SURGERY REMAINS the definitive therapy for adenocarcinoma

Colon Cancer Survival Is Associated With Increasing Number of Lymph Nodes Analyzed: A Secondary Survey of Intergroup Trial INT-0089 By T.E. Le Voyer, ...
Author: Gerald Houston
3 downloads 2 Views 751KB Size
Colon Cancer Survival Is Associated With Increasing Number of Lymph Nodes Analyzed: A Secondary Survey of Intergroup Trial INT-0089 By T.E. Le Voyer, E.R. Sigurdson, A.L. Hanlon, R.J. Mayer, J.S. Macdonald, P.J. Catalano, and D.G. Haller Purpose: To determine the relationship, in patients with adenocarcinoma of the colon, between survival and the number of lymph nodes analyzed from surgical specimens. Patients and Methods: Intergroup Trial INT-0089 is a mature trial of adjuvant chemotherapy for high-risk patients with stage II and stage III colon cancer. We performed a secondary analysis of this group with overall survival (OS) as the main end point. Cause-specific survival (CSS) and disease-free survival were secondary end points. Rates for these outcome measures were estimated using KaplanMeier methodology. Log-rank test was used to compare overall curves, and Cox proportional hazards regression was used to multivariately assess predictors of outcome. Results: The median number of lymph nodes removed at colectomy was 11 (range, one to 87). Of the 3,411 assessable patients, 648 had no evidence of lymph node metas-

tasis. Multivariate analyses were performed on the nodepositive and node-negative groups separately to ascertain the effect of lymph node removal. Survival decreased with increasing number of lymph node involvement (P ⴝ .0001 for all three survival end points). After controlling for the number of nodes involved, survival increased as more nodes were analyzed (P ⴝ .0001 for all three end points). Even when no nodes were involved, OS and CSS improved as more lymph nodes were analyzed (P ⴝ .0005 and P ⴝ .007, respectively). Conclusion: The number of lymph nodes analyzed for staging colon cancers is, itself, a prognostic variable on outcome. The impact of this variable is such that it may be an important variable to include in evaluating future trials. J Clin Oncol 21:2912-2919. © 2003 by American Society of Clinical Oncology.

S

Adjuvant chemotherapy has been shown to improve survival in those patients with locoregional nodal metastases. INT-0089, a large Intergroup trial of adjuvant chemotherapy, evaluated the effects of fluorouracil, levamisole, and leucovorin schedules in patients with high-risk stage II or stage III colorectal cancer.1 This is the largest prospective database for stage II and stage III colon cancer. Because survival was similar for all four arms of the study, this cohort of patients could be evaluated as a single group. We analyzed this population of patients to determine the relationship between the number of lymph nodes recovered and survival.

URGERY REMAINS the definitive therapy for adenocarcinoma of the colon. The principles of surgical resection of colon cancer require en bloc removal of the colon with adequate proximal and distal margins and associated mesentery with draining lymph nodes. The single most important determinant of prognosis in patients with carcinoma of the colon is lymph node involvement. In patients with tumors confined to the bowel wall (stage I or II), the 5-year survival rate is greater than 75%, but this rate decreases to 30% to 60% with lymph node involvement. The number of lymph nodes reported in a colectomy varies widely, and the impact of the number of nodes removed and examined is unknown. This variation in nodal number may be due to variations in surgical technique or to the thoroughness of the pathologist in finding the nodes in the specimen. There have been two opposing views regarding the role of lymphadenectomy in survival. Whether a properly performed lymphadenectomy may produce a therapeutic benefit or whether it is simply a more accurate staging procedure is unknown.

From the Fox Chase Cancer Center, Philadelphia, PA; Cancer and Leukemia Group B, Chicago, IL; Southwest Oncology Group, San Antonio, TX; and Eastern Cooperative Oncology Group, Philadelphia, PA. Submitted May 7, 2002; accepted May 12, 2003. Address reprint requests to Elin R. Sigurdson, MD, Fox Chase Cancer Center; 7701 Burholme Ave, Philadelphia, PA 19111; email: e_sigurdson@ fccc.edu. © 2003 by American Society of Clinical Oncology. 0732-183X/03/2115-2912/$20.00

