Reframing pancreatic cancer diagnosis: metabolic signatures and the challenge of rule-out testing

Pancreatic ductal adenocarcinoma (PDAC) incidence continues to rise globally and is expected to approach 880,000 new cases by 2044, representing an approximately four-fold increase since 1990 (1). In the United States, PDAC is the seventh and ninth most common malignancy in females and males (2), respectively, and is projected to become the leading cause of cancer-related deaths by 2040 (3). Unfortunately, PDAC incidence is rising at an alarming rate in younger populations (4) drawing stark attention to the necessity for improvements in early detection and intervention. Early detection remains uniquely challenging. Most patients present with advanced disease because the pancreas is anatomically difficult to access, pancreatic intraepithelial neoplasias are undetectable without invasive techniques, hallmark symptoms such as unintentional weight loss or abdominal pain typically manifest late in the disease course, and imaging findings are frequently confounded by benign conditions (5-8).

At the same time, nearly one in five individuals have an incidental pancreatic cystic lesion detected on cross-sectional imaging with approximately 80% of these being cystic precursors to PDAC called intraductal papillary mucinous neoplasms (IPMNs) (9,10). While most IPMNs remain indolent, a subset harbor high-grade dysplasia (HGD) or small invasive foci requiring surgical resection. Current radiologic and clinical criteria, including the Kyoto guidelines (11), offer risk stratification for indeterminate pancreatic cysts, but they remain imperfect and early-stage invasive PDACs may go undetected (12) delaying diagnosis until it reaches more advanced-stage disease. While endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) procedures can be used to obtain tissue samples for evaluation, they may miss small foci of invasive PDAC due to sampling error (13,14) and run the risk of cancer cell leakage (15), delaying diagnosis until symptomatic progression. On the other hand, these guidelines can lead to overtreatment of more indolent lesions (12).

Serum CA 19-9, the most widely used biomarker in PDAC, illustrates the diagnostic tension: limited sensitivity in early-stage disease, false elevations in benign pancreatobiliary conditions, and non-expression in Lewis antigen-negative individuals (16). These limitations create an unmet need for minimally invasive adjunctive biomarkers to refine triage in patients with indeterminate pancreatic lesions.

In a recent issue of The Lancet Gastroenterology and Hepatology, Mahajan and colleagues (17) sought to address this unmet need by performing validation of two plasma multi-metabolite signatures in the prospective, multicenter, phase 4 METAPAC study that enrolled 1,370 individuals with a computed tomography (CT)-detected undefined pancreatic lesion. Of 1,129 participants with complete targeted metabolome and CA 19-9 levels, final diagnoses included 489 PDAC, 232 IPMN, 271 non-IPMN cystic lesions, 113 chronic pancreatitis, 11 acute pancreatitis, and 13 extrapancreatic metastases, consistent with the cohort enrichment design.

The authors (17) found that both the improved (i-Metabolic) and minimalistic metabolic (m-Metabolic) signatures performed better than CA 19-9 alone in ruling out PDAC across disease stages, achieving specificities of 90.4% and 93.6%, respectively. Most notably, these findings were upheld even when restricting the PDAC cases to those with resectable tumors (i.e., stages IA-IIB), yielding specificities of 91.7% and 93.6% for the i-Metabolic and m-Metabolic signatures, respectively. Continued performance in early-stage tumors is particularly relevant, as this represents the narrow window during which intervention can meaningfully alter prognosis. The ability of the i-Metabolic and m-Metabolic signatures to maintain high specificity in resectable disease suggests potential utility in distinguishing early invasive PDAC from benign or pre-malignant conditions—an area of persistent diagnostic uncertainty. This potential clinical use is especially pertinent for individuals with imaging-confirmed IPMNs under active surveillance, where the central challenge lies in differentiating indolent cysts from those harboring HGD or early invasion. An adjunctive blood-based metric that refines this distinction could shorten the time to cancer diagnosis and reduce unnecessary surgical resections.

Importantly, these signatures appear to mitigate some of the diagnostic limitations of CA 19-9. While CA 19-9 is frequently elevated in advanced disease, it demonstrates variable sensitivity in early-stage PDAC, with a subset of resectable tumors presenting with normal levels (18). In this context, metabolomic profiling may provide complementary biological information. However, reliance on metabolomic signatures alone could also introduce false reassurance, underscoring the importance of integrating with—rather than substituting for—CA 19-9 within a multimodal framework. A sequential approach in which CA 19-9 serves as an initial screening tool followed by metabolomic refinement in equivocal cases may represent a pragmatic translational pathway.

