Identification of critical radioresistance genes in esophageal squamous cell carcinoma by whole-exome sequencing

Advancing drug-response prediction using multi-modal and -omics machine learning integration (MOMLIN): a case study on breast cancer clinical data

The inherent heterogeneity of cancer contributes to highly variable responses to any anticancer treatments. This underscores the need to first identify precise biomarkers through complex multi-omics datasets that are now available. Although much research has focused on this aspect, identifying biomarkers associated with distinct drug responders still remains a major challenge. Here, we develop MOMLIN, a multi-modal and -omics machine learning integration framework, to enhance drug-response prediction. MOMLIN jointly utilizes sparse correlation algorithms and class-specific feature selection algorithms, which identifies multi-modal and -omics-associated interpretable components. MOMLIN was applied to 147 patients' breast cancer datasets (clinical, mutation, gene expression, tumor microenvironment cells and molecular pathways) to analyze drug-response class predictions for non-responders and variable responders. Notably, MOMLIN achieves an average AUC of 0.989, which is at least 10% greater when compared with current state-of-the-art (data integration analysis for biomarker discovery using latent components, multi-omics factor analysis, sparse canonical correlation analysis). Moreover, MOMLIN not only detects known individual biomarkers such as genes at mutation/expression level, most importantly, it correlates multi-modal and -omics network biomarkers for each response class. For example, an interaction between ER-negative-HMCN1-COL5A1 mutations-FBXO2-CSF3R expression-CD8 emerge as a multimodal biomarker for responders, potentially affecting antimicrobial peptides and FLT3 signaling pathways. In contrast, for resistance cases, a distinct combination of lymph node-TP53 mutation-PON3-ENSG00000261116 lncRNA expression-HLA-E-T-cell exclusions emerged as multimodal biomarkers, possibly impacting neurotransmitter release cycle pathway. MOMLIN, therefore, is expected advance precision medicine, such as to detect context-specific multi-omics network biomarkers and better predict drug-response classifications.

Esophageal Squamous Cancer from 4NQO-Induced Mice Model: CNV Alterations

Squamous esophageal carcinoma is a common pathological type of esophageal carcinoma around the world. The prognosis of esophageal carcinoma is usually poor and diagnosed at late stages. Recently, research suggested that genomic instability occurred in esophageal cells during the development of esophageal squamous cell carcinoma (ESCC). Identifying prognostic and specific genomic characteristics, especially at the early hyperplasia stage, is critical. Mice were given 4-nitroquinoline 1-oxide (4NQO) with drinking water to induce esophageal cancer. The immortalized human esophageal epithelial cell line (NE2) was also treated with 4NQO. We performed histologic analyses, immunofluorescence, and immunohistochemical staining to detect DNA damage at different time points. Whole-exome sequencing was accomplished on the esophagus tissues at different pathological stages to detect single-nucleotide variants and copy number variation (CNV) in the genome. Our findings indicate that all mice were tumor-forming, and a series of changes from simple hyperplasia (ESSH) to intraepithelial neoplasia (IEN) to esophageal squamous cell carcinoma (ESCC) was seen at different times. The expression of γ-H2AX increased from ESSH to ESCC. In addition, mutations of the Muc4 gene were detected throughout the pathological stages. Furthermore, CNV burden appeared in the esophageal tissues from the beginning of ESSH and accumulated more in cancer with the deepening of the lesions. This study demonstrates that mutations caused by the early appearance of DNA damage may appear in the early stage of malignant tissue before the emergence of atypia. The detection of CNV and mutations of the Muc4 gene may be used as an ultra-early screening indicator for esophageal cancer.

Identification of natural product 3, 5-diiodotyrosine as APOBEC3B inhibitor to prevent somatic mutation accumulation and cancer progression

BACKGROUND

The development of cancer is largely dependent on the accumulation of somatic mutations, indicating the potential to develop cancer chemoprevention agents targeting mutation drivers. However, ideal cancer chemoprevention agents that can effectively inhibit the mutation drivers have not been identified yet.

METHODS

The somatic mutation signatures and expression analyses of APOBEC3B were performed in patient with pan-cancer. The computer-aided screening and skeleton-based searching were performed to identify natural products that can inhibit the activity of APOBEC3B. 4-nitroquinoline-1-oxide (4-NQO)-induced spontaneous esophageal squamous cell carcinoma (ESCC) and azoxymethane/dextran sulfate sodium (AOM/DSS)-induced spontaneous colon cancer mouse models were conducted to investigate the influences of APOBEC3B inhibitor on the prevention of somatic mutation accumulation and cancer progression.

