A prognosis for stage IV colorectal cancer is generally poor. As a result, the development of an appropriate treatment strategy for each individual with this disease within a limited time frame is important. Few studies have been made of C-reactive protein (CRP) in stage IV cases of colorectal cancer, so it is unclear whether CRP is a useful prognostic marker for this disease. Thus, the purpose of this study was to clarify the relationship between the preoperative CRP level and the prognosis of stage IV colorectal cancer.
Between April 2007 and December 2015, a total of 384 patients with stage IV colorectal cancer who underwent primary resection were included. Patients were divided into high (HCG) and low (LCG) CRP groups based on a preoperative CRP cutoff value of 1.0 mg/dL. Postoperative short- and long-term results were examined retrospectively.
The 5-year survival rate was 24.6% for HCG and 36.7% for LCG, indicating the survival rate for HCG was lower. The study was limited to patients who were unable to undergo R0 surgery. Preoperative CEA levels were higher in HCG, whereas the postoperative chemotherapy induction rate was lower. HCG also showed a significantly lower survival rate than LCG. Multivariate analysis showed that CRP levels above 1.0 mg/dL, poorly differentiated histopathology, and the absence of chemotherapy were risk factors affecting overall survival.
These results suggest that the preoperative CRP level may be a useful biomarker for the prognosis of incurable stage IV colorectal cancer.
Treatment outcomes for stage IV colorectal cancer are generally poor. For example, in unresectable colorectal cancer, the median survival time without drug therapy is about 8 months.1 Recent advances in drug therapy have extended median survival time to more than 30 months, although this disease remains difficult to cure in most patients.2–4 Because of a limited time with which to act, it is extremely important to formulate an appropriate treatment strategy. However, to date, the prognostication of each patient with stage IV colorectal cancer is thought to be difficult.
A high C-reactive protein (CRP) value in a preoperative examination is suggested to be a poor prognostic factor in various malignant tumors such as gastric,5 lung,6 pancreatic,7,8 and colon cancers.9 However, most of these reports are on radically treated cases, with few reports on stage IV cases.10–14 It is therefore unclear whether the CRP level can be a useful prognostic marker for stage IV colorectal cancer. In addition, stage IV colorectal cancer has a wide variety of medical conditions, including the metastatic organs involved, the number of metastases, and whether it is curable or incurable. The purpose of this study is to clarify the relationship between preoperative CRP levels and the prognosis of stage IV colorectal cancer.
Patients and Methods
Between April 2007 and December 2015, a total of 384 patients with stage IV colorectal cancer underwent primary resection at Saitama Medical University International Medical Center. All study participants provided their informed consent. The study design was approved by the Ethics Committee of Saitama Medical University International Medical Center (no. 19-149). All patients were monitored using a standard Japanese postoperative surveillance program. Patients were divided into either a high CRP group (HCG) with preoperative CRP values of 1.0 mg/dL, or a low CRP group (LCG). Operative outcomes included surgical approach, operating time, blood loss, whether or not an R0 surgical resection was possible, and histopathologic diagnosis. We also assessed postoperative complications, and postoperative hospital stays as short-term clinical outcomes. Five-year cancer-specific and overall survival rates served as long-term oncologic outcomes. In this study, short- and long-term outcomes in HCG and LCG were compared retrospectively.
All statistical analyses were performed using the SPSS software package (SPSS version 25: IBM, Tokyo, Japan). For statistical analysis, we performed χ2 and Mann-Whitney U tests to examine differences between the two groups. The cumulative survival rate was analyzed with Kaplan-Meier and log-rank tests. P < 0.05 was considered a statistically significant difference.
Results
Patient backgrounds are shown in Table 1. There were 107 cases of HCG and 277 cases of LCG. HCG consisted of 73 male and 34 female patients, whereas LCG was made up of 172 males and 105 females. The average ages in each group were 67 and 65 years, respectively. There were no significant differences between HCG and LCG in terms of age, sex, the presence or absence of double or multiple cancers, preoperative American Society of Anesthesiologists physical status (ASA-PS), body mass index (BMI), and history of abdominal surgery. Only preoperative carcinoembryonic antigen (CEA) values were significantly higher in HCG. Table 2 shows short-term postoperative outcomes. A significant difference between HCG and LCG in the rate of laparoscopic surgery, operation time, blood loss, postoperative complication rate, and postoperative hospital stay was not observed. A difference in histopathologic findings between these 2 groups was also not found. However, the number of cases with an R0 surgical resection of distant metastasis was significantly smaller in HCG. Figure 1 shows a Kaplan-Meier survival curve. The 3-year survival rate was 33.9% for HCG and 49.7% for LCG, whereas the 5-year survival rate was 24.6% for HCG and 36.7% for LCG. This indicated a significantly lower survival rate for HCG (P < 0.001). Multivariate analysis of factors affecting the survival rate showed that a CRP value 1.0 (hazard ratio [HR], 0.718; 95% confidence interval [CI], 0.54–0.954; P = 0.022), poor histopathologic type (HR, 0.458; 95% CI, 0.314–0.667; P < 0.001), positive venous invasion (HR, 0.602; 95% CI, 0.389–0.931; P = 0.023), and whether an R0 surgical resection was undertaken (HR, 0.19; 95% CI, 0.132–0.274; P < 0.001) were risk factors (Table 3).
