Short communication
169
Fascin expression is increased in metastatic lesions but does not correlate with progression nor outcome in melanoma Yafeng Maa,e, William J. Fallera, Owen J. Sansoma, Ewan R. Brownb, Tamasin N. Doigc, David W. Meltond and Laura M. Macheskya Levels of the actin bundling protein fascin correlate with invasion and metastasis and reveal prognostic value in many epithelial carcinomas. However, we know very little about the potential role of fascin in melanoma. The purpose of this study is to compare fascin expression in primary melanomas and melanoma metastasis. Fascin expression was examined through the immunohistochemistry of paraffin embedded tissue microarrays including 560 cores of primary tumour and metastasis. Fascin expression was significantly elevated in 48 metastases compared with 254 primary tumours (P = 0.034). In 187 patients with primary melanomas, fascin was not correlated with survival (P = 0.067), whereas low fascin was significantly correlated with the presence of ulceration (P = 0.005). Our results indicate that fascin status does not correlate with progression in melanoma. Upregulated fascin expression was detected in melanoma metastases, but was not correlated to patient outcome. Melanoma Res
Introduction The actin bundling protein fascin-1 (referred to as fascin) crosslinks actin filaments into tight bundles and thus plays an important role in cell migration and invasion [1, 2]. Previous studies indicate fascin is expressed in many human melanoma cells in culture and stabilizes the invasive filopodia (invadopodia) [3], which are believed to promote tumour metastasis [4]. However, in mice, fascin appears to be only transiently expressed in melanoblasts during embryogenesis and to confer enhanced migration and proliferation [5]. Fascin is widely implicated as a prognostic marker of some epithelial cancers [6], including pancreatic cancers [7] and colon cancers [8,9]. However, data supporting a role for fascin in melanoma patients are limited and elusive due to small sample size [10,11] and thus far, studies do not indicate a clear correlation between tumour progression and fascin expression. To determine the involvement of fascin in melanoma, we examined a large human melanoma tissue array.
Methods Melanoma tissue microarrays (TMAs) were constructed as previously described [12]. TMAs were dewaxed, rehydrated and antigen-retrieved in citrate buffer (pH 6.0) as mentioned [5]. TMAs were blocked in peroxidase block (Envision kits, DAKO, Agilent Technologies Inc, Santa Clara, California, USA) and stained with rabbit antifascin (1 : 200, SigmaAldrich, St Louis, Missouri, USA, 2 h, room temperature) in 0960-8931 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
25:169–172 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Melanoma Research 2015, 25:169–172 Keywords: fascin, melanoma, metastasis a
Beatson Institute for Cancer Research, Glasgow, bEdinburgh Cancer Centre, Department of Pathology, NHS Lothian, Western General Hospital, dEdinburgh Cancer Research Centre, MRC Institute of Genetics and Molecular Medicine, University of Edinburgh, MRC Human Genetics Unit, Western General Hospital, Edinburgh, UK and eMedical Oncology Group, Ingham Institute for Applied Medical Research, School of Medicine, University of New South Wales, New South Wales, Australia c
Correspondence to Laura M. Machesky, PhD, Beatson Institute for Cancer Research, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD Scotland, UK Tel: + 44 1413303653; fax: + 44 141 9426521; e-mail: [email protected] Received 16 June 2014 Accepted 10 November 2014
antibody diluent (Dako). After 30 min incubation of peroxidase-labelled polymer (Envision kits), TMAs were washed and incubated in substrate-chromogen (Envision kits) and counter-stained with haematoxylin. Patients and tissue cores were excluded if there was insufficient tissue to score as judged blindly by two independent scientists. In samples where sufficient material was present in duplicate, and in the case of thicker tumours, triplicate or quadruplicate cores were taken. The weighted histoscore is calculated using the following formula: (0 × percentage of negative staining) + (1 × percentage of weak staining) + (2 × percentage of moderate staining) + (3 × percentage of strong staining). This gives a value between 0 and 300. This value represents the staining intensity of each core of TMA. Statistical analyses were carried out using Microsoft Excel 2000 software (Microsoft, Washington, USA) and PASW statistics, version 19.0 (SPSS Inc., Chicago, Illinois, USA). Kaplan–Meier survival analysis was used to analyse the overall survival from the time of surgery. Patients alive at the time of follow-up point were censored. To compare the length of survival between curves, a log-rank test was performed. χ2 and Mann–Whitney (nonparametric) methods were used to test for association of clinicopathological variables with fascin mean value.
