As one of the most lethal cancers, pancreatic cancer is still a frequently occurring disease and remains a therapeutic challenge to humans [18, 19]. Although gemcitabine is a currently and widely used drug in the therapy of pancreatic cancer, various approaches, such as drug delivery system, have to be tried to prolong the plasma half-life of gemcitabine and enhance its bioavailability [20, 21]. As the typical examples, liposome and carbon nanotube have been a success in delivering cancer drugs for pancreatic cancer treatment in recent animal and preclinical trials [19, 22]. Nowadays, a novel carrier system allowing for lower toxic side effects and higher tumor-targeting efficiencies is emphasized, while the high biosafety of the carrier system is also prerequisite [8, 10, 23]. In our previous work, BSA was introduced to act as a drug carrier for gemcitabine loading . We found that GEM-ANPs could result in a sustained release and improved antitumor activity in vitro of gemcitabine. Here, we further exposed human pancreatic carcinoma (PANC-1) to GEM-ANPs and studied cell responses in vitro by cell viability analysis and flow cytometry technique. The loading of gemcitabine on albumin did not reduce the inhibition effect of gemcitabine on PANC-1 metabolism. Moreover, GEM-ANPs with bigger size could even enhance the killing efficacy of gemcitabine in pancreatic carcinoma (Figure 1). GEM-ANPs showed their cell cycle inhibitory property, in the order of 406-nm GEM-ANPs > 110-nm GEM-ANPs > gemcitabine. The higher antiproliferative activity of 406-nm GEM-ANPs could be attributed to the S phase arrest during cell cycle progression (Table 2).
Besides the shorter half-life, the toxic side effects, like increased liver enzymes and leukopenia, have also limited the applications of gemcitabine . Therefore, the blood parameters of rats treated with GEM-ANPs were investigated to assess the reduction effect of albumin loading on gemcitabine toxic side effects. Since the blank nanoparticles could interfere with the growth of cells in vitro, the US Pharmacopoeia limits cell inhibition as no more than 50% for safety . The present study revealed that no significant difference between the ANP treatment group and control group was observed in WBC, RBC, and other parameters of hepatonephric functions, suggesting a satisfactory biocompatibility (Table 1). What was more important was that the high-dose treatment with GEM-ANPs, especially 406-nm GEM-ANPs, could reduce the side effects of gemcitabine (Table 1). In fact, gemcitabine concentration and treatment period were insufficient to induce a relevant blood toxicity in the present study . Our results also demonstrated that gemcitabine loading on 406-nm GEM-ANPs significantly increased the gemcitabine content in the pancreas, liver, and spleen of SD rats compared with the gemcitabine treatment group, but contrary to 110-nm GEM-ANPs (p < 0.05) (Table 3). It is well known that nanospheres are easily taken up by cells of the mononuclear phagocyte system, primarily those located in the reticuloendothelial system-rich organs, such as the liver and spleen . Furthermore, phagocytosis will gradually increase as the size is more than 200 nm . Consequently, it might be one of the reasonable mechanisms for the targeting effect of 406-nm GEM-ANPs in vivo. That was to say, 406-nm GEM-ANPs would enhance the curative effect of gemcitabine in pancreatic cancer. Particularly, literatures have reported that the microvascular permeability of most normal tissues was generally less than 50 nm, but ten times higher in tumor tissues and usually more than 500 nm. For example, Hobbs et al. found that the microvascular permeability of rat hepatoma, fibrosarcoma, and human colon cancer animal model reach 380 to 550, 550 to 780 and 380 to 550 nm, respectively . Yuan et al. also found that the maximum diameter of microvascular permeability in human colon cancer is between 400 and 600 nm . In addition, Desai  and Cortes and Saura  found that albumin nanoparticles could increase albumin receptor, 60-kDa glycoprotein (gp60)-mediated transcytosis, through microvessel endothelial cells in angiogenic tumor vasculature and targets the albumin-binding protein SPARC, which subsequently increased intratumoral accumulation. Therefore, a relatively high antitumor activity of 406-nm GEM-ANPs could be expected due to the passive targeting by EPR effect and gp60-mediated transcytosis [8–10, 23, 32, 33]. Here, the antitumor effects of GEM-ANPs were assessed in vivo using the implanted tumor model of nude mice. We found that the antitumor effect of 406-nm GEM-ANPs was greatest (Figures 2 and 3), with 168.8% inhibitory rate compared to the control. Finally, the slow release of gemcitabine from 406-nm GEM-ANPs could also prolong the drug action, and it might be another possible reason for the higher antitumor activity of GEM-ANPs.