Treatment of Aggressive Fibromatosis: the Experience of a Single Institution

Clinical Oncology (2009) 21: 775e780 doi:10.1016/j.clon.2009.08.012

Original Article

Treatment of Aggressive Fibromatosis: the Experience of a Single Institution M. El-Haddad*, M. El-Sebaie*, R. Ahmad*, E. Khalil*, M. Shahiny, R. Panty, M. Memonz, A. Al-Hebshi*, Y. Khafaga*, M. Al-Shabanah*, A. Allam* *Department of Radiation Oncology, King Faisal Specialist Hospital, Riyadh, Saudi Arabia; yDepartment of Orthopedic Surgery, King Faisal Specialist Hospital, Riyadh, Saudi Arabia; zDepartment of Medical Oncology, King Faisal Specialist Hospital, Riyadh, Saudi Arabia

ABSTRACT: Aims: Aggressive fibromatosis is a locally aggressive infiltrative low-grade tumour, although pathologically benign, and it does not metastasise, yet it can cause serious local distressing symptoms by virtue of local destruction and impairment of local function. The aim of this study was to emphasise the role of radiotherapy and adequate surgery in the treatment of fibromatosis in patients presenting with newly diagnosed or recurrent disease and to analyse our treatment results over 15 years for this rare tumour type. Materials and methods: Fifty-four patients with confirmed diagnosis of aggressive fibromatosis treated at King Faisal Specialist Hospital between 1990 and 2006 were identified from our local cancer registry. Forty-seven patients had surgery: complete resection (R0) in 20 patients, incomplete surgery (R1/2) in 27 patients, and seven patients had biopsy only. Forty-five patients were treated with radiotherapy: 38 patients were treated with postoperative radiotherapy, three patients were treated with preoperative radiotherapy and four patients had radiotherapy as the only treatment. The radiotherapy dose ranged between 45 and 60 Gy (median 50.4 Gy). Three patients did not receive any form of treatment apart from biopsy, but were still included in the final analysis. Results: Fifty-two per cent (28/54 patients) of our patient population had tumour recurrence when first presented to King Faisal Specialist Hospital. The median age was 29.5 years (range 2e63 years). The most common site of involvement was the extremities (28 patients). Among the 54 patients (with primary and recurrent presentation) there were 10 local recurrences, all of which were within the original primary site. The 5-year progression-free survival and overall survival rates for the whole group were 75 and 95%, respectively. Univariate and multivariate Cox regression analysis showed that the depth of invasion significantly affected progression-free survival. Conclusion: Aggressive fibromatosis is effectively treated with surgery and postoperative radiotherapy. Patients first presenting with tumour recurrence may still have local tumour control comparable with newly diagnosed patients. El-Haddad, M. et al. (2009). Clinical Oncology 21, 775—780 ª 2009 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Fibromatosis, local tumour control, radiotherapy, surgery

Introduction Aggressive fibromatosis is a locally aggressive infiltrative low-grade tumour. It virtually never metastasises, but can cause local destruction and organ dysfunction through infiltration of blood vessels and/or nerves. Radical surgery is paramount in local control, but because of the tumour’s infiltrative nature, positive surgical margins and, hence, tumour recurrence are common findings [1,2]. Almost 100 years ago, James Ewing proposed treating inoperable tumours with radiation [3]. At that time, the precise role of radiotherapy was still not clearly defined [4]. Radiotherapy may be given as adjuvant treatment to decrease local recurrence, especially when the surgical margin is positive, or as a single treatment in cases where surgery will result in 0936-6555/09/210775þ06 $36.00/0

severe morbidity or disruption of organ function [5]. Nuyttens et al. [6], in a comparative review between surgery and radiotherapy, found that local control can approach 83% for primary radiotherapy. The aim of the present study was to analyse our treatment results in this rare tumour type and to identify the important prognostic factors that could affect treatment outcome.

Materials and Methods This was a retrospective analysis of patients with primary and recurrent aggressive fibromatosis treated at King Faisal Specialist Hospital (KFSH) between 1990 and 2006. This study was approved by the research accreditation and ethical committee at KFSH in 2007.

