Abstract
Objectives:
The current retrospective study aimed to evaluate the association between combined preoperative and postoperative C-reactive protein-to-albumin ratio, which is correlated with prognosis in different types of malignancies, and prognosis after curative resection in patients with colorectal cancer.
Methods:
This study enrolled 263 patients who underwent curative resection for stage II/III colorectal cancer. C-reactive protein-to-albumin ratio was calculated within 30 days before and 7 days after surgery. Receiver operating characteristic curve analyses were performed to determine the optimal cutoff values of preoperative and postoperative C-reactive protein-to-albumin ratio. The correlations between combined preoperative and postoperative C-reactive protein-to-albumin ratio and prognosis were analyzed.
Results:
The cutoff values of preoperative and postoperative C-reactive protein-to-albumin ratio were 0.223 and 0.813, respectively; higher ratios were significantly associated with poor overall survival, based on the Kaplan-Meier curves (p < 0.001, p = 0.003, respectively). Further, preoperative and postoperative C-reactive protein-to-albumin ratios were correlated with poor progression-free survival (p < 0.001, p = 0.064, respectively). In the multivariate analysis, combined preoperative and postoperative C-reactive protein-to-albumin ratio was an independent predictor of overall survival and progression-free survival (p = 0.012, p = 0.044, respectively). Compared with low preoperative and postoperative C-reactive protein-to-albumin ratio, high ratios of that were significantly associated with poor overall survival (hazard ratio = 3.897, p = 0.006) and progression-free survival (hazard ratio = 2.130, p = 0.029).
Conclusions:
Combined preoperative and postoperative C-reactive protein-to-albumin ratio, useful for prognostic prediction, can be a promising prognostic marker after curative resection in patients with colorectal cancer.
Keywords: colorectal cancer, prognosis, C-reactive protein, albumin
Introduction
Colorectal cancer (CRC) is the third most common malignancy and has the second highest mortality rate[1]. Although there are improvements in surgical procedures and chemotherapy, several patients still have worse outcomes. To further improve treatment outcomes, it is important to not only develop treatment methods but also predict prognosis, which may provide information regarding therapeutic options.
Virchow first discovered the correlation between inflammation and malignancies in 1863[2]. Several studies have shown the detailed mechanisms on the effects of systemic inflammatory response on tumor development and progression[3]. Thus, increasing attention has been paid to the development of simple and cost-effective inflammation-based markers, including C-reactive protein-to-albumin ratio (CAR), lymphocyte-to-monocyte ratio (LMR), neutrophil-to-lymphocyte ratio (NLR), Onodera's prognostic nutritional index (PNI), platelet-to-lymphocyte ratio, and Glasgow coma scale score, for predicting prognosis[4].
While most of the inflammation-based markers have evaluated preoperative host status, the number of reports on the association between postoperative inflammation-based markers and prognosis is limited[5-10]. However, the evaluation of postoperative inflammation-based markers after tumor removal may be informative because these markers can reflect different prognostic factors, such as degree of surgical stress and recovery from surgery and the host initial inflammatory status. Furthermore, since preoperative and postoperative inflammation-based markers may be used to evaluate different factors, combined preoperative and postoperative inflammation-based marker can be a good prognostic marker.
CRP (C-reactive protein) reflects systemic inflammatory response, and there are no reports evaluating postoperative CAR. Therefore, this study focused on CAR, which comprises CRP and albumin. Furthermore, CAR is not calculated using blood cell count alone, which is easily affected by age and sex. Hence, it can predict the prognosis of patients with heterogeneous characteristics such as elderly ones[11]. The predictive value of CAR is comparable or superior to that of other inflammation-based markers in different types of malignancies[11-15]. Herein, we evaluate the prognostic significance of perioperative CAR after curative resection in patients with CRC.
Methods
Study patients
The current study was approved by the institutional review board (approval number: 2022GS-028). This study was conducted in accordance with the Helsinki Declaration using opt-out consent. We retrospectively reviewed consecutive patients who underwent elective curative resection for stage II/III CRC at the Department of Surgery of Suita Municipal Hospital between January 2012 and December 2018. Patients with incomplete laboratory data, multiple primary tumors, and those who underwent simultaneous resection of other organs, emergency surgery, and neoadjuvant radiotherapy or chemotherapy were excluded.
Patients were followed-up. That is, they underwent blood tests including carcinoembryonic antigen (CEA) levels every 3 months and chest and abdominal computed tomography scan every 6 months according to the Japanese Society for Cancer of the Colon and Rectum guidelines[16].
Data collection
We collected data on sex, age, body mass index, primary tumor location (rectum: about up to 12 cm from anal verge), tumor size, histological type, CEA level, preoperative and postoperative CAR, pathological T classification, lymph node metastasis, surgical approach, adjuvant chemotherapy, and postoperative complications (i.e., ≥ grade 2 complications within 30 days of surgery based on the Clavien-Dindo classification system[17]). Open conversion from laparoscopic surgery included open surgery as it could not be determined whether conversion was planned or unexpected due to the retrospective nature. Blood tests were conducted within 30 days before and 7 days after surgery. CAR was calculated as CRP levels divided by serum albumin levels.
