ORIGINAL ARTICLE. and treatment of patients

Transcription

1 ONLINE FIRST ORIGINAL ARTICLE Predictors of Lymph Node Count in Colorectal Cancer Resections Data From US Nationwide Prospective Cohort Studies Teppei Morikawa, MD, PhD; Noriko Tanaka, PhD; Aya Kuchiba, PhD; Katsuhiko Nosho, MD, PhD; Mai Yamauchi, PhD; Jason L. Hornick, MD, PhD; Richard S. Swanson, MD; Andrew T. Chan, MD, MPH; Jeffrey A. Meyerhardt, MD, MPH; Curtis Huttenhower, PhD; Deborah Schrag, MD, MPH; Charles S. Fuchs, MD, MPH; Shuji Ogino, MD, PhD, MS(Epidemiology) Objective: To identify factors that influence the total and negative lymph node counts in colorectal cancer resection specimens independent of pathologists and surgeons. Design: We used multivariate negative binomial regression. Covariates included age, sex, body mass index, family history of colorectal carcinoma, year of diagnosis, hospital setting, tumor location, resected colorectal length (specimen length), tumor size, circumferential growth, TNM stage, lymphocytic reactions and other pathological features, and tumor molecular features (microsatellite instability, CpG island methylator phenotype, long interspersed nucleotide element 1 [LINE-1] methylation, and BRAF, KRAS, and PIK3CA mutations). Setting: Two US nationwide prospective cohort studies. Patients: Patients with rectal and colon cancer (N=918). Main Outcome Measures: The negative and total node counts (continuous). Results: Specimen length, tumor size, ascending colon location, T3N0M0 stage, and year of diagnosis were positively associated with the negative node count (all P.002). Mutation of KRAS might also be positively associated with the negative node count (P=.03; borderline significance considering multiple hypothesis testing). Among node-negative (stages I and II) cases, specimen length, tumor size, and ascending colon location remained significantly associated with the node count (all P.002), and PIK3CA and KRAS mutations might also be positively associated (P=.03 and P=.049, respectively, with borderline significance). Conclusions: This molecular pathological epidemiology study shows that specimen length, tumor size, tumor location, TNM stage, and year of diagnosis are operator-independent predictors of the lymph node count. These crucial variables should be examined in any future evaluation of the adequacy of lymph node harvest and nodal staging when devising individualized treatment plans for patients with colorectal cancer. Arch Surg. Published online April 16, doi: /archsurg Author Affiliations are listed at the end of this article. THE PRESENCE OF LYMPH NODE metastasis has important implications in the prognosis and treatment of patients with colorectal cancer. 1 Observational studies indicate that the number of lymph nodes assessed by pathological examination (in particular, the negative node count) is associated with longer survival in colorectal cancer. 2 Thus, along with disease stage and tumor molecular features, the node count is often used for treatment decision making by oncologists. However, the optimal number of lymph nodes that must be assessed remains controversial. 2-5 Although the average number of lymph nodes evaluated for colorectal cancer has increased in the past decade, it remains uncertain how the node count is influenced by demographic, clinical, and tumor molecular factors. 6,7 The number of recovered lymph nodes may be influenced not only by surgeons and pathologists but also by factors independent of surgeons and pathologists. Those operator-independent factors include tumor location, disease stage, tumor size, host immune response, 8 and tumor molecular features, such as microsatellite instability (MSI) and the CpG island methylator phenotype (CIMP). 9 Molecular features of colorectal cancer and host immune response have been associated with the node count. 9,10 Previous studies have examined the relationship between the recovered node count and various demographic and clini- E1

2 cal features in population-based and hospitalbased studies, but all those studies 11-26,28,29 lacked comprehensive data on specimen length, tumor size, host immune reaction to tumor, and tumor molecular features. Beyond surgeon- and pathologist-related (ie, operator) factors, it is important to identify patient-specific node count predictors (ie, clinical, pathological, or tumor molecular factors) to assess the adequacy of lymph node examination for each patient. To accomplish this aim, a comprehensive database of a large number of colorectal cancer cases with clinical, specimen, pathological, and molecular annotations is needed. We therefore conducted this molecular pathological epidemiology 8,30,31 study using a database of 918 colorectal cancer cases in 2 prospective cohort studies. Considering the overall distribution of the node count, we used negative binomial regression analysis to identify factors associated with the negative and total node counts. Because we used a US nationwide cohort database with clinical, specimen, pathological, and tumor molecular variables (including MSI, CIMP, and KRAS (HGNC 6407), BRAF (HGNC 1097), and PIK3CA (HGNC 8975) mutations), we could assess each node count predictor independent of operator (surgeon and pathologist) factors. METHODS STUDY POPULATION We used the databases of 2 prospective cohort studies: the Nurses Health Study (consisting of women who have been followed up since 1976) and the Health Professionals Follow-up Study (consisting of men who have been followed up since 1986). 