- Email: [email protected]
Tumor budding as a prognostic marker in laryngeal carcinoma
ARTICLE IN PRESS Pathology – Research and Practice 206 (2010) 88–92
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.de/prp
Original Article
Tumor budding as a prognostic marker in laryngeal carcinoma Sulen Sarioglu a,, Cagdas Acara a, Fadime Can Akman b, Nihal Dag b, Cenk Ecevit c, Ahmet Omer Ikiz c, ¨ Head and Neck Tumour Group (DEHNTG) Oguz H. Cetinayak b, Emel Ada d, for Dokuz Eylul a
Department of Pathology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey Department of Radiation Oncology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey c Department of Head and Neck Surgery, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey d Department of Radiology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey b
a r t i c l e in f o
a b s t r a c t
Article history: Received 8 October 2008 Received in revised form 16 July 2009 Accepted 9 September 2009
Tumor budding is recognized as an adverse prognostic factor for a few carcinoma types. We evaluated the importance of this finding in 64 laryngeal carcinoma patients treated with surgery and radiotheraphy, with a median follow up of 53 (6–181) months. Budding was determined by re-evaluating hematoxylin–eosin (H&E) stained tumor sections, and classified as mild, moderate, and marked. Budding was not identified in 14 (21.88%) cases. Mild, moderate, and marked budding was observed in 21 (32.81%), 15 (23.44%), and 14 (21.88%) cases, respectively. Only when cases with marked budding were compared with the others, statistically significant results were obtained. In patients with and without marked budding, local disease-free survival, metastatic disease-free survival, and overall survival were 71% and 97% (p =0.72), 77% and 93% (p =0.038), and 77% and 75% (p =0.71), respectively. Marked budding was identified as a prognostic factor by univariate analysis for distant metastasisfree survival. Multi-variate analysis, by which well-recognized prognostic histopathologic parameters were evaluated, revealed that only the number of metastatic lymph nodes and budding were significantly associated with distant metastasis (p= 0.02 and p= 0.044), respectively. These results suggest that budding might be a valuable prognostic factor, particularly for distant metastasis in laryngeal carcinomas. & 2009 Elsevier GmbH. All rights reserved.
Keywords: Laryngeal carcinoma Budding Prognosis Distant metastasis
Introduction Loss of differentiation and budding or sprouting of the tumor cells has been recognized as an adverse prognostic factor for a few types of carcinomas for nearly 3 decades [3,10]. Tumor budding is a histopathological finding identified by the observation of small cell clusters composed of o5 cells at the invasive tumor margin [6,7]. Colorectal carcinomas have been evaluated extensively for tumor budding. It has been reported that budding is a histopathological finding that is related to lymph node metastasis, poor prognosis, or locoregional recurrence in various series [2,6,7,16,17]. Budding has been evaluated in lung adenocarcinomas, as well as in anal and esophageal squamous cell carcinomas. These studies favor budding as a prognostic factor, albeit as a positive factor for anal carcinomas [8,9,11,14]. Based on the finding
Corresponding author. Tel.: + 90 2324123408; fax: + 90 2322777274; mob.: + 90 5325630283. E-mail address: [email protected] (S. Sarioglu).
0344-0338/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.prp.2009.09.006
that tumor budding is a valuable factor for determining prognostic features in esophageal and anal squamous cell carcinomas as well as adenocarcinomas, the prognostic value of tumor budding in laryngeal carcinomas, in a series of patients treated with surgery and radiotherapy, was evaluated in the present study.
Materials and methods Sixty-four patients with laryngeal carcinoma who were evaluated in this study were all diagnosed and treated at Dokuz Eylul University Hospital and had complete follow-up information. Patients were treated by partial or total laryngectomy and neck dissection, followed by adjuvant radiotheraphy. Ten patients received concomitant cisplatin-based chemotherapy. Age, gender, tumor location, clinical tumor stage and lymph node stage, type of surgery, pathological tumor and lymph node stage, and the number of metastatic lymph nodes, perinodal invasion, tumor grade, treatment scheme with radio/chemotherapy, time and localization of loco/regional recurrence, and distant metastasis were determined from the medical records of the patients. Second
ARTICLE IN PRESS S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
Fig. 1. Tumor budding of the laryngeal squamous cell carcinoma (H&E, original magnification 40).
primary tumors, time of disease-related and all-cause death, and the date of the last follow-up time were recorded. Histopathological evaluation Hematoxylin and eosin (H&E) stained sections of the patients with squamous cell carcinoma were evaluated for the positivity and extent of budding by two pathologists (SS and CA) using light microscopy at high magnification ( 400). Budding was accepted as positive in the presence of small clusters ( r5 cells and Z1 cells) of dedifferentiated tumor cells were identified at the invasive margin at any of the sections (Fig. 1). Furthermore, the largest section of the whole lesion was selected, and budding was classified as mild in the presence of budding at r 13 of the entire margin, moderate in the presence of budding at 13–23 of the margin, and marked in the presence of budding at Z 23 of the invasive margin, as described previously by Kanazawa et al. [7]. Statistical analysis Budding was analyzed by three different methods. First, budding was accepted as positive in the presence of budding in any proportion. Then, cases were evaluated and scored according to the presence of budding as mild, moderate, and marked. Finally, cases with marked budding were compared with the remaining cases. Statistical analysis was conducted by Scientific Package for Social Sciences (SPSS, 11). For comparison of the groups, nonparametric tests, such as a chi-square test and Mann–Whitney U test, were applied. Survival curves were plotted by the Kaplan– Meier method. For comparing survivals of different groups, the log rank test and Cox regression were applied for uni- and multivariate analyses for tumor and lymph node stage, number of metastatic lymph nodes, perinodal invasion, differentiation, and budding.
