Journal of Surgical Oncology 2004;86:152–156
Margins in Extra-Abdominal Desmoid Tumors: A Comparative Analysis ANDREAS LEITHNER, MD,1 MARKUS GAPP,1 KATHARINA LEITHNER, MD,2 ROMAN RADL, MD,1 PETER KRIPPL, MD,3 ALFRED BEHAM, MD,4 AND REINHARD WINDHAGER, MD1* 1 Department of Orthopedic Surgery, Medical University Graz, Graz, Austria 2 Department of Vascular Biology and Thrombosis Research, Medical University Vienna, Vienna, Austria 3 Department of Medicine, Division of Oncology, Medical University Graz, Graz, Austria 4 Institute of Pathology, Medical University Graz, Graz, Austria
Background and Objectives: The main treatment of extra-abdominal desmoid tumors remains surgery, but recurrence rates up to 80% are reported. The impact of microscopic surgical margin status according to the Enneking classification system is discussed controversially. Methods: Therefore, the authors screened the published literature for reliable data on the importance of a wide or radical excision of extra-abdominal desmoid tumors. All studies with more than ten patients, a surgical treatment only, and margin status stated were included. Results: Only 12 out of 49 identified studies fulfilled the inclusion criteria. One hundred fifty-two primary tumors were excised with wide or radical microscopic surgical margins, while in 260 cases a marginal or intralesional excision was performed. In the first group 41 patients (27%) and in the second one 187 patients (72%) developed a recurrence. Therefore, microscopic surgical margin status according to the Enneking classification system is a significant prognostic factor (P < 0.001). Conclusions: The data of this review underline the strategy of a wide or radical local excision as the treatment of choice. Furthermore, as a large number of studies had to be excluded from this analysis, exact microscopic surgical margin status should be provided in future studies in order to allow comparability. J. Surg. Oncol. 2004;86:152–156.
ß 2004 Wiley-Liss, Inc.
KEY WORDS: aggressive fibromatosis; desmoid; surgery; recurrence
INTRODUCTION Extra-abdominal desmoid tumors are often frustrating to treat. Despite their benign nature due to the inability to metastasize, their infiltrative growth and tendency towards local recurrence constitute frequent problems [1]. Furthermore, the etiology has remained unclear [2,3], although a genetic defect in the growth regulation of connective tissue, trauma, and pregnancy have been suspected as causative and growth predicting factors [4]. With an incidence being low with 2–4/million/year [5], until now no randomized trials have been published on the therapy of this tumor. Mainly case reports or small series provided contradictory results on adjuvant or neoadjuvant therapies. Aggravating the situation, authors throughout the literature presented data mixing anatomical groups, although it is known that these groups— ß 2004 Wiley-Liss, Inc.
abdominal, extra-abdominal, and intra-abdominal—have their specific features, including a different growth pattern, age predilection, and recurrence rates [4,6]. For soft tissue tumors various macroscopic and microscopic definitions of surgical margins exist. Enneking established a nomenclature that defines four types of surgical margins based on the relation of the surgical margin to the neoplasm and its reactive zone—intralesional, This work was conducted by the Austrian Cancer Aid/Styria as scientific research work; project no. 05/2001. *Correspondence to: Reinhard Windhager, MD, Department of Orthopedic Surgery, Medical University Graz, Auenbruggerplatz 5, A-8036 Graz, Austria. Fax: þ43 316 385 2957. E-mail:
[email protected] Accepted 13 April 2004 DOI 10.1002/jso.20057 Published online in Wiley InterScience (www.interscience.wiley.com).
Margins in Extra-Abdominal Desmoid Tumors
153
TABLE I. Surgical Margins According to Enneking [7,8] Type
Plane of dissection
Intralesional Marginal Wide Radical
Result
Within lesion Within reactive zone, extracapsular Intracompartimental en bloc with cuff of normal tissue Extracompartimental en bloc with entire compartment
marginal, wide, and radical [7,8] (Table I). As for extraabdominal desmoid tumors contradictory recommendations concerning the adequate surgical procedure exist, the authors screened the published literature for reliable data. MATERIALS AND METHODS Study Identification
A computerized Medline search was conducted for the period from 1969 to 2003 with the use of the following terms and Boolean operators: ‘‘desmoid’’ OR ‘‘aggressive fibromatos*.’’ All studies with more than 10 patients with desmoid tumors were read by two of us. The bibliography of each study was reviewed by two of us for additional relevant studies. All studies with more than 10 patients with extra-abdominal desmoid tumors treated with surgery only and microscopic surgical margin status stated were included in the analysis.
