Deficient HER3 expression in poorly-differentiated colorectal cancer cells enhances gefitinib sensitivity

T2-weighted imaging-based radiomic-clinical machine learning model for predicting the differentiation of colorectal adenocarcinoma

BACKGROUND

The study on predicting the differentiation grade of colorectal cancer (CRC) based on magnetic resonance imaging (MRI) has not been reported yet. Developing a non-invasive model to predict the differentiation grade of CRC is of great value.

AIM

To develop and validate machine learning-based models for predicting the differentiation grade of CRC based on T2-weighted images (T2WI).

METHODS

We retrospectively collected the preoperative imaging and clinical data of 315 patients with CRC who underwent surgery from March 2018 to July 2023. Patients were randomly assigned to a training cohort (n = 220) or a validation cohort (n = 95) at a 7:3 ratio. Lesions were delineated layer by layer on high-resolution T2WI. Least absolute shrinkage and selection operator regression was applied to screen for radiomic features. Radiomics and clinical models were constructed using the multilayer perceptron (MLP) algorithm. These radiomic features and clinically relevant variables (selected based on a significance level of P < 0.05 in the training set) were used to construct radiomics-clinical models. The performance of the three models (clinical, radiomic, and radiomic-clinical model) were evaluated using the area under the curve (AUC), calibration curve and decision curve analysis (DCA).

RESULTS

After feature selection, eight radiomic features were retained from the initial 1781 features to construct the radiomic model. Eight different classifiers, including logistic regression, support vector machine, k-nearest neighbours, random forest, extreme trees, extreme gradient boosting, light gradient boosting machine, and MLP, were used to construct the model, with MLP demonstrating the best diagnostic performance. The AUC of the radiomic-clinical model was 0.862 (95%CI: 0.796-0.927) in the training cohort and 0.761 (95%CI: 0.635-0.887) in the validation cohort. The AUC for the radiomic model was 0.796 (95%CI: 0.723-0.869) in the training cohort and 0.735 (95%CI: 0.604-0.866) in the validation cohort. The clinical model achieved an AUC of 0.751 (95%CI: 0.661-0.842) in the training cohort and 0.676 (95%CI: 0.525-0.827) in the validation cohort. All three models demonstrated good accuracy. In the training cohort, the AUC of the radiomic-clinical model was significantly greater than that of the clinical model (P = 0.005) and the radiomic model (P = 0.016). DCA confirmed the clinical practicality of incorporating radiomic features into the diagnostic process.

CONCLUSION

In this study, we successfully developed and validated a T2WI-based machine learning model as an auxiliary tool for the preoperative differentiation between well/moderately and poorly differentiated CRC. This novel approach may assist clinicians in personalizing treatment strategies for patients and improving treatment efficacy.

Prediction model of poorly differentiated colorectal cancer (CRC) based on gut bacteria

BACKGROUND

The mortality of colorectal cancer is high, the malignant degree of poorly differentiated colorectal cancer is high, and the prognosis is poor.

OBJECTIVE

To screen the characteristic intestinal microbiota of poorly differentiated intestinal cancer.

METHODS

Fecal samples were collected from 124 patients with moderately differentiated CRC and 123 patients with poorly differentiated CRC, and the bacterial 16S rRNA V1-V4 region of the fecal samples was sequenced. Alpha diversity analysis was performed on fecal samples to assess the diversity and abundance of flora. The RDP classifier Bayesian algorithm was used to analyze the community structure. Linear discriminant analysis and Student's t test were used to screen the differences in flora. The PICRUSt1 method was used to predict the bacterial function, and six machine learning models, including logistic regression, random forest, neural network, support vector machine, CatBoost and gradient boosting decision tree, were used to construct a prediction model for the poor differentiation of colorectal cancer.