Results and discussion In an effort to understand fascin expression pattern in melanoma, we performed fascin immunohistochemistry DOI: 10.1097/CMR.0000000000000135
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
170 Melanoma Research
2015, Vol 25 No 2
Fig. 1
(a)
(b)
Metastasis
Primary melanoma
∗ Fascin1 histoscore
300
No stain (0)
200
100
sis ta as M et
Pr im ar y
0
(c) Fascin1 histoscore (mean)
Primary (%)
Metastasis (%)
0 >0, ≤100 >100, ≤200 >200, ≤300 Total cases
10 (3.9) 114 (44.9) 104 (40.9) 26 (10.2) 254
5 (10.4) 13 (27.1) 13 (27.1) 17 (35.4) 48
Weak (1)
Stage (n=302)
Moderate (2)
200
200
100
100
0
0 nc
Ulceration
0.8 Cum survival
1.
Pr
es
1.0
se
en
ce
Kaplan−Meier 50% survival
300
Ab
(d)
300
≤= 1 01 mm −2 m 2. 01 m −4 m m >4 m m
Strong (3)
(f)
∗∗
e
Fascin1 histoscore
(e)
Breslow thickness
0.6 0.4 P=0.067 0.2 0.0 0
2
4 6 8 10 Follow-up (years)
12
Fascin expression does not correlate with tumour stage but is increased in metastases. (a) Fascin IHC analysis of a human tumour array of melanomas and metastasis. Representative images are shown for various staining intensities of primary melanomas and metastasis. No staining, 0; weak staining, 1; moderate staining, 2 and strong staining, 3. Bar in original picture, 100 μm. (b) Box plot of fascin histoscore versus tumour stage (Mann–Whitney U-test,*P = 0.034). Primary tumours exhibit a lower level of fascin expression (n = 254, mean ± SD, 116.6 ± 4.6) versus metastasis (n = 48, mean ± SD, 152.1 ± 15.1). (c) Detailed relationship between fascin histoscore and tumour stage. (d) Kaplan–Meier curve showing the survival of patients with high or low fascin expression. No significant difference was seen at P = 0.05 (Log-rank test P = 0.067). The green line indicates patients with high fascin (histoscore > 114, patient n = 97). The blue line indicates patients with low fascin (histoscore ≤ 113, patient n = 90).‘ + ’ means censored. (e) Box plot of fascin histoscore versus ulceration, **P = 0.005. (f) Box plot of fascin histoscore versus Breslow thickness. IHC, immunohistochemistry.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Increased fascin in melanoma metastatic lesions Ma et al. 171
The relationship between fascin histoscore and ulceration status
The relationship between clinicopathological variables and fascin histoscores
Table 1
Table 2
Ulceration (n = 163) [n (%)] Fascin histoscore (mean) 0 Weak (>0, ≤ 100) Moderate (>100, ≤ 200) Strong (>200, ≤ 300) Total cases
Presence 2 22 14 3
(4.9) (53.7) (34.1) (7.3) 41
Fascin histoscore
Absence 6 46 52 18
(4.9) (37.7) (42.6) (14.8) 122
in a TMA containing a total of 562 cores of primary melanomas and melanoma metastasis (Met) [12]. A total of 302 cases (53.7%) were considered as valid samples (primary melanoma, 254; melanoma Met, 48) (Fig.1a). The fascin staining density was classified as 0 (no staining), 1 (weak staining), 2 (moderate staining), 3 (strong staining) and a histoscore was calculated based on the area of positive staining by two independent investigators [as detailed in the Methods section and (Fig.1a)]. We observed fascin staining primarily in the cytoplasm of melanoma cells. Fascin was absent in 10 of 254 primary tumours and five of 48 melanoma metastases. In all, 35.4% metastases expressed fascin at a high level and generally melanoma metastasis expressed a significantly higher level of fascin compared with primary melanomas (152.1 ± 15.1 vs. 116.6 ± 4.6, Mann–Whitney U-test, P = 0.034) (Fig 1b and c). Fascin levels were not significantly correlated with survival (P = 0.067) in 187 patients with primary tumours (Fig. 1d, green line = higher