Journal of Surgical Oncology 2014;109:234–238
Prognostic Significance of Lymph Node Status in Patients With Metastatic Colorectal Carcinoma Treated With Lymphadenectomy YAN-LONG LIU, MD,1 HAI-TAO XU,
2 MD,
SHI-XIONG JIANG, MD,1 YAN-MEI YANG, MD,3 1 AND BIN-BIN CUI, MD *
1 Department of Colorectal Surgery, The Affiliated 3rd Hospital of Harbin Medical University, Harbin, PR China Department of Hepatobiliary and Pancreatic Surgery, The Affiliated 3rd Hospital of Harbin Medical University, Harbin, PR China 3 Cancer Research Institute, Harbin Medical University, Harbin, PR China
2
Background and Objectives: To test prognostic significance of lymph node status in patients with metastatic colorectal carcinoma (mCRC). Methods: Four hundred ninety six patients diagnosed with synchronous mCRC and treated with lymphadenectomy between 1995 and 2008 were identified and divided into groups pN0, pN1, and pN2 (140 (28.2%) in pN0, 223 (45.0%) in pN1, and 133 (26.8%) in pN2 group) according to their lymph node status. The Kaplan–Meier and Cox regression analyses were used to test associations and independent predictor status of lymph node involvement. Results: The Cox proportional hazards regression showed pN as significantly associated with disease‐specific survival (DSS) both in univariate (HR ¼ 1.609, 95% CI 1.411 to 1.835, P < 0.001) and multivariate (HR ¼ 1.630, 95% CI 1.422 to 1.868, P < 0.001) analyses. The Kaplan–Meier analysis demonstrated that patients with pN2 and pN1 had a significantly worse DSS compared with patients with pN0 tumors (respectively, 17.273 1.020 and 27.145 1.715 vs. 34.992 2.143 months; P < 0.001). In accuracy analyses based on AUC values, nodal status demonstrated the highest accuracy (65.1%) out of all the variables. Conclusions: Our findings indicate that optimal TNM staging for mCRC should incorporate lymph node status to provide a more effective and predictive model.
J. Surg. Oncol.2014;109:234–238. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: lymph nodes; involvement; mCRC; prognosis; TNM; lymphadenectomy
INTRODUCTION TNM staging system adopted by the National Cancer Institute [1] and the American Society of Colon and Rectal Surgeons [2] remains the most important tool for prognosis in colorectal carcinoma (CRC). Up to now, TNM staging is constantly updated in order to increase the predictive accuracy by incorporation of validated favorable or unfavorable prognostic features into the overall assessment process [3]. Stage IV is the most advanced stage of TNM system and includes more complicated treatment strategies, such as conversion therapy, surgery, etc. However, the application of any treatment should prolong the overall patient survival. Thus, it is important to make the most accurate assessment of the patient’s prognosis [4,5]. Currently, stage IV simply includes patients with any T stage or any N stage, and at least one metastatic site (M1). The lymph node involvement, an important parameter, is not taken into account in the stage IV CRC [6]. Therefore, its introduction into more individualized staging is an important and widely debated topic. To date, only a limited significance is ascribed to the lymph node status in all prognostic models applicable to patients with CRC. Here, we decided to evaluate and quantify a prognostic significance of this variable in patients with synchronous metastatic CRC (mCRC). We hypothesized that inclusion of lymph node status may improve the accuracy of predications of cancer‐related mortality.
MATERIALS AND METHODS Patients This study included 496 consecutive patients with synchronous mCRC who had been treated with lymphadenectomy between January 1995 and December 2008. An institutional review board approval was
ß 2014 Wiley Periodicals, Inc.
