Predict Immunotherapy Response Research Articles

Unveiling the role of PANoptosis-related genes in breast cancer: an integrated study by multi-omics analysis and machine learning algorithms.

The heterogeneity of breast cancer (BC) necessitates the identification of novel subtypes and prognostic models to enhance patient stratification and treatment strategies. This study aims to identify novel BC subtypes based on PANoptosis-related genes (PRGs) and construct a robust prognostic model to guide individualized treatment strategies. The transcriptome data along with clinical data of BC patients were sourced from the TCGA and GEO databases. Consensus clustering was performed on 12 PRGs to ascertain potential BC subtypes, and variances in survival, infiltration of immune cells, and functional pathways among them were examined. A prognostic model was generated through 101 combinations of machine learning algorithms and validated across multiple cohorts. The response of patients towards immunotherapy were analyzed using multiple frameworks. Consensus clustering of 12 PRGs identified two distinct BC subtypes, with subtype B exhibiting significantly lower overall survival (OS) rates compared to subtype A. Immune cell infiltration analysis revealed higher immune activity in subtype A. Functional pathway analysis revealed that subtype A exhibited a significant enrichment in immune-related pathways, while subtype B was associated with cell cycle and metabolic processes. An integrated machine learning framework integrating CoxBoost and Random Survival Forest (RSF) algorithms was developed, demonstrating high predictive performance across multiple cohorts. A nomogram combining age and risk score was constructed, showing excellent predictive performance. Immune landscape analysis revealed that the high-risk group exhibited a suppressive tumor immune microenvironment (TIME). Immunotherapy response prediction suggested that low-risk patients were more likely to benefit from PD-1 and CTLA-4 inhibitors. Our study provides a comprehensive framework for BC subtype classification and prognostic prediction, offering valuable insights for personalized treatment strategies.

Investigating the Impact of B Cell-Related Genes on Colorectal Cancer Immunosuppressive Environment and Immunotherapy Evasion.

We aimed to elucidate the prognostic and immunological roles of B cell-related genes in colorectal cancer (CRC). This study comprehensively integrated data from single-cell RNA-sequencing, TCGA, GEO, IMvigor210, GDSC, CancerSEA, HPA, and TISIDB databases to explore prognostic implications and immunological significance of B cell-related gene signature in CRC. We identified seven prognostically significant B cell-related genes for constructing a risk score. Clinical relevance analysis indicated that this risk score served as an independent prognostic factor, with the model accurately predicting patient outcomes. GSEA results implicated the risk score in immune function, cell cycle, and DNA replication. Immune infiltration analysis revealing lower levels of B cells, CD4+ cells, and CD8+ cells in the high-risk group, correlating with decreased immune activity and function. IMvigor210 and TIDE analysis indicated poorer prognosis among high-risk group patients receiving immune therapy. Additionally, the high-risk group exhibited lower sensitivity to immune therapy. Further analysis of drug sensitivity suggested higher resistance to common chemotherapy drugs among high-risk groups. Finally, we identified HSPA1A as the gene with the strongest association with immune and inflammatory responses. Validation of HSPA1A protein expression and prognosis demonstrated elevated expression in CRC compared to normal colorectal tissue, further reinforcing its association with poorer prognosis and higher tumor stage. The risk score exhibited substantial variations in clinical characteristics, functional mechanism, TMB, drug sensitivity, immune cell infiltration, and immune subtype. Our findings may aid in clinical decision-making by shedding light on novel and promising biomarkers for CRC prognosis and immunotherapy response prediction.