fascin score, blue line = lower fascin score). Ulceration and Breslow depth are currently among the two most important prognostic factors in melanoma, with the presence of ulceration and greater Breslow depth predicting poor outcome [13,14]. Surprisingly, low fascin was correlated with the presence of ulceration (P = 0.005) (Fig. 1e and Table 1). Although fascin expression was not correlated with Breslow depth (Pearson’s correlation coefficient r = − 0.142, P = 0.05, Mann–Whitney U-test, P = 0.164). For those primary melanomas, lower fascin expression was associated with higher Breslow depth when Breslow depth was less than 4 mm, whereas fascin expression level was elevated when tumour thickness was more than 4 mm and tumours were tending to metastasis (Fig. 1f). Fascin expression did not correlate to other clinicopathological characteristics, such as age, sex and sun-exposure (Table 2 and data not shown). Neither did fascin level differ within the histological stage of lentigo (139.5 ± 22.5, n = 14), superficial spreading melanoma (116.2 ± 6.4, n = 142), nodular (100.4 ± 13.3, n = 26) and acral (160.3 ± 18.1, n = 5) (Table 2 and data not shown). These data may indicate that fascin expression is increased during metastasis, but that its expression does not correlate with melanoma progression. However, our observation on the relationships between fascin expression and presence of ulceration is different from previous findings in gastrointestinal stromal tumours, where Yamamoto et al. [15] found that high fascin was
Variables
Negative (n = 8)
Low (n = 88)
High (n = 91)
1 7
36 52
36 55
0.635
0 8 0
12 74 1
15 73 3
0.581
1 6 1 0
4 70 14 0
9 66 11 5
0.233
2 5 1 0
24 52 11 1
15 65 10 1
0.024
5 2 1
68 16 4
75 8 8
0.181
Sex Male (n = 73) Female (n = 114) Site Sun-exposed (n = 27) Sun-protected (n = 155) Acral (n = 4) Histology Lentigo (n = 14) SSM (n = 142) Nodular (n = 26) Acral (n = 5) Ulceration Presence (n = 41) Absence (n = 122) Incipient (n = 22) Unknown (n = 2) Status Alive (n = 148) Dead (n = 26) Nonmelanoma death (n = 13)
P
SSM, superficial spreading melanoma.
significantly correlated with mucosal ulceration. The correlation of fascin expression with ulceration could indicate a role for fascin expression in healing and regeneration following damage, as has been postulated in inflammatory bowel diseases [16] and as such could be advantageous to the host rather than the tumour. Yildiz et al. [11] published a small study of 12 malignant melanoma, 20 benign nevi and 12 dysplastic nevi and found that malignant melanomas (25%) express fascin less frequently than benign nevi (95%, P < 0.001) and dysplastic nevi (67%, P = 0.036). Goncharuk et al. [10] examined fascin expression in normal skin and different types of benign and malignant skin tumours and found only 19% of melanoma (3/16) express fascin. Our data on fascin expression levels in a reasonably large number of human melanocytic lesions suggest that fascin is upregulated in melanoma metastasis but levels do not correlate with outcome and are variable. In primary melanoma, low fascin expression levels correlated with ulceration. We previously showed that fascin was transiently upregulated in melanoblasts of mice during embryogenesis and conferred enhanced migration [5]. Thus, expression of fascin in metastatic lesions might correlate with a reactivation of this embryonic migration pattern, but clearly fascin is not limiting for melanoma spread. Our study highlights the complexity of fascin’s involvement in melanoma and suggests that fascin expression is not likely to be useful as a prognostic indicator for melanoma.