32,33 Every 2 years, participants have been sent follow-up questionnaires to update information on potential risk factors and to identify newly diagnosed cancer in themselves and their first-degree relatives. For nonresponders, we searched the National Death Index to discover deaths and to ascertain the causes of death and any diagnosis of cancer. Study physicians reviewed medical records, including pathology reports, and recorded tumor location and pathological TNM (tumor-node-metastasis) stage, the positive and negative node counts, 9 tumor size, circumferential growth along the bowel wall, and resected colorectal length (specimen length). We collected paraffin-embedded tissue blocks from hospitals where patients underwent tumor resections. 33 We collected diagnostic biopsy specimens for patients with rectal cancer who received preoperative treatment to avoid artifacts or bias introduced by treatment. Based on the availability of data on the node count and tumor molecular features, we included a total of 918 colorectal cancer cases diagnosed up to Hospitals where our participants underwent colorectal resections were distributed throughout the United States (Figure 1). Informed consent was obtained from all study subjects. This study was approved by the Human Subjects Committees at Brigham and Women s Hospital and the Harvard School of Public Health. HISTOPATHOLOGICAL EVALUATIONS Tissue blocks from all colorectal cancer cases were evaluated by a pathologist (S.O.). Tumor differentiation was categorized as poor vs well-moderate ( 50% vs 50% glandular areas). The presence and extent of mucinous and/or a signet ring cell component were recorded. Lymphocytic reaction patterns, such as peritumoral lymphocytic reaction and tumor infiltrating lymphocytes, were examined as previously described. 34 A subset of cases (n 100) was reviewed by another pathologist (T.M.), and concordance was as follows: =0.72 for tumor differentiation, Spearman r=0.87 for the percentage of mucin, Spearman r=0.65 for the percentage of signet ring cells, and Spearman r=0.65 for the summation score of peritumoral reaction and tumor infiltrating lymphocytes. PYROSEQUENCING OF KRAS, BRAF, AND PIK3CA AND MSI ANALYSIS We extracted DNA from each tumor, and polymerase chain reaction (PCR) analysis and pyrosequencing targeted for KRAS (codons 12 and 13), 35 BRAF (codon 600), 36 and PIK3CA (exons 9 and 20) 37 were performed. The MSI status was determined using D2S123, D5S346, D17S250, BAT25, BAT26, BAT40, D18S55, D18S56, D18S67, and D18S ,39 We defined MSI-high as the presence of instability in 30% or more of the markers and microsatellite stability/msi-low as instability in 0% to 29% of the markers. ANALYSES OF 2 TYPES OF METHYLATION Sodium bisulfite treatment on DNA and real-time PCR (Methy- Light) assays were validated and performed. 40 We quantified promoter methylation in 8 CIMP-specific markers (CACNA1G [HGNC 1394], CDKN2A [HGNC 1787], CRABP1 [HGNC 2338], IGF2 [HGNC 5466], MLH1 [HGNC 7127], NEUROG1 [HGNC 7764], RUNX3 [HGNC 10473], and SOCS1 [HGNC 19383]). 38,41,42 We defined CIMP-high as 6 or more methylated markers and CIMP-low/0 as 0 to 5 methylated markers according to the previously established criteria. 38 To accurately quantify relatively high long interspersed nucleotide element 1 (LINE-1) methylation levels, we used pyrosequencing. 43,44 STATISTICAL ANALYSIS No. of cases >100 Figure 1. Number of cases analyzed in this study by state. Our patients were distributed throughout the United States, and our results would not have been influenced by any particular surgeon or pathologist, thus increasing the generalizability of our findings. We used SAS software (version 9.1.3; SAS Institute, Inc) for statistical analysis. All P values were 2-sided. Because of multiple hypothesis testing, a P value for significance was adjusted conservatively by Bonferroni correction to.0023 (P=.05/ 22). The 2 test was used to assess the association between categorical variables, and analysis of variance was used to compare continuous variables across categories. E2

3 A No. of Cases B No. of Cases Negative Node Count Total Node Count C Spearman r = P < Specimen Length, cm Negative Node Count Spearman r = 0.21 P <.001 D Negative Node Count Poisson distribution Gamma-Poisson distribution Tumor Size, mm Figure 2. Lymph node counts in 918 colorectal cancers in our 2 US nationwide prospective cohort studies. A, Distribution of the negative node count. B, Distribution of the total node count. Both negative and total node counts approximately follow a gamma-poisson like distribution. In the boxplot above each graph, the vertical line in the middle of each box indicates the median, the diamond indicates the mean, and the left and right borders of the box mark the 25th and 75th percentiles, respectively. The whiskers extending from the left and right ends of the box mark the 5th and 95th percentiles, respectively. The points beyond the whiskers are outliers beyond the 5th or 95th percentile. C, Correlation between the negative node count and specimen length. D, Correlation between the negative node count and tumor size. We adopted multivariate negative binomial regression analysis to assess predictors of the node count because the marginal distribution of the total or negative node count fit the gamma- Poisson like distribution (Figure 2); overdispersion occurred with Poisson generalized linear models. The process of estimation was based on a negative binomial distribution that can be conceptualized as a mixture of a Poisson distribution and a gamma distribution. 