ª 2009 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

776

CLINICAL ONCOLOGY

Patient Data Sets

Table 1 e Patient characteristics

Fifty-four patients with confirmed diagnosis of aggressive fibromatosis treated at KFSH between 1990 and 2006 were identified from our local cancer registry. All pathological specimens referred from outside hospitals were reviewed. Information taken from each patient’s chart included: gender, birthday, tumour location, operations before referral to KFSH, tumour size, surgery date, microscopic marginal status, radiation dose and fractionation, tumour control status, date of recurrence if any, date of last follow-up, and radiation complications if any. Late radiation complications were reported according to the Radiotherapy Oncology Group and European Organization for Research and Treatment of Cancer common toxicity criteria in a retrospective manner.

Characteristic

Patient Characteristics Fifty-two per cent (28/54 patients) of our patient population had tumour recurrence when first presented to KFSH. There were 37 women and 17 men, with a female to male ratio of 2.l: 1. The most common presenting symptom was a palpable mass in 29 patients (53.7%), 17 (31.5%) patients presented with both mass and pain. The median age of the studied population was 29.5 years (range 2e63 years). The duration of symptoms ranged between 1 and 120 months, with a median of 18 months. There were equal numbers of patients with either superficial (27 patients; 50%) or deeply seated tumours (27 patients; 50%). Superficial tumours were defined as lack of any involvement of the superficial fascia in extremity or trunk lesions; deep lesions were defined as: (a) all intraperitoneal, retroperitoneal, and most head and neck lesions; (b) lesions deep to or which involved the superficial fascia, in agreement with the American Joint Committee on Cancer staging 2002 [7]. The most common involved site at presentation was the extremities (28 patients; 51.8%), followed by the trunk (21 patients; 38.8%). The remaining five (9.2%) patients presented with tumours in the head and neck region. Twenty-eight patients had tumours measuring more than 10 cm. Patient characteristics are shown in Table 1.

Surgery Forty-seven patients had surgery: complete resection (R0) in 20 patients (13 patients were considered by our pathologists to have an adequate margin [ R1 cm], seven patients were considered to have a close margin [!1 cm]), incomplete (microscopic or gross residual disease; R1/2) surgery in 27 patients, and seven patients had biopsy only.

Radiotherapy Forty-five (83.4%) patients were treated at KFSH with radiotherapy: 38 (84%) patients received postoperative radiotherapy, three (11%) patients were treated with preoperative radiotherapy and four (8.8%) patients were treated with radiotherapy alone. The radiotherapy dose ranged between 45 and 60 Gy, depending on the tumour volume,

Gender Male Female Median age (range) Presentation Primary Recurrent Tumour size O10 cm %10 cm Resection status R0 R1/R2 Biopsy only Site of tumour Extremity Trunk Head and neck Depth of infiltration Superficial Deep Type of radiotherapy (n ¼ 45) Preoperative Postoperative Single modality treatment Median dose (range) Symptoms Mass Pain Both Others

All patients (n ¼ 54)

17 (31%) 37 (69%) 29.5 (2e63) 26 (48%) 28 (52%) 28 (52%) 26 (48%) 20 (37%) 27 (50%) 7 (13%) 28 (52%) 21 (38%) 5 (10%) 27 (50%) 27 (50%) 3 (6%) 38 (84%) 4 (7.4%) 50.4 Gy (45e60) 29 (54%) 3 (6%) 17 (31%) 5 (9%)

R0, complete resection; R1/2, microscopic or gross residual disease.

with an overall median dose of 50.4 Gy. Most patients (35 patients; 77.7%) were planned conventionally on the simulator using two-dimensional techniques, as most patients presented early in a period where computed tomography planning was not yet implemented as routine practice in our department. All treatment was carried out on linear accelerators, energy ranged between 6 and 18 MV, and/or electron beam, depending on tumour depth and location. The margin around the tumour ranged from 3 to 5 cm in all directions; joints were spared whenever appropriate. Irradiation of the entire circumference of an extremity was avoided, leaving a strip of normal tissue to avoid lymphoedema. Radiotherapy treatment compliance was good, with R90% of patients completing their treatment without interruption. Three patients did not receive any form of treatment apart from biopsy; they had a very short follow-up, but were still included in the final analysis.

