Including Tumor Mutational Burden Research Articles

Integrated machine learning to predict the prognosis of lung adenocarcinoma patients based on SARS-COV-2 and lung adenocarcinoma crosstalk genes.

Viruses are widely recognized to be intricately associated with both solid and hematological malignancies in humans. The primary goal of this research is to elucidate the interplay of genes between SARS-CoV-2 infection and lung adenocarcinoma (LUAD), with a preliminary investigation into their clinical significance and underlying molecular mechanisms. Transcriptome data for SARS-CoV-2 infection and LUAD were sourced from public databases. Differentially expressed genes (DEGs) associated with SARS-CoV-2 infection were identified and subsequently overlapped with TCGA-LUAD DEGs to discern the crosstalk genes (CGs). In addition, CGs pertaining to both diseases were further refined using LUAD TCGA and GEO datasets. Univariate Cox regression was conducted to identify genes associated with LUAD prognosis, and these genes were subsequently incorporated into the construction of a prognosis signature using 10 different machine learning algorithms. Additional investigations, including tumor mutation burden assessment, TME landscape, immunotherapy response assessment, as well as analysis of sensitivity to antitumor drugs, were also undertaken. We discovered the risk stratification based on the prognostic signature revealed that the low-risk group demonstrated superior clinical outcomes (p < 0.001). Gene set enrichment analysis results predominantly exhibited enrichment in pathways related to cell cycle. Our analyses also indicated that the low-risk group displayed elevated levels of infiltration by immunocytes (p < 0.001) and superior immunotherapy response (p < 0.001). In our study, we reveal a close association between CGs and the immune microenvironment of LUAD. This provides preliminary insight for further exploring the mechanism and interaction between the two diseases.

Genomic landscape of superficial malignant peripheral nerve sheath tumor

Superficial malignant peripheral nerve sheath tumors (SF-MPNSTs) are rare cancers and can be difficult to distinguish from spindle cell (SCM) or desmoplastic (DM) melanomas. Their biology is poorly understood. We performed whole-exome sequencing (WES) and RNA sequencing (RNA-seq) on SF-MPNST (n=8) and compared these to cases of SCM (n=7), DM (n=8), and deep MPNST (D-MPNST, n=8). Immunohistochemical staining for H3K27me3 and PRAME was also performed. SF-MPNST demonstrated intermediate features between D-MPNST and melanoma. Patients were younger than those with melanoma, and older than those with D-MPNST; outcome was worse and better respectively. SF-MPNST tumor mutational burden (TMB) was higher than D-MPNST and lower than melanoma; differences were significant only between SF-MPNST and SCM (p = 0.0454) and between D-MPNST and SCM (p = 0.001, Dunn's Kruskal-Wallis post-hoc test). Despite having an overlapping mutational profile in some common cancer-associated genes, the COSMIC mutational signatures clustered DM and SCM together with ultraviolet light exposure signatures (SBS7a, 7b), and SF- and D-MPNST together with defective DNA base excision repair (SBS30, 36). RNA-seq revealed differentially expressed genes between SF-MPNST and SCM (1670 genes), DM (831 genes), and D-MPNST (614 genes), some of which hold promise for development as immunohistochemical markers (SOX8, PLCH1) or aids (MLPH, CALB2, SOX11, TBX4). H3K27me3 immunoreactivity was diffusely lost in most D-MPNSTs (7/8, 88%), but showed variable and patchy loss in SF-MPNSTs (2/8, 25%). PRAME was entirely negative in the majority (0+ in 20/31, 65%), including 11/15 melanomas, and showed no significant difference between groups (p=0.105, Kruskal-Wallis test). Expression of immune cell transcripts was upregulated in melanomas relative to MPNSTs. Next-generation sequencing revealed multiple differential features between SF- MPNST, D-MPNST, SCM, and DM, including tumor mutation burden, mutational signatures, and differentially expressed genes. These findings help advance our understanding of disease pathogenesis and improve diagnostic modalities.

AI-powered genomic mutation signature for predicting immune checkpoint inhibitor therapy outcomes in gastroesophageal cancer: a multi-cohort analysis

BackgroundImmune checkpoint inhibitors (ICIs) have significantly transformed the treatment of gastroesophageal cancer (GEC). However, the lack of reliable prognostic biomarkers hinders the ability to predict patient response to ICI therapy.MethodsIn this study, we engineered and validated a genomic mutation signature (GMS) utilizing an innovative artificial intelligence (AI) algorithm to forecast ICI therapy outcomes in GEC patients. We further explored immune profiles across subtypes through comprehensive multiomics analysis. Our investigation of drug sensitivity data from the Genomics of Drug Sensitivity in Cancer (GDSC) database led to the identification of trametinib as a potential therapeutic agent. We subsequently evaluated trametinib’s efficacy in AGS and MKN45 cell lines using Cell Counting Kit-8 (CCK8) assays and clonogenic experiments.ResultsWe developed a GMS by integrating 297 algorithms, enabling autonomous prognosis prediction for GEC patients. The GMS demonstrated consistent performance across three public cohorts, exhibiting high sensitivity and specificity for overall survival (OS) at 6, 12, and 18 months, as shown by Receiver Operator Characteristic Curve (ROC) analysis. Notably, the GMS surpassed traditional clinical and molecular features, including tumor mutational burden (TMB), programmed death-ligand 1 (PD-L1) expression, and microsatellite instability (MSI), in predictive accuracy. Low-risk samples exhibited elevated levels of cytolytic immune cells and heightened immunogenic potential compared to high-risk samples. Our investigation identified trametinib as a potential therapeutic agent. An inverse correlation was observed between GMS and trametinib IC50. Moreover, the high-risk-derived AGS cell line showed increased sensitivity to trametinib compared to the low-risk-derived MKN45 cell line.ConclusionThe GMS utilized in this study successfully demonstrated the ability to reliably predict the survival advantage for patients with GECs undergoing ICI therapy.