45 Variables initially included in a model were sex, age (continuous), prediagnosis body mass index (continuous), family history of colorectal cancer in any firstdegree relative, year of diagnosis (continuous), hospital setting (academic vs nonacademic), tumor size (continuous), specimen length (continuous), circumferential growth (100% complete vs incomplete), tumor location and TNM stage (categorized as in Table 1 and Table 2), tumor differentiation (poor vs well-moderate), mucinous component (reported as a continuous percentage), signet ring cells (reported as a continuous percentage), peritumoral lymphocytic reaction and tumor infiltrating lymphocytes (absent/minimal vs mild vs moderate vs marked; ordinal), MSI (MSI-high vs microsatellite stability/msi-low), CIMP (CIMP-high vs CIMP-low/0), LINE-1 methylation (continuous), and KRAS, BRAF, and PIK3CA (mutations vs wild-type variants). A backward elimination with a threshold of P=.1 was performed to select variables in the final model except for TNM stage and tumor location, for which all categories were forced into the model. For cases with missing data in a covariate, we carried out 2 separate analyses: the first analysis included all patients, with creation of a categorical indicator for missing responses (missing indicator method; the second analysis included all patients, with missing responses imputed (multiple imputation [MI]). The MI procedure in SAS was used to perform 20 imputations of all variables with missing cases by using the regression method. The results from the regression analysis were then appropriately combined by using the MIANALYZE procedure. RESULTS LYMPH NODE COUNT IN COLORECTAL CANCER RESECTION The total node count showed a skewed (gamma-poisson like) distribution (Figure 2): range, 0 to 54; mean, 12.0; median, 10; and interquartile range, 6 to 16 nodes. The negative node count also showed a skewed (gamma- Poisson like) distribution: range, 0 to 54; mean, 10.5; median, 8; and interquartile range, 4 to 15 nodes. NODE COUNT AND CLINICAL, PATHOLOGICAL, AND MOLECULAR FEATURES Tables 1 and 2 show the clinical, pathological, and molecular features of colorectal cancers according to quartiles of the negative or total node count. The negative and total node counts were both positively associated with specimen length, tumor size, ascending colon location, T3N0M0 stage, MSI status, and CIMP status (all P). MULTIVARIATE ANALYSIS FOR THE NODE COUNT In a multivariate negative binomial regression model, factors independently associated with the negative node count included specimen length, tumor location, TNM stage, year of diagnosis, and tumor size (all P.002) (Table 3). In Table 3, for example, specimens with rectal cancer on average yielded a negative node count of approximately two-thirds (0.67) of that in specimens with ascending colon cancer after controlling for the effects of other variables. A KRAS mutation appeared to be a predictor of the negative node count (P=.03), although multiple hypothesis testing should be considered and the finding confirmed by an independent data set. Factorsindependentlyassociatedwiththetotalnodecount included specimen length, tumor location, TNM stage, and tumor size after controlling for the effects of other variables (all P) (Table 4). A KRAS mutation appeared to be a predictor of the total node count (P=.009), although multiple hypothesis testing should be considered. NODE COUNT IN NODE-NEGATIVE (STAGES I AND II) COLORECTAL CANCER RESECTIONS In patients with stage I or II colorectal cancer, ascending colon tumor location, tumor size, and specimen length were positively associated with the node count after controlling for the effects of other variables (all P.002) (Table 5). Mutations of PIK3CA (P=.03) and KRAS (P=.049) also appeared to predict the node count, although multiple hypothesis testing should be considered. E3

4 Table 1. Demographic and Clinical Features of Colorectal Cancer According to Quartiles of the Negative Node Count and Total Node Count a No. of Negative Lymph Nodes Total No. of Recovered Lymph Nodes Clinical or Molecular Feature Sex All Cases (N = 918) 0-4, Quartile 1 (n = 237) 5-8, Quartile 2 (n = 226) 9-14, Quartile 3 (n = 233) 15, Quartile 4 (n = 222) P Value b 0-6, Quartile 1 (n = 247) 7-10, Quartile 2 (n = 240) 11-15, Quartile 3 (n = 195) 16, Quartile 4 (n = 236) P Value b Male, HPFS 455 (50) 121 (51) 118 (52) 108 (46) 108 (49) 134 (54) 116 (48) 89 (46) 116 (49).60 Female, NHS 463 (50) 116 (49) 108 (48) 125 (54) 114 (51) 113 (46) 124 (52) 106 (54) 120 (51).32 Age, mean (SD), y 67.6 (8.4) 67.3 (7.9) 67.1 (8.2) 67.8 (9.0) 68.2 (8.5) (8.0) 67.5 (8.2) 67.8 (9.4) 67.8 (8.2).50 BMI (40) 91 (39) 79 (37) 102 (45) 90 (41) 91 (38) 93 (40) 85 (45) 93 (40) (42) 108 (46) 97 (45) 87 (39) 86 (39) (45) 94 (41) 78 (41) 97 (42) (17) 34 (15) 40 (19) 37 (16) 43 (20) 41 (17) 45 (19) 25 (13) 43 (18) Family history of colorectal cancer in first-degree relative Negative 718 (80) 193 (83) 161 (74) 183 (81) 181 (83) 194 (81) 174 (75) 154 (82) 196 (84).08 Positive 177 (20) 40 (17) 56 (26) 43 (19) 38 (17) 47 (20) 59 (25) 34 (18) 37 (16).07 Hospital setting Nonacademic 836 (91) 220 (93) 201 (89) 218 (94) 197 (89) 226 (91) 223 (93) 180 (92) 207 (88).14 Academic 82 (9) 17 (7) 25 (11) 15 (6) 25 (11) 21 (9) 17 (7) 15 (8) 29 (12).20 Year of diagnosis Before (38) 99 (42) 91 (40) 87 (37) 74 (33) 94 (38) 103 (43) 74 (38) 80 (34) (37) 89 (38) 96 (42) 75 (32) 77 (35) (41) 85 (35) 66 (34) 84 (36) (25) 49 (21) 39 (17) 71 (30) 71 (32) 51 (21) 52 (22) 55 (28) 72 (31) Resected colorectal specimen length, cm (43) 114 (56) 103 (49) 93 (42) 59 (27) 121 (55) 112 (50) 70 (39) 66 (29) (53) 85 (42) 100 (48) 122 (55) 144 (67) 93 (42) 107 (48) 102 (57) 149 (66) (4) 4 (2) 7 (3) 7 (3) 13 (6) 8 (4) 6 (3) 6 (3) 11 (5) Tumor size, mean 4.4 (2.0) 3.9 (1.8) 4.1 (1.8) 4.7 (2.2) 5.0 (2.1) 3.6 (1.7) 4.4 (2.1) 4.6 (1.9) 5.0 (2.0) (SD), cm Tumor location Cecum 162 (18) 30 (13) 41 (18) 40 (17) 51 (23) 35 (14) 43 (18) 31 (16) 53 (22) Ascending colon 183 (20) 28 (12) 32 (14) 53 (23) 70 (32) 21 (8.5) 41 (17) 46 (24) 75 (32) Transverse colon 98 (11) 23 (10) 20 (9) 30 (13) 25 (11) 22 (9) 27 (11) 24 (13) 25 (32) Descending colon 64 (7) 18 (8) 22 (10) 13 (6) 