Abstract
Background/Aim: The albumin-bilirubin (ALBI) score is a promising tool for the evaluation of the perioperative hepatic function. The present study aimed to evaluate the clinical impact of the preoperative ALBI status in patients with gastric cancer (GC) who received curative treatment. Patients and Methods: The present study included 244 patients who underwent curative treatment for GC between 2005 and 2018. The risk factors for overall survival (OS) and recurrence-free survival (RFS) were identified. Results: Based on the 3- and 5-year OS rates, we set the cut-off value for the ALBI score at -2.7849. The 3- and 5-year OS rates were 87.3% and 80.9%, respectively, in the ALBI-low group, and 66.9% and 60.6% in the ALBI-high group; these differences were statistically significant (p<0.001). The ALBI score was included in the final multivariate analysis model [Hazard ratio (HR)=2.120, 95% confidence interval (CI)=1.177-3.818, p=0.012]. Similar results were observed for RFS. In addition, the ALBI score correlated with the introduction of postoperative adjuvant chemotherapy. Conclusion: The preoperative ALBI score correlated with both the OS and RFS of GC patients as well as the clinical course of adjuvant chemotherapy. Taken together, the ALBI score is a promising prognostic factor for GC.
Gastric cancer (GC) is the third-most common cancer and the second leading cause of cancer-related death in the world (1, 2). Perioperative adjuvant treatment and gastrectomy with lymphadenectomy is the standard treatment for GC (3-5). Although the survival rate of GC is gradually improving, almost half of GC patients develop recurrent disease, even after standard treatment (6, 7). Therefore, it is necessary to develop more aggressive treatments for GC patients who have risk factors for recurrence. Thus far, several prognostic factors, including the metabolic function, nutrition function, and immunological function, have been identified in GC patients (8-10). Among them, the hepatic function is a key prognostic factor. The hepatic function directly reflects the metabolic and synthesis functions (11, 12). However, there are no optimal tools or methods to evaluate the perioperative liver function as a prognostic factor for GC. Recently, the albumin-bilirubin (ALBI) score was developed to evaluate the severity of perioperative liver dysfunction in hepatocellular carcinoma patients (13). The ALBI score consists of the perioperative serum albumin and serum bilirubin levels. The albumin level reflects the synthesis function of the liver, while the bilirubin level reflects the metabolic function of the liver. Therefore, the ALBI score can evaluate the perioperative liver function. In addition, recent studies have demonstrated that the ALBI score is closely related to oncological outcomes in some malignancies (14, 15). Moreover, the ALBI score is easy to measure, highly reproducible, and low-cost. Considering these merits, the ALBI score is a promising and useful prognostic factor for GC in daily clinical practice. However, studies evaluating the clinical impact of the ALBI score in GC have been limited. The present study aimed to evaluate the clinical impact of the preoperative ALBI score in patients with GC who received curative treatment.
Patients and Methods
Patients. Consecutive patients who underwent curative resection for GC at Yokohama City University from 2005 and 2018 were selected according to medical records. Patients with histologically proven adenocarcinoma, that was classified as clinical stage I-III according to the 15th edition of the general rules for GC published by the Japanese Gastric Cancer Association (16), who received R0 resection with radical lymph node dissection were included in the study.
Surgery and adjuvant treatment. All patients underwent gastrectomy with D1+ or D2 lymphadenectomy. Those with pathological II or III disease received postoperative adjuvant chemotherapy for one year. Generally, pathological stage II disease was treated with S-1 monotherapy, whereas pathological stage III disease was treated with S-1 plus docetaxel or capecitabine plus oxaliplatin.
Patient follow-up. At a minimum, patients received physical examinations and hematological tests every three months for five years. Tumor marker levels (carcinoembryonic antigen and CA19-9) were checked every three months for five years. Computed tomography (CT) was performed every 6-12 months for 5 years.
Determination of the albumin-bilirubin score. The ALBI score was calculated using the following formula: ALBI=log10 T-Bil (μmol/l)×0.66+Alb (g/l)×–0.0852. Preoperative blood parameters were collected retrospectively from the patients’ medical records (17).
Statistical analyses. Differences between the ALBI score and clinicopathological parameters were analyzed by a χ2 test. Overall survival (OS) and recurrence-free survival (RFS) curves were calculated by the Kaplan–Meier method. Univariate and multivariate analyses of survival were conducted using a Cox proportional hazards model. Statistical significance was defined as p<0.05. All statistical analyses were conducted using SPSS (v27.0 J Win; SPSS, Chicago, IL, USA).
