkpoint inhibitors, though only a minority of patients achieve long-term survival. These tumours often carry alterations in tumour suppressor genes like TP53, KEAP1, STK11, or NF1, for which pharmacological strategies are still under investigation. This review explores emerging therapeutic opportunities unveiled by multi-omics studies in NSCLCs without actionable genomic alterations. Proteogenomic approaches—integrating genomic, transcriptomic and proteomic data—enable a comprehensive understanding of NSCLC molecular landscapes and signalling network dysregulation, helping to identify distinct tumour subtypes and potential therapeutic targets. These tumours exhibit alterations in cell cycle regulation, DNA repair, immune signalling, epigenetic modulation and metabolic and redox pathways. Although therapies targeting tumour suppressor genes like p53 remain highly anticipated, extending our understanding of the broader molecular landscape in these tumours may reveal novel vulnerabilities and inform the development of novel drugs or combination strategies. This could further advance precision oncology for NSCLC.IntroductionNon-small-cell lung cancer (NSCLC) is the leading cause of cancer-related death worldwide [1]. It encompasses several histological types, including adenocarcinoma (LUAD), squamous cell carcinoma (LUSC), large cell carcinoma and rarer types [2]. NSCLC has a high rate of somatic mutations [3], contributing to its biological complexity and treatment resistance.With advancements in understanding 'oncogene addiction', molecular profiling has become essential for identifying actionable genetic alterations [4,5,6,7,8]. These include several types of EGFR mutations (present in 10–20% of Caucasian populations and up to 50–60% of Asian populations [9, 10]), rearrangements involving ALK, ROS1, RET, NTRK, or NRG1, MET exon 14 skipping mutations and BRAFV600E, KRASG12C, or HER2 mutations. Other driver alterations under clinical investigation are HER2 and MET amplifications. LUAD exhibits a higher prevalence of actionable genomic alterations compared to other histological types, and targeted therapies often represent the first-line treatment when these alterations are detected [11].Tumours with actionable driver alterations represent about 25–30% of NSCLC cases and up to 60% of LUAD [12]. For tumours lacking such mutations, treatment is typically guided by the expression of Programmed Death-Ligand 1 (PD-L1), a marker predictive of response to immune checkpoint inhibitors (ICIs). Despite the advancements in immunotherapy, many patients do not respond or eventually progress due to primary or secondary resistance. Long-term response rates in the range of 20% after first-line immunotherapy alone [13] and 20–30% after chemo-immunotherapy [14, 15] highlight the need to develop additional treatments.The genomic characterisation of NSCLC has been pivotal in identifying therapeutic targets, which have been validated by the clinical efficacy of matched targeted therapies, paving the way for precision oncology [16]. Nonetheless, this approach has notable limitations. Certain genetic alterations remain undruggable, and even when a targetable mutation is present, the corresponding therapy does not always yield a clinical benefit. Additionally, not all genomic changes result in phenotypic consequences, and other molecular mechanisms may play critical roles in tumour development and progression [17,18,19].In recent years, substantial advances in omics technologies, including transcriptomics, proteomics, phosphoproteomics, metabolomics, epigenomics and others, have expanded our capacity to investigate the molecular complexity of cancer [20]. By analyzing multiple molecular layers, these approaches offer complementary perspectives, and their integrative application can allow a more comprehensive characterisation of the biomolecular alterations driving tumorigenesis and the identification of novel therapeutic targets [17, 19, 21].This review article examines the molecular alterations discovered through multi-omics studies in NSCLC, with an emphasis on proteogenomic studies, focusing on tumours lacking known actionable genomic alterations and highlights the potential of multi-omics approaches as research tools to identify new therapeutic targets. Data on tumours with actionable genomic alterations are reported only when they pertain to potential targets not yet utilised in clinical practice. We then provide a brief overview of emerging therapeutic strategies designed to target alterations in tumour suppressor genes and conclude by discussing the challenges associated with integrating multi-omics data, while proposing a potential roadmap for their clinical validation.A systematic literature search was conducted on Pubmed up to April 2025, using various combinations of the following terms, both as free-text and MESH terms: 'Lung Neoplasms', 'Carcinoma, Non-Small-Cell Lung', 'Adenocarcinoma of Lung', 'Carcinoma, Squamous Cell', 'Carcinoma, Large Cell', 'Genomics', 'Transcriptome', 'Proteogenomics', phosphoproteomic (not MESH). Additional material was sought on Scopus, Google Scholar and through manual review of reference lists from relevant articles. The review focused on studies containing original omics data. Articles limited to bioinformatic analyses from public databases, or addressing individual or small sets of genes, or focused specifically on response or resistance to existing therapies, were excluded.NSCLC genomic profilingComprehensive genomic profiling using next-generation sequencing, entailing either a wide targeted gene panel or whole exome sequencing (WES) or whole genome sequencing (WGS), has uncovered a variety of somatic genomic alterations in NSCLC [22,23,24,25,26,27]. They show significant differences not only among histological types [9, 28] but also according to ethnicity [9, 10, 29], gender [30] and smoking history [25]. It is important to note that the criteria for defining significantly mutated genes can vary across studies, complicating direct comparisons. Furthermore, identifying cancer-related genes is particularly challenging in tumours with a high tumour mutational burden (TMB), where the majority of genomic alterations are passengers rather than drivers.Genomic profiling of lung adenocarcinomaIn primary LUAD, driver mutations interest both oncogenes, many of which have clinical relevance being druggable, and tumour suppressor genes [23,24,25, 27, 28]. Among the first are mutations of receptor tyrosine kinases (RTK) [31] from families like ERBB/EGFR, FGFR and MET. Other RTKs, such as ALK, RET, ROS1 and NTRK, are altered by translocations. Further driver oncogenes belong to the downstream signalling pathways and include KRAS, BRAF and PIK3CA. Altered tumour suppressor genes include TP53, KEAP1, STK11 and NF1, among others. Mutations also occur in chromatin-modifying genes, such as ARID1A, ARID2, SETD2, SMARCA4 and MLL3 (KMT2C), leading to epigenomic abnormalities that overlap with genomic alterations, further complicating lung cancer biology. Table 1 provides a list of genes significantly mutated in NSCLCs, described in the work of Campbell et al. [28] and found in OncoKBTM [32, 33], along with potential therapeutic implications reported by OncoKBTM. OncoKBTM provides the levels of evidence for altered genes as biomarkers predicting responsiveness to specific drugs in one or more types of neoplasm; however, this does not necessarily imply that the drug is recommended for clinical use in the specific context being considered. Genes encoding proteins already actionable in clinical practice are not shown. Table 2 provides a list of translocations found in NSCLC, reported by Campbell et al. [28].Table 1 Significantly mutated genes in lung adenocarcinoma, lung squamous cell carcinoma and lung adenocarcinoma without actionable mutations and potentially active targeted drugs.Full size tableTable 2 Fusion genes in lung adenocarcinoma and lung squamous cell carcinoma listed by frequency (only fusions found in ≥5 tumour samples are reported).Full size tableLarge-scale copy number alterations (CNAs) may affect several chromosomal arms [22] and some cases of chromothripsis are reported [22, 27]. Amplifications most often involve NKX2-1 (encoding the Thyroid Transcription Factor 1—TTF1), MYC, TERT, MCL1, while deletions mainly affect CDKN2A, B2M, SMAD4 [22, 27, 28] (Table 3).Table 3 Cancer genes focal amplifications and deletions in lung adenocarcinoma, lung squamous cell carcinoma and lung adenocarcinoma without actionable mutations and potentially active targeted drugs.Full size tableA negative correlation usually occurs between alterations leading to activation of the same pathway, such as those affecting different RTKs, or RTKs and KRAS/STK11, or ATM and TP53 [23, 27, 28], or MYC and MGA [27]. When mutations of KRAS coexist with those in EGFR, they tend to confer resistance to EGFR inhibitors [34]. Co-occurrences have also been described, i.e., between MET amplifications and NF1 mutations, or STK11 and KRAS mutations [28], or MDM2 and CDK4 amplifications. The latter has been suggested to potentially benefit from combined treatment with MDM2 and CDK4 inhibitors [35].Figure 1a, b represent the distribution of the main genomic alterations in LUAD, with the main co-mutations.Fig. 1: Main genomic alterations in primary lung adenocarcinoma.a Counts and co-occurrences of the main genomic alterations in primary lung adenocarcinoma. Downloaded from cBioPortal [189, 190], created from the TCGA PanCancer Atlas case series [27]. Some mutations are grouped by pathway: KEAP1 and NFE2L2; KRAS, BRAF and NF1; RTKs including mutations or fusions of EGFR, ERBB2, MET, ALK, NTRK1-3, RET and ROS1. b Euler diagram representing the percentages of occurrences and co-occurrences of the main genomic alterations in primary lung adenocarcinoma. The diagram (created with R [191], package 'ggvenn') shows the percentages of genomic alterations affecting the most commonly altered genes in lung adenocarcinoma, either with or without concomitant alterations in other frequently altered genes. RTKs include mutations or fusions of EGFR, ERBB2, MET, ALK, NTRK1-3, RET and ROS1. Data are taken from the TCGA PanCancer Atlas case series of lung adenocarcinoma; out of 566 patients in that case series, 83 patients whose tumours do not have any of the considered gene alterations are excluded from the figure. The percentages reported are rough estimates, because the Euler diagram cannot always include all overlapping alterations when considering multiple subgroups.Full size imageThe pathways most affected by genomic alterations in LUAD include: the RTK/RAS/RAF pathway (76% of cases), the PI3K-mTOR pathway (25%), p53 pathway (63%), cell cycle pathways (64%), the oxidative stress pathway (22%) and chromatin remodelling and RNA splicing pathways (49%) [27].Smoking history plays a crucial role in shaping the mutational landscape of LUAD. Smokers tend to have higher TMB and an increased rate of cytosine-to-adenine nucleotide transversions [24, 25, 27]. Smoking-related tumours represent around 70% of lung cancers and appear to decline in parallel with the smoking habit reduction in Western world. However, around 30% of LUAD occur in never smokers and are considered a different disease with a definite natural history and a specific treatment. LUAD in never smokers are more common in East Asia and represent the fifth cause of cancer related deaths worldwide, being more common in young females without smoking exposure. It is considered a different and emerging disease, characterised by oncogene addiction because of somatic mutations, chromosomal rearrangements, increased gene copy number, or gene deletions [36]. Mutations in tumour suppressor genes, including TP53, STK11, KEAP1, NF1, SMARCA4, as well as mutations in KRAS, are enriched in smokers. Alterations in RTKs and PIK3CA are enriched in non-smokers [24, 27]. The frequency of EGFR mutations is higher, and KRAS and BRAF mutations are lower, in Asian populations compared to Caucasians [29].About 62% of the 'The Cancer Genome Atlas' (TCGA) LUAD samples had genomic alterations in known driver oncogenes of the RTK/RAS/RAF pathway, including EGFR, ERBB2, MET, ALK, ROS1, RET, KRAS, NRAS, HRAS, BRAF and MAP2K1 (MEK1). These have been collectively called 'oncogene-positive' tumours. Another 14% showed amplifications of ERBB2 or MET, or mutations in RIT1 or NF1, eventually leading to activation of the same pathway [27]. Overall, 70–80% of LUAD exhibit alterations in the RTK/RAS/RAF pathway. Common and rare alterations affecting this pathway, along with their therapeutic implications, have been described in excellent reviews [8, 37]. Tumours without such alterations, termed 'oncogene-negative', often have loss-of-function mutations in tumour suppressor genes like TP53, KEAP1 and NF1 [27].Lung adenocarcinoma without actionable genomic alterations within the RTK/RAS/RAF pathwayIn an expanded LUAD cohort of TCGA, 15 genes were found to be significantly altered by WES in 'oncogene-negative' tumours [28]. These include regulators of RAS and Rho kinase functions, such as SOS1, RASA1, VAV1 and ARHGAP35, along with amplifications involving FGFR1/WHSC1L1, PDGFRA/KIT/KDR and MAPK1 (ERK2). Overall, 20–30% of LUADs lacked RTK/RAS/RAF pathway alterations (RPA) at WES and were termed RPA(-)E cases [38]. WGS identified RPA in 33% of these tumours, which were missed by WES due to technical challenges or low tumour purity [38]. These included KRAS mutations, amplifications of ARAF, EGFR, MAPK1 and SOS1, deletions of RASA1 and NF1 and NRG1 fusions/mutations.Among the remaining RPA-negative cases at WGS, or RPA(-)G, mutations in tumour suppressor genes such as TP53, STK11, KEAP1 and SMARCA4 were identified, along with mutations in ESR1, BLM and FOXO3, deletions in SETD2 and amplifications involving NKX2-1, KAT6A, CCNE1, MDM2, MYC, MCL1 and MYCL. In noncoding regions, including promoters, enhancers and transcription factor-binding sites, mutations were found near ILF2, CUL2 and TSN. Mutations in the promoter