Gastrointestinal (GI) clinical research has evolved over time. In this article, we explore the latest trends and innovations, including the use of big data and artificial intelligence (AI), which have the potential to revolutionize the design of GI clinical research. To address prevailing clinical trial challenges, understanding barriers from a patient perspective helps inform the introduction of innovations in trial design, e.g., virtual solutions or a multi-stakeholder approach, for more efficient, patient-centric clinical trials. We highlight HER2 inhibitors and FGFR2 genomic alterations as potential biomarkers.
Big Data from Multi-omic, Multi-site Studies Enhance GI Clinical Research
Multi-omics approaches permit the generation and integration of multiple data types on a large scale, offering a comprehensive picture of biological and disease processes. Gastroenterology and hepatobiliary research benefit from this type of analysis due to the enhanced complete investigation into GI and hepatobiliary function, leading to the discovery of connections between these systems and neural, immune, and endocrine cells and gut microbial organisms. Alizadeh et al. (2023) examine how human genetics and genomics, epigenomics, transcriptomics, proteomics, microbiome, and metabolomics have been applied to
- advance biomarker discovery
- develop precision medicine
- generate novel ways to detect GI cancers
- characterize inflammatory bowel disease (IBD).
They also explore how imaging data can assist diagnosis and outcome predictions.
Gastric Cancer Biomarkers
Immunotherapy is a vital treatment option for gastric cancer (GC). Clinical studies are still necessary to delineate molecular pathways and identify new influential factors and biomarkers. For advanced GC, targeted therapy is an important treatment option. Gong et al. (2023) explored the relationship between flavin-containing monooxygenases (FMOs) and peritoneal metastasis (PM) and found a correlation between FMOs and GC. FMOs, therefore, serve as novel markers and potential therapeutic targets for PM in GC.
Big Data and the Shifting Role of Clinical Trials
Clinical studies using pre-existing big data can analyze easily available datasets, rely more on tech than scarce human resources, and factor in comorbidities and medical diversity that affect patient outcomes. In addition, Alizadeh et al. (2023) describe other strengths, including
- the volume and range of data which support the study of uncommon events over long periods of time
- the capacity to simultaneously investigate multiple variables and conduct sensitivity analyses for a thorough analysis of the secondary endpoints and sub-groups that ordinarily may be ignored in randomized clinical trials (RCTs).
Machine Learning
The availability of big data has resulted in the need for AI models such as machine learning (ML) to analyze the immense amount of clinical data. ML has shown significant promise in disease identification and classification by leveraging data.
Kim et al. (2023) studied the application of ML on structured medical data in gastroenterology and found clinical studies on
- the general subject of gastroenterology, including studies focused on the prediction of outcomes in acute diverticulitis, diagnosis of helicobacter pylori, and prediction of patient risk of upper GI lesions;
- GI hemorrhage, including studies on the risk of hospital-based intervention or death in patients with upper GI hemorrhage and prediction of adverse events;
- GC including studies predicting GC risk in patients after Helicobacter pylori eradication, predicting recurrence of GC after operation, and predicting GC metastasis before surgery;
- GI tumors and cancers, including studies predicting risk stratification for GI stromal tumors.
Large Language Models
Large language models have piqued the interest of GI researchers due to their potential to enhance clinical research; however, Kim et al. (2023) caution that they must be handled with care to avoid unintended harm. Kim et al. anticipate that large language models that can perform ML analyses will play an increasingly vital role in GI clinical research.
Human Epidermal Growth Factor Receptor 2 (HER2) Inhibitors
The development of effective HER2 inhibitors, agents that inhibit the HER2 receptor, has improved survival and prognosis in cancer patients, expanding from breast cancer to include GC. In addition, the HER2 protein may be highly expressed in esophageal and gastroesophageal junction (GEJ) cancers. Radford et al. (2023) report that HER2 inhibitor trastuzumab achieved impressive outcomes in patients with HER2+ advanced GC and GEJ adenocarcinoma, and a recent meta-analysis by Wang et al. (2023) concluded that HER2-targeted therapy is a promising treatment for colorectal cancer (CRC). In January 2023, the FDA granted accelerated approval to the combination of targeted drugs trastuzumab and tucatinib for people with advanced CRC that overexpress HER2.