11 (5) 23 (9) 19 (8) 12 (6) 10 (4) Sigmoid colon 228 (25) 73 (31) 63 (28) 55 (24) 37 (17) 77 (31) 60 (25) 48 (25) 43 (18) Rectum 178 (20) 64 (27) 47 (21) 40 (17) 27 (12) 69 (28) 48 (20) 31 (16) 30 (13) Circumferential growth Incomplete 729 (79) 204 (86) 182 (81) 177 (76) 166 (75) 219 (89) 188 (78) 150 (77) 172 (73) % Complete 189 (21) 33 (14) 44 (19) 56 (24) 56 (25) 28 (11) 52 (22) 45 (23) 64 (27) Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HPFS, Health Professionals Follow-up Study; NHS, Nurses Health Study. a Unless otherwise indicated, data are given as number (percentage) indicating the proportion of cases with a specific clinical, pathological, or molecular feature among a given quartile category of the negative or total node count. Denominators vary because of missing data. b P values were calculated by analysis of variance for age, tumor size, and long interspersed nucleotide element 1 methylation and by 2 test for all other variables. Because of multiple hypothesis testing, a P value for significance was adjusted by Bonferroni correction to COMMENT We conducted this study to identify clinical, pathological, and tumor molecular variables that predict the node count in colorectal cancer resections independent of human operator factors (ie, surgeons and pathologists). We found that specimen length, tumor size, T3N0M0 stage, ascending colon tumor location, and year of diagnosis were positively associated with the negative node count. Mutation of KRAS might also be positively associated with the negative node count, but this finding should be confirmed by an independent data set. These results indicate that operator-independent variables influence the node count in colorectal resection and should be examined in any future study that assesses the adequacy of lymph node harvesting and staging. Comprehensive assessment of clinical, pathological, and molecular features is important in cancer research Previous studies have reported that the recovered node count is positively associated with specimen length, 14,18,19,21-24 proximal colon cancer, 13,18,19,25,28 larger tumor size, 11,14,18,20 the number of vascular pedicles, 50 and higher disease stage. 17,18 However, those studies 11,13,14,17-24,28 produced no data on host immune re- E4

5 Table 2. Pathological and Molecular Features of Colorectal Cancer According to Quartiles of the Negative Node Count and Total Node Count a No. of Negative Lymph Nodes Total No. of Recovered Lymph Nodes Clinical or Molecular Feature TNM stage All Cases (N = 918) 0-4, Quartile 1 (n = 237) 5-8, Quartile 2 (n = 226) 9-14, Quartile 3 (n = 233) 15, Quartile 4 (n = 222) P Value b 0-6, Quartile 1 (n = 247) 7-10, Quartile 2 (n = 240) 11-15, Quartile 3 (n = 195) 16, Quartile 4 (n = 236) P Value b T1N0M0 75 (9) 32 (15) 19 (9) 13 (6) 11 (5) 42 (19) 15 (7) 9 (5) 9 (4) T2N0M0 144 (17) 36 (16) 36 (17) 39 (18) 33 (15) 56 (25) 30 (13) 29 (16) 29 (13) T3N0M0 255 (30) 28 (13) 62 (30) 71 (32) 94 (44) 47 (21) 74 (32) 52 (29) 82 (36) T4N0M0 21 (2) 5 (2) 3 (1) 7 (3) 6 (3) 7 (3) 4 (2) 4 (2) 6 (3) T1-2N1M0 40 (5) 13 (6) 9 (4) 15 (7) 3 (1) 15 (7) 8 (3) 14 (8) 3 (1) T3-4xN1M0 140 (16) 24 (11) 38 (18) 41 (19) 37 (17) 25 (11) 37 (16) 37 (20) 41 (18) T, any, N2M0 78 (9) 34 (16) 16 (8) 12 (5) 16 (7) 5 (2) 26 (11) 19 (10) 28 (12) T(any)N(any)M1 110 (13) 47 (21) 26 (12) 22 (10) 15 (7) 25 (11) 36 (16) 18 (10) 31 (13) Tumor differentiation Well-moderate 838 (91) 222 (94) 205 (91) 212 (91) 199 (90) 238 (96) 215 (90) 176 (90) 209 (89).46 Poor 80 (9) 15 (6) 21 (9) 21 (9) 23 (10) 9 (4) 25 (10) 19 (10) 27 (11).01 Mucinous component, % (64) 165 (70) 156 (69) 140 (60) 124 (56) 174 (70) 157 (65) 124 (64) 130 (55) (23) 48 (20) 53 (23) 54 (23) 59 (27) (22) 52 (22) 42 (22) 66 (28) (13) 24 (10) 17 (8) 39 (17) 39 (18) 19 (8) 31 (13) 29 (15) 40 (17) Signet ring cell component, % (92) 219 (92) 206 (91) 214 (92) 206 (93) 231 (94) 218 (91) 181 (93) 215 (91) (7) 15 (6) 19 (8) 12 (5) 14 (6) (6) 19 (8) 11 (6) 16 (7) (1) 3 (1) 1 (0.4) 7 (3) 2 (1) 2 (1) 3 (1) 3 (2) 5 (2) Peritumoral lymphocytic reaction Absent/minimal 107 (12) 35 (15) 23 (11) 22 (10) 27 (13) 29 (12) 27 (12) 18 (9) 33 (14) Mild 672 (76) 180 (79) 173 (80) 172 (77) 147 (69) (81) 177 (78) 143 (75) 159 (70).04 Moderate/marked 104 (12) 14 (6) 21 (10) 30 (13) 39 (18) 17 (7) 23 (10) 29 (15) 35 (15) Tumor infiltrating lymphocytes Absent/minimal 669 (76) 189 (82) 173 (79) 162 (72) 145 (68) 195 (81) 174 (76) 142 (75) 158 (70) Mild 125 (14) 31 (13) 25 (11) 36 (16) 33 (15) (14) 29 (13) 27 (14) 35 (15).02 Moderate/marked 91 (10) 10 (4) 20 (9) 26 (12) 35 (16) 11 (5) 25 (11) 21 (11) 34 (15) MSI status MSI-low/MSS 742 (85) 212 (93) 194 (89) 187 (83) 149 (73) 221 (93) 192 (83) 165 (87) 164 (75) MSI-high 134 (15) 16 (7) 25 (11) 37 (17) 56 (27) 16 (7) 39 (17) 25 (13) 54 (25) CIMP status CIMP-low/0 739 (84) 209 (92) 192 (87) 183 (81) 155 (73) 219 (92) 194 (84) 159 (83) 167 (74) CIMP-high 146 (17) 18 (8) 28 (13) 44 (19) 56 (27) 19 (8) 37 (16) 32 (17) 58 (26) LINE-1 methylation, 62.0 (9.4) 61.1 (10.5) 61.2 (9.4) 62.4 (8.8) 63.5 (8.7) (10.1) 61.2 (9.8) 61.4 (8.5) 63.4 (8.7).12 mean (SD) KRAS mutation Negative 560 (63) 146 (64) 141 (63) 151 (67) 122 (59) 158 (66) 149 (64) 126 (66) 127 (58).40 Positive 322 (37) 82 (36) 82 (37) 74 (33) 84 (41) 81 (34) 84 (36) 65 (34) 92 (42).25 BRAF mutation Negative 760 (86) 207 (90) 199 (90) 189 (85) 165 (80) 222 (92) 197 (85) 164 (87) 177 (80).006 Positive 121 (14) 23 (10) 23 (10) 33 (15) 42 (20) 19 (8) 34 (15) 25 (13) 43 (20).004 PIK3CA mutation Negative 663 (83) 182 (87) 170 (85) 165 (81) 146 (79) 182 (85) 178 (83) 148 (85) 155 (79).19 Positive 136 (17) 28 (13) 31 (15) 39 (19) 38 (21) 33 (15) 36 (17) 26 (15) 41 (21).38 Abbreviations: CIMP, CpG island methylator phenotype; LINE-1, long interspersed nucleotide element 1; MSI, microsatellite instability; MSS, microsatellite stablility. a Unless otherwise indicated, data are given as number (percentage) indicating the proportion of cases with a specific clinical, pathological, or molecular feature among a given quartile category of the negative or total node count. Denominators vary because of missing data. b P values were calculated by analysis of variance for age, tumor size, and LINE-1 methylation and by 2 test for all other variables. Because of multiple hypothesis testing, a P value for significance was adjusted by Bonferroni correction to sponse to tumor or tumor molecular features despite the possible influence of immune reaction and tumor molecular variables on the node count. 