Plasma cell marker, immunoglobulin J polypeptide, predicts early disease-specific mortality in HPV+ HNSCC

From gene modules to gene markers: an integrated AI-human approach selects CD38 to represent plasma cell-associated transcriptional signatures

Background

Knowledge-driven prioritization of candidate genes derived from large-scale molecular profiling data for targeted transcriptional profiling assays is challenging due to the vast amount of biomedical literature that needs to be harnessed. We present a workflow leveraging Large Language Models (LLMs) to prioritize candidate genes within module M12.15, a plasma cell-associated module from the BloodGen3 repertoire, by integrating knowledge-driven prioritization with data-driven analysis of transcriptome profiles.

Methods

The workflow involves a two-step process: (1) high-throughput screening using LLMs to score and rank the 17 genes of module M12.15 based on six predefined criteria, and (2) prioritization employing high-resolution scoring and fact-checking, with human experts validating and refining AI-generated scores.

Results

The first step identified five candidate genes (CD38, TNFRSF17, IGJ, TOP2A, and TYMS). Following human-augmented LLM scoring and fact checking, as part of the second step, CD38 and TNFRSF17 emerged as the top candidates. Next, transcriptome profiling data from three datasets was incorporated in the workflow to assess expression levels and correlations with the module average across various conditions and cell types. It is on this basis that CD38 was prioritized as the top candidate, with TNFRSF17 and IGJ identified as promising alternatives.

Conclusion

This study introduces a systematic framework that integrates LLMs with human expertise for gene prioritization. Our analysis identified CD38, TNFRSF17, and IGJ as the top candidates within the plasma cell-associated module M12.15 from the BloodGen3 repertoire, with their relative rankings varying systematically based on specific evaluation criteria, from plasma cell biology to therapeutic relevance. This criterion-dependent ranking demonstrates the ability of the framework to perform nuanced, multi-faceted evaluations. By combining knowledge-driven analysis with data-driven metrics, our approach provides a balanced and comprehensive method for biomarker selection. The methodology established here offers a reproducible and scalable approach that can be applied across diverse biological contexts and extended to analyze large module repertoires.

Tumor and blood B-cell abundance outperforms established immune checkpoint blockade response prediction signatures in head and neck cancer

BACKGROUND

Immunotherapy has improved the outcomes for some patients with head and neck squamous-cell carcinoma (HNSCC). However, the low and variable response rates observed highlight the need for robust response biomarkers to select patients for treatment.

PATIENTS AND METHODS

We assembled and analyzed a large HNSCC dataset, encompassing 11 clinical cohorts including 1232 patient samples, spanning a variety of disease subtypes and immune checkpoint blockade (ICB) treatment types, tissue sources, data modalities, and timing of measurements. We conducted a comprehensive evaluation of the predictive power of various cell types, traditional biomarkers, and emerging predictors in both blood and tumor tissues of HNSCC patients.

RESULTS

Tumor B-cell infiltration emerged as a strong and robust predictor of both patient survival and ICB response. It outperformed all other established biomarkers of response to ICB, including the tertiary lymphoid structure signature and numerous T-cell-based signatures. B-cell infiltration was associated with a 'hot' antitumor microenvironment that promotes tumor eradication. Furthermore, B-cell levels in peripheral blood mononuclear cells (PBMCs) correlated strongly with tumor B-cell levels and demonstrated high predictive value for ICB response, with high odds ratios (≥7.8) in two independent clinical cohorts.

CONCLUSION

B-cell abundance, whether assessed in PBMCs or tumor tissues, is one of the strongest predictors of ICB response in HNSCC. For translation to patient care, measuring B-cell abundance in PBMCs via cytometry offers a practical and accessible tool for clinical decision making.

B-cell infiltration is associated with survival outcomes following programmed cell death protein 1 inhibition in head and neck squamous cell carcinoma

BACKGROUND

Programmed cell death protein 1 (PD-1) axis blockade has become the mainstay in the treatment of recurrent and/or metastatic (R/M) head and neck squamous cell cancer (HNSCC). Programmed death-ligand 1 (PD-L1) is the only approved biomarker for patient selection; however, response rate is limited even among high expressors. Our primary objective was to investigate the association of immune cell-related biomarkers in the tumor and tumor microenvironment with PD-1 checkpoint inhibitors' outcomes in patients with R/M HNSCC.

PATIENTS AND METHODS

NCT03652142 was a prospective study in nivolumab-treated platinum-refractory R/M HNSCC, aiming to evaluate biomarkers of response to treatment. Tumor biopsies and blood samples were collected from 60 patients at baseline, post-treatment, and at progression. Immune cells in the tumor and stromal compartments were quantified by immunofluorescence using a five-protein panel (CD3, CD8, CD20, FoxP3, cytokeratin). Tertiary lymphoid structures (TLSs), PD-L1 expression, and peripheral blood immune cell composition were also evaluated for associations with outcome. Our findings were validated by gene set enrichment analysis (GSEA) messenger RNA in situ expression data from the same patients, for B-cell- and TLS-associated genes.

