Prognostic significance of retrieved lymph nodes per specimenin resected rectal adenocarcinoma after preoperative chemoradiation therapy

Archives of Medical Research 34 (2003) 281–286

ORIGINAL ARTICLE

Prognostic Significance of Retrieved Lymph Nodes per Specimen in Resected Rectal Adenocarcinoma After Preoperative Chemoradiation Therapy Pedro Luna-Pe´rez,a Sau´l Rodrı´guez-Ramı´rez,a Isabel Alvarado,b Marcos Gutie´rrez de la Barreraa and Sonia Labastidac a

Departamento de Oncologı´a Quiru´rgica, Servicio Colorectal, bDepartamento de Patologı´a, cSeccio´n de Estadı´stica, Hospital de Oncologı´a, Centro Me´dico Nacional Siglo XXI (CMN-SXXI), Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico Received for publication October 8, 2002; accepted February 12, 2003 (02/195).

Background. Histologic examination of a regional lymphadenectomy specimen ordinarily should include 12 or more lymph nodes. However, in specimens from patients who received preoperative chemoradiotherapy this number has not yet been established. Methods. From January 1990 to December 2000, 210 patients with rectal adenocarcinoma located between 0 and 10 cm from anal verge with invasion into perirectal fat, tethered or fixed to the pelvis, diagnosed by computed tomography (CT) scan and/or rectal ultrasound were included. All patients received 45 Gy ⫹ bolus infusion of 5-FU (450 mg/ m2/days 1–5, 28–33 of RT) 4–8 weeks after surgery was performed. Specimens were mapped and sliced. Lymph nodes were studied under clearing or manual techniques. Five-year survival was calculated by Kaplan-Meier method and comparison of groups with logrank test. Multivariate Cox regression analysis was performed to find risk factors affecting local control and survival. Results. There were 126 males and 84 females; mean age was 55.2 years. Low anterior resection was performed in 112 patients, abdominoperineal resection in 85, and pelvic exenteration in 13. Total retrieved lymph nodes numbered 2,554, of which 252 contained metastasis. The group was divided into patients with 1–10 retrieved lymph nodes (n ⫽ 119) and patients with ⱖ11 retrieved lymph nodes (n ⫽ 91). Median follow-up was 49 months. Local recurrence was as follows: 15% in patients with specimens containing 1–10 lymph nodes and conversely 7.4% in those with ⱖ11 ( p ⫽ 0.01). Five-year survival of patients with 1–10 lymph nodes was 48%, whereas for those with ⱖ11 lymph nodes it was 69% ( p ⫽ 0.02). Conclusions. Retrieval of at least 11 lymph nodes in the surgical specimen is not only a powerful tool to properly stage patients with rectal adenocarcinoma treated with preoperative chemoradiotherapy and surgery, but it is also of prognostic relevance in that 5-year survival and local recurrence were better in this group of patients. 쑖 2003 IMSS. Published by Elsevier Science Inc. Key Words: Rectal cancer, Neo-adjuvant chemoradiotherapy, Complete response, Lymph node metastasis, Total mesorectal excision.

Introduction

Address reprint requests to: Pedro Luna-Pe´rez, M.D., Calle Puerto Me´xico #53-101, Col. Roma Sur, 06760 Me´xico, D.F., Me´xico. Phone: (⫹52) (55) 5584-2312; FAX: (⫹52) (55) 5564-8000; E-mail: lunapp@ infosel.net.mx

The most important prognostic factor for survival and local recurrence is tumor stage (1). Of paramount importance is quality of surgical technique together with meticulous histologic study of the surgical specimen to obtain adequate information concerning tumor penetration through rectal

0188-4409/03 $–see front matter. Copyright 쑖 2003 IMSS. Published by Elsevier Science Inc. d o i : 1 0 .1 0 16 / S 0 18 8 - 44 0 9 ( 03 ) 0 00 4 1 -9