Despite the promise of the i- and m-Metabolic signatures, interpretation of these findings requires careful clinical framing. The authors emphasize high specificity and negative predictive value (NPV) as strengths of a “rule-out” tool. Yet, acceptable thresholds for clinical adoption may differ among treating physicians. Some surgical oncologists may prioritize maximizing sensitivity over specificity accepting additional downstream testing to avoid missing an early cancer. Notably, Tab. 3 demonstrates variability in NPV across analytic subsets, underscoring that predictive values are inherently dependent on prevalence.

Selection of metabolomic cutoffs represents an important consideration in optimizing clinical application. As illustrated in Fig. 2, some degree of overlap in metabolite distributions between PDAC and non-malignant conditions is expected and reflects a well-recognized challenge in biomarker development. Future work could explore how alternative threshold strategies—such as prioritizing higher sensitivity versus higher specificity—might influence clinical utility depending on the intended use, whether for reassurance in surveillance settings or for early detection in higher-risk populations. Incorporating perspectives from treating physicians regarding acceptable tradeoffs between false-negative and false-positive rates may further inform how these signatures are best positioned within real-world diagnostic pathways.

The metabolite classes included in the signatures also offer an opportunity for broader contextualization. For example, lysophospholipids have been investigated as candidate biomarkers in breast cancer and other malignancies but have not advanced to routine clinical use (19,20). The current study’s integration of metabolomic markers with CA 19-9 represents a thoughtful advancement toward multimodal risk modeling. As with many promising “omics” approaches, continued validation across independent cohorts, reproducibility across analytic platforms, and demonstration of incremental clinical value beyond existing standards will be important steps in translating these findings into practice (21). The strength of the METAPAC framework provides a solid foundation upon which such future efforts can build.

The systemic metabolic perturbations captured by these signatures also raise intriguing biological questions. Although weight loss was included as a binary variable in Tab. 2, metabolic reprogramming in PDAC is closely intertwined with cachexia and broader tumor-host interactions (22). Future analyses incorporating continuous measures of weight loss, longitudinal weight trajectories, and/or body composition metrics may help clarify whether these signatures primarily capture tumor-specific metabolic alterations or downstream systemic effects associated with cancer-induced wasting. Distinguishing between these mechanisms could be particularly informative for understanding performance in earlier, pre-cachectic disease—an area of substantial unmet clinical need (23).

Further analytical refinement may represent a valuable extension of this important work. As metabolomic datasets mature, examination of individual metabolite distributions may allow exploration of ratio-based metrics—pairing metabolites elevated in PDAC with those enriched in non-malignant conditions—to potentially enhance discriminatory performance. Ratio-based approaches have improved discrimination and robustness in other biomarkers and metabolomic contexts, suggesting that similar secondary modeling strategies could be explored in future analyses of the METAPAC dataset (24,25).

Beyond analytic considerations, thoughtful planning will be important to support successful clinical integration. As with many mass spectrometry–based assays, attention to platform standardization, calibration, and pre-analytical sample handling will help ensure reproducibility across centers (26). Centralized testing infrastructure, similar to established molecular cyst fluid platforms such as PancreaSeq (27), may offer one pathway to maintain analytical consistency. Considerations such as plasma stability, shipping logistics, turnaround time, and accessibility across geographically diverse practice settings will be important factors in designing scalable implementation strategies.

Health economic evaluation represents another natural next step in translation. Demonstrating that incorporation of metabolomic signatures meaningfully reduces unnecessary surgical resections, improves identification of resectable disease, or enhances surveillance efficiency would strengthen the case for payer coverage and broader adoption. Decision-analytic modeling that quantifies downstream clinical impact—including avoided procedures and gains in quality-adjusted life years—may help clarify the value proposition of these assays within multidisciplinary care pathways.

Continued validation in racially and ethnically diverse populations will also be essential to ensure generalizability and equitable benefit. Given well-documented disparities in PDAC incidence, tumor biology, and outcomes (28-31), evaluating performance across heterogeneous cohorts represents an important opportunity to maximize the public health relevance of this work and to ensure that emerging diagnostic tools contribute to narrowing, rather than widening, inequities in care.