obtained to collect patients’ data in a secure database and to report the outcomes. Patients who had received preoperative treatment that may have affected the status of lymph nodes were excluded from this study. The histologic type and staging were determined according to the international TNM staging system [1,2,6]. Lymphadenectomy was routinely performed including the inferior mesenteric, both common and internal iliac, and middle and inferior rectal lymph nodes, according to the standard protocol in our Department. The patients who could be taken care of with simple fistulization were excluded in order to keep data homogeneity. Resection was considered curative if all metastasis nodules could be completely
Grant sponsor: Hospital Research Program; Grant number: JJZ2011‐01.; Grant sponsor: Natural Science Foundation of Heilongjiang Province; Grant number: D201149. The authors declare no conflicts of interest. Yan‐Long Liu and Hai‐tao Xu contributed equally to this study and should be considered co‐first authors. Y.L.L. and H.T.X. developed study design and drafted the manuscript, S.X.J. carried out data acquisition, Y.M.Y. conducted statistical analysis, B.B.C. conceived the study, participated in developing study design, and coordinated the work. All authors read and approved the final manuscript. *Correspondence to: Bin‐Bin Cui, M.D., Department of Colorectal Surgery, The Affiliated 3rd Hospital of Harbin Medical University, 150 Haping Road, Nangang District, Harbin 150040, PR China. Fax: 86‐451‐86298786. E‐mail: [email protected] Received 31 May 2013; Accepted 07 October 2013 DOI 10.1002/jso.23479 Published online 21 January 2014 in Wiley Online Library (wileyonlinelibrary.com).
pN in Metastatic Colorectal Carcinoma removed. The patients were prospectively followed after surgery according to a postoperative surveillance program [1]. Patients with at least 12 assessed nodes were divided into 3 groups according to their lymph node status: pN0 (no regional lymph node metastasis), pN1 (metastasis in 1 to 3 lymph nodes), and pN2 (metastasis in 4 or more lymph nodes). Study outcomes included clinical and pathologic data of the primary tumor and data obtained by patients’ follow‐up (e.g., disease‐specific survival).
Statistical Analysis All variables were analyzed using descriptive statistics. Categorical variables were compared using chi‐square test. Disease‐specific survival (DSS) was modeled using Kaplan–Meier curves, and log‐rank tests were used to analyze the differences between survival curves. Univariate and multivariate survival analyses were carried out using the Cox regression model with DSS as the outcome measure. The P value of
Prognostic Significance of Lymph Node Status in Patients With Metastatic Colorectal Carcinoma Treated With Lymphadenectomy YAN-LONG LIU, MD,1 HAI-TAO XU,
2 MD,
SHI-XIONG JIANG, MD,1 YAN-MEI YANG, MD,3 1 AND BIN-BIN CUI, MD *
1 Department of Colorectal Surgery, The Affiliated 3rd Hospital of Harbin Medical University, Harbin, PR China Department of Hepatobiliary and Pancreatic Surgery, The Affiliated 3rd Hospital of Harbin Medical University, Harbin, PR China 3 Cancer Research Institute, Harbin Medical University, Harbin, PR China
2
Background and Objectives: To test prognostic significance of lymph node status in patients with metastatic colorectal carcinoma (mCRC). Methods: Four hundred ninety six patients diagnosed with synchronous mCRC and treated with lymphadenectomy between 1995 and 2008 were identified and divided into groups pN0, pN1, and pN2 (140 (28.2%) in pN0, 223 (45.0%) in pN1, and 133 (26.8%) in pN2 group) according to their lymph node status. The Kaplan–Meier and Cox regression analyses were used to test associations and independent predictor status of lymph node involvement. Results: The Cox proportional hazards regression showed pN as significantly associated with disease‐specific survival (DSS) both in univariate (HR ¼ 1.609, 95% CI 1.411 to 1.835, P < 0.001) and multivariate (HR ¼ 1.630, 95% CI 1.422 to 1.868, P < 0.001) analyses. The Kaplan–Meier analysis demonstrated that patients with pN2 and pN1 had a significantly worse DSS compared with patients with pN0 tumors (respectively, 17.273 1.020 and 27.145 1.715 vs. 34.992 2.143 months; P < 0.001). In accuracy analyses based on AUC values, nodal status demonstrated the highest accuracy (65.1%) out of all the variables. Conclusions: Our findings indicate that optimal TNM staging for mCRC should incorporate lymph node status to provide a more effective and predictive model.