282

Luna-Pe´rez et al. / Archives of Medical Research 34 (2003) 281–286

wall and number and location of lymph nodes. Total mesorectal excision (2) and use of clearing or manual techniques to properly identify lymph nodes (3–5) makes it possible to achieve these goals. The International Union Against Cancer (IUCC) established that histologic examination of regional lymphadenectomy in specimens not receiving preoperative radiotherapy requires 12 or more lymph nodes (6). The main purpose of this study was to find the prognostic significance of retrieved lymph nodes in specimens of locally advanced rectal adenocarcinoma treated with preoperative chemoradiotherapy plus surgical resection with total mesorectal excision. Materials and Methods Five hundred ninety patients with a mean age of 58.7 ⫾ 14.4 years were treated between January 1990 and December 1999 at the IMSS Oncology Hospital at the 21st Century National Medical Center (CMN-SXXI) in Mexico City. Of these, 210 patients with histologically proven rectal adenocarcinoma were treated with preoperative chemoradiotherapy. Tumor location was between 0 and 10 cm from anal verge with invasion into perirectal fat, tethered or fixed to neighboring pelvic organs or to pelvis wall. Distance between anal verge and distal limit of tumor was determined by rigid proctoscopy with patients placed in jackknife position. Pre-treatment evaluation included medical history, physical examination, complete blood cell count, chemistry profile, determination of carcinoembryonic antigen, chest xray, endorectal ultrasound, computed tomography (CT) of abdomen, pelvis, and perineum, and endorectal ultrasound. Colonoscopy was performed in all patients except in those with rectal tumor stenosis. Inclusion criteria for this study were the following: tumor penetration through muscularis propia and perirectal fat or metastatic lymph nodes; tumors attached to neighboring pelvic organs either tethered or fixed to pelvic sidewall; age ⬍75 years; ECOG performance status 0–2; white blood cell count of at least 4,000/mm2; platelet count of at least 100,000/mm2; normal liver and renal function tests, and curative surgical resections (R0). Exclusion criteria for this study were distant metastatic disease found during surgery, pathologic report without lymph nodes, patients with ⬍12 months of follow-up (except those who died of disease), and patients lost during follow-up. Scheduled treatment. Radiation therapy was delivered with an 8-Mev linear accelerator using either a two- or threefield technique with patients placed in prone position with distended bladder. Top of field was placed at midpoint of the body at L5. Lateral borders were located 1 cm outside bony pelvis and inferior margin was at anal verge. A dose of 45 Gy was administered at 1.8 Gy/day for 5 days per week during 5 consecutive weeks. 5-Fluorouracil at doses of 450

mg/m2 was also administered as bolus infusion 1 h prior to administration of radiotherapy on days 1–5 and 28–32. Four weeks after completion of chemoradiotherapy, restaging procedures were performed and included the following: physical examination; CT of abdomen, pelvis, and perineum; chest x-ray; complete blood cell count; biochemical profile, and rectosigmoidoscopy or full colonoscopy. In all patients, surgery was performed 4–6 weeks after chemoradiotherapy completion. Abdominoperineal resection was performed on patients with tumors confined to rectum and located between 0 and 5–6 cm from anal verge. Low anterior resection was performed on patients with tumors confined to rectal wall located between 6 and 10 cm from anal verge or lower. Ultralow anterior resection with complete mesorectal excision and colo-anal anastomosis was performed on selected patients with tumor located between 3 and 5 cm from anal verge and with clinical response to induction therapy. Posterior wall vaginectomy was performed on some females if tumor had attachment to vagina. Posterior or total pelvic exenteration was performed when a large tumor was attached to uterus, bladder, prostate, or seminal vesicles. Specimens were mapped and sliced. Lymph nodes were retrieved using clearing or manual technique (4,5); lymph nodes were bisected or sectioned at 2–3-mm intervals; paraffin-embedded, formalin-fixed tissue blocks were cut at a thickness of 4 µm, stained with hematoxylin and eosin, and staged using American Joint Committee on Cancer (AJCC) tumor node metastasis (TNM) classification (7). During the first postoperative year, all patients were examined at 2-month intervals, during the second year at 3month intervals, at 4-month intervals during the third year, and thereafter at least annually. Follow-up evaluation included physical examination, liver function tests, carcinoembryonic antigen (CEA), chest x-ray, liver ultrasound, and annual colonoscopy. For abnormalities in liver function tests or levels of CEA ⬎10 ng, CT scan of abdomen and pelvis was performed. For purposes of this study, the last followup was carried out when the patient died or in June 2001, whichever occurred first. Local recurrence was defined as originating in abdominal surgical area, pelvic sidewall, and/or regional lymph nodes and was biopsy-proven or corroborated by CT. All other recurrences were defined as distant. Cancer-specific survival was analyzed by Kaplan-Meier method. Differences in survival rates were analyzed using log-rank test. Multivariate Cox proportional hazard regression model was fitted to explore relative risk (RR) of local recurrence and cancer-related death. Significance level was p ⬍0.05 in all statistical tests. Results There were 126 males and 84 females; mean age was 55.2 ⫾ 13.9 years. Tumor location was as follows: 0–5 cm

