Supplementary data: 1) Search terms used: a. Fascin-1 Fascin/Fascin(s)/Fascin(1) MICROFILAMENT PROTEIN(S) Actin-bundling protein(s) Actin-binding protein(s) FSCN(1) Fascin homology(1) b. Breast Cancer Breast Cancer BREAST NEOPLASM(S) CARCINOMA,INTRADUCTAL,NONINFILTRATING CARCINOMA,DUCTAL,BREAST Breast tumor(s) Breast carcinoma(s)/carcinoma(s) of the breast Breast adenoma(s) c. Colorectal Cancer COLORECTAL NEOPLASM(s) Colorectal cancer(s) Colorectal carcinoma(s) Colorectal tumour(s) COLONIC NEOPLASM(s) Colon neoplasm(s) Colon tumour(s) Colon cancer(s) RECTAL NEOPLASM(s) Rectal cancer(s) Rectal tumour(s) Rectum cancer(s) Rectum tumour(s) d. Gastric Cancer STOMACH NEOPLASM(S) Gastric neoplasm(s) Stomach cancer(s) Gastric cancer(s) stomach carcinoma(s) Gastric carcinoma(s) Stomach tumo(u)r(s) Gastric tumo(u)r(s) 1 e. Lung Cancer LUNG NEOPLASM(S) Lung tumour(s) Lung Cancer(s) Pulmonary neoplasm(s) Pulmonary cancer(s) NON SMALL CELL LUNG CARCINOMA(S) NSCLC Non small cell lung cancer(s) SMALL CELL LUNG CARCINOMA(S) small cell lung cancer(s) SCLC Oat cell carcinoma(s) of Lung BRONCHIAL NEOPLASM(S) Bronchi/Bronchial tumour(s) Bronchi/Bronchial Cancer(s) Bronchial carcinoid(s) BRONCHOGENIC CARCINOMA(S) Bronchial carcinoma BRONCHIOLO-ALVEOLAR ADENOCARCINOMA(S) Bronchiolo-alveolar carcinoma(s) Bronchiolar carcinoma(s) Alveolar adenocarcinoma(s) Alveolar carcinoma(s) f. Oesophageal Cancer Esophageal Neoplasm(s) Oseophageal Neoplasm(s) Esophagus Neoplasm(s) Oseophagus Neoplasm(s) Esophageal cancer(s) Oesophageal cancer(s) Esophagus cancer(s) Oesophagus cancer(s) Esophageal carcinoma(s) Oseophageal carcinoma(s) Esophagus carcinoma(s) Oseophagus carcinoma(s) 2 2) Scoring methodologies used to assess immunohistochemical staining of fascin-1: a. Different scoring systems have been used to assess the immunohistochemical data on fascin-1. Some studies measured the intensity of fascin-1 staining and/or the percentage of immunoreactive cells, while other studies multiplied the intensity of fascin-1 staining by the percentage of immunoreactive cells. For papers [1,2] which presented separate data for immunoreactivity or intensity scores, we chose to extract the immunoreactivity scores. This is because a recent study [3] showed immunoreactivity scores alone to be a sufficient measure for estimating the association of fascin-1 with survival. b. We pooled data using alternative categorisation methods if we did not agree with the scoring methods used by the individual studies. For example, Takikita et al [4] and Pang et al [5] categorised a score of ≤3 as negative. As an overall score of 3 could possibly include fascin1-positive cells, we pooled data by using the alternative categorisation of dichotomising the results as low (overall score of ≤3) and high (score of ≥4). 3) Method used to derive hazard ratios from the Kaplan-Meier curve analysis: We extracted data from the Kaplan-Meier curve using the method by Tierney et al [6]. For their method, the Kaplan-Meier curve was divided into a number of time intervals and a number of iterative calculations were carried out to estimate the number of events during each interval and the number of patients who were: i) event-free at the start of the interval; ii) censored during the interval; and iii) at risk during the interval. HRs and their variance were then calculated for each time interval. The HRs and variance for the whole curve was derived by combining the estimates across time intervals. We extracted data from the Kaplan-Meier Curves analysis of the study by Tsai et al [7] by using the method of Tierney et al [6]. The fulle sample set included 100 patients. The Kaplan-Meier curve included 75 patients and we did not have data on the proportion of fascin-1 positive or negative tumour specimens in these 75 patients. We estimated the log relative risk from the study by assuming that the proportion of fascin-1 positive versus negative patients was the same in the 75 patients included in the survival analysis as in the full sample of 100 patients. 