2912

PATIENTS AND METHODS

Eligibility INT-0089 opened for enrollment in August 1988 and closed in July 1992 after meeting its accrual goal. In all, 3,759 patients were entered onto the study, of which 3,561 had eligible cases. Patients not receiving assigned therapy (50 patients) and who were otherwise eligible for entry onto the study were included in the final analyses on an intention-to-treat basis. Accrual to this trial was such that 3,322 patients, 88% of the total, were entered onto the study between January 1990 and July 1992. The study was conducted with plans for two interim analyses at 60% and 80% information, with a final analysis planned for 100% information. The final analysis of the trial’s arms was completed in November 1997, and the information was presented at the annual meeting of the American Society of Clinical Oncology in 1998.1 Patients were eligible for the adjuvant trial if they had histologic proof of adenocarcinoma of the colon and complete resection of the primary tumor with no gross or microscopic evidence of residual disease. The tumor must have been resected en bloc if it was adherent to any associated organs or structures. The gross inferior margin of the tumor must have been above the peritoneal reflection to exclude a rectal primary. The primary tumor must have shown evidence of poor prognosis, with either nodal involvement

Journal of Clinical Oncology, Vol 21, No 15 (August 1), 2003: pp 2912-2919 DOI: 10.1200/JCO.2003.05.062

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2913

COLON CANCER SURVIVAL: A SECONDARY SURVEY Table 1.

Demographics for Node-Positive and NodeNegative Groups

Men, n Women, n Median age, years Lymph nodes recovered Median Range

Node-Positive

Node-Negative

1,515 1,248 63.9

334 314 62.3

13 1-87

13 1-59

proctoscopic examination and barium enema or a colonoscopy was performed at the 6-month and 12-month visit and then at least every 24 months.

Statistical Methods

Fig 1. orally.

Treatment schema. IV, intravenous; FU, fluorouracil; h, hour; PO,

(Dukes’ C) or Dukes’ B2 (into or through the serosa) and evidence of perforation or obstruction. Regional or mesenteric implants, if present, must have been resected en bloc. Patients with discontiguous or distant disease were ineligible. Patients with synchronous colonic primaries were eligible and were classified according to the most advanced stage lesion. Patient ineligibility criteria were concurrent radiation or chemotherapy, prior fluorouracil therapy, or prior radiation or chemotherapy for the same malignant disease. Patients with a concurrent or previous malignant tumor within the previous 3 years (except for superficial squamous or basal cell carcinoma of the skin or in situ carcinoma of the cervix) were ineligible. Patients must not have had any other major medical illnesses that would have prevented them from receiving the protocol chemotherapy. Patients were entered onto the study and randomly assigned to a treatment arm between 21 and 35 days postoperatively. Patients were treated at institutions affiliated with the following cooperative groups: Eastern Cooperative Oncology Group, Southwest Oncology Group, Cancer and Leukemia Group B, and the Australasia Gastrointestinal Trials Group.

Randomization Patients were stratified according to the extent of bowel-wall involvement, nodal stage, presence of perforation or obstruction, and presence of resected peritoneal implants. Patients with more than one simultaneously diagnosed colonic primary were classified according to the most advanced-stage lesion.

Treatment All patients underwent standard surgical resection to achieve a complete extirpation of the tumor; the operation performed was determined retrospectively from the operative report and the pathology report. Postoperatively, patients were randomly assigned to one of four fluorouracil-based chemotherapy arms, which are depicted in the schema shown in Figure 1.

Follow-Up Patients were followed up after completion of their chemotherapy every 3 months, for one year after their resection, at which point they were followed up on a 6-month basis. After 5 years, patients were seen annually. During their follow-up visits, patients gave their histories and underwent physical examination. Blood work was obtained to evaluate their complete blood cell count as well as to perform liver function tests and an optional carcinoembryonic antigen test. A chest x-ray was obtained every 6 months. A

End points for our secondary analysis of INT-0089 were overall survival, cause-specific survival, and disease-free survival. Rates for these outcome measures were estimated using Kaplan-Meier methodology2 measuring time from the date of surgery to the date of death, date of death secondary to colon cancer, and the date of failure at any site, respectively. Censoring in accordance with Kaplan-Meier methodology was performed for those patients who had not experienced the end point. The log-rank test3 was used to compare overall curves, and Cox proportional hazards regression4 was used to multivariately assess predictors of outcome. Covariates considered in this analysis were age (continuous), sex (dichotomous), the total number of lymph nodes recovered (continuous), the total number of positive lymph nodes (continuous), T stage (T1 and T2 v T3 v T4), tumor differentiation (well v moderate v poor), tumor type (adenocarcinoma v colloid v signet cell v other), and adjuvant therapy regimen. These covariates were chosen prospectively. Heping Zhang’s recursive partitioning5 was used to determine optimal breakpoints for the numbers of lymph nodes recovered as presented in the univariate setting.