In the present study, an R0 surgical resection was performed in 122 cases (31.8%). Eighteen cases (16.8%) had an R0 resection in HCG, and 104 cases (37.5%) in LCG. The proportion of R0 resections was significantly higher in LCG. The background and short-term postoperative outcomes of patients with an R0 resection are shown in Tables 4 and 5. A significant difference between HCG and LCG in background factors, including preoperative CEA value, was not noted. In terms of surgical treatment, the proportion of laparoscopic surgeries was lower in HCG. Accordingly, the amount of estimated blood loss was also larger in HCG. However, a significant difference was not noted between the 2 groups in regard to the postoperative complication rate, the length of postoperative hospital stays, and histopathologic evaluation. As for long-term survival, there was no significant difference in overall survival between the 2 groups in patients with an R0 resection (Fig. 2).
We also examined only those patients who underwent palliative surgery. A total of 262 patients were unable to undergo R0 surgical resection in this series. Similar to the results of the overall stage IV study, many cases existed in HCG that showed a high CEA level within the patient background. Differences did not exist concerning age, sex, ASA–PS, BMI, and other variables listed in Table 6. In postoperative short-term results, as shown in Table 7, postoperative chemotherapy was introduced to 67.4% of patients in HCG and 83.8% of patients in LCG (P = 0.002). Comparing overall survival rates, HCG had significantly lower survival rates than LCG (Fig. 3). Multivariate analysis showed that CRP values above 1.0 mg/dL, poorly differentiated histopathology, and the absence of chemotherapy were risk factors affecting overall survival (Table 8).
Discussion
CRP is an acute-phase protein that is synthesized and secreted by the liver and is recognized as part of the inflammatory response.15 CRP is useful for predicting complications after the intestinal resection of malignant tumors and in the prognosis of patients who have undergone surgical resection of malignant tumors.7–9,16–19 Increased CRP is observed in many conditions such as infection, inflammation, malignant tumors, and trauma; previous studies have suggested a link between inflammation and cancer. In addition to CRP, other indicators of inflammation include the Glasgow prognostic score, calculated from CRP and albumin, the neutrophil-to-lymphocyte ratio, interleukin-6, matrix metalloproteinase-9, and other variables; their relationship with the degree of malignant tumor progression and prognosis has been described.20,21 Other studies have demonstrated that the CEA value correlates with the prognosis of colorectal cancer.22,23 In the present study, patients in HCG with palliative surgery showed a significantly high CEA. However, HCG patients who underwent an R0 surgical resection did not show a significantly high CEA, and CEA was not found to be a prognostic factor in our study.
In short-term outcomes, univariate analysis revealed that a small number of people could perform an R0 surgical resection with HCG. In stage IV colorectal cancer, the ability to perform R0 surgery is also a prognostic factor.24 Few cases exist in which an R0 surgical resection is possible in HCG, resulting in a poor prognosis. In addition to the high CRP value in multivariate analysis, the inability to perform an R0 surgical resection was also a factor that reduced the survival rate.
In this study, HCG patients with palliative surgery showed a significantly lower chemotherapy induction rate. Prechemotherapy CRP levels help predict prognosis in patients with colorectal cancer receiving oxaliplatin-based chemotherapy. Chemotherapy is the most important treatment for unresectable colorectal cancer, suggesting that the low induction rate of chemotherapy in HCG may contribute to the decrease in survival rate observed.
Our results indicate that preoperative CRP levels were not useful for a subsequent prognosis in patients with an R0 surgical resection. Therefore, it is important to aim for an R0 surgical resection, regardless of the CRP level, if possible. If curative surgical resection cannot be performed in patients with high preoperative CRP levels and a poor prognosis, it is important to consider treatment strategies. CRP levels before treatment may be a useful biomarker for prudent decision-making if difficulty exists in performing an R0 surgical resection in patients with stage IV colorectal cancer.
Conclusion
These results suggest that the preoperative CRP level may be a useful biomarker for the prognosis of incurable stage IV colorectal cancer.
Acknowledgments
The authors thank the experts at BioMed Proofreading for English copy editing.