Results The study population consisted of 60 (93.75%) males and 4 (6.25%) females. The median age of the patients was 56 years (range, 36–86 years). Eight (12.5%) patients underwent partial and 56 patients (87.5%) underwent total laryngectomy. All patients had squamous cell carcinoma, but two cases were of the basaloid
89
variant. Twenty-two cases (34.38%) were supraglottic, 29 (45.31%) were glottic, 4 (6.25%) were subglottic, and 9 (14.06%) were transglottic. Pathological T stages were as follows: T2 for 5 patients (7.8%), T3 for 28 patients (43.8%), and T4 for 30 patients (46.9%). Thirty patients (46.9%) had no have lymph node metastasis (pN0), and the numbers of pN1, pN2, and pN3 cases were 13 (20.3%), 19 (29.7%), and 1 (0.16%), respectively. Seventeen patients (26.6%) had perinodal invasion. All patients were considered for post-operative radiotherapy for the following reasons: pT stages 3 and 4, pN(+ ), patients with pT2 a nd a positive surgical margin, and/or subglottic extension. All patients received adjuvant radiotherapy, and 10 received chemotherapy. Budding was not identified in 14 (21.88%) patients. The numbers of cases with mild, moderate, and marked scores of budding were 21 (32.81%), 15 (23.44%), and 14 (21.88%), respectively (Table 1). The median follow-up of the patients was 53 months (range, 6–181 months). Four (6.25%) patients had locoregional recurrence, and 10 patients (15.6%) had distant metastases. The 5-year overall, locoregional disease-free, metastatic disease-free, and disease-specific survival rates were 76%, 94%, 86%, and 92%, respectively. Budding scores of none, mild, moderate, and marked for local disease free survival were 100%, 95%, 83%, and 100%, respectively (p = 0.29). For distant metastasis free survival, the above mentioned budding scores were 100%, 89%, 69%, and 84%, respectively (p = 0.15), and for overall survival, the budding scores were 90%, 68%, 67%, and 84%, respectively (p = 0.4). When cases with marked budding, observed at Z 23 of the invasive margin of tumor, were compared with the remaining cases, local disease-free survival, metastatic disease-free survival, and overall survival in patients with and without marked budding were 71% and 97% (p =0.72), 77% and 93% (p =0.038; Fig. 2), and 77% and 75% (p = 0.71), respectively. Univariate analysis, by which well-known histopathological and clinical prognostic factors were evaluated, revealed that clinical and pathological T stage grade, perinodal invasion, and budding were not significantly associated with overall survival, disease-specific survival, and locoregional disease-free survival, whereas marked budding was significantly associated with metastatic disease-free survival (p =0.038). Based on the results of univariate analysis, marked budding was identified as a prognostic factor in this laryngeal carcinoma series for distant metastasis-free survival. Multi-variate analysis, by which well-recognized prognostic histopathologic parameters, including the above-mentioned parameters for univariate analysis, were evaluated, revealed that only the number of metastatic lymph nodes and budding were significantly associated with distant metastasis (p= 0.02 and p= 0.044, respectively; Table 2).
Discussion Our results suggest that marked budding of dedifferentiated tumor cells at the invasive margin of laryngeal carcinomas may have prognostic significance for distant metastasis in patients treated with surgery and radiotherapy. Since the number of cases with locoregional recurrence was small in this series, the results regarding the insignificance of budding in this aspect are debatable. Budding appears to be associated with the nuclear location of b-catenin, which is related to E-cadherin aberrations, along with loss of the expression of epithelial cell adhesion molecule (Ep-CAM). These changes contribute to the loss of intercellular
ARTICLE IN PRESS 90
S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
Table 1 Tumor budding patterns with clinical and histopathological tumor characteristics. Characteristics
No budding N (%)
Mild budding N (%)
Moderate budding N (%)
Marked budding N (%)
Total N (%)
Gender Male Female
14 (23.33) 0
19 (31.67) 2 (3.12)
14 (23.33) 1 (1.56)
13 (21.67) 1 (1.56)
60 (93.75) 4 (6.25)
Primary tumor site Supraglottic Glottic Subglottic Transglottic
6 (27.27) 5 (17.24) 0 3 (33.33)
9 9 1 2
Tumor grade I II III
7 (28) 5 (20.83) 2 (13.33)
9 (36) 7 (29.17) 5 (33.33)
5 (20) 8 (33.33) 2 (13.33)
4 (16) 4 (16.67) 6 (40)
25 (39.06) 24 (36.92) 15 (23.08)
Clinical T stage T II T III T IV
3 (23.08) 10 (30.30) 1 (5.55)
2 (15.38) 12 (36.36) 7 (38.89)
5 (38.46) 5 (15.15) 5 (27.78)
3 (23.08) 6 (18.18) 5 (27.78)
13 (20.31) 33 (51.56) 18 (28.13)
Clinical N stage N0 N1 N2–3
12 (25.53) 1 (25) 1 (7.69)
17 (36.17) – 4 (30.77)
8 (17.02) 1 (25) 6 (46.15)
10 (21.28) 2 (50) 2 (15.38)
47 (73.44) 4 (6.25) 13 (20.31)
Pathological T stage T II T III T IV
1 (20) 7 (25.93) 6 (18.75)
2 (40) 9 (33.33) 10 (31.25)
1 (20) 4 (14.81) 10 (31.25)
1 (20) 8 (29.63) 6 (18.75)
5 (7.81) 27 (42.19) 32 (50)
Pathological N stage N0 N1 N2–3
10 (32.26) 2 (15.38) 2 (10)
10 (32.26) 4 (30.77) 7 (35)
4 (12.90) 5 (38.46) 6 (30)
7 (22.58) 2 (15.38) 5 (25)
31 (48.44) 13 (20.31) 20 (31.25)
Pericapsular invasion No Yes
13 (27.66) 1 (5.88)
16 (34.04) 5 (29.41)
7 (14.89) 8 (47.06)
11 (23.40) 3 (17.65)
47 (73.44) 17 (26.56)
(40.91) (31.03) (25) (22.22)
1 9 2 3
(4.54) (31.03) (50) (33.33)
6 6 1 1
(27.27) (20.69) (25) (11.11)
22 29 4 9
(34.38) (43.31) (6.25) (14.06)
Cumulative Metastatic Disease Free Survival Rate
100
80
60
Budding not marked budding marked budding
40
20
0 0
30
60
90
120
150
180
Months Fig. 2. Distant metastasis-free survival.