Demographics Data from 3,557 patients from INT-0089 were available for review. Of these records, 146 were excluded because of missing data regarding the exact numbers of lymph nodes removed or discrepancy with surgical dates or procedure, thus impacting survival calculations. The remaining 3,411 patients constitute this analysis. An analysis of survival end points in our cohort showed no difference from the original INT-0089 population. Median follow-up of the population was 79 months (range, 1 to 131 months). The cohort analyzed consisted of 1,849 men and 1,562 women with a median age of 63.7 years (range, 15 to 90 years). Table 1 reveals the demographics according to nodal status, and Figure 2 reveals the demographics according to T stage. As would be expected by the criterion for entry onto the adjuvant trial, the majority of patients were node-positive and had an advanced T stage. The node-positive and node-negative groups were similar in regard to the mean and median number of lymph nodes recovered (13 and 11 lymph nodes, respectively). Distribution of patients by the site of the primary tumor revealed that 1,392 (40.8%) of the tumors were right-sided, 259 (7.6%) originated from the transverse colon, 178 (5.2%) originated from the splenic flexure, and 215 (6.3%) originated from the descending colon, whereas 1,276 patients (37.4%) were found to have a cancer that originated from the sigmoid colon. Figure 3 reveals the distribution of lesions in the 3,411 patients analyzed for this study. The distribution of operations performed is reflected in Table 2. The types of resection are separated by nodal status.

RESULTS

Multivariate analyses were performed separately on the nodenegative and node-positive groups to ascertain the effect of lymph node removal. Tables 3 and 4 depict the stepwise Cox

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2914

LE VOYER ET AL

Fig 2. Node-positive and node-negative patients according to T stage.

multivariate regression model results for the node-negative and node-positive patients, respectively. In the lymph node–positive group, age, T stage, tumor grade, tumor differentiation, the number of positive lymph nodes, and the total number of nodes recovered were highly significant predictors of survival. When the same covariates were analyzed for the lymph node–negative group, only age and the number of nodes recovered were found to statistically affect survival. These covariates and their impacts on overall survival, cause-specific survival, and disease-free survival are depicted in Table 5. Survival decreased with increasing lymph node involvement (P ⫽ .0001 for all three end points). After controlling for the number of nodes involved, survival increased as the number of nodes recovered increased (P ⫽ .0001 for all three end points). Even when no nodes were involved, overall and cause-specific survival improved as more nodes were recovered (P ⫽ .0005 and P ⫽ .0071, respectively). In both the node-positive and node-negative groups, the improvement in survival with more nodes recovered was on a continuum. Using recursive partitioning, the breakpoints for the number of nodes recovered were identified, and Kaplan-Meier

survival curve were generated. The American Joint Committee on Cancer recognizes that survival diminishes with increasing lymph node involvement and defines patients with more than three lymph nodes positive as having N2 disease. We separated the node-positive cohort into those patients with one to three positive nodes and those with four or more involved nodes. The overall survival curve for patients with one to three positive nodes is depicted in Figure 4. Survival improved as the number of lymph nodes recovered increased. An absolute 23% improvement in 5-year survival was achieved if more than 40 lymph nodes were identified as compared with those patients who had 10 or fewer nodes. This absolute improvement in survival persisted and, in fact, increased as time progressed. A similar benefit was noted in the N2 patients. The 5-year survival rate of a patient with four or more positive nodes could be improved by 20% (from 51% to 71%) if more than 35 lymph nodes were analyzed in the pathologic specimen. These survival differences are depicted in Figure 5. Figure 6 reveals the overall survival curves for patients without evidence of nodal metastases. Survival was 73% if 10 or fewer nodes were recovered,

Fig 3.

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

Locations of primary tumors.

2915

COLON CANCER SURVIVAL: A SECONDARY SURVEY Table 2.

