The pretreatment Controlling Nutritional Status (CONUT) score is an independent prognostic factor in patients undergoing resection for colorectal cancer

The Controlling Nutritional Status (CONUT) score is a marker of nutrition and is associated with poor survival in various kinds of cancers. However, no reports have yet compared risk factors for colorectal cancer recurrence using a nutritional index. We assessed the predictive value of the CONUT score compared with the modified Glasgow Prognostic Score (mGPS) and Prognostic Nutritional Index (PNI) in colorectal cancer (CRC) patients. We performed a retrospective cohort study of the medical records of 336 consecutive patients with stage I-I I I CRC who underwent curative resection at a single institution in 2012–2017. Univariate and multivariate analyses were conducted to identify prognostic factors associated with relapse-free survival (RFS) and overall survival (OS). The low CONUT score group exhibited higher RFS and longer OS compared to the high CONUT score group (82.2% vs. 63.3%, p = 0.002 and 95.5% and 86.2%, p = 0.005, respectively). The Akaike’s information criterion values of each index for RFS and OS were superior in CONUT score (723.71 and 315.46, respectively) compared to those of PNI (726.95 and 316.52) and mGPS (728.15 and 318.07, respectively). The CONUT score was found to be a good predictor of RFS and OS in patients with resectable CRC.


AIC (Akaike's information criterion) of each index model. The AIC values of each index for RFS
were 723.713 for the CONUT score, 726.948 for the PNI, 728.146 for the mGPS. According to this comparison, the CONUT score had the best goodness-of-fit, followed by the PNI and mGPS. In OS, the AIC values of each index for RFS were 315.459 for the CONUT score, 316.516 for the PNI, 318.074 for the mGPS; thus, the CONUT score had the best goodness-of-fit, followed by the PNI and mGPS.
correlation between albumin, lymphocyte count and cRp. The correlation between albumin concentrations and CRP and the correlation between albumin concentrations and TLC were compared. As a result, CRP was R = 0.436, TLC was R = 0.15579, and CRP was more correlated with albumin concentrations (Fig. 4).     www.nature.com/scientificreports/

Discussion
We evaluated and compared the prognostic impacts of the CONUT score, the PNI, and the mGPS in 301 patients with colorectal cancer, and our analyses revealed that only the CONUT score was an independent prognostic factor for RFS and OS, and it was superior to the other nutritional-based markers in terms of predictive ability for prognosis before initial treatment.
The CONUT score is known as a nutritional assessment index. Recently, the CONUT score has been reported to help estimate postoperative recurrence rates and predict the prognosis of patients with gastrointestinal (GI) cancer. The CONUT score is composed of three common factors: the serum albumin level, which is a nutritional index, the peripheral lymphocyte count, and the total cholesterol concentration. Serum albumin has been reported to correlate with tumor necrosis because inflammatory cytokines reduce albumin synthesis 9 . As a result, hypoalbuminemia is reported to be a risk factor for the increased prevalence of high-grade colorectal cancer 10 . Peripheral lymphocytes, which play an important role in the immune response to tumors, are known to indicate an individual's immunological and nutritional status 11 . We reported that an elevated neutrophil-tolymphocyte ratio on postoperative day 7 is a significant independent predictor of reduced RFS for patients with stage II colorectal cancer 12 . Lymphocytes, the main cells of the body's immunity, make an immune response to tumor cells, so a reduction in lymphocytes leads to a reduced ability to block tumors in the body. Therefore, high NLR levels are effective for tumor survival and lead to poor prognosis. Hypercoagulability of blood is leading for metastasis of malignant tumors 13,14 . The total cholesterol concentration is known as an indicator of a patient's reserve calories 15 . Other studies indicated that low serum cholesterol levels were associated with a poorer prognosis in patients with various cancers [16][17][18] . It remains unclear why low serum cholesterol levels are associated with a poor prognosis. The total cholesterol levels have been reported to correlate with tumor progression as tumor tissue reduces the body's plasma cholesterol levels and caloric intake 19 . Therefore, hypocholesterolemia is not considered to be the cause of cancer but is thought to be caused by cancer 18,20 .
We observed herein that the CONUT score was associated with age, pT category, and early complications. Elderly patients often have poor nutritional status compared to younger people. The number of early complications after surgery may have been affected by our patients' poor nutritional status.
