J Korean Surg Soc 2011;80:10-15 DOI: 10.4174/jkss.2011.80.1.10 원 저 1 cm 이하의침윤성유방암환자에서겨드랑이림프절전이에대한예측인자 계명대학교의과대학외과학교실유방내분비분과 강선희ㆍ조지형 Predictive Factors Affecting Axillary Lymph Node Metastasis in Patients with Invasive Breast Carcinoma of 1 cm or Less Sun Hee Kang, Jihyung Cho Breast Endocrine Division, Department of Surgery, Keimyung University School of Medicine, Daegu, Korea Purpose: Although screening MMG leads to increase of early small breast cancer, axillary lymph node metastasis is still an important prognsotic factor in these patients. The aim of this study is to evaluate the incidence and predictors for axillary lymph node metastasis in patients with invasive breast carcinoma of 1 cm or less. Methods: A retrospective analysis was made of 144 patients who underwent resection of primary tumor and axillary procedures between January 1999 and August 2009 for breast cancer of 1 cm or less in size. Patients were divided into two groups according to axillary node metastasis and clinicopathologic factors including age, palpable mass during physical examination, location of tumor, multifocality, tumor size, histologic type, extensive in situ component, histologic grade, nuclear grade, lymphovascular invasion, hormonal receptor status, and C-erbB-2 status were compared. Results: Twenty-eight (19.4%) patients of all 144 patients had metastasis in the axillary lymph node. Three variables such as multifocality (P=0.023), histologic high grade (P=0.033), presence of lymphovascular invasion (P=0.002) were found to be significant in univariate analysis. In a multivariate logistic regression analysis, however, multifocality (P=0.022) and presence of lymphovascular invasion (P=0.007) were independent predictors of axillary lymph node metastasis. Conclusion: The incidence of axillary lymphnode metastasis of breast cancer 1 cm or less in size was 19.4%. Although the size of invasive breast carcinoma is less than 1 cm, if the tumor presented lymphovascular invasion or multifocality, axillary lymph node dissection might prove better than sentinel node biopsy. (J Korean Surg Soc 2011;80:10-15) Key Words: Breast cancer, Less than 1 cm size, Axillary node metastasis, Predictive factor 중심단어 : 유방암, 1 cm 이하크기, 겨드랑이림프절전이, 예측인자 서 론 책임저자 : 강선희, 대구시중구동산동 194 700-712, 계명대학교의과대학외과학교실 Tel: 053-250-8027, Fax: 053-250-7322 Email : shkang9002@dsmc.or.kr 접수일 :2010 년 9 월 19 일, 게재승인일 :2010 년 11 월 17 일본연제는 2008 년대한외과학회학술대회에서구연발표하였음. 한국인의유방암발생률은지난과거에비해급속도로증가하면서, 특히무증상의상태로건강검진상발견되는상피내암또는침윤성조기유방암이많은부분을차지하게되었다. 최근보고에의하면종양의크기가 2 cm 이하이 10