Results
Patients. A total of 244 patients were included in this study. The cut-off value for the ALBI score was set to –2.7849 according to the 3- and 5-year OS rates and previous reports (Table I). Patients were divided into ALBI-low [<–2.7849; n=127 (52%)] and ALBI-high [>–2.7849; n=117 (48%)] groups according to the cut-off value. The analysis of the patient background characteristics revealed statistically significant differences between the ALBI-low and ALBI-high groups in the following variables: median age (66 years vs. 71 years, p<0.001), preoperative hemoglobin (13.7 g/dl vs. 11.9 g/dl, p<0.001), and preoperative C-reactive protein (0.11 mg/dl vs. 0.65 mg/dl, p<0.001).
Comparison of survival rates stratified by patient characteristics.
Patient survival. The 3-year and 5-year OS rates in the ALBI-low group were 87.3% and 80.9%, respectively, while those in the ALBI-high group were 66.9% and 60.6%. These differences were statistically significant (p<0.001) (Figure 1). The clinicopathological factors were categorized as shown in Table II and their prognostic significance was analyzed. The univariate analyses of factors associated with OS demonstrated that the ALBI score, pathological T status, pathological N status, histological type, and perioperative surgical complications were significant prognostic factors. Thus, the ALBI score was included in the final multivariate analysis model [Hazard ratio (HR)=2.120; 95% confidence interval (CI)=1.177-3.818, p=0.012]. The 3-year and 5-year RFS rates in the ALBI-low group were 84.1% and 78.2%, respectively, whereas those in the ALBI-high group were 64.2% and 52.3%. These differences were statistically significant (p<0.001) (Figure 2). The clinicopathological factors were categorized as shown in Table III and their prognostic significance was analyzed. The univariate analyses of factors associated with OS and RFS demonstrated that the ALBI score, pathological T status, pathological N status, lymphatic invasion, and perioperative surgical complications were significant prognostic factors. The ALBI score was also selected for inclusion in the final multivariate analysis model (HR=1.778, 95%CI=1.033-6.644, p=0.038). A comparison of the sites of recurrence between the ALBI-low and ALBI-high groups demonstrated a significance difference in peritoneal recurrence (6.3% vs. 16.2%, p=0.013) (Table IV).
Overall survival in the ALBI-high and ALBI-low groups. ALBI: Albumin-bilirubin.
Uni- and multivariate Cox proportional hazards analysis of clinicopathological factors for overall survival.
Uni- and multivariate Cox proportional hazards analysis of clinicopathological factors for recurrence-free survival.
Recurrence-free survival in the ALBI-high and ALBI-low groups. ALBI: Albumin-bilirubin.
Patterns of recurrence according to systemic inflammation score.
The perioperative clinical courses of the patients in the ALBI-low and ALBI-high groups. The rates of postoperative surgical complications in the ALBI-low and ALBI-high groups did not differ to a statistically significant extent (37.4% vs. 38.1%, p=0.913). Furthermore, the rates of anastomotic leakage (7.2% vs. 9.5%, p=0.511), pneumonia (7.2% vs. 4.0%, p=0.260), and abdominal abscess (2.2% vs. 3.8%, p=0.444) were similar between the two groups. The ALBI score did not affect the incidence of postoperative surgical complications.
In contrast, a statistically significant difference was observed in the percentage of patients who required postoperative adjuvant chemotherapy, with 28.1% of the patients in the ALBI-low group and 41.9% of the patients in the ALBI-high group requiring postoperative adjuvant chemotherapy, which amounted to a statistically significant difference (p=0.024). However, the percentage of patients who received postoperative adjuvant chemotherapy in the ALBI-high group was significantly lower than that in the ALBI-low group (89.7% vs. 65.9%, p=0.010). The results suggest that the ALBI score affected the introduction of postoperative adjuvant chemotherapy.
Discussion
The present study aimed to evaluate the clinical impact of the preoperative ALBI score in GC patients who received curative treatment. The major finding was that the preoperative ALBI score affected both OS and RFS in GC patients. In addition, preoperative ALBI score also affected the clinical course of adjuvant chemotherapy. Considering these, ALBI is one of the promising prognostic factors for GC patients.