RESULTS

High pre-treatment density of stromal B cells was associated with prolonged progression-free survival (PFS) (P = 0.011). This result was validated by GSEA, as stromal enrichment with B-cell-associated genes showed association with response to nivolumab. PD-L1 positivity combined with high B-cell counts in stroma defined a subgroup with significantly longer PFS and overall survival (P = 0.013 and P = 0.0028, respectively).

CONCLUSIONS

Increased B cells in pre-treatment HNSCC biopsy samples correlate with prolonged benefit from PD-1-based immunotherapy and could further enhance the predictive value of PD-L1 expression.

Transcriptomic analysis of mRNAs in human whole blood identified age-specific changes in healthy individuals

Older age is one of the most important shared risk factors for multiple chronic diseases, increasing the medical burden to contemporary societies. Current research focuses on identifying aging biomarkers to predict aging trajectories and developing interventions aimed at preventing and delaying the progression of multimorbidity with aging. Here, a transcriptomic changes analysis of whole blood genes with age was conducted. The age-related whole blood gene-expression profiling datasets were downloaded from the Gene Expression Omnibus (GEO) database. We screened the differentially expressed genes (DEGs) between healthy young and old individuals and performed functional enrichment analysis. Cytoscape with Cytohubba and MCODE was used to perform an interaction network of DEGs and identify hub genes. In addition, ROC curves and correlation analysis were used to evaluate the accuracy of hub genes. In total, we identified 29 DEGs between young and old samples that were enriched mainly in immunoglobulin binding and complex, humoral immune response, and immune response-activating signaling pathways. In combination with the PPI network and topological analysis, 4 hub genes (IGLL5, Jchain, POU2AF1, and Bach2) were identified. Pearson analysis showed that the expression changes of these hub genes were highly correlated with age. Among them, 3 hub genes (IGLL5, POU2AF1, and Bach2) were identified with good accuracy (AUC score > 0.7), indicating that these genes were the best indicators of age. Together, our results provided potential biomarkers IGLL5, POU2AF1, and Bach2 to identify individuals at high early risk of age-related disease to be targeted for early interventions and contribute to understanding the molecular mechanisms in the progression of aging.

IGJ suppresses breast cancer growth and metastasis by inhibiting EMT via the NF‑κB signaling pathway

Breast cancer metastasis is the primary cause of mortality of patients with breast cancer. The present study aimed to explore the role and underlying mechanisms of IGJ in the invasion and metastasis of breast cancer. The Cancer Genome Atlas database was utilized to analyze the differential gene expression profiles in patients with breast cancer with or without metastasis; the target gene, joining chain of multimeric IgA and IgM (JCHAIN, also known as IGJ, as referred to herein), with significant expression and with prognostic value was screened. The expression levels of IGJ in human breast cancer paired tissues and cell lines were detected using reverse transcription‑quantitative PCR and western blot analysis. IGJ differential expression was detected in paired human breast cancer tissues using immunohistochemistry. The role of IGJ in breast cancer was verified using CCK‑8, invasion and migration assays, and scratch tests in vivo and in vitro. Further exploration of the role and mechanism of IGJ in breast cancer was conducted through Gene Set Enrichment Analysis, Kyoto Encyclopedia of Genes and Genomes analysis, western blot analysis and immunofluorescence experiments. Through the analysis of gene expression profiles, it was found that IGJ was poorly expressed in patients with breast cancer with metastasis compared to patients with non‑metastatic breast cancer. The overexpression of IGJ was associated with an improved distant metastasis‑free survival and overall survival (OS). COX multivariate regression analysis demonstrated that IGJ was an independent prognostic factor for the OS and relapse‑free survival of patients with breast cancer. In comparison to healthy breast cancer adjacent tissues and cell lines, IGJ was poorly expressed in breast cancer tissues and cell lines (P<0.05). Further analyses indicated that the overexpression of IGJ suppressed the proliferation, invasion and metastasis of breast cancer cells in vivo and in vitro by inhibiting the occurrence of epithelial‑to‑mesenchymal transition (EMT) and suppressing the nuclear translocation of p65. Finally, rescue experiments indicated that IGJ restricted the proliferation and metastasis of breast cancer cells by regulating the NF‑κB signaling pathway. On the whole, the present study demonstrates that IGJ suppresses the invasion and metastasis of breast cancer by inhibiting both the occurrence of EMT and the NF‑κB signaling pathway. These findings may provide novel biomarkers and potential therapeutic targets for the treatment of metastatic breast cancer.

Multifactorial estimation of clinical outcome in HPV-associated oropharyngeal squamous cell carcinoma via automated image analysis of routine diagnostic H&E slides and neural network modelling

OBJECTIVE

Routine haematoxylin and eosin (H&E) photomicrographs from human papillomavirus-associated oropharyngeal squamous cell carcinomas (HPV + OpSCC) contain a wealth of prognostic information. In this study, we developed a high content image analysis (HCIA) workflow to quantify features of H&E images from HPV + OpSCC patients to identify prognostic features and predict patient outcomes.

