Baseline characteristics
A total of 202 cases suffering GBC were recruited, the mean age was 68.54±11.02 years, 68.3% were female, and the mean AAPR reached 0.40±0.22. Furthermore, 46% cases had low AAPR with GBC according to diagnosis. Table 1 lists the fundamental features of survivor and non-survivor cases. Compared with survivor cases, non-survivor cases were older, with lower BMI, hemoglobin, albumin and AAPR. Between the survivor and non-survivor cases, large CA19-9 (P<0.001) and large CEA (P=0.006) were noticeably different. In terms of tumor characteristics, lymph node metastasis (P=0.046), tumor differentiation (P<0.001), advanced TNM stage (P<0.001) and advanced infiltration depth T (P<0.001) noticeably impacted the death of GBC. Moreover, the AAPR was conducive to splitting participants into two groups, group with high AAPR (n=109), and group with low AAPR (n=93) (Table 2). In contrast to high AAPR cases, the low AAPR cases were older, with lower hemoglobin, albumin. Higher ALP, NLR (P<0.001), CA19-9 (P=0.028), CEA (P=0.029), incidence of gallstones (P=0.048), advanced infiltration depth T (P=0.033), advanced TNM stage (P=0.099) and larger tumor size (P=0.009) evidently impacted low GNRI levels.
Kaplan-Meier survival analysis
Of the 202 cases, 94 died, and 108 continued to live. In terms of the OS (P<0.001) (Fig. 1) and RFS (P<0.001) (Fig. 2), the prognosis was noticeably worse in cases exhibiting low AAPR levels than in cases with high AAPR levels. Cases in the high AAPR level group developed a middle OS time of 38 months (95% CI 28.35-47.94), and those in the low AAPR group developed a middle OS of 26 months (95% CI 21.03-30.97). The recurrence rate during following-up here reached 52.0% (105 cases), and the middle illness -free survival in this series reached 24 months. The middle RFS was noticeably smaller in the low AAPR group than in the high AAPR group (12months and 35 months, separately; P<0.01).
Univariate and multivariate analysis
Table 3 lists the OS-relevant parameters when GBC radical surgical process is completed in line with univariate and multivariate Cox proportional hazard modes. For the univariate study, age, serum CA19-9>37, CEA>5, hemoglobin level, lymph node metastasis, TNM III+IV, ineffective differentiation, invasive depth III+IV, GNRI<100, AAPR<0.4 exhibited relationships to low OS. In line with the results of multivariate analysis, age (HR, 1.030, 95%CI, 1.001-1.061; P=0.047), hemoglobin (HR, 1.040, 95%CI, 1.000-1.049; P=0.043), ineffective differentiation (HR, 2.073, 95%CI, 1.081-4.381; P=0.031), TNM III+IV (HR, 7.853, 95%CI, 1.648-37.421; P=0.010), GNRI (HR, 2.312, 95%CI, 1.119-4.777; P=0.024) and AAPR (HR, 2.247, 95%CI, 1.483-3.404; P=0.028) adversely influenced OS. Table 4 lists a multiple-variate and univariate Cox proportional hazard regression mode in terms of RFS. Multivariate analysis identified three adverse prognosis elements, lymph node metastases (HR, 2.605, 95%CI, 1.163-6.896; P=0.026), GNRI (HR, 4.881, 95%CI, 2.367-10.063; P<0.001) and AAPR (HR, 1.640, 95%CI, 1.037,3.210; P=0.041), which affect RFS.
Prognosis implication of AAPR
To compare the predictability of the mortality and recurrence rate among GBC cases, ROC curves of ALP, albumin, AAPR were drawn (Fig 3). AAPR displayed the maximal area under the ROC curves, with statistical significance (AUC=0.706 for mortality, 95%CI, 0.633-0.779; P<0.001, AUC=0.711 for recurrence rate, 95%CI, 0.641-0.781; P<0.001). The optimal cutoffs of GNRI for predicted mortality and recurrence rate reached 0.40 and 0.41 separately.