Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study

Transcription

1 Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study E.S Van Der Zaag., C.J. Buskens, N. Kooij, H. Akol, H.M. Peters, W.H. Bouma, W.A. Bemelman To cite this version: E.S Van Der Zaag., C.J. Buskens, N. Kooij, H. Akol, H.M. Peters, et al.. Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study. EJSO - European Journal of Surgical Oncology, WB Saunders, 2009, 35 (10), pp < /j.ejso >. <hal > HAL Id: hal Submitted on 16 Jan 2011 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

2 Accepted Manuscript Title: Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study Authors: E.S van der Zaag., C.J. Buskens, N. Kooij, H. Akol, H.M. Peters, W.H. Bouma, W.A. Bemelman PII: S (09) DOI: /j.ejso Reference: YEJSO 2791 To appear in: European Journal of Surgical Oncology Received Date: 16 July 2008 Revised Date: 22 November 2008 Accepted Date: 2 February 2009 Please cite this article as: van der Zaag ES, Buskens CJ, Kooij N, Akol H, Peters HM, Bouma WH, Bemelman WA. Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study, European Journal of Surgical Oncology (2009), doi: /j.ejso This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

3 1 Improving staging accuracy in colon and rectal cancer by sentinel lymph node mapping: a comparative study. E.S. van der Zaag 1, C.J. Buskens 1, N. Kooij 2, H. Akol 3, H.M. Peters 2, W.H. Bouma 1, W.A. Bemelman 4. Departments of Surgery 1, Pathology 2, and Gastroenterology 3, Gelre Ziekenhuizen, Apeldoorn, The Netherlands. Department of Surgery 4, Academic Medical Center, Amsterdam, The Netherlands Running title: sentinel lymph node procedure in colorectal cancer Corresponding author: E.S. van der Zaag Department of Surgery Gelre Ziekenhuizen Albert Schweitzerlaan DZ Apeldoorn The Netherlands Telephone: address: e.van.der.zaag@gelre.nl

4 2 ABSTRACT Aim: To compare the predictive value of sentinel lymph node (SN) mapping between patients with colon and rectal cancer. Patients & Methods: An ex vivo SN procedure was performed in 100 patients with colon and 32 patients with rectal cancer. If the sentinel node was negative, immunohistochemical analyses using two different antibodies against cytokeratins (Cam5.2, and CK 20) and one antibody against BerEp-4 were performed to detect occult tumour cells. Isolated tumour cells (< 0.2mm) were discriminated from micrometastases (0.2-2mm). Results: A SN was identified in 117 patients (89%), and accurately predicted nodal status in 106 patients (accuracy 91%). Both sensitivity and negative predictive value was higher in colon carcinomas than in rectal carcinomas (83% versus 57%, p=0.06 and 93% versus 65%, p=0.002 respectively). In patients with extensive lymph node metastases the SN procedures was unsuccessful. Eleven of the 13 unsuccessful SN procedures were performed in patients with rectal cancer who had pre-operative radiotherapy. After immunohistochemical analysis 21 of the 73 N0 patients had occult tumour cells in their SN; 8 patients had micrometastases and 13 patients had isolated tumour cells. Conclusion: SN mapping accurately predicts nodal status in patients with colonic cancer. Immunohistochemical analysis demonstrates micrometastatic disease in eight out of 73 N0 patients, with a true upstaging rate of 11%. SN mapping is less reliable in patients with rectal cancer after pre-operative radiotherapy. Key words: Sentinel lymph node mapping, rectal cancer, colon cancer.

5 3 INTRODUCTION Lymph node status is the most significant predictor for recurrence and survival in colorectal cancer patients. However, 10-25% of patients with stage I or II (absence of lymph node metastases; according to the American Joint Committee on Cancer (AJCC)) will have local recurrences or distant metastases within five years [1]. Metastatic nodes can be missed if an insufficient number of lymph nodes has been examined, or if occult metastases have not been identified by standard histopathological examination. Current data show that adequate staging of colon cancer requires examination of at least 12 nodes per patients [2]. However, to achieve an 85% probability of true node negativity even 40 lymph nodes should be examined [3]. This has prompted the use of sentinel lymph node (SN) mapping for colorectal cancer in an effort to select the nodes most likely to harbour metastases. This technique accurately predicted nodal status in more than 90%, and resulted in upstaging 10 to 30% of patients with stage I or II disease to stage III disease [4, 5], thereby making these patients candidates for adjuvant chemotherapy. The rectum is anatomically quite different from the colon, with a more bulky mesentery and an infraperitoneal location that makes the in vivo visualisation of the SN procedure difficult. Therefore, ex vivo SN mapping is generally performed in these carcinomas. In the Netherlands, preoperative short-course radiotherapy is widely used in rectal cancer because it lowers the local recurrence rate [6]. Although radiotherapy does not alter tumour and node classification, it can induce fibrosis and cause lymphoid tissue to disappear [7, 8]. In contrast to breast cancer, where subsequent lymph nodes will only be harvested when the SN shows metastases, the main purpose of SN mapping in patients with colorectal cancer is to use this technique for a more concentrated pathologic assessment of lymph nodes that are most likely to harbour metastases. Occult tumour cells have been detected in up to