RESULTS

There was no significant difference in fecal flora alpha diversity between moderately and poorly differentiated colorectal cancer (P > 0.05). The bacteria that accounted for a large proportion of patients with poorly differentiated and moderately differentiated colorectal cancer were Blautia, Escherichia-Shigella, Streptococcus, Lactobacillus, and Bacteroides. At the genus level, there were nine bacteria with high abundance in the poorly differentiated group, including Bifidobacterium, norank_f__Oscillospiraceae, Eisenbergiella, etc. There were six bacteria with high abundance in the moderately differentiated group, including Megamonas, Erysipelotrichaceae_UCG-003, Actinomyces, etc. The RF model had the highest prediction accuracy (100.00% correct). The bacteria that had the greatest variable importance in the model were Pseudoramibacter, Megamonas and Bifidobacterium.

CONCLUSION

The degree of pathological differentiation of colorectal cancer was related to gut flora, and poorly differentiated colorectal cancer had some different bacterial flora, and intestinal bacteria can be used as biomarkers for predicting poorly differentiated CRC.

An Intratumor Heterogeneity-Related Signature for Predicting Prognosis, Immune Landscape, and Chemotherapy Response in Colon Adenocarcinoma

Background

Colon adenocarcinoma (COAD) is a frequent malignancy of the digestive system with a poor prognosis and high mortality rate worldwide. Intratumor heterogeneity (ITH) is associated with tumor progression, poor prognosis, immunosuppression, and therapy resistance. However, the relationship between ITH and prognosis, the immune microenvironment, and the chemotherapy response in COAD patients remains unknown, and this knowledge is urgently needed.

Methods

We obtained clinical information and gene expression data for COAD patients from The Cancer Genome Atlas (TCGA) database. The DEPTH2 algorithm was utilized to evaluate the ITH score. X-tile software was used to determine the optimal cutoff value of the ITH score. The COAD patients were divided into high- and low-ITH groups based on the cutoff value. We analyzed prognosis, tumor mutation burden (TMB), gene mutations, and immune checkpoint expression between the high- and low-ITH groups. Differentially expressed genes (DEGs) in the high- and low-ITH groups were subjected to Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. We performed univariate Cox regression and least absolute shrinkage and selection operator (LASSO) regression analyses to screen the prognosis-related genes for the construction of an ITH-related prognostic signature. The nomogram was used to predict the overall survival (OS) of COAD patients. The protein–protein interaction (PPI) network was constructed by using the GeneMANIA database. Principal component analysis (PCA) and single-sample gene set enrichment analysis (ssGSEA) were employed to explore the differences in biological pathway activation status between the high- and low-risk groups. The proportion and type of tumor-infiltrating immune cells were evaluated by the CIBERSORT and ESTIMATE algorithms. Additionally, we assessed the chemotherapy response and predicted small-molecule drugs for treatment. Finally, the expression of the prognosis-related genes was validated by using the UALCAN database and Human Protein Atlas (HPA) database.

Results

The OS of the high-ITH group was worse than that of the low-ITH group. A positive correlation between ITH and TMB was identified. In subgroups stratified by age, gender, and tumor stage, the OS of the low-ITH group remained better than that of the high-ITH group. There were dramatic differences in the mutated genes, single nucleotide variant classes, variant types, immune checkpoints and cooccurring and mutually exclusive mutations of the DEGs between the high- and low-ITH groups. Based on the DEGs between the high- and low-ITH groups, we constructed a five-gene signature consisting of CEACAM5, ENO2, GABBR1, MC1R, and SLC44A4. The COAD patients were divided into high- and low-risk groups according to the median risk score. The OS of the high-risk group was worse than that of the low-risk group. The nomogram was used to accurately predict the 1-, 3- and 5-year OS of COAD patients and showed good calibration and moderate discrimination ability. The stromal score, immune score, and ESTIMATE score of the high-risk group were significantly higher than those of the low-risk group, whereas tumor purity showed the opposite trend. The patients classified by the risk score had distinguishable sensitivity to chemotherapeutic drugs. Finally, two public databases confirmed that CEACAM5 and SLC44A4 were upregulated in normal tissues compared with COAD tissues, and ENO2, GABBR1, and MC1R were upregulated in COAD tissues compared with normal tissues.

Conclusion

Overall, we identified an ITH-related prognostic signature for COAD that was closely related to the tumor microenvironment and chemotherapy response. This signature may help clinicians make more personalized and precise treatment decisions for COAD patients.