Machine‐learning‐based integration of tumor microenvironment features predicting immunotherapy response

ABSTRACTImmunotherapy has revolutionized cancer treatment in recent years, yet non‐responsiveness of immunotherapy remains a challenge for cancer treatment. Therefore, the prediction method for potential clinical benefits of patients from immunotherapy is urgently needed. This study aims to develop an effective clinical practice assistance tool to evaluate the potential clinical benefits and therapy responsiveness of patients undergoing immunotherapy. We developed an immunotherapy resistance score (IRS), which performed well compared with conventional immunotherapy response indicators across different immunotherapy cohorts. Tumor microenvironment (TME) analysis showed that both immune and nonimmune features collectively impact immunotherapy responsiveness. Thus, IRS was constructed based on the TME features using machine learning approaches. The clinical application potential of IRS has been demonstrated in our in‐house Harbin Medical University (HMU) cohort and an external validation cohort. Furthermore, we analyzed the correlation between IRS and pathways related to cancer therapy targets to explore the application potential of IRS in comprehensive cancer therapy. In conclusion, IRS is a robust tool for predicting patient immunotherapy prognosis, which has great potential to promote precise clinical therapy.

Prediction of immunotherapy response in nasopharyngeal carcinoma: a comparative study using MRI-based radiomics signature and programmed cell death ligand 1 expression score.

To compare an MRI-based radiomics signature with the programmed cell death ligand 1 (PD-L1) expression score for predicting immunotherapy response in nasopharyngeal carcinoma (NPC). Consecutive patients with NPC who received immunotherapy between January 2019 and June 2022 were divided into training (n = 111) and validation (n = 66) sets. Tumor radiomics features were extracted from pretreatment MR images. PD-L1 combined positive score (CPS) was calculated using immunohistochemistry. The least absolute shrinkage and selection operator (LASSO) algorithm was used for feature selection and radiomics signature construction. Receiver operating characteristic (ROC) curve analysis was performed to assess prediction performance. A total of eleven radiomics features with the greatest discrimination capability were identified by the LASSO algorithm to construct the radiomics signature. In predicting patients with objective response to immunotherapy, radiomics score (Rd-score) yielded a significantly higher area under the ROC curve than that of CPS in both the training (0.790 vs. 0.645, p = 0.025) and the validation (0.735 vs. 0.608, p = 0.038) sets. Multivariate analysis identified the Rd-score as an independent influencing factor in predicting immunotherapy response (odds ratio = 19.963, p < 0.001). Kaplan-Meier analysis indicated that patients with Rd-score ≥ 0.5 showed longer progression-free survival than patients with Rd-score < 0.5 (log-rank p < 0.01). An MRI-based radiomics signature demonstrated greater efficacy than the PD-L1 expression score in predicting immunotherapy response in patients with NPC. Question How does an MRI-based radiomics signature compare with the programmed cell death ligand 1 expression score for predicting immunotherapy response in nasopharyngeal carcinoma? Findings The MRI-based radiomics signature demonstrated superior predictive value compared with programmed cell death ligand 1 expression score in identifying immunotherapy responders. Clinical relevance MRI-based radiomics are a promising novel noninvasive tool for predicting immunotherapy outcomes in nasopharyngeal carcinoma.

Identification of a metabolic-immune signature associated with prognosis in colon cancer and exploration of potential predictive efficacy of immunotherapy response

The role of metabolic reprogramming of the tumor immune microenvironment in cancer development and immune escape has increasingly attracted attention. However, the predictive value of differences in metabolism-immune microenvironment on the prognosis of colon cancer (CC) and the response to immunotherapy have not been elucidated. The aim of this study was to investigate changes in metabolism and immune profile of CC and to identify a reliable signature for predicting prognosis and therapeutic response. The metabolism and immune-related differential genes in CC were screened out by differential gene expression analysis. A metabolism and immune related prognostic signature was established by the least absolute shrinkage and selection operator (LASSO) Cox algorithm. The training cohort with 417 patients from The Cancer Genome Atlas (TCGA) database and the validation cohort of 232 patients from GSE17538 were used to confirm the robustness of the prognostic signature. Immunohistochemical staining scores were used to assess gene expression levels in our clinical samples. Gene ontology (GO) analysis, gene set enrichment analysis (GSEA), single nucleotide variation (SNV) analysis, immune infiltration and immune factors analysis were used to explore the characteristics of patients with different subtypes. Multiple cancer immunotherapy datasets were used to assess the response of patients with different subtypes to immune checkpoint inhibitors. We established the Metabolism and Immune-Related Prognostic Score (MIRPS) based on six genes (CD36, PCOLCE2, SCG2, CALB2, STC2, CLDN23) to predict the prognosis of CC patients. We found a correlation between MIRPS and the malignant phenotype, microsatellite subtype, mutation load, and immune escape in CC. Tumors with high MIRPS presented a higher tumor mutation load and a more prominent immunosuppressive microenvironment. This subset of patients may potentially respond well to immune checkpoint inhibitor therapy. MIRPS may be used as a novel prognostic tool for CC and have potential value for immunotherapy response prediction.