Clinical Interest in Exome-Based Analysis of Somatic Mutational Signatures for Non-Small Cell Lung Cancer.

Non-small cell lung cancer (NSCLC) remains the leading cause of cancer-related mortality. This study investigates the clinical interest of whole exome sequencing (WES) for analyzing somatic mutational signatures in patients with advanced or metastatic NSCLC treated with the current standard of care. Exome sequencing data and clinical characteristics from 132 patients with advanced or metastatic NSCLC were analyzed. Somatic mutational signatures including single base substitutions (SBSs), double base substitutions (DBSs), and copy number signatures were evaluated. Structural variations including tumor mutational burden (TMB), the number of neoantigens, TCR clonality, homologous recombination deficiency (HRD), copy number alterations (CNAs), and microsatellite instability (MSI) score were determined. The association between these genomic features, NSCLC subtypes, and patient outcomes (progression-free and overall survival) was evaluated. Exome sequencing offers valuable insights into somatic mutational signatures in NSCLC. This study identified specific signatures associated with a poor response to immune checkpoint inhibitor (ICI) therapy and chemotherapy, potentially aiding treatment selection and identifying patients unlikely to benefit from these approaches.

Pathologic and Molecular Features of Perihilar Cholangiocarcinoma Based on U-P Point Division.

The Japanese Society of Hepato-Biliary-Pancreatic Surgery guidelines propose a classification scheme that differs from the Union for International Cancer Control (UICC) system, in which the anatomic U-P point is the boundary between intrahepatic cholangiocarcinoma and perihilar cholangiocarcinoma (PCC). To investigate whether this classification system improves clinicopathologic and genomic differentiation. Fifty-eight PCC cases defined by the UICC system were collected and classified into intrahepatic PCC (IPCC) and extrahepatic PCC (EPCC) categories using U-P point division. They were analyzed by next-generation sequencing using a panel that targeted 425 cancer-related genes. The IPCC group exhibited a significant larger tumor size compared with the EPCC group (4.67 ± 2.44 cm versus 2.50 ± 0.91 cm, P = .002). The mutation frequency of KRAS proto-oncogene, GTPase (KRAS) Q61 was also significantly higher in the IPCC group than in the EPCC group (16.7% versus 0.0%, P = .03). There were no statistically significant differences in other pathologic features or genomic characteristics, including tumor mutation burden and microsatellite instability. Significant differences in gene mutation rates, such as phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA; 0.0% versus 15.8%, P = .01) and tumor protein p53 (TP53; 34.5% versus 63.2%, P = .04), were observed between PCC and adjacent biliary tract cancers. This study offers valuable insight into the clinicopathologic and genomic features of PCC. It is proposed that the U-P point division may have limited potential to refine the characterization of PCC regarding these features, and that the UICC classification system can readily demonstrate the molecular specificity of PCC.

Uncovering molecular features driving lung adenocarcinoma heterogeneity in patients who formerly smoked

BackgroundAn increasing proportion of lung adenocarcinoma (LUAD) occurs in patients even after they have stopped smoking. Here, we aimed to determine whether tobacco smoking induced changes across LUADs from patients who formerly smoked correspond to different biological and clinical factors.MethodsRandom forest models (RFs) were trained utilizing a smoking associated signature developed from differentially expressed genes between LUAD patients who had never smoked (NS) or currently smoked (CS) from TCGA (n = 193) and BCCA (n = 69) cohorts. The RFs were subsequently applied to 299 and 131 formerly smoking patients from TCGA and MSKCC cohorts, respectively. FS were RF-classified as either CS-like or NS-like and associations with patient characteristics, biological features, and clinical outcomes were determined.ResultsWe elucidated a 123 gene signature that robustly classified NS and CS in both RNA-seq (AUC = 0.85) and microarray (AUC = 0.92) validation test sets. The RF classified 213 patients who had formerly smoked as CS-like and 86 as NS-like from the TCGA cohort. CS-like and NS-like status in formerly smoking patients correlated poorly with patient characteristics but had substantially different biological features including tumor mutational burden, number of mutations, mutagenic signatures and immune cell populations. NS-like formerly smoking patients had 17.5 months and 18.6 months longer overall survival than CS-like patients from the TCGA and MSKCC cohorts, respectively.ConclusionsPatients who had formerly smoked with LUAD harbor heterogeneous tumor biology. These patients can be divided by smoking induced gene expression to inform prognosis and underlying biological characteristics for treatment selection.