Follow-up Schedule All patients were seen 2 months after the end of their treatment, then every 3 months for the first 2 years, every 6

TREATMENT OF AGGRESSIVE FIBROMATOSIS

months in the third year and then yearly thereafter. The determination of local outcome was based on a clinical examination, magnetic resonance imaging (MRI), and/or computed tomography. MRI was requested every 6 months in the first 2 years then yearly, or whenever recurrence or disease progression was clinically suspected. Recurrence was determined by a radiographic (computed tomography or MRI) report of increased tumour size after treatment, or regrowth of tumour in cases with no gross disease at the end of treatment.

Statistical Methods The local failure rate was estimated using the Kaplane Meier method [8]. The Log-rank test was used to identify the variable prognostic factors affecting treatment outcome, as well as a comparison of survival curves. Potential prognostic factors for progression-free survival were assessed for statistical significance using univariate and multivariate Cox regression analysis. Progression-free survival was taken from the date of diagnosis until the date of progression and overall survival was defined as from the date of diagnosis until the date of last follow-up. Data analysis was carried out using statistical software SAS version 9.1.3 (SAS institute, North Carolina, USA).

Results

777

analysis, patients with completely resected tumours did better, but this was not statistically significant. Local tumour control was 75% for patients with tumours in the extremities vs 76% for all other sites. Progression-free survival for male patients was significantly better than that for female patients (P ¼ 0.016), but the number of men was almost half that of women patients (17 vs 37) with less events. Patients with a tumour size %10 cm had better local tumour control than patients with larger tumours, with a 5-year local control rate of 85% vs 62% (P ¼ 0.77). Patients who presented with relapse did as well as primary cases, even numerically better (79% vs 69%). Superficial tumours fared better than deep tumours and this reached statistical significance in univariate as well as multivariate analyses. In the current study, the dose of radiotherapy R50.4 Gy did not affect local tumour control. Progressionfree survival for the group of patients treated by surgery plus postoperative radiotherapy was 76% at 5 years and 66% at 10 years. Six patients of our patient population were treated by surgery alone; two had complete surgery (R0) and four had incomplete surgery, (R1: two patients; R2: two patients). Radiotherapy was not given for two patients because of their age (2 and 8 years), but it is not clear from the chart why the other four patients did not receive radiotherapy, although they had a positive margin. Two patients (out of six) had tumour recurrence, with a progression-free survival rate of 33% at 5 years, and were salvaged by resurgery.

Overall Outcome For the whole group (taking all patients’ primary and recurrent cases at presentation), with a median follow-up of 7.3 years, there were 10 (18.5%) local recurrences, all of which were within the original primary site or within the irradiated field. The actuarial local control rate at 5 years was 75% and at 10 years 67% (Fig. 1). The univariate regression analysis of factors potentially affecting local control is shown in Table 2. In the univariate

1.0

Cum Survival

0.8

0.6

0.4

0.2

0.0 0.00

2.00

4.00

6.00

8.00

10.00

Years of Follow-Up Fig. 1 e Progression-free survival for the whole group (n ¼ 54 patients).

Table 2 e Local control results for all 54 patients Prognostic factor

Number of patients (n ¼ 54)

Resection margin status R0 R1/R2 Tumour site Extremities Other site Depth of invasion Superficial Deep Gender Female Male Size O10 cm %10 cm Relapse at presentation Yes No Dose of radiationy %50.4 Gy O50.4 Gy

5-year PFS (%)

P value*

20 27

79 71

0.8

28 26

75 76

0.97

27 27

85 62

0.010

37 17

74 83

0.016

28 26

62 85

0.77

28 26

79 69

0.33

35 10

76 85

0.39

R0, complete resection; R1/2, microscopic or gross residual disease. *Univariate analysis using the Log-rank test. yForty-five patients only.

778

CLINICAL ONCOLOGY

Treatment of Relapse Of 10 relapsed patients, four were salvaged by resurgery, one patient received radiotherapy at the time of relapse, one patient had surgery and radiotherapy, one patient received chemotherapy, and another patient received tamoxifen, followed by sulindac, two other patients refused further treatment. Patients salvaged with surgery, surgery and radiotherapy or radiotherapy alone had a longterm local tumour control with a follow-up of more than 5 years.