adhesions [4]. Initially, the invasive margin of the tumor had attracted more attention, and budding was evaluated at the invasive margin [19–21]. However, Morodomi et al. [10] have noted that budding can be identified in biopsy specimens and is closely related to nodal metastasis. This suggests that budding is
not only a feature of the invasive tumor margin, but also a common feature of both superficial and deep portions of tumors with aggressive behavior, along with well-known carcinogenesis steps, such as destruction of the basement membrane or loss of intercellular adhesions [3]. Guzin´ska-Ustymowicz [5] has
ARTICLE IN PRESS S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
91
Table 2 The results of the multivariate analysis concerning prognostic factors. Prognostic factors
Clinical stage: II III IV Pathological stage: II III IV Number of pathologic lymph nodes: N0 r3 43 Peri-nodal invasion: ( ) (+) Grade: I II III Budding: Marked Non-marked or negative
Univariate analysis for
Multivariate analysis for
OS
LRDFS
DMFS
OS
LRDFS
DMFS
p = 0.14
p = 0.59
p= 0.49
p= 0.5
p = 0.49
p = 0.49
p = 0.61
p = 0.6
p= 0.69
p= 0.6
p = 0.36
p = 0.53
p = 0.07
p = 0.56
p= 0.07
p = 0.018
p = 0.61
p = 0.021
p = 0.07
p = 0.93
p= 0.29
p= 0.8
p = 0.4
p = 0.3
p = 0.43
p = 0.44
p= 0.91
p= 0.22
p = 0.84
p = 0.67
p = 0.71
p = 0.71
p= 0.039
p= 0.9
p = 0.64
p = 0.044
OS= overall survival; LRDFS=loco-regional disease free survival; DMFS = distant metastasis free survival.
reported increased aggressive behavior of colorectal carcinoma in the presence of MMP-9 and cathepsin B expression at the site of budding. This might be the reason for laryngeal carcinomas, since the expression of MMP increases with progression of carcinoma [5,15,22]. Budding can be identified by observing H&E stained tumor sections using light microscopy. Different methods of quantification have been applied in different series. We preferred using the criteria described by Kanazawa et al. [7], considering the extent of budding at the invasive tumor margin in H&E stained sections. Ueno et al. [19–21] have evaluated the region with most budding, and the budding was scored as I, II, III, and IV, according to the number of budding units as 0–4, 5–9, 10–19, and 420, respectively. The same method was applied in different series [8,16]. Roh et al. [14] have divided the cases into 2 groups (cases with o5 or 45 budding under a 20 objective at the tumor region with the most extensive budding). Tadahiko et al. [17] have counted the budding units at the invasive margin of the whole tumor in one section, and this was probably possible as they only evaluated a series of T1 colorectal carcinomas, i.e., a group of relatively small tumors. Some authors have evaluated tumor budding in sections stained by immunohistochemical methods. Strong expression of the laminin-5 gamma 2-chain in the budding cells has been used for identifying tumor budding in anal carcinomas [11]. The method for assessing tumor budding has not yet been standardized, albeit the method we preferred in this series of laryngeal carcinomas is of prognostic significance for distant metastases. Oosterkamp et al. [12] have reported that pericapsular invasion of metastatic lymph nodes is the most important risk factor for distant metastasis, followed by lymph node metastasis and T4 tumor stage, in a series of 165 patients with laryngeal carcinoma. In our series, we also identified the number of metastatic lymph
nodes as an independent factor for determining distant metastasis, along with marked tumor budding. Tumor-related histopathologic features of prognostic or predictive value include tumor type, pathologic T stage, grade, lymphovascular invasion, perineural invasion, and infiltrative growth pattern for laryngeal carcinomas [1,13,18]. The results in this series suggest that budding might be a valuable prognostic factor, particularly for distant metastasis in laryngeal carcinomas.