Statistical analysis
Continuous variables were presented as median and range. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cutoff values of preoperative and postoperative CAR. Categorical variables were compared using the Fisher's exact probability test, and continuous variables were compared using the Mann-Whitney U test. Kaplan-Meier curves and log-rank tests were utilized to analyze survival data. A multivariate analysis was performed using the Cox proportional hazard model to determine the predictive factors of overall survival (OS) and relapse-free survival (RFS). P values of <0.05 were considered significant. Statistical analyses were conducted using JMP software (version 13; SAS Institute, Cary, NC, the USA).
Results
Clinicopathological characteristics of the participants
Table 1 shows the characteristics of 263 patients at a median follow-up period of 50.0 months. The current study comprised 146 male and 117 female, with a median age of 72 years. The median preoperative and postoperative CAR levels were 0.055 and 0.397, respectively. In total, 120 patients had positive lymph node metastasis, and 143 did not. Moreover, 51 (19.4%) patients presented with postoperative complications. There was an extremely low correlation between the preoperative and postoperative CAR (r = 0.127, p = 0.040) (Figure 1).
Table 1.
Clinicopathological Characteristics of the Patients.
| Characteristics of the patients | Total |
|---|---|
| n = 263 | |
| Sex, n (%) | |
| Male | 146 (55.5) |
| Female | 117 (44.5) |
| Age (range), years | 72 (41–100) |
| BMI (range), kg/m2 | 21.97 (13.31–36.96) |
| Primary tumor location, n (%) | |
| Colon | 228 (86.7) |
| Rectum | 35 (13.3) |
| Tumor size (range), mm | 50 (15–130) |
| Histological type, n (%) | |
| Well/moderately differentiated | 241 (91.6) |
| Others | 22 (8.4) |
| CEA level, ng/mL | |
| <5 | 147 (56.3) |
| >5 | 114 (43.7) |
| Preoperative CAR (range) | 0.054 (0.039–7.919) |
| Postoperative CAR (range) | 0.397 (0.050–6.748) |
| Tumor depth, n (%) | |
| T1–3 | 212 (80.6) |
| T4 | 51 (19.4) |
| Lymph node metastasis, n (%) | |
| Negative | 120 (45.6) |
| Positive | 143 (54.4) |
| Surgical approach, n (%) | |
| Open | 35 (13.3) |
| Laparoscopic | 228 (86.7) |
| Adjuvant chemotherapy, n (%) | |
| No | 148 (56.3) |
| Yes | 115 (43.7) |
| Complications (>grade 2 based on the CD classification system) | |
| Absent | 212 (80.6) |
| Present | 51 (19.4) |
BMI, body mass index; CEA, carcinoembryonic antigen; CAR C-reactive protein-to-albumin ratio
Figure 1.
Correlation between preoperative and postoperative CAR. CAR C-reactive protein-to-albumin ratio
Correlations between preoperative and postoperative CAR and clinicopathological characteristics
According to the ROC analyses, the cutoff values of preoperative and postoperative CAR were 0.223 and 0.813, respectively (Figure 2). Based on these values, the participants were classified under the high and low CAR groups preoperative and postoperatively. A high preoperative CAR was associated with a tumor size of ≥ 50 mm (p < 0.001), tumor depth (T4) (p = 0.006), open surgery (p = 0.001), and postoperative complication (p = 0.017). Among the 35 open surgery patients, 18 patients were open conversion from laparoscopic surgery. A high postoperative CAR was correlated with age ≥75 years (p = 0.001), primary tumor location (rectum) (p = 0.030), open surgery (p = 0.030), and postoperative complication (p < 0.001) (Table 2).
Figure 2.
Receiver operating characteristics curves of preoperative (a) and postoperative (b) CAR of overall survival in patients with colorectal cancer after curative resection. CAR C-reactive protein-to-albumin ratio
Table 2.
Association between the Preoperative and Postoperative C-Reactive Protein-to-Albumin Ratio and Clinicopathological Characteristics.