UBA5 inhibition restricts lung adenocarcinoma via blocking macrophage M2 polarization and cisplatin resistance

UBA5, a ubiquitin-like activated enzyme involved in ufmylation and sumoylation, presents a viable target for pancreatic and breast cancer treatments, yet its role in lung adenocarcinoma (LUAD) remains underexplored. This study reveals UBA5's tumor-promoting effect in LUAD, as evidenced by its upregulation in patients and positive correlation with TNM stages. Elevated UBA5 levels predict poor outcomes for these patients. Pharmacological inhibition of UBA5 using DKM 2–93 significantly curtails the growth of A549, H1299, and cisplatin-resistant A549 (A549/DDP) LUAD cells in vitro. Additionally, UBA5 knockdown via shRNA lentivirus suppresses tumor growth both in vitro and in vivo. High UBA5 expression adversely alters the tumor immune microenvironment, affecting immunostimulators, MHC molecules, chemokines, receptors, and immune cell infiltration. Notably, UBA5 expression correlates positively with M2 macrophage infiltration, the predominant immune cells in LUAD. Co-culture experiments further demonstrate that UBA5 knockdown directly inhibits M2 macrophage polarization and lactate production in LUAD. Moreover, in vivo studies show reduced M2 macrophage infiltration following UBA5 knockdown. UBA5 expression is also associated with increased tumor heterogeneity, including tumor mutational burden, microsatellite instability, neoantigen presence, and homologous recombination deficiency. Experiments indicate that UBA5 overexpression promotes cisplatin resistance in vitro, whereas UBA5 inhibition enhances cisplatin sensitivity in both in vitro and in vivo settings. Overall, these findings suggest that targeting UBA5 inhibits LUAD by impeding cancer cell proliferation, M2 macrophage polarization, and cisplatin resistance.

Concomitant genomic features stratify prognosis to patients with advanced EGFR mutant lung cancer.

This study aimed to explore the clinical significance of genomics features including tumor mutation burden (TMB) and copy number alteration (CNA) for advanced EGFR mutant lung cancer. We retrospectively identified 1378 patients with advanced EGFR mutant lung cancer and next-generation sequencing tests from three cohorts. Multiple co-occurring genomics alternations occurred in a large proportion (97%) of patients with advanced EGFR mutant lung cancers. Both TMB and CNA were predictive biomarkers for these patients. A joint analysis of TMB and CNA found that patients with high TMB and high CNA showed worse responses to EGFR-TKIs and predicted worse outcomes. TMBhighCNAhigh, as a high-risk genomic feature, showed predictive ability in most of the subgroups based on clinical characteristics. These patients had larger numbers of metastatic sites, and higher rates of EGFR copy number amplification, TP53 mutations, and cell-cycle gene alterations, which showed more potential survival gain from combination treatment. Furthermore, a nomogram based on genomic features and clinical features was developed to distinguish prognosis. Genomic features could stratify prognosis and guide clinical treatment for patients with advanced EGFR mutant lung cancer.

Comprehensive pan‑cancer analysis of MTDH for human tumor prognosis and as an immunological biomarker including breast and kidney cancer.

Metadherin (MTDH), initially discovered in primary astrocytes of the human fetus through rapid subtraction hybridization and labeled as astrocyte elevated gene-1, represents a widely recognized oncogene present in multiple types of cancers. However, the role of MTDH in different types of cancer remains unclear. To address this, a comprehensive analysis of MTDH across various types of cancers was conducted by utilizing multiple databases such as The Cancer Genome Atlas. The present analysis discovered that MTDH exhibits differential expression in different types of cancer and is associated with important factors including tumor mutational burden and microsatellite instability. These findings highlighted the significance of MTDH in the tumor microenvironment and its involvement in the development of immune cells in specific cancers. Furthermore, the results of the present study indicated that the expression of MTDH is strongly correlated with clinical prognosis, mutations and immune cell infiltration. MTDH could serve as a potential indicator of patient prognosis and potentially play a role in modulating the immune system. Given its potential as a novel immunological checkpoint, MTDH may be a viable target for tumor immunotherapy.

Comprehensive genomic and immune profiling of non-V600 BRAF-mutated melanomas: Implications for therapeutic strategies beyond BRAF/MEK inhibitors.

