Related Experiment Video
Updated: Aug 16, 2026
06:46
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
CAVE-Onc: Graph-constrained agentic validation for cross-domain contradictions in CDISC oncology submissions
1Harrisburg University of Science and Technology, Harrisburg, Pennsylvania, United States of America.
Plos One
|August 14, 2026
Summary
CAVE-Onc enhances clinical trial data validation by detecting cross-domain contradictions missed by standard tools. This graph-based approach improves data integrity for oncology studies using RECIST criteria.
Area of Science:
- Clinical Data Management
- Bioinformatics
- Oncology Clinical Trials
Background:
- Current CDISC Rules Engine (CORE) validation for Study Data Tabulation Model (SDTM) submissions has limitations in detecting cross-domain contradictions.
- Imperative, domain-scoped architectures cannot identify inconsistencies, such as between RECIST 1.1 overall response and lesion status, leading to undetected errors.
- This gap impacts the reliability of oncology clinical trial data, particularly concerning response evaluation criteria.
Purpose of the Study:
- To introduce CAVE-Onc, a novel two-layer graph-constrained agentic validation engine.
- To augment existing CORE validation with declarative SHACL and LangGraph-based agents for enhanced data integrity.
- To address the limitations of current tools in detecting complex, cross-domain contradictions in SDTM data.
Main Methods:
- Developed CAVE-Onc, operating on an RDF knowledge graph derived from XPT datasets.
- Implemented Layer 1 with 111 SHACL shapes, including CORE rules, RECIST 1.1 derivations, and cross-domain constraints.
- Utilized a LangGraph-based agent (Layer 3) for specific tasks like RECIST Table 7 verification.
Main Results:
- CAVE-Onc's Layer 1 and CORE produced disjoint validation flag sets on clean data (Jaccard index = 0.004).
- CAVE-Onc successfully detected all 20 injected contradiction archetypes (100%), with 19 via SHACL-SPARQL and 1 via the L3 agent.
- Compared to industry validators (CDISC CORE, Pinnacle 21 FDA) which detected 0/10 cross-domain RECIST contradictions, CAVE-Onc detected 10/10 (McNemar p=0.002).
Conclusions:
- Graph-based validation, as implemented in CAVE-Onc, significantly augments industry-standard tools like CORE.
- CAVE-Onc demonstrates superior expressiveness in detecting cross-domain contradictions, particularly for RECIST criteria in oncology trials.
- The proposed validation engine improves clinical trial data quality and reliability by addressing limitations in current validation approaches.
Related Concept Videos
01:21Cancer Survival Analysis
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
01:05Cancer-Critical Genes II: Tumor Suppressor Genes
Genes usually encode proteins necessary for the proper functioning of a healthy cell. Mutations can often cause changes to the gene expression pattern, thereby altering the phenotype.
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
01:05Cancer-Critical Genes II: Tumor Suppressor Genes
Genes usually encode proteins necessary for the proper functioning of a healthy cell. Mutations can often cause changes to the gene expression pattern, thereby altering the phenotype.
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
02:18Cancer
Cancers arise due to mutations in genes involved in the regulation of cell division, which leads to unrestricted cell proliferation. Modern science and medicine have made great strides in the understanding and treatment of cancer, including eradicating cancer in some patients. However, there is still no cure for cancer. This is largely due to the fact that cancer is a large group of many diseases.
01:33Cancer-Critical Genes I: Proto-oncogenes
Genes usually encode proteins necessary for the proper functioning of a healthy cell. Mutations can often cause changes to the gene expression pattern, thereby altering the phenotype.
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
When the function of certain critical genes, especially those involved in cell cycle regulation and cell growth signaling cascades, gets disrupted, it upsets the cell cycle progression. Such cells with unchecked cell cycles start proliferating uncontrollably and eventually develop into tumors.
Such genes that act...
01:25Formulating and Validating Nursing Diagnosis II
Nursing diagnoses represent a problem validated by major defining characteristics. There are four categories of nursing diagnoses: problem-focused, risk, health promotion or wellness, and syndrome. The anatomy of a nursing diagnosis includes three components: problem statement or diagnostic label, defining characteristics, and related factors.
Risk nursing diagnoses represent clinical judgments of an individual, family, or community more vulnerable to developing the health problem than others...
Risk nursing diagnoses represent clinical judgments of an individual, family, or community more vulnerable to developing the health problem than others...