References [1] F.C. Akman, N. Dag, O.U. Ataman, C. Ecevit, A.O. Ikiz, I. Arslan, S. Sarıoglu, E. Ada, M. Kinay, The impact of treatment center on the outcome of patients with laryngeal cancer treated with surgery and radiotherapy, Eur. Arch. Otorhinolaryngol. 265 (2008) 1245–1255. [2] H.J. Choi, K.J. Park, J.S. Shin, M.S. Roh, H.C. Kwon, H.S. Lee, Tumor budding as a prognostic marker in stage-III rectal carcinoma, Int. J. Colorectal Dis. 22 (2007) 863–868. [3] H. Gabbert, Mechanisms of tumor invasion: evidence from in vivo observations, Cancer Metastasis Rev. 4 (1985) 293–309. [4] M.J. Gosens, L.C. van Kempen, C.J. van de Velde, J.H. van Krieken, I.D. Nagtegaal, Loss of membranous Ep-CAM in budding colorectal carcinoma cells, Mod. Pathol. 20 (2007) 221–232. [5] K. Guzin´ska-Ustymowicz, MMP-9 and cathepsin B expression in tumor budding as an indicator of a more aggressive phenotype of colorectal cancer (CRC), Anticancer Res. 26 (2006) 1589–1594. [6] K. Hase, C. Shatney, D. Johnson, M. Trollope, M. Vierra, Prognostic value of tumor ‘‘budding’’ in patients with colorectal cancer, Dis. Colon Rectum 36 (1993) 627–635. [7] H. Kanazawa, H. Mitomi, Y. Nishiyama, I. Kishimoto, N. Fukui, Watanabe. Nakamura, Tumour budding at invasive margins and outcome in colorectal cancer, Colorectal Dis. 10 (2007) 41–47. [8] M. Koike, Y. Kodera, Y. Itoh, G. Nakayama, M. Fujiwara, N. Hamajima, A. Nakao, Multivariate analysis of the pathologic features of esophageal squamous cell cancer: tumor budding is a significant independent prognostic factor, Ann. Surg. Oncol. 15 (2008) 1977–1982. [9] Y. Moriya, T. Niki, T. Yamada, Y. Matsuno, H. Kondo, S. Hirohashi, Increased expression of laminin-5 and its prognostic significance in lung adenocarcinomas of small size, Cancer 91 (2001) 1129–1141.
ARTICLE IN PRESS 92
S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
[10] T. Morodomi, H. Isomoto, K. Shirouzu, K. Kakegawa, K. Irie, M. Morimatsu, An index for estimating the probability of lymph node metastasis in rectal cancers. Lymph node metastasis and the histopathology of actively invasive regions of cancer, Cancer 63 (1989) 539–543. [11] P.J. Nilsson, C. Rubio, C. Lenander, G. Auer, B. Glimelius, Tumour budding detected by laminin-5 {gamma} 2-chain immunohistochemistry is of prognostic value in epidermoid anal cancer, Ann. Oncol. 16 (2005) 893–898. [12] S. Oosterkamp, J.M. de Jong, P.L. Van den Ende, J.J. Manni, C. Dehing-Oberije, B. Kremer, Predictive value of lymph node metastases and extracapsular extension for the risk of distant metastases in laryngeal carcinoma, Laryngoscope 116 (2006) 2067–2070. [13] A. Ozdek, S. Sarac, M.U. Akyol, O.F. Unal, A. Sungur, Histopathological predictors of occult lymph node metastases in supraglottic squamous cell carcinomas, Eur. Arch. Otorhinolaryngol. 257 (2000) 389–392. [14] M.S. Roh, J.I. Lee, P.J. Choi, Tumor budding as a useful prognostic marker in esophageal squamous cell carcinoma, Dis. Esophagus. 17 (2004) 333–337. [15] S. Sarioglu, E. Ozer, F. Kirimca, B. Sis, U. Pabuc- cuoglu, Matrix metalloprotei˘ nase-2 expression in laryngeal preneoplastic and neoplastic lesions, Pathol. Res. Pract. 197 (2001) 483–486. [16] D.K. Sohn, H.J. Chang, J.W. Park, D.H. Choi, K.S. Han, C.W. Hong, K.H. Jung, D.Y. Kim, S.B. Lim, H.S. Choi, S.Y. Jeong, Histopathological risk factors for lymph
[17]
[18]
[19]
[20]
[21]
[22]
node metastasis in submucosal invasive colorectal carcinoma of pedunculated or semipedunculated type, J. Clin. Pathol. 60 (2007) 912–915. M. Tadahiko, H. Matsuoka, M. Sugiyama, N. Abe, A. Sakamoto, Y. Atomi, Actual number of tumor budding as a new tool for the individualization of treatment of T1 colorectal carcinomas, J. Gastroenterol. Hepatol. 21 (2006) 1115–1121. L.D. Thompson, B.M. Wenig, D.K. Heffner, D.R. Gnepp, Exophytic and papillary squamous cell carcinomas of the larynx: a clinicopathologic series of 104 cases, Otolaryngol. Head Neck Surg. 120 (1999) 718–724. H. Ueno, A.M. Jones, K.H. Wilkinson, J.R. Jass, I.C. Talbot, Histological categorisation of fibrotic cancer stroma in advanced rectal cancer, Gut 53 (2004) 581–586. H. Ueno, J. Murphy, J.R. Jass, H. Mochizuki, I.C. Talbot, Tumour ‘budding’ as an index to estimate the potential of aggressiveness in rectal cancer, Histopathology 40 (2002) 127–132. H. Ueno, H. Mochizuki, E. Shinto, Y. Hashiguchi, K. Hase, I.C. Talbot, Histological indices in biopsy specimens for estimating the probability of extended local spread in rectal cancer, Cancer 94 (2002) 2882–2891. M. Xie, Y. Sun, Y. Li, Expression of matrix metalloproteinases in supraglottic carcinoma and its clinical implication for estimating lymph node metastases, Laryngoscope 114 (2004) 2243–2248.
