Tumour budding detected by laminin-5 {gamma}2-chain immunohistochemistry is of prognostic value in epidermoid anal cancer

P. J. Nilsson1,*, C. Rubio2, C. Lenander1, G. Auer2 and B. Glimelius2,3

1 Centre of Gastrointestinal Disease, Ersta Hospital, Stockholm; 2 Department of Oncology and Pathology, Karolinska University Hospital, SE 171 76 Stockholm; 3 Department of Oncology, Radiology and Clinical Immunology, Akademiska Sjukhuset, SE 751 85 Uppsala, Sweden

*Correspondence to: Dr P. J. Nilsson, Centre of Gastrointestinal Disease, Ersta Hospital, PO Box 4622, SE 116 91, Stockholm, Sweden. Tel: +46-8-714-61-00; Fax: +46-8-714-66-80; Email: per.nilsson{at}ersta.se


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Background:: Markers for guidance with regard to individual prognosis and treatment planning are sought in epidermoid anal cancer. This study assessed the prognostic and predictive value of tumour budding.

Patients and methods:: From a population-based consecutive series of patients who were prospectively recorded, it was possible to investigate 209 (76%) of the pretreatment biopsies. Immunohistochemistry with a monoclonal antibody for the {gamma}2 chain of laminin-5 was used to detect tumour budding (defined as dissociated single cancer cells or clusters of up to five cells).

Results:: Tumour budding was detected in 104 (50%) of the 209 samples. No significant correlation was found between tumour budding and clinicopathological characteristics. Patients with tumour budding had a statistically significantly better 5-year overall survival rate compared with patients lacking tumour budding (74% versus 64%, P <0.05). Albeit not statistically significant, other outcome variables such as tumour-specific survival, recurrence after initial complete response and rate of distant metastases, were all in favour of patients with tumour budding. Multivariate analysis reveals tumour budding as an independent positive prognostic factor.

Conclusions:: Tumour budding detected by laminin-5 immunohistochemistry may be of prognostic value in the treatment of epidermoid anal cancer. However, further studies are needed to clarify the possible clinical implications.

Key words: anus neoplasm, radiotherapy, prognosis, tumour budding, laminin


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Primary treatment of epidermoid anal cancer is radiotherapy. Chemotherapy is generally added in locally advanced cases but whether this should also be the case in smaller tumours is still unclear [1Go]. Radical surgery is reserved for poor responders to, or failures after, (chemo)radiation [2Go]. Although the proportion of patients undergoing surgery is rather small, the morbidity associated with surgery on these heavily pretreated patients is considerable [3Go]. A more individualised therapy aiming at cure with least possible morbidity could be offered to patients, were there reliable prognostic and predictive markers. For instance, poor responders to radiotherapy could be advised to undergo surgery at an earlier stage, thereby reducing the risk of post-operative complications induced by curative radiotherapy.

The presence of single cancer cells or clusters of up to five cancer cells just ahead of the invasive front of an invasive carcinoma has been described as tumour budding [4Go]. In colorectal adenocarcinomas, tumour budding is associated with poor prognosis [5Go–7Go]. Studies of the prognostic importance of tumour budding in squamous cell carcinomas in general, and in epidermoid anal cancer in particular, are infrequent.

Tumour budding is possible to detect in routine staining [6Go]. However, there are several reports describing immunostaining of laminin-5 (Ln-5) in tumour buds [8Go–11Go], which greatly facilitates the detection of budding cancer cells. The presence of Ln-5 immunostaining in tumour buds has also been shown in squamous cell carcinoma [12Go–14Go]. In non-neoplastic tissue, laminins are a group of extracellular matrix proteins mainly localised in the basement membrane of various epithelial tissues. The isoform laminin-5, and in particular the {gamma}2 chain, is preferentially expressed in the cytoplasm of cancer cells along the advancing edge of a tumour and has been associated with cancer growth and invasion [8Go, 9Go, 12Go]. Ln-5 {gamma}2 chain has been associated with the formation of tumour buds [15Go] and has also been proposed as an aid in identifying micrometastasis in tissues surrounding carcinoma [11Go].