e21544 Background: The development of targeted therapies ( BRAF plus MEK inhibitors) and immune checkpoint inhibitors (ICIs) has significantly improved the outcomes of patients with advanced melanoma, particularly those harboring BRAF V600 mutations (Class 1). However, melanomas with non-V600 BRAF mutations (classes 2 and 3) remain a subset where effective therapeutic options beyond ICIs are lacking. In this study, we investigated the distinct genomic and immune profile of non-V600 BRAF melanomas to further understand the mechanism of resistance to BRAF/MEK inhibitors and explore other pathways of potential therapeutic targets. Methods: A retrospective cohort of 175 melanoma samples was evaluated by comprehensive genomic and immune profiling, including tumor mutational burden (TMB) and the expression of 395 immune genes. Tumor specimens were classified as BRAFClass1, BRAFClass2, and BRAFClass3. TMB (high ≥10 Mut/Mb) was obtained by DNA sequencing. Gene expression signatures of tumor inflammation and cell proliferation were determined by targeted RNA-sequencing. We used chi-square and Wilcoxon rank-sum tests to determine the association of BRAF mutational status to genomic and immune correlates. Results: The median age of diagnosis was 69 years, with 57% of patients being male. A total of 152 (86.9%), 13 (7.4%), and 10 (5.7%) patients had BRAFClass1, BRAFClass2, and BRAFClass3 alterations, respectively. Genomic alterations in TP53, NF1, MET and ROS1 were more prevalent in BRAFClass2 and BRAFClass3 as opposed to BRAFClass1. BRAFClass2 and BRAFClass3 tumors combined had significantly higher tumor inflammation (median = 55.38; p &lt; 0.05) compared to BRAFClass1 (median = 41.79) as well as higher TMB (median = 54.8; p &lt; 0.001) compared to BRAFClass1 (median = 9.75). BRAFClass3 tumors showed significantly higher cellular proliferation (median = 47.5; p &lt; 0.05) compared to BRAFClass1 (median = 33.5) and BRAFClass2 (median = 31.8) tumors, as well as higher TMB (median = 71.5; p &lt; 0.01) compared to BRAFClass1 (median = 10.9). When looking at differentially expressed genes, CDK1 was upregulated in BRAFClass3 compared to BRAFClass1 and BRAFClass2, and IL12B was upregulated in BRAFClass2 compared to BRAFClass1. Conclusions: Our study showed that the non-Class 1 BRAF (V600E or K) mutated tumors have higher immune response suggesting a favorable response to ICIs. Additionally, we found that BRAFClass2 and BRAFClass3 tumors more commonly harbor alterations in NF1 and TP53, supporting the prior understanding that BRAF non-V600 tumors have a more aggressive clinical course. Comprehensive molecular analysis offers insight into the immunogenicity and distinct oncogenic drivers associated with unfavorable clinical outcomes, which could guide therapeutic development beyond BRAF and MEK inhibitors in BRAF non-V600 melanomas.

A pan-cancer analysis of SMARCA4 alterations and the unique clinicogenomic characteristics associated with SMARCA4 mutation types.

3073 Background: SWI/SNF complexes are regulators of cell lineage and alterations in these proteins occur in &gt;20% of solid tumors with SMARCA4being the most common. SMARCA4 mutations are associated with worse clinical outcomes in various cancer types but are also vulnerable to novel SMARCA2 inhibitors currently being evaluated in clinical trials. We examined SMARCA4 alterations across cancer types to understand their interplay with other genomic abnormalities and clinical consequences. Methods: We analyzed the Memorial Sloan Kettering (MSK) institutional cohort profiled by MSK-IMPACT tumor-normal sequencing. SMARCA4 mutations were annotated based on OncoKB and literature. FACETS resolved zygosity and genomic metrics of complexity including tumor mutational burden (TMB) and fraction genome altered (FGA). The pan-cancer landscape of somatic SMARCA4 mutations was mapped by both category (Class I vs II) and zygosity (mono- vs biallelic). DISCOVER was employed to evaluate mutational signatures and frequency-adjusted co-occurring and mutually exclusive mutations. We then extracted progression-free (PFS) and overall survival (OS) to systemic therapy using natural language processing. Results: We identified 3,385 tumors from 3,049 patients across cancer types with a total 3,965 oncogenic somatic SMARCA4 alterations. Copy number estimation was successful for 2,343 (69%). Non-small cell lung cancer (NSCLC) was the most abundant SMARCA4-altered histology ( n=493, 7.1% of all NSCLC), and the majority (87%) were biallelic. Other common histologies were colorectal ( n=163, 2.9%), bladder ( n=122, 5.5%), uterine ( n=88, 4.3%), pancreatic ( n=81, 3.2%) and esophagogastric ( n=92, 4%). SMARCA4biallelic mutated tumors had higher chromosomal instability compared to WT (FGA = 0.53 vs 0.47; p&lt;0.001), while monoallelic had lower (FGA = 0.3; p&lt;0.001). Monoallelic SMARCA4 altered tumors had a higher TMB (10.5) than both WT (4.4) and biallelic cancers (6.1; p&lt;0.001). Monoallelic and biallelic SMARCA4 cancers exhibited distinct mutational signatures in most tumor types. Patients with biallelic SMARCA4 mutations had worse outcomes with both immune checkpoint blockade (ICB) and platinum chemotherapy across in certain tumor types, whereas monoallelic uterine cancers were associated with improved outcomes with ICB. Conclusions: This pan-cancer clinicogenomic analysis demonstrates that monoallelic and biallelic SMARCA4-mutated tumors have distinguishable genomic features. In addition, biallelic SMARCA4 alterations correlated with poor outcomes across different cancer-types and classes of therapy, while improved outcomes to ICB in monoallelic SMARCA4alterations were potentially due to higher representation of MSI-H tumors in this subset. Our findings suggest SMARCA4 zygosity may be important biomarkers for ongoing clinical trials in SMARCA4-deficient tumors.