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.de/prp
Original Article
Tumor budding as a prognostic marker in laryngeal carcinoma Sulen Sarioglu a,, Cagdas Acara a, Fadime Can Akman b, Nihal Dag b, Cenk Ecevit c, Ahmet Omer Ikiz c, ¨ Head and Neck Tumour Group (DEHNTG) Oguz H. Cetinayak b, Emel Ada d, for Dokuz Eylul a
Department of Pathology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey Department of Radiation Oncology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey c Department of Head and Neck Surgery, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey d Department of Radiology, Medical School, Dokuz Eyl¨ ul University, 35340 Inciralti, Izmir, Turkey b
a r t i c l e in f o
a b s t r a c t
Article history: Received 8 October 2008 Received in revised form 16 July 2009 Accepted 9 September 2009
Tumor budding is recognized as an adverse prognostic factor for a few carcinoma types. We evaluated the importance of this finding in 64 laryngeal carcinoma patients treated with surgery and radiotheraphy, with a median follow up of 53 (6–181) months. Budding was determined by re-evaluating hematoxylin–eosin (H&E) stained tumor sections, and classified as mild, moderate, and marked. Budding was not identified in 14 (21.88%) cases. Mild, moderate, and marked budding was observed in 21 (32.81%), 15 (23.44%), and 14 (21.88%) cases, respectively. Only when cases with marked budding were compared with the others, statistically significant results were obtained. In patients with and without marked budding, local disease-free survival, metastatic disease-free survival, and overall survival were 71% and 97% (p =0.72), 77% and 93% (p =0.038), and 77% and 75% (p =0.71), respectively. Marked budding was identified as a prognostic factor by univariate analysis for distant metastasisfree survival. Multi-variate analysis, by which well-recognized prognostic histopathologic parameters were evaluated, revealed that only the number of metastatic lymph nodes and budding were significantly associated with distant metastasis (p= 0.02 and p= 0.044), respectively. These results suggest that budding might be a valuable prognostic factor, particularly for distant metastasis in laryngeal carcinomas. & 2009 Elsevier GmbH. All rights reserved.
Keywords: Laryngeal carcinoma Budding Prognosis Distant metastasis
Introduction Loss of differentiation and budding or sprouting of the tumor cells has been recognized as an adverse prognostic factor for a few types of carcinomas for nearly 3 decades [3,10]. Tumor budding is a histopathological finding identified by the observation of small cell clusters composed of o5 cells at the invasive tumor margin [6,7]. Colorectal carcinomas have been evaluated extensively for tumor budding. It has been reported that budding is a histopathological finding that is related to lymph node metastasis, poor prognosis, or locoregional recurrence in various series [2,6,7,16,17]. Budding has been evaluated in lung adenocarcinomas, as well as in anal and esophageal squamous cell carcinomas. These studies favor budding as a prognostic factor, albeit as a positive factor for anal carcinomas [8,9,11,14]. Based on the finding
Corresponding author. Tel.: + 90 2324123408; fax: + 90 2322777274; mob.: + 90 5325630283. E-mail address: [email protected] (S. Sarioglu).
0344-0338/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.prp.2009.09.006
that tumor budding is a valuable factor for determining prognostic features in esophageal and anal squamous cell carcinomas as well as adenocarcinomas, the prognostic value of tumor budding in laryngeal carcinomas, in a series of patients treated with surgery and radiotherapy, was evaluated in the present study.
Materials and methods Sixty-four patients with laryngeal carcinoma who were evaluated in this study were all diagnosed and treated at Dokuz Eylul University Hospital and had complete follow-up information. Patients were treated by partial or total laryngectomy and neck dissection, followed by adjuvant radiotheraphy. Ten patients received concomitant cisplatin-based chemotherapy. Age, gender, tumor location, clinical tumor stage and lymph node stage, type of surgery, pathological tumor and lymph node stage, and the number of metastatic lymph nodes, perinodal invasion, tumor grade, treatment scheme with radio/chemotherapy, time and localization of loco/regional recurrence, and distant metastasis were determined from the medical records of the patients. Second
ARTICLE IN PRESS S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
Fig. 1. Tumor budding of the laryngeal squamous cell carcinoma (H&E, original magnification 40).
primary tumors, time of disease-related and all-cause death, and the date of the last follow-up time were recorded. Histopathological evaluation Hematoxylin and eosin (H&E) stained sections of the patients with squamous cell carcinoma were evaluated for the positivity and extent of budding by two pathologists (SS and CA) using light microscopy at high magnification ( 400). Budding was accepted as positive in the presence of small clusters ( r5 cells and Z1 cells) of dedifferentiated tumor cells were identified at the invasive margin at any of the sections (Fig. 1). Furthermore, the largest section of the whole lesion was selected, and budding was classified as mild in the presence of budding at r 13 of the entire margin, moderate in the presence of budding at 13–23 of the margin, and marked in the presence of budding at Z 23 of the invasive margin, as described previously by Kanazawa et al. [7]. Statistical analysis Budding was analyzed by three different methods. First, budding was accepted as positive in the presence of budding in any proportion. Then, cases were evaluated and scored according to the presence of budding as mild, moderate, and marked. Finally, cases with marked budding were compared with the remaining cases. Statistical analysis was conducted by Scientific Package for Social Sciences (SPSS, 11). For comparison of the groups, nonparametric tests, such as a chi-square test and Mann–Whitney U test, were applied. Survival curves were plotted by the Kaplan– Meier method. For comparing survivals of different groups, the log rank test and Cox regression were applied for uni- and multivariate analyses for tumor and lymph node stage, number of metastatic lymph nodes, perinodal invasion, differentiation, and budding.