Lipid Metabolism Related Gene ACSL3 as a Biomarker for Predicting Immunotherapy Outcomes in Lung Adenocarcinoma.

Investigate the role of lipid metabolism in the tumor immune microenvironment (TIME) of lung adenocarcinoma (LUAD) and identify vital lipid metabolism-related genes (LMRGs) that contribute to immunotherapy outcomes. 1130 LUAD patients were acquired utilizing public databases. Multiple algorithms were used to analyze the contribution of lipid metabolism in TIME. Importantly, cell lines, clinical samples (52 patients in surgery cohort and 36 in immunotherapy cohort), animal models, RNA-seq, experiments in protein and mRNA levels were conducted for identifying and validating key biomarker in LUAD immunotherapy. A prognostic signature comprising 33 LMRGs was developed and validated as an effective predictor of prognosis and TIME, with a C-index of 0.766 (95% CI: 0.729-0.804). Additionally, we identified Acyl-CoA Synthetase Long Chain Family Member 3 (ACSL3) as a potential biomarker for immunotherapy prognosis. The expression of ACSL3 was verified in 88 clinical tissues from LUAD patients, which indicated that elevated ACSL3 expression was correlated with worse progression-free survival (PFS) (p<0.001) and overall survival (OS) (p=0.008). Subsequent experiments revealed that knockdown of ACSL3 in vivo enhanced the efficacy of immunotherapy, potentially through increasing interferon α secretion, as indicated by Bulk RNA-seq and ELISA analysis, thereby promoting the infiltration of anti-tumor immune cells. The study established a model that accurately predicts immunotherapy response, prognosis, and TIME dynamics in LUAD patients. Notably, the pivotal role of ACSL3 in driving tumor progression and immune evasion was uncovered, offering novel insights into the optimization of immunotherapy strategies for LUAD.

Predictive value of NUF2 for prognosis and immunotherapy responses in pan-cancer.

To investigate the association of NUF2 expression with tumor prognosis and its regulatory role in tumor microenvironment. We analyzed NUF2 expression, its prognostic value, and is immune-related functions across different cancer types using datasets from the Human Protein Atlas (HPA), TCGA, GTEx, CCLE, and TIMER. RT-qPCR, Western blotting, and immunohistochemistry were used to detect NUF2 expression in liver cancer cell lines and tissue and blood samples from patients with liver cancer. GO, KEGG, and GSEA analyses were conducted to explore the molecular mechanisms of NUF2 and its related genes, and a competitive endogenous RNA (ceRNA) network for NUF2 in liver cancer was constructed. NUF2 expression was upregulated in the tumor tissues of 27 cancers and was associated with clinical stages in several cancers. High NUF2 expressions were correlated with poor overall survival, disease-specific survival, progression-free survival, and disease-free survival of cancer patients. NUF2 expression levels were positively correlated with tumor mutational burden, microsatellite instability, infiltrating immune cells, immune cell marker genes and immune checkpoint genes in different cancers. RT-qPCR, Western blotting, and immunohistochemistry confirmed that NUF2 expression was upregulated in liver cancer cell lines and tumor tissues and blood samples of liver cancer patients, and was decreased significantly after operation. GO, KEGG and GSEA analyses indicated that NUF2 was involved in chromosome segregation and cell cycle and was associated with glycine, serine and threonine metabolism. NUF2 expression is upregulated in 27 cancers and is associated with clinical stage and poor prognosis in some malignancies. NUF2 expression is closely correlated with immune cell infiltration in different cancers, suggesting its potential value for predicting immunotherapy response in these cancers.

