Systemic abrogation of TGF-β signaling results in tumor reduction through cytotoxic T lymphocytes activity in a mouse model. The administration of polysaccharide-Kureha (PSK) into tumor-bearing mice also showed tumor regression with reduced TGF-β. However, there have been no studies regarding the PSK administration to cancer patients and the association with plasma TGF-β. PSK (3 g/day) was administered as a neoadjuvant therapy for 2 weeks before surgery. In total, 31 advanced gastric cancer (AGC) patients were randomly assigned to group A (no neoadjuvant PSK; n=14) or B (neoadjuvant PSK therapy; n=17). Plasma TGF-β was measured pre- and postoperatively. The allocation factors were clinical stage (cStage) and gender. Plasma TGF-β ranged from 1.85-43.5 ng/ml (average, 9.50 ng/ml) in AGC, and 12 patients (38.7%) had a high value, >7.0 ng/ml. These patients were largely composed of poorly-differentiated adenocarcinoma with pathological stage III/IV. All the six elevated cases in group B showed a significant reduction of plasma TGF-β (from 21.6 to 4.5 ng/ml, on average), whereas this was not exhibited in group A. The cases within the normal limits of TGF-β remained unchanged irrespective of PSK treatment. Analysis of variance showed a statistically significant reduction in the difference of plasma TGF-β between groups A and B (P=0.019). PSK reduced the plasma TGF-β in AGC patients when the levels were initially high. The clinical advantage of PSK may, however, be restricted to specific histological types of AGC. Perioperative suppression of TGF-β by PSK may antagonize cancer immune evasion and improve patient prognosis in cases of AGC.