6 4 40% of histologically negative nodes in patients with colorectal cancers. The AJCC nowadays recommends discrimination of occult tumour cells in micrometastases and isolated tumour cells, since the latter do not seem to have prognostic significance.[9,10] Detection of occult tumour cells can be done using immunohistochemistry or polymerase chain reaction (PCR) techniques. However, since PCR techniques do not allow morphological determination of identified staining and will detect isolated tumour cells, micrometastases, and benign marker-positive mesothelial cells with equal sensitivity, immunohistochemistry is the preferred technique and most frequently used.[11-13] The aim of this study was to investigate the accuracy of SN-mapping in patients with colorectal cancer. Specifically we want to analyse whether differences in anatomy and preoperative treatment of rectal cancer affect the predictive value of the SN procedure in rectal cancer in comparison to patients with colon cancer. In addition, the incidence of micrometastatic disease in histologically negative nodes was assessed, and the ability to refine staging by immunohistochemistry compared to conventional histopathological examination.

7 5 MATERIAL AND METHODS Patients Patients with colorectal cancer who were operated on with curative intent between November 2006 and May 2008 were considered for inclusion in the study. Exclusion criteria were invasion of other organs (T4 carcinomas), or two adjacent colorectal carcinomas. Five patients with locally advanced rectal cancer undergoing neoadjuvant chemoradiotherapy were excluded since this treatment downstages tumour and lymph node status, thereby introducing possible false negative findings. Six patients who underwent a colon resection for benign disease (e.g. Crohn s disease or diverticulosis), were used as a negative control group. Resection was either performed via an open or laparoscopic approach. Patients with rectal cancer underwent total mesorectal excision (TME) after preoperative radiotherapy comprising five fractions of five Gray. Patients were generally operated on within five days, but no later than 10 days after the last fraction of radiation. A small number of patients with rectal cancer did not receive radiotherapy due to imminent occlusion or the pre-operative diagnosis of villous adenoma with high grade dysplasia. The study was done in accordance with the guidelines of the local ethics committee. Sentinel lymph node mapping An ex vivo SN technique was used in all patients. After standard resection, ml (depending on the volume of the tumour) patent blue V (Guerbet, Gorinchem, the Netherlands) was injected around the tumour with the colonic specimen left intact, after which the injection site was gently massaged. For colon carcinomas, the mesocolon was inspected and the first one to four blue lymph nodes were identified as SNs and either dissected or marked with a suture. For rectal carcinomas, the identification of blue nodes was

8 6 done at the department of pathology to keep the circumferential resection margin intact. In order not to have the results affected by this discrepancy in SN identification, the specimen was sent to the pathologist in fresh state immediately after resection. Pathological examination Immediately after resection, the fresh specimen was processed using a standardised protocol. Tumours were staged according to the AJCC TNM classification 2002 [9]. All lymph nodes (SNs and non-sns) identified by the pathologist were collected in separate boxes and marked according to location, then cut in two with both sides stained with haematoxylin and eosin and evaluated for tumour involvement. If the SN was negative for metastases, serial sectioning was performed at three levels of 500μm intervals. Three sections of each level were stained with three monoclonal antibodies. The anti-epithelial cell antibody Ber-EP4 (DAKO, The Netherlands), was combined with two anti-cytokeratin antibodies: the anti-ck20 antibody, with its expression limited to gastrointestinal epithelial cells (Euro Diagnostica, Arnhem, The Netherlands), and the anticytokeratin marker Cam5.2 directed against cytokeratin 8 and 18 expressed in all epithelial cells (Becton and Dickinson, Alphen aan den Rijn, The Netherlands) [14]. Appropriate positive and negative controls were added on each automated run, to confirm the sensitivity and specificity of the antibodies (sections of colonic carcinoma tissue served as positive controls; negative controls were obtained by omitting the primary antibody).

