Optimization Of N Staging System For Colorectal Cancer

PART ⅠThe prognostic impact of number-indexes of lymph nodes in stage Nlc colorectal cancer patients:To subdivide the Nlc categoryBackground:Total number of lymph node(TLN),number of negative lymph node(NLN)and log odds of positive lymph nodes(LODDS)were considered as relatively good prognostic indicators of colorectal cancer.However,the prognostic value of TLN,NLN and LODDS in stage N1c colorectal patients remained unknown.Purpose:This study was aimed to evaluate the prognostic impact of number-indexes of lymph nodes in stage N1c colorectal cancer(CRC)patients.Methods:The clinicopathologic data of CRC patients with stage pTxN1cM0 who initially underwent radical surgery in Cancer Hospital,Chinese Academy of Medical Sciences/National Cancer center(NCC)and patients with stage pTxN0-2bM0 who initially underwent radical surgery in the Surveillance,Epidemiology and End Results(SEER)database from January 2010 to December 2015 were retrospectively analyzed.The prognostic significance of number-indexes of lymph nodes in stage N1c CRC patients were explored.We used the X-tile software to determine the optimal cut-off value of numberindexes of lymph nodes.The survival curves were draw by Kaplan-Meier method and we used log-rank test to compared the survival difference.Univariate and multivariate analysis of prognostic factors were performed by using Cox proportional hazard regression model.Results:A total of 70320 CRC patients with stage pTxN0-2bM0 were enrolled in this study,and 1165 patients were stage N1c,including 85 patients of NCC cohort and 1080 patients of SEER cohort.According to TLN or NLN,stage N1c patients were divided into two groups:stage N1c1(TLN or NLN≥13)and stage N1c2(TLN or NLN<13).In the NCC cohort,the 5-year overall survival(OS)rates of stage N1c1 and N1c2 patients were 80.9%and 53.3%,respectively(P=0.002).In the SEER cohort,the 5-year OS rate was 64.7%for stage N1c1 patients compared with 52.2%for stage N1c2 patients(P<0.001).Based on the value of LODDS,stage N1c patients were divided into two groups:LODDS1(LODDS≤1.43)and LODDS2(LODDS>-1.43).In the NCC cohort,the 5-year OS rate was 80.9%for patients in the group LODDS1,compared with 53.3%for those in the group LODDS2(P=0.002).In the SEER cohort,the 5-year OS rate was 64.7%for patients in the LODDS1 group,compared with 52.2%for those in the LODDS2 group(P<0.001).In the SEER cohort,there was no significantly survival difference between stage N1c1 and N1b patients,the 5-year OS rate was 64.7%and 66.6%,respectively(P=0.902).Moreover,no significantly survival difference was also found between stage N1c2 and N2a patients,the 5-year OS rate was 52.2%and 58.1%,respectively(P=0.058).The results of Cox univariate and multivariate analysis both demonstrated that TLN or NLN and LODDS were associated with OS significantly in both the NCC cohort and SEER cohort.Conclusion:TLN or NLN and LODDS were all independent prognostic factors of OS.The clinicians could use these number-indexes of lymph nodes for prognostic stratification and make the different adjuvant therapeutic schemes for CRC patients with stage Nlc.PART ⅡA comprehensive evaluation of lymph node staging and to subdivide the N2b category in colorectal cancerBackground:In colorectal cancer,≥7 lymph nodes(LN)metastases were currently defined as stage N2b,but there was considerable heterogeneity in the prognosis of patients with stage N2b.Objective:We aimed to subdivide the N2b category in CRC patients based on the number of metastatic LNs.Methods:The clinicopathologic data of CRC patients with stage pTxN2bM0 who initially underwent radical surgery in Cancer Hospital,Chinese Academy of Medical Sciences/National Cancer center(NCC)and patients with stage pTxN0-2bM0-1 in the Surveillance,Epidemiology and End Results(SEER)database between January 2010 and December 2015 were retrospectively analyzed.We used the X-tile software to determine the optimal cut-off value of the number of positive lymph node(PLN).The survival curves were draw by using Kaplan-Meier method.Log-rank test was used to evaluate the survival difference.Univariate and multivariate analysis of prognostic factors were performed by using Cox proportional hazard regression model.Results:A total of 68335 CRC patients