Survival Analysis Overall survival at 5 years was 95%. Two patients died, one of extensive intra-abdominal disease and the other of unrelated cause.

Complications Of 54 patients, 11 (20.4%) had grade 2 fibrosis, lower limb oedema grade 2 was reported in one patient, three patients developed neurological complications; postoperative foot drop (one), lower limb weakness related to muscle resection (one) as well as nerve entrapment from the tumour, and one patient developed Horner’s syndrome. None of our patients developed a second malignancy.

Discussion Aggressive fibromatosis is a locally aggressive tumour, with a local recurrence rate that may reach as high as 70e90%, especially after incomplete resection [9,10]. The treatment of fibromatosis usually consists of adequate surgery with a wide surgical margin (the definition of this margin is still contradictory), plus or minus postoperative radiotherapy [1,11,12]. Chemotherapy, nonsteroidal anti-inflammatory drugs or tamoxifen have been used, with limited published data [13,14]. In case surgery is considered dangerous or devastating for the patient [15], radiotherapy can still be used preoperatively or as a single treatment modality [14]. One of the major discussion points in fibromatosis is what should be the proper management for patients with fibromatosis after surgery; should we give immediate postoperative radiation or should we wait and see, considering the fact that the tumour may not progress or even may regress in certain situations [16,17]. In MGH, observation is favoured for patients who have microscopically positive margins after surgery for primary desmoid tumours. This conservative approach is based on the fact that not all patients with complete gross resection and a pathological positive margin may fail; also that salvage treatment is quite successful [18]. Local recurrence for aggressive fibromatosis may cause severe morbidity [19]. Local tumour control in our patient population was 75% at 5 years for the whole group; this may be lower than that reported by other investigators [4,20], but our results may be affected by the presence of more censored

patients. Also, our patients presented with large tumour size (more than 50% with a tumour size O10 cm). For the 38 patients who received postoperative radiotherapy, progression-free survival was 76%. Almost in accordance with the results of 41 patients treated with surgery and postoperative radiotherapy at Mass General Hospital (MGH), local recurrence occurred in 10 patients, multivariate analysis indicating that radiation had a significant effect on local tumour control [4]. Jelinek et al. [21], in a series of 54 patients, reported that postoperative radiotherapy was the only significant prognostic factor for local control; their 5-year local control rate for 35 patients who received postoperative radiotherapy was 81%. Results of studies addressing postoperative radiotherapy are contradictory [21,22]. Most have been retrospective, which may have affected the results. Some investigators have suggested that postoperative radiotherapy is indicated in the case of positive margins [19,22], but the effect of a positive margin is contradictory. In general, most investigators found that a positive margin is a negative prognostic factor [4,6]. Reitamo et al. [15,23] found a low relapse rate for patients with incomplete resections and multivariate analysis did not show a significant association of marginal status and local control. Interestingly, the type and aggressiveness of surgical intervention may not influence local control [24,25]. In the current series there was a difference in progression-free survival between patients treated with R0 vs R1/2 resection at 5 years (79 and 71%, respectively), but this difference did not reach statistical significance (P ¼ 0.8). Although more than half the patients were in relapse at first presentation, their progression-free survival and overall survival rates were comparable with those with primary presentation, even numerically better. This contradicts the results of Zlotecki et al. [24,25] who found, in a retrospective review of 65 patients, that 11 of 13 local recurrences occurred in relapsed patients, and the only significant prognostic factor was primary vs recurrent presentation. The authors suggested that tumour spillage, extensive fibrosis and more infiltration may be the cause of poorer outcome. Nuyttens et al. [6] carried out a comparative review of 22 studies, published between 1983 and 1998, on the management of fibromatosis. In total, 780 tumours were included in these studies. They found that primary presentation fared better than recurrent tumour in respect to local control (62% vs 47%). Most of our patients had surgery at local hospitals, where surgical facilities are suboptimal. Adequate surgery may still be needed for reasonable tumour control, which may explain the high recurrence rate at the time of referral. Local control rates vary at different sites (superficial, deep, extremities, others), probably reflecting the greater ease of obtaining wide surgical margins in certain anatomical areas rather than any inherent biological difference. In our group of patients there was a statistically significant difference in local tumour control between superficial/ deep tumours (P ¼ 0.01). This was not consistent with the MD Anderson data for 115 patients, as they could not find a significant effect of tumour site on local tumour control [26]. Tumour size may play some role; in our data set, local