The aims of this present study were to evaluate the prognostic and predictive potential of tumour budding detected by means of Ln-5 {gamma}2-chain immunohistochemistry in a large, population-based, consecutive series of epidermoid anal carcinomas.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients and treatment
All patients with epidermoid anal carcinomas in the Stockholm Health Care Region (population 1.9 million) have been prospectively recorded since the mid-1980s. Between the years 1985 and 2000, 308 patients with invasive epidermoid anal cancer were diagnosed. Treatment with curative intent was given to 276 (90%) of the 308 patients. Clinical data and treatment results from this cohort of patients have been reported [16Go]. Although treatment has been given according to defined guidelines, different approaches are present in the series. However, among patients with locally advanced tumours, i.e. tumours with a maximum diameter of at least 4 cm or node positive tumours, two well-defined treatment groups can be distinguished. From 1985 until 1992, external beam radiotherapy with or without concomitant bleomycin was used for these patients. Introduced in 1989, and with a gradually increased implementation, neoadjuvant platinum-based chemotherapy in three courses, followed by external beam irradiation was used. Surgery in the form of an abdominoperineal resection has been used for poor responders to the first course of radiotherapy of 40–46 Gy, and as salvage therapy for residual or recurrent tumours after full dose radiotherapy (60–64 Gy). Pretreatment data on patients are given in Table 1. The study was approved by the Regional Ethics Committee.


View this table:
[in this window]
[in a new window]
 
Table 1. Pretreatment data on all patients treated with curative intent between 1985–2000 and on the investigated cohort

 
Tissue samples
From the cohort of 276 patients treated with curative intent, it was possible to retrieve 230 (83%) formalin-fixed, paraffin wax-embedded blocks containing the original pretreatment diagnostic biopsy. Twenty-one blocks contained material of inadequate quantity or quality and could not be analysed; hence biopsies from 209 (76%) patients were investigated.

Immunohistochemistry
The monoclonal antibody 6C12 (mAb 6C12) was produced by hybridoma techniques. In brief, GST-laminin {gamma}2-III fusion protein containing amino acid residues 391–567 of human laminin {gamma}2 chain was used as an antigen. Balb/c mice were immunised four times after which mouse spleen cells were fused with mouse myeloma cells (P3X63Ag8.653). The production of the anti-laminin {gamma}2 antibody by the hybridoma clones was screened by the aid of immunohistology, ELISA and western blotting. The best clone was picked up and cloned again (single-cell cloning) to ensure that the produced hybridoma cell line was monoclonal. For immunohistochemistry, IgG was purified from the hybridoma cell culture media using Protein G Sepharose column (Amersham Pharmacia Biotec). The antibody specificity was checked using purified full-length human {gamma}1 and {gamma}2 chains that were electrophoresed by SDS-PAGE. The electrophoresed proteins were stained with Coomassie blue to confirm the presence of {gamma}1- and {gamma}2-chain proteins. The protein preparations were transferred to nylon membranes and immunoreacted with the purified mAb 6C12 generated against the recombinant GST-laminin {gamma}2-III fusion protein and specific reaction only with the {gamma}2 chain could be observed.