| Preoperative CAR | Postoperative CAR | |||||
|---|---|---|---|---|---|---|
| High CAR group (n = 63) | Low CAR group (n = 200) | p value | High CAR group (n = 78) | Low CAR group (n = 185) | p value | |
| Characteristics of the participants | ||||||
| Sex, n (%) | 0.565 | 0.136 | ||||
| Male | 37 (58.7) | 109 (54.5) | 49 (62.8) | 97 (52.4) | ||
| Female | 26 (41.3) | 91 (45.5) | 29 (37.2) | 88 (47.6) | ||
| Age (years), n (%) | 0.148 | 0.001 | ||||
| <75 | 29 (46.0) | 114 (57.0) | 30 (38.5) | 113 (61.1) | ||
| >75 | 34 (54.0) | 86 (43.0) | 48 (61.5) | 72 (38.9) | ||
| BMI (kg/m2), n (%) | 0.291 | 1.000 | ||||
| <25 | 53 (84.1) | 155 (77.5) | 62 (79.5) | 146 (78.9) | ||
| >25 | 10 (15.9) | 45 (22.5) | 16 (20.5) | 39 (21.1) | ||
| Primary tumor location, n (%) | 0.673 | 0.030 | ||||
| Colon | 56 (88.9) | 172 (86.0) | 62 (79.5) | 166 (89.7) | ||
| Rectum | 7 (11.1) | 28 (14.0) | 16 (20.5) | 19 (10.3) | ||
| Tumor size (mm), n (%) | <0.001 | 0.496 | ||||
| <50 | 13 (20.6) | 101 (50.8) | 31 (39.7) | 83 (45.1) | ||
| >50 | 50 (79.4) | 98 (49.2) | 47 (60.3) | 101 (54.9) | ||
| Histological type, n (%) | 0.747 | 0.627 | ||||
| Well/moderately differentiated | 57 (90.5) | 184 (92.0) | 73 (93.6) | 168 (90.8) | ||
| Others | 6 (9.5) | 16 (8.0) | 5 (6.4) | 17 (9.2) | ||
| CEA level (ng/mL), n (%) | 0.187 | 1.000 | ||||
| <5 | 30 (48.4) | 117 (58.8) | 43 (55.8) | 104 (56.5) | ||
| >5 | 32 (51.6) | 82 (41.2) | 34 (44.2) | 80 (43.5) | ||
| Tumor depth, n (%) | 0.006 | 0.609 | ||||
| T1–3 | 43 (68.3) | 169 (84.5) | 61 (78.2) | 151 (81.6) | ||
| T4 | 20 (31.7) | 31 (15.5) | 17 (21.8) | 34 (18.4) | ||
| Lymph node metastasis, n (%) | 0.247 | 0.416 | ||||
| Negative | 33 (52.4) | 87 (43.5) | 39 (50.0) | 81 (43.8) | ||
| Positive | 30 (47.6) | 113 (56.5) | 39 (50.0) | 104 (56.2) | ||
| Surgical approach, n (%) | 0.001 | 0.030 | ||||
| Open | 17 (27.0) | 18 (9.0) | 16 (20.5) | 19 (10.3) | ||
| Laparoscopic | 46 (73.0) | 182 (91.0) | 62 (79.5) | 166 (89.7) | ||
| Adjuvant chemotherapy, n (%) | 0.194 | 0.104 | ||||
| No | 40 (63.5) | 108 (54.0) | 50 (64.1) | 98 (53.0) | ||
| Yes | 23 (36.5) | 92 (46.0) | 28 (35.9) | 87 (47.0) | ||
| Complications (>grade 2 based on the CD classification system), n (%) | 0.017 | <0.001 | ||||
| Absent | 44 (69.8) | 168 (84.0) | 45 (57.7) | 167 (90.3) | ||
| Present | 19 (30.2) | 32 (16.0) | 33 (42.3) | 18 (9.7) | ||
BMI, body mass index; CEA, carcinoembryonic antigen; CAR, C-reactive protein-to-albumin ratio
Survival analysis according to preoperative and postoperative CAR
Based on the survival curves, the high preoperative CAR group had significantly lower 5-year OS and RFS rates than the low preoperative CAR group (69.5% vs. 88.0%; p < 0.001; 55.4% vs. 75.4%; p = 0.003, respectively) (Figure 3a, b). The high postoperative CAR group had lower OS and PFS rates than the low postoperative CAR group (72.7% vs. 87.1%; p < 0.001; 63.5% vs. 73.8%; p = 0.064, respectively) (Figure 3c, d).
Figure 3.
Overall survival (a) and progression-free survival (b) according to preoperative CAR and overall survival (c) and progression-free survival (d) according to postoperative CAR. CAR C-reactive protein-to-albumin ratio
Survival analysis according to the combined preoperative and postoperative CAR
According to the combined preoperative and postoperative CAR, the patients were divided into three groups, which were as follows: low CAR group (both preoperative and postoperative patients with low CAR), intermediate CAR group (preoperative patients with high CAR and postoperative patients with low CAR or preoperative patients with low CAR and postoperative patients with high CAR), and high CAR group (both preoperative and postoperative patients with high CAR). Table 3 shows the association between the combined preoperative and postoperative CAR and clinicopathological features. Age (p = 0.005), tumor size (p = 0.005), surgical approach (p < 0.001), and postoperative complication (p < 0.001) were associated with the combined preoperative and postoperative CAR. Figure 4 shows the survival curves of OS and PFS. The high CAR group had a lower OS than the low CAR group and intermediate CAR groups (p < 0.001, p = 0.0130, respectively). The intermediate CAR group had a lower OS than the low CAR group (p = 0.006). The high CAR group had a lower PFS than the low and intermediate CAR groups (p < 0.001, p = 0.041, respectively).
Table 3.
Associations between Combined Preoperative and Postoperative C-Reactive Protein-to-Albumin Ratio and Clinicopathological Characteristics.