Operations Performed for Node-Positive and NodeNegative Groups

Operation Performed

Node-Positive

Node-Negative

Right hemicolectomy Left hemicolectomy Anterior resection Segmental resection Subtotal colectomy Total colectomy Other

1,199 375 368 583 83 42 113

276 106 47 134 41 5 39

Results of Stepwise Cox Multivariate Regression Models for NodeNegative Patients

Covariate

OS (200/648 events) Age Sex Nodes removed Stage Differentiation T1 T2 T3 Adjuvant regimen CSS (123/648 events) Age Sex Nodes removed Stage Differentiation T1 T2 T3 Adjuvant regimen DFS (164/648 events) Age Sex Nodes removed Stage Differentiation T1 T2 T3 Adjuvant regimen

Results of Stepwise Cox Multivariate Regression Models for NodePositive Patients

Covariate

whereas that number improved to 87%, an absolute 14% improvement in survival, if more than 20 lymph nodes were identified. Furthermore, this absolute improvement in overall survival increased to 20% at 8 years. The cause-specific survival curves for the same three groups are revealed in Figures 7, 8, and 9, and a similar pattern of improved survival is noted as more lymph nodes are recovered. Among the node-positive patients, there was a significant difference in time to relapse related to the number of nodes recovered. In both the N1 and the N2 patients, disease-free survival increased as more nodes were removed. These curves are depicted in Figures 10 and 11. The node-negative group did not have a significant difference in disease-free survival that Table 3.

Table 4.

Parameter Estimate

P

Hazard Ratio

0.03374 0.26171 ⫺0.03246 0.15198 0.24136 0.12777 0.26971 0.61999 0.26816

⬍ .0001 .0704 .0005 .2252 .0711 .5962 .6531 .5423 .1701

1.034 1.299 0.968 1.164 1.273 1.136 1.310 1.859 1.308

0.01429 0.14517 ⫺0.03196 0.22726 0.22419 0.17573 0.45562 1.29348 0.33225

.0876 .4283 .0071 .1659 .1941 .5571 .5362 .2107 .1823

1.014 1.156 0.969 1.255 1.251 1.192 1.577 3.645 1.394

0.00955 0.15336 ⫺0.01604 0.33974 0.13327 0.18661 0.59912 0.76823 0.55005

.1759 .3344 .0947 .0194 .3693 .4645 .3203 .4532 .0160

1.010 1.166 0.984 1.405 1.143 1.205 1.821 2.156 1.733

Abbreviations: OS, overall survival; CSS, cause-specific survival; DFS, disease-free survival.

OS (1224/2763 events) Age Sex Nodes removed Stage Differentiation No. of positive nodes T1 T2 T3 Adjuvant regimen CSS (944/2763 events) Age Sex Nodes removed Stage Differentiation No. of positive nodes T1 T2 T3 Adjuvant regimen DFS (1076/2763 events) Age Sex Nodes removed Stage Differentiation No. of positive nodes T1 T2 T3 Adjuvant regimen

Parameter Estimate

P

Hazard Ratio

0.01852 0.17955 ⫺0.02558 0.32502 0.13400 0.09604 0.11039 0.72067 ⫺0.09060 ⫺0.02348

⬍.0001 .0021 ⬍.0001 ⬍.0001 .0114 ⬍.0001 .2508 .0029 .8982 .7627

1.019 1.197 0.975 1.384 1.143 1.101 1.117 2.056 0.913 0.977

0.00687 0.12463 ⫺0.02991 0.44461 0.15013 0.11091 0.15747 0.67500 0.19517 ⫺0.05533

.0285 .0599 ⬍.0001 ⬍.0001 .0124 ⬍.0001 .1434 .0111 .7831 .5309

1.007 1.133 0.971 1.560 1.162 1.117 1.171 1.964 1.216 0.946

0.00202 0.12626 ⫺0.02985 0.39148 0.10140 0.10385 0.17013 0.45512 ⫺0.08138 ⫺0.02854

.4816 .0418 ⬍.0001 ⬍.0001 .0710 ⬍.0001 .0872 .0759 .9086 .7316

1.002 1.135 0.971 1.479 1.107 1.109 1.185 1.576 0.922 0.972

Abbreviations: OS, overall survival; CSS, cause-specific survival; DFS, disease-free survival.