of ILF2, leading to its overexpression, can affect DNA repair and resistance to DNA-damaging agents [39]. RPA(-)G tumours exhibited a high TMB. They also displayed complex structural variants [38], leading to gene amplification and overexpression.Overall, deep genomic characterisation of RPA(-)G LUADs shows that they are heterogeneous but share some common biological features, including enrichment for TP53, KEAP1 and SMARCA4 mutations and a high TMB. It remains unclear whether these tumours represent a distinct biological entity or rely on rarer mechanisms (e.g., epigenetic) of activation of the RTK/RAS/RAF pathway. Candidate drivers include ILF2 mutations, amplifications of genes acting downstream of RAS/RAF, such as MYC, or loss of tumour suppressors [38].Rare fusions involving genes, such as EGFR, FGFR, MET, HER2, BRAF, NRG and others were also observed in a small subset of treatment-naïve NSCLC cases, potentially offering new therapeutic options [40].Figure 2 represents the gene alterations in the pathways more frequently altered among LUADs in the TCGA case series [27], considering only cases without actionable driver genomic alterations in the RTK/RAS/RAF pathway.Fig. 2: Gene alterations in the pathways more frequently altered in primary lung adenocarcinoma lacking targetable driver alterations in the RTK/RAS/RAF pathway.Figure downloaded from cBioPortal, representing data from the TCGA PanCancer Atlas case series, excluding cases with mutations in EGFR, ERBB2, MET, KRAS, BRAF, or fusions of ALK, NTRK1-3, RET, ROS1 (any alterations reported for these genes are represented exclusively by CNAs).Full size imageGenomic profiling of lung squamous cell carcinomaThe genomic profile of LUSC differs significantly from that of LUAD [28]. LUSC has a high rate of genomic alterations, including mutations, rearrangements and CNAs. According to a TCGA study, frequently mutated genes include TP53 (mutated in about 90% of cases) and CDKN2A (about 70% of cases), as well as PTEN, PIK3CA, KEAP1 and RB1 [26] (a more comprehensive list is provided in Table 1). Further significantly mutated genes include, among others, HRAS, EGFR, SMARCA4 and BRAF, though the spectrum of EGFR mutations in LUSC differs from that in LUAD.Amplifications are commonly found in genes like SOX2, FGFR1, CCND1, PDGFRA, EGFR and MYC, while deletions affect CDKN2A, FOXP1 and PTEN (a more comprehensive list is provided in Table 3). Although LUSC shows a high number of somatic rearrangements, actionable gene fusions are rare, with only one NTRK2-TP63 fusion reported in the TCGA series [28].NSCLC transcriptomic profilingGene expression profiling can distinguish the histological types of lung cancer [41], identify subclasses [41,42,43,44] and stratify patients according to prognosis [43,44,45]. Integrating gene expression profiling with the mutational status of key genes allows further refinement of LUAD’s subclassification [46].The TCGA study identified three transcriptional subtypes of LUAD: the terminal respiratory unit (TRU), proximal-inflammatory (PI) and proximal-proliferative (PP) [27]. The TRU subtype is enriched with EGFR mutations and RTKs fusions, while PI is characterised by NF1 and TP53 co-mutations, and PP is associated with KRAS mutations and STK11 loss. The PI and PP subtypes are typically found in smokers and show higher TMB, while TRU is more common in non-smokers and women.In an enlarged TCGA case series, five transcriptional subtypes (S1–S5) of LUAD were identified via consensus clustering and their features were investigated by integrating multiple data sources [47]. S5 is closely related to the TRU subtype and S4 to the PP subtype, while the PI subtype was split into S1, S2 and S3. Pathway analysis revealed distinct characteristics: S1 had a low immune/inflammatory signature, S2 was enriched in epithelial-mesenchymal transition (EMT) and cell-adhesion signatures, S3 had high immune/inflammatory and proliferation signatures, S4 showed high proliferation, and S5 had a low proliferation signature linked to longer survival. Each subtype was associated with specific gene alterations: EGFR in S2 and S5, KRAS, STK11 and KEAP1 in S4, TP53 and NF1 in S3. Vulnerability data highlighted alterations of CDK4 in S3 and of CDK6 and CCND3 in S4, making them potentially sensitive to CDK4/6 inhibitors. In addition to the immune/inflammatory signature and CD274 (PD-L1) amplification and overexpression, S3 exhibited frequent MET amplification and overexpression, suggesting a combination treatment of MET inhibitors and ICIs, as well as a combination of CDK4/6 inhibitors and ICIs. The S3 subtype was found to be a stronger predictor of progression-free survival than PD-L1 expression in an independent cohort of LUAD patients treated with ICIs, highlighting its potential clinical utility as a predictive marker. In contrast, mutations in KEAP1 and/or STK11, enriched in the S4 subtype, have been associated in several studies with resistance to anti-PD-1 and anti-PD-L1 [48,49,50,51,52], which can be overcome by using combinations of these drugs with an anti-CTLA4 [53]. Indeed, alterations in KEAP1 and/or STK11, which are especially frequent in KRAS-mutated tumours, promote an immunosuppressive tumour microenvironment, enriched in suppressive myeloid cells and depleted in CD8+ cytotoxic T cells, but with relative sparing of CD4+ effector T cells. The latter are sensitive to anti-CTLA4, and the combination of an anti-CTLA4 with an anti-PD-1 or anti-PD-L1 induces suppressive myeloid cells reprogramming into inducible nitric oxide synthase-expressing tumoricidal phenotypes and recruits circulating neutrophils [53]. Co-mutations in KRAS or TP53 can further affect these phenotypes [54, 55].For LUSC, four subtypes were identified: primitive, classical, secretory and basal, each with distinct genetic profiles, prognosis and normal cell type correspondence [26, 56]. They are characterised by enrichment of genes involved in proliferation, metabolism, immune response and cell adhesion processes, respectively.The classical subtype is marked by overexpression of SOX2, PIK3CA and the ΔN-isoform of p63 and by alterations in KEAP1, NFE2L2 (encoding NRF2) and PTEN. The primitive subtype shows frequent RB1 and PTEN alterations, and the basal subtype NF1 alterations.Transcriptomic profiling holds promise for predicting treatment responses and outcomes. Numerous transcriptomic predictors have been developed to predict responses to ICIs [57], with some specifically applied to NSCLC [58, 59]. A notable example is the T-effector/interferon-γ-associated gene expression signature, which reveals immune activation through the levels of transcripts of PD-L1, CXCL9 and IFNγ. High expression of this signature correlates with benefit from atezolizumab in first- and second-line therapy of NSCLC, particularly in patients lacking RTK/RAS/RAF genomic alterations [60,61,62].Other gene signatures focus on identifying tumours with significant activation of specific pathways. For instance, a 46-gene signature predicts the activation of the KEAP1/NFE2L2(NRF2) antioxidant pathway [63]. The associated K1N2 score not only reliably indicates the presence of KEAP1/NFE2L2 mutations but also outperforms mutational analysis in predicting patient survival.Proteogenomics of NSCLC and pathways activation and targetingProteogenomic studies integrate genomic and proteomic data, alongside other omics like DNA methylation, RNA-seq and post-translational modifications (PTMs) such as protein phosphorylation, to provide a functional analysis of the cellular signalling networks and identify relevant therapeutic targets [21]. Most proteogenomic studies in NSCLC [64] are not specific for tumours with or without actionable genomic alterations but identify subgroups of tumours that are related to these categories.Proteogenomic analysis of lung adenocarcinomaProteomics provides quite different information from transcriptomics on cellular phenotypes. The strength of the correlation between mRNA and protein levels for a given gene depends, at least in part, on post-transcriptional (e.g., splicing, microRNA-mediated regulation) and post-translational (e.g., ubiquitination, protein degradation) mechanisms. These regulatory processes can vary across tumour subgroups, influencing signalling networks and tumour behaviour. In LUAD, gene-specific mRNA–protein correlation patterns have been used to identify dysregulated genes in tumours with early recurrence compared to those that remain relapse-free [65].Phosphoproteomic analyses of LUAD have revealed that activation of key growth pathways like RAS/MAPK and PI3K/AKT/mTOR are only partially driven by genetic alterations in the same pathways [27]. Drugs targeting specific proteins within a pathway may also be effective when the pathway is activated by alterations in other proteins, even if the direct target is not mutated. One such example is the CDK4/6 inhibitor palbociclib, which has shown efficacy in NSCLC patients with KRAS mutations [66] but performed poorly in cases with CDKN2A mutations or in LUSCs with CCND1-3/CDK4 amplifications [67]. Such drug sensitivities can be attributed to signalling network rewiring resulting from the genetic and epigenetic alterations of a specific tumour [68].A comprehensive proteogenomic study of 110 treatment-naïve LUAD samples from patients of different ancestries identified four distinct multi-omics clusters (C1–C4), which partly overlap with the previously identified transcriptomic subtypes PI, PP and TRU, but subdivide the second into two distinct clusters [69]. These clusters are characterised by different genetic and molecular features. C1 includes cases enriched with TP53 mutations, CpG island methylator phenotype (CIMP)-high status and a high TMB, C2 is characterised by wild-type TP53 and EGFR and intermediate CIMP, C3 is enriched for Vietnamese patients and STK11 mutations, while C4 is enriched with EGFR mutations and EML4-ALK fusions, primarily affecting Chinese and female patients. Each cluster also displays distinct pathway activations, such as immune signalling in C1, Rho GTPases and haemostasis/platelet in C2, histone deacetylase and cell cycle in C3 and MAPK1/MAPK3, MECP2 regulation, surfactant metabolism and chromatin organisation in C4. The study identified new potentially oncogenic gene fusions, some of which activate oncogenes like PTK2, PDGFRA and GSK3B, or disrupt tumour suppressors like STK11, STK4 and ATM. Phosphoproteomic analyses linked these fusions to specific activated pathways, providing insights into potential therapeutic targets. Analyses of paired tumours and normal tissues identified increased expression and altered PTMs of proteins linked to cell cycle progression, glycolysis and MYC targets in LUAD samples. NPM1 and MKI67 showed increased phosphorylation and Histone 2B and EP300 increased acetylation in tumours and five proteins (GFPT1, BZW2, PDIA4, P4HB and PMM2) were consistently upregulated, suggesting their potential as biomarkers or drug targets.CIMP-high tumours, frequent in the C1 cluster, are characterised by CpG islands hypermethylation in the promoter regions of multiple genes, leading to reduced expression of several genes associated with lung cancer development and prognosis, such as CLDN18, ANK1 and PTPRCAP. Potential therapeutic approaches for CIMP-high tumours include DNA demethylating agents and histone deacetylase inhibitors, though their efficacy in lung cancer remains unproven.Proteogenomic analyses from the same study revealed several therapeutic vulnerabilities in LUAD, including the upregulation of mismatch repair and DNA damage response proteins in TP53-mutated tumours. These may predict sensitivity to ICIs and to PARP inhibitors, respectively, though results of PARP inhibitors in NSCLC have been unsatisfactory [70,71,72,73,74,75,76]. Additionally, TP53 mutations were associated with increased expression of EZH2, a lysine methyltransferase that methylates histone H3, inducing chromatin condensation and resistance to DNA-damaging agents, suggesting a potential role for EZH2 inhibitors [77]. Other driver mutations, such as in SMARCA4 and STK11, were linked to increased expression or phosphorylation of SMAD2 and SMAD4, proteins involved in the TGF-β pathway, suggesting potential sensitivity to the TGF-βR1 inhibitor vactosertib [78]. KEAP1 mutations lead to increased levels of NRF2, which is involved in antioxidant responses through activation of mTOR signalling. Inhibitors targeting TORC1/2 have shown preclinical activity in lung cancer models with KEAP1/NRF2 pathway alterations [79]. KRAS mutations were linked to the activation of SOS1 and early data suggest that SOS1 inhibitors may enhance the efficacy of KRASG12C inhibitors in LUAD [80]. The SHP2 protein tyrosine phosphatase, encoded by PTPN11, is frequently phosphorylated in EGFR mutant and in ALK fusion-positive LUAD, leading to activation of the MAPK pathway. SHP2 inhibitors have shown promise in preclinical studies of EGFR-mutant and ALK fusion-positive tumours with activation of MAPK pathway [81, 82]. The study identified several hyperphosphorylated kinases that are known drug targets in other cancers. These included PRKCD in KRAS-mutant tumours, BRAF in TP53-mutant tumours and WEE1 in EML4-ALK fusion-positive tumours. Additionally, 27 other putatively druggable hyperphosphorylated kinases, whose inhibitors have not yet been approved by the FDA, were also identified [69].In the same study, the immune landscape of LUAD was classified into three clusters: hot-tumour-enriched (HTE), cold-tumour-enriched (CTE) and normal adjacent tissue (NAT)-enriched. HTE tumours show upregulation of members of multiple immune-related pathways, including PD-1, PD-L1, CTLA4, FOXP3, and IDO1, suggesting the utility of combination immunotherapy. In contrast, CTE tumours showed upregulation of epithelial barrier components, contributing to immune suppression, while NAT-enriched tumours showed intermediate immune signatures. STK11 mutant tumours show strong immune downregulation associated with a neutrophil degranulation signature. The