TREATMENT OF AGGRESSIVE FIBROMATOSIS

control in 28 patients with tumour O10 cm was only 62% vs 85% for tumours %10 cm. Data from MD Anderson patients showed that tumours smaller than 5 cm had better 10-year local tumour control than tumours larger than 5 or 10 cm (90, 69 and 62%, respectively) and this was statistically significant (P ¼ 0.01). Primary radiotherapy may be preferred to surgery in cases where surgery may result in major functional loss or after recurrences after at least one attempted wide resection. Ballo et al. [5] reported one of the largest series in 1998. Twenty-one of these patients received primary radiotherapy for inoperable disease. The local control rate for gross disease was 75% and the mean dose given was 55 Gy (range 50e75 Gy). Of these 21 patients, 16 responded to treatment and five relapsed within the treatment field. Four of our patients were treated with radiotherapy alone, and none of them relapsed at a median follow-up of more than 5 years (5.3 years, the number is still very small in comparison with others). In the German Cooperative Group on Radiotherapy for Benign Diseases in their National Patterns of Care Study published in 2005 [27], satisfactory data for long-term clinical evaluation were reported for 345 patients from 19 institutions. The radiotherapy doses given ranged from 36 to 65 Gy, with a median dose of 60 Gy. About 60% of patients received primary radiotherapy alone and achieved a local control rate of 81.4%. No clear doseeresponse relationship could be established, which is similar to our finding as we did not see a significant value for doses higher than 50.4 Gy, but the number of patients was smaller. Spear et al. [4], in a study of 107 patients with fibromatosis, found that younger patients had worse outcomes, especially for lesions in the plantar region. Taking 18 years as the cut-off age, we could not find any correlation between age and local tumour control in our patient data set (P ¼ 0.73). No firm conclusions regarding the optimal management of desmoid tumours can be drawn because of the small number of patients and retrospective nature of published series, including our study. The patients presented here represent a heterogeneous study population for whom management strategies and radiation dose were individualised at the discretion of providers, rather than being based on standardised treatment algorithms. The European Organization for Research and Treatment of Cancer phase II trial (now closed) may put some light on the role of radiotherapy in inoperable aggressive fibromatosis. We will probably still need more trials to address the role of radiotherapy in the postoperative setting.

Conclusion Overall survival for aggressive fibromatosis is usually high and most deaths may not be related to the disease itself. Local control rates for primary vs recurrent disease were comparable in our data set. Radiotherapy and adequate surgery are very effective in radical treatment for aggressive fibromatosis, but still more data are needed to define when and who should be treated with radiotherapy.

779

Acknowledgment. We would like to thank Dr Abdelmoneim Eldaly from the Biostatistics Department for his help with the statistics for this study. Author for correspondence: M. El-Haddad, Kasr Al-Ainy Oncology Center, Cairo University, 115 Abd Al Aziz Al Soud St., Al Manyel. 11451, Cairo, Egypt. Tel: þ20-10-666-11-65; Fax: þ202-23930-888; E-mail: [email protected], mostafa.elhaddad@nem rock.edu.eg Received 28 January 2009; received in revised form 11 August 2009; accepted 12 August 2009