could be attributed to the number of lymph nodes recovered and is depicted in Fig 12. DISCUSSION

Multiple prognostic factors have been identified that affect the survival of patients with colon cancer. The presence of lymph node involvement is undoubtedly the most important prognostic factor and has long been recognized. Dukes6 in 1932 recognized the importance of nodal status and incorporated this factor into his staging system for rectal cancer. In 1939, Simpson and Mayo7 applied Dukes’ staging system to patients with colon cancer. All subsequent staging systems for colon cancer, including the current tumor-node-metastasis system, have continued to recognize the impact of involved lymph nodes. The adverse effect of lymph node involvement on survival has led to the routine use of adjuvant chemotherapy for those patients found to have regional metastases.8 However, the relationship between the number of lymph nodes removed or analyzed and patient survival has not been previously studied. Our study reveals that the number of lymph nodes analyzed is a significant variable that affects survival in both node-negative and node-positive patients. Regardless of the

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2916

LE VOYER ET AL Table 5. P Values for Multivariant Analysis Node-Negative Patients

Node-Positive Patients

Covariate

OS

CSS

DFS

OS

CSS

DFS

Age T stage Tumor type Differentiation No. of lymph nodes recovered No. of positive lymph nodes Adjuvant regimen

.0001 .2252 .5440 .0711 .0005 N/A .1701

.0876 .1659 .2108 .1941 .0071 N/A .1823

.1759 .0194 .3203 .3693 .0947 N/A .0160

.0001 .0001 .0029 .0014 .0001 .0001 .3121

.0285 .0001 .0111 .0123 .0001 .0001 .1115

.4816 .0001 .0759 .0001 .0001 .0001 .1803

Abbreviations: OS, overall survival; CSS, cause-specific survival; DFS, disease-free survival.

number of positive nodes, survival improved as more nodes were removed. If a patient was classified as node-negative, survival was also affected positively with an increase in the number of nodes analyzed. Two mechanisms might explain our results. The first mechanism may relate to the surgeon as a technician. Perhaps those patients who each had more nodes identified in their specimen had a more complete excision of their tumor and the draining nodes. In this case, it could be hypothesized that the surgeon was, in fact, a controlling variable in outcome. A surgeon who performed a more complete excision with removal of more lymph nodes might also be expected to perform a better technical operation with a complete in continuity resection of involved structures. Improved surgical technique may include improved intraoperative staging, to exclude stage IV patients. As such, there would be less likelihood of leaving tumor behind and thus affecting survival. The concept of the surgeon as a variable in outcomes is not new. Luft et al9 proposed that hospitals with greater volumes, greater surgeon experience, or both could impact the outcome of particular types of surgery, such as open-heart surgery. The survival of patients undergoing pancreatic resection for malignancy has been correlated with hospital volume and thus surgeon experience. Lieberman et al10 reviewed the outcome of patients undergoing either a pancreaticoduodenectomy or total pancreatectomy for malignancy in the state of New York. In this study, hospitals were grouped by case volume over an 8-year period of

study. When adjusted for differences in patient characteristics, such as age, sex, race, admission status, and year of surgery, the mortality rates were 18.9% for low-volume centers (10 to 50 cases over the period of study) and 5.5% for high-volume centers (⬎ 80 cases). Simunovic et al11 showed similar volume outcome results for patients undergoing resection of a pancreatic malignancy in the Canadian health care system. The outcome of patients undergoing resection for rectal cancer has also been linked to the volume of practice or the specific training of the surgeon. Lane et al12 performed a survey of the members of the Association of Coloproctology of Great Britain and Ireland and found that outcome and local recurrence decreased when surgery was performed by a specialist. The impact of specialization was noted only for patients with Dukes’ C cancer, and a decreased local failure rate (11% v 26%) translated into an improved 3-year survival rate. Although surgeon variability may have played a role in the outcome, another explanation for our results is also possible. It is likely that those patients with more nodes identified in their specimen had a more complete examination by the pathologist and, therefore, a more accurate staging. Survival in colon cancer is influenced not only by the presence of positive nodes but also by the number of positive nodes. The American Joint Committee on Cancer has defined a breakpoint for classifying a patient as N1 or N2 based on the presence of more than three positive

Fig 4. Overall survival for patients with N1 disease. LNR, lymph nodes recovered.