Sun Hee Kang and Jihyung Cho:Predictor for Axillary Node Metastsis in Breast Cancer of 1 cm or Less 11 며림프절전이가없는 1기유방암이 37.9% 를차지하여과거 10년전에비하여 93.4% 의증가율을기록하였다.(1) 이들의수술적치료는감시림프절생검을포함한유방보존술이주류를이루어유방의미용적효과를최대화하고, 겨드랑이합병증을줄여기능적손실을피할수있다. 감시림프절생검술이초기유방암에서고식적겨드랑이림프절박리술을대신할수있다는문헌들 (2,3) 과영상의학적기술의발달로수술전에림프절전이상태를보다정확하게진단함으로써 (4-6) 겨드랑이림프절박리술의빈도는점점줄어들고있다. 그러나 T1 유방암에서겨드랑이림프절전이율은약 3 37%(7-11) 으로보고되는데림프절전이에영향을주는가장중요한인자는종양의크기이다. 즉, 5 mm 이하의 T1a 종양은림프절전이가약 0 11% 이나, 10 mm 이상의종양은 29 36% 로증가한다.(12-15) 본연구는 T1 종양중에서도크기가 1 cm 이하인유방암환자를대상으로이들의겨드랑이림프절전이에영향을주는임상병리학적인자를연구하였다. 방법 1999년 8월부터 2009년 8월까지침윤성유방암으로본원에서수술치료를받은환자들중종양의병리학적크기가 1 cm 이하이면서감시림프절생검술또는겨드랑이림프절박리술을받은총 144명을대상으로의무기록과병리조직결과를바탕으로후향적조사를하였다. 전이된림프절개수와는상관없이 1개이상전이된군과전이가없는군으로나누어임상병리학적특징을비교하였다. 분석에사용된인자들은나이, 신체검사상종양의촉지성, 종양의위치, 종양의크기, 다발성, 조직학적유형, 광범위상피내암 (extensive in situ component, EIC) 존재, 조직학적등급, 핵등급, 림프관및혈관침범, 에스트로겐수용체 (Estrogen receptor, ER), 프로게스테론수용체 (Progesteron receptor, PR), C-erbB-2의발현등이다. 종양의크기는 7th AJCC (American Joint Committee on Cancer) 의분류에따라침윤성병변의크기가 0.1 cm 이하인 T1mic, 0.1 cm 초과 0.5 cm 이하의 T1a, 0.5 cm 초과 1.0 cm 이하의 T1b 환자로나누었고, 종양의다발성 (multiple lesion) 은병리학적결과지를근거로구역 (quadrant) 의분포와는관계없이두개이상의침윤성병변이절제한유방조직내에존재하는경우로정의하였다. 종양의위치는병리학적결과지에근거하여상외측과나머지구역으로구분하였다. 조직학적등급과핵등급은 Bloom-Richardson grading system에따라점수를매겨등급화하고 1, 2등급을저등급, 3등급을고등급으로두단계분류하였다. ER, PR, C-erbB-2 발현상태는면역조직화학염색법 (immunohistochemistry) 을이용하여 ER, PR은핵염색이 10% 이상일때양성, C-erbB-2는세포막전체염색이 10% 이상일때양성으로정의하였다. ER 검사에서사용된항체는 NCL-ER-6F11 (Novocastra laboratories Ltd. Newcastle Upon Tyne, UK) 이며, PR은 NCL-PGR (Novocastra laboratories Ltd., Newcastle Upon Tyne, UK) 항체를사용하였다. ER 또는 PR 둘중한개라도양성이면호르몬수용체양성으로, 두수용체모두음성이면호르몬수용체음성으로분류하였다. C-erbB-2에대한검사는 2007년이전에는 NCL- CB11 (Novocastra laboratories Ltd, Newcastle Upon Tyne, UK) 를사용하였고, 이후에는 BenchMark R XT (Ventana medical system, USA) 자동면역염색기기를이용하였다. 일차항체는 polyclonal rabbit anti-human c-erbb-2 oncoprotein (A0485, DakoCtyomation,Glostrup, Denmark) 을 1:1,000으로희석하여사용하였다. 겨드랑이림프절전이유무에따른두군의비교에서단변량분석은변수의특성에따라 chi-square test 또는 independent t test를이용하였고, 다변량분석은 multiple logistic regression test를시행하였다. SPSS version 17.0 (SPSS Inc., Chicago, IL, USA) 을사용하여신뢰구간 (confidence interval) 95%, P-value 0.05 이하일때통계적의미가있다고정의하였다. 결과 겨드랑이림프절전이는전체 144명의환자중 28명 (19.4%) 에서관찰되었다. 평균나이는 49.9세이며, 신체검사상에촉지성종양이 75명 (52.4%) 이고, 상외측에종양이위치하였던환자는 60명 (24.7%) 이었다. 다발성종양은 11명 (7.6%) 에서관찰되었으며, 종양의평균크기는 6.8 mm이고, T1mic 17명 (11.8%), T1a 29명 (20.1%), T1b 98명 (68.1%) 이었다. 침윤성관암이 125명 (86.8%), 침윤성엽상암이 5명 (3.5%), 기타조직유형이 14명 (9.7%) 이었다. EIC를보인경우는 102명 (70.8%) 이며, 조직학적등급이 3등급이었던경우는 45명 (36%), 핵등급이 3등급이었던경우는 58명 (46%) 이었다. 림프관및혈관침범은 27명 (23.5%) 에서관찰되었다. 호르몬수용체양성은 104명 (73.3%), 음성은 38명 (26.4%), C-erbB-2 과발현은 26명 (19%) 에서나타났다 (Table 1).