study identified several cancer-testis antigens that are immunogenic and could be useful to develop tumour vaccines [69].A proteogenomic study on 103 LUAD cases in Chinese patients revealed that 50% had EGFR mutations. The study identified three proteomic subtypes: one linked to cellular environment and metabolism pathways, a mixed subtype with most EGFR mutations and a subtype enriched in proliferation with a higher TMB. In total, 11 potential drug targets were identified, including IMPDH2 and GAPDH, which are FDA-approved [83]. Despite East-Asian LUAD often occurs in non-smokers with EGFR mutations, TP53 mutations were found in a third of cases in Taiwan patients and were associated with activation of proteins involved in DNA repair, presenting further therapeutic opportunities [84]. The APOBEC mutational signature, linked to immunotherapy benefits, was common in younger women without EGFR mutations and was associated with the activation of CDK1, CDK2 and Aurora Kinase B as potential targets. Additionally, matrix metalloproteinases, especially MMP11, were linked to poor survival, representing another therapeutic target [84].In a proteogenomic study on 87 LUAD cases in the U.S., multi-omics clustering confirmed the transcriptomic subtypes TRU, PI and PP [85]. While the TRU subtype is enriched in EGFR mutations, the PI subtype showed enhanced IFN-γ signalling and PD-L1 and CTLA4 expression, making it potentially responsive to ICIs. The PP subtype is characterised by activation of several CDKs and of MAP2K7 and by metabolic alterations affecting glycolysis and glutaminolysis. As SMARCA4 inactivation is known to be synthetic lethal with CDK4, PP tumours with SMARCA4 mutations and high CDK4 activity may respond well to CDK4/6 inhibitors. They are also potentially vulnerable to glutaminase inhibitors in STK11-KEAP1-KRAS-mutant cases or to stearoyl-coenzyme A desaturase inhibitors in combination with ferroptosis inducers in STK11-KEAP1 co-mutant cases.Proteogenomic studies can provide valuable insights into the molecular characteristics and therapeutic opportunities in lung cancers related to specific aetiological factors. One such study focused on 169 never-smoking females from the Xuanwei area in China, where exposure to coal smoke is the primary cause of LUAD [86]. The study identified benzo[a]pyrene, a polycyclic aromatic hydrocarbon, as the main carcinogenic agent. LUADs linked to this carcinogen exhibited unique molecular features, including specific EGFR G719C/A/D/S (G719X) mutations, which were present in 20% of the cohort. These mutations were associated with upregulation of components of the MAPK signalling pathway, such as MAP2K2 (MEK) and MAPK3 (ERK1), as well as kinases involved in cell cycle regulation, including CDK2, AURKB, CSNK1A1 and CDK4. Notably, these kinases are targets of drugs that are either already approved or currently in clinical development. Afatinib and osimertinib are recommended as the preferred first-line therapies for LUADs with these specific EGFR mutations [11]. However, the study suggests that combining these agents with inhibitors targeting the upregulated downstream kinases may provide a strategy to overcome resistance and improve treatment outcomes.Another area of proteogenomics investigation concerns pre-invasive or minimally invasive lesions, to shed light on mechanisms of carcinogenesis, develop diagnostic and prognostic tools and identify therapeutic targets [87,88,89].Lung adenocarcinoma without actionable genomic alterationsSome proteogenomic studies have been conducted specifically on LUADs lacking clinically actionable genomic alterations. One such multi-omics study focused on non-coding regions of DNA [90]. A significant finding was the lack of enhancer activity in the MAML2 gene, linked to translocations (forming fusion genes such as CRTC1-MAML2, a known oncogenic driver in mucoepidermoid carcinoma), enhancer mutations, DNA methylation or histone acetylation changes, which downregulated MAML2. This led to downregulation of members of the NOTCH and WNT/β-catenin pathways, while ERBB2 was overexpressed. Patients with low CD302, FAT4 and FOXN3 expression, associated with MAML2 downregulation, had worse overall survival and showed upregulation of molecules like PLK1, UBE2C and LYPD3, which might represent therapeutic targets. Epigenetic therapies could potentially restore MAML2 expression in some cases.A study on Korean patients with EGFR- and ALK-wildtype LUAD revealed elevated oestrogen receptor (ER) signalling, especially in never-smoker with STK11 mutations, while KRAS mutations were associated with elevated ER signalling regardless of smoking status [91]. Deletions in chromosomes 14 and 21, DNA hypomethylation of genes LLGL2 and ST14 and SRC overexpression were also associated with increased ER signalling. The SRC inhibitor saracatinib showed activity in STK11-mutant and especially in STK11/ERBB2 co-mutated LUAD cell lines, representing a potential therapy in this setting, while selective ER modulators like tamoxifen were inactive.A proteogenomic study of 99 never-smoking Korean patients with EGFR- and ALK-wildtype LUAD identified four molecular subgroups with distinct clinical outcomes, based on previously defined tumour and microenvironment signatures applied to transcriptome and proteome data [92]. The Proliferation-high (P) subgroup, with a worse prognosis, is marked by TP53 and ARID1A mutations, upregulation of proliferation-related genes, high Ki67 levels and low immune activation. It is enriched for multiple actionable kinases, including CDK2, CDK5, polo-like kinases and ATR and shows greater dependency on CDK9 at vulnerability screens. The Immune-high (I) subgroup shows high immune cell infiltration, particularly B-cells, and upregulation of immune checkpoints (e.g., PD-1, TIGIT, CTLA4), cytokines (e.g., CXCL13, CD27) and chemokines (e.g., CCL5). It shows dependency on TRAF2, a mediator of resistance to ICIs [93]. The Angiogenesis-high (A) subgroup features TP53 and KRAS co-mutations and upregulation of proangiogenic factors (FGF2, CXCL12, PDGFB, LGALS3), often secreted by stromal cells. It showed pronounced dependency on GRB2, encoding an adaptor protein for RTKs. Lastly, the Metabolism (M) subgroup shows upregulation of metabolic enzymes involved in oxidative phosphorylation, lipid and carbon metabolism, upregulation of signalling molecules such as ERBB3, ICK and ARAF, partial CD8+ T-cell suppression and high expression of CERS4, an enzyme involved in sphingolipid metabolism and linked to anti-PD-1 responses [94]. It shows dependency on metabolic genes such as ACACA, involved in mitochondrial fatty acid synthesis. According to the PRISM repurposing dataset [95], potentially active drugs include digitoxin and the histone deacetylase tacedinaline in subgroup P, the lysophosphatidic acid receptor antagonist KI16425 in subgroup I, the growth hormone secretagogue ibutamoren (MK-677) in subgroup A and the 3-phosphoglycerate-inhibitor veterinary anthelmintic clorsulon for the M subgroup. These findings reveal the molecular diversity within EGFR- and ALK-wildtype LUADs in never smokers and suggest subgroup-specific vulnerabilities for tailored therapies.KRAS mutant lung adenocarcinomaSome proteogenomic studies have focused on KRAS-mutant LUADs. A proteomic study considering only EGFR wildtype cases found that approximately one-third of KRAS-mutant LUADs had higher ERK pathway activation compared to the KRAS wildtype counterpart [96]. Cross-talks were observed between KRAS effectors and the AKT/mTOR pathway, along with correlations to RTKs phosphorylation. Additionally, 18% of KRAS-mutant tumours exhibited increased phosphorylation of ER alpha, supporting potential treatment strategies targeting multiple pathways.Further integrative analysis classified KRAS-mutant tumours into three subgroups based on co-mutations with STK11, TP53 and CDKN2A/B, each showing distinct genomic characteristics and potential treatment vulnerabilities, such as sensitivity to HSP90 inhibitors in KRAS/STK11 co-mutated LUAD cell lines [54]. ICIs could be effective for tumours in the KRAS/TP53 co-mutated subgroup, which shows expression of PD-1/PD-L1 and CTLA4.Finally, an analysis integrated cancer cell line dependencies, gene actionability and patient genomic data to identify EGLN1 as a therapeutic target, especially in KRAS-mutated LUAD [97].TTF1-negative lung adenocarcinomaTTF1-negative LUADs have fewer actionable mutations and poorer outcomes. A proteogenomic study revealed that these LUADs are enriched with KEAP1 mutations, leading to increased NRF2 expression, which may be a potential therapeutic target [98, 99]. The study also found increased expression of DNA repair enzymes (e.g., CHK1), cell cycle molecules (e.g., cyclin B1) and reduced MAPK and PI3K/mTOR signalling. LUAD cell lines with low TTF1 expression were more sensitive to DNA repair-targeting drugs like PARP and ATM inhibitors but showed reduced sensitivity to paclitaxel and pemetrexed. Additionally, TTF1-negative LUAD cell lines often overexpress SRGN, encoding the proteoglycan serglycin, which promotes PD-L1 expression, proinflammatory cytokine production and increased tumour invasiveness and represents a potential therapeutic target [100].Proteogenomic analysis of lung squamous cell carcinomaA multi-omics study of LUSC identified three primary proteomic subtypes: inflamed, redox and mixed [101]. Each subtype has distinct characteristics and potential therapeutic implications. The inflamed subtype is characterised by an intense inflammatory infiltrate, involving neutrophils, lymphocytes, monocytes, regulatory T cells or myeloid-derived suppressor cells. These tumours show elevated expression of neutrophil-associated proteins, extracellular matrix proteins and PD-1 and tend to harbour fewer mutations in key genes and fewer CNAs compared to other subtypes. They often present tertiary lymphoid structures (TLSs), which are associated with better prognosis. Combination immune therapies could be particularly effective for the inflamed subtype, and it has been suggested that adding an anti-CD33 drug, such as gemtuzumab ozogamicin, might enhance therapeutic responses by eliminating immunosuppressive myeloid cells. The redox subtype is distinguished by alterations in metabolic oxidation-reduction processes. It shows elevated expression of aldo-keto reductase and alcohol dehydrogenase enzymes, along with high rates of CNAs, particularly amplifications in chromosome regions 3q2 (harbouring TP63, SOX2 and PIK3CA) and 2q3 (containing NFE2L2). Mutations in NFE2L2 and KEAP1 are also prevalent, with 84% of redox tumours exhibiting alterations in these genes. Based on public RNA interference and CRISPR knockout screenings data, three genes—TP63, PSAT1 and TFRC—were identified as promising therapeutic targets for this subtype. TP63 encodes the ΔNp63 protein, which has oncogenic functions and regulates glutathione metabolism, while PSAT1, whose expression is induced by NRF2, is involved in serine biosynthesis and associated with poor prognosis. TFRC encodes the transferrin receptor 1, involved in ferroptosis, a form of programmed cell death controlled by NRF2. This subtype has metabolic vulnerabilities, including altered serine biosynthesis, glycolysis and reactive oxygen species (ROS) production, potentially making these processes promising therapeutic targets. The mixed subtype, the smallest of the three, is characterised by expression of proteins linked to the WNT/β-catenin pathway and an enrichment of APC mutations.In a proteomic study comparing LUSC to LUAD, a shift from cap-dependent to cap-independent translation was highlighted as a key feature of LUSC [102]. Under stress conditions like hypoxia or ROS, mTOR activity decreases, leading to suppression of cap-dependent translation and an increased reliance on cap-independent translation. This process facilitates the production of oncogenic proteins like HIF1α, MYC, VEGFA and BCL-2, which support tumour growth, angiogenesis and cell survival.In another thorough proteogenomic study, multi-omics clustering identified five subtypes of LUSC: Basal-Inclusive (B-I), Epithelial-to-Mesenchymal Transition-Enriched (EMT-E), Classical, Inflamed-Secretory (I-S) and Proliferative-Primitive (P-P) [103]. The B-I subtype displayed overexpression of TROP2, a target of antibody-drug conjugates like sacituzumab govitecan, while the EMT-E subtype was marked by activation of targetable proteins such as PDGFRB and ROR2. Importantly, the study identified loss of cell cycle inhibitors, such as CDKN2A/p16INK4a and RB1, due to genetic, epigenetic, or unknown reasons, as a universal feature of LUSC, and amplification of CCND or CDK4/6 genes was also frequent. CDK4/6 inhibitors have shown limited success in LUSC, but a subset of patients with high RB phosphorylation levels might benefit from these drugs, as demonstrated in LUSC cell lines, where RB phosphorylation levels were more predictive of response than alterations in CCND1, CDK2NA and RB1. Several other key signalling pathways and proteins were identified as potential therapeutic targets. These include the NRF2 pathway, frequently activated through mutations in NFE2L2, CUL3, or KEAP1 in the classical subtype, but sometimes also in the absence of those mutations and possibly because of CDK5 upregulation. The squamous differentiation marker SOX2, frequently co-amplified with TP63 in the classical subtype, is generally considered undruggable, but was positively correlated with the chromatin modifiers KDM1A (LSD1), KDM3A and EZH2, whose inhibition leads to SOX2 downregulation. Conversely, LUSC cases with low ΔNp63α expression show upregulation of survivin (BIRC5) and are potentially responsive to survivin inhibitors [103]. Expression of the immune checkpoints PD-1, PD-L1, CTLA4 and IDO1 is prominent in the I-S subtype, predicting responsiveness to ICIs. Additionally, upregulation of Rho GTPase signalling in immune cells and of CSF1R predominantly in macrophages, along with dysregulation of other immune-related proteins, represents