Acknowledgements The authors thank Colin Nixon of Beatson Histology Services for his excellent help with TMA processing.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
172
Melanoma Research
2015, Vol 25 No 2
Conflicts of interest
Y.M., L.M.M. and O.J.S. are funded by a Core CRUK grant. For the remaining authors there are no conflicts of interest.
9
References 1 2 3
4 5
6
7
8
Adams JC. Roles of fascin in cell adhesion and motility. Curr Opin Cell Biol 2004; 16:590–596. Jayo A, Parsons M. Fascin: a key regulator of cytoskeletal dynamics. Int J Biochem Cell Biol 2010; 42:1614–1617. Li A, Dawson JC, Forero-Vargas M, Spence HJ, Yu X, König I, et al. The actinbundling protein fascin stabilizes actin in invadopodia and potentiates protrusive invasion. Curr Biol 2010; 20:339–345. Machesky LM, Li A. Fascin: Invasive filopodia promoting metastasis. Commun Integr Biol 2010; 3:263–270. Ma Y, Li A, Faller WJ, Libertini S, Fiorito F, Gillespie DA, et al. Fascin 1 is transiently expressed in mouse melanoblasts during development and promotes migration and proliferation. Development 2013; 140:2203–2211. Tan VY, Lewis SJ, Adams JC, Martin RM. Association of fascin-1 with mortality, disease progression and metastasis in carcinomas: a systematic review and meta-analysis. BMC Med 2013; 11:52. Li A, Morton JP, Ma Y, Karim SA, Zhou Y, Faller WJ, et al. Fascin is regulated by slug, promotes progression of pancreatic cancer in mice, and is associated with patient outcomes. Gastroenterology 2014; 146:1386–1396. Hashimoto Y, Skacel M, Lavery IC, Mukherjee AL, Casey G, Adams JC. Prognostic significance of fascin expression in advanced colorectal cancer:
10 11 12
13
14 15
16
an immunohistochemical study of colorectal adenomas and adenocarcinomas. BMC Cancer 2006; 6:241. 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. Dis Markers 2007; 23:153–160. Goncharuk VN, Ross JS, Carlson JA. Actin-binding protein fascin expression in skin neoplasia. J Cutan Pathol 2002; 29:430–438. Yildiz L, Kefeli M, Aydin O, Kandemir B. Fascin expression in melanocytic lesions of the skin. Eur J Dermatol 2009; 19:445–450. Brown ER, Doig T, Anderson N, Brenn T, Doherty V, Xu Y, et al. Association of galectin-3 expression with melanoma progression and prognosis. Eur J Cancer 2012; 48:865–874. Balch CM, Soong SJ, Gershenwald JE, Thompson JF, Reintgen DS, Cascinelli N, et al. Prognostic factors analysis of 17 600 melanoma patients: validation of the American Joint Committee on Cancer Melanoma Staging System. J Clin Oncol 2001; 19:3622–3634. Dickson PV, Gershenwald JE. Staging and prognosis of cutaneous melanoma. Surg Oncol Clin N Am 2011; 20:1–17. Yamamoto H, Kohashi K, Fujita A, Oda Y. Fascin-1 overexpression and miR-133b downregulation in the progression of gastrointestinal stromal tumor. Mod Pathol 2013; 26:563–571. Qualtrough D, Smallwood K, Littlejohns D, Pignatelli M. The actin-bundling protein fascin is overexpressed in inflammatory bowel disease and may be important in tissue repair. BMC Gastroenterol 2011; 11:14.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
169
Fascin expression is increased in metastatic lesions but does not correlate with progression nor outcome in melanoma Yafeng Maa,e, William J. Fallera, Owen J. Sansoma, Ewan R. Brownb, Tamasin N. Doigc, David W. Meltond and Laura M. Macheskya Levels of the actin bundling protein fascin correlate with invasion and metastasis and reveal prognostic value in many epithelial carcinomas. However, we know very little about the potential role of fascin in melanoma. The purpose of this study is to compare fascin expression in primary melanomas and melanoma metastasis. Fascin expression was examined through the immunohistochemistry of paraffin embedded tissue microarrays including 560 cores of primary tumour and metastasis. Fascin expression was significantly elevated in 48 metastases compared with 254 primary tumours (P = 0.034). In 187 patients with primary melanomas, fascin was not correlated with survival (P = 0.067), whereas low fascin was significantly correlated with the presence of ulceration (P = 0.005). Our results indicate that fascin status does not correlate with progression in melanoma. Upregulated fascin expression was detected in melanoma metastases, but was not correlated to patient outcome. Melanoma Res
