Artificial Intelligence-Assisted Amphiregulin and Epiregulin IHC Predicts Panitumumab Benefit in RAS Wild-Type Metastatic Colorectal Cancer

Prognostic and predictive biomarkers for anti-EGFR monoclonal antibody therapy in RAS wild-type metastatic colorectal cancer: a systematic review and meta-analysis

BACKGROUND

RAS mutations affect prognosis in patients with metastatic colorectal cancer (mCRC) and have been identified as strong negative predictive markers for anti-epidermal growth factor receptor monoclonal antibody (anti-EGFR mAb) therapy, but many tumors containing wild-type RAS genes still do not respond to these therapies. Some additional biomarkers may have prognostic or predictive roles, but conclusions remain controversial.

METHODS

We performed a meta-analysis and systematic review of randomized controlled trials comparing anti-EGFR mAb therapy with alternative therapy that investigated the prognostic and predictive impact of additional biomarkers in RAS wild-type (wt) mCRC patients. Hazard ratios (HRs) and 95% confidence intervals (CIs) for progression-free survival (PFS) and overall survival (OS) and odds ratios (ORs) for objective response rate (ORR) were calculated. The prognostic value of biomarkers was investigated by separately pooling HR and OR for different treatment groups in an individual study. The predictive value was assessed by pooling study interactions between treatment effects and biomarker subgroups.

RESULTS

Thirty publications reporting on eighteen trials were selected, including a total of 13,507 patients. In prognostic analysis, BRAF mutations were associated with poorer PFS [HRs = 3.76 (2.47-5.73) and 2.69 (1.82-3.98)] and OS [HRs = 2.66 (1.95-3.65) and 2.45 (1.55-3.88)] in both the experimental and control arms; low miR-31-3p expression appeared to have longer PFS and OS. In terms of predictive effect, a lack of response to anti-EGFR therapy was observed in patients with BRAF mutant tumors (Pinteraction < 0.01 for PFS). Patients with tumors with any mutation in the KRAS/NRAS/BRAF/PIK3CA gene also showed similar results compared with all wild-type tumors (Pinteraction for PFS, OS, and ORR were < 0.01, < 0.01 and 0.01, respectively). While low miR-31-3p expression could predict PFS (Pinteraction = 0.01) and OS (Pinteraction = 0.04) benefit. The prognostic and predictive value regarding PIK3CA mutations, PTEN mutations or deletions, EGFR, EREG/AREG, HER2, HER3, and HER4 expression remains uncertain.

CONCLUSIONS

In RAS wt mCRC patients receiving EGFR-targeted therapy, BRAF mutation is a powerful prognostic and therapy-predictive biomarker, with no effect found for PIK3CA mutation, PTEN mutation or deletion, but the combined biomarker KRAS/NRAS/BRAF/PIK3CA mutations predict resistance to anti-EGFR therapy. Low miR-31-3p expression may have positive prognostic and therapy predictive effects. Evidence on the prognostic and predictive roles of EGFR and its ligands, and HER2/3/4 is insufficient.

Associations between AI-Assisted Tumor Amphiregulin and Epiregulin IHC and Outcomes from Anti-EGFR Therapy in the Routine Management of Metastatic Colorectal Cancer

PURPOSE

High tumor production of the EGFR ligands, amphiregulin (AREG) and epiregulin (EREG), predicted benefit from anti-EGFR therapy for metastatic colorectal cancer (mCRC) in a retrospective analysis of clinical trial data. Here, AREG/EREG IHC was analyzed in a cohort of patients who received anti-EGFR therapy as part of routine care, including key clinical contexts not investigated in the previous analysis.

EXPERIMENTAL DESIGN

Patients who received panitumumab or cetuximab ± chemotherapy for treatment of RAS wild-type mCRC at eight UK cancer centers were eligible. Archival formalin-fixed paraffin-embedded tumor tissue was analyzed for AREG and EREG IHC in six regional laboratories using previously developed artificial intelligence technologies. Primary endpoints were progression-free survival (PFS) and overall survival (OS).

RESULTS

A total of 494 of 541 patients (91.3%) had adequate tissue for analysis. A total of 45 were excluded after central extended RAS testing, leaving 449 patients in the primary analysis population. After adjustment for additional prognostic factors, high AREG/EREG expression (n = 360; 80.2%) was associated with significantly prolonged PFS [median: 8.5 vs. 4.4 months; HR, 0.73; 95% confidence interval (CI), 0.56-0.95; P = 0.02] and OS [median: 16.4 vs. 8.9 months; HR, 0.66 95% CI, 0.50-0.86; P = 0.002]. The significant OS benefit was maintained among patients with right primary tumor location (PTL), those receiving cetuximab or panitumumab, those with an oxaliplatin- or irinotecan-based chemotherapy backbone, and those with tumor tissue obtained by biopsy or surgical resection.

CONCLUSIONS

High tumor AREG/EREG expression was associated with superior survival outcomes from anti-EGFR therapy in mCRC, including in right PTL disease. AREG/EREG IHC assessment could aid therapeutic decisions in routine practice. See related commentary by Randon and Pietrantonio, p. 4021.