The PNI, which is the nutritional index calculated using the peripheral lymphocyte count and the serum albumin level, was reported to be associated with the survival of CRC patients 21 . The GPS has been used in several different oncological field and could, due to the routine assessment of albumin and CRP, be assessed in all aspects of the prospective care of patients with cancer 22,23 . In the present study, although the PNI and the CONUT score were found to be predictive factors for RFS and OS in the univariate analyses, the results of the multivariate analysis demonstrated that only the CONUT score was an independent prognostic factor for these patients with CRC.
In addition, the CONUT score more accurately predicted the survival of the CRC patients than the PNI and the mGPS. Although the CONUT score. PNI, and mGPS have common factors for nutrition, they led to different results. We thus investigated the reasons why the CONUT score was superior to the PNI and mGPS in predicting the patients' prognoses. Notably, an additional factor evaluated in the CONUT score which is not included in the mGPS or the PNI is the total cholesterol concentrations. The mGPS is composed of albumin and CRP, and PNI is composed of serum albumin concentrations and TLC. Serum albumin concentrations was included in all scores, and the correlation between albumin and TLC and CRP was lower in TLC (Fig. 4). It may be diagnosing a factor associated with recurrence that is different from serum albumin concentrations. It was considered that the diagnostic value increased because the cholesterol variable was further increased in the group separated by serum albumin concentrations and TLC.
Our study is that post operative early comlications were risk factors of recurrence. It has been reported that the risk of recurrence increases with anastomosis leakage, which is one of the postoperative complications 24 . There are also reports that perioperative blood transfusion blood and subsequent development of postoperative infectious complications may be associated with a poor prognosis 25 . In our study, 8 cases of anastomosis leakage and 7 cases of early postoperative complications accompanied by blood transfusion were considered to have influenced this.
This study has its limitations. First, this study was retrospective in design and included patients from a single institution. Secondly, patients underwent a variety of invasive surgical procedures that resulted in higher mortality and morbidity, which this study did not consider. Third, there is no consensus regarding the CONUT score cut-off value, and this makes it difficult to use the CONUT score in clinical settings. We selected the CONUT cut-off value herein by performing an ROC analysis. The CONUT score is a non-specific marker of nutrition, which implies that another systemic disease can affect the CONUT score. Our findings need further review and validation in more CRC patients.
conclusion Pretreatment CONUT score in CRC patients was demonstrated to be an independent predictor of RFS and OS. It was superior to PNI and mGPS in predicting the prognosis of patients who underwent curative surgery for CRC. The cost of measuring the CONUT score is very low and can easily be obtained using laboratory data from routine clinical practice. We recommend that patients calculate their CONUT score regularly before undergoing curative surgery. This could be a useful indicator for prognosis in CRC. www.nature.com/scientificreports/ Informed consent was obtained from all participants, the reporting of our research is in accordance with the STROBE guidelines 27 . The variables assessed included patient age, sex, tumor location, histological grade, TNM (tumor, nodes, metastasis) stage, personal medical history, preoperative laboratory data, early complications, and follow-up status. The CONUT score and the PNI were calculated on based on the patient's preoperative laboratory data using. Early complications included anastomotic leakage surgical site infection, colitis, ileus, urinary tract infection, respiratory infection, and dysuria.