9 Previous studies that examined the relationship between MSI and the node count lacked specimen length and molecular variables besides MSI. 51,52 In contrast to all previous studies that examined potential predictors of the node count, 11,13,14,17-24,28 we have used a US nationwide cohort database with well-annotated clinical, specimen, pathological, and tumor molecular data, including MSI, E5

6 Table 3. Multivariate Negative Binomial Regression Analysis to Predict the Negative Lymph Node Count in Colorectal Cancer Resections a Variable in the Final Model to Predict the Negative Node Count Adjusted Fold Change in Mean Negative Node Count by a Given Variable (95% CI) Class-Level Test P Value Overall Test Specimen length, per 10-cm increase 1.11 ( ) Tumor location Ascending colon Cecum 0.90 ( ).19 Transverse colon 0.79 ( ).012 Descending colon 0.57 ( ) Sigmoid colon 0.72 ( ) Rectum 0.67 ( ) TNM stage T3N0M0 T1N0M ( ).002 T2N0M ( ).06 T4N0M ( ).08 T1-2N1M ( ).005 T3-4N1M ( ).07 T(any)N2M ( ) T(any)N(any)M ( ) Year of diagnosis, for 5-y increase as a unit 1.10 ( ).001 Tumor size, for 5-cm increase as a unit 1.24 ( ).002 Age at diagnosis, for 10-y increase as a unit 0.92 ( ).009 Circumferential growth, 100% complete vs 1.17 ( ).02 incomplete KRAS mutation vs wild-type 1.13 ( ).03 a Variables (potential predictors) initially included in a regression model were the variables listed in the table, sex, body mass index, family history of colorectal cancer, hospital setting, tumor differentiation, mucinous component, signet ring cells, peritumoral lymphocytic reaction, tumor infiltrating lymphocytes, CpG island methylator phenotype, microsatellite instability, long interspersed nucleotide element 1 methylation, and BRAF and PIK3CA mutations. A backward elimination with a threshold of P =.1 was used to select variables in the final model. The adjusted fold change in mean node count by a given variable represents the exponential of the adjusted coefficient derived by the final model. Because of multiple hypothesis testing, a P value for significance was adjusted by Bonferroni correction to CIMP, and KRAS, BRAF, and PIK3CA mutations, all of which are potential predictors of the node count. With regard to the influence of medical care quality or socioeconomic status on the lymph node count, 12,53 academic hospital status 5,14,25 and the degree of practicing experience of the surgeons 11,18 and pathologists 11,13,17,18,54 have been associated with the node count (for review, see Storli et al 55 ). However, all but three 14,18,24 of those previous studies on medical care quality or socioeconomic status 5,11,13,16,17,53,54 lacked data on specimen length. Our ability to use the database of 2 US nationwide prospective cohort studies to assess operator-independent predictors of the node count provided advantages. First, cohort participants who developed cancer were treated at hospitals throughout the United States (Figure 1) and were more representative of colorectal cancer cases in the general US population than one might expect of patients in 1 to a few hospitals. Second, because of our study design, any particular surgeon or pathologist could not have influenced our results, increasing the generalizability of our findings. Third, our rich molecular pathological epidemiology 8,30,31 database enabled us to simultaneously assess a number of variables and to adjust for potential confounding. One weakness of our study is that participants of our cohort studies were US health professionals and predominantly non-hispanic white individuals, thus constituting a rather homogeneous group, and the studies lacked other occupational and ethnic groups. One of the primary reasons for selecting health professionals as subjects in the cohort studies was that they have a good understanding of various diseases as well as of the value of the cohort studies, which increases the reliability and completeness of questionnaire-based follow-up and data collection. Second, we excluded a subset of cancer cases without available tumor tissue, which might cause bias. Nonetheless, the tumor specimen procurement rate has been 60% to 70% of attempts, and a previous study has shown that there is no substantial demographic or clinical difference between cases with and without tumor tissue analyzed. 32 Our ultimate goal is to determine how many nodes must be harvested to attain optimal care when devising an individualized treatment plan in each case. To achieve this goal, we need to assemble an adequate database with prospective follow-up to record detailed outcome data, preferably in a trial setting. We do not have enough data now to recommend a specific number of nodes that should be examined for optimal patient care. Nonetheless, our unique data set, which has provided strong evidence of the effects of specimen length, tumor size, tumor location, and TNM stage on the node count, will likely serve as a guide for future trials. In conclusion, our study has shown that specimen length, tumor size and location, and TNM stage are predictors of E6