Lymph Node Number Seen After Postoperative Chemoradiotherapy

from anal verge (n ⫽ 108) and between 5.1 and 10 cm (n ⫽ 102). Histologic differentiation was well (n ⫽ 64), moderate (n ⫽ 136), and poor (n ⫽ 10). Surgeries performed were low anterior resection (n ⫽ 112), abdominoperineal resection (n ⫽ 85), and pelvic exenteration (n ⫽ 13). Specimens were studied by modified clearing technique (n ⫽ 74) and manual technique (n ⫽ 136). Overall number of retrieved lymph nodes in 210 specimens was 2,554 (average 12.1/specimen, range 1–59); mean of 22 lymph nodes was obtained with use of fat clearing technique, whereas mean of eight lymph nodes was obtained in specimens studied with manual technique ( p ⫽ 0.001); 252 (9.8%) of these contained metastases. Lymph node metastasis was observed in 72 of 210 specimens (34.2%). Postoperative pathologic stages are shown in Table 1. Patients were grouped according to number of retrieved lymph nodes per specimen: those with 1–10 lymph nodes (n ⫽ 119) and those with ⱖ11 lymph nodes (n ⫽ 91). Median follow-up was 46 months. Local recurrences were observed in 25 patients (11.8%) at median of 17 months (range 7–53 months). Overall 5-year cancer-specific survival was 57.4%. Five-year cancer-specific survival in the former group was 48%, and conversely 69.6% in the latter group ( p ⫽ 0.02) as shown in Figure 1. Local recurrences according to pathologic tumor stage are shown in Table 2. Patients with pT0–2,N0,M0 had 1% of local recurrences, whereas there were 12.2 and 19.7% of patients with pT3–4,N0,M0 and pT2–4, N⫹, M0, respectively ( p ⫽ 0.001). Local recurrence according to retrieval of lymph nodes/specimen was as follows: 18/119 (15.1%) in patients with 1–10 lymph nodes, and conversely 7/91 (7.4%) in those with ⱖ11 lymph nodes (p ⫽ 0.01). Distribution of metastatic lymph nodes according to retrieval of lymph nodes was as follows: 1–10 lymph nodes 39/119 (32.8%), whereas in patients with ⱖ11 lymph nodes it was 33/91 (36.3%) ( p ⫽ 0.35). Number of specimens with N⫹ Table 1. Postradiated pathologic stage classificationa Stage

Number of patients (%)

0 PT0,N0,M0

11 (5.2)

PT1,N0,M0 PT2,N0,M0

12 (5.7) 34 (16.2)

PT3,N0,M0 PT4,N0,M0 III PT0–2,N1,M0 PT3,N1–2,M0 PT4,N1–2,M0 Total

73 (34.8) 9 (4.3)

I

II

a

AJCC TNM classification.

10 56 5 210

(4.8) (26.6) (2.4) (100)

283

Figure 1. Five-year survival as a function of retrieved lymph nodes.

(metastatic lymph nodes) according to number of retrieved lymph nodes and AJCC classification was as follows: N1– 2 in 1–10 retrieved lymph nodes were 22 and 17, respectively. In the group with ⱖ11 lymph nodes, there were 20 and 13, respectively ( p ⫽ 0.45). Five-year survival in 1–10 lymph-nodes group and in the ⱖ11 lymph group according to N⫹ (1–2) status was as follows: N1 (28 and 69%, respectively) ( p ⫽ 0.04), whereas in N2 it was (0 and 19%, respectively) (p ⫽ 0.74) (Figure 2). Risk factors for local recurrences are shown in Table 3, while risk factors for cancer-related deaths are depicted in Table 4.