12 J Korean Surg Soc. Vol. 80, No. 1 Table 1. Clinicopathologic characteristics Variables Number (%) Age (year, mean±sd) 49.96±9.5 Palpable tumor (n=143) Palpable 75 (52.4) Non-palpable 68 (47.6) Location of tumor (n=128) Upper outer quadrant 60 (42.7) Others 68 (47.2) Multifocality of tumor (n=144) Single 133 (92.4) Multiple 11 (7.6) Size of tumor (mm, mean±sd) 6.7±3 T stage (n=144) T1mic 17 (11.8) T1a 29 (20.1) T1b 98 (68.1) Histologic type (n=144) Invasive ductal carcinoma 125 (86.8) Invasive lobular carcinoma 5 (3.5) Others 14 (9.7) Extensive in situ component (n=144) Present 102 (70.8) Absent 42 (29.2) Histologic grade (n=125) Low 80 (64.0) High 45 (36.0) Nuclear grade (n=126) Low 68 (54.0) High 58 (46.0) Lymphovascular invasion (n=115) Present 27 (23.5) Absent 88 (76.5) Hormonal receptor status (n=142) Positive 104 (73.3) Negative 38 (26.7) C-erbB-2 status (n=137) Negative 111 (81.0) Positive 26 (19.0) 단변량분석에서겨드랑이림프절전이에영향을주는임상병리학적특징은다발성종양 (17.9% vs 5.2% P=0.023), 높은조직학적등급 (53.8% vs 31.3% P=0.033), 림프관및혈관침범 (46.2% vs 16.9% P=0.002) 이었다. 나이, 종양의촉지여부, 종양의위치, 종양의크기, 조직학적유형, 광범위상피내암의존재여부, 핵등급, ER, PR, C-erbB-2의발현정도는통계학적유의성이없었다 (Table 2). 다변량분석은단변량분석에서유의성을보인림프관및혈관침범, 다발성, 조직학적등급등의세가지인자로시행하였다. 그결과겨드랑이림프절전이에영향을주는 Table 2. Univariate analysis according to axillary lymph node metastasis Variables Negative node group Positive node group P-value Age (year, mean±sd) 50.66±9.9 47.04±7.03 0.070* Palpable tumor Palpable 56 (48.7%) 19 (67.9%) 0.069 Non-palpable 59 (51.3%) 9 (32.1%) Location of tumor Upper outer quadrant 47 (44.8%) 13 (50%) 0.631 Others 58 (55.2%) 13 (50%) Multifocality of tumor Single 110 (94.8%) 23 (82.1%) 0.023 Multiple 6 (5.2%) 5 (17.9%) Size of tumor (mean, mm) 6.5±3.2 7.4±2.3 0.140* T stage T1mic 17 (14.7%) 0 (0%) 0.096 T1a 23 (19.8%) 6 (21.4%) T1b 76 (65.5%) 22 (78.6%) Histologic type Invasive ductal carcinoma 98 (84.5%) 27 (96.5%) 0.231 Invasive lobular carcinoma 5 (4.3%) 0 (0%) Others 13 (11.2%) 1 (3.5%) Extensive in situ component Present 83 (71.6%) 19 (67.9%) 0.699 Absent 33 (28.4%) 9 (32.1%) Histologic grade Low 68 (68.7%) 12 (46.2%) 0.033 High 31 (31.3%) 14 (53.8%) Nuclear grade Low 57 (57%) 11 (42.3%) 0.181 High 43 (43%) 15 (57.7%) Lymphovascular invasion Positive 15 (16.9%) 12 (46.2%) 0.002 Negative 74 (84.1%) 14 (53.8%) Hormonal receptor Positive 84 (73.7%) 20 (71.4%) 0.809 Negative 30 (26.3%) 8 (28.6%) C-erbB-2 Negative 90 (81.8%) 21 (77.8%) 0.631 Positive 20 (18.2%) 6 (22.2%) *independent t- test; Pearson s chi-square test; Positive = ER and/or PR positive; Negative = ER and PR negative. 독립적인자는종양의다발성과림프관및혈관침범이었다. 다발성종양은상대위험도 5.93으로단발성종양에비해약 6배더높은위험도를보여주었고 (P=0.022), 림프관및혈관침범또한상대위험도가 4.12로그렇지않은군에비해약 4배의높은위험도를보여주었다 (P=0.007). 그러나단변량분석에서의미가있었던조직학적등급은다변량분석에서는통계적의미가없었다 (P=0.236)(Table 3).