References 1 Acker JC, Bossen EH, Halperin EC. The management of desmoid tumors. Int J Radiat Oncol Biol Phys 1993;26:851e858. 2 Faulkner LB, Hajdu SI, Kher U, et al. Pediatric desmoid tumor: retrospective analysis of 63 cases. J Clin Oncol 1995;13: 2813e2818. 3 Ewing J. Neoplastic disease. Philadelphia: WB Saunders; 1928. 4 Spear MA, Jennings LC, Mankin HJ, et al. Individualizing management of aggressive fibromatoses. Int J Radiat Oncol Biol Phys 1998;40:637e645. 5 Ballo MT, Zagars GK, Pollack A. Radiation therapy in the management of desmoid tumors. Int J Radiat Oncol Biol Phys 1998;42:1007e1014. 6 Nuyttens JJ, Rust PF, Thomas Jr CR, Turrisi 3rd AT. Surgery versus radiation therapy for patients with aggressive fibromatosis or desmoid tumors: a comparative review of 22 articles. Cancer 2000;88:1517e1523. 7 Greenberg D, McIntyre H, Ramsaroop R, Arthur J, Harman J. Aggressive fibromatosis of the breast: a case report and literature review. Breast J 2002;8:55e57. 8 Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457e481. 9 Rock MG, Pritchard DJ, Reiman HM, Soule EH, Brewster RC. Extra-abdominal desmoid tumors. J Bone Joint Surg Am 1984; 66:1369e1374. 10 Markhede G, Lundgren L, Bjurstam N, Berlin O, Stener B. Extraabdominal desmoid tumors. Acta Orthop Scand 1986;57:1e7. 11 Goy BW, Lee SP, Fu YS, Selch MT, Eilber F. Treatment results of unresected or partially resected desmoid tumors. Am J Clin Oncol 1998;21:584e590. 12 Miralbell R, Suit HD, Mankin HJ, Zuckerberg LR, Stracher MA, Rosenberg AE. Fibromatoses: from postsurgical surveillance to combined surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1990;18:535e540. 13 Mukherjee A, Malcolm A, de la Hunt M, Neal DE. Pelvic fibromatosis (desmoid)etreatment with steroids and tamoxifen. Br J Urol 1995;75:559e560. 14 Catton CN, O’Sullivan B, Bell R, Cummings B, Fornasier V, Panzarella T. Aggressive fibromatosis: optimisation of local management with a retrospective failure analysis. Radiother Oncol 1995;34:17e22. 15 Reitamo JJ. The desmoid tumor. IV. Choice of treatment, results, and complications. Arch Surg 1983;118:1318e1322. 16 Mitchell G, Thomas JM, Harmer CL. Aggressive fibromatosis: evidence for a stable phase. Sarcoma 1998;2:149e154. 17 Stoeckle E, Coindre JM, Longy M, et al. A critical analysis of treatment strategies in desmoid tumours: a review of a series of 106 cases. Eur J Surg Oncol 2009;35:129e134. 18 Suit H, Spiro I. Radiation treatment of benign mesenchymal disease. Semin Radiat Oncol 1999;9:171e178.

780

CLINICAL ONCOLOGY

19 Posner MC, Shiu MH, Newsome JL, Hajdu SI, Gaynor JJ, Brennan MF. The desmoid tumor. Not a benign disease. Arch Surg 1989;124:191e196. 20 Suit HD. Radiation dose and response of desmoid tumors. Int J Radiat Oncol Biol Phys 1990;19:225e227. 21 Jelinek JA, Stelzer KJ, Conrad E, et al. The efficacy of radiotherapy as postoperative treatment for desmoid tumors. Int J Radiat Oncol Biol Phys 2001;50:121e125. 22 Merchant NB, Lewis JJ, Woodruff JM, Leung DH, Brennan MF. Extremity and trunk desmoid tumors: a multifactorial analysis of outcome. Cancer 1999;86:2045e2052. 23 Reitamo JJ, Scheinin TM, Hayry P. The desmoid syndrome. New aspects in the cause, pathogenesis and treatment of the desmoid tumor. Am J Surg 1986;151:230e237.

24 Zlotecki RA, Scarborough MT, Morris CG, et al. External beam radiotherapy for primary and adjuvant management of aggressive fibromatosis. Int J Radiat Oncol Biol Phys 2002;54:177e181. 25 Stockdale AD, Cassoni AM, Coe MA, et al. Radiotherapy and conservative surgery in the management of musculo-aponeurotic fibromatosis. Int J Radiat Oncol Biol Phys 1988;15: 851e857. 26 Guadagnolo BA, Zagars GK, Ballo MT. Long-term outcomes for desmoid tumors treated with radiation therapy. Int J Radiat Oncol Biol Phys B 2008;71(2):441e447. 27 Micke O, Seegenschmiedt MH. Radiation therapy for aggressive fibromatosis (desmoid tumors): results of a national Patterns of Care Study. Int J Radiat Oncol Biol Phys 2005;61: 882e891.