Disulfidptosis-related gene signatures as prognostic biomarkers and predictors of immunotherapy response in HNSCC

Disulfidptosis-related gene signatures as prognostic biomarkers and predictors of immunotherapy response in HNSCC

Aptamer-Proximity Ligation Coupled with Rolling Circle Amplification Strategy for an Ultrasensitive Analysis of Tumor-Derived Extracellular Vesicles PD-L1.

Tumor-derived extracellular vesicles (T-EVs) PD-L1 are an important biomarker for predicting immunotherapy response and can help us understand the mechanism of resistance to immunotherapy. However, this is due to the interference from a large proportion of nontumor-derived EVs. It is still challenging to accurately analyze T-EVs PD-L1 in complex human fluids. Herein, a simple and ultrasensitive method based on the dual-aptamer-proximity ligation assay (PLA)-guided rolling circle amplification (RCA) for the analysis of T-EVs PD-L1 was developed. First, dual aptamers with strong binding affinity were utilized for the recognition of EpCAM and PD-L1 on EVs, and then the aptamer-based PLA occurred. With the aid of the high signal amplification ability of RCA guided by the dual-aptamer-based PLA and efficient magnetic separation, the biosensor could realize highly sensitive quantification of EpCAM and PD-L1 dual-positive EVs with a low detection limit of 7.5 particles/μL. In addition, this method based on the aptamer-PLA-guided RCA was used to discriminate cancer patients from healthy donors with 100% accuracy without additional purification. Overall, this strategy might provide a practical tool for the analysis of multiple proteins on EVs, exhibiting great potential in early cancer diagnosis and treatment.

OSBPL3 modulates the immunosuppressive microenvironment and predicts therapeutic outcomes in pancreatic cancer

BackgroundPancreatic cancer is characterized by a complex tumor microenvironment that hinders effective immunotherapy. Identifying key factors that regulate the immunosuppressive landscape is crucial for improving treatment strategies.MethodsWe constructed a prognostic and risk assessment model for pancreatic cancer using 101 machine learning algorithms, identifying OSBPL3 as a key gene associated with disease progression and prognosis. We integrated multi-dataset analyses, single-cell transcriptomic data, and functional experiments to explore the role of OSBPL3 in pancreatic cancer.ResultsOur risk prediction model, developed using machine learning algorithms, demonstrated high predictive accuracy across multiple datasets. Notably, the “rf” algorithm model showed an AUC of 1 in the training set and AUCs of 0.887 and 0.977 in two validation datasets. Ridge regression analysis identified OSBPL3 as a core prognostic feature gene. High OSBPL3 expression in pancreatic cancer samples was associated with immunosuppressive characteristics, including reduced CD8 + T cell infiltration and increased immunosuppressive cell populations such as Treg cells and M2 macrophages. Transcriptomic sequencing following OSBPL3 knockdown revealed enrichment of immune-related pathways, suggesting OSBPL3’s influence on the immune microenvironment. Single-cell analyses further confirmed OSBPL3’s role in shaping the immunosuppressive landscape by modulating Treg cells and M2 macrophages. Additionally, OSBPL3 expression was linked to resistance to immunotherapy, with high OSBPL3 expression associated with lower Immunophenoscore (IPS) scores, indicating poor responsiveness to immunotherapy.ConclusionsOur study reveals OSBPL3 as a critical regulator of the immunosuppressive microenvironment in pancreatic cancer and a potential therapeutic target. Targeting OSBPL3 may enhance the efficacy of immunotherapy and improve patient outcomes. Further research is warranted to explore OSBPL3 as a biomarker for predicting immunotherapy response and as a potential therapeutic target in combination with anti-PD1 therapy.