9 7 Occult tumour cells Tumour deposits within lymph nodes were classified and staged according to the revised guidelines set by the AJCC [10]. Tumour cell deposits between 0.2mm and 2.0mm were referred to as micrometastases and those smaller than 0.2mm as isolated tumour cells. False-positive non-neoplastic haematopoeitic cells (e.g. reticular cells and plasma cells which can also show staining for cytokeratins), were discriminated from isolated tumour cells by microscopic morphological and nuclear differences. The immunohistochemically stained slides were evaluated blind and independently by two experienced pathologists, who were unaware of the clinical data. Statistical analsysis The following definitions were used for calculation: - Identification rate (%): number of patients with successfully retrieved SN X 100 / number of patients enrolled. - Accuracy (%): number of patients with correct prediction of nodal status X 100 / number of patients with successfully retrieved SN. - Sensitivity (%): number of patients with tumour involved SN X 100 / number of patients with successfully retrieved SN and macrometastases in any lymph node. - Negative predictive value (%): number of nodal negative patients with successfully retrieved SN X 100 / (number of nodal negative patients + number of false-negative patients). Results are expressed as percentages or mean ± SD. All statistical analyses were performed using SPSS version 14.0 (SPSS INC., Chicago, IL, USA). The association between clinicopathological features and detection rate, negative predictive value, sensitivity, and the

10 8 presence of micrometastases was analysed using Student's t-test (continuous variables) or Chi-squared test (categorical variables). P-values of 0.05 or less were considered statistically significant.

11 9 RESULTS Patient and tumour characteristics A total of 100 patients with colon cancer and 32 patients with rectal cancer were included. Of the 32 patients with a rectal cancer, 23 received pre-operative short-course radiotherapy. Baseline clinicopathological characteristics were comparable for both groups, except for the mean number of lymph nodes and SNs analysed per patient which was higher for patients with colon cancer (Table 1). Predictive value of SN mapping A SN was identified in 117 of the 132 patients. Identification of at least one SN was possible in 92 of 100 patients with colon cancer and in 25 of the 32 patients with rectal cancer (p=0.03). Overall, the SN accurately predicted nodal status in 106 patients with 11 false negative SN procedures (accuracy 91%). There was a significant discrepancy in staging accuracy for colon and rectal carcinomas (95% versus 76%, respectively; p=0.005). The sensitivity of the SN procedure was also higher in colon carcinomas (83%) than in rectal carcinomas (57%), although this was not statistically significant (p=0.06). The SN was negative in 67 patients with colon cancer and 17 patients with rectal cancer. In the colon carcinoma group five patients had a false negative procedure, whereas six patients in the rectal carcinoma group had a false negative procedure, resulting in negative predictive values of 93% and 65% respectively (p=0.002) (Table 2). In both groups more unidentified or false negative SNs were seen with an increasing number of positive lymph nodes (p=0.01) and with depth of tumour infiltration (T-stage), although the latter was not significant. Eleven of the 13 unsuccessful or false negative SN procedures were performed in patients with rectal cancer who had pre-operative short-course radiotherapy (p=0.05) (Table 2a and 2b).

12 10 Identification of micrometastatic disease in N0 carcinomas A total of 62 colon cancer patients and 11 rectal cancer patients classified as nodal negative by routine H&E staining with successfully retrieved SN underwent serial step sectioning and focussed immunohistochemical analysis of the SN. In eight patients a micrometastasis was detected in the sinus of the SN, resulting in an upstaging rate of 11%. Another 13 patients revealed isolated tumour cells, resulting in an overall upstaging rate of 29%. There was no significant difference in overall upstaging between patients with colon cancer (27%) or rectal cancer (20%). The pathologists agreed in 67 of the 73 patients (92%) All micrometastases were identified by both pathologists independently. In six patients there was a disagreement about the occurrence of isolated tumour cells. The presence of micrometastases was significantly related to lymphangio invasion (p<0.0001), whereas this correlation could not be found for the occurrence of isolated tumour cells and lymphangio invasion (p=0.4). Although six out of the eight patients with micrometastases had a tumour invading the adventitia (T3), no correlation with depth of tumour infiltration and the presence of micrometastases could be demonstrated (p=0.7) (Table 3). No occult tumour cells were found in the SN of the negative control group.