Discussion The most important prognostic factor determining patient survival after curative resection for rectal cancer is tumor stage. Strong evidence supports that in rectal cancer, surgically treated tumor penetration to perirectal fat and presence of nodal metastases are the most important factors in determining whether an individual is a candidate for adjuvant therapy (8–10). In patients treated with preoperative therapy, response evaluated in the surgical specimen has been identified as an independent prognostic factor (11). Unfortunately, there is no consensus to administration of postoperative chemotherapy. The American Joint Committee on Cancer and the International Union Against Cancer distinguish between patients with 1–3 metastatic lymph nodes and those with four or more (6,7). Undetected metastatic lymph node may contribute to lack of multidisciplinary treatment Table 2. Local recurrence according to postradiated pathologic stagea Stage

Number of patients

pT0–2,N0,M0 pT3–4,N0,M0 pT0–2,N⫹,M0 Total a

AJCC TNM classification.

57 82 71 210

Local recurrence (%) 1 10 14 25

(1.8) (12.2) (19.7) (11.9)

284

Luna-Pe´rez et al. / Archives of Medical Research 34 (2003) 281–286 Table 4. Cox regression analysis for cancer-related death Covariate Stage II Stage III Retrieved ⱖ11 lymph nodes/specimen

β

RR

p

1.4 2.2 ⫺0.6

4.3 9.5 0.5

0.0007 0.00001 0.008

RR ⫽ relative risk.

Figure 2. Five-year survival as a function of metastatic lymph node status and retrieved lymph nodes.

in patients with high recurrence risk. To define this question, a number of investigators have sought to determine the number of lymph nodes that must be retrieved to reflect accurately status of regional basin (12,13). Number of lymph nodes assessed pathologically is the result of a combination of quality of surgical technique and meticulous histologic study of surgical specimen to obtain adequate tumor stage information. Total mesorectal excision (2) and use of clearing or manual techniques to identify lymph nodes (3–5) makes it possible to achieve this goal. Kotanagi et al. (14) reported a mean of 23 lymph nodes retrieved per specimen in patients with rectal cancer without neoadjuvant therapy performing intensive manual dissection. Andreola et al. (4) confirmed the previous observation and reported a mean of 36 lymph nodes per specimen using manual dissection. Scott and Grace (15) reported that at least 13 lymph nodes should be harvested; when this was not possible, the authors suggested use of fat clearing technique to obtain a minimum of 13 lymph nodes. Herrera and Villarreal (16) performed this technique and reported an average of 34 lymph nodes per specimen. Our group (5) retrieved a mean of 30.9 lymph nodes using modified clearing technique. Wong et al. (17) suggested that at least 14 lymph nodes should be studied after resection of T2 or T3 colorectal carcinoma; Tepper et al. (18) confirmed this observation. Goldstein et al. (13) recommended that at least 17 lymph

Table 3. Cox regression analysis of risk factors for local recurrences Covariate Stage II Stage III Well-differentiated Tumor location (0-5 cm)a Retrieved ⱖ11 lymph nodes/specimen RR ⫽ relative risk. a From the anal verge.

β

RR

p

2.4 3.2 ⫺1.0 1.2 ⫺0.8

11.0 25.4 0.3 3.4 0.4

0.002 0.0001 0.01 0.01 0.04

nodes should be recovered in specimens with colorectal carcinoma. Wichmann et al. (19) demonstrated that administration of preoperative chemoradiotherapy had an influence on lymph node retrieval after resection of rectal cancer. They demonstrated a mean of 19 lymph nodes/specimen in patients treated with surgery alone and a mean of 13 lymph nodes in patients treated with preoperative chemoradiotherapy. Cawthorn et al. (20) reported a series of 272 patients randomized into a trial of preoperative radiotherapy at nine centers in the UK. Their results demonstrated that usage of clearing technique or meticulous assessment of surgical specimen identified not only a greater number of lymph nodes (mean ⫽ 23 and 13, respectively) but also more lymph nodes containing metastases than centers not employing these techniques (mean ⫽ 10.5) ( p ⫽ 0.001). This statement is consistent with results of the current study in which a mean of 22 lymph nodes was obtained with use of fat clearing technique or intensive dissection, whereas a mean of eight lymph nodes was obtained in specimens studied with manual technique ( p ⫽ 0.001). Some authors reported alternative methods to improve staging of lymph nodes contained in surgical specimen, such as serial sections stained with hematoxylin and eosin, immunohistochemical examination, or reverse transcriptasepolymerase chain reaction (PCR) techniques. However, the biological significance of these techniques remains controversial (21–23), perhaps due to diversity in defining micrometastatic disease, the retrospective nature of these series, and the fact that most series based their studies on patients with stage II in an attempt to upstage. In colon cancer, intraoperative dye-direct mapping identifies the sentinel lymph node that has the highest likelihood of harboring metastasis if present; furthermore, using the previously mentioned techniques could enhance the ability to detect metastasis and accurately predict status of regional node basin (24). However, to date there is limited prospective data on the prognostic role of micrometastases. Furthermore, the AJCC cancer staging manual makes a clear distinction between micrometastasis and isolated tumor cells. Patients whose lymph nodes contained isolated tumor cells are classified as N0 and patients with nodal micrometastasis (0.2–2 mm), N1. Despite the fact that these statements do not apply to postradiated specimens, identification of sentinel lymph nodes with ex vivo dye technique (25) and selectively performed intensive workup of these nodes with immunohistochemical stains probably results in better staging than routine histopathologic assessment of all retrieved lymph