A vision-language foundation model for precision oncology.

Clinical decision-making is driven by multimodal data, including clinical notes and pathological characteristics. Artificial intelligence approaches that can effectively integrate multimodal data hold significant promise in advancing clinical care1,2. However, the scarcity of well-annotated multimodal datasets in clinical settings has hindered the development of useful models. In this study, we developed the Multimodal transformer with Unified maSKed modeling (MUSK), a vision-language foundation model designed to leverage large-scale, unlabelled, unpaired image and text data. MUSK was pretrained on 50 million pathology images from 11,577 patients and one billion pathology-related text tokens using unified masked modelling. It was further pretrained on one million pathology image-text pairs to efficiently align the vision and language features. With minimal or no further training, MUSK was tested in a wide range of applications and demonstrated superior performance across 23 patch-level and slide-level benchmarks, including image-to-text and text-to-image retrieval, visual question answering, image classification and molecular biomarker prediction. Furthermore, MUSK showed strong performance in outcome prediction, including melanoma relapse prediction, pan-cancer prognosis prediction and immunotherapy response prediction in lung and gastro-oesophageal cancers. MUSK effectively combined complementary information from pathology images and clinical reports and could potentially improve diagnosis and precision in cancer therapy.

Radiomics signature for dynamic monitoring of tumor inflamed microenvironment and immunotherapy response prediction

BackgroundThe efficacy of immune checkpoint inhibitors (ICIs) depends on the tumor immune microenvironment (TIME), with a preference for a T cell-inflamed TIME. However, challenges in tissue-based assessments via biopsies have...

PD-L1 as a Biomarker in Gastric Cancer Immunotherapy.

Combining chemotherapy with immune checkpoint inhibitors (ICIs) that target the programmed death-1 (PD-1) protein has been shown to be a clinically effective first-line treatment for human epidermal growth factor receptor 2 (HER2)-negative and -positive advanced or metastatic gastric cancer (GC). Currently, PD-1 inhibitors combined with chemotherapy are the standard treatment for patients with HER2-negative/positive locally advanced or metastatic GC. Programmed death-ligand 1 (PD-L1) expression, as assessed using immunohistochemistry (IHC), is a crucial biomarker for predicting response to anti-PD-1/PD-L1 agents in various solid tumors, including GC. In GC, the PD-L1 IHC test serves as a companion or complementary diagnostic test for immunotherapy, and an accurate interpretation of PD-L1 status is essential for selecting patients who may benefit from immunotherapy. However, PD-L1 IHC testing presents several challenges that limit its reliability as a biomarker for immunotherapy. In this review, we provide an overview of the current practices of immunotherapy and PD-L1 testing in GC. In addition, we discuss the clinical challenges associated with PD-L1 testing and its future use as a biomarker for immunotherapy. Finally, we present prospective biomarkers currently under investigation as alternative predictors of immunotherapy response in GC.

Predicting response to PD-1 Inhibitors in head and neck squamous cell carcinomas using peripheral blood inflammatory markers

Immune checkpoint inhibitor (ICI) treatment has the potential to induce durable disease remission. However, the current combined positive score (CPS) is insufficient accurate for predicting which patients will benefit from it. In the present study, a real-world retrospective study was conducted on 56 patients of HNSCC who received ICI treatment. Then the treatment that patient received and levels of pre-treatment blood inflammatory markers (NLR, MLR and PLR) were identified to develop a model for predicting immunotherapy response. Notably, the model achieved an area under the curve (AUC) of 0.877 (95% CI 0.769-0.985) , providing a larger net benefit than the CPS marker (AUC=0.614, 95% CI 0.466-0.762). Furthermore, the internal validation of the prediction model showed a C-index of 0.835. Patients with high score of the model would get improved PFS than those with low score. Therefore, the prediction model for patients with local advanced or R/M HNSCC receiving ICI treatment, which represented an better efficient prediction of immunotherapy response than CPS marker.