further potential targets in the I-S subtype. The analysis of CNAs in this study showed that WHSC1L1 belongs to the same amplicon as FGFR1 and may be the critical driver oncogene, thus representing a potential therapeutic target and explaining the failure of anti-FGFR1 drugs in this context. Lastly, LUSC tumours showed upregulation of protein kinases such as EGFR, SRC and MAPK14, and EGFR phosphorylation correlated with ligand abundance rather than with gene amplification [103].Proteogenomic studies including multiple histological types of NSCLCSome proteogenomic studies involved an analysis of both LUAD and LUSC cases. A multi-omics cluster of clusters analysis of 1,023 NSCLC cases from TCGA identified nine tumour subtypes, three with predominantly LUSC and six with predominantly LUAD cases [104]. LUSC subtypes showed increased SOX2 amplification and expression, as well as p63 and KRT5/6 overexpression and more PTEN losses, while LUAD subtypes had higher expression of NKX2-1 and KRT7 and more STK11 alterations with decreased expression. LUAD subtypes also showed elevated mTOR and MAPK pathways phosphorylation, suggesting potential therapeutic responsiveness. Immune checkpoint activation and cancer-testis antigen expression were present in both LUSC and LUAD subtypes, offering therapeutic opportunities.In a study of 141 NSCLC samples, including all major histological types, six proteome subtypes were identified [105]. Subtypes 1–4 mostly included LUADs, subtype 5 included neuroendocrine neoplasms and subtype 6 comprised LUSCs. Subtypes were further analyzed based on immune cell infiltration, metabolic pathways and mutations. Subtypes 2 and 3 had higher immune infiltration, subtype 5 had the highest proliferation rates, and subtype 1 had the lowest. The predominant mutations affected EGFR in subtype 1, STK11, KEAP1 and SMARCA4 in subtype 4, RB1 in subtype 5 and TP53 in subtype 6. This aligns with the network analysis findings of metabolic pathways activation in subtype 4, E2F1/MYC signalling in subtype 5 and p53 signalling in subtype 6. While subtypes 2 and 3 showed high levels of T cell and B cell infiltrates, respectively, the immune-cold subtypes 4 and 6 were characterised by low immune activity despite neoantigen expression. These subtypes expressed immune-inhibitory ligands, such as FGL1 (binding to LAG-3) and B7-H4 (binding to activated T-cells), which could be treated with anti-LAG-3 and anti-B7-H4 therapies. Subtype 4, associated with STK11 mutations, showed activation of mTOR signalling and potential vulnerability to mTOR inhibitors combined with LAG-3/FGL1 checkpoint inhibitors.An integrative multi-omics analysis of 229 Korean patients with NSCLC identified five molecular subtypes, validated through prior multi-omics studies [106]. The cohort included LUADs (61%), LUSCs (27%) and other histological types (12%) across early and advanced stages. Subtype 1, termed 'metabolic', was predominantly composed of LUAD cases in female patients. This subtype was enriched with EGFR and TP53 mutations, CDKN2A copy number loss and frequent whole genome doubling (WGD) events. It also exhibited upregulation of proteins involved in metabolic pathways. Subtype 2, referred to as 'alveolar-like', consisted mainly of LUADs with EGFR mutations but low frequencies of TP53 mutations and WGD events. This subtype displayed activation of IL-33 and NOTCH pathways and was associated with the best survival. Subtype 3, labelled 'proliferative', primarily included LUSCs from male smokers. It had the highest frequency of WGD events, frequent TP53 and PIK3CA mutations and amplifications on chromosome 3q involving SOX2 and other cancer-related genes. This subtype was enriched in cell-cycle related pathways, including E2F/MYC targets, G2M checkpoint and CDKs. Among the upregulated proteins was XPO1, whose inhibitor selinexor demonstrated antitumor activity in WGD-positive LUSC organoids. Subtype 4, termed 'hypoxic', included all histological types and was associated predominantly with metastatic cases and the poorest prognosis. It was characterised by activation of hypoxia, PI3K-AKT and neutrophil degranulation pathways. This subtype showed upregulation of CSNK2A1 and GSK3B, known to activate the PI3K-AKT pathway, as well as SLK (phosphorylated by CSNK2A1 and involved in apoptosis) and LRRFIP1 (promoting EMT), both conferring a poor prognosis. Subtype 5, called 'immunogenic', showed frequent tumour-infiltrating lymphocytes (TILs)-associated patterns, enrichment in KRAS mutations and in immune-related pathways such as TNFα signalling via NF-kB. Analysis of the tumour immune microenvironment revealed three immune clusters: HTE, CTE and NAT-enriched. While HTE tumours were generally associated with better prognosis, this advantage was diminished in cases with regulatory T-cell (Treg) enrichment. Subtype 5 tumours, often enriched in HTE patterns, also frequently exhibited Tregs and neutrophils, reducing survival outcomes. Upregulated immunomodulators in HTE tumours included SLAMF7, associated with the presence of SMARCA4 mutations and a target of the monoclonal antibody elotuzumab. Furthermore, subtype 5 was characterised by cryptic MHC class I-associated peptides [107], noncanonical neoantigens derived from noncoding transcripts, pseudogenes, or untranslated regions of mRNA. These neoantigens were positively correlated with HTE status and demonstrated prognostic significance. From a therapeutic perspective, subtype 5 showed the largest benefit from adjuvant chemotherapy or chemoradiation, unlike the other subtypes. Additionally, ICIs could provide benefit for this subtype.Single-cell analyses assessing tumour and microenvironment heterogeneityNSCLC is characterised by significant intratumor heterogeneity, affecting prognosis and treatment response. The TRACERx study revealed that, on average, each NSCLC harbours 4.2 truncal and 2.8 subclonal driver mutations, with 77% of tumours showing at least one WGD event and 19% showing at least one subclonal WGD event [108, 109]. In LUAD, mutations in RTKs, MYC and NRF2 pathways are early events under truncal selection. Other mutations in key cancer genes like STK11, TP53 and KRAS may act as either truncal or subclonal mutations. Mutations in chromatin remodelling and NOTCH pathways or in genes like PTEN, RUNX1 and SMAD are often subclonal and involved in later stages of tumour evolution. In LUSC, subclonal selection frequently affects mutations in a different gene set, including ATM, KEAP1, NFE2L2 and PIK3CA.Single-cell RNA sequencing (scRNA-seq) has revealed distinct cellular compositions and interactions in NSCLC subtypes, including different spectra of stromal and immune cells [110] and different expression of immune checkpoint molecules (e.g., TIM3 and TIGIT in LUAD, CD96 in LUSC and LILRB1/2 in both tumour types), highlighting different potential therapeutic targets [111]. A combined analysis including scRNA-seq and bulk RNA-seq identified AT2 cells as dominant malignant cells in LUAD and basal cells in LUSC [112], with distinct oncogenic drivers. NKX2-1 emerged as a key regulator of AT2 cells in LUAD, whereas KLF5 and MYC were identified as key transcription factors in basal cells in LUSC. Other genes overexpressed and with potential oncogenic roles are AZGP1 and S100A13 in AT2 cells in LUAD and PPT1 and KPNA2 in basal cells in LUSC, all promoting cell proliferation and representing potential therapeutic targets.Single-cell transcriptomic and multiomic analyses, often combined with spatial analysis, are promising tools to improve the prediction of response to ICIs, considering both tumour features and the stromal and immune cells landscape [113]. In patients treated with neoadjuvant nivolumab, scRNA-seq showed lower expression of genes involved in cytolytic programs and upregulation of immune checkpoints in TILs specific for mutation-associated neoantigens, compared with TILs not specific for neoantigens [114]. Furthermore, TILs from patients who achieved a major pathological response (MPR) showed higher expression of genes associated with effector and memory functions and lower expression of exhaustion markers compared to TILs from patients who did not achieve an MPR. Beyond highlighting the high tumour and stromal/immune cell heterogeneity in LUAD, spatial analyses found increased T-regulatory cells and decreased cytotoxic T cells and antigen-presenting cells in normal lung parenchyma adjacent to the tumour compared to distant parenchyma [115]. A proteomic signature derived from stromal areas has been found to be a stronger predictor of response to ICIs than a signature from the tumour compartment [116]. The presence of activated TLSs [117], composed mainly of germinal centre B cells, effector memory CD4 T cells and follicular helper T cells, as well as the presence of stem-immunity hubs [118], including mainly stem-like TCF7+PD-1+CD8+ T cells, activated CCR7+LAMP3+ dendritic cells and CCL19+ fibroblasts, have been shown to predict responses to ICIs. Moreover, the proportion of collagen type XI alpha 1 chain-positive (COL11A1+) cancer-associated fibroblasts (CAFs) was significantly higher in LUAD from non-responders than in responders to neoadjuvant chemo-immunotherapy and non-responders showed an increase in monocytes/macrophages and dendritic cells in post-treatment surgical samples compared to baseline biopsies [117]. COL11A1+ CAFs co-localise with SPP1+ macrophages at tumour borders and their interaction promotes the production of collagen by CAFs, that obstacles the contacts between tumour cells and cytotoxic immune cells, leading to immune exclusion.Specific features of metastatic NSCLCStudies on metastatic NSCLC reveal a different mutational landscape compared to primary tumours, with higher rates of EGFR alterations, which could be due to referral bias. An analysis of 860 patients with recurrent or metastatic LUAD, mostly pretreated, identified actionable somatic alterations in 87% of cases [119]. Most alterations involved the RTKs/RAS/MAPK pathway, the PI3K/AKT/mTOR pathway, or BRCA1/2. Twelve percent of the tumours lacked any known actionable somatic alterations and were classified as the 'unknown mitogenic driver' subset. They were enriched for alterations in TP53, STK11, KEAP1, KMT2D and PDGFRA, as well as other alterations related to smoking history.The TRACERx study revealed that metastatic lesions often harbour unique mutations not found in primary tumours, but only 33% of these are driver mutations, affecting genes like TP53, KMT2D, STK11, SMARCA4, FAT1, NF1, RBM10, PIK3CA, ARID1A, CUX1, FBXW7, EGFR, ARHGAP35 [120]. Metastasis-unique mutations can arise from therapeutic pressure, such as those in PMS1 linked to platinum-based chemotherapy.A study of over 2500 samples, including primary and metastatic LUAD, found that metastasis-unique actionable oncogenic alterations occurred in only 4% of metastases and were often linked to resistance mechanisms developed in patients treated with RTK inhibitors [121].Targeting alterations in tumour suppressor genesA critical issue in the treatment of NSCLC is targeting alterations in key tumour suppressor genes. TP53, the most frequently mutated gene in NSCLC [26, 27], exemplifies the complexity of tumour molecular mechanisms [122, 123]. Missense mutations often lead to conformational changes in p53, resulting in varying degrees of loss of function, but in some cases also leading to gain-of-function effects, typically of a non-canonical oncogenic nature, or to increased immunogenicity. Truncating mutations are usually associated with a complete loss of function. Reduced expression and function of p53 may stem from overexpression of its inhibitors MDM2 or MDM4, due to gene amplification or post-translational modifications, or from deletion of CDKN2A, which encodes p14ARF, an MDM2 inhibitor. Occasionally, aberrant p53 conformation and function are observed even in TP53 wild-type tumours, either in cancer cells or in stromal cells such as the CAFs supporting tumour growth [123]. Conversely, certain structural p53 mutants may, under specific conditions, regain a normal conformation [124]. Thus, genomic characterisation alone may be insufficient to identify all the alterations in the p53 pathway.A variety of therapeutic strategies aimed at restoring normal p53 functions are under preclinical and clinical development [122, 123]. In TP53 wild-type tumours, inhibitors of the p53–MDM2 interaction, such as idasanutlin, have thus far yielded limited efficacy and notable toxicity, but newer molecules are currently in development [123]. Other inhibitors, such as kevetrin, which blocks the E3 ligase activity of MDM2 and prevents p53 degradation [125] and the stapled peptide sulanemadlin, which mimics the N-terminal domain of p53 and binds to MDM2 to inhibit its activity [126], are also being studied. Proteolysis-targeting chimeras (PROTACs) [122, 127], bifunctional molecules that simultaneously bind a target protein, such as MDM2 and an E3 ubiquitin ligase to induce proteasomal degradation, are effective at sub-stoichiometric concentrations and are progressing through clinical development. In TP53-mutated tumours, drug development efforts are focused on restoring proper p53 conformation and transcriptional activity. In this context, eprenetapopt is in advanced clinical development with some promising results [128, 129] and recently, rezatapopt, a small molecule that selectively reactivates p53 carrying the Y220C mutation, has demonstrated strong preclinical activity and is now undergoing clinical evaluation [130]. Another class of investigational drugs are those promoting translational readthrough of nonsense-mutant TP53, including certain aminoglycoside and macrolide antibiotics and 5-fluorouridine, which enable ribosomes to bypass premature termination codons and produce a full-length functional protein [123]. Numerous other therapeutic strategies focused on p53 are under clinical investigation, including gene therapy, vaccines, synthetic small interfering RNA oligonucleotides, bispecific antibodies targeting mutant