Introduction

Colorectal cancer (CRC) is the third most commonly cancer which seriously affects the health of people worldwide [1]. In recent decades, advances of surgical procedures and chemotherapy have decreased the mortality rate of CRC [2, 3], however 20–30% of patients with advanced CRC still experience postoperative recurrence even after curative resection, leading to poor prognosis [4,5,6]. Therefore, it is critically important to identify a prognostic factor that allow more accurate prediction of poor survival in patients undergoing curative resection for CRC.

Cholinesterase (ChE) is one of classical nutritional markers [7], which is synthesized by the liver and secreted into the blood [8]. Previous studies have shown that low serum ChE levels have been associated with several clinical conditions, such as liver dysfunction, heart failure, and inflammation bowel disease [9,10,11], because serum ChE levels tend to decrease in the stress and inflammatory response sensitively [12, 13]. Furthermore, serum ChE level has been reported to be decreased in patients with advanced cancer, which can reflect cancer-related inflammation [14,15,16]. However, the prognostic role of ChE in patients with CRC has not been fully investigated compared with albumin and systemic inflammatory responses.

The aim of the present study is to evaluate the prognostic value of ChE for prognosis after curative resection for CRC, compared with other nutritional factors and systemic inflammatory responses.

Materials and methods

Patient selection

Between March 2014 and December 2020, 216 patients with CRC of stage I to III underwent curative colorectal resection (R0 resection). Of these, 42 patients were excluded (16 for obstruction or perforation as emergency, 3 for intervention for a synchronous metastatic lesion, 4 for local recurrence, 4 for multiple or double cancer, and 15 for insufficient data), remaining 174 patients for the present study. We performed a retrospective review of a prospectively maintained database of patients. The present study was approved by the Ethics Committee of Tokyo General Hospital (No. 21–3).

Assessment of the nutrition and systemic inflammatory response

Hemogram and chemistry profile were routinely evaluated within 4 weeks before surgery. We measured albumin and ChE which is a classical nutritional marker, as previously described [17]. Additionally, the systemic inflammatory response was assessed as follows: neutrophil to lymphocyte ratio (NLR) [18], prognostic nutritional index (PNI) [19], and C-reactive protein to albumin ratio (CAR) [20]. NLR was calculated by dividing the neutrophil count (per mm3) by the lymphocyte count (per mm3) [18]. PNI was calculated as follows: 10 × serum albumin level (g/dl) + 0.005 × total peripheral lymphocyte count (per mm3) [19]. CAR was calculated by dividing the serum C-reactive protein (CRP) level (mg/dl) by serum albumin level (g/dl) [20].

Treatment and patient follow-up

Postoperative complications were defined as those occurring within 30 days after primary surgery. Patients with Clavien-Dindo grade II or higher complications were included in the complication group [21]. Postoperatively, the patients were followed up every 3 months for 3 years, and then every 6 months until 5 years. At each follow-up, tumor biomarkers, including serum carcinoembryonic antigen (CEA) and carbohydrate antigen (CA) 19–9, were measured. Contrast-enhanced computed tomography or magnetic resonance imaging and colonoscopy were performed during the follow-up period according to the NCCN Clinical Practice Guidelines in Oncology [22].

Statistical analysis

All statistical analyses were conducted using EZR software version 1.51 (Saitama Medical Center, Jichi Medical University, Japan), and all P-values were two-sided with α level of 0.05.

Data are expressed as a median value. Continuous and categorical variables were compared using the Mann–Whitney U or chi-square tests, as appropriate. The optimal cutoff value of clinical continuous variables was determined by a receiver-operating characteristic (ROC) analyses to overall survival event. Clinical continuous variables were classified into two groups for Cox proportional-hazard regression models using the cutoff value, and above or below the standard value.