Lymph Node Number Seen After Postoperative Chemoradiotherapy

nodes. To support this proposal, Wong et al. (26) recently reported that volume of metastases is not as important as overall number of lymph nodes involved. Results of the current study suggest that the final stage is dependent on number of retrieved lymph nodes and is a powerful prognostic covariate for local recurrence and survival as demonstrated in multivariate analysis. These results are consistent with our previous report (27). In AJCC and International Union Against Cancer (IUCC) stage classifications (6,7) concerning nonradiated rectal cancer, lymph nodes are classified according to number involved with metastatic tumor. In patients with colon cancer, 5-year survival rate with 1–3 positive nodes (N1) and with four or more positive nodes (N2) was 66 and 37%, respectively (8). In the current series, significant differences in 5year survival were found in N1 positive group of patients according to number of lymph nodes retrieved per specimen. In N1 group with 1–10 lymph nodes, it was 28%, whereas it was 69% in the group with ⱖ11 lymph nodes ( p ⫽ 0.04). However, no difference was found in patients with N2. These results corroborate the number of retrieved lymph nodes per specimen properly staged and support the fact that appropriate staging could allow patients to be stratified for more aggressive adjuvant chemotherapy therapy such as the combination of 5-FU continuous infusion ⫹ CPT-11 or oxaliplatinum in patients with metastatic lymph nodes after preoperative chemoradiotherapy. Furthermore, the group of undersampled patients should be considered for adjuvant chemotherapy. Other investigators also have validated that number of metastatic lymph nodes is a prognostic factor. Jass et al. (28) reported 379 patients with surgically treated rectal cancer. Five-year survival among patients with a score of III (number of metastatic lymph nodes [1–4 vs. ⱖ4], character of invasive margin, peritumoral lymphocytic infiltration, and local spread) was 56%, whereas in patients with a score of 4–5 the rate was 27%. Bognel et al. (29) reported 339 patients treated with surgery; 161 received radiotherapy either pre- or postoperatively or in combination. Five-year disease-free survival with 1–2 lymph node metastases was 46, and 26% with ⱖ3. Preoperative chemoradiotherapy is associated with downsizing and downstaging locally advanced rectal carcinoma. Furthermore, some investigators have associated complete response in primary tumor and presence of lymph node metastases as important prognostic factors for both local recurrence and survival. Onaitis et al. (30) reported 141 patients with median follow-up of 27 months and found that the sole independent covariate for local recurrence and survival was presence of lymph node metastases. These findings are consistent with results obtained in the current series, in which the most powerful indicator for local recurrence and survival was tumor stage as demonstrated by number of lymph node metastases retrieved in postradiated surgical specimen. Recently, Goldstein (31) reported that

285

the probability of identifying a single lymph node metastasis was 0.25, and 0.46 when 12 or 18 lymph nodes were retrieved, respectively. Cserni et al. (32) reported no cut-off value on number of examined lymph nodes for adequate staging; they suggested that to obtain adequate staging, as many lymph nodes as possible should be retrieved and assessed. Because to date there is no strong evidence of the prognostic role of exhaustive examination of sentinel nodes in rectal cancer treated with preoperative chemoradiotherapy, these results support that an intensive search for lymph nodes in the specimen is the goal to obtain proper staging and not a therapeutic effect. Node-negative patients with a large number of retrieved lymph nodes can be considered truly node-negative, whereas patients with a small number of retrieved lymph nodes could be considered understaged. In conclusion, retrieval of at least 11 lymph nodes in surgical specimen is a powerful tool to stage patients with rectal adenocarcinoma treated with chemoradiotherapy and surgery due to the fact that proper staging is the main predictor for local recurrence and survival.