Pan-cancer analysis of CDKN2A alterations identifies a subset of gastric cancer with a cold tumor immune microenvironment

BackgroundAlthough CDKN2A alteration has been explored as a favorable factor for tumorigenesis in pan-cancers, the association between CDKN2A point mutation (MUT) and intragenic deletion (DEL) and response to immune checkpoint inhibitors (ICIs) is still disputed. This study aims to determine the associations of CDKN2A MUT and DEL with overall survival (OS) and response to immune checkpoint inhibitors treatment (ICIs) among pan-cancers and the clinical features of CDKN2A-altered gastric cancer.MethodsThis study included 45,000 tumor patients that underwent tumor sequencing across 33 cancer types from four cohorts, the MSK-MetTropism, MSK-IMPACT, OrigiMed2020 and TCGA cohorts. Clinical outcomes and genomic factors associated with response to ICIs, including tumor mutational burden, copy number alteration, neoantigen load, microsatellite instability, tumor immune microenvironment and immune-related gene signatures, were collected in pan-cancer. Clinicopathologic features and outcomes were assessed in gastric cancer. Patients were grouped based on the presence of CDKN2A wild type (WT), CDKN2A MUT, CDKN2A DEL and CDKN2A other alteration (ALT).ResultsOur research showed that CDKN2A-MUT patients had shorter survival times than CDKN2A-WT patients in the MSK MetTropism and TCGA cohorts, but longer OS in the MSK-IMPACT cohort with ICIs treatment, particularly in patients having metastatic disease. Similar results were observed among pan-cancer patients with CDKN2A DEL and other ALT. Notably, CDKN2A ALT frequency was positively related to tumor-specific objective response rates to ICIs in MSK MetTropism and OrigiMed 2020. Additionally, individuals with esophageal carcinoma or stomach adenocarcinoma who had CDKN2A MUT had poorer OS than patients from the MSK-IMPACT group, but not those with adenocarcinoma. We also found reduced levels of activated NK cells, T cells CD8 and M2 macrophages in tumor tissue from CDKN2A-MUT or DEL pan-cancer patients compared to CDKN2A-WT patients in TCGA cohort. Gastric cancer scRNA-seq data also showed that CDKN2A-ALT cancer contained less CD8 T cells but more exhausted T cells than CDKN2A-WT cancer. A crucial finding of the pathway analysis was the inhibition of three immune-related pathways in the CDKN2A ALT gastric cancer patients, including the interferon alpha response, inflammatory response, and interferon gamma response.ConclusionsThis study illustrates the CDKN2A MUT and DEL were associated with a poor outcome across cancers. CDKN2A ALT, on the other hand, have the potential to be used as a biomarker for choosing patients for ICI treatment, notably in esophageal carcinoma and stomach adenocarcinoma.

Prognostic analysis of hepatocellular carcinoma based on cuproptosis -associated lncRNAs

ObjectivesCuproptosis represents an innovative type of cell death, distinct from apoptosis, driven by copper dependency, yet the involvement of copper apoptosis-associated long non-coding RNAs (CRLncRNAs) in hepatocellular carcinoma (HCC) remains unclear. This study is dedicated to unveiling the role and significance of these copper apoptosis-related lncRNAs within the context of HCC, focusing on their impact on both the development of the disease and its prognosis.MethodsWe conducted an analysis of gene transcriptomic and clinical data for HCC cases by sourcing information from The Cancer Genome Atlas database. By incorporating cuproptosis-related genes, we established prognostic features associated with cuproptosis-related lncRNAs. Furthermore, we elucidated the mechanism of cuproptosis-related lncRNAs in the prognosis and treatment of HCC through comprehensive approaches, including Lasso and Cox regression analyses, survival analyses of samples, as well as examinations of tumor mutation burden and immune function.ResultsWe developed a prognostic model featuring six cuproptosis-related lncRNAs: AC026412.3, AC125437.1, AL353572.4, MKLN1-AS, TMCC1-AS1, and SLC6A1-AS1. This model demonstrated exceptional prognostic accuracy in both training and validation cohorts for patients with tumors, showing significantly longer survival times for those categorized in the low-risk group compared to the high-risk group. Additionally, our analyses, including tumor mutation burden, immune function, Gene Ontology, Kyoto Encyclopedia of Genes and Genomes pathway enrichment, and drug sensitivity, further elucidated the potential mechanisms through which cuproptosis-associated lncRNAs may influence disease outcome.ConclusionsThe model developed using cuproptosis-related long non-coding RNAs (lncRNAs) demonstrates promising predictive capabilities for both the prognosis and immunotherapy outcomes of tumor patients. This could play a crucial role in patient management and the optimization of immunotherapeutic strategies, offering valuable insights for future research.

Survival in metastatic microsatellite-stable colorectal cancer correlated with tumor mutation burden and mutations identified by next-generation sequencing.

Next-generation sequencing (NGS) identifies mutations and molecular abnormalities within tumors, including tumor mutation burden (TMB). If a solid tumor has high TMB, immune checkpoint inhibitors (ICIs) are approved as an option for treatment. Studies have been inconclusive regarding how effective ICI are in treating patients with colorectal cancer (CRC), and it is unclear if high TMB is a good prognostic marker for CRC. We collected data from NGS of CRC and correlated survival to both TMB and mutations of interest, as well as investigated the efficacy of ICI. This was a retrospective cohort analysis at a single institution, collecting NGS data from January 2018 to December 2020 in patients with CRC who were microsatellite-stable (MSS), n=161. Demographics, clinical data, and results from NGS were collected, and a survival analysis looking at TMB and selected mutations of interest was performed. Patients who were treated with ICI were assessed in a descriptive subset analysis. Patients with CRC who were MSS and had high TMB trended towards worse survival [hazard ratio (HR) =1.38] though the result was not significant (P=0.28). Survival was significantly worse in patients with a KRAS mutation (HR =1.71, P=0.04) and/or a CDKN2A mutation (HR =4.45, P<0.001). In this study population, 12 patients with high TMB had treatment with ICI, with nine of these patients having shorter progression-free survival (PFS) between 0.7 and 4.1 months, and three patients having longer PFS of 26.3, 24.7, and 13.2 months. High TMB in MSS CRC did not show statistical difference in outcome. Mutations in KRAS and/or CDKN2A correlated with worse prognosis. Some patients with MSS CRC and high TMB responded to ICI, though there is a need to identify a better biomarker to predict which patients will have a good response to ICI therapy.