p53, targeted T cell receptor–T cell therapies and agents that exploit synthetic lethality in tumours harbouring TP53 mutations [122, 123].Another frequently mutated tumour suppressor gene in NSCLC is KEAP1, which is part of the NRF2 signalling pathway. The transcription factor NRF2 is a master regulator of cellular homoeostasis, controlling redox balance and various aspects of cellular metabolism [99]. KEAP1 acts as an adaptor protein, linking NRF2 to the Cullin3–RBX1 E3 ubiquitin ligase complex, leading to NRF2 ubiquitination and proteasomal degradation. This mechanism maintains low levels of NRF2 under non-stressed conditions. Electrophilic compounds and ROS can modify specific cysteine residues in KEAP1, inducing a conformational change that disrupts its adaptor function. As a result, NRF2 accumulates and activates the transcription of stress-response genes to restore cellular homoeostasis. Prolonged NRF2 activation may also result from KEAP1 sequestration in autophagosomes in response to various stimuli.The NRF2 pathway, through its antioxidant and detoxifying activities, plays a protective role in normal cells by preventing neoplastic transformation and NRF2 agonists are being studied in cancer prevention [131]. However, in certain tumours this pathway is constitutively activated due to gain-of-function mutations in NFE2L2 (encoding NRF2) or loss-of-function alterations in KEAP1 or CUL3, promoting cell proliferation and drug resistance. NRF2 activation has been associated with poor response to several cancer therapies, including chemotherapy, radiotherapy, RTK inhibitors and ICIs [132,133,134,135,136]. NRF2 also plays a pivotal role in the metabolic reprogramming of cancer cells, redirecting glucose and glutamine utilisation toward anabolic pathways required to sustain proliferation. Among its effects, NRF2 upregulates the expression of enzymes involved in glutathione biosynthesis and inhibits ferroptosis, a form of programmed cell death driven by iron-dependent lipid peroxidation. By promoting glutamine uptake and increasing tumour cell dependence on external glutamine sources, NRF2 enhances sensitivity to glutaminase inhibitors, some of which are in clinical trials, albeit with limited results at present [137, 138].Several NRF2 inhibitors, including small molecules, stapled peptides and natural products, have shown antitumor activity and synergism with other therapies in preclinical studies [139]. However, their mechanisms of action have not been confirmed in some cases, and no such compounds are currently in clinical development. Among new promising candidates are PROTACS [140] and the so-called molecular glues [141] such as R16, which binds specifically to a crevice in mutant KEAP1 and restores its binding affinity for NRF2 [142]. Targeting NRF2 epigenetic regulation and PTMs are other potential strategies [143]. In addition, NR0B1, an orphan nuclear receptor expressed in KEAP1-mutant NSCLC, has been identified as potential druggable target [144], while tumours with overactivation of NRF2 show selective vulnerability to inhibitors of respiratory complex I [145]. The TORC1/2 inhibitor TAK-228 has also shown promise in treating NSCLC with NRF2 pathway alterations [79], and combination therapies with glutaminase inhibitors may help overcome resistance in LUSC [146].An additional tumour suppressor gene frequently mutated in NSCLC is STK11, which encodes the liver kinase B1 (LKB1) protein, a master kinase that regulates cell polarity, metabolism, proliferation and migration [147]. Unlike most kinases, which are activated by phosphorylation, LKB1 is allosterically activated through its interaction with the STE-20-related kinase adaptor protein (STRAD) and Mouse protein 25 (MO25), which together form a heterotrimeric complex with LKB1. LKB1 phosphorylates and activates AMP-activated protein kinase (AMPK). AMPK is also affected by the ratio of intracellular AMP to ATP, acting as an energy sensor, activated by AMP and deactivated by ATP. Activation of the LKB1–AMPK pathway stimulates catabolic processes and antagonises anabolic processes. It works, among other mechanisms, by inhibiting mTORC1, but also by inducing autophagy. Additionally, LKB1 activates a number of AMPK-related kinases through which it regulates pathways such as the Hippo pathway and EMT.AMPK agonists are currently under investigation as potential anticancer therapies. The most extensively studied compound is metformin, which increases the cytoplasmic AMP: ATP ratio, thereby activating AMPK [148, 149]. However, the LKB1–AMPK pathway may also exert pro-tumorigenic effects, and its downregulation can enhance the efficacy of certain anticancer treatments. Therefore, the development of drugs targeting the LKB1–AMPK pathway requires a deeper understanding of the specific effects of its activation and inhibition across different cellular contexts [147].Drugs targeting other tumour suppressor genes frequently altered in NSCLC, such as NF1 and SMARCA4, are being developed [150, 151].A roadmap to clinical implementationThe development of multi-omics assays faces several challenges related to the standardisation of omics techniques, data analysis and interpretation and clinical implementation.Each omics technique must adhere to appropriate quality standards, from sample preparation to the technical execution of the assay and the data pre-processing required for analysis [16, 152,153,154,155]. Data formats differ significantly across omics platforms, ranging from qualitative to discrete or continuous quantitative data [156]. Analytical approaches span from classical statistical techniques to machine learning methods, although the boundary between these disciplines is often blurred.Any omics data analysis must first deal with the 'curse of dimensionality,' whereby the high number of measured variables relative to the limited number of patients hinders parameter estimation in classical statistical models. This can dilute data correlations and complicate the identification of significant predictors, thereby reducing the performance of even machine learning models [157]. Additional issues include the need to correct for multiple testing [158] and the increased risk of overfitting as the number of analyzed variables grows [159].As a result, dimensionality reduction techniques are often a necessary first step, using either linear or nonlinear methods [160, 161]. The choice between these techniques can significantly impact analytical outcomes and requires careful judgement. The study objective further dictates the choice between supervised analyses, aiming to identify differences between predefined categories of tumours or patients based on phenotype or outcome and unsupervised analyses, aiming to identify subgroups of patients or tumour subtypes based on biomolecular profile similarity within-subgroups and dissimilarity between-subgroups [159, 161].There are various strategies for integrating data from different omics platforms [156, 161,162,163,164]. Early integration involves merging data from all omics types into a single large matrix and applying supervised or unsupervised analyses appropriate to the study objectives. Late integration involves analyzing each omics data type separately, then combining the results. Many approaches fall between these two extremes and are referred to as intermediate integration strategies.Some analytical methods incorporate existing biological knowledge, such as functional genomics or functional proteomics, into omics data analysis, generally yielding results that are more accurate and biologically interpretable [165]. With different types of analyses, it is possible to reconstruct molecular networks involved in biological processes—such as gene regulatory networks, protein-protein interaction networks and others—and use them for data integration and interpretation [166]. Specific computational methods have been elaborated for single-cell transcriptomic and multiomic studies, including methods for spatial analysis [154, 155, 167].Starting from a reduced set of variables, selected from multi-omics data using statistical and machine learning techniques, other analytical methods typical of systems biology can further characterise biological processes and potential interventions. These include, among others, logical models, aiming to capture the qualitative behaviour of biological systems and the causal influences between variables and models based on systems of differential equations, including stochastic ones, attempting to quantitatively describe the mechanisms underlying biological processes [168, 169]. These models require in-depth biological knowledge but can offer insightful interpretations and accurate predictions, potentially enabling the development of patient-specific models testable in biological systems such as organoids or patient-derived xenografts [170, 171]. The use of these models might require measuring only a limited number of biomolecular variables in individual patients, selected based on tumour type, through a form of targeted multi-omics analysis.Clinical application of predictive multi-omics models requires evidence of clinical utility, which must follow the established criteria for developing biomarkers [172, 173]. These include demonstrating analytical validity (accuracy, reproducibility, reliability), clinical validity (ability to identify distinct patient subgroups based on biology or outcomes) and clinical utility, i.e., evidence that using the biomarker to guide therapy improves patients’ outcomes, or maintains similar outcomes with reduced toxicity and/or costs, compared to standard clinic-pathological criteria. The highest level of evidence comes from randomised clinical trials specifically designed to demonstrate the biomarker’s clinical utility in the intended setting.Various clinical trial designs have been developed for this purpose, including adaptive trials that dynamically identify the patients’ subset benefiting from the tested treatment [174]. While prospective demonstration of the utility of transcriptomic predictors in breast cancer required large and long-duration trials [175, 176], increasingly accurate predictors may significantly reduce the number of patients needed for clinical validation. Moreover, as each tumour is biologically unique, n-of-1 trials may be appropriate [177,178,179], wherein each patient receives the treatment (mono- or poly-pharmacologic) deemed optimal for their specific tumour. Meta-analyses of individualised therapy outcomes from multiple n-of-1 trials can then be compared with standard treatment results. Additional opportunities may help optimise clinical research. Mathematical models that simulate tumour growth and its reduction in response to treatment, calibrated with individual patient data, can serve as patients’ digital twins. These models have the potential to enable personalised clinical trials by simulating tumour evolution in response to various treatment options, thereby supporting the identification of the most effective therapy and allowing for the adjustment of specific treatment parameters [180]. They can be integrated with data-driven models, generating predictions through machine learning algorithms [181], as well as with biology-based models that describe biological mechanisms through systems of differential equations.ConclusionsTargeted genomic profiling is a standard in the diagnostic workup of NSCLC and essential for guiding treatment in patients with actionable mutations. This approach has transformed the management of NSCLC and laid the groundwork for precision oncology. However, for tumours lacking actionable alterations, treatment relies primarily on ICIs with or without chemotherapy, despite suboptimal predictive biomarkers and heterogeneous response rates.Genomic information alone is often insufficient to identify optimal therapeutic targets. The integration of diverse omics technologies—genomics, transcriptomics, proteomics, epigenomics and others—offers a more comprehensive view of tumour biology. Proteomics, in particular, provides insight into cellular phenotypes and the activity of intracellular signalling networks through protein-level information and post-translational modifications. These integrated approaches enable tumour subtyping, elucidate signalling derangements and can support the identification of therapeutic targets [17, 21].As described throughout this review, NSCLCs lacking currently actionable mutations exhibit a complex molecular landscape. This includes rare but targetable alterations (Tables 1–3), epigenetic modifications under investigation and frequent mutations in tumour suppressor genes that remain non-targetable yet are intensively studied. The strength of the multi-omics approach lies in its ability to reveal mechanistic links across molecular layers. For instance, BZW2, which has limited prognostic relevance [182], was found to be upregulated in proteogenomic analyses [69] and later shown to promote LUAD progression and represent a potential therapeutic target [183,184,185]. The observed discordance between mRNA and protein levels has helped clarify the effects of some specific KEAP1 mutations on NRF2 activation in LUAD [69]. In LUSC, activation of the NRF2 pathway was identified even in the absence of direct mutations within the pathway itself [103], broadening the spectrum of potential use of NRF2 pathway inhibitors.Single-cell and spatial multi-omics analyses further dissect tumour and stromal cell interactions, highlighting predictors of response to ICIs [113] and supporting their use in this and other future applications.Despite their promise, multi-omics approaches remain primarily investigational. Their translation to clinical practice requires technical standardisation, selection of relevant variables and rigorous clinical validation. Some molecular layers, such as noncoding RNAs, epigenetics and metabolomics, are still underexplored and need better integration.Several NSCLC alterations affect non-actionable genes, including tumour suppressors and oncogenes like MYC [186]. Research into targeting these pathways remains critical to expanding therapeutic options.In conclusion, genomic profiling remains central to both clinical decision-making and research. While many alterations may not yet inform treatment choices, they are essential for discovering new targets. Multi-omics strategies can further deepen our understanding of NSCLC biology, with the potential to reveal novel therapeutic targets and resistance mechanisms, thereby supporting precision oncology. With rigorous validation, multi-omics analysis can pave the way for new advances in personalised cancer care.ReferencesKocarnik JM, Compton K, Dean FE, Fu W, Gaw BL, Harvey JD, et al. Cancer incidence, mortality, years of life lost, years lived with disability and disability-adjusted life years for 29 cancer groups from 2010 to 2019. JAMA Oncol 2022;8:420–44.PubMed Google Scholar WHO classification of tumours editorial board. Thoracic tumours. WHO classification of tumours. 