Pancreatic cancer (PC) stands for the most intractable malignancy.Gemcitabine (GEM) is one of the few approved first-line treatments for PC patients.The fast clearance, demanded high dosage, and existence of drug resistance have, nevertheless, posed not only a significant limitation to its clin. efficacy but also serious toxicity concerns.KRASG12D mutation is identified as a key driver in many PC patients, and its expression level shows a correlation with drug resistance and mortality.Here, we explored KRASG12D siRNA-gemcitabine oligonucleotide chimeras (siKRAS-Gn) as a dual prodrug that was designed to specifically silence KRASG12D gene and sensitize PC cells to GEM for the synergistic treatment of PC. siKRAS-Gn conjugates with 1, 2, 3, 4, or 5 units of GEM were synthesized and delivered using cRGD-decorated polymersomes.Interestingly, the proapoptotic activity of siKRAS-Gn was shown to highly depend on the number of GEM, in which three GEM units (siKRAS-G3) were found to be optimal and induced strong apoptosis of PANC-1 cells (apoptosis rate: 64.2%).In contrast, minimal cell apoptosis was discerned for siKRAS, siKRAS-G1, siKRAS-G2, and free GEM (9-fold of that in siKRAS-G3). siKRAS-G3, while showing similar KRAS mRNA silencing ability to siKRAS, markedly enhanced the downregulation of KRAS protein in vitro and in vivo.Accordingly, siKRAS-G3 significantly outperformed siKRAS and siScramble-G3 in both tumor inhibition and survival benefits.The targeted delivery of the siKRAS-gemcitabine prodrug conjugate has emerged as an appealing treatment for pancreatic cancer.