7 Table 4. Multivariate Negative Binomial Regression Analysis to Predict the Total Number of Recovered Nodes in Colorectal Cancer Resections a Variable in the Final Model to Predict the Total Node Count Adjusted Fold Change in Mean Total Node Count by a Given Variable (95% CI) Class-Level Test P Value Overall Test Specimen length, per 10-cm increase 1.09 ( ) Tumor location Ascending colon Cecum 0.86 ( ).04 Transverse colon 0.82 ( ).02 Descending colon 0.61 ( ) Sigmoid colon 0.73 ( ) Rectum 0.71 ( ) TNM stage T3N0M0 T1N0M ( ).002 T2N0M ( ).03 T4N0M ( ).08 T1-2N1M ( ).12 T3-4N1M ( ).92 T(any)N2M ( ).02 T(any)N(any)M ( ).38 Tumor size, per 5-cm increase 1.29 ( ) Year of diagnosis, per 5-y increase 1.07 ( ).004 KRAS mutation vs wild-type 1.13 ( ).009 Age at diagnosis, per 10-y increase 0.94 ( ).03 Circumferential growth, 100% complete 1.14 ( ).03 vs incomplete Peritumoral lymphocytic reaction, per 1-unit increase in severity 1.08 ( ).09 a Variables (potential predictors) initially included in a regression model were the variables listed in the table, sex, body mass index, family history of colorectal cancer, hospital setting, tumor differentiation, mucinous component, signet ring cells, tumor infiltrating lymphocytes, CpG island methylator phenotype, microsatellite instability, long interspersed nucleotide element 1 methylation, and BRAF and PIK3CA mutations. A backward elimination with a threshold of P =.1 was used to select variables in the final model. The adjusted fold change in mean node count by a given variable represents the exponential of the adjusted coefficient derived by the final model. Because of multiple hypothesis testing, a P value for significance was adjusted by Bonferroni correction to Table 5. Multivariate Negative Binomial Regression Analysis to Predict the Node Count in Stage I and II Colorectal Cancer Resections a Variable in the Final Model to Predict the Total Node Count Tumor location Ascending colon Adjusted Fold Change in Mean Total Node Count by a Given Variable (95% CI) Class-Level Test Cecum 0.91 ( ).33 Transverse colon 0.77 ( ).02 Descending colon 0.66 ( ).001 Sigmoid colon 0.64 ( ) Rectum 0.66 ( ) P Value Overall Test Tumor size, per 5-cm increase 1.29 ( ).002 Specimen length, per 10-cm increase 1.09 ( ).002 TNM stage T3N0M0 T1N0M ( ).003 T2N0M ( ) T4N0M ( ).17 PIK3CA mutation vs wild-type 1.21 ( ).03 KRAS mutation vs wild-type 1.15 ( ).049 a Variables (potential predictors) initially included in a regression model were the variables listed in the table, sex, age, body mass index, family history of colorectal cancer, year of diagnosis, hospital setting, circumferential growth, tumor differentiation, mucinous component, signet ring cells, peritumoral lymphocytic reaction, tumor infiltrating lymphocytes, CpG island methylator phenotype, microsatellite instability, long interspersed nucleotide element 1 methylation, and BRAF mutation. A backward elimination with a threshold of P =.1 was used to select variables in the final model. The adjusted fold change in mean node count by a given variable represents the exponential of the adjusted coefficient derived by the final model. Because of multiple hypothesis testing, a P value for significance was adjusted by Bonferroni correction to E7

8 the lymph node count in colorectal cancer resections independent of operator (surgeon and pathologist) factors. In addition, some tumor molecular features, such as KRAS mutation, might influence the node count but must be confirmed by an independent data set. Our data suggest that these clinical, pathological, specimen, and molecular variables should be examined as crucial elements in any future evaluation of the adequacy of lymph node examination for patients with colorectal cancer. Accepted for Publication: January 30, Published Online: April 16, doi: /archsurg Author Affiliations: Department of Medical Oncology, Dana-Farber Cancer Institute and Harvard Medical School (Drs Morikawa, Kuchiba, Yamauchi, Meyerhardt, Schrag, Fuchs, and Ogino), Departments of Pathology (Drs Hornick and Ogino) and Surgery (Dr Swanson) and Channing Laboratory, Department of Medicine (Drs Chan and Fuchs), Brigham and Women s Hospital and Harvard Medical School, Gastrointestinal Unit, Massachusetts General Hospital (Dr Chan), and Department of Biostatistics, Harvard School of Public Health (Dr Huttenhower), Boston, Massachusetts; National Surgical Adjuvant Breast and Bowel Project Operations and Biostatistics Center, Pittsburgh, Pennsylvania (Dr Tanaka); and First Department of Internal Medicine, Sapporo Medical University, Sapporo, Japan (Dr Nosho). Correspondence: Shuji Ogino, MD, PhD, MS(Epidemiology), Department of Medical Oncology, Center for Molecular Oncologic Pathology, Dana-Farber Cancer Institute, Brigham and Women s Hospital, 450 Brookline Ave, Room JF-215C, Boston, MA (shuji_ogino@dfci.harvard.edu). Author Contributions: Drs Morikawa, Tanaka, Kuchiba, and Nosho contributed equally. Drs Tanaka and Ogino had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Swanson, Schrag, Fuchs, and Ogino. Acquisition of data: Morikawa, Nosho, Yamauchi, Chan, Schrag, Fuchs, and Ogino. Analysis and interpretation of data: Morikawa, Tanaka, Kuchiba, Hornick, Swanson, Chan, Meyerhardt, Huttenhower, Schrag, Fuchs, and Ogino. Drafting of the manuscript: Morikawa, Tanaka, Nosho, Yamauchi, Schrag, Fuchs, and Ogino. Critical revision of the manuscript for important intellectual content: Morikawa, Kuchiba, Hornick, Swanson, Chan, Meyerhardt, Huttenhower, Schrag, Fuchs, and Ogino. Statistical analysis: Tanaka, Kuchiba, Nosho, Huttenhower, Schrag, and Fuchs, Obtained funding: Chan, Schrag, Fuchs, and Ogino. Administrative, technical, and material support: Morikawa, Fuchs, and Ogino. Study