5th ed. IARC; 2021.Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 2013;499:214–8.PubMed PubMed Central Google Scholar Weinstein BI. Addiction to oncogenes—the Achilles heel of cancer. Science. 2002;297:64–5.Google Scholar Paez JG, Jänne PA, Lee JC, Tracy S, Greulich H, Gabriel S, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science. 2004;304:1497–500.PubMed Google Scholar Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med. 2004;350:2129–39.PubMed Google Scholar Pao W, Miller V, Zakowski M, Doherty J, Politi K, Sarkaria I, et al. EGF receptor gene mutations are common in lung cancers from 'never smokers' and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci USA. 2004;101:13306–11.PubMed PubMed Central Google Scholar Herbst RS, Morgensztern D, Boshoff C. The biology and management of non-small cell lung cancer. Nature. 2018;553:446–54.PubMed Google Scholar Dearden S, Stevens J, Wu Y-L, Blowers D. Mutation incidence and coincidence in non-small-cell lung cancer: meta-analyses by ethnicity and histology (mutMap). Ann Oncol. 2013;24:2371–6.PubMed PubMed Central Google Scholar Chen J, Yang H, Teo ASM, Amer LB, Sherbaf FG, Tan CQ, et al. Genomic landscape of lung adenocarcinoma in East Asians. Nat Genet. 2020;52:177–86.PubMed Google Scholar NCCN clinical practice guidelines in oncology. Non-small cell lung cancer (Version 4.2025) (2025).Kris MG, Johnson BE, Berry LD, Kwiatkowski DJ, Iafrate AJ, Wistuba II, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311:1998–2006.PubMed PubMed Central Google Scholar De Castro G, Kudaba I, Wu YL, Lopes G, Kowalski DM, Turna HZ, et al. Five-year outcomes with pembrolizumab versus chemotherapy as first-line therapy in patients with non-small-cell lung cancer and programmed death ligand-1 tumor proportion score ≥ 1% in the KEYNOTE-042 study. J Clin Oncol. 2023;41:1986–91.PubMed Google Scholar Garassino MC, Gadgeel S, Speranza G, Felip E, Esteban E, Dómine M, et al. Pembrolizumab plus pemetrexed and platinum in nonsquamous non-small-cell lung cancer: 5-year outcomes from the phase 3 KEYNOTE-189 study. J Clin Oncol. 2023;41:1992–8.PubMed PubMed Central Google Scholar Novello S, Kowalski DM, Luft A, Gümüş M, Vicente D, Mazières J, et al. Pembrolizumab plus chemotherapy in squamous non-small-cell lung cancer: 5-year update of the phase III KEYNOTE-407 study. J Clin Oncol. 2023;41:1999–2006.PubMed PubMed Central Google Scholar Berger MF, Mardis ER. The emerging clinical relevance of genomics in cancer medicine. Nat Rev Clin Oncol. 2018;15:353–65.PubMed PubMed Central Google Scholar Rodriguez H, Pennington SR. Revolutionizing precision oncology through collaborative proteogenomics and data sharing. Cell. 2018;173:535–9.PubMed PubMed Central Google Scholar Yi S, Lin S, Li Y, Zhao W, Mills GB, Sahni N. Functional variomics and network perturbation: connecting genotype to phenotype in cancer. Nat Rev Genet. 2017;18:395–410.PubMed PubMed Central Google Scholar Rodriguez H, Zenklusen JC, Staudt LM, Doroshow JH, Lowy DR. The next horizon in precision oncology: proteogenomics to inform cancer diagnosis and treatment. Cell. 2021;184:1661–70.PubMed PubMed Central Google Scholar Dai X, Shen L. Advances and trends in omics technology development. Front Med. 2022;9:911861.Mani DR, Krug K, Zhang B, Satpathy S, Clauser KR, Ding L, et al. Cancer proteogenomics: current impact and future prospects. Nat Rev Cancer. 2022;22:298–313.PubMed Google Scholar Weir BA, Woo MS, Getz G, Perner S, Ding L, Beroukhim R, et al. Characterizing the cancer genome in lung adenocarcinoma. Nature. 2007;450:893–8.PubMed PubMed Central Google Scholar Ding L, Getz G, Wheeler DA, Mardis ER, McLellan MD, Cibulskis K, et al. Somatic mutations affect key pathways in lung adenocarcinoma. Nature. 2008;455:1069–75.PubMed PubMed Central Google Scholar Imielinski M, Berger AH, Hammerman PS, Hernandez B, Pugh TJ, Hodis E, et al. Mapping the hallmarks of lung adenocarcinoma with massively parallel sequencing. Cell. 2012;150:1107–20.PubMed PubMed Central Google Scholar Govindan R, Ding L, Griffith M, Subramanian J, Dees ND, Kanchi KL, et al. Genomic landscape of non-small cell lung cancer in smokers and never-smokers. Cell. 2012;150:1121–34.PubMed PubMed Central Google Scholar Hammerman PS, Voet D, Lawrence MS, Voet D, Jing R, Cibulskis K, et al. Comprehensive genomic characterization of squamous cell lung cancers. Nature. 2012;489:519–25.Google Scholar Collisson EA, Campbell JD, Brooks AN, Berger AH, Lee W, Chmielecki J, et al. Comprehensive molecular profiling of lung adenocarcinoma: the cancer genome atlas research network. Nature. 2014;511:543–50.Google Scholar Campbell JD, Alexandrov A, Kim J, Wala J, Berger AH, Pedamallu CS, et al. Distinct patterns of somatic genome alterations in lung adenocarcinomas and squamous cell carcinomas. Nat Genet. 2016;48:607–16.PubMed PubMed Central Google Scholar Zhang XC, Wang J, Shao GG, Wang Q, Qu X, Wang B, et al. Comprehensive genomic and immunological characterization of Chinese non-small cell lung cancer patients. Nat Commun. 2019;10:1–12.Google Scholar Tokumo M, Toyooka S, Kiura K, Shigematsu H, Tomii K, Aoe M, et al. The relationship between epidermal growth factor receptor mutations and clinicopathologic features in non-small cell lung cancers. Clin Cancer Res. 2005;11:1167–73.PubMed Google Scholar Davies H, Hunter C, Smith R, Stephens P, Greenman C, Bignell G, et al. Somatic mutations of the protein kinase gene family in human lung cancer. Cancer Res. 2005;65:7591–5.PubMed Google Scholar Chakravarty D, Gao J, Phillips S, Kundra R, Zhang H, Wang J, et al. OncoKB: a precision oncology knowledge base. JCO Precis Oncol. 2017;PO.17.00011.Suehnholz SP, Nissan MH, Zhang H, Kundra R, Nandakumar S, Lu C, et al. Quantifying the expanding landscape of clinical actionability for patients with cancer. Cancer Discov. 2024;14:49–65.PubMed Google Scholar Pao W, Wang TY, Riely GJ, Miller VA, Pan Q, Ladanyi M, et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS Med. 2005;2:e17.PubMed PubMed Central Google Scholar Zhou J, Sanchez-Vega F, Caso R, Tan KS, Brandt WS, Jones GD, et al. Analysis of tumor genomic pathway alterations using broad-panel next-generation sequencing in surgically resected lung adenocarcinoma. Clin Cancer Res. 2019;25:7475–84.PubMed PubMed Central Google Scholar LoPiccolo J, Gusev A, Christiani DC, Jänne PA. Lung cancer in patients who have never smoked—an emerging disease. Nat Rev Clin Oncol. 2024;21:121–46.PubMed PubMed Central Google Scholar Harada G, Yang SR, Cocco E, Drilon A. Rare molecular subtypes of lung cancer. Nat Rev Clin Oncol. 2023;20:229–49.PubMed PubMed Central Google Scholar Carrot-Zhang J, Yao X, Devarakonda S, Deshpande A, Damrauer JS, Silva TC, et al. Whole-genome characterization of lung adenocarcinomas lacking the RTK/RAS/RAF pathway. Cell Rep. 2021;34:108707.PubMed PubMed Central Google Scholar Marchesini M, Ogoti Y, Fiorini E, Aktas Samur A, Nezi L, D’Anca M, et al. ILF2 is a regulator of RNA splicing and DNA damage response in 1q21-amplified multiple myeloma. Cancer Cell. 2017;32:88–100.e6.PubMed PubMed Central Google Scholar Seker-Cin H, Tay TKY, Kazdal D, Kluck K, Ball M, Neumann O, et al. Analysis of rare fusions in NSCLC: genomic architecture and clinical implications. Lung Cancer. 2023;184:107317.Garber ME, Troyanskaya OG, Schluens K, Petersen S, Thaesler Z, Pacyna-Gengelbach M, et al. Diversity of gene expression in adenocarcinoma of the lung. Proc Natl Acad Sci USA. 2001;98:13784–9.PubMed PubMed Central Google Scholar Bhattacharjee A, Richards WG, Staunton J, Li C, Monti S, Vasa P, et al. Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses. Proc Natl Acad Sci USA. 2001;98:13790–5.PubMed PubMed Central Google Scholar Tomida S, Koshikawa K, Yatabe Y, Harano T, Ogura N, Mitsudomi T, et al. Gene expression-based, individualized outcome prediction for surgically treated lung cancer patients. Oncogene. 2004;23:5360–70.PubMed Google Scholar Hayes DN, Monti S, Parmigiani G, Gilks CB, Naoki K, Bhattacharjee A, et al. Gene expression profiling reveals reproducible human lung adenocarcinoma subtypes in multiple independent patient cohorts. J Clin Oncol. 2006;24:5079–90.PubMed Google Scholar Beer DG, Kardia SLR, Huang C-C, Giordano TJ, Levin AM, Misek DE, et al. Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nat Med. 2002;8:816–24.PubMed Google Scholar Takeuchi T, Tomida S, Yatabe Y, Kosaka T, Osada H, Yanagisawa K, et al. Expression profile-defined classification of lung adenocarcinoma shows close relationship with underlying major genetic changes and clinicopathologic behaviors. J Clin Oncol. 2006;24:1679–88.PubMed Google Scholar Roh W, Geffen Y, Cha H, Miller M, Anand S, Kim J, et al. High-resolution profiling of lung adenocarcinoma identifies expression subtypes with specific biomarkers and clinically relevant vulnerabilities. Cancer Res. 2022;82:3917–31.PubMed PubMed Central Google Scholar Koyama S, Akbay EA, Li YY, Aref AR, Skoulidis F, Herter-Sprie GS, et al. STK11/LKB1 deficiency promotes neutrophil recruitment and proinflammatory cytokine production to suppress T-cell activity in the lung tumor microenvironment. Cancer Res. 2016;76:999–1008.PubMed PubMed Central Google Scholar Skoulidis F, Goldberg ME, Greenawalt DM, Hellmann MD, Awad MM, Gainor JF, et al. STK11/LKB1 mutations and PD-1 inhibitor resistance in KRAS-mutant lung adenocarcinoma. Cancer Discov. 2018;8:822–35.PubMed PubMed Central Google Scholar Rizvi H, Sanchez-Vega F, La K, Chatila W, Jonsson P, Halpenny D, et al. Molecular determinants of response to anti-programmed cell death (PD)-1 and anti-programmed death-ligand 1 (PD-L1) blockade in patients with non-small-cell lung cancer profiled with targeted next-generation sequencing. J Clin Oncol. 2018;36:633–41.PubMed PubMed Central Google Scholar Ricciuti B, Arbour KC, Lin JJ, Vajdi A, Vokes N, Hong L, et al. Diminished efficacy of programmed death-(ligand)1 inhibition in STK11- and KEAP1-mutant lung adenocarcinoma is affected by KRAS mutation status. J Thorac Oncol. 2022;17:399–410.PubMed Google Scholar Ricciuti B, Garassino MC. Precision immunotherapy for STK11/KEAP1-mutant NSCLC. J Thorac Oncol. 2024;19:877–82.PubMed Google Scholar Skoulidis F, Araujo HA, Do MT, Qian Y, Sun X, Cobo AG, et al. CTLA4 blockade abrogates KEAP1/STK11-related resistance to PD-(L)1 inhibitors. Nature. 2024;635:462–71.Skoulidis F, Byers LA, Diao L, Papadimitrakopoulou VA, Tong P, Izzo J, et al. Co-occurring genomic alterations define major subsets of KRAS-mutant lung adenocarcinoma with distinct biology, immune profiles and therapeutic vulnerabilities. Cancer Discov. 2015;5:861–78.Google Scholar Scalera S, Mazzotta M, Corleone G, Sperati F, Terrenato I, Krasniqi E, et al. KEAP1 and TP53 frame genomic, evolutionary and immunologic subtypes of lung adenocarcinoma with different sensitivity to immunotherapy. J Thorac Oncol. 2021;16:2065–77.PubMed Google Scholar Wilkerson MD, Yin X, Hoadley KA, Liu Y, Hayward MC, Cabanski CR, et al. Lung squamous cell carcinoma mRNA expression subtypes are reproducible, clinically important and correspond to normal cell types. Clin Cancer Res. 2010;16:4864–75.PubMed PubMed Central Google Scholar Havel JJ, Chowell D, Chan TA. The evolving landscape of biomarkers for checkpoint inhibitor immunotherapy. Nat Rev Cancer. 2019;19:133–50.PubMed PubMed Central Google Scholar Safa H, Abu Rous F, Belani N, Borghaei H, Gadgeel S, Halmos B. Emerging biomarkers in immune oncology to guide lung cancer management. Target Oncol. 2023;18:25–49.PubMed Google Scholar Sholl LM. Biomarkers of response to checkpoint inhibitors beyond PD-L1 in lung cancer. Mod Pathol. 2022;35:66–74.PubMed Google Scholar Fehrenbacher L, Spira A, Ballinger M, Kowanetz M, Vansteenkiste J, Mazieres J, et al. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. Lancet. 2016;387:1837–46.PubMed Google Scholar Socinski MA, Jotte RM, Cappuzzo F, Orlandi F, Stroyakovskiy D, Nogami N, et al. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N. Engl J Med. 2018;378:2288–301.PubMed Google Scholar Kowanetz M, Zou W, Mccleland M, Gandara DR, Gadgeel S, Rittmeyer A, et al. Pre-existing immunity measured by Teff gene expression in tumor tissue is associated with atezolizumad efficacy in NSCLC. J. Thorac. Oncol. 2017;S1817–8. https://doi.org/10.1016/j.jtho.2017.09.485.Arolt C, Dugan M, Wild R, Richartz V, Holz B, Scheel AH, et al. KEAP1/NFE2L2 pathway signature outperforms KEAP1/NFE2L2 mutation status and reveals alternative pathway-activating mutations in NSCLC. J Thorac Oncol. 2023;18:1550–67.PubMed Google Scholar Vavilis T, Petre ML, Vatsellas G, Ainatzoglou A, Stamoula E, Sachinidis A, et al. Lung cancer proteogenomics: shaping the future of clinical investigation. Cancers. 2024;16:1–17.Google Scholar Sharpnack MF, Ranbaduge N, Srivastava A, Cerciello F, Codreanu SG, Liebler DC, et al. Proteogenomic analysis of surgically resected lung adenocarcinoma. J Thorac Oncol. 2018;13:1519–29.PubMed PubMed Central Google Scholar Middleton G, Fletcher P, Popat S, Savage J, Summers Y, Greystoke A, et al. The national lung matrix trial of personalized therapy in lung cancer. Nature. 