KLRD1 (CD94): A Prognostic Biomarker and Therapeutic Candidate in Head and Neck Squamous Cell Carcinoma.

Background: Killer Cell Lectin Like Receptor D1 (KLRD1) plays a crucial role in antitumor immunity. However, its expression patterns across various cancers, its relationship with patient prognosis, and its potential as an immunotherapy target remain inadequately understood. Methods: We analyzed KLRD1 expression across various cancer types using multi-omics data from The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), and Gene Expression Omnibus (GEO) databases, correlating it with patient prognosis. Single-cell RNA sequencing data were employed to further explore KLRD1 expression in natural killer (NK) cells and exhausted CD8+ T cells (CD8Tex). Functional enrichment analyses using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) identified the biological processes and pathways associated with KLRD1. Immune infiltration analysis, conducted via CIBERSORT, assessed the relationship between KLRD1 expression and immune cell infiltration within the tumor microenvironment. Furthermore, the Tracking Tumor Immunophenotype (TIP) meta-server and Easier tool were employed to assess the role of KLRD1 in the cancer immunity cycle and to predict immunotherapy responses. Drug sensitivity was predicted using tools like CellMiner and the Genomics of Drug Sensitivity in Cancer (GDSC) database to explore the link between KLRD1 expression and responsiveness to various anticancer drugs. Results: KLRD1 exhibits significant differential expression and strong prognostic value across cancers, particularly as an independent prognostic factor in head and neck squamous cell carcinoma (HNSC). Single-cell analysis revealed high expression of KLRD1 in NK and CD8Tex cells, indicating its critical role in antitumor immune responses. Functional enrichment analyses showed that KLRD1 is involved in several immune-related signaling pathways, including NK cell-mediated cytotoxicity and T cell receptor pathways. Immune infiltration analysis further confirmed a positive correlation between KLRD1 expression and the infiltration of various immune cells. Moreover, higher KLRD1 expression in HNSC is associated with enhanced immune pathway activity, increased sensitivity to cell division inhibitors, and the identification of arachidonyltrifluoromethane as a potential compound to counteract its oncogenic effects. Conclusion: In HNSC, KLRD1 is a key prognostic marker and potential target for personalized immunotherapy.

A Distinguished Roadmap of Fibroblast Senescence in Predicting Immunotherapy Response and Prognosis Across Human Cancers.

The resistance of tumors to immune checkpoint inhibitors (ICI) may be intricately linked to cellular senescence, although definitive clinical validation remains elusive. In this study, comprehensive pan-cancer scRNA-seq analyses identify fibroblasts as exhibiting the most pronounced levels of cellular senescence among tumor-associated cell populations. To elucidate this phenomenon, a fibroblast senescence-associated transcriptomic signature (FSS), which correlated strongly with protumorigenic signaling pathways and immune dysregulation that fosters tumor progression, is developed. Leveraging the FSS, the machine learning (ML) framework demonstrates exceptional accuracy in predicting ICI response and survival outcomes, achieving superior area under curve (AUC) values across validation, testing, and in-house cohorts. Strikingly, FSS consistently outperforms established signatures in predictive robustness across diverse cancer subtypes. From an integrative analysis of 17 CRISPR/Cas9 libraries, CDC6 emerges as a pivotal biomarker for pan-cancer ICI response and prognostic stratification. Mechanistically, experimental evidence reveals that CDC6 in tumor cells orchestrates fibroblast senescence via TGF-β1 secretion and oxidative stress, subsequently reprogramming the tumor microenvironment and modulating ICI response. These findings underscore the translational potential of targeting fibroblast senescence as a novel therapeutic strategy to mitigate immune resistance and enhance antitumor efficacy.

