Original Research Article - (2026) Volume 15, Issue 5
Received: 06-Oct-2026, Manuscript No. jcct-26-194849;
Editor assigned: 08-Oct-2026, Pre QC No. P-194849;
Reviewed: 12-Oct-2026, QC No. Q-194849;
Revised: 19-Oct-2026, Manuscript No. R-194849;
Published:
26-Oct-2026
Citation: Jacob Gabayan, Yousuf Altameemi and Ali
Muhammad MD, MBA. âLeveraging Artificial Intelligence to Structure Nextgeneration
Sequencing Data for Precision Oncology Research: A Real-world
Pancreatic Ductal Adenocarcinoma Liquid Biopsy Experienceâ J Cancer Clin
Trials 15 (2026): 388.
Copyright: © 2026 Muhammad A. This is an open-access article distributed
under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
Precision oncology generates increasingly complex molecular datasets that can be difficult to organize and analyze efficiently. We evaluated 212
real-world pancreatic ductal adenocarcinoma liquid biopsy datasets and explored AI-assisted workflows to transform unstructured next-generation
sequencing outputs into structured research datasets. The analysis characterized 99 unique biomarkers and genes and demonstrates the potential
operational value of AI-enabled data structuring for research, biomarker assessment and future clinical trial matching.
Pancreatic ductal adenocarcinoma • Artificial Intelligence • Next-generation Sequencing • Liquid Biopsy • Comprehensive Genomic Profiling • Precision Oncology • Biomarkers • Clinical Research • Real-world Data • Beverly Hills Cancer Center • Clinical Research Advisors LLC
NGS testing has transformed oncology; however, manual extraction of molecular findings remains labor intensive. This project sought to characterize the PDAC genomic landscape while examining how AI-assisted tools can support organization, standardization and analysis of high-volume genomic information.
Retrospective review of PDAC records identified within the Beverly Hills Cancer Center and its Clinical Research Advisors satellite sites’ liquid biopsy dataset. Records containing pancreatic cancer terminology were reviewed and aggregated. AI-assisted workflows were conceptually utilized to standardize biomarker data, support dataset structuring and facilitate downstream analysis. The analyzed database contained 212 PDAC-related genomic data sets (Table 1 and Table 2).
| Metric | Value |
|---|---|
| PDAC-related datasets | 212 |
| Unique biomarkers/genes | 99 |
Table 1: Dataset Summary.
| Biomarker/Gene | Count |
|---|---|
| DNMT3A | 170 |
| TP53 | 140 |
| MSI-High | 120 |
| Tumor Mutational Burden (TMB) | 90 |
| TET2 | 90 |
| KRAS | 80 |
| Tumor Fraction | 80 |
| ATM | 70 |
| PPM1D | 40 |
| UGT1A1 | 40 |
Table 2: Top Biomarkers/Genes.
AI-assisted extraction and standardization outputs were reviewed against the source genomic reports before analysis. Biomarker names, specimen-level observations, duplicate records and derived counts were checked using a predefined quality-control process, with discrepancies resolved through manual review [1-4].
The molecular landscape was dominated by canonical driver alterations including KRAS, TP53, SMAD4 and CDKN2A. AI-assisted organization enabled rapid aggregation of molecular observations into searchable datasets and visual analytics. Figure 1 presents the most frequently observed biomarkers/ genes in the PDAC subset (Figure 1) [5-6].
AI-assisted approaches reduced dependence on manual abstraction, improved consistency of nomenclature, enabled scalable data review and provide a framework for future clinical trial selection on institutional level and potential clinical trial matching decisions. Rather than replacing physician judgment, AI functioned as a force multiplier supporting efficient data organization and analysis [7-9].
This retrospective analysis was derived from a single network dataset and is subject to limitations associated with accession-level analyses, free-text diagnoses and heterogeneous testing patterns. Clinical outcomes, survival endpoints and treatment response data were not evaluated at this point although with the encouraging results of this project we are planning future analysis of clinical outcomes, survival endpoints and treatment response rates related to different therapeutic indications and treatment approaches.
The future of precision oncology depends not only on improved diagnostics but also on improved information management. This report suggests that AIassisted data structuring can complement traditional oncology research by enabling efficient handling of increasingly complex genomic datasets.
As a private, academic, community-based cancer center, Beverly Hills Cancer Center provides state-of-the-art, multidisciplinary cancer care while supporting leading clinical trials and oncology research. By integrating specialized treatment services with clinical research operations, Beverly Hills Cancer Center offers patients access to innovative investigational therapies and advances the development of new approaches to cancer treatment.
Clinical research is conducted through Beverly Hills Cancer Center, the main Beverly Hills research location of Clinical Research Advisors LLC and five affiliated oncology satellite sites located in West Hollywood, Encino, Korea Town, West Covina and Downtown Los Angeles. This integrated research network is supported by multidisciplinary investigators and experienced clinical research professionals who coordinate patient identification, informed consent, protocol-directed assessments, treatment and follow-up across the greater Los Angeles area.
The authors declare no potential conflicts of interest related to this work.
Google Scholar, Cross Ref, Indexed at
Google Scholar, Cross Ref, Indexed at
Google Scholar, Cross Ref, Indexed at
Google Scholar, Cross Ref, Indexed at
Google Scholar, Cross Ref, Indexed at
Google Scholar, Cross Ref, Indexed at
Journal of Cancer Clinical Trials received 95 citations as per Google Scholar report