Targeting epiregulin in the treatment-damaged tumor microenvironment restrains therapeutic resistance

The tumor microenvironment (TME) represents a milieu enabling cancer cells to develop malignant properties, while concerted interactions between cancer and stromal cells frequently shape an "activated/reprogramed" niche to accelerate pathological progression. Here we report that a soluble factor epiregulin (EREG) is produced by senescent stromal cells, which non-cell-autonomously develop the senescence-associated secretory phenotype (SASP) upon DNA damage. Genotoxicity triggers EREG expression by engaging NF-κB and C/EBP, a process supported by elevated chromatin accessibility and increased histone acetylation. Stromal EREG reprograms the expression profile of recipient neoplastic cells in a paracrine manner, causing upregulation of MARCHF4, a membrane-bound E3 ubiquitin ligase involved in malignant progression, specifically drug resistance. A combinational strategy that empowers EREG-specific targeting in treatment-damaged TME significantly promotes cancer therapeutic efficacy in preclinical trials, achieving response indices superior to those of solely targeting cancer cells. In clinical oncology, EREG is expressed in tumor stroma and handily measurable in circulating blood of cancer patients post-chemotherapy. This study establishes EREG as both a targetable SASP factor and a new noninvasive biomarker of treatment-damaged TME, thus disclosing its substantial value in translational medicine.

Use of Personalized Biomarkers in Metastatic Colorectal Cancer and the Impact of AI

Colorectal cancer is a major cause of cancer-related death worldwide and is correlated with genetic and epigenetic alterations in the colonic epithelium. Genetic changes play a major role in the pathophysiology of colorectal cancer through the development of gene mutations, but recent research has shown an important role for epigenetic alterations. In this review, we try to describe the current knowledge about epigenetic alterations, including DNA methylation and histone modifications, as well as the role of non-coding RNAs as epigenetic regulators and the prognostic and predictive biomarkers in metastatic colorectal disease that can allow increases in the effectiveness of treatments. Additionally, the intestinal microbiota’s composition can be an important biomarker for the response to strategies based on the immunotherapy of CRC. The identification of biomarkers in mCRC can be enhanced by developing artificial intelligence programs. We present the actual models that implement AI technology as a bridge connecting ncRNAs with tumors and conducted some experiments to improve the quality of the model used as well as the speed of the model that provides answers to users. In order to carry out this task, we implemented six algorithms: the naive Bayes classifier, the random forest classifier, the decision tree classifier, gradient boosted trees, logistic regression and SVM.

Leveraging Comprehensive Cancer Registry Data to Enable a Broad Range of Research, Audit and Patient Support Activities

Simple Summary

Registry data has the potential to support a broad range of research, audit and education initiatives. Here, we describe the experience and learnings of a series of large multi-institutional cancer registries that leverage real-world clinical data for a range of purposes, that informs the conduct and output of each registry in a virtuous cycle. Lessons learnt include the need for careful and continuous curation of information being collected, regular database updates, and the need for a continued focus on data quality. As a standalone resource, each registry has supported numerous projects, but linkage with external datasets with patients in common has enhanced the research potential. Multiple projects have linked registry data with matched tissue specimens to support the discovery and valiation of prognostic and predictive markers in the tumour and blood specimens. Registry-based biomarker trials have been successfully supported, generating novel and practice-changing data. Registry-based clinical trials, particularly studies exploring the best use of drug options are now complementing the research conducted in traditional clinical trials. More recent projects supported by the registries include health economic studies, personalised patient education material, and increased consumer engagement, including consumer entered data.

Abstract

Traditional cancer registries have often been siloed efforts, established by single groups with limited objectives. There is the potential for registry data to support a broad range of research, audit and education initiatives. Here, we describe the establishment of a series of comprehensive cancer registries across the spectrum of common solid cancers. The experience and learnings of each registry team as they develop, implement and then use collected data for a range of purposes, that informs the conduct and output of other registries in a virtuous cycle. Each registry is multi-site, multi-disciplinary and aims to collect data of maximal interest and value to a broad range of enquiry, which would be accessible to any researcher with a high-quality proposal. Lessons learnt include the need for careful and continuous curation of data fields, with regular database updates, and the need for a continued focus on data quality. The registry data as a standalone resource has supported numerous projects, but linkage with external datasets with patients in common has enhanced the audit and research potential. Multiple projects have linked registry data with matched tissue specimens to support prognostic and predictive biomarker studies, both validation and discovery. Registry-based biomarker trials have been successfully supported, generating novel and practice-changing data. Registry-based clinical trials, particularly randomised studies exploring the optimal use of available therapy options are now complementing the research conducted in traditional clinical trials. More recent projects supported by the registries include health economic studies, personalised patient education material, and increased consumer engagement, including consumer entered data.

