Deep Learning Predicts Imminent Tumor Progression in Advanced Pancreatic Adenocarcinoma Using Serial CT Scans During Chemotherapy

Abstract

Advanced pancreatic ductal adenocarcinoma (PDAC) often progresses rapidly during chemotherapy despite initial assessments of stable disease or partial response by Response Evaluation Criteria in Solid Tumors (RECIST 1.1), underscoring the limitations of the current methods for predicting short-term progressive disease (PD). To address this, the study developed a spatiotemporal deep learning framework that integrates convolutional and long short-term memory (LSTM) neural networks to dynamically predict PD at the next follow-up visit using serial computed tomography (CT) scans and baseline clinical variables. The model was trained on a retrospective cohort of 243 patients (415 predicted events, defined as temporal sequences for the next follow-up PD prediction) and evaluated across internal, external, and prospective cohorts. The model achieved area under the curve (AUC) values of 0.77, 0.76, and 0.74, respectively. Performance remained robust across chemotherapy regimens (AG or Gemcitabine-based, FOLFIRINOX, and SOXIRI; AUC 0.68–0.79), PD subtypes (target lesion growth vs. new metastases; AUC 0.72 vs. 0.77), and baseline disease stages (locally advanced vs. metastatic; AUC 0.85 vs. 0.71). This framework enables the noninvasive, real-time prediction of imminent PD in advanced PDAC, facilitating timely treatment modification. Its validated generalizability and reliance on routine clinical data underscore its potential for seamless integration into chemotherapy management.

Keywords

deep learning / longitudinal prediction / pancreatic ductal adenocarcinoma / progressive disease / serial CT

References

[1] S. Lei, Y. Mao, Q. Yang, et al., “Trends in Pancreatic Cancer Incidence, Prevalence, and Survival Outcomes by Histological Subtypes: A Retrospective Cohort Study,” Gastroenterology Report 13 (2025): goaf030.

[2] R. L. Siegel, T. B. Kratzer, A. N. Giaquinto, et al., “Cancer Statistics, 2025,” CA: A Cancer Journal for Clinicians 75 (2025): 10–45.

[3] T. Conroy, F. Desseigne, M. Ychou, et al., “FOLFIRINOX Versus gemcitabine for Metastatic Pancreatic Cancer,” New England Journal of Medicine 364 (2011): 1817–1825.

[4] D. D. Von Hoff, T. Ervin, and F. P. Arena, “Increased Survival in Pancreatic Cancer With Nab-Paclitaxel Plus Gemcitabine,” New England Journal of Medicine 369 (2013): 1691–1703.

[5] A. J. Grossberg, L. C. Chu, C. R. Deig, et al., “Multidisciplinary Standards of Care and Recent Progress in Pancreatic Ductal Adenocarcinoma,” CA: A Cancer Journal for Clinicians 70 (2020): 375–403.

[6] M. Del Chiaro, T. Sugawara, S. D. Karam, et al., “Advances in the Management of Pancreatic Cancer,” Bmj 383 (2023): e073995.

[7] T. Conroy, P. Pfeiffer, V. Vilgrain, et al., “Pancreatic Cancer: ESMO Clinical Practice Guideline for Diagnosis, Treatment and Follow-up,” Annals of Oncology 34 (2023): 987–1002.

[8] S. De Dosso, A. R. Siebenhüner, T. Winder, et al., “Treatment Landscape of Metastatic Pancreatic Cancer,” Cancer Treatment Reviews 96 (2021): 102180.

[9] E. A. Eisenhauer, P. Therasse, J. Bogaerts, et al., “New Response Evaluation Criteria in Solid Tumours: Revised RECIST Guideline (Version 1.1),” European Journal of Cancer 45 (2009): 228–247.

[10] L. Lu, L. Dercle, B. Zhao, et al., “Deep Learning for the Prediction of Early On-Treatment Response in Metastatic Colorectal Cancer From Serial Medical Imaging,” Nature Communications 12 (2021): 6654.

[11] A. Dohan, B. Gallix, B. Guiu, et al., “Early Evaluation Using a Radiomic Signature of Unresectable Hepatic Metastases to Predict Outcome in Patients With Colorectal Cancer Treated With FOLFIRI and Bevacizumab,” Gut 69 (2020): 531–539.

[12] Q. Yang, Y. Mao, Y. Han, et al., “Noninvasive Computed Tomography-Based Quantification of Tumor Fibrosis Predicts Pancreatic Cancer Response to Gemcitabine/Nab-Paclitaxel,” Research 8 (2025): 0937.