Representative 4 µm paraffin wax-embedded sections were cut and subjected to the standard horseradish peroxidase avidin-biotin complex technique (VECTASTAIN® Elite ABC kit, cat PK-6100; Vector Laboratories Inc., Burlingame, CA). The sections were first deparaffinised in xylene, rehydrated in an ethanol series and rinsed in water. The sections were treated enzymatically with Pronase 0.05% (Sigma P5147) for 15 min at 20–22 °C and rinsed. Following the addition of 0.5% H2O2 to the slides, they were incubated at room temperature for 20 min and then rinsed in water. To prevent unspecific staining, 40 µl 1% bovine serum albumin (BSA) in 1x phosphate buffered saline (PBS) was added to each slide, which were then incubated for 30 min at room temperature in a humid chamber. The mAb 6C12 (2.7 mg/ml) was diluted 1:100 in 1% BSA/1 x PBS, and 40 µl was added to each slide. After incubation overnight at 4 °C, biotinylated anti-mouse IgG (Vector), diluted 1:200, was added to the sections for 30 min, followed by incubation in the avidin–biotin–peroxidase complex for 30 min. Tris buffered saline (pH 7.6) was used for washing in between steps. Diaminobenzidine tetrahydrochloride was used to develop the peroxidase reaction. The slides were counterstained in Mayer's hematoxylin for 4–6 min, dehydrated, cleaned in xylene and mounted in a xylene-soluble mounting medium.

Evaluation
All slides were evaluated by a senior pathologist (CR) who was blinded to the clinical data and therapeutic outcome. Tumour budding was defined as dissociated single cancer cells or clusters of up to five cancer cells with cytoplasmic Ln-5 immunostaining, ahead of the invasive front. Fifty randomly selected slides were re-evaluated by another investigator (PJN) with a 96% concurrence regarding the presence of tumour budding.

Statistical methods
To assess the correlation between proportions, the chi-square test and the Mann–Whitney U-test were used where appropriate. Survival curves were plotted using the method of Kaplan and Meier. Comparison between survivals in the different groups was by the log rank test. Cox proportional hazards model was used to perform uni- and multivariate analyses on parameters related to survival. T-stage, N-stage, histology and tumour budding were included in the Cox analyses. P values <0.05 were considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In samples containing normal anorectal glandular mucosa, the immunostaining was restricted to the basement membrane (Figure 1A). In tumour tissue, the immunoreactivity was found in the cytoplasm of the tumour cells, particularly along the invasive front (Figure 1B). In 170 (81%) of the 209 examined samples cytoplasmic staining was detected in >1% cancer cells. Tumour budding, i.e. the presence of dissociated clusters of up to five cancer cells with cytoplasmic Ln-5 positivity, was present in 104 (50%) of the 209 samples (Figure 1C and 1D), whereas no tumour budding could be detected in the remaining 105 slides. No significant correlation was found between the presence of tumour budding and the clinicopathological characteristics listed in Table 2. In addition, Table 2 summarizes the treatments delivered to the investigated cohort of 209 patients. Although neoadjuvant chemoradiation was delivered to a slightly larger proportion of patients in the group lacking tumour budding, no statistically significant differences were seen between patients with or without tumour budding regarding the therapeutic approach.



View larger version (123K):
[in this window]
[in a new window]
 
Figure 1. (A) Transitional mucosa showing continuous laminin-5 {gamma}2-chain expression along the basement membrane. (B) Detail of invasive epidermoid anal carcinoma. Note laminin-5 {gamma}2-chain expression in the cytoplasm of cells along the invasive front. (C) Low-power view of an invasive squamous carcinoma of the anus. Note intense immunostaining in tumour buds originating at the invasive front. (D) Low-power view of an invasive epidermoid anal carcinoma. Intense laminin-5 {gamma}2-chain expression in dissociated tumour buds ahead the invasive front.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Clinicopathological characteristics and therapeutic approaches among patients with and without tumor budding, respectively (n=209)

 
Patients having tumour budding at the invasive front had a statistically significantly better 5-year overall survival rate compared with patients with no tumour buds (74% versus 64%, P <0.05 log rank) (Figure 2). Albeit not statistically significant, there was also a difference in tumour-specific survival with a 5-year survival rate of 78% in the group with tumour budding compared with 68% where no tumour budding was observed (P=0.11). The complete response rate (CR), including surgery when performed, was similar in the two groups (94% versus 91%, P=0.43). There was a difference in recurrence rates after an initial CR between the groups with and without tumour budding but it did not reach statistical significance (14% versus 24%, P=0.09). Also, the numerical differences concerning isolated local failures, either as residual tumour after treatment or as recurrence (14% versus 21%, P=0.22) and the rate of distant metastases (5% versus 10%, P=0.19) were not statistically significant.