METHODS

First, we have developed an open-source HCIA tool for single-cell segmentation and classification of H&E images. Subsequently, we have used our HCIA tool to analyse a set of 889 images from diagnostic H&E slides in a retrospective cohort of HPV + OpSCC patients with favourable (FO, n = 60) or unfavourable (UO, n = 30) outcomes. We have identified and measured 31 prognostic features which were quantified in each sample and used to train a neural network (NN) model to predict patient outcomes.

RESULTS

Univariate and multivariate statistical analyses revealed significant differences between FO and UO patients in 31 and 17 variables, respectively (P < 0.05). At the single-image level, the NN model had an overall accuracy of 72.5% and 71.2% in recognising FO and UO patients when applied to test or validation sets, respectively. When considering 10 images per patient, the accuracy of the NN model increased to 86.7% in the test set.

CONCLUSION

Our open-source H&E analysis workflow and predictive models confirm previously reported prognostic features and identifies novel factors which predict HPV + OpSCC outcomes with promising accuracy. Our work supports the use of machine learning in digital pathology to exploit clinically relevant features in routine diagnostic pathology without additional biomarkers.

Immunological Aspects of Human Papilloma Virus-Related Cancers Always Says, “I Am like a Box of Complexity, You Never Know What You Are Gonna Get”

The human papillomavirus (HPV) can cause different cancers in both men and women. The virus interferes with functions of the cervix, vulva, vagina, anus in the anogenital area, breast, and head and neck cancer due to the local lesions. The tumors lead to death if not treated as a result of distant metastasis to internal organs and brain. Moreover, HPV attenuates the immune system during chronic infection and releases viral antigens into the tumor microenvironment. The tumors know how difficult is to win the battle with a strong united army of immune cells that are equipped with cytokines and enzymes. They confuse the immune cells with secreting viral antigens. The immune system is equipped with cytokines, a complement system, antibodies, and other secretory proteins to overcome the foreign invaders and viral antigens. However, the majority of the time, tumors win the battle without having all the equipment of the immune cells. Thus, in this review, we describe the recent progression in cellular and humoral immunity studies during the progression of HPV-related cancers. First of all, we describe the role of B, plasmoid cells, and B regulatory cells (Breg) in their functions in the tumor microenvironment. Then, different subtypes of T cells such as T CD8, CD4, T regulatory (Treg) cells were studied in recently published papers. Furthermore, NK cells and their role in tumor progression and prevention were studied. Finally, we indicate the breakthroughs in immunotherapy techniques for HPV-related cancers.

Cancer Progression Mediated by CAFs Relating to HCC and Identification of Genetic Characteristics Influencing Prognosis

BACKGROUND

Hepatocellular carcinoma (HCC) is one of the most common malignancies, and although there are several treatment options, the overall results are not satisfactory. Cancer-associated fibroblasts (CAFs) can promote cancer progression through various mechanisms.

METHODS

HCC-associated mRNA data were sourced from The Cancer Genome Atlas database (TCGA) and International Cancer Genome Consortium (ICGC) database. First, the differentially expressed CAF-related genes (CAF-DEGs) were acquired by difference analysis and weighted gene coexpression network analysis (WGCNA). Moreover, a CAF-related risk model was built by Cox analysis. Kaplan-Meier (K-M) curves and receiver operating characteristic (ROC) curves were utilized to evaluate the validity of this risk model. Furthermore, enrichment analysis of differentially expressed genes (DEGs) between the high- and low-risk groups was executed to explore the functions relevant to the risk model. Furthermore, this study compared the differences in immune infiltration, immunotherapy, and drug sensitivity between the high- and low-risk groups. Finally, we verified the mRNA expression levels of selected prognostic genes by quantitative real-time polymerase chain reaction (qRT-PCR).

RESULTS

107 CAF-DEGs were identified in the HCC samples, and five prognosis-related genes (ACTA2, IGJ, CTHRC1, CXCL12, and LAMB1) were obtained by Cox analysis and utilized to build a CAF-related risk model. K-M analysis illustrated a low survival in the high-risk group, and ROC curves revealed that the risk model could accurately predict the 1-, 3-, and 5-year overall survival (OS) of HCC patients. In addition, Cox analysis demonstrated that the risk score was an independent prognostic factor. Enrichment analysis illustrated that DEGs between the high- and low-risk groups were related to immune response, amino acid metabolism, and fatty acid metabolism. Furthermore, risk scores were correlated with the tumor microenvironment, CAF scores, and TIDE scores, and CAF-related marker genes were positively correlated with all five model genes. Notably, the risk model was relevant to the sensitivity of chemotherapy drugs. Finally, the results of qRT-PCR demonstrated that the expression levels of 5 model genes were in accordance with the analysis.

CONCLUSION

A CAF-related risk model based on ACTA2, IGJ, CTHRC1, CXCL12, and LAMB1 was built and could be utilized to predict the prognosis and treatment of HCC.