Introduction The actin bundling protein fascin-1 (referred to as fascin) crosslinks actin filaments into tight bundles and thus plays an important role in cell migration and invasion [1, 2]. Previous studies indicate fascin is expressed in many human melanoma cells in culture and stabilizes the invasive filopodia (invadopodia) [3], which are believed to promote tumour metastasis [4]. However, in mice, fascin appears to be only transiently expressed in melanoblasts during embryogenesis and to confer enhanced migration and proliferation [5]. Fascin is widely implicated as a prognostic marker of some epithelial cancers [6], including pancreatic cancers [7] and colon cancers [8,9]. However, data supporting a role for fascin in melanoma patients are limited and elusive due to small sample size [10,11] and thus far, studies do not indicate a clear correlation between tumour progression and fascin expression. To determine the involvement of fascin in melanoma, we examined a large human melanoma tissue array.
Methods Melanoma tissue microarrays (TMAs) were constructed as previously described [12]. TMAs were dewaxed, rehydrated and antigen-retrieved in citrate buffer (pH 6.0) as mentioned [5]. TMAs were blocked in peroxidase block (Envision kits, DAKO, Agilent Technologies Inc, Santa Clara, California, USA) and stained with rabbit antifascin (1 : 200, SigmaAldrich, St Louis, Missouri, USA, 2 h, room temperature) in 0960-8931 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
25:169–172 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Melanoma Research 2015, 25:169–172 Keywords: fascin, melanoma, metastasis a
Beatson Institute for Cancer Research, Glasgow, bEdinburgh Cancer Centre, Department of Pathology, NHS Lothian, Western General Hospital, dEdinburgh Cancer Research Centre, MRC Institute of Genetics and Molecular Medicine, University of Edinburgh, MRC Human Genetics Unit, Western General Hospital, Edinburgh, UK and eMedical Oncology Group, Ingham Institute for Applied Medical Research, School of Medicine, University of New South Wales, New South Wales, Australia c
Correspondence to Laura M. Machesky, PhD, Beatson Institute for Cancer Research, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD Scotland, UK Tel: + 44 1413303653; fax: + 44 141 9426521; e-mail: [email protected] Received 16 June 2014 Accepted 10 November 2014
antibody diluent (Dako). After 30 min incubation of peroxidase-labelled polymer (Envision kits), TMAs were washed and incubated in substrate-chromogen (Envision kits) and counter-stained with haematoxylin. Patients and tissue cores were excluded if there was insufficient tissue to score as judged blindly by two independent scientists. In samples where sufficient material was present in duplicate, and in the case of thicker tumours, triplicate or quadruplicate cores were taken. The weighted histoscore is calculated using the following formula: (0 × percentage of negative staining) + (1 × percentage of weak staining) + (2 × percentage of moderate staining) + (3 × percentage of strong staining). This gives a value between 0 and 300. This value represents the staining intensity of each core of TMA. Statistical analyses were carried out using Microsoft Excel 2000 software (Microsoft, Washington, USA) and PASW statistics, version 19.0 (SPSS Inc., Chicago, Illinois, USA). Kaplan–Meier survival analysis was used to analyse the overall survival from the time of surgery. Patients alive at the time of follow-up point were censored. To compare the length of survival between curves, a log-rank test was performed. χ2 and Mann–Whitney (nonparametric) methods were used to test for association of clinicopathological variables with fascin mean value.