View larger version (14K):
[in this window]
[in a new window]
 
Figure 2. Overall survival curves (Kaplan–Meier) of patients with epidermoid anal cancer according to presence or absence of tumour budding (n=209).

 
In the Cox proportional hazards multivariate analysis on survival, T-stage 3–4 and absence of tumour budding were both independent variables of poor prognosis (Table 3). A total of 115 patients constituted a subset of patients with locally advanced tumours having a diameter of at least 4 cm and/or positive lymph nodes treated in accordance with guidelines. Twenty-two patients (19%) were poor responders to the (chemo)radiotherapy and underwent surgery after the first course of irradiation. This proportion did not differ between tumours showing budding or not. Among the 93 patients who completed the non-surgical therapy including two courses of irradiation, the rate of CR was 89% in patients treated with neoadjuvant chemoradiotherapy compared with 65% among patients treated with radiotherapy with or without bleomycin (P <0.01). There was no difference in the proportion of patients who responded with CR to the (chemo)radiotherapy, whether tumour budding was present or not (81% versus 80%) (data not illustrated).


View this table:
[in this window]
[in a new window]
 
Table 3. Cox proportional hazards analyses on overall survival (n=209)

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In the present study immunohistochemistry with a monoclonal antibody for the {gamma}2 chain of Ln-5 was used to detect tumour budding in 209 cases of epidermoid anal carcinoma. Among patients where tumour budding was present, the overall 5-year survival rate was significantly better, compared to those in whom no tumour budding was detected. Although not statistically significant, other outcome variables such as tumour-specific survival, failure rate after initial CR, isolated local failure rate and rate of distant metastases, were all in favour of patients with tumour budding. In the multivariate analysis, the presence of tumour budding was an independent positive prognostic factor. These results suggest that tumour budding detected by Ln-5 immunohistochemistry may be of prognostic value in epidermoid anal cancer.

In contrast to these results, tumour budding is generally regarded as an indicator of poor prognosis [5Go–7Go]. However, previous investigations on budding have, for the most part, concerned adenocarcinomas and not squamous cell cancers. The treatment of colorectal adenocarcinomas is generally surgical, whereas our cohort of patients with epidermoid anal cancers has been treated mainly with radiotherapy. This difference in therapeutic approach may have an influence on the prognostic impact of tumour budding.

Tumour budding is a representation of the dissociation of the invasive front of a tumour and is related to the invasive and metastatic capability of a tumour [5Go]. Epidermoid anal carcinomas generally have a low capability of distant metastasis, although locally aggressive and with frequent nodal spread [17Go]. In the investigated cohort of patients, 50% showed tumour budding but the rate of distant metastasis was only 7%. In addition, no correlation between lymph node engagement and presence of tumour budding was found in the present study. It is conceivable that tumour budding also represents other intrinsic tumour characteristics that are related to long-term outcome, i.e. survival, but not the immediate response to therapy.

Earlier reports on Ln-5 expression in various squamous cell carcinomas have generally described Ln-5 positivity as an indicator of poor prognosis [13Go, 14Go, 18Go]. However, in those studies, immunoreactivity for the {gamma}2 chain of Ln-5 per se, and not the presence or absence of tumour budding was evaluated. The {gamma}2 chain of Ln-5 has been associated with the formation of tumour buds [15Go]. Giannelli et al. [11Go] suggested that Ln-5 can be used to detect micrometastasis in peritumoural tissue and in the present study, immunohistochemistry was used for the detection of tumour budding. Thus, the results from previous investigations on Ln-5 in different squamous cell carcinomas are not directly comparable to those obtained in the present study. However, the full meaning of Ln-5 expression in various tumour types remains to be elucidated.