Results The study population consisted of 60 (93.75%) males and 4 (6.25%) females. The median age of the patients was 56 years (range, 36–86 years). Eight (12.5%) patients underwent partial and 56 patients (87.5%) underwent total laryngectomy. All patients had squamous cell carcinoma, but two cases were of the basaloid
89
variant. Twenty-two cases (34.38%) were supraglottic, 29 (45.31%) were glottic, 4 (6.25%) were subglottic, and 9 (14.06%) were transglottic. Pathological T stages were as follows: T2 for 5 patients (7.8%), T3 for 28 patients (43.8%), and T4 for 30 patients (46.9%). Thirty patients (46.9%) had no have lymph node metastasis (pN0), and the numbers of pN1, pN2, and pN3 cases were 13 (20.3%), 19 (29.7%), and 1 (0.16%), respectively. Seventeen patients (26.6%) had perinodal invasion. All patients were considered for post-operative radiotherapy for the following reasons: pT stages 3 and 4, pN(+ ), patients with pT2 a nd a positive surgical margin, and/or subglottic extension. All patients received adjuvant radiotherapy, and 10 received chemotherapy. Budding was not identified in 14 (21.88%) patients. The numbers of cases with mild, moderate, and marked scores of budding were 21 (32.81%), 15 (23.44%), and 14 (21.88%), respectively (Table 1). The median follow-up of the patients was 53 months (range, 6–181 months). Four (6.25%) patients had locoregional recurrence, and 10 patients (15.6%) had distant metastases. The 5-year overall, locoregional disease-free, metastatic disease-free, and disease-specific survival rates were 76%, 94%, 86%, and 92%, respectively. Budding scores of none, mild, moderate, and marked for local disease free survival were 100%, 95%, 83%, and 100%, respectively (p = 0.29). For distant metastasis free survival, the above mentioned budding scores were 100%, 89%, 69%, and 84%, respectively (p = 0.15), and for overall survival, the budding scores were 90%, 68%, 67%, and 84%, respectively (p = 0.4). When cases with marked budding, observed at Z 23 of the invasive margin of tumor, were compared with the remaining cases, local disease-free survival, metastatic disease-free survival, and overall survival in patients with and without marked budding were 71% and 97% (p =0.72), 77% and 93% (p =0.038; Fig. 2), and 77% and 75% (p = 0.71), respectively. Univariate analysis, by which well-known histopathological and clinical prognostic factors were evaluated, revealed that clinical and pathological T stage grade, perinodal invasion, and budding were not significantly associated with overall survival, disease-specific survival, and locoregional disease-free survival, whereas marked budding was significantly associated with metastatic disease-free survival (p =0.038). Based on the results of univariate analysis, marked budding was identified as a prognostic factor in this laryngeal carcinoma series for distant metastasis-free survival. Multi-variate analysis, by which well-recognized prognostic histopathologic parameters, including the above-mentioned parameters for univariate analysis, were evaluated, revealed that only the number of metastatic lymph nodes and budding were significantly associated with distant metastasis (p= 0.02 and p= 0.044, respectively; Table 2).
Discussion Our results suggest that marked budding of dedifferentiated tumor cells at the invasive margin of laryngeal carcinomas may have prognostic significance for distant metastasis in patients treated with surgery and radiotherapy. Since the number of cases with locoregional recurrence was small in this series, the results regarding the insignificance of budding in this aspect are debatable. Budding appears to be associated with the nuclear location of b-catenin, which is related to E-cadherin aberrations, along with loss of the expression of epithelial cell adhesion molecule (Ep-CAM). These changes contribute to the loss of intercellular
ARTICLE IN PRESS 90
S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
Table 1 Tumor budding patterns with clinical and histopathological tumor characteristics. Characteristics
No budding N (%)
Mild budding N (%)
Moderate budding N (%)
Marked budding N (%)
Total N (%)
Gender Male Female
14 (23.33) 0
19 (31.67) 2 (3.12)
14 (23.33) 1 (1.56)
13 (21.67) 1 (1.56)
60 (93.75) 4 (6.25)
Primary tumor site Supraglottic Glottic Subglottic Transglottic
6 (27.27) 5 (17.24) 0 3 (33.33)
9 9 1 2
Tumor grade I II III
7 (28) 5 (20.83) 2 (13.33)
9 (36) 7 (29.17) 5 (33.33)
5 (20) 8 (33.33) 2 (13.33)
4 (16) 4 (16.67) 6 (40)
25 (39.06) 24 (36.92) 15 (23.08)
Clinical T stage T II T III T IV
3 (23.08) 10 (30.30) 1 (5.55)
2 (15.38) 12 (36.36) 7 (38.89)
5 (38.46) 5 (15.15) 5 (27.78)
3 (23.08) 6 (18.18) 5 (27.78)
13 (20.31) 33 (51.56) 18 (28.13)
Clinical N stage N0 N1 N2–3
12 (25.53) 1 (25) 1 (7.69)
17 (36.17) – 4 (30.77)
8 (17.02) 1 (25) 6 (46.15)
10 (21.28) 2 (50) 2 (15.38)
47 (73.44) 4 (6.25) 13 (20.31)
Pathological T stage T II T III T IV
1 (20) 7 (25.93) 6 (18.75)
2 (40) 9 (33.33) 10 (31.25)
1 (20) 4 (14.81) 10 (31.25)
1 (20) 8 (29.63) 6 (18.75)
5 (7.81) 27 (42.19) 32 (50)
Pathological N stage N0 N1 N2–3
10 (32.26) 2 (15.38) 2 (10)
10 (32.26) 4 (30.77) 7 (35)
4 (12.90) 5 (38.46) 6 (30)
7 (22.58) 2 (15.38) 5 (25)
31 (48.44) 13 (20.31) 20 (31.25)
Pericapsular invasion No Yes
13 (27.66) 1 (5.88)
16 (34.04) 5 (29.41)
7 (14.89) 8 (47.06)
11 (23.40) 3 (17.65)
47 (73.44) 17 (26.56)
(40.91) (31.03) (25) (22.22)
1 9 2 3
(4.54) (31.03) (50) (33.33)
6 6 1 1
(27.27) (20.69) (25) (11.11)
22 29 4 9
(34.38) (43.31) (6.25) (14.06)
Cumulative Metastatic Disease Free Survival Rate
100
80
60
Budding not marked budding marked budding
40
20
0 0
30
60
90
120
150
180
Months Fig. 2. Distant metastasis-free survival.