Results and discussion In an effort to understand fascin expression pattern in melanoma, we performed fascin immunohistochemistry DOI: 10.1097/CMR.0000000000000135
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
170 Melanoma Research
2015, Vol 25 No 2
Fig. 1
(a)
(b)
Metastasis
Primary melanoma
∗ Fascin1 histoscore
300
No stain (0)
200
100
sis ta as M et
Pr im ar y
0
(c) Fascin1 histoscore (mean)
Primary (%)
Metastasis (%)
0 >0, ≤100 >100, ≤200 >200, ≤300 Total cases
10 (3.9) 114 (44.9) 104 (40.9) 26 (10.2) 254
5 (10.4) 13 (27.1) 13 (27.1) 17 (35.4) 48
Weak (1)
Stage (n=302)
Moderate (2)
200
200
100
100
0
0 nc
Ulceration
0.8 Cum survival
1.
Pr
es
1.0
se
en
ce
Kaplan−Meier 50% survival
300
Ab
(d)
300
≤= 1 01 mm −2 m 2. 01 m −4 m m >4 m m
Strong (3)
(f)
∗∗
e
Fascin1 histoscore
(e)
Breslow thickness
0.6 0.4 P=0.067 0.2 0.0 0
2
4 6 8 10 Follow-up (years)
12
Fascin expression does not correlate with tumour stage but is increased in metastases. (a) Fascin IHC analysis of a human tumour array of melanomas and metastasis. Representative images are shown for various staining intensities of primary melanomas and metastasis. No staining, 0; weak staining, 1; moderate staining, 2 and strong staining, 3. Bar in original picture, 100 μm. (b) Box plot of fascin histoscore versus tumour stage (Mann–Whitney U-test,*P = 0.034). Primary tumours exhibit a lower level of fascin expression (n = 254, mean ± SD, 116.6 ± 4.6) versus metastasis (n = 48, mean ± SD, 152.1 ± 15.1). (c) Detailed relationship between fascin histoscore and tumour stage. (d) Kaplan–Meier curve showing the survival of patients with high or low fascin expression. No significant difference was seen at P = 0.05 (Log-rank test P = 0.067). The green line indicates patients with high fascin (histoscore > 114, patient n = 97). The blue line indicates patients with low fascin (histoscore ≤ 113, patient n = 90).‘ + ’ means censored. (e) Box plot of fascin histoscore versus ulceration, **P = 0.005. (f) Box plot of fascin histoscore versus Breslow thickness. IHC, immunohistochemistry.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Increased fascin in melanoma metastatic lesions Ma et al. 171
The relationship between fascin histoscore and ulceration status
The relationship between clinicopathological variables and fascin histoscores
Table 1
Table 2
Ulceration (n = 163) [n (%)] Fascin histoscore (mean) 0 Weak (>0, ≤ 100) Moderate (>100, ≤ 200) Strong (>200, ≤ 300) Total cases
Presence 2 22 14 3
(4.9) (53.7) (34.1) (7.3) 41
Fascin histoscore
Absence 6 46 52 18
(4.9) (37.7) (42.6) (14.8) 122
in a TMA containing a total of 562 cores of primary melanomas and melanoma metastasis (Met) [12]. A total of 302 cases (53.7%) were considered as valid samples (primary melanoma, 254; melanoma Met, 48) (Fig.1a). The fascin staining density was classified as 0 (no staining), 1 (weak staining), 2 (moderate staining), 3 (strong staining) and a histoscore was calculated based on the area of positive staining by two independent investigators [as detailed in the Methods section and (Fig.1a)]. We observed fascin staining primarily in the cytoplasm of melanoma cells. Fascin was absent in 10 of 254 primary tumours and five of 48 melanoma metastases. In all, 35.4% metastases expressed fascin at a high level and generally melanoma metastasis expressed a significantly higher level of fascin compared with primary melanomas (152.1 ± 15.1 vs. 116.6 ± 4.6, Mann–Whitney U-test, P = 0.034) (Fig 1b and c). Fascin levels were not significantly correlated with survival (P = 0.067) in 187 patients with primary tumours (Fig. 1d, green line = higher