[13] J. Cheng, Y. Pan, W. Huang, et al., “Differentiation Between Immune Checkpoint Inhibitor-Related and Radiation Pneumonitis in Lung Cancer by CT Radiomics and Machine Learning,” Medical Physics 49 (2022): 1547–1558.

[14] R. Huang, X. Meng, X. Zhang, et al., “Artificial Intelligence-Driven Change Redefining Radiology Through Interdisciplinary Innovation,” Interdisciplinary Medicine 3, no. 1 (2025): e20240063.

[15] W. He, X. Wu, Z. Jin, et al., “Generative Artificial Intelligence in Medical Imaging: Current Landscape, Challenges, and Future Directions,” Interdisciplinary Medicine 3 (2025): e20250024.

[16] J. Li, H. Huang, D. Ni, et al., “MUVF-YOLOX: A Multi-Modal Ultrasound Video Fusion Network for Renal Tumor Diagnosis,” International Conference on Medical Image Computing and Computer-Assisted Intervention (2023): 642–651.

[17] L. Zhou, L. Ouyang, B. Guo, et al., “Enhanced Prediction of Preoperative Mesenteric Lymph Node Metastasis in Colorectal Cancer Using Machine Learning With CT-Based Data,” MedComm-Future Medicine 3 (2024): e100.

[18] Q. Yang, Y. Mao, H. Xie, et al., “Identifying Outcomes of Patients With Advanced Pancreatic Adenocarcinoma and RECIST Stable Disease Using Radiomics Analysis,” JCO Precision Oncology 6 (2022): e2100362.

[19] J. Yao, Y. Shi, K. Cao, et al., “DeepPrognosis: Preoperative Prediction of Pancreatic Cancer Survival and Surgical Margin via Comprehensive Understanding of Dynamic Contrast-Enhanced CT Imaging and Tumor-Vascular Contact Parsing,” Medical Image Analysis 73 (2021): 102150.

[20] J. Yao, K. Cao, Y. Hou, et al., “Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study,” Annals of Surgery 278 (2023): e68–e79.

[21] S. P. Madani, M. Mirza-Aghazadeh-Attari, A. Mohseni, et al., “Value of Radiomics Features Extracted From Baseline Computed Tomography Images in Predicting Overall Survival in Patients With Nonsurgical Pancreatic Ductal Adenocarcinoma: Incorporation of a Radiomics Score to a Multiparametric Nomogram to Predict 1-Year Overall Survival,” Journal of Gastrointestinal Surgery 29 (2025): 101882.

[22] W. Lu, G. Wu, X. Miao, et al., “The Radiomics Nomogram Predicts the Prognosis of Pancreatic Cancer Patients With Hepatic Metastasis After Chemoimmunotherapy,” Cancer Immunology, Immunotherapy 73 (2024): 87.

[23] H. Nasief, C. Zheng, D. Schott, et al., “A Machine Learning Based Delta-Radiomics Process for Early Prediction of Treatment Response of Pancreatic Cancer,” NPJ Precision Oncology 3 (2019): 25.

[24] T. F. Stoop, R. T. Theijse, L. W. F. Seelen, et al., “Preoperative Chemotherapy, Radiotherapy and Surgical Decision-Making in Patients With Borderline Resectable and Locally Advanced Pancreatic Cancer,” Nature Reviews Gastroenterology & Hepatology 21 (2024): 101–124.

[25] P. T. Chen, T. Wu, P. Wang, et al., “Pancreatic Cancer Detection on CT Scans With Deep Learning: A Nationwide Population-Based Study,” Radiology 306 (2023): 172–182.

[26] K. Cao, Y. Xia, J. Yao, et al., “Large-Scale Pancreatic Cancer Detection via Non-Contrast CT and Deep Learning,” Nature Medicine 29 (2023): 3033–3043.

[27] N. Sushentsev, L. Rundo, L. Abrego, et al., “Time Series Radiomics for the Prediction of Prostate Cancer Progression in Patients on Active Surveillance,” European Radiology 33 (2023): 3792–3800.

[28] Y. Xu, A. Hosny, R. Zeleznik, et al., “Deep Learning Predicts Lung Cancer Treatment Response From Serial Medical Imaging,” Clinical Cancer Research 25 (2019): 3266–3275.

[29] Y. LeCun, Y. Bengio, and G. Hinton, “Deep Learning,” Nature 521 (2015): 436–444.