| Low CAR group | Intermediate CAR group | High CAR group | p value | |
|---|---|---|---|---|
| Sex, n (%) | 0.350 | |||
| Male | 78 (52.4) | 50 (57.5) | 18 (66.7) | |
| Female | 71 (47.6) | 37 (42.5) | 9 (33.3) | |
| Age (years), n (%) | 0.005 | |||
| <75 | 93 (62.4) | 41 (47.1) | 9 (33.3) | |
| >75 | 56 (37.6) | 46 (52.9) | 18 (66.7) | |
| BMI (kg/m2), n (%) | 0.134 | |||
| <25 | 113 (75.8) | 75 (86.2) | 20 (74.1) | |
| >25 | 36 (24.2) | 12 (13.8) | 7 (25.9) | |
| Primary tumor location, n (%) | 0.521 | |||
| Colon | 132 (88.6) | 74 (85.1) | 22 (81.5) | |
| Rectum | 17 (11.4) | 13 (14.9) | 5 (18.5) | |
| Tumor size (mm), n (%) | 0.005 | |||
| <50 | 77 (52.0) | 30 (34.5) | 7 (25.9) | |
| >50 | 71 (48.0) | 57 (65.5) | 20 (74.1) | |
| Histological type, n (%) | 0.535 | |||
| Well/moderately differentiated | 135 (90.6) | 82 (94.3) | 24 (88.9) | |
| Others | 14 (9.4) | 5 (5.8) | 3 (11.1) | |
| CEA level (ng/mL), n (%) | 0.127 | |||
| <5 | 84 (56.8) | 53 (60.9) | 10 (38.5) | |
| >5 | 64 (43.2) | 34 (39.1) | 16 (61.5) | |
| Tumor depth, n (%) | 0.075 | |||
| T1–3 | 126 (84.6) | 68 (78.2) | 18 (66.7) | |
| T4 | 23 (15.4) | 19 (21.8) | 9 (33.3) | |
| Lymph node metastasis, n (%) | 0.325 | |||
| Negative | 62 (41.6) | 44 (50.6) | 14 (51.9) | |
| Positive | 87 (58.4) | 43 (49.4) | 13 (48.1) | |
| Surgical approach, n (%) | <0.001 | |||
| Open | 11 (7.4) | 15 (17.2) | 9 (33.3) | |
| Laparoscopic | 138 (92.6) | 72 (82.7) | 18 (66.7) | |
| Adjuvant chemotherapy, n (%) | 0.108 | |||
| No | 78 (52.4) | 50 (57.5) | 20 (74.1) | |
| Yes | 71 (47.6) | 37 (42.5) | 7 (25.9) | |
| Complications (>grade 2 based on the CD classification system), n (%) | <0.001 | |||
| Absent | 137 (92.0) | 61 (70.1) | 14 (51.9) | |
| Present | 12 (8.0) | 26 (29.9) | 13 (48.1) |
BMI, body mass index; CEA, carcinoembryonic antigen; CAR, C-reactive protein-to-albumin ratio
Figure 4.
Overall survival (a) and progression-free survival (b) according to combined preoperative and postoperative CAR (*, p = 0.006; **, p = 0.013; ***, p < 0.001; ****, p = 0.041). CAR C-reactive protein-to-albumin ratio
Predictive factors of prognosis
Table 4 shows the correlations between clinicopathological factors and OS. According to the univariate analysis, age (p = 0.002), lymph node metastasis (p = 0.028), postoperative complications (p = 0.016), and combined preoperative and postoperative CAR (p < 0.001) were significantly associated with OS. Multivariate analysis revealed that age (p = 0.007), lymph node metastasis (p = 0.017), and combined preoperative and postoperative CAR (p = 0.013) were significantly correlated with OS. Table 5 shows the correlations between clinicopathological factors and PFS. According to the univariate analysis, age (p = 0.029), CEA level (p < 0.001), lymph node metastasis (p = 0.003), and combined preoperative and postoperative CAR (p = 0.005) were significantly associated with PFS. Multivariate analysis revealed that CEA levels (p = 0.005), lymph node metastasis (p = 0.005), and combined preoperative and postoperative CAR (p = 0.044) were significantly correlated with PFS.
Table 4.
Univariate and Multivariate Analyses of the Predictive Factors of Overall Survival.
| Characteristics of the participants | Univariate analysis | Multivariate analysis | |||||
|---|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | ||
| Sex (male) | 1.234 | 0.640–2.379 | 0.530 | ||||
| Age (>75 years) | 3.057 | 1.531–6.103 | 0.002 | 2.726 | 1.313–5.661 | 0.007 | |
| BMI (>25 mg/m2) | 1.478 | 0.730–2.992 | 0.278 | ||||
| Primary tumor location (rectum) | 1.427 | 0.627–3.250 | 0.397 | ||||
| Tumor size (>50 mm) | 1.063 | 0.551–2.052 | 0.856 | ||||
| Histological type (others) | 1.335 | 0.265–2.117 | 0.586 | ||||
| CEA level (>5 ng/mL) | 1.792 | 0.928–3.461 | 0.082 | 1.640 | 0.792–3.393 | 0.183 | |
| Tumor depth (T4) | 1.096 | 0.481–2.497 | 0.827 | ||||
| Lymph node metastasis (positive) | 2.258 | 1.093–4.666 | 0.028 | 2.520 | 1.181–5.376 | 0.017 | |
| Surgical approach (open) | 1.987 | 0.908–4.347 | 0.089 | 1.140 | 0.450–2.883 | 0.783 | |
| Adjuvant chemotherapy (no) | 1.525 | 0.784–2.967 | 0.214 | ||||
| Complications | 2.327 | 1.169–4.635 | 0.016 | 1.744 | 0.778–3.906 | 0.177 | |
| Combined CAR | 0.013 | ||||||
| Low CAR | 1.000 | Reference | <0.001 | 1.000 | Reference | ||
| Intermediate CAR | 2.731 | 1.276–5.845 | 0.010 | 2.638 | 1.153–6.033 | 0.022 | |
| High CAR | 7.025 | 3.017–16.357 | <0.001 | 3.897 | 1.470–10.330 | 0.006 | |
BMI, body mass index; CEA, carcinoembryonic antigen; CAR, C-reactive protein-to-albumin ratio
Table 5.
