Abstract
Background: This study aimed to determine whether the prealbumin prognostic score (PPS), a novel indicator using prealbumin instead of albumin in the modified Glasgow Prognostic Score (mGPS), is a better predictive marker postoperatively in patients with gastric cancer. Methods: This retrospective study included consecutive patients who underwent radical gastrectomy for primary pStages I-III gastric cancer between 2006 and 2017. The cutoff values for preoperative prealbumin and C-reactive protein (CRP) were 22 mg/dL and 0.5 mg/dL, respectively. According to the prealbumin and CRP levels, a PPS of zero was defined as both being above the cutoff value, of one as either being below the cutoff value, and of two as both being below the cutoff value. Results: Of the 4663 patients, 3421 (73.4%) had a score of zero, 984 (21.1%) had a score of one, and 258 (5.5%) had a score of two. The higher the PPS, the poorer the overall survival [OS] (p < 0.001). When comparing OS by the PPS in patients with an mGPS of zero, a PPS of one indicated poorer OS than a PPS of zero (p < 0.001). In the multivariate analysis, PPSs of one (hazard ratio [HR]: 1.603; 95% confidence interval [CI]: 1.378-1.866; p < 0.001) and two (HR: 1.322; 95% CI: 1.055-1.656; p = 0.015) were independent poor prognostic factors for OS. Conclusions: The PPS, which is based on a combination of prealbumin and CRP levels, can identify a wider range of patients with poor OS than mGPS in patients with gastric cancer after gastrectomy.