fascin score, blue line = lower fascin score). Ulceration and Breslow depth are currently among the two most important prognostic factors in melanoma, with the presence of ulceration and greater Breslow depth predicting poor outcome [13,14]. Surprisingly, low fascin was correlated with the presence of ulceration (P = 0.005) (Fig. 1e and Table 1). Although fascin expression was not correlated with Breslow depth (Pearson’s correlation coefficient r = − 0.142, P = 0.05, Mann–Whitney U-test, P = 0.164). For those primary melanomas, lower fascin expression was associated with higher Breslow depth when Breslow depth was less than 4 mm, whereas fascin expression level was elevated when tumour thickness was more than 4 mm and tumours were tending to metastasis (Fig. 1f). Fascin expression did not correlate to other clinicopathological characteristics, such as age, sex and sun-exposure (Table 2 and data not shown). Neither did fascin level differ within the histological stage of lentigo (139.5 ± 22.5, n = 14), superficial spreading melanoma (116.2 ± 6.4, n = 142), nodular (100.4 ± 13.3, n = 26) and acral (160.3 ± 18.1, n = 5) (Table 2 and data not shown). These data may indicate that fascin expression is increased during metastasis, but that its expression does not correlate with melanoma progression. However, our observation on the relationships between fascin expression and presence of ulceration is different from previous findings in gastrointestinal stromal tumours, where Yamamoto et al. [15] found that high fascin was
Variables
Negative (n = 8)
Low (n = 88)
High (n = 91)
1 7
36 52
36 55
0.635
0 8 0
12 74 1
15 73 3
0.581
1 6 1 0
4 70 14 0
9 66 11 5
0.233
2 5 1 0
24 52 11 1
15 65 10 1
0.024
5 2 1
68 16 4
75 8 8
0.181
Sex Male (n = 73) Female (n = 114) Site Sun-exposed (n = 27) Sun-protected (n = 155) Acral (n = 4) Histology Lentigo (n = 14) SSM (n = 142) Nodular (n = 26) Acral (n = 5) Ulceration Presence (n = 41) Absence (n = 122) Incipient (n = 22) Unknown (n = 2) Status Alive (n = 148) Dead (n = 26) Nonmelanoma death (n = 13)
P
SSM, superficial spreading melanoma.
significantly correlated with mucosal ulceration. The correlation of fascin expression with ulceration could indicate a role for fascin expression in healing and regeneration following damage, as has been postulated in inflammatory bowel diseases [16] and as such could be advantageous to the host rather than the tumour. Yildiz et al. [11] published a small study of 12 malignant melanoma, 20 benign nevi and 12 dysplastic nevi and found that malignant melanomas (25%) express fascin less frequently than benign nevi (95%, P < 0.001) and dysplastic nevi (67%, P = 0.036). Goncharuk et al. [10] examined fascin expression in normal skin and different types of benign and malignant skin tumours and found only 19% of melanoma (3/16) express fascin. Our data on fascin expression levels in a reasonably large number of human melanocytic lesions suggest that fascin is upregulated in melanoma metastasis but levels do not correlate with outcome and are variable. In primary melanoma, low fascin expression levels correlated with ulceration. We previously showed that fascin was transiently upregulated in melanoblasts of mice during embryogenesis and conferred enhanced migration [5]. Thus, expression of fascin in metastatic lesions might correlate with a reactivation of this embryonic migration pattern, but clearly fascin is not limiting for melanoma spread. Our study highlights the complexity of fascin’s involvement in melanoma and suggests that fascin expression is not likely to be useful as a prognostic indicator for melanoma.