Optimizing T cell inflamed signature through a combination biomarker approach for predicting immunotherapy response in NSCLC

The introduction of anti-PD-1/PD-L1 therapies revolutionized treatment for advanced non-small cell lung cancer (NSCLC), yet response rates remain modest, underscoring the need for predictive biomarkers. While a T cell inflamed gene expression profile (GEP) has predicted anti-PD-1 response in various cancers, it failed in a large NSCLC cohort from the Stand Up To Cancer-Mark (SU2C-MARK) Foundation. Re-analysis revealed that while the T cell inflamed GEP alone was not predictive, its performance improved significantly when combined with gene signatures of myeloid cell markers. These additional signatures, however, showed negative contributions to prediction, hinting at immune alterations affecting therapy. Based on this, we proposed a combination biomarker approach that integrates the T cell inflamed GEP with immune-altered signatures, derived from the SU2C-MARK cohort using a machine-learning approach, as novel biomarkers. These signatures consisted of genes highly expressed in myeloid and stromal cells. We then assessed the predictive ability of these combined biomarkers in six independent cancer cohorts treated with anti-PD-1. The combined biomarkers demonstrated enhanced performance in NSCLC and gastric cancer cohorts, but not in melanoma cohorts. Our study introduces new biomarkers for predicting anti-PD-(L)1 response in NSCLC and offers mechanistic insights into treatment efficacy.

Bioinformatics-based modeling of lung squamous cell carcinoma prognosis and prediction of immunotherapy response

Lung squamous cell carcinoma (LUSC) is a subtype of non-small cell lung cancer. It has a grim prognosis for patients, primarily because the disease often remains asymptomatic in its early stages. As a result, it is frequently diagnosed at an advanced stage, limiting treatment options. This underscores the importance of studying potential biomarkers and developing personalized treatment strategies. In this study, we used an advanced bioinformatics approach, integrating two authoritative databases, NCBI's GEO and TCGA, to perform a large-scale cross-platform gene expression analysis. To deeply mine the gene expression data of a large number of lung squamous carcinoma samples, we used a screening strategy based on median absolute deviation to select genes that differed significantly in multiple datasets. The expression variations of these genes between normal and cancerous tissues provided us with valuable clues revealing key molecules that may be involved in the disease process. Through rigorous statistical tests, we identified 36 genes that were significantly associated with patient survival, and further constructed a model using Cox proportional risk model containing 11 key genes (MRPL40, GABPB1AS1, PTPN3, SNCA, PYGB, RAP1, VDR, PHPT1, KIAA0100, TBC1D30, CYP7B1) in a risk prediction model. The prediction model not only reflects the strong correlation between gene expression and LUSC prognosis, but also provides clinicians with an effective tool to predict patients' survival prospects. In the future, this model is expected to guide the development of individualized treatment plans, thereby improving the quality of life and overall prognosis of patients.

Deep Learning Model for Predicting Immunotherapy Response in Advanced Non−Small Cell Lung Cancer

