Predictive model using CT scans, blood tests, and pathology to forecast pancreatic cancer outcomes
Official title: A Study on a Predictive Model for Efficacy and Prognosis of Pancreatic Carcinoma Based on Multimodal Data
Run by Shanghai Zhongshan Hospital · NCT07561814 · Data updated 2026-11-12
This project will try to see if combining CT scans, blood tests, and pathology reports can better predict outcomes for adults who had surgery for pancreatic cancer.
Quick facts
| Study type | Observational |
|---|---|
| Ages | 18 Years and up |
| Sex | All |
| Enrollment | 1030 (estimated) |
| Sponsor | Shanghai Zhongshan Hospital Academic / other |
| Locations | 1 site (Shanghai, Shanghai Municipality) |
| Status | Recruiting |
| Started | December 9, 2025 |
| Main results expected | July 31, 2029 |
| Study ends | December 31, 2029 |
| Registry record updated | July 2, 2026 |
| Trial ID | NCT07561814 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Adults (age ≥18) with histologically or cytologically confirmed pancreatic cancer who had surgical treatment at a participating center and have available pre- and postoperative CT images, clinical data, pathology reports, and blood test results are ideal candidates.
✕Probably not a fit
Patients without the required imaging or pathology data, those with other severe comorbidities, minors, pregnant or breastfeeding women, or people unable to consent or complete follow-up are unlikely to benefit.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
This observational project will use preoperative and postoperative contrast-enhanced CT images, routine blood tests, and pathology reports from patients who underwent surgery for pancreatic cancer to build a multimodal predictive model using deep learning methods. Researchers will extract clinical and imaging features from existing medical records without adding tests or treatments and will conduct telephone follow-up about every three months for up to three years to capture outcomes. The main goals are to improve estimates of overall survival and risk of recurrence by integrating multiple data types. Data quality checks and exclusion of cases with excessive missing data or extreme outliers will be applied before modeling.
Why it matters
Potential benefit: If successful, the model could help clinicians give more accurate survival and recurrence estimates and better personalize follow-up and treatment planning.
How similar studies have gone: Previous radiomics and multimodal machine-learning studies in pancreatic and other cancers have shown promising but mixed results, so the approach is supported by preliminary evidence but not yet standard practice.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * Age ≥ 18 years. * Histologically or cytologically confirmed diagnosis of pancreatic cancer. * Availability of preoperative and postoperative contrast-enhanced CT images at the participating center; patient records include corresponding clinical data, postoperative pathological reports, and preoperative and postoperative blood test results. * Underwent surgical treatment for pancreatic cancer at a participating center of this study. Exclusion Criteria: * Age \< 18 years (minors). * Presence of other severe diseases (e.g., severe liver or kidney failure, cardiovascular or cerebrovascular diseases, malignancies other than pancreatic cancer); pregnant or breastfeeding women; individuals with mental illness; * patients unable to comply with follow-up or provide informed consent. * Presence of significant outliers (e.g., laboratory values exceeding 10 times the normal range without clinically reasonable explanation) or samples with excessive missing data.
Where it's running
-
Zhongshan Hospital Affiliated to Fudan University
Shanghai, Shanghai Municipality, China Recruiting
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Integrated genomic and clinical database for cancer biomarker discovery
Researchers are building a combined genomic and clinical database at Samsung Medical Center to test biomarkers for cancer diagnosis, treatment choice, and predicting how well treatments work.
-
Using contrast-enhanced ultrasound to check for local pancreatic cancer recurrence after surgery
This pilot test uses contrast-enhanced ultrasound to see if it can help detect return of pancreatic cancer after surgery in patients being watched with CT scans.
-
Combination of ivermectin with standard skin-protective care for daraxonrasib-related skin toxicity in pancreatic cancer
This study tests whether adding ivermectin to standard preventive skin care helps people with pancreatic cancer who develop skin side effects from daraxonrasib.
-
Becotatug Vedotin Plus Anti-PD-1 and S-1 in EGFR-Positive Advanced Pancreatic Cancer
This study tests a three-drug regimen (becotatug vedotin, an anti-PD-1 therapy, and S-1) in adults with EGFR-positive advanced pancreatic cancer who have not responded to at least one prior therapy.
-
Blood protein biomarkers to help diagnose pancreatic cancer in adults
This study tests whether blood protein biomarkers can help detect pancreatic ductal adenocarcinoma and tell it apart from chronic pancreatitis and mucinous pancreatic lesions, and whether combining markers with CA 19-9 improves accuracy.
-
Lm-LLO-TT and Gemcitabine in Unresectable Pancreatic Cancer (Phase 1)
This early-stage study tests whether giving Lm-LLO-TT directly into the belly with low-dose gemcitabine is safe and tolerable for adults with previously treated, unresectable or metastatic pancreatic cancer.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.