Sun Hee Kang and Jihyung Cho:Predictor for Axillary Node Metastsis in Breast Cancer of 1 cm or Less 13 Table 3. Multivariate analysis of variables predicting axillary lymph node metastasis Variables β-coefficient HR* 95% CI P-value Presence of lymphovascular invasion 1.416 4.122 1.47 11.54 0.007 Multifocality of tumor 1.782 5.939 1.30 27.15 0.022 Histologic high grade 0.598 1.818 0.68 4.88 0.236 *HR = hazard ratio; CI = confidence interval. 고찰 한국인유방암의시대적발병률에따르면, 과거 10년전에비해조기유방암의비율이월등히높아졌다.(1) 특히선별검사로서유방촬영술이대중화됨에따라덩이 (lump) 를형성하지않고미세석회화 (microcalcification) 만을보이는단계에서침윤성유방암이진단되기도하고, 초음파또는자기공명영상 (MRI) 장비의발달로신체검사상촉지되지않은작은크기의종양도쉽게찾을수있게되었다. 이러한영상의학적도구의진보는수술하기전에겨드랑이림프절전이의유무를예측하기도하는데, 특히 PET CT를사용하여다양한연구결과들이있다. Veronesi 등 (4) 과 Lovrics 등 (5) 은 FDG-PET 결과와감시림프절생검을비교하여 96 97% 의특이도 (specificity) 와 88 97% 의양성예측도 (positive predictive value) 를보고하였다. 즉, 술전에 FDG-PET 결과림프절전이가없다면감시림프절생검을실시하여림프절병기를결정하고, 반대로 FDG-PET 상에열점 (hot uptake) 이보인다면감시림프절생검보다는오히려고식적액와림프절박리술을바로실시하기를권유하였다. Ahn 등 (6) 은 FDG-PET CT와초음파를추가한다면더정확하게겨드랑이림프절전이상태를예측할수있다고밝혔다. T1 유방암에서수술후겨드랑이림프절전이가발견되는경우는 3 37% 로보고된다.(7-11) Markopoulos 등 (13) 은겨드랑이림프절박리술을시행한 195명을대상으로 T1a 종양에서는림프절전이가관찰되지않으나, T1b는 25.7%, T1c는 33.8% 의높은전이율을보여겨드랑이림프절박리술을생략할수있는군은 T1a 뿐이며종양의크기가 0.5 cm 이상의유방암환자는모두겨드랑이수술이필요하다고주장하였다. Chen 등 (15) 은감시림프절생검을실시한 T1 환자 257명을대상으로 T1a는 13.5%, T1b는 20.4%, T1c 는 35.9% 의겨드랑이림프절전이를보이고다변량분석결과가장큰영향을주는독립인자는종양의크기였다. Rivadeneira 등 (16) 은 1 cm 이하크기유방암환자 919명을 대상으로겨드랑이림프절박리술을실시하여약 18.0% 의액와림프절전이율을보고하고 1 cm 이하유방암에서도겨드랑이수술을반드시시행할것을주장하였다. 이들은또한종양의크기를 1 mm 단위의연속변수를사용하여다변량분석을실시하였는데종양의크기가증가할수록더높은겨드랑이전이율을보였다. 본연구또한 1 cm 이하유방암환자를대상으로겨드랑이림프절전이율은약 19.4% 로비슷하였으나종양의크기에따른림프절전이율은통계학적의미가없었다. 종양의크기외에 T1 유방암환자의림프절전이에영향을주는임상병리학적특징들은여러문헌에서다양하게보고되지만, 높은조직학적등급 (8,9,14,16) 림프혈관침범 (9-12,14-16,17) 에스트로겐수용체음성, 삼중복음성 (Triple negative),(17) 젊은연령,(8,14,16) 촉지성,(9,12,14,17) 종양의다발성등 (18) 으로요약된다. 본연구에서도단변량분석결과겨드랑이림프절전이에영향을주는인자는다발성종양, 림프혈관침범, 조직학적등급으로나타났으나다변량분석에서는조직학적등급은통계학적의미가없었다. Guarnieri 등 (19) 은조직학적등급과림프혈관침범의유무에따라 T1a, T1b 유방암에서수술을두단계로실시하도록권유하였다. 즉, 종양절제를우선시행하여병리학적결과혈관침범이없고, 조직학적등급이낮다면겨드랑이수술을피할수있고, 그렇지않다면두번째수술로서겨드랑이수술을권유하였다. 초음파와자기공명영상 (MRI) 의발달로종양의주덩어리 (main tumor) 외에같은구역또는다른구역에서우연히발견되는위성병변 (satellite leision) 때문에유방보존술이불가능한경우를임상에서종종접하게된다. 다발성유방암에서림프절전이율은종양의크기에따라다양하지만 T1 크기로제한하였을때약 48 61% 로단발성종양이약 35 38% 에비해높은전이율을보였다.(20,21) Coombs와 Boyages(22) 는다발성종양에서림프절전이율이높은것은종양의부피 (burden) 와연관있음을설명하고,