2020;583:807–12.PubMed PubMed Central Google Scholar Edelman MJ, Redman MW, Albain KS, McGary EC, Rafique NM, Petro D, et al. SWOG S1400C (NCT02154490)-A phase II study of palbociclib for previously treated cell cycle gene alteration-positive patients with stage IV squamous cell lung cancer (lung-MAP substudy). J Thorac Oncol. 2019;14:1853–9.PubMed PubMed Central Google Scholar Pawson T, Warner N. Oncogenic re-wiring of cellular signaling pathways. Oncogene. 2007;26:1268–75.PubMed Google Scholar Gillette MA, Satpathy S, Cao S, Dhanasekaran SM, Vasaikar SV, Krug K, et al. Proteogenomic characterization reveals therapeutic vulnerabilities in lung adenocarcinoma. Cell. 2020;182:200–25.e35.PubMed PubMed Central Google Scholar Yang S-R, Gedvilaite E, Ptashkin R, Chang J, Ziegler J, Mata DA, et al. Microsatellite instability and mismatch repair deficiency define a distinct subset of lung cancers characterized by smoking exposure, high tumor mutational burden and recurrent somatic MLH1 inactivation. J Thorac Oncol. 2024;19:409–24.Ramalingam SS, Thara E, Awad MM, Dowlati A, Haque B, Stinchcombe TE, et al. JASPER: phase 2 trial of first-line niraparib plus pembrolizumab in patients with advanced non-small cell lung cancer. Cancer. 2022;128:65–74.PubMed Google Scholar Ramalingam SS, Novello S, Guclu SZ, Bentsion D, Zvirbule Z, Szilasi M, et al. Veliparib in combination with platinum-based chemotherapy for first-line treatment of advanced squamous cell lung cancer: a randomized, multicenter phase III study. J Clin Oncol. 2021;39:3633–44.PubMed PubMed Central Google Scholar Owonikoko TK, Redman MW, Byers LA, Hirsch FR, Mack PC, Schwartz LH, et al. Phase 2 study of talazoparib in patients with homologous recombination repair-deficient squamous cell lung cancer: lung-MAP substudy S1400G. Clin Lung Cancer. 2021;22:187–94.e1.PubMed PubMed Central Google Scholar Zhao G, Feng E, Liu Y. Efficacy and safety of veliparib combined with traditional chemotherapy for treating patients with lung cancer: a comprehensive review and meta-analysis. PeerJ. 2023;11:e16402.PubMed PubMed Central Google Scholar Garcia-Campelo R, Arrieta O, Massuti B, Rodriguez-Abreu D, Granados ALO, Majem M, et al. Combination of gefitinib and olaparib versus gefitinib alone in EGFR mutant non-small-cell lung cancer (NSCLC): a multicenter, randomized phase II study (GOAL). Lung Cancer. 2020;150:62–9.PubMed Google Scholar Ahn M-J, Bondarenko I, Kalinka E, Cho BC, Sugawara S, Gálffy G, et al. Durvalumab in combination with olaparib versus durvalumab alone as maintenance therapy in metastatic NSCLC: the phase 2 ORION study. J Thorac Oncol. 2023;18:1594–606.PubMed Google Scholar Kuser-Abali G, Gong L, Yan J, Liu Q, Zeng W, Williamson A, et al. An EZH2-mediated epigenetic mechanism behind p53-dependent tissue sensitivity to DNA damage. Proc Natl Acad Sci USA. 2018;115:3452–7.PubMed PubMed Central Google Scholar Jin CH, Krishnaiah M, Sreenu D, Subrahmanyam VB, Rao KS, Lee HJ, et al. Discovery of N-((4-([1,2,4]triazolo[1,5-a]pyridin-6-yl)-5-(6-methylpyridin-2-yl)-1H-imidazol-2-yl)methyl)-2-fluoroaniline (EW-7197): a highly potent, selective and orally bioavailable inhibitor of TGF-β type I receptor kinase as cancer immunotherapeutic. J Med Chem. 2014;57:4213–38.PubMed Google Scholar Paik PK, Fan P-D, Qeriqi B, Namakydoust A, Daly B, Ahn L, et al. Targeting NFE2L2/KEAP1 mutations in advanced NSCLC with the TORC1/2 inhibitor TAK-228. J Thorac Oncol. 2023;18:516–26.PubMed Google Scholar Daley BR, Sealover NE, Sheffels E, Hughes JM, Gerlach D, Hofmann MH, et al. SOS1 inhibition enhances the efficacy of KRASG12C inhibitors and delays resistance in lung adenocarcinoma. Cancer Res. 2025;85:118–33.Chen Y-NP, LaMarche MJ, Chan HM, Fekkes P, Garcia-Fortanet J, Acker MG, et al. Allosteric inhibition of SHP2 phosphatase inhibits cancers driven by receptor tyrosine kinases. Nature. 2016;535:148–52.PubMed Google Scholar Richards CE, Elamin YY, Carr A, Gately K, Rafee S, Cremona M, et al. Protein tyrosine phosphatase non-receptor 11 (PTPN11/Shp2) as a driver oncogene and a novel therapeutic target in non-small cell lung cancer (NSCLC). Int J Mol Sci. 2023;24:10545.Xu JY, Zhang C, Wang X, Zhai L, Ma Y, Mao Y, et al. Integrative proteomic characterization of human lung adenocarcinoma. Cell. 2020;182:245–.e17.PubMed Google Scholar Chen YJ, Roumeliotis TI, Chang YH, Chen CT, Han CL, Lin MH, et al. Proteogenomics of non-smoking lung cancer in East Asia delineates molecular signatures of pathogenesis and progression. Cell. 2020;182:226–44.e17.PubMed Google Scholar Soltis AR, Bateman NW, Liu J, Nguyen T, Franks TJ, Zhang X, et al. Proteogenomic analysis of lung adenocarcinoma reveals tumor heterogeneity, survival determinants and therapeutically relevant pathways. Cell Rep. Med. 2022;3:100819.PubMed PubMed Central Google Scholar Zhang H, Liu C, Wang S, Wang Q, Feng X, Jiang H, et al. Proteogenomic analysis of air-pollution-associated lung cancer reveals prevention and therapeutic opportunities. eLife. 2024;13:95453.Zhang C, Zhang J, Xu FP, Wang YG, Xie Z, Su J, et al. Genomic landscape and immune microenvironment features of preinvasive and early invasive lung adenocarcinoma. J Thorac Oncol. 2019;14:1912–23.PubMed PubMed Central Google Scholar Nishimura T, Nakamura H, Tan KT, Zhuo DW, Fujii K, Koizumi H, et al. A proteogenomic profile of early lung adenocarcinomas by protein co-expression network and genomic alteration analysis. Sci Rep. 2020;10:1–13.Google Scholar Mazzilli SA, Rahal Z, Rouhani MJ, Janes SM, Kadara H, Dubinett SM, et al. Translating premalignant biology to accelerate non-small-cell lung cancer interception. Nat Rev Cancer. 2025;25:379–92.PubMed Google Scholar Asada K, Kaneko S, Takasawa K, Shiraishi K, Shinkai N, Shimada Y, et al. Multi-omics and clustering analyses reveal the mechanisms underlying unmet needs for patients with lung adenocarcinoma and identify potential therapeutic targets. Mol Cancer. 2024;23:182.Park SJ, Ju S, Goh SH, Yoon BH, Park JL, Kim JH, et al. Proteogenomic characterization reveals estrogen signaling as a target for never-smoker lung adenocarcinoma patients without EGFR or ALK alterations. Cancer Res. 2024;84:1491–503.PubMed Google Scholar Kim H, Lee W, Kim Y, Lee SJ, Choi W, Lee GK, et al. Proteogenomic characterization identifies clinical subgroups in EGFR and ALK wild-type never-smoker lung adenocarcinoma. Exp Mol Med. 2024;56:2082–95.Siegmund D, Wagner J, Wajant H. TNF receptor associated factor 2 (TRAF2) signaling in cancer. Cancers. 2022;14:4055.Wang J, Li RZ, Wang WJ, Pan HD, Xie C, Yau LF, et al. CERS4 predicts positive anti-PD-1 response and promotes immunomodulation through Rhob-mediated suppression of CD8+Tim3+ exhausted T cells in non-small cell lung cancer. Pharm Res. 2023;194:106850.Google Scholar Tsherniak A, Vazquez F, Montgomery PG, Weir BA, Kryukov G, Cowley GS, et al. Defining a cancer dependency map. Cell. 2017;170:564–25.e16.PubMed PubMed Central Google Scholar Baldelli E, Bellezza G, Haura EB, Crinó L, Cress WD, Deng J, et al. Functional signaling pathway analysis of lung adenocarcinomas identifies novel therapeutic targets for KRAS mutant tumors. Oncotarget. 2015;6:32368–79.PubMed PubMed Central Google Scholar Reggiani F, Sauta E, Torricelli F, Zanetti E, Tagliavini E, Santandrea G, et al. An integrative functional genomics approach reveals EGLN1 as a novel therapeutic target in KRAS mutated lung adenocarcinoma. Mol Cancer. 2021;20:1–6.Google Scholar Cardnell RJG, Behrens C, Diao L, Fan YH, Tang X, Tong P, et al. An integrated molecular analysis of lung adenocarcinomas identifies potential therapeutic targets among TTF1-negative tumors, including DNA repair proteins and Nrf2. Clin Cancer Res. 2015;21:3480–91.PubMed PubMed Central Google Scholar Zhang DD. Thirty years of NRF2: advances and therapeutic challenges. Nat Rev Drug Discov. 2025;24:421–44.Tanaka I, Dayde D, Tai MC, Mori H, Solis LM, Tripathi SC, et al. SRGN-triggered aggressive and immunosuppressive phenotype in a subset of TTF-1-negative lung adenocarcinomas. J Natl Cancer Inst. 2022;114:290–301.PubMed Google Scholar Stewart PA, Welsh EA, Slebos RJC, Fang B, Izumi V, Chambers M, et al. Proteogenomic landscape of squamous cell lung cancer. Nat Commun. 2019;10:1–17.Google Scholar Nishimura T, Fujii K, Nakamura H, Naruki S, Sakai H, Kimura H, et al. Protein co-expression network-based profiles revealed from laser-microdissected cancerous cells of lung squamous-cell carcinomas. Sci Rep. 2021;11:1–15.Google Scholar Satpathy S, Krug K, Jean Beltran PM, Savage SR, Petralia F, Kumar-Sinha C, et al. A proteogenomic portrait of lung squamous cell carcinoma. Cell. 2021;184:4348–71.e40.PubMed PubMed Central Google Scholar Chen F, Zhang Y, Parra E, Rodriguez J, Behrens C, Akbani R, et al. Multiplatform-based molecular subtypes of non-small-cell lung cancer. Oncogene. 2017;36:1384–93.PubMed Google Scholar Lehtiö J, Arslan T, Siavelis I, Pan Y, Socciarelli F, Berkovska O, et al. Proteogenomics of non-small cell lung cancer reveals molecular subtypes associated with specific therapeutic targets and immune-evasion mechanisms. Nat Cancer. 2021;2:1224–42.PubMed PubMed Central Google Scholar Song KJ, Choi S, Kim K, Hwang HS, Chang E, Park JS, et al. Proteogenomic analysis reveals non-small cell lung cancer subtypes predicting chromosome instability and tumor microenvironment. Nat Commun. 2024;15:10164.Starck SR, Shastri N. Nowhere to hide: unconventional translation yields cryptic peptides for immune surveillance. Immunol Rev. 2016;272:8–16.PubMed PubMed Central Google Scholar Frankell AM, Dietzen M, Al Bakir M, Lim EL, Karasaki T, Ward S, et al. The evolution of lung cancer and impact of subclonal selection in TRACERx. Nature. 2023;616:525–33.PubMed PubMed Central Google Scholar Jamal-Hanjani M, Wilson GA, McGranahan N, Birkbak NJ, Watkins TBK, Veeriah S, et al. Tracking the evolution of non-small-cell lung cancer. N. Engl J Med. 2017;376:2109–21.PubMed Google Scholar Wu F, Fan J, He Y, Xiong A, Yu J, Li Y, et al. Single-cell profiling of tumor heterogeneity and the microenvironment in advanced non-small cell lung cancer. Nat Commun. 2021;12:1–11.Google Scholar De Zuani M, Xue H, Park JS, Dentro SC, Seferbekova Z, Tessier J, et al. Single-cell and spatial transcriptomics analysis of non-small cell lung cancer. Nat Commun. 2024;15:4388.Zhang L, Zhang Y, Wang C, Yang Y, Ni Y, Wang Z, et al. Integrated single-cell RNA sequencing analysis reveals distinct cellular and transcriptional modules associated with survival in lung cancer. Signal Transduct Target Ther. 2022;7:1–13.Google Scholar Vaccaro A, Rahal Z, Kadara H, Cascone T. A roadmap to precision immunotherapy for early-stage non-small cell lung cancer. Cancer Discov. 2025;15:884–9.PubMed Google Scholar Caushi JX, Zhang J, Ji Z, Vaghasia A, Zhang B, Hsiue EHC, et al. Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature. 2021;596:126–32.PubMed PubMed Central Google Scholar Sinjab A, Han G, Treekitkarnmongkol W, Hara K, Brennan PM, Dang M, et al. Resolving the spatial and cellular architecture of lung adenocarcinoma by multiregion single-cell sequencing. Cancer Discov. 2021;11:2506–23.PubMed PubMed Central Google Scholar Song X, Xiong A, Wu F, Li X, Wang J, Jiang T, et al. Spatial multi-omics revealed the impact of tumor ecosystem heterogeneity on immunotherapy efficacy in patients with advanced non-small cell lung cancer treated with bispecific antibody. J Immunother Cancer. 2023;11:e006234.Yan Y, Sun D, Hu J, Chen Y, Sun L, Yu H, et al. Multi-omic profiling highlights factors associated with resistance to immunochemotherapy in non-small-cell lung cancer. Nat Genet. 2025;57:126–39.PubMed Google Scholar Chen JH, Nieman LT, Spurrell M, Jorgji V, Elmelech L, Richieri P, et al. Human lung cancer harbors spatially organized stem-immunity hubs associated with response to immunotherapy. Nat Immunol. 2024;25:644–58.PubMed PubMed Central Google Scholar Jordan EJ, Kim HR, Arcila ME, Barron D, Chakravarty D, Gao JJ, et al. Prospective comprehensive molecular characterization of lung adenocarcinomas for efficient patient matching to approved and emerging therapies. Cancer Discov. 2017;7:596–609.PubMed PubMed Central Google Scholar Al Bakir M, Huebner A, Martínez-Ruiz C, Grigoriadis K, Watkins TBK, Pich O, et al. The evolution of non-small cell lung cancer metastases in TRACERx. Nature. 2023;616:534–42.PubMed PubMed Central Google Scholar Lengel HB, Mastrogiacomo B, Connolly JG, Tan KS, Liu Y, Fick CN, et al. Genomic mapping of metastatic organotropism in lung adenocarcinoma. Cancer Cell. 2023;41:970–85.e3.PubMed PubMed Central Google Scholar Peuget S, Zhou X, Selivanova G. Translating p53-based therapies for cancer into the clinic. Nat Rev Cancer. 2024;24:192–215.PubMed Google Scholar Tuval A, Strandgren C, Heldin A, Palomar-Siles M, Wiman KG. Pharmacological reactivation of p53 in the era of precision anticancer medicine. Nat Rev Clin Oncol. 2024;21:106–20.PubMed Google Scholar Rivlin N, Katz S, Doody M, Sheffer M, Horesh S, Molchadsky A, et al. Rescue of embryonic stem cells from cellular transformation by proteomic stabilization of mutant p53 and conversion into WT conformation. Proc Natl Acad Sci USA. 2014;111:7006–11.PubMed PubMed Central Google Scholar Napolitano R, De Matteis S, Carloni S, Bruno S, Abbati G, Capelli L, et al. Kevetrin induces apoptosis in TP53 wild‑type and mutant acute myeloid leukemia cells. Oncol Rep. 2020;44:1561–73.PubMed PubMed Central Google Scholar Ingelshed K, Melssen MM, Kannan P, Chandramohan A, Partridge AW, Jiang L, et al. MDM2/MDMX inhibition by sulanemadlin synergizes with anti-programmed death 1 immunotherapy in wild-type p53 tumors. IScience. 2024;27:109862.PubMed PubMed Central Google Scholar Li K, Crews CM. PROTACs: past, present and future. Chem Soc Rev. 2022;51:5214–36.PubMed PubMed Central Google Scholar Park H, Shapiro GI, Gao X, Mahipal A, Starr J, Furqan M, et al. Phase Ib study of eprenetapopt (APR-246) in combination with pembrolizumab in patients with advanced or metastatic solid tumors. ESMO Open. 2022;7:100573.PubMed PubMed Central Google Scholar Garcia-Manero G, Goldberg AD, Winer ES, Altman JK, Fathi AT, Odenike O, et al. Eprenetapopt combined with venetoclax and azacitidine in TP53-mutated acute myeloid leukaemia: a phase 1, dose-finding and expansion study. Lancet Haematol. 