Acknowledgements The authors thank Colin Nixon of Beatson Histology Services for his excellent help with TMA processing.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
172
Melanoma Research
2015, Vol 25 No 2
Conflicts of interest
Y.M., L.M.M. and O.J.S. are funded by a Core CRUK grant. For the remaining authors there are no conflicts of interest.
9
References 1 2 3
4 5
6
7
8
Adams JC. Roles of fascin in cell adhesion and motility. Curr Opin Cell Biol 2004; 16:590–596. Jayo A, Parsons M. Fascin: a key regulator of cytoskeletal dynamics. Int J Biochem Cell Biol 2010; 42:1614–1617. Li A, Dawson JC, Forero-Vargas M, Spence HJ, Yu X, König I, et al. The actinbundling protein fascin stabilizes actin in invadopodia and potentiates protrusive invasion. Curr Biol 2010; 20:339–345. Machesky LM, Li A. Fascin: Invasive filopodia promoting metastasis. Commun Integr Biol 2010; 3:263–270. Ma Y, Li A, Faller WJ, Libertini S, Fiorito F, Gillespie DA, et al. Fascin 1 is transiently expressed in mouse melanoblasts during development and promotes migration and proliferation. Development 2013; 140:2203–2211. Tan VY, Lewis SJ, Adams JC, Martin RM. Association of fascin-1 with mortality, disease progression and metastasis in carcinomas: a systematic review and meta-analysis. BMC Med 2013; 11:52. Li A, Morton JP, Ma Y, Karim SA, Zhou Y, Faller WJ, et al. Fascin is regulated by slug, promotes progression of pancreatic cancer in mice, and is associated with patient outcomes. Gastroenterology 2014; 146:1386–1396. Hashimoto Y, Skacel M, Lavery IC, Mukherjee AL, Casey G, Adams JC. Prognostic significance of fascin expression in advanced colorectal cancer:
10 11 12
13
14 15
16
an immunohistochemical study of colorectal adenomas and adenocarcinomas. BMC Cancer 2006; 6:241. 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. Dis Markers 2007; 23:153–160. Goncharuk VN, Ross JS, Carlson JA. Actin-binding protein fascin expression in skin neoplasia. J Cutan Pathol 2002; 29:430–438. Yildiz L, Kefeli M, Aydin O, Kandemir B. Fascin expression in melanocytic lesions of the skin. Eur J Dermatol 2009; 19:445–450. Brown ER, Doig T, Anderson N, Brenn T, Doherty V, Xu Y, et al. Association of galectin-3 expression with melanoma progression and prognosis. Eur J Cancer 2012; 48:865–874. Balch CM, Soong SJ, Gershenwald JE, Thompson JF, Reintgen DS, Cascinelli N, et al. Prognostic factors analysis of 17 600 melanoma patients: validation of the American Joint Committee on Cancer Melanoma Staging System. J Clin Oncol 2001; 19:3622–3634. Dickson PV, Gershenwald JE. Staging and prognosis of cutaneous melanoma. Surg Oncol Clin N Am 2011; 20:1–17. Yamamoto H, Kohashi K, Fujita A, Oda Y. Fascin-1 overexpression and miR-133b downregulation in the progression of gastrointestinal stromal tumor. Mod Pathol 2013; 26:563–571. Qualtrough D, Smallwood K, Littlejohns D, Pignatelli M. The actin-bundling protein fascin is overexpressed in inflammatory bowel disease and may be important in tissue repair. BMC Gastroenterol 2011; 11:14.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