Univariate and Multivariate Analyses of the Predictive Factors of Progression-Free Survival.
| Characteristics of the participants | Univariate analysis | Multivariate analysis | |||||
|---|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | ||
| Sex (male) | 1.210 | 0.759–1.929 | 0.423 | ||||
| Age (>75 years) | 1.676 | 1.055–2.663 | 0.029 | 1.602 | 0.994–2.584 | 0.053 | |
| BMI (>25 mg/m2) | 1.403 | 0.838–2.347 | 0.198 | ||||
| Primary tumor location (rectum) | 1.145 | 0.617–2.127 | 0.668 | ||||
| Tumor size (>50 mm) | 1.145 | 0.716–1.830 | 0.571 | ||||
| Histological type (others) | 1.052 | 0.456–2.427 | 0.906 | ||||
| CEA level (>5 ng/mL) | 2.218 | 1.384–3.557 | <0.001 | 2.005 | 1.232–3.263 | 0.005 | |
| Tumor depth (T4) | 1.506 | 0.884–2.566 | 0.132 | ||||
| Lymph node metastasis (positive) | 2.151 | 1.304–3.547 | 0.003 | 2.069 | 1.242–3.447 | 0.005 | |
| Surgical approach (open) | 1.386 | 0.746–2.575 | 0.302 | ||||
| Adjuvant chemotherapy (No) | 1.126 | 0.708–1.790 | 0.616 | ||||
| Complications | 1.403 | 0.815–2.415 | 0.222 | ||||
| Combined preoperative and postoperative CAR | 0.005 | 0.044 | |||||
| Low CAR | 1.000 | Reference | 1.000 | Reference | |||
| Intermediate CAR | 1.480 | 0.887–2.468 | 0.133 | 1.640 | 0.974–2.762 | 0.063 | |
| High CAR | 2.901 | 1.525–5.518 | 0.001 | 2.130 | 1.079–4.201 | 0.029 | |
BMI, body mass index; CEA, carcinoembryonic antigen; CAR, C-reactive protein-to-albumin ratio
Discussion
The current study investigated the correlation between CAR and prognosis after curative resection for stage II/III CRC. Results showed that a high preoperative and postoperative CAR was correlated with poor prognosis. Furthermore, combined preoperative and postoperative CAR could stratify prognosis and better extract populations with poor or favorable prognosis. To the best of our knowledge, this is the first study that assessed the prognostic value of combined preoperative and postoperative CAR.
CAR is an inflammation-based marker calculated using CRP and serum albumin concentrations, which are easy-to-measure and cost-effective in clinical settings. This marker has been used in acute cases such as medical admission and death in patients with sepsis[18,19]. However, recently, CAR was found to have a prognostic value in different types of malignant tumors[11-15,20]. Although the actual mechanisms for the association between CAR and prognosis are not clearly understood, the possible explanations are as follows: First, elevated CRP levels are associated with poor prognosis in different types of malignancies[21-24]. CRP is an acute-phase protein synthesized in hepatocytes and is regulated by inflammatory cytokines such as interleukin-1, interleukin-6, and tumor necrosis factor[25]. These inflammatory cytokines promote tumor growth, invasion, and metastasis. Therefore, elevated CRP levels indicate the production of inflammatory cytokines and, consequently, poor prognosis. Second, decreased albumin levels are associated with poor prognosis[26-28]. Serum albumin, which is the most abundant protein in the plasma protein, represents host nutrition. Hypoalbuminemia reflects not only undernourishment but also nutritional impairment, which is associated with inflammatory and immunosuppressive conditions[29]. These conditions enhance cancer progression, and they are associated with decreased quality of life[27]. Moreover, if systemic inflammatory response occurs, which then causes decreased albumin levels, the CRP level increases[15]. Taken together, preoperative CAR is correlated with prognosis by reflecting not only inflammation but also immunonutritional status. As previously reported, the current study showed that high preoperative CAR was associated with poor prognosis.
Recent studies have focused on the association between not only preoperative but also postoperative inflammatory status and prognosis. The underlying mechanism for the effect of postoperative inflammation on cancer prognosis has not been validated yet. However, this phenomenon has possible explanations. First, surgery-related inflammation is involved in immunosuppression, prothrombotic change, and imbalance of pro-/anti-tumorigenic cytokine that influences tumor progression[30-32]. Changes in host status, which is favorable for the development of tumor triggered by inflammatory response, activates residual tumor cell growth during the postoperative period. Second, inflammatory response itself enhances cancer stem cell proliferation[2,33]. Inflammation-related tumor effects may worsen prognosis, and this then supports the notion that postoperative complication and elevated postoperative CRP levels are correlated with poor prognosis[34-39]. Based on these findings, we investigated the association between postoperative CAR and prognosis. Results showed that high postoperative CAR was associated with poor prognosis. Further, it was associated with old age, primary tumor location (rectum), open surgery, and postoperative complication. This finding suggests that high postoperative CAR could predict poor prognosis as a result of reflecting a prolonged higher inflammatory state induced by surgical stress, slow recovery, and additional stress.
