Preoperative transferrin level is a novel prognostic marker for colorectal cancer

Abstract Aim This study investigated whether preoperative serum transferrin, a rapid‐turnover protein, was associated with prognosis after colorectal cancer (CRC) resection. Methods We evaluated preoperative transferrin, which was calculated as iron and unsaturated iron‐binding capacity, in 501 patients who underwent surgery for Stage I–III CRC. Transferrin level was directly proportional to total iron‐binding capacity (TIBC), and TIBC < 250 μg/dl was defined as low transferrin. The associations between transferrin and prognosis were evaluated in univariate and multivariate Cox proportional hazards analyses. Results Fifty‐eight of 501 patients (11.5%) had low transferrin. In these patients, low transferrin was significantly associated with high age, female gender, low body mass index (<18.5), high white blood cell count, low total protein, low albumin, high C‐reactive protein, low hemoglobin, and low neutrophil/lymphocyte ratio. In the univariate analysis, low transferrin was associated with shorter relapse‐free survival (RFS) (hazard ratio [HR] 2.180, 95% confidence interval [CI] 1.417‐3.354, P < .001), overall survival (OS) (HR 2.930, 95% CI 1.784‐4.811, P < .001), and cancer‐specific survival (CSS) (HR 2.122, 95% CI 1.053‐4.275, P = .035). In multivariate analysis, high age (P < .001), Glasgow Prognostic Score (P = .009), and low transferrin (HR 2.336, 95% CI 1.173‐4.654, P = .011) were independently associated with shorter OS, and depth of invasion pT4 (P = .015), presence of lymph node metastasis (P = .001), low hemoglobin (P = .034), and low transferrin (HR 2.638, 95% CI 1.113‐5.043, P = .025) were independently associated with shorter CSS. Conclusions Preoperative serum transferrin in Stage I–III CRC patients was identified as a novel prognostic marker by univariate and multivariate analyses.


| INTRODUC TI ON
Colorectal cancer (CRC) is the third most commonly diagnosed malignant disease in men and the second in women worldwide. 1 The most effective treatment for resectable CRC is surgical resection with lymph node (LN) dissection and adjuvant chemotherapy after surgery in Stage III patients, 2 although many patients, even in this group, suffered from recurrence or cancer-specific mortality after curative surgery and adjuvant chemotherapy.
Prognostic factors have been analyzed to predict CRC patient survival, to investigate novel perioperative strategies, and follow-up with the aim of improving CRC patients' survival. Inflammatory and nutritional parameters were associated with patient prognosis in various cancers. 3,4 Many inflammatory and nutritional parameters have been analyzed and scoring systems were constructed to estimate the prognosis in CRC patients. 5 C-reactive protein (CRP), albumin, lymphocytes, and cholesterol are measured in usual clinical practice, and these parameters have been evaluated using the Glasgow Prognostic Score, which consists of CRP and albumin and predicts the survival of CRC patients 6 ; nutritional scoring systems, prognostic nutritional index, 7 and controlling nutritional status 8 have also been associated with prognosis for cancer patients.
Transferrin, prealbumin, and retinal-binding protein are rapid-turnover proteins. The half-life of albumin is 21 days, but the half-lives of transferrin, prealbumin, and retinal-binding protein are 7 days, 1.9 days, and 12 hours, respectively 9 . These parameters were measured after surgery to estimate real-time nutritional status, 10 but because they were not usually measured before surgery, the association between preoperative values for these parameters and the short-term or long-term outcome of CRC patients who underwent surgery remains unclear. On the other hand, transferrin level before surgery can be estimated, because transferrin is directly proportional to total iron-binding capacity. 9 The aim of this study was to evaluate the relationship between preoperative serum transferrin status and the short-term and longterm outcomes in Stage I-III CRC patients who underwent colorectal resection. The results provide a novel insight into the association between preoperative transferrin status and prognosis of CRC patients, and we propose that transferrin should be measured to evaluate preoperative nutritional status and the need for nutritional support.  11 Transferrin was estimated by adding the value of iron to unsaturated iron-binding capacity. 12 The lower limit of the normal range of total iron-binding capacity (TIBC) was 250 μg/dl, and TIBC < 250 μg/dl was therefore defined as low transferrin. Clinical data, including age, gender, body mass index, depth of invasion (pT), presence of metastatic LN (pN), pathological type, lymphatic invasion, and vascular invasion, were retrospectively available for 501 patients. Laboratory measurements included carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), white blood cells (WBC), serum total protein, albumin, CRP, hemoglobin, platelets, peripheral neutrophils, and lymphocytes. Each cutoff value was defined based on the recommendations of the measuring kits our institute adopted. The cut-off values of hemoglobin in males and females were 13.0 and 12.0 g/dl, respectively, which were defined based on the World Health Organization's definitions. Neutrophil-tolymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR), as inflammatory markers, were evaluated because the association between these markers and metastatic CRC patient prognosis has been reported. Cut-off values of NLR and PLR were determined as 5 and 150, respectively, in accordance with previous studies. 5 Glasgow prognostic score (GPS) was defined based on the presence of hypoalbuminemia (<35 g/l) and elevated CRP (>10 mg/l): if both were abnormal, the score was 2; if either one or the other were abnormal, the score was 1; if neither were abnormal, the score was 0 in accordance with a previous report. 13 Complication after surgery was classified in accordance with the Clavien-Dindo classification. 14 Written informed consent was obtained from all the patients for the treatments.