13 11 DISCUSSION Predicitive value of SN mapping This study demonstrates that SN mapping can reliably predict nodal status in patients with colon carcinomas. A significantly lower identification rate, accuracy, and negative predictive value of SN procedures was found for patients with rectal cancer particularly after radiotherapy. A recent review describes varying accuracy rates of SN lymph node mapping with accuracy rates between 78 and 100% and sensitivity rates varying between 25 and 100% [6]. Interestingly, most studies with disappointing results included both patients with colon and rectal cancer [15-18]. Our findings of the inferior results of the SN procedure in patients with rectal cancer can be an explanation for the conflicting results in previous studies. There are several possible reasons why the SN technique is less reliable in patients with rectal cancer. In contrast to the mesorectum, the mesocolon can easily be inspected from two sides after resection which facilitates identification of the SN. In addition, the mesorectum has a perirectal fascia and this circumferential resection margin should be kept intact during the resection for adequate pathological examination and to reduce the risk of local recurrence, making SN identification more difficult [6, 19]. In our study, SN identification of rectal carcinomas was done at the department of pathology immediately after resection. Not directly identifying SNs after patent blue injection harbours the risk of increasing the number of sentinel nodes by including second echelon lymph nodes [20]. Analysis of a large number of SNs by serial step immunohistochemistry is impractical. However, our study demonstrates that even after postponed SN identification in rectal carcinoma, the number of SNs is still significantly lower than for colon carcinomas. Another possible reason why the SN identification is unsuccessful in patients with rectal cancer is pre-operative radiotherapy. Radiotherapy has been demonstrated to induce fibrosis, causing the already small rectal lymph nodes to disappear [7, 8, 21]. This might alter

14 12 lymphatic flow and lead to false-positive staining of non-sns. In addition, it can also explain the significant lower number of SNs found in rectal carcinomas leading to decreased accuracy. So far, only a few other studies with a small number of patients have concentrated on SN mapping in rectal carcinomas after neo-adjuvant therapy with similar disappointing results [22, 23]. These results suggest that a different approach should be sought for SN mapping for patients with rectal cancer after pre-operative radiotherapy. Factors influencing false negative results Our study demonstrates better results of SN mapping in patients with colon cancer, with a sensitivity of 83% and a negative predictive value of 93%. In literature however, there is a persistent controversy over the value of SN lymph node mapping in these patients. False negative findings are attributable to a number of factors including extent of the disease, mapping technique, timing and method of pathological processing, number of SN lymph nodes evaluated, and method of ultrastaging. Especially local tumour invasion of adjacent lymph nodes has been indicated as a predictor of false negative SN mapping [24, 25], which is confirmed by our data in which a strong correlation between false negative results and number of involved lymph nodes was found. In contrast to our study, other studies also suggest a negative association with tumour volume and number of SNs identified [25, 26]. However, these false negative procedures can be regarded as technical failures. Viehl et al. identified dye volume as a significant predictor of successful SN identification and demonstrated that when the volume is adjusted to tumour size (0.5 ml per 1.0 cm tumour) technical failures can be prevented [24]. Therefore, standardisation of working definitions, mapping technique and pathological processing are critical to the success of SN lymph node mapping for patients with colon cancer. We used the ex vivo SN procedure since only this technique can be performed easily in rectal carcinomas. The ex vivo technique has been demonstrated to be as accurate as the in

15 13 vivo technique [27, 28]. In addition, it facilitates standardised, uniform specimen processing and assessment, recognising that the principal limitation of this technique is its inability to detect the few (2-10%) cases with aberrant lymphatic drainage [28, 29]. Detection of occult tumour cells Although almost all studies analysing SN mapping for colorectal cancer demonstrate the detection of micrometastases, the clinical significance of this micrometastatic disease is still unknown. A number of studies with respect to this subject have been published with conflicting results. Six of the 12 reported studies found that the presence of nodal micrometastases correlates with significantly worse survival or recurrent disease, whereas the other six show no relation with clinical outcome parameters [5]. Comparison of results between studies is hampered by the substantial variation in detection method, staining protocols and antibodies. Reverse transcription-polymerase chain reaction (RT-PCR) is more sensitive in detecting marker-positive cells and less subject to sampling error and interobserver variation when compared to immunohistochemistry [11, 12]. However, this technique does not allow morphological determination, and will detect isolated tumour cells, hyperplastic epithelial cells and benign marker-positive mesothelial cells with equal sensitivity. In addition, polyclonal antibodies are frequently used, which have been demonstrated to be associated with more false-positive staining [13]. To prevent false positive staining, immunohistochemistry with three different monoclonal antibodies was used to detect micrometastatic disease in this study. Another explanation for the conflicting data is the variation in terminology and definitions of micrometastatic disease. In some studies all nodes with positive immunohistochemical findings are included, while more recent studies consider isolated tumour cells different from micrometastases, as suggested by the revised 6 th edition of the TNM system by the AJCC, since clusters of cells smaller than 0.2 mm seem to have a different level of clinical significance [30]. Although our study was not designed to comment