Furthermore, combined preoperative and postoperative CAR could stratify prognosis. Notably, stratification is good in extracting poor and favorable prognosis. Based on preoperative CAR assessment alone, 200 patients presented with a low CAR and 63 with a high CAR. The 5-year OS rates were 88.0% and 69.5%, respectively. The 5-year-PFS rates were 75.4% and 55.4%, respectively. Meanwhile, according to the combined preoperative and postoperative CAR assessment, 149 patients were included in the low CAR group and 27 in the high CAR group. The 5-year OS rates were 89.0% and 57.1%, respectively. The 5-year-PFS rates were 76.5% and 47.2%, respectively. Increased discriminability may be achieved by adding an intermediate CAR group. Based on the low preoperative CAR and the high postoperative CAR, these patients developed inflammation postoperatively. Meanwhile, the high preoperative CAR and the low postoperative CAR suggested that the patients were free from tumor-related inflammation via tumor elimination. In fact, the intermediate CAR group is between the low and high CAR groups based on the Kaplan-Meier curves. Several studies have evaluated the association between changes in the levels of perioperative inflammation-based markers (including PNI, LMR, and NLR) and prognosis[5-10]. However, the current study first showed the association between combined preoperative and postoperative CAR and prognosis.
Previous reports have commonly evaluated postoperative markers at approximately 1 month after surgery. This could represent the host initial status without surgical stress and tumor influence. Murakami et al. and Miyatani et al., who evaluated the inflammatory status 1 month after surgery, showed a positive correlation between preoperative and postoperative markers in PNI (r = 0.59) and NLR (r = 0.38)[7,8], respectively. Even after tumor resection, the characteristics of postoperative inflammatory status may not differ significantly from the preoperative inflammatory status. In contrast, evaluation 7 days after surgery in the current study could strongly reflects the degree of surgical stress rather than nature of host status. The current study had only an extremely weak correlation between preoperative and postoperative CAR (r = 0.13). Hence, the combination of preoperative and postoperative CAR can be a combination of two factors with different characteristics. The best timing for evaluation or whether the timing differs based on the marker should still be determined. However, combining early postoperative CAR with preoperative CAR has good prognostic predictability. Predicting prognosis in the early postoperative period could be advantageous for subsequent treatments such as adjuvant chemotherapy selection.
Recent molecular biological approaches, such as ctDNA and immunoscoring, likely play an important role in predicting recurrence[40,41]. Particularly, research on ctDNA-based postoperative adjuvant chemotherapy selection is rapidly progressing[42-44]. Preclinical studies of these molecular approaches have determined the relationship between recurrence and its predictors. Conversely, as mentioned above, it is difficult to elucidate the mechanism by which CAR influences prognosis because CAR is considered a complex indicator of host immune and inflammatory status. However, considering that recurrence may result from multiple factors, such as host, tumor, and surgical factors, CAR may serve as a complement to molecular biological approaches. Furthermore, CAR is particularly useful in low-income countries due to its affordability.
The current study had several limitations. First, it was retrospective in nature, and a relatively small number of patients from a single center were included. Second, underlying diseases likely to affect blood test results, including liver cirrhosis, chronic renal failure, and infections, were not taken into consideration. Third, blood test on postoperative day 7 might not be the best for predicting prognosis. Yamamoto et al. showed that the highest CRP level after surgery had the highest area under the ROC curve and was closely associated with poor prognosis compared with other measurement timings[45]. Therefore, in the current study, the highest CRP during the postoperative period may be more reliable than that on postoperative day 7. Therefore, further investigations including the timing of postoperative blood test should be conducted to validate the study results.
Conclusion
Combined preoperative and postoperative CAR can be a prognostic predictor after curative resection in patients with CRC. Preoperative CAR may represent host immunonutritional status under tumoral conditions. Meanwhile, postoperative CAR can mainly reflect surgery-related stress. Combined preoperative and postoperative CAR can be obtained easily and a useful biological marker with consideration of long-term outcomes.
Conflicts of Interest
There are no conflicts of interest.
Author Contributions
K Tamai: study conception and design, acquisition of data, analysis and interpretation of data, drafting of manuscript; H Hirose: study conception and design, acquisition of data, analysis and interpretation of data, critical revision of manuscript; S Okamura: analysis and interpretation of data, drafting of manuscript; Y Akazawa: acquisition of data; M Koh: acquisition of data; K Hayashi: acquisition of data; Y Katsura: study conception and design, critical revision of manuscript; C Ebisui: analysis and interpretation of data, critical revision of manuscript; M Yano: critical revision of manuscript. All authors approved the final version for publication, agreed to be accountable for all aspects of the work, and ensure that any questions related to the accuracy or integrity of any part of the work will be appropriately investigated and resolved.
Approval by Institutional Review Board (IRB)
Approval code issued by the institutional review board: 2022GS-028
Name of the institution(s) that granted the approval
Suita Municipal Hospital
Acknowledgements
The authors would like to thank Enago (https://www.enago.jp) for English language editing.