[30] M. He, Z. F. Chen, S. Liu, et al., “Deep Learning Model Based on Multi-Lesion and Time Series CT Images for Predicting the Benefits From Anti-HER2 Targeted Therapy in Stage IV Gastric Cancer,” Insights Imaging 15 (2024): 59.

[31] W. Lee, H. J. Park, H. J. Lee, et al., “Preoperative Data-Based Deep Learning Model for Predicting Postoperative Survival in Pancreatic Cancer Patients,” International Surgery Journal 105 (2022): 106851.

[32] J. H. Lee, J. Shin, J. H. Min, et al., “Preoperative Prediction of Early Recurrence in Resectable Pancreatic Cancer Integrating Clinical, Radiologic, and CT Radiomics Features,” Cancer Imaging 24 (2024): 6.

[33] S. van Roessel, G. G. Kasumova, J. Verheij, et al., “International Validation of the Eighth Edition of the American Joint Committee on Cancer (AJCC) TNM Staging System in Patients with Resected Pancreatic Cancer,” JAMA Surgery 153 (2018): e183617.

[34] S. Lei, Y. Mao, T. Qin, et al., “Neutrophil-to-Apolipoprotein A1 Ratio Predicted the Efficiency of Chemotherapy in Advanced Pancreatic Cancer,” Pancreatology 25, no. 3 (2025): 417–425.

[35] M. Piper, R. B. Ross, J. Hu, et al., “Vasculitis, CA19-9, and Perineural Invasion Differentially Predict Response and Surgical Outcome in Pancreatic Ductal Adenocarcinoma,” International Journal of Radiation and Oncology in Biology and Physics 116 (2023): 627–639.

[36] J. L. Humphris, D. K. Chang, A. L. Johns, et al., “The Prognostic and Predictive Value of Serum CA19.9 in Pancreatic Cancer,” Annals of Oncology 23 (2012): 1713–1722.

[37] L. Laurent, D. Sefrioui, A. L. Bignon, et al., “CA19.9 Decrease >15% Is a Predictor of Favourable Outcome in Patients Treated for Advanced Pancreatic Carcinoma: Analysis of Two Independent Cohorts,” HPB 21 (2019): 582–588.

[38] E. G. Chiorean, D. D. Von Hoff, and M. Reni, “CA19-9 Decrease at 8 Weeks as a Predictor of Overall Survival in a Randomized Phase III Trial (MPACT) of Weekly Nab-Paclitaxel Plus Gemcitabine Versus Gemcitabine Alone in Patients With Metastatic Pancreatic Cancer,” Annals of Oncology 27 (2016): 654–660.

[39] S. Tsai, B. George, D. Wittmann, et al., “Importance of Normalization of CA19-9 Levels Following Neoadjuvant Therapy in Patients With Localized Pancreatic Cancer,” Annals of Surgery 271 (2020): 740–747.

[40] R. A. L. Willems, J. Konings, D. Huskens, et al., “Altered Whole Blood Thrombin Generation and Hyperresponsive Platelets in Patients With Pancreatic Cancer,” Journal of Thrombosis and Haemostasis 22 (2024): 1132–1144.

[41] T. Hank, U. Hinz, T. Reiner, et al., “A Pretreatment Prognostic Score to Stratify Survival in Pancreatic Cancer,” Annals of Surgery 276 (2022): e914–e922.

[42] V. Giannakeas, J. Kotsopoulos, J. D. Brooks, et al., “Platelet Count and Survival After Cancer,” Cancers 14 (2022): 549.

[43] G. Perri, L. Prakash, W. Qiao, et al., “Response and Survival Associated With First-Line FOLFIRINOX vs Gemcitabine and Nab-Paclitaxel Chemotherapy for Localized Pancreatic Ductal Adenocarcinoma,” JAMA Surgery 155 (2020): 832–839.

[44] E. Perez, F. Strub, and H. D. Vries, “FiLM: Visual Reasoning With a General Conditioning Layer,” Proceedings of the AAAI Conference on Artificial Intelligence 32 (2018), https://doi.org/10.1609/aaai.v32i1.11671.

[45] R. Fluss, D. Faraggi, and B. Reiser, “Estimation of the Youden Index and Its Associated Cutoff Point,” Biometrical Journal 47 (2005): 458–472.

Rights & permissions

2026 The Author(s). MedComm published by Sichuan International Medical Exchange & Promotion Association (SCIMEA) and John Wiley & Sons Australia, Ltd.