16 14 on the prognostic significance of micrometastatic disease, the relation between micrometastases and lymphangio invasion does suggest a clinical relevance, especially since a correlation between lymphangio invasion and isolated tumour cells could not be established. Conclusion We show that SN mapping can accurately predict nodal status in patients with colon cancer. Immunohistochemical analysis of the SN detects micometastases in 11% of N0 patients, possibly identifying a patient group that may benefit from adjuvant therapy. The results suggest that SN mapping with focussed histopathological examination should be considered as a mandatory step towards optimal staging in patients with colon cancer. In contrast, SN mapping is less reliable in patients with rectal cancer who underwent TME after pre-operative short course radiotherapy. To improve staging in rectal cancer, other studies should be performed investigating other techniques or tracers. Further studies must focus on univocal definitions of occult tumour cells and survival analysis of patients with accurately staged stage I and II colorectal cancer.

17 15 References 1. International multicentre pooled analysis of B2 colon cancer trials (IMPACT B2). J Clin Oncol 1999; 17: Goldstein NS, Sanford W, Copffey M, et al. Lymph node recovery from colorectal resection specimens removed for adenocarcinoma. Am J Clin Pathol 1996; 106: Joseph NE, Sigurdson ER, Hanlon AL, Wang H, Mayer RJ, MacDonald JS et al. Accuracy of determining nodal negativity in colorectal cancer on the basis of the number of nodes retrieved on resection. Ann Surg Oncol 2003; 10: Saha S, Wiese D, Badin J, Beutler T, Nora D, Ganatra BK, et al. Technical details of sentinel lymph node mapping in colorectal cancer and its impact on staging. Ann Surg Oncol 2000; 7: De Haas RJ, Wicherts DA, Hobbelink MG, Borel Rinkes IH, Schipper ME, Van der Zee JA, Van Hillegersberg R. Sentinel lymph node mapping in colon cancer: current status. Ann Surg Oncol 2007; 14: Kapiteijn E, Marijnen CA, Nagtegaal ID, Putter H, Steup WH, Wiggers T, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001; 345: Marijnen CA, Nagtegaal ID, Klein KE, Hermans J, Van de Velde CJ, Leer JW, et al. No downstaging after short-term preoperative radiotherapy in rectal cancer patients. J Clin Oncol 2001; 19: Nagtegaal ID, Marijnen CA, Kranenbarg EK, Mulder-Stapel A, Hermans J, Van de Velde CJ, et al. Short-term preoperative radiotherapy interferes with the determination of pathological parameters in rectal cancer. J Pathol 2002; 197:20-7.

18 16 9. Greene FL, Page DL, Fleming ID. AJCC Cancer staging manual (ed 6). New York, NY, Springer, Singletary SE, Greene FL, Sobin LH. Classification of isolated tumor cells: Clarification of the 6 th edition of the American Joint Committee on Cancer Staging manual. Cancer 2003; 98: Kelder W, Braat AE, Van den berg A, Platteel I, Hollema H, Groen H, Plukker J. Value of RT-PCR analysis of sentinel nodes in determining the pathological nodal status in colon cancer. Anticancer Res 2007; 27: Doekhie FS, Kuppen PJ, Peeters KC, Mesker WE, van Soest RA, Morreau H, van de Velde CJ, Tanke HJ, Tollenaar RA. Prognostic relevance of occult tumour cells in lymph nodes in colorectal cancer. Eur J Surg Oncol. 2006;32: Borgen E, Beiske K, Trachsel S, Nesland JM, Kvalheim G, Herstad TK, et al. Immunocytochemical detection of isolated epithelial cells in bone marrow: nonspecific staining and contribution by plasma cells directly reactive to alkaline phosphatase. J Pathol 1998;185: Buskens CJ, Ten Kate FJ, Obertop H, Izbicki JR, van Lanschot JJ. Analysis of micrometastatic disease in histologically negative lymph nodes of patients with adenocarcinoma of the distal esophagus or gastric cardia. Dis. Esophagus 2008; 21: Joosten JJ, Strobbe LJ, Wauters CA, et al. Intraoperative lymphatic mapping and the sentinel node concept in colorectal carcinomas. Br J Surg 1999; 86: Saha S, Bilchik A, Wiese D, et al. Ultrastaging of colorectal cancer by sentinel lymph node mapping technique a multicenter trial. Ann Surg Oncol 2001; 8:94S-98S. 17. Wong JH, Steineman S, Calderia C, Bowles J, Namiki T. Ex vivo sentinel node mapping in carcinoma of the colon and rectum. Ann Surg 2001; 233:

19 Tiffet O, Kaczmarek D, Chambonniere ML, Guillan T, Baccot S, Prevot N, Baqeacu et al. Combining radioisotopic and blue-dye technique does not improve the falsenegative rate in sentinel lymph node mapping for colorectal cancer. Dis Colon Rectum 2007; 50 : Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet 1986; ii: Gandy CP, Biddlestone LR, Roe AM, O Leary DP. Intra-operative injection of patent blue V to facilitate nodal staging in colorectal cancer. Colorectal Dis 2002; 4: Topor B, Acland R, Kolodko V, et al. Mesorectal lymph nodes: their location and distribution within the mesorectum. Dis Colon Rectum 2003; 46: Baton O, Lasser P, Sabourin JC, Boige V, Duvillard P, Elias D, Malka D, Ducreux M, Pocard M. Ex vivo lymph node study for rectal adenocarcinoma : preliminary study. World J Surg 2005; 29: Braat AE, Oosterhuis JW, De Vries JE, Tollenaar RA. Lymphatic staging in colorectal cancer: pathologic, molecular, and sentinel node techniques. Dis Colon Rectum 2005; 48: Viehl CT, Hamel CT, Marti WR, et al. Identification of sentinel lymph nodes in colon cancer depends on the amount of dye injected relative to tumor size. World J Surg 2003;27: Stojadinovic A, Nissan A, Protic M, Adair CF, Prus D, et al. Prospective randomized study comparing sentinel lymph node evaluation with standard pathlogic evaluation for the staging of colon carcinoma. Ann Surg 2007; 245: Stojadinovics A, Allen PJ, Protic M, et al. Colon sentinel lymph node mapping: practical surgical applications. J Am Coll Surg 2005; 201:

20 Teuch JJ, Pessaux P, Fiore FD, et al. Sentinel node mapping in coloncarcinoma : invivo versus ex-vivo approach. Eur J Surg Oncol 2006 ; 32 : 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 2001; 8: Patten LC, Berger DH, Rodriquez-Bigas M, et al. A prospective evaluation of radiocolloid and immunohistochemical staining in colon carcinoma lymphatic mapping. Cancer 2004; 100: Messerini L, Cianchi F, Cortesini C, Comin CE. Incidence and prognostic significance of occult tumor cells in lymph nodes from patients with stage IIA colorectal carcinoma. Hum Pathol 2006; 37:

21 1 Table 1 Baseline characteristics of the 132 included patients with colorectal cancer. Characteristics Colon (n=100) Rectum (n=32) p-value Age (yrs) 70.4± ± Gender Male Female Type of resection Open Laparoscopic AJCC Classification Dukes Dukes Dukes Total lymph nodes identified 16.3±8 12.4± Total SN identified 2.3±3 1.1±1 0.04

22 2 Table 2a The predictive value of SN mapping in percentages Overall Colon cancer Rectal cancer p-value N=132 n=100 n=32 Detection rate Accuracy Sensitivity Negative predictive value

23 3 Table 2b Factors influencing SN accuracy in 100 patients with colon cancer and 32 patients with rectal cancer SN correct SN not identified/ incorrect Colon cancer N=100 N=87 N=13 p-value Depth of tumour invasion a T T T Number of positive nodes b N N N2 9 5 Tumour size mean cm Rectal cancer N=32 N=19 N=13 Depth of tumour invasion T T2 8 0 T3 8 9 Number of positive nodes N N1 6 4 N2 2 5 Tumour size mean cm Pre-operative radiotherapy No Yes 12 11

24 4 Table 3 Correlation of SN micrometastases and clinicopathological findings in 73 conventional N0 patients with colorectal cancer. No micrometastases (n=65) Micrometastases (n=8) p-value Depth of tumour invasion T T T Differentiation grade Well Moderate 48 6 Poor 10 2 Lymphangio-invasion No Yes 2 5