Abstract 4951: Integrative multi-omic machine learning model predicts neoadjuvant immunotherapy response using molecular data and deep learning-derived features from digital pathology

Abstract Background: Over the past decade, there has been an exponential accumulation of digital patient data through molecular profiling and digitization of pathology slides. However, integrating this to advance diagnostics and cancer patient care remains a significant challenge. By integrating information from diverse modalities to capture aspects of the tumor microenvironment (TME) and immune landscape, we aim to improve patient stratification for immunotherapy. Methods: We collected data from baseline biopsies from patients with palpable melanoma stage III (n=195) who received neoadjuvant immunotherapy followed by surgical resection of largest lymph node metastasis. The different modalities included clinical information, whole-genome sequencing (WGS), RNAseq, laboratory tests of blood, haematoxylin and eosin (H&amp;E) images, and PDL1 immunohistochemistry (IHC) images. Cell detection and classification models were trained to identify tumor-infiltrating lymphocytes and tumor cells on H&amp;E whole-slide images (WSI), deriving spatial features to quantitatively describe the tumor-immune interaction. Various biomarkers, including tumor mutational burden, BRAF status, expression of inflammatory gene signatures, systemic neutrophil-to-lymphocyte ratio, and tumor-lymphocyte spatial measures, were collectively integrated using machine learning models (random forest, elastic-net regularized logistic regression, CatBoost, support vector machine (SVM), and an ensemble classifier combined from the other models) within a 5x5-fold nested cross-validation scheme to predict pathologic response. Results: Pathologic response to immunotherapy is associated with multi-omics biomarkers which were integrated using machine learning. The ensemble approach showed that when all modalities were combined a higher AUC (AUC=0.69, 0.61-0.77) could be achieved than for the clinical modality alone. RNA plus clinical emerged as the most predictive standalone modality with an AUC of 0.62 (0.50-0.75). The overall highest AUC was achieved using a SVM (AUC=0.72, 0.66-0.82). Conclusions: Immunotherapy efficacy depends on diverse tumor microenvironment components, which were comprehensively assessed through data integration. A multi-omic predictive model holds promise for refining personalized and effective immunotherapy strategies. Citation Format: Lindsay V. Leek, Vanessa Botha, Vipul A. Baxi, Jonas Teuwen, Hugo M. Horlings, Scott D. Chasalow, Joris van de Haar, Lodewyk F. Wessels, Auditi DebRoy, Emile E. Voest. Integrative multi-omic machine learning model predicts neoadjuvant immunotherapy response using molecular data and deep learning-derived features from digital pathology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4951.

Abstract 2543: Evaluating biomarkers of immunotherapy response in a real-world metastatic NSCLC cohort

Abstract Background: Non-small cell lung cancer (NSCLC) is the leading cause of global cancer deaths, since a large proportion of cases are initially diagnosed at advanced stages. Therapeutic developments have shifted first-line (1L) therapy in the metastatic setting to immunotherapy (IO)-based regimens, albeit with heterogeneity in response. Patients harboring KEAP1 and STK11 mutations have been reported to exhibit both poorer outcomes and IO treatment resistance. Recent work suggests gene expression signatures which characterize mutant phenotypes may better identify those with compromised IO response. Leveraging a molecular real-world dataset of metastatic NSCLC patients treated with 1L IO, we aimed to evaluate gene expression signatures representing KEAP1 and STK11-mutated phenotypes and their impact on real-world outcomes. Methods: Metastatic NSCLC patients who received 1L anti-PD(L)1 as a monotherapy or in combination were identified from the multimodal real-world database of Tempus Labs, Inc. All patients in the analytical cohort had tissue biopsies sequenced on Tempus targeted DNA and whole transcriptome RNA assays within 90 days prior to the start of the 1L IO regimen. Patients were negative for actionable mutations with record duration of at least 30 days from 1L initiation. KEAP1 and STK11 mutants were identified with likely pathogenic and/or pathogenic mutations as annotated by Tempus bioinformatic pipelines. Gene expression signatures related to these mutations were obtained from recently published literature. Evaluation of mutations or signatures as genomic markers for IO non-response were conducted using real-world progression-free survival (rwPFS) and overall survival (rwOS). Results: A total of 332 patients were included in the study (45% IO monotherapy, 55% IO combination therapy). Previously validated biomarkers, including tumor mutation burden, T-cell inflamed GEP signature, and PD(L)1 expression, were associated with prolonged rwPFS. KEAP1 mutations were observed in 9% of analytical cohort with a higher proportion in non-squamous (11%) compared with squamous (4%) samples. For STK11, 10% were identified as mutants with a similar histology distribution (12% vs. 3%). Neither KEAP1 mutations nor related gene signatures evaluated in this study were significantly associated with real-world survival outcomes. Patients harboring STK11 mutations trended towards poor outcomes; furthermore, those with higher STK11 inactivation signature score were associated with shorter rwPFS and rwOS. Conclusions: STK11 inactivation phenotype described by a gene expression signature was able to distinguish patients with significantly poorer survival outcomes. This finding would need validation in independent cohorts. This real-world cohort study adds supporting evidence for the utility of gene expression signatures encompassing mutant phenotypes for patient stratification. Citation Format: Katherina C. Chua, Assieh Saadatpour, Ariel Chen, Aisha Hasan, Bevan Emma Huang, Stewart Bates. Evaluating biomarkers of immunotherapy response in a real-world metastatic NSCLC cohort [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2543.