The following patients were excluded: (a) those who received adjuvant chemotherapy, (b) those with multiple primary malignancies , and (c) those with familial adenomatous polyposis or Lynch syndromes. Histopathological,clinical and laboratory data were obtained from the patients and blood sampling test was conducted within 1-2 weeks prior to the surgery. Total of 35 patients (neoadjuvant therapy 7, adenomatous polyposis 2, multiple primary malignancies 5 and data not available 21) were excluded (Fig. 5). conUt score. Each patient's CONUT score was calculated using his/her data of serum albumin, total peripheral lymphocyte count, and total cholesterol concentrations, based on a previous report that used preoperative serum samples 28 . Albumin concentrations ≥ 3.5, 3.5 > and > 3.0, 2.99 > and ≥ 2.5, and < 2.5 g/dL were scored as 0, 2, 4, and 6 points, respectively; (2) total lymphocyte count ≥ 1,600, 1599-1,200, 1,199-800, and < 800/ mm 3 were scored as 0, 1, 2, and 3 points, respectively; and (3) total cholesterol concentrations ≥ 180, 140-179, 100-139, and < 100 mg/dL were scored as 0, 1, 2, and 3 points, respectively. The CONUT score was defined as the sum of items (1), (2), and (3). The CONUT score thus ranges from 0 to 12, with higher scores indicating a worse nutritional status. mGPS score. The GPS score is as follows: hypoalbuminemia (< 35 g/L) and the elevated C-reactive protein (CRP) (> 10 mg/L) was given a score of 2. Hypoalbuminemia (< 35 g/L) or CRP > 10 mg/L was score of 1. Albumin concentrations ≥ 3.5 g/dL and CRP ≤ 10 mg/L was score of 0 9 . However, hypoalbuminemia alone was not associated with reduced survival, so mGPS was created. The mGPS score 2 is hypoalbuminemia (< 35 g/L) and the elevated CRP (> 10 mg/L), score 1 was the elevated CRP (> 10 mg/L), score 0 is hypoalbuminemia (< 35 g/L) or serum albumin level (> 35 g/L) and the serum CRP lebel (< 10 mg/L) 29 . pni. The PNI, which is calculated using serum albumin and the peripheral lymphocyte count, is a simple and useful score for predicting the prognosis of patients with various cancers 30 . The PNI was calculated using the following formula: PNI = serum albumin level (g/dL) + 5 × total lymphocyte count 21 . Onodera reported that this index provided an accurate, quantitative estimate of operative risk 31 . In general, resection and anastomosis of the gastrointestinal tract can be safely practiced when the index is > 45. The same procedure may be dangerous when the index is between 45 and 40. When the PNI is < 40, this surgery may be contraindicated 31 . In other studies, the cutoff value of PNI was 40, which is common, We thus set the cut-off value for the PNI as 40 21,32,33 .

AIC (Akaike information criterion).
The AIC is a popular method for comparing the adequacy of multiple, possibly nonnested models 34 . A statistic that evaluates the predictability of a statistical model using the difference between the observed value and the theoretical value. The smaller the value, the better the fit 35 .
Follow-up. Surgical resection was defined as radical when there was no evidence of the tumor clearance and distant metastasis was both histologically and macroscopically complete. The patients were followed up every 3 months for the first 3 years, every 6 months for the next 2 years, and once annually thereafter. At each followup, all of the patients underwent a physical examination as well as the measurement of serum carcinoembryonic antigen (CEA) and CA19-9 (carbohydrate antigen . They also underwent full colonoscopies at 1-2 years after the surgery (retum cancer was every 1 year after surgery). Chest-abdominal computed tomography scans www.nature.com/scientificreports/ were generally obtained every 6 months. Recurrence was defined as the emergence of a radiological, clinical, and/or pathological diagnosis of cancer cells locally or distant from the original position.
Statistical analysis. Relapse-free survival (RFS) and overall survival (OS) were calculated from the date of the patient underwent surgery to that of recurrence or death, using the Kaplan-Meier method. A Cox regression analysis was performed to identify factors that are significantly associated with RFS or OS. Probability (p)-values ≤ 0.05 were considered significant. The Pearson product-moment correlation coefficient was used for the bivariate correlation. All statistical analyses were performed using JMP 14 software (SAS, Cary, NC, USA). We conducted a receiver operating characteristic (ROC) curve analysis to determine the cut-off values for the CONUT score. At each value, an ROC curve was created by plotting the sensitivity and specificity of each result under study. The score closest to the point with both maximum sensitivity and specificity was chosen as the cutoff value, and the largest number of tumors was correctly classified for clinical outcome. There was a minor difference between the optimal cut-off values of RFS and those of OS, and we therefore used the cut-off values of the RFS for the OS in order to maintain consistency and prevent confusion.
The clinicopathological factors examined were as follows: sex, age, cancer location site (right side vs. left side), histology [tub 1 and tub 2 vs. others (por and pap etc.)], preoperative CONUT score, mGPS score, PNI, presence/ absence of vascular invasion, presence/absence of lymph invasion pT category (T1, T2 vs. ≥ T3), pN category, tumor size, preoperative CEA level, preoperative CA19-9 level, and early complications. ethical approval. The present study was conducted in accord with the Declarations of Helsinki and was approved by the Ethics Committee of the Teikyo University (approval date, 23 August 2016, Registration Number; 16-032).