We evaluated the prognostic significance of ChE in patients with CRC of stage I to III. Univariate and multivariate Cox proportional-hazard regression models were used to estimate the hazard ratios (HR) for disease-free and overall survival. The multivariable Cox regression model included sex (male vs. female), age (≥ 65 vs. < 65 years), surgical approach (laparoscopic vs. open), tumor location (rectum vs. colon), pathological depth of invasion (T3 or T4 vs. T1 or T2), pathological lymph node metastasis (positive vs. negative), serum albumin (< 3.5 vs. ≥ 3.5 g/dl), serum ChE (< 247 vs. ≥ 247 U/l), NLR (≥ 2.64 vs. < 2.64), PNI (< 45.2 vs. ≥ 45.2), CAR (≥ 0.119 vs. < 0.119), serum CEA (≥ 6.0 vs. < 6.0 ng/ml), serum CA19-9 (≥ 37 vs. < 37 U/ml), and postoperative complications (yes vs. no). A backward elimination stepwise approach was conducted with a threshold P-value of 0.05 to select variables for the final model. The Kaplan–Meier method was used to estimate cumulative survival probabilities, and the differences between groups were compared using the log-rank test.

Results

Patients’ characteristics

Demographic and preoperative characteristics of patients (n = 174) are shown in Table 1. The median age was 74 (23–98) years and 97 patients were male. Fifty-nine patients (33.9%) developed postoperative complications and laparoscopic surgery was performed in 108 patients (62.1%). In the present study, the median follow-up time was 3.1 years and the 5-year overall survival rate after resection for CRC was 83.3%.

Table 1 Patient characteristics
Full size table

Survival curve in patients after curative resection for CRC according to preoperative serum ChE levels

The optimal cutoff value of serum ChE levels determined by a ROC analysis was 247 U/l, and the area under the curve (AUC) was 0.684 (95% confidence interval [CI]: 0.578–0.790). Preoperative serum ChE levels were significantly associated with disease-free survival (Fig. 1a; log-rank test, P < 0.01) and overall survival (Fig. 1b; log-rank test, P < 0.01) after curative resection for CRC.

Fig. 1
figure 1

Kaplan–Meier curves according to preoperative serum ChE levels for disease-free survival (a) and overall survival (b) after curative resection for colorectal cancer. Preoperative serum ChE levels were significantly associated with disease-free survival (P < 0.01) and overall survival (P < 0.01)

Full size image

Univariate and multivariate analyses of clinicopathologic variables in relation to disease-free survival after curative resection for CRC

Table 2 lists the relationship between the clinical variables and disease-free survival after curative resection for CRC. In univariate analysis, disease-free survival was significantly worse in patients with depth of invasion T3 or T4 (P = 0.029), lymph node metastasis (P < 0.01), albumin < 3.5 g/dl (P = 0.011), ChE < 247 g/dl (P < 0.01), PNI < 45.2 (P = 0.026), CAR ≥ 0.119 (P < 0.01), CEA ≥ 6.0 ng/ml (P < 0.01), and postoperative complication occurrence (P = 0.037). In multivariate analysis, lymph node metastasis (P = 0.011) and ChE < 247 g/dl (P < 0.01) were independent predictors of disease-free survival.

Table 2 Univariate and multivariate analyses of factors associated with disease-free survival
Full size table

Univariate and multivariate analyses of clinicopathologic variables in relation to overall survival after curative resection for CRC

Table 3 lists the relationship between the clinical variables and overall survival after curative resection for CRC. In univariate analysis, overall survival was significantly worse in patients with lymph node metastasis (P < 0.01), albumin < 3.5 g/dl (P < 0.01), ChE < 247 g/dl (P < 0.01), PNI < 45.2 (P = 0.020), CAR ≥ 0.119 (P < 0.01), CEA ≥ 6.0 ng/ml (P < 0.01), and CA19-9 ≥ 37 ng/ml (P = 0.017). In multivariate analysis, lymph node metastasis (P = 0.013), ChE < 247 g/dl (P < 0.01), and CA19-9 ≥ 37 ng/ml (P = 0.022) were independent predictors of overall survival.

Table 3 Univariate and multivariate analyses of factors associated with overall survival
Full size table

Patients’ characteristics according to preoperative serum ChE levels

The clinical perioperative characteristics of patients according to ChE levels are shown in Table 4. Body mass index (BMI) (P< 0.01), surgical approach (laparoscopic) (P < 0.01), serum albumin levels (P < 0.01), and PNI (P < 0.01) were significantly lower in the low-ChE group. On the other hand, age (P < 0.01), NLR (P = 0.021), CAR (P < 0.01), and distant metastasis (P < 0.01) were significantly higher in the low-ChE group. There was no difference in local recurrence (P = 0.79) between the two groups.