Fig 5. Overall survival for patients with N2 disease. LNR, lymph nodes recovered.

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2917

COLON CANCER SURVIVAL: A SECONDARY SURVEY

Fig 6. Overall survival for patients with N0 disease. LNR, lymph nodes recovered.

Fig 8. Cause-specific survival for patients with N2 disease. LNR, lymph nodes recovered.

nodes. An inaccurate pathologic examination might miss positive nodes and thereby understage a patient. To have an accurate assessment, the mesocolon must be examined carefully. The number of nodes found in a specimen depends on the size and completeness of the surgical resection, the care taken by the pathologist to identify the nodes, and the examination method used by the pathologist. Hermanek et al13 determined that the average number of nodes present in a standard colon resection after use of a fat-clearance technique was 47. Without the aid of a fat-clearance technique, the authors found 31 nodes on average. Although this number is almost one third less, it is still significantly greater than the mean of 13 identified in both our node-positive and node-negative patients. The number of nodes that must be examined to yield a reliable assessment of the nodal status of a patient has been debated. In 1990, the Working Party Report to the World Congresses of Gastroenterology reviewed this topic and came to a consensus recommendation that at least 12 nodes must be sampled to ade-

quately stage a patient.14 However, several more recent studies have brought this number into question. Goldstein et al15 found that the number of node-positive patients continued to increase until 17 to 20 nodes had been examined, leading to the conclusion that a minimum of 17 nodes should be analyzed. Wong et al16 found that node-negative patients had significantly fewer nodes examined than did node-positive patients (14 v 20 nodes), and to achieve a nodal positivity rate commensurate with the National Cancer Data Bank, at least 14 nodes should be examined. It is clear that to accurately stage a patient with colon cancer it is better to evaluate as many nodes as possible. Our data would indicate that for node-negative patients, at least 20 nodes should be identified. This number may be somewhat falsely elevated because of the high-risk population studied. Nevertheless, survival diminishes when fewer than this number of nodes are removed or examined. This result might be explained by the phenomenon of stage migration, but it also may be that patients who have this many nodes in their specimen actually receive a benefit from a more

Fig 7. Cause-specific survival for patients with N1 disease. LNR, lymph nodes recovered.

Fig 9. Cause-specific survival for patients with N0 disease. LNR, lymph nodes recovered.

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2918

LE VOYER ET AL

Fig 10. Disease-free survival for patients with N1 disease. LNR, lymph nodes recovered.

complete excision. Because all patients in this study received chemotherapy, there can be no bias in survival based on the fact that a falsely node-negative patient did not receive the potential benefit of adjuvant therapy. The likelihood of identifying a positive node increases as more nodes are examined. The difficulty comes in finding all the nodes in a specimen. It would seem obvious that the larger nodes would be the easiest to find. Studies have looked at nodal size as a possible predictor of pathologic status, and no correlation could be found.16,17 Herrera-Ornelas et al18 used a fat-dissolving technique to identify lymph nodes present within the specimen mesentery and found that 64% of the positive nodes were less than 5 mm in size. Furthermore, of those positive nodes smaller than 5 mm, 45% were ⱕ 3 mm in size. The fat-dissolving technique, however, is labor-intensive and costly. Ratto et al19 described an approach to handling the mesentery that added only minimal time and significantly increased the yield of identified nodes. In his study, conventional methods of visual inspection and palpation were

Fig 11. Disease-free survival for patients with N2 disease. LNR, lymph nodes recovered.

Fig 12. Disease-free survival for patients with N0 disease. LNR, lymph nodes recovered.

