Mode
Text Size
Log in / Sign up

Systematic review and meta-analysis on pre-diagnostic weight loss in pancreatic cancer

Systematic review and meta-analysis on pre-diagnostic weight loss in pancreatic cancer
Photo by Geoff Moore / Unsplash
Key Takeaway
Consider pre-diagnostic weight loss as a potential biomarker for pancreatic cancer, but recognize the association is modest and requires validation.

This is a systematic review and meta-analysis examining pre-diagnostic weight loss in pancreatic cancer patients compared to matched controls. The analysis included 12,081 cases and 367,678 controls. The main synthesized finding was a standardized mean difference (SMD) of 0.51 (95% CI 0.31-0.71) for weight loss in people with pancreatic cancer, indicating a modest association. A BMI reduction equivalent of 2.26 kg/m2 was also reported.

The authors note significant heterogeneity, influenced by country, publication year, cancer type, and diabetes status. They acknowledge that weight loss may serve as a useful non-invasive biomarker for early detection, especially in those with new-onset diabetes, but emphasize the need for further validation.

Limitations include the observational nature of the included studies and the noted heterogeneity. Practice relevance is restrained, suggesting that quantified weight loss data could enhance predictive algorithms for earlier detection, but causation cannot be inferred from this association.

Study Details

Study typeMeta analysis
EvidenceLevel 1
PublishedMay 2026
View Original Abstract ↓
Pancreatic cancer, characterised by high mortality and late diagnosis, urgently requires improved early detection methods. Unintentional weight loss, reported by 70-75% of patients prior to diagnosis, may serve as a key early marker. This systematic review and meta-analysis evaluated the extent of pre-diagnostic weight loss in pancreatic cancer patients compared with matched controls. Thirteen studies encompassing 12,081 cases and 367,678 controls were analysed, revealing a medium standardised mean difference (SMD = 0.51, [95% CI 0.31-0.71]) equivalent to a 2.26 kg/m2 BMI reduction in people with pancreatic cancer. Significant heterogeneity was observed, influenced by country, publication year, cancer type and diabetes status. Weight loss may serve as a useful non-invasive biomarker for early pancreatic cancer detection, especially in those with new-onset diabetes. Using quantified weight loss data could enhance the accuracy of predictive algorithms, allowing their inclusion in EMR based detection pipelines, ultimately aiding earlier detection and improving survival outcomes.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.