References
- 1.Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a Cancer Journal for Clinicians. 2021 May; 71(3): 209-49. [DOI] [PubMed] [Google Scholar]
- 2.Coussens LM, Werb Z. Inflammation and cancer. Nature. 2002 Dec; 420(6917): 860-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Diakos CI, Charles KA, McMillan DC, et al. Cancer-related inflammation and treatment effectiveness. Lancet Oncol. 2014 Oct; 15(11): e493-503. [DOI] [PubMed] [Google Scholar]
- 4.Dupré A, Malik HZ. Inflammation and cancer: what a surgical oncologist should know. European Journal of Surgical Oncology. 2018 May; 44(5): 566-70. [DOI] [PubMed] [Google Scholar]
- 5.Ni HH, Lu Z, Huang X, et al. Combining pre- and postoperative lymphocyte-C-reactive protein ratios can better predict hepatocellular carcinoma prognosis after partial hepatectomy. Journal of Inflammation Research. 2022 Apr; 15: 2229-41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Song Q, Wu JZ, Jiang HF, et al. The postoperative lymphocyte to monocyte ratio change predicts poor clinical outcome in patients with esophageal squamous cell carcinoma undergoing curative resection. Disease Markers. 2020 Apr; 2020: 1451864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Miyatani K, Saito H, Kono Y, et al. Combined analysis of the pre- and postoperative neutrophil-lymphocyte ratio predicts the outcomes of patients with gastric cancer. Surgery Today. 2018 Mar; 48(3): 300-7. [DOI] [PubMed] [Google Scholar]
- 8.Murakami Y, Saito H, Kono Y, et al. Combined analysis of the preoperative and postoperative prognostic nutritional index offers a precise predictor of the prognosis of patients with gastric cancer. Surgery Today. 2018 Apr; 48(4): 395-403. [DOI] [PubMed] [Google Scholar]
- 9.Shibutani M, Maeda K, Nagahara H, et al. The prognostic significance of the postoperative prognostic nutritional index in patients with colorectal cancer. BMC Cancer. 2015 Jul; 15: 521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tamai M, Kiuchi J, Kuriu Y, et al. Clinical impact of postoperative prognostic nutritional index in colorectal cancer patients undergoing adjuvant chemotherapy. American Journal of Cancer Research. 2021 Oct; 11(10): 4947-55. [PMC free article] [PubMed] [Google Scholar]
- 11.Tamai K, Okamura S, Makino S, et al. C-reactive protein/albumin ratio predicts survival after curative surgery in elderly patients with colorectal cancer. Updates in Surgery. 2022 Feb; 74(1): 153-62. [DOI] [PubMed] [Google Scholar]
- 12.Liu Z, Jin K, Guo M, et al. Prognostic value of the CRP/Alb ratio, a novel inflammation-based score in pancreatic cancer. Annals of Surgical Oncology. 2017 Feb; 24(2): 561-8. [DOI] [PubMed] [Google Scholar]
- 13.Shibutani M, Maeda K, Nagahara H, et al. Prognostic significance of the preoperative ratio of C-reactive protein to albumin in patients with colorectal cancer. Anticancer Research. 2016 Mar; 36(3): 995-1001. [PubMed] [Google Scholar]
- 14.Ishizuka M, Nagata H, Takagi K, et al. Clinical significance of the C-reactive protein to albumin ratio for survival after surgery for colorectal cancer. Annals of Surgical Oncology. 2016 Mar; 23(3): 900-7. [DOI] [PubMed] [Google Scholar]
- 15.Kinoshita A, Onoda H, Imai N, et al. The C-reactive protein/albumin ratio, a novel inflammation-based prognostic score, predicts outcomes in patients with hepatocellular carcinoma. Annals of Surgical Oncology. 2015 Mar; 22(3): 803-10. [DOI] [PubMed] [Google Scholar]
- 16.Hashiguchi Y, Muro K, Saito Y, et al. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2019 for the treatment of colorectal cancer. International Journal of Clinical Oncology. 2020 Jan; 25(1): 1-42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Annals of Surgery. 2004 Aug; 240(2): 205-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fairclough E, Cairns E, Hamilton J, et al. Evaluation of a modified early warning system for acute medical admissions and comparison with C-reactive protein/albumin ratio as a predictor of patient outcome. Clinical Medicine. 2009 Feb; 9(1): 30-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ranzani OT, Zampieri FG, Forte DN, et al. C-reactive protein/albumin ratio predicts 90-day mortality of septic patients. PLOS ONE. 2013; 8(3): e59321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wu M, Zhou Y, Chen Q, et al. Prognostic role of pretreatment C-reactive protein to albumin ratio in urological cancers: a systematic review and meta-analysis. Frontiers in Oncology. 2022 Apr; 12: 879803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Falconer JS, Fearon KC, Ross JA, et al. Acute-phase protein response and survival duration of patients with pancreatic cancer. Cancer. 1995; 75(8): 2077-82. [DOI] [PubMed] [Google Scholar]
- 22.Wigmore SJ, Mcmahon AJ, Sturgeon CM, et al. Acute-phase protein response, survival and tumour recurrence in patients with colorectal cancer. British Journal of Surgery. 2001 Feb; 88(2): 255-60. [DOI] [PubMed] [Google Scholar]
- 23.Karakiewicz PI, Hutterer GC, Trinh QD, et al. C-reactive protein is an informative predictor of renal cell carcinoma-specific mortality: a European study of 313 patients. Cancer. 2007 Sep; 110(6): 1241-7. [DOI] [PubMed] [Google Scholar]