used to identify nodes. One group had the entire specimen fixed whole. The second group had the specimen divided into three parts. The first was the colon, opened along its antimesenteric border, with an adjacent 2 cm of mesentery. The second specimen was the mesentery with principal draining vessels, and the third specimen was the remaining mesentery. All specimens in the second group were pinned to cork boards and then fixed. The second group had a mean of 29.6 nodes identified, compared with only 11.3 in the first group. Furthermore, the number of node-positive patients was significantly greater (37.5% v 30.2%) in the second group. Notably, despite the upstaging of patients in the second group, overall survival was improved. This study shows that it is possible to improve the yield of lymph nodes from specimens if some additional time is taken to prepare them. Identifying all the nodes is one problem facing the pathologist. A potential second problem is being able to identify a positive node. The technology of sentinel lymph node biopsy allows the pathologist to concentrate efforts on the analysis of a relative few lymph nodes with the use of multiple sections and standard staining with hematoxylin and eosin as well as special immunohistochemical stains to improve the likelihood of identifying a positive node. This technique, which has been applied to breast cancer and melanoma, has also been studied in patients with colon cancer.20 The sentinel node accurately reflected the status of the nodes 94% of the time. Aberrant nodal drainage was identified in 9% of the patients, causing the conduct of the case to be altered. Sentinel node technology may aid in accurately staging patients, as well as assisting the surgeon in identifying nodes and mesentery that must be included in the specimen. One of the important implications of our study is that the number of lymph nodes analyzed is a prognostic variable for outcome. This variable may be a surrogate marker for those patients who have had a complete resection of all tumor-bearing tissue, for improved intraoperative staging, or for improved pathologic staging. This variable is of such significance that it should be considered in the stratification and analysis of future clinical trials.

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

2919

COLON CANCER SURVIVAL: A SECONDARY SURVEY

REFERENCES 1. Haller DG, Catalano PJ, Macdonald JS, et al: Fluorouracil (FU), leucovorin and levamisole adjuvant therapy for colon cancer: Five year report of INT-0089. Proc Am Soc Clin Oncol 17:256a, 1998 (abstr) 2. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958 3. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley, 1980 4. Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972 5. Zhang HP, Singer B: Recursive Partitioning in Health Sciences. New York, NY, Springer, 1999 6. Dukes CE: The classification of cancer of the rectum. J Pathol Bacteriol 35:323-332, 1932 7. Simpson WC, Mayo CW: The mural penetration of the carcinoma cell in the colon: Anatomic and clinical study. Surg Gynecol Obstet 68:872-877, 1939 8. Moertel CG: Chemotherapy for colorectal cancer. N Engl J Med 330:1136-1142, 1994 9. Luft H, Bunker J, Enthoven A: Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med 301:1364-1369, 1979 10. Lieberman M, Kilburn H, Lindsey M, et al: Relation of perioperative deaths to hospital volume among patients undergoing pancreatic resection for malignancy. Ann Surg 222:638-645, 1995 11. Simunovic M, To T, Theriault M, et al: Relation between hospital surgical volume and outcome for pancreatic resection for neoplasm in a publicly funded health care system. CMAJ 160:643-648, 1999

12. Lane R, Thompson M, Whatley P, et al: Effect of specialization on outcome in patients having surgery for rectal cancer. Dis Colon Rectum 42:A6, 1999 13. Hermanek P, Giedl J, Dworak O: Two programs for examination of regional lymph nodes in colorectal carcinoma with regard to the new pN classification. Path Res Pract 185:867-873, 1989 14. Fielding LP, Arsenault PA, Chapuis PH, et al: Working Report to the World Congresses of Gastroenterology, Sydney 1990. J Gastroenterol Hepatol 6:325-344, 1991 15. Goldstein NS, Weldon S, CoffeyM, et al: Lymph node recovery from colorectal resection specimens removed for adenocarcinoma: Trends over time and a recommendation for a minimum number of lymph nodes to be removed. Am J Clin Pathol 106:209-216, 1996 16. Wong JH, Severino R, Honnebier MB, et al: Number of nodes examined and staging accuracy in colorectal carcinoma. J Clin Oncol 17:2896-2900, 1999 17. Mo¨nig SP, Baldus SE, Zirbes TK, et al: Lymph node size and metastatic infiltration in colon cancer. Ann Surg Oncol 6:579-581, 1999 18. Herrera-Ornelas L, Justiniano J, Castillo N, et al: Metastases in small nodes from colon cancer. Arch Surg 122:1253-1256, 1987 19. Ratto C, Sofo L, Ippoliti M, et al: Accurate lymph-node detection in colorectal specimens resected for cancer is of prognostic significance. Dis Colon Rectum 42:143-158, 1999 20. Wood TF, Saha S, Morton DL, et al: Validation of lymphatic mapping in colorectal cancer: In vivo, ex vivo, and laparoscopic techniques. Ann Surg Oncol 8:150-157, 2001

Downloaded from ascopubs.org by 37.44.207.149 on January 19, 2017 from 037.044.207.149 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

Suggest Documents