FGFR2 (Fibroblast growth factor receptor 2)
Dysregulation of FGFR2 drives several cancers, and as such, FGFR2 is a promising drug target. FGFR2 genomic alterations are considered potential biomarkers of the therapeutic response of FGFR2 inhibitors. Genomic alterations in FGFR2 include amplification of FGFR2 in up to 10% of gastric cancers and FGFR2 mutations in less than 5% of gastric cancer.
Genetics, Trial Design, and Personalized Medicine
Researchers have identified molecular biomarkers with prognostic and predictive implications to understand GC heterogeneity better and develop targeted therapies. The rise of precision medicine targeting biomarkers has led to new trial designs that depart from the sequential phases I, II, and III of conventional trial designs. Master protocols can, e.g., evaluate histology-specific therapies targeting a common oncogenic mutation or screen for multiple biomarkers. Clinical trials have shifted from tumor-type-centered to gene-directed and histology-agnostic trials, with adaptive designs and personalized treatment strategies tailored to biomarker profiles. According to Noor & Raine (2023), adopting trial designs to improve efficiency, such as master protocol trials, has increasingly been advocated for IBD.
Patient Recruitment & AI
Patient recruitment is impacted by the variability and subjectivity of endoscopic disease scoring by referring physicians. The benefits of AI and computer vision-driven algorithms, objectivity and speed, occur by improving the enrollment process and building more precise and quantifiable clinical endpoints to evaluate drug candidates. Digital solutions automate the interpretation of colonoscopy images and videos, allowing for standardized endoscopic disease scores for each patient.
Patient recruitment is also affected by low awareness of trials as an option and exacerbated by the growing number and complexity of clinical trials. Trial search tools have gaps such as complex navigation, limited search functionality, overly complex data for non-specialists, and unclear next steps. Jordan et al. (2021) developed a novel AI-based trial search tool to restructure trial information, match patients to trials, and ultimately increase access to and understanding of GI clinical trials. They found that democratizing trial information increases trial accrual and patient satisfaction while reducing disparities.
Virtual Health Solutions & Patient-centric Trials
The efficiency of GI clinical trials can be improved by using virtual health solutions to enhance patient experience. For example, clinical trial protocols for IBD often require face-to-face visits and monitoring, hospital-based medication administration, and paper-based forms and questionnaires. Noor & Siegel (2023) studied the application of virtual innovations in IBD clinical trials, including digital invitations using large patient registries, remote consent and recruitment (eConsent), virtual visits, remote patient monitoring and data collection (EDC), remote medication delivery and administration, remote clinical trial monitoring, and routinely collected health data for long-term follow-up. In addition to improving data quality, adopting such virtual tech and ePRO (electronic patient-reported outcome) drives patient centricity, impacts patient satisfaction and adherence, and secures long-term committed patients.
Multi-stakeholder Approach
As GI clinical trials often place a substantial burden on patients, researchers advise including patient feedback from early on during the study design process. Input from a diverse group of patients and multiple diverse sites facilitates protocol design, operational risk mitigation, and the execution of efficient clinical trials. To improve the development of drugs for children and adolescents with IBD, Croft et al. (2023) propose the collaborative action of an international multi-stakeholder group to coordinate efforts to accelerate access to new drugs, explore the use of extrapolation from adult data and prioritize different classes of investigational drugs.
Summary
GI clinical research has evolved over time. We explore the latest trends and innovations, including the incorporation of new techniques like big data and AI, innovative trial designs like virtual and patient-centric trials, and studies at the genomic level.
Vial
Vial is a next-generation, technology-first contract research organization (CRO) building towards a more efficient future for clinical trials. By deploying technology at every step, we are driving efficiencies in speed and cost savings for innovative biotech companies of all sizes. The Vial Gastroenterology CRO delivers faster, better, and cheaper gastroenterology trial results for biotech sponsors. Our mission is to empower scientists to discover groundbreaking scientific therapeutics that help people live happier, healthier lives. Learn more about our work in GI clinical trials.