3 4) References of papers excluded from the meta-analysis and reasons for exclusion 1. Compared fascin-1 in normal tissues versus tumours in the same patients [8,9] 2. Carried out studies in adenomas [10], pleomorphic carcinomas [11] or neuroendocrine tumours [12]. 3. Did not report on relevant exposure [13] 4. Did not present outcomes on cancer-specific mortality, disease-free survival or metastasis [3,14-18]. 5. Did not present results for survival or metastasis [19-23] 6. Duplicate abstract or study [24-27] 7. Inability to categorise fascin-1 expression [28] 8. Unable to obtain one paper either electronically or after contacting the authors [29] 5) Method used to derive 2-yr mortality risk ratio for the study by Ozerhan et al [30]: For the study by Ozerhan et al [30], two-year survival rates for each fascin-1 category was published in the results section. This was multiplied by the total number of patients in each fascin-1 category to estimate the number of events at two years. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Pelosi G, Pastorino U, Pasini F, Maissoneuve P, Fraggetta F, Iannucci A, Sonzogni A, De Manzoni G, Terzi A, Durante E et al: Independent prognostic value of fascin immunoreactivity in stage I nonsmall cell lung cancer. British Journal of Cancer 2003, 88(4):537-547. Hashimoto Y, Ito T, Inoue H, Okumura T, Tanaka E, Tsunoda S, Higashiyama M, Watanabe G, Imamura M, Shimada Y: Prognostic significance of fascin overexpression in human esophageal squamous cell carcinoma. Clinical Cancer Research 2005, 11(7):2597-2605. Chan C, Jankova L, Fung CLS, Clarke C, Robertson G, Chapuis PH, Bokey L, Lin BPC, Dent OF, Clarke S: Fascin Expression Predicts Survival After Potentially Curative Resection of Nodepositive Colon Cancer. American Journal of Surgical Pathology 2010, 34(5):656-666. Takikita M, Hu N, Shou JZ, Giffen C, Wang QH, Wang C, Hewitt SM, Taylor PR: Fascin and CK4 as biomarkers for esophageal squamous cell carcinoma. Anticancer Research 2011, 31 (3):945-952. Pang Q, Zhang X, Zhang J, Li Y, Wu W, Wang H, Cui A, Zhang Y: Correlative studies on expression of CXCR4, CXCR12, Fascin and E-cadherin in colorectal carcinoma with lymphatic metastasis. [Chinese]. Chinese Journal of Clinical Oncology 2009, 36 (16):926-929+933. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR: Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 2007, 8:16. Tsai WC, Jin JS, Chang WK, Chan DC, Yeh MK, Cherng SC, Lin LF, Sheu LF, Chao YC: Association of cortactin and fascin-1 expression in gastric adenocarcinoma: Correlation with clinicopathological parameters. Journal of Histochemistry and Cytochemistry 2007, 55 (9):955-962. Gao X, Wu DH: Fascin expression in human epithelial tumors and its clinical significance. [Chinese]. Nan Fang Yi Ke Da Xue Xue Bao = Journal of Southern Medical University 2008, 28 (6):953-955. Xue LY, Zou SM, Zheng S, Xie YQ, Wen P, Liu XY, Lin DM, Lu N: [Expression of fascin and CK14 in different histological types of cancer and its differential diagnostic significance]. Zhonghua Zhong Liu Za Zhi 2010, 32(11):838-844. Qualtrough D, Singh K, Banu N, Paraskeva C, Pignatelli M: The actin-bundling protein fascin is overexpressed in colorectal adenomas and promotes motility in adenoma cells in vitro. Br J Cancer 2009, 101(7):1124-1129. Pelosi G, Fraggetta F, Nappi O, Pastorino U, Maisonneuve P, Pasini F, Iannucci A, Solli P, Musavinasab HS, De Manzoni G et al: Pleomorphic carcinomas of the lung show a selective 4 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. distribution of gene products involved in cell differentiation, cell cycle control, tumor growth, and tumor cell motility - A clinicopothologic and immunohistochemical study of 31 cases. American Journal of Surgical Pathology 2003, 27(9):1203-1215. Pelosi G, Pasini F, Fraggetta F, Pastorino U, Iannucci A, Maisonneuve P, Arrigoni G, De Manzoni G, Bresaola E, Viale G: Independent value of fascin immunoreactivity for predicting lymph node metastases in typical and atypical pulmonary carcinoids. Lung Cancer 2003, 