2023;10:e272–83.PubMed Google Scholar Puzio-Kuter AM, Xu L, McBrayer MK, Dominique R, Li HH, Fahr BJ, et al. Restoration of the tumor suppressor function of Y220C-mutant p53 by rezatapopt, a small molecule reactivator. Cancer Discov. 2025;15:1159–79.Pouremamali F, Pouremamali A, Dadashpour M, Soozangar N, Jeddi F. An update of Nrf2 activators and inhibitors in cancer prevention/promotion. Cell Commun Signal. 2022;20:1–16.Google Scholar Goeman F, De Nicola F, Scalera S, Sperati F, Gallo E, Ciuffreda L, et al. Mutations in the KEAP1-NFE2L2 pathway define a molecular subset of rapidly progressing lung adenocarcinoma. J Thorac Oncol. 2019;14:1924–34.PubMed Google Scholar Krall EB, Wang B, Munoz DM, Ilic N, Raghavan S, Niederst MJ, et al. KEAP1 loss modulates sensitivity to kinase-targeted therapy in lung cancer. Elife 2017;6:e18970.Nadal E, Palmero R, Muñoz-Pinedo C. Mutations in the antioxidant KEAP1/NRF2 pathway define an aggressive subset of NSCLC resistant to conventional treatments. J Thorac Oncol. 2019;14:1881–3.PubMed Google Scholar Frank R, Scheffler M, Merkelbach-Bruse S, Ihle MA, Kron A, Rauer M, et al. Clinical and pathological characteristics of KEAP1- and NFE2L2-mutated non-small cell lung carcinoma (NSCLC). Clin Cancer Res. 2018;24:3087–96.PubMed Google Scholar Jeong Y, Hoang NT, Lovejoy A, Stehr H, Newman AM, Gentles AJ, et al. Role of KEAP1/NRF2 and TP53 mutations in lung squamous cell carcinoma development and radiation resistance. Cancer Discov. 2017;7:86–101.PubMed Google Scholar Cyriac R, Lee K. Glutaminase inhibition as potential cancer therapeutics: current status and future applications. J Enzyme Inhib Med Chem. 2024;39:2290911.Gouda MA, Voss MH, Tawbi H, Gordon M, Tykodi SS, Lam ET, et al. A phase I/II study of the safety and efficacy of telaglenastat (CB-839) in combination with nivolumab in patients with metastatic melanoma, renal cell carcinoma and non-small-cell lung cancer. ESMO Open. 2025;10:104536.PubMed PubMed Central Google Scholar Lv B, Xing S, Wang Z, Zhang A, Wang Q, Bian Y, et al. NRF2 inhibitors: recent progress, future design and therapeutic potential. Eur J Med Chem. 2024;279:116822.PubMed Google Scholar Ji J, Ma S, Zhu Y, Zhao J, Tong Y, You Q, et al. ARE-PROTACs enable Co-degradation of an Nrf2-MafG heterodimer. J Med Chem. 2023;66:6070–81.PubMed Google Scholar Baek K, Schulman BA. Molecular glue concept solidifies. Nat Chem Biol. 2020;16:2–3.PubMed Google Scholar Aboulkassim T, Tian X, Liu Q, Qiu D, Hancock M, Wu JH, et al. A NRF2 inhibitor selectively sensitizes KEAP1 mutant tumor cells to cisplatin and gefitinib by restoring NRF2-inhibitory function of KEAP1 mutants. Cell Rep.2023;42:113104.PubMed Google Scholar Yang X, Liu Y, Cao J, Wu C, Tang L, Bian W, et al. Targeting epigenetic and post-translational modifications of NRF2: key regulatory factors in disease treatment. Cell Death Discov. 2025;11:1–16.Google Scholar Bar-Peled L, Kemper EK, Suciu RM, Vinogradova EV, Backus KM, Horning BD, et al. Chemical proteomics identifies druggable vulnerabilities in a genetically defined cancer. Cell. 2017;171:696–709.e23.PubMed PubMed Central Google Scholar Weiss-Sadan T, Ge M, Hayashi M, Gohar M, Yao CH, de Groot A, et al. NRF2 activation induces NADH-reductive stress, providing a metabolic vulnerability in lung cancer. Cell Metab. 2023;35:487–503.e7.PubMed PubMed Central Google Scholar Momcilovic M, Bailey ST, Lee JT, Fishbein MC, Braas D, Go J, et al. The GSK3 signaling axis regulates adaptive glutamine metabolism in lung squamous cell carcinoma. Cancer Cell. 2018;33:905–21.e5.PubMed PubMed Central Google Scholar Trelford CB, Shepherd TG. LKB1 biology: assessing the therapeutic relevancy of LKB1 inhibitors. Cell Commun Signal. 2024;22:1–20.Google Scholar Chaudhary S, Kulkarni A. Metformin: past, present and future. Curr Diab Rep. 2024;24:119–30.PubMed Google Scholar Mallik R, Chowdhury TA. Metformin in cancer. Diabetes Res Clin Pract. 2018;143:409–19.PubMed Google Scholar Giraud JS, Bièche I, Pasmant É, Tlemsani C. NF1 alterations in cancers: therapeutic implications in precision medicine. Expert Opin Investig Drugs. 2023;32:941–57.PubMed Google Scholar Malone HA, Roberts CWM. Chromatin remodellers as therapeutic targets. Nat Rev Drug Discov. 2024;23:661–81.PubMed Google Scholar Stark R, Grzelak M, Hadfield J. RNA sequencing: the teenage years. Nat Rev Genet. 2019;20:631–56.PubMed Google Scholar Kennedy JJ, Abbatiello SE, Kim K, Yan P, Whiteaker JR, Lin C, et al. Demonstrating the feasibility of large-scale development of standardized assays to quantify human proteins. Nat Methods. 2014;11:149–55.PubMed Google Scholar Baysoy A, Bai Z, Satija R, Fan R. The technological landscape and applications of single-cell multi-omics. Nat Rev Mol Cell Biol. 2023;24:695–713.PubMed Google Scholar Wu X, Yang X, Dai Y, Zhao Z, Zhu J, Guo H, et al. Single-cell sequencing to multi-omics: technologies and applications. Biomark Res. 2024;12:1–28.Google Scholar Cai Z, Poulos RC, Liu J, Zhong Q. Machine learning for multi-omics data integration in cancer. IScience. 2022;25:103798.PubMed PubMed Central Google Scholar Altman N, Krzywinski M. The curse(s) of dimensionality this-month. Nat Methods. 2018;15:399–400.PubMed Google Scholar Yamada R, Okada D, Wang J, Basak T, Koyama S. Interpretation of omics data analyses. J Hum Genet. 2021;66:93–102.PubMed Google Scholar Ahmad A, Fröhlich H. Integrating heterogeneous omics data via statistical inference and learning techniques. Genom Comput Biol. 2016;2:32.Google Scholar Meng C, Zeleznik OA, Thallinger GG, Kuster B, Gholami AM, Culhane AC. Dimension reduction techniques for the integrative analysis of multi-omics data. Brief Bioinform. 2016;17:628–41.PubMed PubMed Central Google Scholar Correa-Aguila R, Alonso-Pupo N, Hernández-Rodríguez EW. Multi-omics data integration approaches for precision oncology. Mol Omics. 2022;18:469–79.PubMed Google Scholar Rappoport N, Shamir R. Multi-omic and multi-view clustering algorithms: review and cancer benchmark. Nucleic Acids Res. 2018;46:10546–62.PubMed PubMed Central Google Scholar Duan R, Gao L, Gao Y, Hu Y, Xu H, Huang M, et al. Evaluation and comparison of multi-omics data integration methods for cancer subtyping. 2021;17:e1009224.Ritchie MD, Holzinger ER, Li R, Pendergrass SA, Kim D. Methods of integrating data to uncover genotype-phenotype interactions. Nat Rev Genet. 2015;16:85–97.PubMed Google Scholar Li W, Ballard J, Zhao Y, Long Q. Knowledge-guided learning methods for integrative analysis of multi-omics data. Comput Struct Biotechnol J. 2024;23:1945–50.PubMed PubMed Central Google Scholar Jiang W, Ye W, Tan X, Bao YJ. Network-based multi-omics integrative analysis methods in drug discovery: a systematic review. BioData Min. 2025;18:27.Vandereyken K, Sifrim A, Thienpont B, Voet T. Methods and applications for single-cell and spatial multi-omics. Nat Rev Genet. 2023;24:494–515.PubMed Google Scholar Le Novere N. Quantitative and logic modelling of molecular and gene networks. Nat Rev Genet. 2015;16:146–58.PubMed PubMed Central Google Scholar Kolch W, Halasz M, Granovskaya M, Kholodenko BN. The dynamic control of signal transduction networks in cancer cells. Nat Rev Cancer. 2015;15:515–27.PubMed Google Scholar Fey D, Halasz M, Dreidax D, Kennedy SP, Hastings JF, Rauch N, et al. Signaling pathway models as biomarkers: patient-specific simulations of JNK activity predict the survival of neuroblastoma patients. Sci Signal. 2015;8:1–16.Google Scholar Eduati F, Jaaks P, Wappler J, Cramer T, Merten CA, Garnett MJ, et al. Patient‐specific logic models of signaling pathways from screenings on cancer biopsies to prioritize personalized combination therapies. Mol Syst Biol. 2020;16:1–13.Google Scholar Hayes DF. ScienceDirect biomarker validation and testing. Mol Oncol. 2015;9:950–6.Google Scholar Hayes DF. Defining clinical utility of tumor biomarker tests: a clinician’s viewpoint. J Clin Oncol. 2021;39:238–48.PubMed Google Scholar Simon R. Review of statistical methods for biomarker-driven clinical trials. JCO Precis Oncol. 2019;3:1–9.Cardoso F, van’t Veer LJ, Bogaerts J, Slaets L, Viale G, Delaloge S, et al. 70-gene signature as an aid to treatment decisions in early-stage breast cancer. N. Engl J Med. 2016;375:717–29.PubMed Google Scholar Sparano JA, Gray RJ, Makower DF, Pritchard KI, Albain KS, Hayes DF, et al. Adjuvant chemotherapy guided by a 21-gene expression assay in breast cancer. N. Engl J Med. 2018;379:111–21.PubMed PubMed Central Google Scholar Lillie EO, Patay B, Diamant J, Issell B, Topol EJ, Schork NJ. The n-of-1 clinical trial: the ultimate strategy for individualizing medicine? Per Med. 2011;8:161–73.PubMed PubMed Central Google Scholar Schork NJ. Time for one-person trials. Nature. 2015;520:609–11.PubMed Google Scholar Yurkovich JT, Tian Q, Price ND, Hood L. A systems approach to clinical oncology uses deep phenotyping to deliver personalized care. Nat Rev Clin Oncol. 2020;17:183–94.PubMed Google Scholar Yankeelov TE, Hormuth DA, Lima EABF, Lorenzo G, Wu C, Okereke LC, et al. Designing clinical trials for patients who are not average. IScience. 2024;27:108589.PubMed Google Scholar Lorenzo G, Ahmed SR, Hormuth DA, Vaughn B, Kalpathy-Cramer J, Solorio L, et al. Patient-specific, mechanistic models of tumor growth incorporating artificial intelligence and big data. Annu Rev Biomed Eng. 2024;26:529–60.PubMed Google Scholar Chiou J, Chang YC, Jan YH, Tsai HF, Yang CJ, Huang MS, et al. Overexpression of BZW1 is an independent poor prognosis marker and its down-regulation suppresses lung adenocarcinoma metastasis. Sci Rep. 2019;9:1–11.Google Scholar Wang M, He T, Meng D, Lv W, Ye J, Cheng L, et al. BZW2 modulates lung adenocarcinoma progression through glycolysis-mediated IDH3G lactylation modification. J Proteome Res. 2023;22:3854–65.PubMed Google Scholar Jin K, Li Y, Wei R, Liu Y, Wang S, Tian H. BZW2 promotes malignant progression in lung adenocarcinoma through enhancing the ubiquitination and degradation of GSK3β. Cell Death Discov. 2024;10:1–11.Google Scholar Mo Y, Feng X, Su J, Chen G, Xian L. BZW2 is a potential regulator of non-small cell lung cancer progression. Gene. 2024;935:149055.PubMed Google Scholar Dhanasekaran R, Deutzmann A, Mahauad-Fernandez WD, Hansen AS, Gouw AM, Felsher DW. The MYC oncogene—the grand orchestrator of cancer growth and immune evasion. Nat Rev Clin Oncol. 2022;19:23–36.PubMed Google Scholar de Bono J, Mateo J, Fizazi K, Saad F, Shore N, Sandhu S, et al. Olaparib for metastatic castration-resistant prostate cancer. N. Engl J Med. 2020;382:2091–102.PubMed Google Scholar Fizazi K, Azad AA, Matsubara N, Carles J, Fay AP, De Giorgi U, et al. First-line talazoparib with enzalutamide in HRR-deficient metastatic castration-resistant prostate cancer: the phase 3 TALAPRO-2 trial. Nat Med. 2024;30:257–64.PubMed Google Scholar Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2:401–4.PubMed Google Scholar Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6:pl1.R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. n.d. 2023.Download referencesAcknowledgementsWe would like to thank Dr. Moriah Heller Nissan, OncoKB Scientific Relations Manager, for reviewing Tables 1 and 3 and for her guidance on drug evidence levels assignment.FundingThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Open access funding provided by Università degli Studi di Trieste within the CRUI-CARE Agreement.Author informationAuthors and AffiliationsDepartment of Medical, Surgical and Health Sciences, University of Trieste, Trieste, ItalyAndrea Rocca, Barbara Ruaro, Marco Confalonieri & Daniele GeneraliIRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) ‘Dino Amadori’, Meldola, ItalyLucio CrinòFunctional Cell Biology Laboratory, International Centre for Genetic Engineering and Biotechnology, Trieste, ItalyLuca BragaPulmonology Unit, Azienda Sanitaria Universitaria Giuliano Isontina (ASUGI), Trieste, ItalyFrancesco Salton, Barbara Ruaro, Marco Confalonieri & Paola ConfalonieriBreast and Brain Unit, ASST Cremona, Cremona, ItalyDaniele GeneraliAuthorsAndrea RoccaView author publicationsSearch author on:PubMed Google ScholarLucio CrinòView author publicationsSearch author on:PubMed Google ScholarLuca BragaView author publicationsSearch author on:PubMed Google ScholarFrancesco SaltonView author publicationsSearch author on:PubMed Google ScholarBarbara RuaroView author publicationsSearch author on:PubMed Google ScholarMarco ConfalonieriView author publicationsSearch author on:PubMed Google ScholarDaniele GeneraliView author publicationsSearch author on:PubMed Google ScholarPaola ConfalonieriView author publicationsSearch author on:PubMed Google ScholarContributionsAndrea Rocca, Lucio Crinò, Marco Confalonieri and Daniele Generali conceived the manuscript. Andrea Rocca, Luca Braga, Paola Confalonieri, Francesco Salton and Barbara Ruaro performed the literature review. Andrea Rocca wrote the first manuscript draft. All authors discussed the article contents, revised the draft and approved the final version of the manuscript.Corresponding authorCorrespondence to Andrea Rocca.Ethics declarationsCompeting interestsThe authors declare the following competing interests: Daniele Generali declares personal fees for educational events by Novartis, Lilly, Pfizer, Daiichy-Sankyo, Roche; research funds from Astrazeneca, Novartis and LILT. Andrea Rocca has been salaried through a research fund provided by Novartis to the University of Trieste. The other authors have nothing to declare.Declaration of generative AI and AI-assisted technologies in the writing processDuring the preparation of this work the authors used ChatGPT in order to improve readability of some sentences or paragraphs. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Rights and permissionsOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.Reprints and permissionsAbout this article