J. Surg. Oncol.2014;109:234–238. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: lymph nodes; involvement; mCRC; prognosis; TNM; lymphadenectomy
INTRODUCTION TNM staging system adopted by the National Cancer Institute [1] and the American Society of Colon and Rectal Surgeons [2] remains the most important tool for prognosis in colorectal carcinoma (CRC). Up to now, TNM staging is constantly updated in order to increase the predictive accuracy by incorporation of validated favorable or unfavorable prognostic features into the overall assessment process [3]. Stage IV is the most advanced stage of TNM system and includes more complicated treatment strategies, such as conversion therapy, surgery, etc. However, the application of any treatment should prolong the overall patient survival. Thus, it is important to make the most accurate assessment of the patient’s prognosis [4,5]. Currently, stage IV simply includes patients with any T stage or any N stage, and at least one metastatic site (M1). The lymph node involvement, an important parameter, is not taken into account in the stage IV CRC [6]. Therefore, its introduction into more individualized staging is an important and widely debated topic. To date, only a limited significance is ascribed to the lymph node status in all prognostic models applicable to patients with CRC. Here, we decided to evaluate and quantify a prognostic significance of this variable in patients with synchronous metastatic CRC (mCRC). We hypothesized that inclusion of lymph node status may improve the accuracy of predications of cancer‐related mortality.
MATERIALS AND METHODS Patients This study included 496 consecutive patients with synchronous mCRC who had been treated with lymphadenectomy between January 1995 and December 2008. An institutional review board approval was
ß 2014 Wiley Periodicals, Inc.
obtained to collect patients’ data in a secure database and to report the outcomes. Patients who had received preoperative treatment that may have affected the status of lymph nodes were excluded from this study. The histologic type and staging were determined according to the international TNM staging system [1,2,6]. Lymphadenectomy was routinely performed including the inferior mesenteric, both common and internal iliac, and middle and inferior rectal lymph nodes, according to the standard protocol in our Department. The patients who could be taken care of with simple fistulization were excluded in order to keep data homogeneity. Resection was considered curative if all metastasis nodules could be completely
Grant sponsor: Hospital Research Program; Grant number: JJZ2011‐01.; Grant sponsor: Natural Science Foundation of Heilongjiang Province; Grant number: D201149. The authors declare no conflicts of interest. Yan‐Long Liu and Hai‐tao Xu contributed equally to this study and should be considered co‐first authors. Y.L.L. and H.T.X. developed study design and drafted the manuscript, S.X.J. carried out data acquisition, Y.M.Y. conducted statistical analysis, B.B.C. conceived the study, participated in developing study design, and coordinated the work. All authors read and approved the final manuscript. *Correspondence to: Bin‐Bin Cui, M.D., Department of Colorectal Surgery, The Affiliated 3rd Hospital of Harbin Medical University, 150 Haping Road, Nangang District, Harbin 150040, PR China. Fax: 86‐451‐86298786. E‐mail: [email protected] Received 31 May 2013; Accepted 07 October 2013 DOI 10.1002/jso.23479 Published online 21 January 2014 in Wiley Online Library (wileyonlinelibrary.com).
pN in Metastatic Colorectal Carcinoma removed. The patients were prospectively followed after surgery according to a postoperative surveillance program [1]. Patients with at least 12 assessed nodes were divided into 3 groups according to their lymph node status: pN0 (no regional lymph node metastasis), pN1 (metastasis in 1 to 3 lymph nodes), and pN2 (metastasis in 4 or more lymph nodes). Study outcomes included clinical and pathologic data of the primary tumor and data obtained by patients’ follow‐up (e.g., disease‐specific survival).
Statistical Analysis All variables were analyzed using descriptive statistics. Categorical variables were compared using chi‐square test. Disease‐specific survival (DSS) was modeled using Kaplan–Meier curves, and log‐rank tests were used to analyze the differences between survival curves. Univariate and multivariate survival analyses were carried out using the Cox regression model with DSS as the outcome measure. The P value of