The power of heteronemin in cancers

Heteronemin (Haimian jing) is a sesterterpenoid-type natural marine product that is isolated from sponges and has anticancer properties. It inhibits cancer cell proliferation via different mechanisms, such as reactive oxygen species (ROS) production, cell cycle arrest, apoptosis as well as proliferative gene changes in various types of cancers. Recently, the novel structure and bioactivity evaluation of heteronemin has received extensive attention. Hormones control physiological activities regularly, however, they may also affect several abnormalities such as cancer. L-Thyroxine (T₄), steroid hormones, and epidermal growth factor (EGF) up-regulate the accumulation of checkpoint programmed death-ligand 1 (PD-L1) and promote inflammation in cancer cells. Heteronemin suppresses PD-L1 expression and reduces the PD-L1-induced proliferative effect. In the current review, we evaluated research and evidence regarding the antitumor effects of heteronemin and the antagonizing effects of non-peptide hormones and growth factors on heteronemin-induced anti-cancer properties and utilized computational molecular modeling to explain how these ligands interacted with the integrin αvβ3 receptors. On the other hand, thyroid hormone deaminated analogue, tetraiodothyroacetic acid (tetrac), modulates signal pathways and inhibits cancer growth and metastasis. The combination of heteronemin and tetrac derivatives has been demonstrated to compensate for anti-proliferation in cancer cells under different circumstances. Overall, this review outlines the potential of heteronemin in managing different types of cancers that may lead to its clinical development as an anticancer agent.

MiR-323a regulates ErbB3/EGFR and blocks gefitinib resistance acquisition in colorectal cancer

The rapid onset of resistance to epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) limits its clinical utility in colorectal cancer (CRC) patients, and pan-erb-b2 receptor tyrosine kinase (ErbB) treatment strategy may be the alternative solution. The aim of this study was to develop a possible microRNA multi-ErbB treatment strategy to overcome EGFR-TKI resistance. We detect the receptor tyrosine kinase activity in gefitinib-resistant colorectal cancer cells, ErbB3/EGFR is significantly activated and provides a potential multi-ErbB treatment target. MiR-323a-3p, a tumor suppressor, could target both ErbB3 and EGFR directly. Apoptosis is the miR-323a-3p inducing main biological process by functional enrichment analysis, and The EGFR and ErbB signaling are the miR-323a-3p inducing main pathway by KEGG analysis. MiR-323a-3p promotes CRC cells apoptosis by targeting ErbB3-phosphoinositide 3-kinases (PI3K)/PKB protein kinase (Akt)/glycogen synthase kinase 3 beta (GSK3β)/EGFR-extracellular regulated MAP kinase (Erk1/2) signaling directly. And miR-323a-3p, as a multi-ErbBs inhibitor, increase gefitinib sensitivity of the primary cell culture from combination miR-323a-3p and gefitinib treated subcutaneous tumors. MiR-323a-3p reverses ErbB3/EGFR signaling activation in gefitinib-resistant CRC cell lines and blocks acquired gefitinib resistance.

Cyclooxygenase-2 mediates gefitinib resistance in non-small cell lung cancer through the EGFR/PI3K/AKT axis

Gefitinib is a potent inhibitor of EGFR and represents the front-line treatment for non-small cell lung cancer (NSCLC) therapeutics. However, NSCLC patients are prone to develop acquired resistance through as yet, undefined mechanisms of resistance. Here, we investigated the role of COX-2 during gefitinib resistance in NSCLC cells and revealed its underlying mechanism(s) of action. We report the upregulation of COX-2 in gefitinib-resistant NSCLC tissues and cells, which is associated with poor prognosis. In vitro assays in NSCLC cells (PC9/GR) showed that COX-2 facilitates gefitinib resistance in NSCLC cells through its effects on P-gp, MRP1, and BCRP, and cancer cell migration and invasion. In vivo, COX-2 silencing could repress tumor growth. We found that the overexpression of COX-2 enhances the transcription of MMP-2, MMP-7, and MMP-9 which mediates PI3K-AKT activation. In summary, we demonstrate that COX-2 mediates the gefitinib resistance of NSCLC cells through its interaction with EGFR and the PI3K-AKT axis. This highlights COX-2 as a novel molecular target for NSCLC.