42(2):203-213. Hashimoto Y, Skacel M, Adams JC: Association of loss of epithelial syndecan-1 with stage and local metastasis of colorectal adenocarcinomas: an immunohistochemical study of clinically annotated tumors. Bmc Cancer 2008, 8:185. Chikman B, Lavy R, Tolstov G, Habler L, Kapiev A, Halevy A, Sandbank J: Significance of Fascin expression in high grade breast cancer. Annals of Surgical Oncology 2010, 17:S57-S57. Kanber Y, Xu YH, Hartmann DP, Azumi N: Fascin is overexpressed in poorly differentiated colonic carcinomas: Possible prognostic indicator of colonic cancer. American Journal of Clinical Pathology 2004, 122(4):642-642. Kefeli M, Sengul AT, Yildiz L, Baris S, Basoglu A, Kandemir B: EMMPRIN and fascin expression in non-small cell lung carcinoma. Central European Journal of Medicine 2010, 5(6):659-665. Sarrio D, Rodriguez-Pinilla SM, Hardisson D, Cano A, Moreno-Bueno G, Palacios J: Epithelialmesenchymal transition in breast cancer relates to the basal-like phenotype. Cancer Research 2008, 68(4):989-997. Erdogan G, Pestereli HE, Colak T, Karaveli FS, Akaydin M: Fascin expression in invasive ductal carcinoma of breast. Turk Patoloji Dergisi/Turkish Journal of Pathology 2010, 26 (2):130-135. Grothey A, Hashizume R, Sahin AA, McCrea PD: Fascin, an actin-bundling protein associated with cell motility, is upregulated in hormone receptor negative breast cancer. British Journal of Cancer 2000, 83(7):870-873. del Rosario A, Goncharuk VN, Tran TA, Sheehan CE, Ross JS: Fascin immunoreactivity correlates with tumor grade in non-small cell lung cancer (NSCLC). Laboratory Investigation 2001, 81(1):218A-218A. Zheng HC, Zheng YS, Xia P, Xu XY, Xing YN, Takahashi H, Guan YF, Takano Y: The pathobiological behaviors and prognosis associated with Japanese gastric adenocarcinomas of pure WHO histological subtypes. Histology and Histopathology 2010, 25(4):445-452. Goncharuk V, Arun B, Middleton L, Sahin A: Assessment of actin-binding protein fascia expression in stage I and II invasive breast cancer. A significant correlation with aggressive clinical behavior. Laboratory Investigation 2003, 83(1):30A-30A. Lee JH, Bae BH, Lee JM, Kim JM, Hwang JY, Goo JM, Jang JS, Roh MW, Choi SR: The prognostic significance of fascin expression in colorectal carcinoma. Journal of Gastroenterology and Hepatology 2011, Conference: Asian Pacific Digestive Week 2011 Singapore Singapore. Conference Start: 20111001 Conference End: 20111004. Conference Publication: (var.pagings). 26:225. Rodriguez-Pinilla SM: Fascin and caveolin expression are associated with the basal-like subtype of invasive breast carcinoma. Virchows Archiv 2005, 447(2):244-245. Xue L, Hu N, Song Y, Zou S, Shou J, Qian L, Ren L, Lin D, Tong T, He Z et al: Tissue microarray analysis reveals a tight correlation between protein expression pattern and progression of esophageal squamous cell carcinoma. BMC Cancer 2006, 6(296). Xie JJ: Involvement of CYR61 and CTGF in the Fascin-Mediated Proliferation and Invasiveness of Esophageal Squamous Cell Carcinomas Cells (vol 176, pg 939, 2010). American Journal of Pathology 2010, 176(5):939-951. Grothey A, Hashizume R, Sahin AA, McCrea PD: Fascin, an actin-bundling protein associated with cell motility, is upregulated in hormone receptor negative breast cancer. European Journal of Cancer 1999, 35:S200-S200. Tsai WC, Chao YC, Sheu LF, Chang JL, Nieh S, Jin JS: Overexpression of fascin-1 in advanced colorectal adenocarcinoma: Tissue microarray analysis of immunostaining scores with clinicopathological parameters. Disease Markers 2007, 23(3):153-160. Nese N, Kandiloglu AR, Simsek G, Lekili M, Ozdamar A, Catalkaya A, Coskun T: Comparison of the desmoplastic reaction and invading ability in invasive ductal carcinoma of the breast and prostatic adenocarcinoma based on the expression of heat shock protein 47 and fascin. Analytical & Quantitative Cytology & Histology 2010, 32(2):90-101. Ozerhan IH, Ersoz N, Onguru O, Ozturk M, Kurt B, Cetiner S: Fascin expression in colorectal carcinomas. Clinics 2010, 65(2):157-164. 5