Table 4 Clinicopathological variables according to preoperative serum ChE levels in patients undergoing curative resection for CRC
Full size table

Subgroup analysis according to the tumor-node classification

Subgroup analysis was performed to further investigate the predictive value of preoperative serum ChE levels. In the subgroup of T3-4 patients, the low-ChE group had worse disease-free survival and overall survival than the high-ChE group. Similar associations were observed in the subgroup of T1-2 (Fig. 2a–d). Also, in the subgroup of lymph node metastasis-positive patients, the low-ChE group had worse disease-free survival and overall survival than the high-ChE group. Similar associations were observed in the subgroup of lymph node metastasis-negative patients (Fig. 3a–d).

Fig. 2
figure 2

Kaplan–Meier curves according to preoperative serum ChE levels for disease-free survival and overall survival in the subgroups of T3-4 or T1-2 patients. In the subgroup of T3-4 patients, the low-ChE group had worse disease-free survival (a) and overall survival (b) than the high-ChE group. Similar associations were observed in the subgroup of T1-2 patients (c and d)

Full size image
Fig. 3
figure 3

Kaplan–Meier curves according to preoperative serum ChE levels for disease-free survival and overall survival in the subgroup of lymph node metastasis-positive or -negative patients. In the subgroup of lymph node metastasis-positive patients, the low-ChE group had worse disease-free survival (a) and overall survival (b) than the high-ChE group. Similar associations were observed in the subgroup of lymph node metastasis negative patients (c and d)

Full size image

Discussion

We showed the significant association between preoperative serum ChE levels and poor prognosis in patients who had underwent curative resection for CRC. Multivariable analysis revealed that low ChE levels were the independent risk factor for worse disease-free survival and overall survival. Furthermore, subgroup analyses revealed that the low-ChE group had worse disease-free survival and overall survival than the high-ChE group regardless of T and N stage. Also, the low-ChE group had more distant metastasis and higher systemically inflammatory response compared with the high-ChE group. These findings suggest that preoperative serum ChE levels, which is a classical nutrition marker, can be one of the prognostic values compared with the other systemic inflammatory response in patients who underwent colorectal resection for CRC.

ChE has been reported as predictor of postoperative prognosis in various cancers, including gastric [23] and pancreatic cancer [24]. In consistent with the previous studies, our finding revealed that serum ChE levels can be a significant predictor in patients with CRC. The mechanism that preoperative serum ChE levels are sensitive for predicting poor prognosis in patients with CRC is unclear. However, there are some reasons for determining that ChE is a valuable factor compared with the other systemic inflammatory response. Preoperative serum ChE levels can reflect nutritional status more acutely than albumin, because the half-life of ChE is shorter than albumin (3–11 vs. 12–21 days) to reflect preoperative general condition more sensitively [7, 25,26,27]. In fact, serum ChE levels were more accurately than albumin, PNI, and CAR for assessing poor prognosis in the present study. Furthermore, low serum ChE levels can reflect the elevation of inflammation status [15]. In patients with cancer or frailty, cytokine such as interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α), which were negatively correlated with serum ChE levels, significantly increase to activate signal transducer and activator of transcription 3 (STAT3) and nuclear factor-kappa B (NF-kB) pathway in cancer [15, 16]. These inflammatory cascades have been reported to block the synthesis of ChE and albumin in favor of acute-phase protein synthesis [15, 28], and promote cancer-related inflammation, epithelial-to-mesenchymal transition (EMT)-mediated invasion, and metastasis in CRC [29, 30]. These evidence support our findings that several systemic inflammatory responses and distant metastasis were negatively associated with serum ChE levels.

The present study has several limitations. The most important limitation was the retrospective observational study and the confounding factors cannot be completely ruled out. Serum ChE levels might be confounded by several factors such as inflammatory, nutritional status, and cancer stage, which may have influenced the results. Also, since the present study was conducted at a single institution with limited sample size. Further prospective studies with the large sample size are needed to robustly assess the prognostic value of ChE and validate our findings.

Conclusion

Preoperative low serum ChE levels predicted worse disease-free survival and overall survival in patients with CRC. Routine preoperative measurement of preoperative serum ChE levels in patients undergoing curative resection for CRC would be useful for assessing their nutritional-inflammatory status and predicting postoperative outcome.