References 1. Dukes CE. Cancer of the rectum. An analysis of 1,000 cases. J Pathol Bacteriol 1940;50:527–539. 2. MacFarlane JK, Ryall RDH, Heald RJ. Mesorectal excision for rectal cancer. Lancet 1993;341:457–460. 3. Herrera L, Justiniano J, Castillo N, Petrelli NJ, Stulc JP, Mittelman A. Metastases in small lymph nodes from colon cancer. Arch Surg 1987;122:1253–1256. 4. Andreola S, Leo E, Belli F, Bufalino R, Tomasic G, Lavarino C, Baldini MT, Meroni E. Manual dissection of adenocarcinoma of the lower third of the rectum specimens for detection of lymph nodes smaller than 5 mm. Cancer 1996;77:607–612. 5. Sa´nchez W, Luna-Pe´rez P, Alvarado I, Labastida S, Herrera L. Modified clearing technique to identify lymph node metastases in post-irradiated surgical specimens from rectal adenocarcinoma. Arch Med Res 1996;27:31–36. 6. Sobin LH, Wittekind C, editors. IUCC TNM classification of malignant tumors. Colon and rectum. 6th ed. New York: Wiley-Liss & Sons, Inc., International Union Against Cancer;2002. pp. 72–76. 7. Fleming ID, Cooper JS, Henson DE, Hutter RVP, Kennedy BJ, Murphy GP, O’Sullivan B, Sobin LH, Yarbro JW, editors. AJCC cancer staging handbook. Colon and rectum. 5th ed. Philadelphia, PA, USA: Lippincott-Raven, American Joint Committee on Cancer;1998. pp. 81–88. 8. Cohen AM, Tremiterra S, Candela F, Thaler HT, Sigurdson ER. Prognosis of node-positive colon cancer. Cancer 1991;67:1859–1861. 9. Krook JE, Moertel CG, Gunderson LL, Wieand HS, Collins RT, Beart RW, Kubista TP, Poon MA, Meyers WC, Mailliard JA, Twito DI, Morton RF, Veeder MH, Witzig TE, Cha S, Vidyarthi SC. Effective surgical adjuvant for high risk rectal carcinoma. N Engl J Med 1991;324:709–715. 10. O’Connell MJ, Martenson JA, Wieand HS, Krook JE, Macdonald JS, Haller DG, Mayer RJ, Gunderson LL, Rich TA. Improving adjuvant therapy for rectal cancer by combining protracted-infusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 1994; 331:502–507. 11. Janjan NA, Abbruzzese J, Pazdur R, Khoo VS, Cleary K, Dubrow R, Ajani J, Rich TA, Goswitz MS, Evetts PA, Allen PK, Lynch PM, Skibber JM. Prognostic implications of response to preoperative

286

12.

13.

14.

15.

16.

17.

18.

19.

20. 21.