- 24.Allin KH, Nordestgaard BG. Elevated C-reactive protein in the diagnosis, prognosis, and cause of cancer. Critical Reviews in Clinical Laboratory Sciences. 2011 Aug; 48(4): 155-70. [DOI] [PubMed] [Google Scholar]
- 25.Mantovani A, Allavena P, Sica A, et al. Cancer-related inflammation. Nature. 2008 Jul; 454(7203): 436-44. [DOI] [PubMed] [Google Scholar]
- 26.Oñate-Ocaña LF, Aiello-Crocifoglio V, Gallardo-Rincón D, et al. Serum albumin as a significant prognostic factor for patients with gastric carcinoma. Annals of Surgical Oncology. 2007 Feb; 14(2): 381-9. [DOI] [PubMed] [Google Scholar]
- 27.Gupta D, Lis CG. Pretreatment serum albumin as a predictor of cancer survival: A systematic review of the epidemiological literature. Nutrition Journal. 2010 Dec; 9: 69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lai CC, You JF, Yeh CY, et al. Low preoperative serum albumin in colon cancer: a risk factor for poor outcome. International Journal of Colorectal Disease. 2011 Apr; 26(4): 473-81. [DOI] [PubMed] [Google Scholar]
- 29.Suzuki S, Shibata M, Gonda K, et al. Immunosuppression involving increased myeloid-derived suppressor cell levels, systemic inflammation and hypoalbuminemia are present in patients with anaplastic thyroid cancer. Molecular and Clinical Oncology. 2013 Nov; 1(6): 959-64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mualla NM, Hussain MR, Akrmah M, et al. The impact of postoperative complications on long-term oncological outcomes following curative resection of colorectal cancer (stage I-iii): a systematic review and meta-analysis. Cureus. 2021 Jan; 13(1): e12837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Onuma AE, Zhang H, Gil L, et al. Surgical stress promotes tumor progression: a focus on the impact of the immune response. Journal of Clinical Medicine. 2020 Dec; 9(12): 4096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chen Z, Zhang P, Xu Y, et al. Surgical stress and cancer progression: the twisted tango. Molecular Cancer. 2019 Sep; 18(1): 132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hirsch HA, Iliopoulos D, Struhl K. Metformin inhibits the inflammatory response associated with cellular transformation and cancer stem cell growth. Proceedings of the National Academy of Sciences. 2013 Jan; 110(3): 972-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Matsubara D, Arita T, Nakanishi M, et al. The impact of postoperative inflammation on recurrence in patients with colorectal cancer. International Journal of Clinical Oncology. 2020 Apr; 25(4): 602-13. [DOI] [PubMed] [Google Scholar]
- 35.Saito T, Kurokawa Y, Miyazaki Y, et al. Which is a more reliable indicator of survival after gastric cancer surgery: postoperative complication occurrence or C-reactive protein elevation? Journal of Surgical Oncology. 2015 Dec; 112(8): 894-9. [DOI] [PubMed] [Google Scholar]
- 36.Shibutani M, Nakao S, Maeda K, et al. Inflammation caused by surgical stress has a negative impact on the long-term survival outcomes in patients with colorectal cancer. Anticancer Research. 2020 Jun; 40(6): 3535-42. [DOI] [PubMed] [Google Scholar]
- 37.Lu J, Xu BB, Xue Z, et al. Perioperative CRP: a novel inflammation-based classification in gastric cancer for recurrence and chemotherapy benefit. Cancer Medicine. 2021 Jan; 10(1): 34-44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ito H, Are C, Gonen M, et al. Effect of postoperative morbidity on long-term survival after hepatic resection for metastatic colorectal cancer. Annals of Surgery. 2008 Jun; 247(6): 994-1002. [DOI] [PubMed] [Google Scholar]
- 39.Law WL, Choi HK, Lee YM, et al. The impact of postoperative complications on long-term outcomes following curative resection for colorectal cancer. Annals of Surgical Oncology. 2007 Sep; 14(9): 2559-66. [DOI] [PubMed] [Google Scholar]
- 40.Kasi PM, Fehringer G, Taniguchi H, et al. Impact of circulating tumor dna-based detection of molecular residual disease on the conduct and design of clinical trials for solid tumors. JCO Precision Oncology. 2022; 6: e2100181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Pagès F, Mlecnik B, Marliot F, et al. International validation of the consensus Immunoscore for the classification of colon cancer: a prognostic and accuracy study. Lancet. 2018; 391(10135): 2128-39. [DOI] [PubMed] [Google Scholar]
- 42.Taniguchi H, Nakamura Y, Kotani D, et al. CIRCULATE-Japan: circulating tumor DNA-guided adaptive platform trials to refine adjuvant therapy for colorectal cancer. Cancer Science. 2021; 112(7): 2915-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Tie J, Cohen JD, Lahouel K, et al. Circulating tumor dna analysis guiding adjuvant therapy in stage ii colon cancer. New England Journal of Medicine. 2022; 386(24): 2261-72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kotani D, Oki E, Nakamura Y, et al. Molecular residual disease and efficacy of adjuvant chemotherapy in patients with colorectal cancer. Nature Medicine. 2023; 29(1): 127-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Yamamoto M, Saito H, Uejima C, et al. Prognostic value of the combination of pre- and postoperative C-reactive protein in colorectal cancer patients. Surgery Today. 2018 Nov; 48(11): 986-93. [DOI] [PubMed] [Google Scholar]