14 J Korean Surg Soc. Vol. 80, No. 1 다발성종양에서 T 병기는최대직경이아니라각각종양의직경을합해야한다고주장하였다. 다발성종양에있어서겨드랑이수술또한여전히논란이되고있다. Giard 등 (23) 은다발성유방암의감시림프절생검에대한다기관전향적연구로서, T1 종양일때감시림프절위음성율이 13.6% 로감시림프절생검보다는고식적겨드랑이림프절박리술을권유하였다. 그러나, 다른후향적보고에서는다발성종양의감시림프절생검의위음성율은 7 8% 로단발성종양과유사하므로감시림프절생검술을실시하여도좋다고결론지었다.(24,25) 본연구에서는다발성종양이단발성종양에비해상대위험도 5.92로서림프절전이의위험이매우높았다. 이는종양의크기가작더라도초음파또는 MRI 상에다발성병변이의심되거나, 수술전에조직학적으로진단된다발성유방암이라면감시림프절생검보다는고식적겨드랑이림프절박리술이더유익하리라사료된다. 결론적으로 1 cm 이하크기의유방암에서술전검사상림프관및혈관침범또는다발성의소견이있다면감시림프절생검보다고식적겨드랑이림프절박리술을고려하는것이좋겠다. 향후다기관전향적연구를통해 1 cm 이하유방암의겨드랑이림프절전이에대한위험인자를선별할필요가있을것이다. REFERENCES 1) Ko SS, The Korean Breast Cancer Society. Chronological changing patterns of clinical characteristics of Korean breast patients during 10 years (1996-2006) using nationwide breast cancer registration on -line program: Biannual update. J Surg Oncol 2008;98:318-23. 2) Veronesi U, Paganelli G, Viale G, Luini A, Surrida S, Galimverti V, et al. Sentinel - lymph-node biopsy as a staging procedure in breast cancer: update of a randomized controlled study. Lancet Oncol 2006;7:983-90. 3) Martelli G, Boracchi P, Michaela de P, Pilotti S, Oriana S, Zucali R, et al. A randomized trial comparing axillary dissection to no axillary dissection in older patients with T1N0 breast cancer: Results after 5 years of follow-up. Ann Surg 2005;242:1-6. 4)Veronesi U, De Cicco C, Galimberti VE, Fernandez JR, Rotmensz N, Viale G, et al. A comparative study on the value of FDG-PET and sentinel node biopsy to identify occult axillary metastases. Ann Oncol 2007;18:473-8. 5) Lovrics PJ, Chen V, Coates G, Cornacchi SD, Goldsmith CH, Law C. A prospective evaluation of positron emission tomography scanning, sentinel lymph node biopsy, and standard axillary dissection for axillary staging in patients with early stage breast cancer. Ann Surg Oncol 2004;11:846-53. 6) Ahn JH, Son EJ, Kim JA, Youk JH, Kim EK, Kwak JY, et al. The role of ultrasonography and FDG-PET in axillary lymph node staging of breast cancer. Acta Radiol 2010;51: 859-65. 7) Kang HS, Noh DY, Youn YK, Oh SK, Choe KJ. The predictors of axillary node metastasis in 2 cm or less breast cancer univariate and multivariate analysis. J Korean Breast Cancer Soc 1999;2:7-13. 8) Mustafa IA, Bland KI. Indications for axillary dissection in T1 breast cancer. Ann Surg Oncol 1998;5:4-8. 9) Silverstein MJ, Skinner KA, Lomis TJ. Predicting axillary nodal positivity in 2282 patients with breast carcinoma. World J Surg 2001;25:767-72. 10) Yip CH, Taib NA, Tan GH, Ng KL, Yoong BK, Choo WY. Predictors of axillary lymph node metastases in breast cancer: Is there a role for minimal axillary surgery? World J Surg 2009;33:54-7. 11) Kim TH, Bae JW, Kim F, Lee JB, Son GS, Koo BH. Factors related with axillary lymph nodes metastases in T1 invasive ductal carcinomas of the breast. J Breast Cancer 2006;9:31-5. 12) Fein DA, Fowble BL, Hanlon AL, Hooks MA, Hoffman JP, Sigurdson ER, et al. Identification of women with T1-T2 breast cancer at low risk of positive axillary nodes. J Surg Oncol 1997;65:34-9. 13) Markopoulos C, Kouskos E, Gogas H, Mandas D, Kakisis J, Gogas J. Factors affecting axillary lymph node metastases in patients with T1 breast carcinoma. Am Surg 2000;66:1011-3. 