In the treatment of epidermoid anal cancer the aims are cure and, whenever possible, sphincter preservation. When non-surgical therapy fails and sphincter preservation is impossible, an abdominoperineal resection is generally undertaken. In patients where this is the case, the previous radiation therapy renders patients susceptible to post-operative complications. Predictors of radiation therapy response could facilitate planning of a therapeutic approach [19Go] and curative radiotherapy at a high dose, which has a risk of increasing surgical complications, could be avoided if the chances of tumour control are low. When exploring the possibilities of evaluating the predictive potential of tumour budding, either in the entire material or in a subset of patients with locally advanced tumours, no correlation could be obtained. Bearing in mind the relatively small number of patients investigated with respect to this, the present study does not suggest a predictive potential of tumour budding regarding tumour response to (chemo)radiotherapy.

Previously, different prognostic markers in epidermoid anal cancer have been investigated. For example, Holm et al. [20Go] found shorter overall survival for p21-negative tumours in a series of 94 epidermoid anal cancer patients and Wong et al. [21Go] reported an independent prognostic value for p53 in 49 patients treated with radiotherapy, 5-fluorouracil and mitomycin C. However, in a recent review of earlier studies on prognostic factors in epidermoid anal cancer, it was concluded that results so far were unable to provide reliable guidance in treatment decisions [22Go]. In the present study, the possible role of tumour budding as a prognostic factor has been investigated in a comparatively large number of patients with epidermoid anal cancer and it can be concluded that tumour budding detected by Ln-5 immunohistochemistry can be of prognostic value. However, further studies are needed to clarify the possible clinical implications.


    Acknowledgements
 
An abstract containing part of the information here was presented at the 3rd International Meeting on Cancer Molecular Markers, organized by EORTC/NCI, in Brussels, Belgium, 18–20 April 2004.

This work was supported by the Public Health and Medical Service Committee of Stockholm County Council, the Swedish Cancer Society and the Cancer Society in Stockholm. The authors are grateful to Inga Maurin and Ulla Aspenblad for excellent technical assistance.

Received for publication September 28, 2004. Accepted for publication January 7, 2005.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
1. Clark MA, Hartley A, Geh JI. Cancer of the anal canal. Lancet Oncol 2004; 5: 149–157.[CrossRef][ISI][Medline]

2. Ryan DP, Compton CC, Mayer RJ. Cancer of the anal canal. N Engl J Med 2000; 342: 792–800.[Free Full Text]

3. Nilsson PJ, Svensson C, Goldman S, Glimelius B. Salvage abdominoperineal resection in anal epidermoid cancer. Br J Surg 2002; 89: 1425–1429.[CrossRef][ISI][Medline]

4. Hase K, Shatney C, Johnson D, Trollope M, Vierra M. Prognostic value of tumor ‘budding’ in patients with colorectal cancer. Dis Colon Rectum 1993; 36: 627–635.[ISI][Medline]

5. Ueno H, Murphy J, Jass JR, Mochizuki H, Talbot IC. Tumor ‘budding’ as an index to estimate the potential of aggressiveness in rectal cancer. Histopathology 2002; 40: 127–132.[CrossRef][ISI][Medline]

6. Jass JR, Barker M, Fraser L et al. APC mutation and tumour budding in colorectal cancer. J Clin Pathol 2003; 56: 69–73.[Abstract/Free Full Text]

7. Okuyama T, Oya M, Ishikawa H. Budding as a useful prognostic marker in pT3 well- or moderately-differentiated rectal adenocarcinoma. J Surg Oncol 2003; 83: 42–47.[CrossRef][ISI][Medline]