supervision: Swanson, Huttenhower, Schrag, Fuchs, and Ogino. Financial Disclosure: None reported. Funding/Support: This work was supported by grants P01 CA87969, P01 CA55075, P50 CA (Dr Fuchs), R01 CA (Dr Ogino), and R01 CA (Dr Chan) from the National Institutes of Health; the Bennett Family Fund for Targeted Therapies Research; and the Entertainment Industry Foundation through the National Colorectal Cancer Research Alliance. Disclaimer: The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health. Role of the Sponsor: The funding agencies had no role in the design of the study; in the collection, analysis, or interpretation of the data; in the decision to submit the manuscript for publication; or in the writing of the manuscript. Additional Contributions: We thank the participants and staff of the Nurses Health Study and the Health Professionals Follow-up Study for their valuable contributions, as well as the following state cancer registries for their help: Alabama, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Florida, Georgia, Idaho, Illinois, Indiana, Iowa, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Nebraska, New Hampshire, New Jersey, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, Tennessee, Texas, Virginia, Washington, and Wyoming. REFERENCES 1. Compton CC. Optimal pathologic staging: defining stage II disease. Clin Cancer Res. 2007;13(22, pt 2):6862s-6870s. 2. Chang GJ, Rodriguez-Bigas MA, Skibber JM, Moyer VA. Lymph node evaluation and survival after curative resection of colon cancer: systematic review. J Natl Cancer Inst. 2007;99(6): Hammond ME, Fitzgibbons PL, Compton CC, et al; Cancer Committee and Conference Participants. College of American Pathologists Conference XXXV: solid tumor prognostic factors which, how and so what? summary document and recommendations for implementation. Arch Pathol Lab Med. 2000;124(7): Nelson H, Petrelli N, Carlin A, et al; National Cancer Institute Expert Panel. Guidelines 2000 for colon and rectal cancer surgery. J Natl Cancer Inst. 2001;93 (8): Bilimoria KY, Bentrem DJ, Stewart AK, et al. Lymph node evaluation as a colon cancer quality measure: a national hospital report card. J Natl Cancer Inst. 2008; 100(18): Parsons HM, Tuttle TM, Kuntz KM, Begun JW, McGovern PM, Virnig BA. Association between lymph node evaluation for colon cancer and node positivity over the past 20 years. JAMA. 2011;306(10): Porter GA, Urquhart R, Bu J, Johnson P, Rayson D, Grunfeld E. Improving nodal harvest in colorectal cancer: so what [published online October 4, 2011]? Ann Surg Oncol. doi: /s Ogino S, Galon J, Fuchs CS, Dranoff G. Cancer immunology analysis of host and tumor factors for personalized medicine. Nat Rev Clin Oncol. 2011;8(12): Ogino S, Nosho K, Irahara N, et al. Negative lymph node count is associated with survival of colorectal cancer patients, independent of tumoral molecular alterations and lymphocytic reaction. Am J Gastroenterol. 2010;105(2): George S, Primrose J, Talbot R, et al; Wessex Colorectal Cancer Audit Working Group. Will Rogers revisited: prospective observational study of survival of 3592 patients with colorectal cancer according to number of nodes examined by pathologists. Br J Cancer. 2006;95(7): Morris EJ, Maughan NJ, Forman D, Quirke P. Identifying stage III colorectal cancer patients: the influence of the patient, surgeon, and pathologist. J Clin Oncol. 2007;25(18): Wong SL, Ji H, Hollenbeck BK, Morris AM, Baser O, Birkmeyer JD. Hospital lymph node examination rates and survival after resection for colon cancer. JAMA. 2007; 298(18): Lemmens VE, van Lijnschoten I, Janssen-Heijnen ML, Rutten HJ, Verheij CD, Coebergh JW. Pathology practice patterns affect lymph node evaluation and outcome of colon cancer: a population-based study. Ann Oncol. 2006;17(12): Wright FC, Law CH, Last L, et al. Lymph node retrieval and assessment in stage II colorectal cancer: a population-based study. Ann Surg Oncol. 2003;10(8): Grote T, Hughes AH, Rimmer CC, Less DA, Abernethy AP; Multidisciplinary Gastrointestinal Tumor Board, Derrick L. Davis Forsyth Regional Cancer Center. E8

9 Targeting lymph node retrieval and assessment in stage II colon cancer: a quality outcome community-based cancer center study. J Oncol Pract. 2008;4(2): Cone MM, Shoop KM, Rea JD, Lu KC, Herzig DO. Ethnicity influences lymph node resection in colon cancer. J Gastrointest Surg. 2010;14(11): Evans MD, Barton K, Rees A, Stamatakis JD, Karandikar SS. The impact of surgeon and pathologist on lymph node retrieval in colorectal cancer and its impact on survival for patients with Dukes stage B disease. Colorectal Dis. 2008;10 (2): Valsecchi ME, Leighton J Jr, Tester W. Modifiable factors that influence colon cancer lymph node sampling and examination. Clin Colorectal Cancer. 2010; 9(3): Shen SS, Haupt BX, Ro JY, Zhu J, Bailey HR, Schwartz MR. Number of lymph nodes examined and associated clinicopathologic factors in colorectal carcinoma. Arch Pathol Lab Med. 2009;133(5): Thorn CC, Woodcock NP, Scott N, Verbeke C, Scott SB, Ambrose NS. What factors affect lymph node yield in surgery for rectal cancer? Colorectal Dis. 2004; 6(5): Norwood MG, Sutton AJ, West K, Sharpe DP, Hemingway D, Kelly MJ. Lymph node retrieval in colorectal cancer resection specimens: national standards are achievable, and low numbers are associated with reduced survival. Colorectal Dis. 2010;12(4): Gelos M, Gelhaus J, Mehnert P, et al. Factors influencing lymph node harvest in colorectal surgery. Int J Colorectal Dis. 2008;23(1): Neufeld D, Bugyev N, Grankin M, et al. Specimen length as a perioperative surrogate marker for adequate lymphadenectomy in colon cancer: the surgeon s role. Int Surg. 2007;92(3): Stocchi L, Fazio VW, Lavery I, Hammel J. Individual surgeon, pathologist, and other factors affecting lymph node harvest in stage II colon carcinoma: is a minimum of 12 examined lymph nodes sufficient? Ann Surg Oncol. 