Luna-Pe´rez et al. / Archives of Medical Research 34 (2003) 281–286 infusional chemoradiation in locally advanced rectal cancer. Radiother Oncol 1999;51:153–160. Fielding LP, Aresnault PA, Chapuis PH, Dent O, Gathright B, Hardcastle JD, Hermanek P, Jass JR, Newland RC. Clinicopathological staging for colorectal cancer: an International documentation system (IDS) and an International comprehensive anatomical terminology (ICAT). J Gastroenterol Hepatol 1991;6:325–344. Goldstein NS, Sanford W, Coffey M, Layfield LJ. Lymph node recovery from colorectal resection specimens removed for adenocarcinoma. Trends over time and a recommendation for minimum number of lymph nodes to be recovered. Am J Clin Pathol 1996;106:209–216. Kotanagi H, Fukuoka T, Shibata Y, Yoshioka T, Aizawa O, Saito Y, Tur GE, Koyama K. The size of regional lymph nodes does not correlate with the presence or absence of metastasis in lymph nodes in rectal cancer. J Surg Oncol 1993;54:252–254. Scott KWM, Grace RH. Detection of lymph node metastases in colorectal carcinoma before and after fat clearance. Br J Surg 1989;76: 1165–1167. Herrera L, Villarreal JR. Incidence of metastases from rectal adenocarcinoma in small lymph nodes detected by a clearing technique. Dis Colon Rectum 1992;35:783–788. Wong JH, Severino R, Honnebier MB, Tom P, Namiki TS. Number of nodes examined and staging accuracy in colorectal carcinoma. J Clin Oncol 1997;17:2896–2900. Tepper JE, O’Connell MJ, Niedzwiecki D, Hollis D, Compton C, Benser AB 3rd, Cummings B, Gunderson L, Macdonald JS, Mayer RJ. Impact of nodes retrieved on outcome in patients with rectal cancer. J Clin Oncol 2001;19:157–163. Wichmann MW, Mu¨ller C, Meyer G, Strauss T, Hornung HM, LauWerner U, Angele MK, Schildberg FW. Effect of preoperative radiochemotherapy on lymph node retrieval after resection of rectal cancer. Arch Surg 2002;137:206–210. Cawthorn SJ, Gibbs NM, Marks CG. Clearance technique for the detection of lymph nodes in colorectal cancer. Br J Surg 1986;73:58–60. Adell G, Boeryd B, Franlund B, Sjodahl R, Hakansson L. Occurrence and prognostic importance of micrometastases in regional lymph nodes in Dukes’ B colorectal carcinoma: an immunohistochemical study. Eur J Surg 1996;162:637–642.

22. Liefers GJ, Clenton-Jensen AM, Van de Velde CJH, Hermans J, Van Krieken JH, Cornelisse CJ, Tollenaar RA. Micrometastases and survival in stage II colorectal cancer. New Engl J Med 1998;339:223–228. 23. Tschmelitsch J, Klimstra DS, Cohen AM. Lymph node micrometastases do not relapse in stage II colon cancer. Ann Surg Oncol 2000;7:60–608. 24. Bilchik AJ, Nora DT. Lymphatic mapping of nodal micrometastasis in colon cancer: putting the cart before the horse? Ann Surg Oncol 2002;9:529–531. 25. Fitzgerald TL, Khalifa MA, Law CHL, Smith AJ. Ex vivo sentinel lymph node biopsy in colorectal cancer: a feasibility study. J Surg Oncol 2002;80:27–32. 26. Wong JH, Steinemann S, Paul T, Morita S, Tauchi-Nishi P. Volume of lymphatic metastases does not independently influence prognosis in colorectal cancer. J Clin Oncol 2002;20:1506–1511. 27. Luna-Pe´rez P, Trejo-Valdivia B, Labastida S, Garcı´a-Alvarado S, Rodrı´guez DF, Delgado S. Prognostic factors in patients with locally advanced rectal adenocarcinoma treated preoperative radiotherapy and surgery. World J Surg 1999;23:1069–1075. 28. Jass JR, Love SB, Northover JMA. A new prognostic classification of rectal cancer. Lancet 1987;i:1303–1306. 29. Bognel C, Rekacewicz C, Mankarios H, Pignon JP, Elias D, Duvillard P, Prade M, Ducreux M, Kac J, Rougier P. Prognostic value of neural invasion in rectal carcinoma: a multivariate analysis of 339 patients with curative resection. Eur J Cancer 1995;31A:894–898. 30. Onaitis MW, Noone RB, Hartwig M, Morse M, Jowell P, McGrath K, Lee C, Anscher MS, Clary B, Mantyh C, Pappas TN, Ludwig K, Seigler HF, Tyler DS. Neoadjuvant chemoradiation for rectal cancer: analysis of clinical outcomes from 13-year institutional experience. Ann Surg 2001;233:778–785. 31. Goldstein NS. Lymph node recoveries from 2427 pT3 colorectal resection specimens spanning 45 years: recommendations for a minimum number of recovered lymph nodes based on predictive probabilities. Am J Surg Pathol 2002;26:179–189. 32. Cserni G, Vinh-Hung V, Burzykowski T. Is there a minimum number of lymph nodes that should be histologically assessed for a reliable nodal staging of T3N0M0 colorectal carcinomas? J Surg Oncol 2002;81:63–69.