14) Brenin DR, Manasseh DM, El-Tamer M, Troxel A, Schnabel F, Ditkoff BA, et al. Factors correlating with lymph node metastases in patients with T1 breast cancer. Ann Surg Oncol 2001;8:432-7. 15) Chen M, Palleschi S, Khoynezhad A, Gecelter G, Marini CP, Simms HH. Role of primary breast cancer characteristics in predicting positive sentinel lymph node biopsy results. A multivariate analysis. Arch Surg 2002;137:606-10. 16) Rivadeneira DE, Simmons RM, Christos PJ, Hanna K, Daly JM, Osborne MP. Predictive factors associated with axillary lymph node metastases in T1a and T1b breast carcinomas: Analysis in more than 900 patients. J Am Coll Surg 2000;191:1-8. 17) Lee JH, Kim SH, Suh YJ, Shim BY, Kim HK. Predictors of axillary lymph node metastases (ALNM) in a Korean population with T1-2 breast carcinoma: Triple negative breast cancer has a high incidence of ALNM irrespective of the tumor size. Cancer Res Treat 2010;42:30-6. 18) Bevilacqua J, Cody H 3rd, MacDonald KA, Tan LK, Borgen PI, Van Zee KJ. A model for predicting axillary node metastases based on 2000 sentinel node procedures and tumor position. EJSO 2002;28:490-500.
Sun Hee Kang and Jihyung Cho:Predictor for Axillary Node Metastsis in Breast Cancer of 1 cm or Less 15 19) Guarnieri A, Nerri A, Correale PP, Lottini M, Testa M, Mariani F, et al. Prediction of lymph node status by analysis of prognostic factors and possible indications for elective axillary dissection in T1 breast cancers. Eur J Surg 2001;167: 255-9. 20) Cabioglu N, Ozmen V, Kaya H, Tuzlali S, Igci A, Muslumanoglu M, et al. Increased lymph node positivity in multifocal and multicentric breast cancer. J Am Coll Surg 2009;208: 67-74. 21) Andea AA, Bouwman D, Wallis T, Visscher DW. Correlation of tumor volume and surface area with lymph node status in patients with multifocal/multicentric breast carcinoma. Cancer 2004;100:20-7. 22) Coombs NJ, Boyages J. Multifocal and multicentric breast cancer: Does each focus matter? J Clin Oncol 2005;23: 7497-502. 23) Giard S, Chauvet MP, Penel N, Mignotte H, Martel P, Tunon De Lara C, et al. Feasibility of sentinel lymph node biopsy in multiple unilateral synchronous breast cancer: results of a French prospective multi-institutional study (IGASSU0502). Ann Oncol 2010;21:1630-5. 24) Kim HJ, Lee JS, Park EH, Choi SL, Lim WS, Chang MA, et al. Sentinel node biopsy in patients with multiple breast cancer. Breast Cancer Res Treat 2008;109:503-6. 25) Ferrari A, Dionigi P, Rovera F, Boni L, Limonta G, Garancini S, et al. Multifocality and multicentricity are not contraindications for sentinel lymph node biopsy in breast cancer surgery. World J Surg Oncol 2006;4:79-86.