adhesions [4]. Initially, the invasive margin of the tumor had attracted more attention, and budding was evaluated at the invasive margin [19–21]. However, Morodomi et al. [10] have noted that budding can be identified in biopsy specimens and is closely related to nodal metastasis. This suggests that budding is
not only a feature of the invasive tumor margin, but also a common feature of both superficial and deep portions of tumors with aggressive behavior, along with well-known carcinogenesis steps, such as destruction of the basement membrane or loss of intercellular adhesions [3]. Guzin´ska-Ustymowicz [5] has
ARTICLE IN PRESS S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
91
Table 2 The results of the multivariate analysis concerning prognostic factors. Prognostic factors
Clinical stage: II III IV Pathological stage: II III IV Number of pathologic lymph nodes: N0 r3 43 Peri-nodal invasion: ( ) (+) Grade: I II III Budding: Marked Non-marked or negative
Univariate analysis for
Multivariate analysis for
OS
LRDFS
DMFS
OS
LRDFS
DMFS
p = 0.14
p = 0.59
p= 0.49
p= 0.5
p = 0.49
p = 0.49
p = 0.61
p = 0.6
p= 0.69
p= 0.6
p = 0.36
p = 0.53
p = 0.07
p = 0.56
p= 0.07
p = 0.018
p = 0.61
p = 0.021
p = 0.07
p = 0.93
p= 0.29
p= 0.8
p = 0.4
p = 0.3
p = 0.43
p = 0.44
p= 0.91
p= 0.22
p = 0.84
p = 0.67
p = 0.71
p = 0.71
p= 0.039
p= 0.9
p = 0.64
p = 0.044
OS= overall survival; LRDFS=loco-regional disease free survival; DMFS = distant metastasis free survival.
reported increased aggressive behavior of colorectal carcinoma in the presence of MMP-9 and cathepsin B expression at the site of budding. This might be the reason for laryngeal carcinomas, since the expression of MMP increases with progression of carcinoma [5,15,22]. Budding can be identified by observing H&E stained tumor sections using light microscopy. Different methods of quantification have been applied in different series. We preferred using the criteria described by Kanazawa et al. [7], considering the extent of budding at the invasive tumor margin in H&E stained sections. Ueno et al. [19–21] have evaluated the region with most budding, and the budding was scored as I, II, III, and IV, according to the number of budding units as 0–4, 5–9, 10–19, and 420, respectively. The same method was applied in different series [8,16]. Roh et al. [14] have divided the cases into 2 groups (cases with o5 or 45 budding under a 20 objective at the tumor region with the most extensive budding). Tadahiko et al. [17] have counted the budding units at the invasive margin of the whole tumor in one section, and this was probably possible as they only evaluated a series of T1 colorectal carcinomas, i.e., a group of relatively small tumors. Some authors have evaluated tumor budding in sections stained by immunohistochemical methods. Strong expression of the laminin-5 gamma 2-chain in the budding cells has been used for identifying tumor budding in anal carcinomas [11]. The method for assessing tumor budding has not yet been standardized, albeit the method we preferred in this series of laryngeal carcinomas is of prognostic significance for distant metastases. Oosterkamp et al. [12] have reported that pericapsular invasion of metastatic lymph nodes is the most important risk factor for distant metastasis, followed by lymph node metastasis and T4 tumor stage, in a series of 165 patients with laryngeal carcinoma. In our series, we also identified the number of metastatic lymph
nodes as an independent factor for determining distant metastasis, along with marked tumor budding. Tumor-related histopathologic features of prognostic or predictive value include tumor type, pathologic T stage, grade, lymphovascular invasion, perineural invasion, and infiltrative growth pattern for laryngeal carcinomas [1,13,18]. The results in this series suggest that budding might be a valuable prognostic factor, particularly for distant metastasis in laryngeal carcinomas.