In the present study, there were significant differences in OS and RFS between the ALBI-high and ALBI-low groups. In addition, HR was 2.120 for OS and 1.778 for RFS. Previously, limited studies showed similar results in GC. For example, Kanda et al. evaluated the survival impacts of the ALBI score in 283 locally advanced GC patients who received radical gastrectomy (18). They divided the patients into an ALBI-high group (ALBI score >–2.60; n=228) and an ALBI-low group (ALBI score <–2.60; n=55). The comparison of survival between the ALBI-high and ALBI-low group revealed significant differences in 5-year disease-specific survival (66% vs. 83%, p=0.0014) and 5-year disease-free survival (59% vs. 82%, p=0.0004). The overall recurrence rate was 17% in the ALBI-high group and 38% in the ALBI-low group, which was a significant difference (p=0.0008). Moreover, Zue et al. clarified the survival impact of the ALBI score in 243 patients with locally advanced GC who received radical gastrectomy (19). They divided the patients into an ALBI-high group (ALBI score >–2.60; n=141) and an ALBI-low group (ALBI score <–2.34; n=102). The median survival time was 33.8 months in the ALBI-high group and 39.8 months in the ALBI-low group (p<0.001). Furthermore, there was a significant difference in the median survival times of the ALBI-high and ALBI-low groups when patients were grouped according to the TNM stage. In patients with stage II disease, the median survival time was 33.9 months in the ALBI-high group and 41.9 months in the ALBI-low group (p=0.043). In patients with stage III disease, the median survival time was 24.1 months in the ALBI-high group and 40.4 months in the ALBI-low group (p<0.001). Similar results were observed in other gastrointestinal cancers. Taken together, the ALBI score may be correlated with the survival of patients with GC.
There are some possible reasons why the ALBI status correlates with the survival of GC patients. First, the ALBI status might affect the clinical course of adjuvant treatment. In the present study, the ALBI-high group included more patients who required adjuvant treatment in comparison to the ALBI-low group (28.1% vs. 41.9%, p=0.024). However, more patients in the ALBI-low group actually received adjuvant treatment in comparison to the ALBI-high group (89.7% vs. 65.9%, p=0.010). The ALBI status had some clinical impact on the introduction of adjuvant treatment. In addition, Miwa et al. evaluated the clinical relation between the ALBI status and the tolerability of postoperative adjuvant treatment in 98 GC patients who received S-1 adjuvant chemotherapy (20). They divided the patients into an ALBI-high group (ALBI score >–2.696; n=17) and an ALBI-low group (ALBI score <–2.696; n=81). The duration of S-1 adjuvant chemotherapy in the ALBI-high group was significantly shorter than that in the ALBI-low group (p=0.001). In addition, ALBI was an independent risk factor for the continuation of S-1 adjuvant chemotherapy (odd ratios: 10.3, 95%CI=2.33-45.8, p=0.002). Considering these findings, the preoperative ALBI status may correlate with the clinical course of adjuvant treatment in GC. A second possible reason was that the preoperative ALBI score had a clinical association with the occurrence of postoperative surgical complications. Zhu et al. compared the incidence of postoperative surgical complications according to the preoperative ALBI status (19). They found that the incidence of surgical complications was 32.6% in their ALBI-high group and 16.7% in their ALBI-low group (p=0.005). In addition, the rate of severe complications (Clavien-Dindo grade ≥III) was 11.3% in the ALBI-high group and 3.9% in the ALBI-low group (p=0.038). The authors concluded that the preoperative ALBI status could predict postoperative complications in GC. However, the mechanisms of the association between the ALBI status and survival of GC patients are unclear. Further studies should focus on this issue.
The present study had some limitations. First, this was a retrospective study that was conducted at a single institution between 2010 and 2017. Therefore, there may be a selection bias and time bias in the present study. Second, the optimal cut-off value of the ALBI score was unclear. In the present study, we set a cut-off value of –2.7849 according to the survival rate. Previous studies set the cut-off values of –2.696, –2.60, and –2.34 according to the ROC curves and X-tile software program (18-20). The differences in ALBI can be attributed to the number patients, patient background factors, and the methods used to evaluate the ALBI score. For the ALBI score to be applied as a prognostic factor in GC, it is necessary to establish the optimal cut-off value and methods of evaluation.
In conclusion, the preoperative ALBI score correlated with both OS and RFS in GC patients as well as the clinical course of adjuvant chemotherapy. Taken together, the ALBI score is a promising prognostic factor for GC.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 21K08688.
Footnotes
↵* These Authors contributed equally to this study.
Authors’ Contributions
TA, MJ, and KK made substantial contributions to the concept and design. TA, MJ, KK, HT, AT, KK2 (Kazuki Kano), IH, YM, HC, KH, TI, SO, TO, NY, TO, and YR made substantial contributions to the acquisition, analysis, and interpretation of the data. TA, MJ, HT, TO, NY, and YR were involved in drafting the article or revising it critically for important intellectual content. TA, MJ, KK, and TO gave their final approval of the version to be published.
Conflicts of Interest
The Authors declare no conflicts of interest in association with the present study.
- Received May 26, 2022.
- Revision received June 17, 2022.
- Accepted June 21, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.