RESULTS

Here, we discovered that the cytidine deaminase APOBEC3B correlated somatic mutations were widely observed in a variety of cancers, and its overexpression indicated poor survival. SMC247 (3, 5-diiodotyrosine), as a source of kelp iodine without side effects, could strongly bind APOBEC3B (K_D=65 nM) and effectively inhibit its deaminase activity (IC₅₀=1.69 µM). Interestingly, 3, 5-diiodotyrosine could significantly reduce the clusters of mutations, prevent the precancerous lesion progression, and prolong the survival in 4-NQO-induced spontaneous ESCC and AOM/DSS-induced spontaneous colon cancer mouse models. Furthermore, 3, 5-diiodotyrosine could reduce colitis, increase the proportion and function of T lymphocytes via IL-15 in tumor microenvironment. The synergistic cancer prevention effects were observed when 3, 5-diiodotyrosine combined with PD-1/PD-L1 blockade.

CONCLUSIONS

This is the first prove-of-concept study to elucidate that the natural product 3, 5-diiodotyrosine could prevent somatic mutation accumulation and cancer progression through inhibiting the enzymatic activity of APOBEC3B. In addition, 3, 5-diiodotyrosine could reduce the colitis and increase the infiltration and function of T lymphocytes via IL-15 in tumor microenvironment. 3, 5-diiodotyrosine combined with PD-1/PD-L1 blockade could elicit synergistic cancer prevention effects, indicating a novel strategy for both prevent the somatic mutation accumulation and the immune-suppressive microenvironment exacerbation.

Association of MUC19 Mutation With Clinical Benefits of Anti-PD-1 Inhibitors in Non-small Cell Lung Cancer

Although anti-PD-1 inhibitors exhibit impressive clinical results in non-small cell lung cancer (NSCLC) cases, a substantial percentage of patients do not respond to this treatment. Moreover, the current recommended biomarkers are not perfect. Therefore, it is essential to discover novel molecular determinants of responses to anti-PD-1 inhibitors. We performed Whole Exome Sequencing (WES) in a cohort of 33 Chinese NSCLC patients. Patients were classified into the durable clinical benefit (DCB) and no durable benefit (NDB) groups. Infiltrating CD8⁺ cells in the tumor microenvironment (TME) were investigated by immunohistochemistry. We also used public datasets to validate our results. In our cohort, good clinical responses to anti-PD-1 inhibitors were more pronounced in younger patients with lower Eastern Cooperative Oncology Group (ECOG) scores and only extra-pulmonary metastasis. More importantly, we identified a novel MUC19 mutation, which was significantly enriched in DCB patients (P = 0.015), and MUC19-mutated patients had a longer progression-free survival (PFS) (hazard ratio = 0.3, 95% CI 0.1-0.9; P = 0.026). Immunohistochemistry results indicated that the MUC19 mutation was associated with increased infiltration by CD8⁺ T cells in the TME (P = 0.0313). When combining MUC19 mutation with ECOG scores and intra-pulmonary metastasis status, patients with more positive predictors had longer PFS (P = 0.003). Furthermore, MUC19 mutation was involved in immune responses and associated with a longer PFS in the Memorial Sloan-Kettering Cancer Center (MSKCC) cohort. Collectively, we identified that MUC19 mutations were involved in immune responses, and NSCLC tumors harboring mutated MUC19 exhibited good responses to anti-PD-1 inhibitors.

Inhibition of SOCS6 confers radioresistance in esophageal squamous cell carcinoma

Esophageal cancer is one of the most common cancer of the digestive system and radiotherapy is widely applied in advanced esophageal cancer treatment, however radioresistance (RR) is one of the major reasons for radiotherapy failure. There is limited knowledge on the mechanisms that cause RR, here we identify suppressors of cytokine signaling 6 (SOCS6) is a negative regulator of radioresistance in ESCC cells. SOCS6 deficiency in ESCC cells conferred radioresistance in vitro and in vivo by increasing radiation-induced G2/M arrest, DNA damage repair and inhibiting radiation-induced apoptosis. Moreover, the transcriptome sequencing analysis demonstrates that the transcription of SOCS6 was partially p53-dependent. Importantly we found that highly correlated SOCS6 and P53 express lower in RR esophageal cancer tissues compare with radiosensitive ones. Collectedly our study uncovers that SOCS6, as a downstream effector of p53, is a key regulator involved in the radioresistance of ESCC.