Finally, integration within existing clinical algorithms remains a promising area for further study. Modeling how metabolomic signatures perform when combined with radiologic features within the Kyoto framework (11) could clarify their real-world utility. The potential application may extend beyond ruling out overt invasive PDAC to refining risk stratification for HGD in IPMNs, thereby moving the field closer to biologically informed interception.

In summary, Mahajan and colleagues provided rigorous validation of two plasma metabolic signatures that enhance discrimination of PDAC (including in early-stage, resectable disease) from benign pancreatic conditions with high specificity and accuracy. The cohort enrichment design allows for their findings to be directly applicable to high-risk patient populations with the greatest need for refined diagnostic tools. Their work meaningfully advances the field toward multimodal diagnostic refinement. Continued efforts focused on standardization, economic evaluation, equitable validation, and integration into structured care pathways will help translate these promising findings into clinical impact. Metabolomic biomarkers hold considerable promise within a precision diagnostic ecosystem for pancreatic oncology. Ultimately, their influence on practice will depend not only on statistical performance, but on clinical utility within the complex decision-making environment that defines management of indeterminate pancreatic lesions.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Translational Gastroenterology and Hepatology. The article did not undergo external peer review.

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2026-0045/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Li X, Zhang Y, Yan Z, et al. Global, regional and national burden of pancreatic cancer and its attributable risk factors from 2019 to 2021, with projection to 2044. Front Oncol 2024;14:1521788. [Crossref] [PubMed]
2. Siegel RL, Kratzer TB, Giaquinto AN, et al. Cancer statistics, 2025. CA Cancer J Clin 2025;75:10-45. [Crossref] [PubMed]
3. Rahib L, Wehner MR, Matrisian LM, et al. Estimated Projection of US Cancer Incidence and Death to 2040. JAMA Netw Open 2021;4:e214708. [Crossref] [PubMed]
4. Gaddam S, Abboud Y, Oh J, et al. Incidence of Pancreatic Cancer by Age and Sex in the US, 2000-2018. JAMA 2021;326:2075-7. [Crossref] [PubMed]
5. Chang VT, Sandifer C, Zhong F. GI Symptoms in Pancreatic Cancer. Clin Colorectal Cancer 2023;22:24-33. [Crossref] [PubMed]
6. Park W, Chawla A, O'Reilly EM. Pancreatic Cancer: A Review. JAMA 2021;326:851-62. [Crossref] [PubMed]
7. Takeda Y, Saiura A, Takahashi Y, et al. Asymptomatic Pancreatic Cancer: Does Incidental Detection Impact Long-Term Outcomes? J Gastrointest Surg 2017;21:1287-95. [Crossref] [PubMed]
8. Takikawa T, Kikuta K, Hamada S, et al. Clinical features and prognostic impact of asymptomatic pancreatic cancer. Sci Rep 2022;12:4262. [Crossref] [PubMed]
9. Vilela A, Quingalahua E, Vargas A, et al. Global Prevalence of Pancreatic Cystic Lesions in the General Population on Magnetic Resonance Imaging: A Systematic Review and Meta-analysis. Clin Gastroenterol Hepatol 2024;22:1798-1809.e6. [Crossref] [PubMed]
10. Lahat G, Ben Haim M, Nachmany I, et al. Pancreatic incidentalomas: high rate of potentially malignant tumors. J Am Coll Surg 2009;209:313-9. [Crossref] [PubMed]
11. Ohtsuka T, Fernandez-Del Castillo C, Furukawa T, et al. International evidence-based Kyoto guidelines for the management of intraductal papillary mucinous neoplasm of the pancreas. Pancreatology 2024;24:255-70. [Crossref] [PubMed]
12. Srinivasan N, Teo JY, Chin YK, et al. Systematic review of the clinical utility and validity of the Sendai and Fukuoka Consensus Guidelines for the management of intraductal papillary mucinous neoplasms of the pancreas. HPB (Oxford) 2018;20:497-504. [Crossref] [PubMed]