PAC-5 Gene Expression Signature for Predicting Prognosis of Patients with Pancreatic Adenocarcinoma

Pancreatic adenocarcinoma (PAC) is one of the most aggressive malignancies. Intratumoural molecular heterogeneity impedes improvement of the overall survival rate. Current pathological staging system is not sufficient to accurately predict prognostic outcomes. Thus, accurate prognostic model for patient survival and treatment decision is demanded. Using differentially expressed gene analysis between normal pancreas and PAC tissues, the cancer-specific genes were identified. A prognostic gene expression model was computed by LASSO regression analysis. The PAC-5 signature (LAMA3, E2F7, IFI44, SLC12A2, and LRIG1) that had significant prognostic value in the overall dataset was established, independently of the pathological stage. We provided evidence that the PAC-5 signature further refined the selection of the PAC patients who might benefit from postoperative therapies. SLC12A2 and LRIG1 interacted with the proteins that were implicated in resistance of EGFR kinase inhibitor. DNA methylation was significantly involved in the gene regulations of the PAC-5 signature. The PAC-5 signature provides new possibilities for improving the personalised therapeutic strategies. We suggest that the PAC-5 genes might be potential drug targets for PAC.

Association between the overexpression of Her3 and clinical pathology and prognosis of colorectal cancer: A meta-analysis

BACKGROUND

This study aimed to investigate the association between the overexpression of human epidermal growth factor receptor-3 (Her3) and the clinicopathological parameters and survival of patients with colorectal cancer (CRC).

METHODS

Relevant studies on the overexpression of Her3 (measured by immunohistochemistry) and overall survival (OS) in patients with CRC were searched for in PubMed, EMBASE, and Cochrane Library. Published data were extracted and computed into odds ratios (ORs) for assessing the association of Her3 overexpression with tumor differentiation, tumor node metastasis (TNM) stage, position of colon cancer, sex, and age. Prognostic data were computed into hazard ratios (HRs) for OS.

RESULTS

Eight studies including 1716 patients with CRC were included in this meta-analysis. The results revealed a significant association between Her3 overexpression and tumor differentiation [OR = 2.38; 95% confidence interval (95% CI): 1.76-3.22; P < .001], TNM tumor stage (OR = 0.71; 95% CI: 0.53-0.96; P = .03), and position of colon cancer (OR = 1.71; 95% CI: 1.28-2.27; P < .001). While patients with Her3 overexpression demonstrated a worse tumor response (OR = 0.31; 95% CI: 0.16-0.60; P < .001) and OS after treatment with cetuximab (HR = 1.86; 95% CI: 1.24-2.79; P = .003), they demonstrated better OS after symptomatic treatment (HR = 0.65; 95% CI: 0.50-0.85; P = .002). Her3 overexpression was not associated with sex (OR = 1.03; 95% CI: 0.83-1.28; P = .79), age (OR = 0.96; 95% CI: 0.75-1.24; P = .77), colon or rectum site (OR = 0.79; 95% CI: 0.44-1.43; P = .44), and total OS (HR = 1.09; 95% CI: 0.69-1.72; P = .72).

CONCLUSION

Her3 expression is associated with the clinical pathology and prognosis of CRC, which explains the nonefficacy of cetuximab treatment in patients with CRC.

Gefitinib enhances sensitivity of endometrial cancer cells to progestin therapy via dual-specificity phosphatase 1

In this study, we investigated if Gefitinib, an epidermal growth factor receptor (EGFR) inhibitor, augments endometrial cancer (EC) therapy with medroxyprogesterone acetate (MPA). Combined treatment with Gefitinib plus MPA decreased the proliferation and invasiveness of the Ishikawa and RL952 EC cell lines more effectively than MPA treatment alone. Moreover, combined treatment with Gefitinib plus MPA reduced growth of EC xenografts in Balb/c nude mice more than either Gefitinib or MPA alone. The therapeutic efficacy of combined Gefitinib plus MPA treatment was dependent on expression of dual-specificity phosphatase 1 (DUSP1). DUSP1 knockdown in Ishikawa cells treated with Gefitinib plus MPA showed greater proliferation and invasiveness than parental Ishikawa cells treated similarly. EC cells treated with the combination of Gefitinib plus MPA also showed DUSP1-dependent reductions in phospho-ERK1/2 and increases in E-Cadherin. Thus, Gefitinib appears to DUSP1-dependently enhance the therapeutic efficacy of progestin in EC cells.