Only a small fraction of patients with advanced non-small cell lung cancer (NSCLC) respond to immune checkpoint inhibitor (ICI) treatment. For optimal personalized NSCLC care, it is imperative to identify patients who are most likely to benefit from immunotherapy. To develop a supervised deep learning-based ICI response prediction method; evaluate its performance alongside other known predictive biomarkers; and assess its association with clinical outcomes in patients with advanced NSCLC. This multicenter cohort study developed and independently validated a deep learning-based response stratification model for predicting ICI treatment outcome in patients with advanced NSCLC from whole slide hematoxylin and eosin-stained images. Images for model development and validation were obtained from 1 participating center in the US and 3 in the European Union (EU) from August 2014 to December 2022. Data analyses were performed from September 2022 to May 2024. Monotherapy with ICIs. Model performance measured by clinical end points and objective response rate (ORR) differentiation power vs other predictive biomarkers, ie, programmed death-ligand 1 (PD-L1), tumor mutational burden (TMB), and tumor-infiltrating lymphocytes (TILs). A total of 295 581 image tiles from 958 patients (mean [SD] age, 66.0 [10.6] years; 456 [48%] females and 502 [52%] males) treated with ICI for NSCLC were included in the analysis. The US-based development cohort consisted of 614 patients with median (IQR) follow-up time of 54.5 (38.2-68.1) months, and the EU-based validation cohort, 344 patients with 43.3 (27.4-53.9) months of follow-up. The ORR to ICI was 26% in the developmental cohort and 28% in the validation cohort. The deep learning model's area under the receiver operating characteristic curve (AUC) for ORR was 0.75 (95% CI, 0.64-0.85) in the internal test set and 0.66 (95% CI, 0.60-0.72) in the validation cohort. In a multivariable analysis, the deep learning model's score was an independent predictor of ICI response in the validation cohort for both progression-free (hazard ratio, 0.56; 95% CI, 0.42-0.76; P < .001) and overall survival (hazard ratio, 0.53; 95% CI, 0.39-0.73; P < .001). The tuned deep learning model achieved a higher AUC than TMB, TILs, and PD-L1 in the internal set; in the validation cohort, it was superior to TILs and comparable with PD-L1 (AUC, 0.67; 95% CI, 0.60-0.74), with a 10-percentage point improvement in specificity. In the validation cohort, combining the deep learning model with PD-L1 scores achieved an AUC of 0.70 (95% CI, 0.63-0.76), outperforming either marker alone, with a response rate of 51% compared to 41% for PD-L1 (≥50%) alone. The findings of this cohort study demonstrate a strong and independent deep learning-based feature associated with ICI response in patients with NSCLC across various cohorts. Clinical use of this deep learning model could refine treatment precision and better identify patients who are likely to benefit from ICI for treatment of advanced NSCLC.

Multi-omics analysis and experiments uncover the function of cancer stemness in ovarian cancer and establish a machine learning-based model for predicting immunotherapy responses.

The heterogeneity of cancer makes it challenging to predict its response to immunotherapy, highlighting the need to find reliable biomarkers for assessment. The sophisticated role of cancer stemness in mediating resistance to immune checkpoint inhibitors (ICIs) is still inadequately comprehended. Genome-scale CRISPR screening of RNA sequencing data from Project Achilles was utilized to pinpoint crucial genes unique to Ovarian Cancer (OV). Thirteen publicly accessible OV transcriptomic datasets, seven pan-cancer ICI transcriptomic cohorts, and one single-cell RNA dataset from melanoma patients treated with PD-1 were utilized to scale a novel cancer stemness index (CSI). An OV single-cell RNA dataset was amassed and scrutinized to uncover the role of Small Nuclear Ribonucleoprotein Polypeptide E (SNRPE) in the tumor microenvironment (TME). Vitro experiments were performed to validate the function of SNRPE in promoting proliferation and migration of ovarian cancer. Through the analysis of extensive datasets on ovarian cancer, a specific gene set that impacts the stemness characteristics of tumors has been identified and we unveiled a negative correlation between cancer stemness, and benefits of ICI treatment in single cell ICI cohorts. This identified gene set underpinned the development of the CSI, a groundbreaking tool leveraging advanced machine learning to predict prognosis and immunotherapy responses in ovarian cancer patients. The accuracy of the CSI was further confirmed by applying PD1/PD-L1 ICI transcriptomic cohorts, with a mean AUC exceeding 0.8 for predicting tumor progression and immunotherapy benefits. Remarkably, when compared to existing immunotherapy and prognosis markers, CSI exhibited superior predictive capabilities across various datasets. Interestingly, our research unveiled that the amplification of SNRPE contribute to remodeling the TME and promoting the evasion of malignant cells from immune system recognition and SNRPE can server as a novel biomarker for predicting immunotherapy response. A strong relationship between cancer stemness and the response to immunotherapy has been identified in our study. This finding provides valuable insights for devising efficient strategies to address immune evasion by targeting the regulation of genes associated with cellular stemness.