13. Sun L, Li Y, Song Q, et al. The factors that influence the diagnostic accuracy and sample adequacy of EUS-guided tissue acquisition for the diagnosis of solid pancreatic lesions. Endosc Ultrasound 2024;13:183-9. [Crossref] [PubMed]
14. Farrell JJ. Prevalence, Diagnosis and Management of Pancreatic Cystic Neoplasms: Current Status and Future Directions. Gut Liver 2015;9:571-89. [Crossref] [PubMed]
15. Archibugi L, Ponz de Leon Pisani R, Petrone MC, et al. Needle-Tract Seeding of Pancreatic Cancer after EUS-FNA: A Systematic Review of Case Reports and Discussion of Management. Cancers (Basel) 2022;14:6130. [Crossref] [PubMed]
16. Liu C, Deng S, Jin K, et al. Lewis antigen‑negative pancreatic cancer: An aggressive subgroup. Int J Oncol 2020;56:900-8. [Crossref] [PubMed]
17. Mahajan UM, Oehrle B, Goni E, et al. Validation of two plasma multimetabolite signatures for patients at risk of or with suspected pancreatic ductal adenocarcinoma (METAPAC): a prospective, multicentre, investigator-masked, enrichment design, phase 4 diagnostic study. Lancet Gastroenterol Hepatol 2025;10:634-47. [Crossref] [PubMed]
18. Sun J, Awad MA, Hwang J, et al. Molecular Biomarkers for the Diagnosis and Prognostication of Pancreatic Ductal Adenocarcinoma. J Pers Med 2025;15:236. [Crossref] [PubMed]
19. Murph M, Tanaka T, Pang J, et al. Liquid chromatography mass spectrometry for quantifying plasma lysophospholipids: potential biomarkers for cancer diagnosis. Methods Enzymol 2007;433:1-25. [Crossref] [PubMed]
20. Zhang Q, Yang X, Wang Q, et al. "Modeling-Prediction" Strategy for Deep Profiling of Lysophosphatidic Acids by Liquid Chromatography-Mass Spectrometry: Exploration Biomarkers of Breast Cancer. J Chromatogr A 2020;1634:461634. [Crossref] [PubMed]
21. Dunn WB, Broadhurst D, Begley P, et al. Procedures for large-scale metabolic profiling of serum and plasma using gas chromatography and liquid chromatography coupled to mass spectrometry. Nat Protoc 2011;6:1060-83. [Crossref] [PubMed]
22. Porporato PE. Understanding cachexia as a cancer metabolism syndrome. Oncogenesis 2016;5:e200. [Crossref] [PubMed]
23. Park MA, Whelan CJ, Ahmed S, et al. Defining and Addressing Research Priorities in Cancer Cachexia through Transdisciplinary Collaboration. Cancers (Basel) 2024;16:2364. [Crossref] [PubMed]
24. Ganna A, Salihovic S, Sundström J, et al. Large-scale metabolomic profiling identifies novel biomarkers for incident coronary heart disease. PLoS Genet 2014;10:e1004801. [Crossref] [PubMed]
25. Rifai N, Gillette MA, Carr SA. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat Biotechnol 2006;24:971-83. [Crossref] [PubMed]
26. Gika HG, Theodoridis GA, Earll M, et al. Does the mass spectrometer define the marker? A comparison of global metabolite profiling data generated simultaneously via UPLC-MS on two different mass spectrometers. Anal Chem 2010;82:8226-34.
27. Paniccia A, Polanco PM, Boone BA, et al. Prospective, Multi-Institutional, Real-Time Next-Generation Sequencing of Pancreatic Cyst Fluid Reveals Diverse Genomic Alterations That Improve the Clinical Management of Pancreatic Cysts. Gastroenterology 2023;164:117-133.e7. [Crossref] [PubMed]
28. Fuko CD, Magacha HM, Noah G, et al. Ethnic/Racial Disparities in Pancreatic Cancer Mortality Across the United States: A National Inpatient Sample Database Analysis. Cureus 2025;17:e77389. [Crossref] [PubMed]
29. Liu W, Rao L, Qiao Z, et al. Global disparities in the burden of pancreatic cancer (1990-2021): insights from the 2021 Global Burden of Disease study. BMC Cancer 2025;25:722. [Crossref] [PubMed]
30. Mandal S, Teslow EA, Huang M, et al. Molecular Differences in Pancreatic Ductal Adenocarcinomas from Black versus White Patients. Cancer Res Commun 2025;5:128-37. [Crossref] [PubMed]
31. Samaan JS, Abboud Y, Oh J, et al. Pancreatic Cancer Incidence Trends by Race, Ethnicity, Age and Sex in the United States: A Population-Based Study, 2000-2018. Cancers (Basel) 2023;15:870. [Crossref] [PubMed]