HER3 expression is correlated to distally located and low-grade colon cancer

Background HER3 is a member of the human epidermal growth factor receptor complex (EGFR, HER2, HER3 and HER4). It has been investigated as a prognostic biomarker in colorectal cancer but is sparingly studied in colon cancer. HER3 can affect cellular proliferation, differentiation and migration in oncogenesis through ligand binding and activation of intracellular signal pathways. Recently, we found that expression of cell surface HER3 can be detected at a high extent in primary colorectal tumors, lymph node and liver metastases and that it correlated with poor prognosis. This large, explorative, retrospective study evaluates the prognostic value of HER3 in colon cancer and the association of HER3 to tumor location.

MATERIAL AND METHODS

Immunohistochemical detection with a monoclonal HER3 antibody in primary colon tumors of stage II and III, from 521 patients, was performed. Results HER3 was expressed at high levels in 67% of the colon tumors. High HER3 expression was associated with distal tumor location (p < 0.0001) and low-grade tumor (p < 0.0001). In the group of patients with distal colon cancer (230/521), HER3 expression correlated to shorter disease-free survival (DFS) (p = 0.03) in the univariate analysis and in the multivariate analysis, a hazard ratio of 0.56 (95% CI 0.31-0.99) (p = 0.047) was observed. Conclusion In this explorative, retrospective study, high HER3 expression in colon cancer was associated to distal colon location and low-grade tumor. High HER3 expression was of prognostic value according to DFS in distal colon cancer in univariate and multivariate analysis. We could not find a significant value of HER3 expression with respect to overall survival (OS).

Dual targeting of HER3 and EGFR in colorectal tumors might overcome anti-EGFR resistance

Multiple genetic alterations have been associated with resistance to anti-EGFR therapy in metastatic colorectal cancer (CRC) patients. Research has been mainly focused on driver mutations in KRAS, NRAS, BRAF and PI3K. However, recent evidence suggests a crucial role for non-genetic mechanisms in conferring resistance to anti-EGFR therapy. Specifically, the HER3 receptor is capable of heterodimerizing with multiple EGFR family members resulting in downstream activation of the PI3K and MAPK pathways. Monoclonal antibodies targeted against the HER3 receptor are being investigated in clinical trials; however, preliminary data has shown limited clinical activity. Thus, given the relevance of the HER3 receptor in activating downstream effector pathways and in conferring resistance to anti-EGFR therapy, the therapeutic targeting of HER3 in combination with primary drivers of the tumor is also being investigated. Here, we review the role of HER3 as a promoter of clinical resistance to EGFR therapy and discuss therapeutic approaches that could potentially overcome this resistance.

Mechanisms of resistance to EGFR tyrosine kinase inhibitors

Since the discovery that non-small cell lung cancer (NSCLC) is driven by epidermal growth factor receptor (EGFR) mutations, the EGFR tyrosine kinase inhibitors (EGFR-TKIs, e.g., gefitinib and elrotinib) have been effectively used for clinical treatment. However, patients eventually develop drug resistance. Resistance to EGFR-TKIs is inevitable due to various mechanisms, such as the secondary mutation (T790M), activation of alternative pathways (c-Met, HGF, AXL), aberrance of the downstream pathways (K-RAS mutations, loss of PTEN), impairment of the EGFR-TKIs-mediated apoptosis pathway (BCL2-like 11/BIM deletion polymorphism), histologic transformation, ATP binding cassette (ABC) transporter effusion, etc. Here we review and summarize the known resistant mechanisms to EGFR-TKIs and provide potential targets for development of new therapeutic strategies.