References [1] F.C. Akman, N. Dag, O.U. Ataman, C. Ecevit, A.O. Ikiz, I. Arslan, S. Sarıoglu, E. Ada, M. Kinay, The impact of treatment center on the outcome of patients with laryngeal cancer treated with surgery and radiotherapy, Eur. Arch. Otorhinolaryngol. 265 (2008) 1245–1255. [2] H.J. Choi, K.J. Park, J.S. Shin, M.S. Roh, H.C. Kwon, H.S. Lee, Tumor budding as a prognostic marker in stage-III rectal carcinoma, Int. J. Colorectal Dis. 22 (2007) 863–868. [3] H. Gabbert, Mechanisms of tumor invasion: evidence from in vivo observations, Cancer Metastasis Rev. 4 (1985) 293–309. [4] M.J. Gosens, L.C. van Kempen, C.J. van de Velde, J.H. van Krieken, I.D. Nagtegaal, Loss of membranous Ep-CAM in budding colorectal carcinoma cells, Mod. Pathol. 20 (2007) 221–232. [5] K. Guzin´ska-Ustymowicz, MMP-9 and cathepsin B expression in tumor budding as an indicator of a more aggressive phenotype of colorectal cancer (CRC), Anticancer Res. 26 (2006) 1589–1594. [6] K. Hase, C. Shatney, D. Johnson, M. Trollope, M. Vierra, Prognostic value of tumor ‘‘budding’’ in patients with colorectal cancer, Dis. Colon Rectum 36 (1993) 627–635. [7] H. Kanazawa, H. Mitomi, Y. Nishiyama, I. Kishimoto, N. Fukui, Watanabe. Nakamura, Tumour budding at invasive margins and outcome in colorectal cancer, Colorectal Dis. 10 (2007) 41–47. [8] M. Koike, Y. Kodera, Y. Itoh, G. Nakayama, M. Fujiwara, N. Hamajima, A. Nakao, Multivariate analysis of the pathologic features of esophageal squamous cell cancer: tumor budding is a significant independent prognostic factor, Ann. Surg. Oncol. 15 (2008) 1977–1982. [9] Y. Moriya, T. Niki, T. Yamada, Y. Matsuno, H. Kondo, S. Hirohashi, Increased expression of laminin-5 and its prognostic significance in lung adenocarcinomas of small size, Cancer 91 (2001) 1129–1141.
ARTICLE IN PRESS 92
S. Sarioglu et al. / Pathology – Research and Practice 206 (2010) 88–92
[10] T. Morodomi, H. Isomoto, K. Shirouzu, K. Kakegawa, K. Irie, M. Morimatsu, An index for estimating the probability of lymph node metastasis in rectal cancers. Lymph node metastasis and the histopathology of actively invasive regions of cancer, Cancer 63 (1989) 539–543. [11] P.J. Nilsson, C. Rubio, C. Lenander, G. Auer, B. Glimelius, Tumour budding detected by laminin-5 {gamma} 2-chain immunohistochemistry is of prognostic value in epidermoid anal cancer, Ann. Oncol. 16 (2005) 893–898. [12] S. Oosterkamp, J.M. de Jong, P.L. Van den Ende, J.J. Manni, C. Dehing-Oberije, B. Kremer, Predictive value of lymph node metastases and extracapsular extension for the risk of distant metastases in laryngeal carcinoma, Laryngoscope 116 (2006) 2067–2070. [13] A. Ozdek, S. Sarac, M.U. Akyol, O.F. Unal, A. Sungur, Histopathological predictors of occult lymph node metastases in supraglottic squamous cell carcinomas, Eur. Arch. Otorhinolaryngol. 257 (2000) 389–392. [14] M.S. Roh, J.I. Lee, P.J. Choi, Tumor budding as a useful prognostic marker in esophageal squamous cell carcinoma, Dis. Esophagus. 17 (2004) 333–337. [15] S. Sarioglu, E. Ozer, F. Kirimca, B. Sis, U. Pabuc- cuoglu, Matrix metalloprotei˘ nase-2 expression in laryngeal preneoplastic and neoplastic lesions, Pathol. Res. Pract. 197 (2001) 483–486. [16] D.K. Sohn, H.J. Chang, J.W. Park, D.H. Choi, K.S. Han, C.W. Hong, K.H. Jung, D.Y. Kim, S.B. Lim, H.S. Choi, S.Y. Jeong, Histopathological risk factors for lymph
[17]
[18]
[19]
[20]
[21]
[22]
node metastasis in submucosal invasive colorectal carcinoma of pedunculated or semipedunculated type, J. Clin. Pathol. 60 (2007) 912–915. M. Tadahiko, H. Matsuoka, M. Sugiyama, N. Abe, A. Sakamoto, Y. Atomi, Actual number of tumor budding as a new tool for the individualization of treatment of T1 colorectal carcinomas, J. Gastroenterol. Hepatol. 21 (2006) 1115–1121. L.D. Thompson, B.M. Wenig, D.K. Heffner, D.R. Gnepp, Exophytic and papillary squamous cell carcinomas of the larynx: a clinicopathologic series of 104 cases, Otolaryngol. Head Neck Surg. 120 (1999) 718–724. H. Ueno, A.M. Jones, K.H. Wilkinson, J.R. Jass, I.C. Talbot, Histological categorisation of fibrotic cancer stroma in advanced rectal cancer, Gut 53 (2004) 581–586. H. Ueno, J. Murphy, J.R. Jass, H. Mochizuki, I.C. Talbot, Tumour ‘budding’ as an index to estimate the potential of aggressiveness in rectal cancer, Histopathology 40 (2002) 127–132. H. Ueno, H. Mochizuki, E. Shinto, Y. Hashiguchi, K. Hase, I.C. Talbot, Histological indices in biopsy specimens for estimating the probability of extended local spread in rectal cancer, Cancer 94 (2002) 2882–2891. M. Xie, Y. Sun, Y. Li, Expression of matrix metalloproteinases in supraglottic carcinoma and its clinical implication for estimating lymph node metastases, Laryngoscope 114 (2004) 2243–2248.