Comparison of manual and automated digital image analysis systems for quantification of cellular protein expression

OBJECTIVE

Quantifying protein expression in immunohistochemically stained histological slides is an important tool for oncologic research. The use of computer-aided evaluation of IHC-stained slides significantly contributes to objectify measurements. Manual digital image analysis (mDIA) requires a user-dependent annotation of the region of interest (ROI). Others have built-in machine learning algorithms with automated digital image analysis (aDIA) and can detect the ROIs automatically. We aimed to investigate the agreement between the results obtained by aDIA and those derived from mDIA systems.

METHODS

We quantified chromogenic intensity (CI) and calculated the positive index (PI) in cohorts of tissue microarrays (TMA) using mDIA and aDIA. To consider the different distributions of staining within cellular sub-compartments and different tumor architecture our study encompassed nuclear and cytoplasmatic stainings in adenocarcinomas and squamous cell carcinomas.

RESULTS

Within all cohorts, we were able to show a high correlation between mDIA and aDIA for the CI (p<0.001) along with high agreement for the PI. Moreover, we were able to show that the cell detections of the programs were comparable as well and both proved to be reliable when compared to manual counting.

CONCLUSION

mDIA and aDIA show a high correlation in acquired IHC data. Both proved to be suitable to stratify patients for evaluation with clinical data. As both produce the same level of information, aDIA might be preferable as it is time-saving, can easily be reproduced, and enables regular and efficient output in large studies in a reasonable time period.

The Role of EREG/EGFR Pathway in Tumor Progression

Aberrant activation of the epidermal growth factor receptor (EGFR/ERBB1) by erythroblastic leukemia viral oncogene homolog (ERBB) ligands contributes to various tumor malignancies, including lung cancer and colorectal cancer (CRC). Epiregulin (EREG) is one of the EGFR ligands and is low expressed in most normal tissues. Elevated EREG in various cancers mainly activates EGFR signaling pathways and promotes cancer progression. Notably, a higher EREG expression level in CRC with wild-type Kirsten rat sarcoma viral oncogene homolog (KRAS) is related to better efficacy of therapeutic treatment. By contrast, the resistance of anti-EGFR therapy in CRC was driven by low EREG expression, aberrant genetic mutation and signal pathway alterations. Additionally, EREG overexpression in non-small cell lung cancer (NSCLC) is anticipated to be a therapeutic target for EGFR-tyrosine kinase inhibitor (EGFR-TKI). However, recent findings indicate that EREG derived from macrophages promotes NSCLC cell resistance to EGFR-TKI treatment. The emerging events of EREG-mediated tumor promotion signals are generated by autocrine and paracrine loops that arise from tumor epithelial cells, fibroblasts, and macrophages in the tumor microenvironment (TME). The TME is a crucial element for the development of various cancer types and drug resistance. The regulation of EREG/EGFR pathways depends on distinct oncogenic driver mutations and cell contexts that allows specific pharmacological targeting alone or combinational treatment for tailored therapy. Novel strategies targeting EREG/EGFR, tumor-associated macrophages, and alternative activation oncoproteins are under development or undergoing clinical trials. In this review, we summarize the clinical outcomes of EREG expression and the interaction of this ligand in the TME. The EREG/EGFR pathway may be a potential target and may be combined with other driver mutation targets to combat specific cancers.

p130Cas Is Correlated with EREG Expression and a Prognostic Factor Depending on Colorectal Cancer Stage and Localization Reducing FOLFIRI Efficacy

p130 Crk-associated substrate (p130Cas) is associated with poor prognosis and treatment resistance in breast and lung cancers. To elucidate p130Cas functional and clinical role in colorectal cancer (CRC) progression/therapy resistance, we performed cell culture experiments and bioinformatic/statistical analyses of clinical data sets. p130Cas expression was associated with poor survival in the cancer genome atlas (TCGA) data set. Knockdown/reconstitution experiments showed that p130Cas drives migration but, unexpectedly, inhibits proliferation in CRC cells. TCGA data analyses identified the growth factor epiregulin (EREG) as inversely correlated with p130Cas. p130Cas knockdown and simultaneous EREG treatment further enhanced proliferation. RNA interference and EREG treatment experiments suggested that p130Cas/EREG limit each other’s expression/activity. Inverse p130Cas/EREG Spearman correlations were prominent in right-sided and earlier stage CRC. p130Cas was inducible by 5-fluorouracil (5-FU) and FOLFIRI (folinic acid, 5-FU, irinotecan), and p130Cas and EREG were upregulated in distant metastases (GSE121418). Positive p130Cas/EREG correlations were observed in metastases, preferentially in post-treatment samples (especially pulmonary metastases). p130Cas knockdown sensitized CRC cells to FOLFIRI independent of EREG treatment. RNA sequencing and gene ontology analyses revealed that p130Cas is involved in cytochrome P450 drug metabolism and epithelial-mesenchymal transition. p130Cas expression was associated with poor survival in right-sided, stage I/II, MSS (microsatellite stable), or BRAF-mutated CRC. In summary, p130Cas represents a prognostic factor and potential therapeutic target in CRC.