Abstract 4311: The interplay of SYNCRIP deficiency, APOBEC activity, and extrachromosomal DNA in castration-resistant prostate cancer drug resistance

Abstract Background: Prostate cancer, particularly in its castration-resistant form (CRPC), frequently develops resistance to androgen receptor (AR)-targeted therapies. This resistance involves complex factors, including tumor mutational burden and heterogeneity. APOBEC proteins, associated with mutagenesis and DNA damage, are implicated in this process. Our study centers on SYNCRIP, a key regulator of APOBEC activity, and investigates its influence on extrachromosomal DNA (ecDNA) in CRPC. Objective: This research aims to explore how SYNCRIP deficiency leads to increased APOBEC activity, and to examine its possible implications for ecDNA formation and its contribution to drug resistance and tumor heterogeneity in CRPC. Methods: Genomic and transcriptomic analyses, such as whole exome sequencing and single-cell RNA-seq, were performed on CRPC cell lines and patient-derived samples, with an emphasis on SYNCRIP-deficient models. The study focused on understanding the relationship between elevated APOBEC activity due to SYNCRIP deficiency and potential ecDNA changes, evaluating their effect on AR-targeted drug sensitivity and overall tumor behavior. Results: SYNCRIP was identified as a critical factor in resistance to AR therapy in both in vitro and in vivo models. Its deficiency resulted in increased therapy resistance and amplified APOBEC-mediated mutagenesis. Functional CRISPR screening identified eight key drivers of resistance in SYNCRIP-deficient tumors. Single-cell RNA sequencing revealed APOBEC-driven mutagenesis as a key factor in promoting intratumoral heterogeneity and resistance. Notably, in SYNCRIP-deficient CRPC samples, increased APOBEC activity and significant DNA damage were observed, prompting an investigation into a possible correlation between resistance and ecDNA presence. Conclusion: The study uncovers a significant link between SYNCRIP deficiency and increased APOBEC activity in CRPC. While it also hints at a potential connection with ecDNA alterations, further research is necessary to establish this relationship firmly. This interplay contributes to the complexity of drug resistance in CRPC, highlighting the need for novel therapeutic targets. The dynamics among SYNCRIP, APOBEC activity, and ecDNA offer valuable insights into the molecular underpinnings of therapy resistance in prostate cancer, suggesting new directions for treatment strategies. Citation Format: Xiaoling Li, Su Deng, Choushi Wang, Yunguan Wang, Ping Mu. The interplay of SYNCRIP deficiency, APOBEC activity, and extrachromosomal DNA in castration-resistant prostate cancer drug resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4311.

A community challenge to predict clinical outcomes after immune checkpoint blockade in non-small cell lung cancer

BackgroundPredictive biomarkers of immune checkpoint inhibitor (ICI) efficacy are currently lacking for non-small cell lung cancer (NSCLC). Here, we describe the results from the Anti–PD-1 Response Prediction DREAM Challenge, a crowdsourced initiative that enabled the assessment of predictive models by using data from two randomized controlled clinical trials (RCTs) of ICIs in first-line metastatic NSCLC.MethodsParticipants developed and trained models using public resources. These were evaluated with data from the CheckMate 026 trial (NCT02041533), according to the model-to-data paradigm to maintain patient confidentiality. The generalizability of the models with the best predictive performance was assessed using data from the CheckMate 227 trial (NCT02477826). Both trials were phase III RCTs with a chemotherapy control arm, which supported the differentiation between predictive and prognostic models. Isolated model containers were evaluated using a bespoke strategy that considered the challenges of handling transcriptome data from clinical trials.ResultsA total of 59 teams participated, with 417 models submitted. Multiple predictive models, as opposed to a prognostic model, were generated for predicting overall survival, progression-free survival, and progressive disease status with ICIs. Variables within the models submitted by participants included tumor mutational burden (TMB), programmed death ligand 1 (PD-L1) expression, and gene-expression–based signatures. The best-performing models showed improved predictive power over reference variables, including TMB or PD-L1.ConclusionsThis DREAM Challenge is the first successful attempt to use protected phase III clinical data for a crowdsourced effort towards generating predictive models for ICI clinical outcomes and could serve as a blueprint for similar efforts in other tumor types and disease states, setting a benchmark for future studies aiming to identify biomarkers predictive of ICI efficacy.Trial registration: CheckMate 026; NCT02041533, registered January 22, 2014.CheckMate 227; NCT02477826, registered June 23, 2015.