2011;18(2): Senthil M, Trisal V, Paz IB, Lai LL. Prediction of the adequacy of lymph node retrieval in colon cancer by hospital type. Arch Surg. 2010;145(9): Lan YT, Lin JK, Jiang JK, Chang SC, Liang WY, Yang SH. Significance of lymph node retrieval from the terminal ileum for patients with cecal and ascending colonic cancers. Ann Surg Oncol. 2011;18(1): Lagoudianakis E, Pappas A, Koronakis N, et al. Lymph node harvesting in colorectal carcinoma specimens. Tumori. 2011;97(1): Chou JF, Row D, Gonen M, Liu YH, Schrag D, Weiser MR. Clinical and pathologic factors that predict lymph node yield from surgical specimens in colorectal cancer: a population-based study. Cancer. 2010;116(11): Tekkis PP, Smith JJ, Heriot AG, Darzi AW, Thompson MR, Stamatakis JD; Association of Coloproctology of Great Britain and Ireland. A national study on lymph node retrieval in resectional surgery for colorectal cancer. Dis Colon Rectum. 2006;49(11): Ogino S, Stampfer M. Lifestyle factors and microsatellite instability in colorectal cancer: the evolving field of molecular pathological epidemiology. J Natl Cancer Inst. 2010;102(6): Ogino S, Chan AT, Fuchs CS, Giovannucci E. Molecular pathological epidemiology of colorectal neoplasia: an emerging transdisciplinary and interdisciplinary field. Gut. 2011;60(3): Chan AT, Ogino S, Fuchs CS. Aspirin and the risk of colorectal cancer in relation to the expression of COX-2. N Engl J Med. 2007;356(21): Morikawa T, Kuchiba A, Yamauchi M, et al. Association of CTNNB1 ( -catenin) alterations, body mass index, and physical activity with survival in patients with colorectal cancer. JAMA. 2011;305(16): Ogino S, Nosho K, Irahara N, et al. Lymphocytic reaction to colorectal cancer is associated with longer survival, independent of lymph node count, microsatellite instability, and CpG island methylator phenotype. Clin Cancer Res. 2009; 15(20): Ogino S, Kawasaki T, Brahmandam M, et al. Sensitive sequencing method for KRAS mutation detection by pyrosequencing. J Mol Diagn. 2005;7(3): Ogino S, Kawasaki T, Kirkner GJ, Loda M, Fuchs CS. CpG island methylator phenotype low (CIMP-low) in colorectal cancer: possible associations with male sex and KRAS mutations. J Mol Diagn. 2006;8(5): Nosho K, Kawasaki T, Ohnishi M, et al. PIK3CA mutation in colorectal cancer: relationship with genetic and epigenetic alterations. Neoplasia. 2008;10(6): Nosho K, Irahara N, Shima K, et al. Comprehensive biostatistical analysis of CpG island methylator phenotype in colorectal cancer using a large populationbased sample. PLoS One. 2008;3(11):e3698. doi: /journal.pone Ogino S, Nosho K, Kirkner GJ, et al. CpG island methylator phenotype, microsatellite instability, BRAF mutation and clinical outcome in colon cancer. Gut. 2009; 58(1): Ogino S, Kawasaki T, Brahmandam M, et al. Precision and performance characteristics of bisulfite conversion and real-time PCR (MethyLight) for quantitative DNA methylation analysis. J Mol Diagn. 2006;8(2): Weisenberger DJ, Siegmund KD, Campan M, et al. CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet. 2006;38(7): Ogino S, Cantor M, Kawasaki T, et al. CpG island methylator phenotype (CIMP) of colorectal cancer is best characterised by quantitative DNA methylation analysis and prospective cohort studies. Gut. 2006;55(7): Ogino S, Nosho K, Kirkner GJ, et al. A cohort study of tumoral LINE-1 hypomethylation and prognosis in colon cancer. J Natl Cancer Inst. 2008;100(23): Irahara N, Nosho K, Baba Y, et al. Precision of pyrosequencing assay to measure LINE-1 methylation in colon cancer, normal colonic mucosa, and peripheral blood cells. J Mol Diagn. 2010;12(2): Agresti A. Categorical Data Analysis. 2nd ed. New York, NY: Wiley Interscience; Hayanga AJ, Mukherjee D, Chang D, et al. Teaching hospital status and operative mortality in the United States: tipping point in the volume-outcome relationship following colon resections? Arch Surg. 2010;145(4): Nicolini A, Ferrari P, Duffy MJ, et al. Intensive risk-adjusted follow-up with the CEA, TPA, CA19.9, and CA72.4 tumor marker panel and abdominal ultrasonography to diagnose operable colorectal cancer recurrences: effect on survival. Arch Surg. 2010;145(12): Cone MM, Herzig DO, Diggs BS, et al. Dramatic decreases in mortality from laparoscopic colon resections based on data from the Nationwide Inpatient Sample. Arch Surg. 2011;146(5): Piessen G, Muscari F, Rivkine E, et al; FRENCH (Fédération de Recherche en Chirurgie). Prevalence of and risk factors for morbidity after elective left colectomy: cancer vs noncomplicated diverticular disease. Arch Surg. 2011;146(10): Nash GM, Row D, Weiss A, et al. A predictive model for lymph node yield in colon cancer resection specimens. Ann Surg. 2011;253(2): Eveno C, Nemeth J, Soliman H, et al. Association between a high number of isolated lymph nodes in T1 to T4 N0M0 colorectal cancer and the microsatellite instability phenotype. Arch Surg. 2010;145(1): Belt EJ, Te Velde EA, Krijgsman O, et al. High lymph node yield is related to microsatellite instability in colon cancer [published online October 12, 2011]. Ann Surg Oncol. doi: /s McBride RB, Lebwohl B, Hershman DL, Neugut AI. Impact of socioeconomic status on extent of lymph node dissection for colon cancer. Cancer Epidemiol Biomarkers Prev. 2010;19(3): Ostadi MA, Harnish JL, Stegienko S, Urbach DR. Factors affecting the number of lymph nodes retrieved in colorectal cancer specimens. Surg Endosc. 2007; 21(12): Storli K, Lindboe CF, Kristoffersen C, Kleiven K, Søndenaa K. Lymph node harvest in colon cancer specimens depends on tumour factors, patients and doctors, but foremost on specimen handling. APMIS. 2011;119(2): E9