Pancreatic and Esophageal cancers are highly aggressive with high mortality and limited treatment, causing over 466,000 and 544,100 deaths worldwide in 2020 respectively. This highlights the urgent need for safer,and effective anticancer agents. Catalpol, a natural iridoid glycoside, shows anticancer potential, but due to its poor drug-like properties it requires structural modification. This study investigates pyrazole-modified catalpol derivatives as dual inhibitors for these cancers using Quantitative Structure Activity Relationship (QSAR) modelling, molecular docking, and pharmacokinetic studies. We analyzed fourteen pyrazole-modified catalpol derivatives with reported IC50values against four cancer cell lines(BxPC-3, PANC-1, Eca109, and EC9706). The molecules were optimized using DensityFunctional Theory (DFT), and 2D molecular descriptors were calculated using PaDEL. QSAR models were developed by utilizing a Genetic Function Algorithm (GFA) and Multiple Linear Regression (MLR) and validated using statistical metrics such as R2, Q2, R2adj, and R2pred. Docking studies targeted VEGFR-2 and BRAF V600E kinases using AutoDockVina, while ADMET and drug-likeness properties were predicted using SwissADME and pkCSM tools. The external validation R2pred values for BxPC-3, PANC-1, Eca109, and EC9706 cell lines were 0.9412, 0.9535, 0.9981, and 0.9935, respectively. Among the derivatives, compound 3k showed the highest binding affinity for VEGFR-2 (- 8.18 kcal/mol) and BRAF (- 8.64 kcal/mol), surpassing the control drugs etoposide (- 8.00 kcal/mol) and dabrafenib (- 8.15 kcal/mol) respectively. ADMET analysis confirmed good intestinal absorption, limited blood-brain barrier penetration, non-toxicity, acceptable total clearance, and compliance with Lipinski's rule. Overall, the study suggests that pyrazole-modified catalpol derivatives, especially compound 3k, are promising multi-target inhibitors for pancreatic and esophageal cancers, justify further in-vitro and in-vivo studies.