Cancer testis antigen burden (CTAB): a novel biomarker of tumor-associated antigens in lung cancer

BackgroundCancer-testis antigens (CTAs) are tumor antigens that are normally expressed in the testes but are aberrantly expressed in several cancers. CTA overexpression drives the metastasis and progression of lung cancer, and is associated with poor prognosis. To improve lung cancer diagnosis, prognostic prediction, and drug discovery, robust CTA identification and quantitation is needed. In this study, we examined and quantified the co-expression of CTAs in lung cancer to derive cancer testis antigen burden (CTAB), a novel biomarker of immunotherapy response.MethodsFormalin fixed paraffin embedded (FFPE) tumor samples in discovery cohort (n = 5250) and immunotherapy and combination therapy treated non-small cell lung cancer (NSCLC) retrospective (n = 250) cohorts were tested by comprehensive genomic and immune profiling (CGIP), including tumor mutational burden (TMB) and the mRNA expression of 17 CTAs. PD-L1 expression was evaluated by IHC. CTA expression was summed to derive the CTAB score. The median CTAB score for the discovery cohort of 170 was applied to the retrospective cohort as cutoff for CTAB “high” and “low”. Biomarker and gene expression correlation was measured by Spearman correlation. Kaplan–Meier survival analyses were used to detect overall survival (OS) differences, and objective response rate (ORR) based on RECIST criteria was compared using Fisher’s exact test.ResultsThe CTAs were highly co-expressed (p < 0.05) in the discovery cohort. There was no correlation between CTAB and PD-L1 expression (R = 0.011, p = 0.45) but some correlation with TMB (R = 0.11, p = 9.2 × 10–14). Kaplan–Meier survival analysis of the immunotherapy-treated NSCLC cohort revealed better OS for the pembrolizumab monotherapy treated patients with high CTAB (p = 0.027). The combination group demonstrated improved OS compared to pembrolizumab monotherapy group (p = 0.04). The pembrolizumab monotherapy patients with high CTAB had a greater ORR than the combination therapy group (p = 0.02).ConclusionsCTA co-expression can be reliably measured using CGIP in solid tumors. As a biomarker, CTAB appears to be independent from PD-L1 expression, suggesting that CTAB represents aspects of tumor immunogenicity not measured by current standard of care testing. Improved OS and ORR for high CTAB NSCLC patients treated with pembrolizumab monotherapy suggests a unique underlying aspect of immune response to these tumor antigens that needs further investigation.

POLE variant phenotypic classification to enable identification of patients with colorectal cancer (CRC) who may benefit from immune checkpoint inhibition (ICI).

194 Background: Pathogenic POLE (p POLE) variants interfere with the exonuclease activity of POLε during DNA replication, resulting in somatic ultramutation and conferring susceptibility to ICI. Historical classification of p POLE relies on localization to the exonuclease domain (ED). We present the landscape of p POLE variants in CRC according to a proposed phenotypic classification model. Methods: Comprehensive genomic profiling was performed on tissue (TBx) and liquid (LBx) biopsies during the course of clinical care. A POLE-specific variant classification model including tumor mutational burden (TMB), mutational signatures, and germline frequency was used to differentiate p POLE from benign POLE (b POLE) variants. Results: POLE variant status was evaluated in 455,965 TBx and 41,804 LBx pan-tumor. POLE alterations, including p POLE and variants of unknown significance, were detected in 3.9% of samples with 7,404 discrete alleles observed. After applying the classification model, 30 variants were classified as p POLE, representing 3.6% (751/20,601) of observed POLE variants. All p POLE were localized to the ED and a majority were missense substitutions (29/30). There were an additional 620 ED variants categorized as b POLE. CRC had the second highest prevalence of p POLE variants (0.5%, 302/60,547). p POLE variants were associated with ultramutation in CRC where the median tissue TMB was 183.0 mut/Mb, 1-2 orders of magnitude higher than for CRC harboring b POLE variants (7.8 mut/Mb) or without POLE variants (3.5 mut/Mb). 32.2% of p POLE CRC also had evidence of high microsatellite instability (MSI-H). Combined p POLE + MSI-H was associated with markedly elevated TMB compared to either signature alone (344.4 versus 53.1 and 151.3 mut/Mb for b POLE + MSI-H and p POLE + MS-Stable, respectively). A small minority of CRC tumors with p POLE variants (7/298, 2.3%) had low TMB (&lt;10 mut/Mb). However, this was associated with a low p POLE variant allele frequency (VAF; median 2.8% versus 24.0% for tumors w/ TMB ≥10, p &lt; .0001) pointing to low tumor purity and likely an underestimation of TMB. While targetable alterations (e.g., homologous recombination repair [HR] gene variants) were enriched in p POLE tumors, these tumors had a low rate of HR deficiency (HRD) as determined by a computational algorithm (HRDsig), suggesting these are likely incidental monoallelic variants. Conclusions: A POLE phenotype-specific classification model identified p POLE variants associated with ultramutation in CRC. Co-occurrence of p POLE and MSI-H in CRC is common and has a synergistic effect leading to even higher TMB. We hypothesize that p POLE detection predicts ultramutation and response to ICI even when TMB is under-detected due to low tumor purity. Caution is advised for therapy selection based on co-alterations in the setting of p POLE due to an abundance of incidental variants.