were enrolled in this study,including 240 patients of NCC cohort;65189 stage M0 patients and 2906 stage M1 patients of SEER cohort.The optimal cut-off value of PLN was 13 which was performed by the X-tile software.According to PLN,stage N2b patients were divided into two group’s:stage N2b#(7≤PLN<13)and stage N3(PLN≥13).In the NCC cohort,the 5-year overall survival(OS)rates of stage N2b#and N3 patients were 66.0%and 45.7%,respectively(P<0.001).In the SEER cohort,the 5-year cancer-specific survival(CSS)rate was 57.1%for stage N2b#patients compared with 40.2%for stage N3 patients;the 5-year OS rates of stage N2b#and N3 patients were 52.1%and 35.3%,respectively(both P<0.001).The results of multivariate Cox analysis demonstrated that optimized stage pN(the stage pN after subdividing the stage N2b)was the independent prognosis factor of OS in the NCC cohort(HR=1.869,95%CI:1.253-2.787,P=0.002);and optimized stage pN was also the independent prognosis indicator of CSS in the SEER cohort(N3:N0,HR=8.170,95%CI:7.298-9.146,P<0.001).There was no survival difference between TxN3M0 and TxN02b#M0(5-year CSS rate:40.2%vs 30.1%,P=0.050;5-year OS rate:35.3%vs 27.8%,P=0.358).Conclusion:Clinically,we could use PLN for prognostic stratification in stage N2b colorectal cancer patients.The prognosis of colorectal patients with stage N2b#(7≤PLN<13)and stage N3(PLN≥13)was significantly different,and different intensities of adjuvant therapy should be adopted.PART ⅢThe minimum number of lymph nodes examined for rectal cancer with neoadjuvant therapyBackground:Neoadjuvant therapy decreased the number of examined lymph nodes(LNs)and positive lymph nodes after the radical resection of rectal cancer.However,for patients with or without neoadjuvant therapy,the current American Joint Committee on Cancer(AJCC)guideline both recommended that 12 or more LNs should be examined in rectal cancer radical surgery.Purpose:The purpose of this study was to identify the minimum number of lymph nodes examined for rectal cancer patients with neoadjuvant therapy.Methods:We retrospectively collected and analyzed the clinicopathologic data of rectal cancer patients with NCRT and received radical resection in Cancer Hospital,Chinese Academy of Medical Sciences/National Cancer center(NCC)from January 2004 to December 2015 and patients who were diagnosed as rectal cancer and received neoadjuvant therapy followed by surgery in the Surveillance,Epidemiology,and End Results(SEER)database between January 2010 and December 2015.The NCC and SEER database was as the development cohort and validation cohort,respectively.The optimal cut-off value of the examined LNs was determined by the software of X-tile.Kaplan-Meier method and log-rank test were used to evaluate the prognosis.And we used Cox regression analysis to identify the risk factors of prognosis.Results:A total of 6634 patients were enrolled in this study,including 391 patients of NCC cohort and 6243 patients of SEER cohort.In the NCC cohort,there was no survival difference between patients with<12 examined LNs and patients with≥12 examined LNs,the 5-year disease-free survival(DFS)rate was 72.0%and 76.5%,respectively(P=0.099).Furthermore,the optimal cut-off value of examined LNs was 7 which was determined by the X-tile software.The 5-year DFS rate was 65.1%for patients with<7 examined LNs compared with 76.3%for those with≥7 examined LNs in the NCC cohort(P=0.028).In the multivariate Cox analysis,the number of examined LNs(the cut-off value was 7)was the independent prognostic factor of DFS[hazard ratio(HR)=3.255,95%confidence interval(CI):1.796-5.897,P<0.001].In the SEER cohort,the 5-year cancer-specific survival(CSS)rate was 81.2%for patients with<12 examined LNs compared with 83.1%for those with≥12 examined LNs(P=0.095).The rate of 5-year CSS was 79.1%for patients with<7 examined LNs compared with 83.0%for those with≥7 examined LNs(P=0.037).The results of multivariate Cox analysis demonstrated that the number of examined LNs(the cut-off value was 7)was the independent prognostic factor associated with CSS(HR=1.606,95%CI:1.308-1.970,P<0.001).Conclusion:For rectal cancer patients with NCRT follow by radical surgery,the minimum number of LNs examined might be 7,and the current standard(≥12 LNs should be examined)might be inappropriate.