Tumor at margin Tumor satellites May leave skip lesions No residual
(4) extra-abdominal tumor, (5) no other treatment than surgery. If one of the five was missing, the data were classified as not acceptable for this analysis. Microscopic surgical margins had to be classified according to the Enneking classification system or described in comparable terms. For example, Markhede’s ‘‘inadequate’’ and ‘‘probably adequate’’ excisions were classified as marginal/intralesional for this analysis [9]. In one case the first author of a published study was contacted for further data in order to be able to exclude two abdominal cases from a larger series of extremity and trunk tumors [10]. According to Enneking an amputation without providing microscopic surgical margin status was not defined per se as radical [8]. Therefore, these unclear cases were excluded from the analysis. Data on treatment of patients with a first recurrence were analyzed separately. To underline the importance of the study of the margins by the surgeon and the pathologist, margin status is referred as microscopic surgical margin status in the current study.
Data Extraction
For each of the eligible studies, the relevant data were abstracted by one of us and were rechecked for accuracy by two others. Specifically, the following data had to be stated in the article: (1) primary or recurrent tumor, (2) microscopic surgical margin status, (3) recurrence rate,
Data Analysis
Prior to analyzing the data, we developed hypotheses regarding the association between the surgical margin and the recurrence rate. Specifically, we hypothesized that (1) a marginal or intralesional resection of a primary
TABLE II. Articles With Extra-Abdominal Desmoid Tumors Included Patients
Primary tumors
Recurrent tumors
Author
Year
Total
Included
R0a
Rec.
%
R1/2b
Rec.
%
Dalen [12] Enzinger [11] Hunt [13] Karakousis [10] Kofoed [14] Markhede [9] Mehrotra [15] Miralbell [16] Pignatti [17] Pritchard [18] Rao [20] Rock [19] Total included Total excluded
2003 1967 1960 1993 1985 1986 2000 1990 2000 1996 1987 1984
30 30 29 9 15 53 36 16 83 50 20 193
22 29 26 3 15 39 17 16 29 34 10 172 412 152
8 6 18 3 3 21 8
0 3 3 1 0 3 1
0 50 17 33 0 14 13
14 23 8
10 14 4
71 61 50
18 13 6 48 152
5 2 0 23 41
28 15 0 48 27
12 18 9 16 11 21 4 124 260
9 13 5 4 5 10 1 112 187
75 72 56 25 45 48 25 90 72
R0a
Rec.
%
R1/2b
Rec.
%
11 2 10
3 2 0
27 100 0
7 3
6 1
86 33
44 67
28 33
64 49
50 60
46 53
92 88
Year of publication, number of patients, patients included, microscopical surgical margin status, and recurrences. R0 (wide/radical). b R1/2 (marginal/intralesional). a
154
Leithner et al.
tumor would lead to a higher recurrence rate at primary resection, (2) following resection of a recurrent tumor there would be no difference in the recurrence rate despite of a difference in microscopic surgical margin status. Chi-square test was used to assess the statistical significance, a P < 0.05 was considered significant. For the statistical analyses we used SPSS 11.0.1 for Windows (SPSS, Inc., Chicago, IL). RESULTS Forty-nine studies on 1,966 patients with primary extra-abdominal desmoid tumors were identified. Of these, 12 studies with 412 primary cases and 127
recurrent cases were included in the final analysis [9– 20] (Table II). Thirty-seven studies or 1,402 patients and 152 patients from the 12 included studies had to be excluded due to one or several of the following reasons: no exact data concerning microscopic surgical margin status; mixing of subgroups (abdominal/intra-abdominal/ extra-abdominal); mixing of different therapy regimes (e.g., radiotherapy); includable patients number lower than 10 (primary and recurrent cases); no follow-up given or