| Treatment strategy and follow-up evaluation
The treatment strategy followed the Japanese colorectal cancer guidelines, 2 which recommend surgery with LN dissection for Stage I-III CRC. Patients were followed up at 3-month intervals.
Recurrence was confirmed by clinical examinations, including computed tomography (CT). Tumor marker levels were measured every 3 months for 5 years after surgery. CT scanning studies that included the neck to the pelvis were performed at least twice a year for 3 years after surgery.

| Statistical analysis
The association of transferrin with recorded clinical and pathological characteristics was determined by chi-squared and Fisher's exact tests. All P values were two-sided; P < .05 was considered significant.
Analysis of risk factors for survival included age, gender, depth of invasion, LN metastasis, hemoglobin, GPS, NLR, PLR, and transferrin. Mortality was estimated from relapse-free survival (RFS), overall survival (OS), and cancer-specific survival (CSS). The log-rank test was used in the survival analysis; the Kaplan-Meier method was used to assess cumulative survival. Cox proportional hazards regression models were utilized to calculate hazard ratio (HR) and 95% confidence interval (CI). We performed multivariate Cox proportional hazards

| Association between GPS and survival
GPS was reported as a prognostic factor after colorectal surgery. In our study, the CRC patients with preoperative GPS 1 and GPS 2 had shorter PFS (P < .001), OS (P < .001), and CSS (P = .014) than those with preoperative GPS 0 (Supplementar y

| Association of transferrin with blood transfusion and postoperative complication
The association of the prognosis of CRC patients after surgery with blood transfusion and postoperative complication was previously reported. 16,17 Preoperative and intraoperative blood transfusions were more frequently performed for patients with low transferrin compared with those with normal transferrin (P = .0124 and 0.0063, respectively). Postoperative complication was evaluated with the Clavien-Dindo classification. The transferrin level was not significantly associated with anastomotic leakage (P = .4609); however, patients with low transferrin tended to suffer from postoperative complication with Clavien-Dindo classification ≥ 3 compared to those with normal transferrin (P = .0696) ( Table 5 and Figure 2).  Five-year relapse-free survival rates of hypoalbuminemia and normal albumin patients were 78.9% and 73.5%, respectively (HR 1.28, 95% CI 1.04-1.56, P = .020), and five-year OS rates were 78.0%
The study indicated that albumin was strongly associated with OS. 19 CRP and albumin were prognostic factors for CRC patients in another study. 20 Several scoring systems have shown that these inflammatory and nutritional factors are associated with CRC patient prognosis. One of these, the Glasgow Prognostic Score (GPS), is based on the combination of CRP and serum albumin levels. CRC patients with elevated GPS or modified GPS were significantly associated with shorter OS (HR 2.20, 95% CI 1.88-2.57, P < .001) and shorter CSS (HR 1.86, 95% CI 1.59-2.17, P < .001) in a meta-analysis. 21 In our study, GPS was evaluated as an inflammatory and nutritional marker in multivariate analysis, because GPS includes albumin and CRP. GPS was associated with RFS and OS, although it was not independently associated with CSS. However, low transferrin was independently associated with shorter CSS as well as shorter RFS and OS in multivariate analysis.
Transferrin is a rapid-turnover protein and is synthesized in the liver. It is one of the nutrition markers. A preoperative immunonutrition pharmaceutics diet and prebiotics in patients with gastrointestinal cancer increased immunoglobulin as well as transferrin. Nutritional status may be tightly associated with immunity.
Improving nutritional status and immunity reduce the incidence of postoperative complications and infections. 22,23 The preoperative serum transferrin level is a possible predictive marker of postoperative pneumonia after esophageal surgery. 24 Serum transferrin was also predictive of spontaneous closure in patients with gastrointestinal cutaneous fistulas. 25 These reports suggested that transferrin may be a surrogate marker of immunity and wound healing as well as nutritional status. Furthermore, the IL-6, IL-8, VEGF-A, and midkine cytokines were elevated in cachectic patients in gastroesophageal cancers. Transferrin was decreased in these cachectic patients. 26 TA B L E 5 Association between transferrin and blood transfusion and postoperative complications

F I G U R E 2
The association between postoperative complication and preoperative transferrin, in accordance with the Clavien-Dindo classification