8. Pyke C, Salo S, Ralfkiær E, Rømer J, Danø K, Tryggvason K. Laminin-5 is a marker of invading cancer cells in some human carcinomas and is coexpressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas. Cancer Res 1995; 55: 4132–4139.[Abstract]

9. Sordat I, Bosman F, Dorta G et al. Differential expression of laminin-5 subunits and integrin receptors in human colorectal neoplasia. J Pathol 1998; 185: 44–52.[CrossRef][ISI][Medline]

10. Lenander C, Habermann JK, Öst Å et al. Laminin-5 {gamma}2 chain expression correlates with unfavorable prognosis in colon carcinomas. Anal Cell Pathol 2001; 22: 201–209.[ISI][Medline]

11. Giannelli G, Fransvea E, Bergamini C, Marinosci F, Antonaci S. Laminin-5 chains are expressed differentially in metastatic and nonmetastatic hepatocellular carcinoma. Clin Cancer Res 2003; 9: 3684–3691.[Abstract/Free Full Text]

12. Kosmehl H, Berndt A, Strassburger S et al. Distribution of laminin and fibronectin isoforms in oral mucosa and oral squamous cell carcinoma. Br J Cancer 1999; 81: 1071–1079.[CrossRef][ISI][Medline]

13. Hellman K, Hellström A-C, Silfverswärd C et al. Cancer of the vagina: Laminin-5{gamma}2 chain expression and prognosis. Int J Gynecol Cancer 2000; 10: 391–396.[CrossRef][ISI][Medline]

14. Yamamoto H, Itoh F, Iku S, Hosokawa M, Imai K. Expression of the {gamma}2 chain of laminin-5 in the invasive front is associated with recurrence and poor prognosis in human esophageal squamous cell carcinoma. Clin Cancer Res 2001; 7: 896–900.[Abstract/Free Full Text]

15. Masaki T, Matsuoka H, Sugiyama M et al. Laminin-5 gamma 2 chain and matrix metalloproteinase-2 may trigger colorectal carcinoma invasiveness through formation of budding tumor cells. Anticancer Res 2003; 23: 4113–4119.[ISI][Medline]

16. Nilsson PJ, Svensson C, Goldman S, Ljungqvist O, Glimelius B. Epidermoid anal cancer: A review of a population-based series of 308 consecutive patients treated according to prospective protocols. Int J Radiat Oncol Biol Phys 2005; 61: 92–102.[CrossRef][ISI][Medline]

17. Myerson RJ, Karnell LH, Menck HR. The national cancer database report on carcinoma of the anus. Cancer 1997; 80: 805–815.[CrossRef][ISI][Medline]

18. Ono Y, Nakanishi Y, Ino Y et al. Clinicopathologic significance of laminin-5 {gamma}2 chain expression in squamous cell carcinoma of the tongue. Cancer 1999; 85: 2315–2321.[CrossRef][ISI][Medline]

19. Swedish Cancer Society Investigation Group. Turesson I, Carlsson J, Brahme A, Glimelius B, Zackrisson B, Stenerlow B. Biological response to radiation therapy. Acta Oncol 2003; 42: 92–106.[CrossRef][ISI][Medline]

20. Holm R, Skovlund E, Skomedal H, Flørenes VA, Tanum G. Reduced expression of p21WAF1 is an indicator of malignant behaviour in anal carcinomas. Histopathology 2001; 39: 43–49.[CrossRef][ISI][Medline]

21. Wong CS, Tsao MS, Sharma V, Chapman WB, Pintilie M, Cummings BJ. Prognostic role of p53 protein expression in epidermoid carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 1999; 45: 309–314.[ISI][Medline]

22. Fenger C. Prognostic factors in anal carcinoma. Pathology 2002; 34: 573–578.[CrossRef][ISI][Medline]





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
16/6/893    most recent
mdi179v1
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Nilsson, P. J.
Articles by Glimelius, B.
PubMed
PubMed Citation
Articles by Nilsson, P. J.
Articles by Glimelius, B.