- Email: [email protected]
Pediatric aggressive fibromatosis of the head and neck: a 20-year retrospective review
Journal of Pediatric Surgery (2008) 43, 1596–1604
www.elsevier.com/locate/jpedsurg
Review
Pediatric aggressive fibromatosis of the head and neck: a 20-year retrospective review Alok Sharma a,⁎, Bo-Yee Ngan b , George K.B. Sándor c , Paolo Campisi a , Vito Forte a a
Department of Otolaryngology Surgery, Hospital for Sick Children, Toronto, Ontario, Canada Department of Pathology, Hospital for Sick Children, Toronto, Ontario, Canada c Department of Oral Surgery, Hospital for Sick Children, Toronto, Ontario, Canada b
Received 16 October 2007; revised 1 February 2008; accepted 1 February 2008
Key words: Pediatric; Fibromatosis; Head and neck; Diagnosis; Surgical management; Treatment; APC; β-catenin
Abstract Aggressive fibromatosis in children is a rare, benign condition that is locally infiltrative and destructive. It often presents as a rapidly growing, painless lump in the head and neck region. To date, only small series and case reports have been reported, and the management of the condition remains unclear. Recently, nuclear β-catenin expression has been suggested as a tumor-specific marker for aggressive fibromatosis (desmoid). Aim: The aims of the study were to review our experience of the presentation, management, and treatment outcome of pediatric aggressive fibromatosis in the head and neck and to identify the presence of the desmoid tumor marker β-catenin within this population. Method: The study was conducted as a retrospective case review of children diagnosed with aggressive fibromatosis in the head and neck for a period of 20 years and a review of the literature. Pathologic review of the original tumor specimens was undertaken for evidence of positive tumor margins and presence of nuclear β-catenin expression. Results: A total of 10 patients (6 males, 4 females) were identified. The age at presentation ranged from 12 months to 14 years. In total, 8 patients were treated with surgery alone. This included 7 patients with extension of the tumor to the resection margin; all had good long-term outcomes with no disease progression. Two patients received chemoradiotherapy, one as primary treatment, and the other as adjuvant treatment after gross incomplete resection. Both resulted in poor outcomes requiring further treatments.Within our series of pediatric fibromatosis, only 4 cases (40%) had positive results for any nuclear β-catenin expression, and 6 (60%) of 10 patients had negative results for β-catenin. Conclusion: Our experience is that total gross resection and preservation of form and function is of higher priority than achieving a negative resection margin. Pediatric fibromatosis though aggressive is still a benign condition, and careful thought should be taken before considering adjuvant chemoradiotherapy. Nuclear β-catenin expression should not be considered a specific tumor marker for pediatric aggressive fibromatosis of the head and neck. Pediatric aggressive fibromatosis in this region may be a distinct subtype of desmoid tumor from its adult form. © 2008 Elsevier Inc. All rights reserved.
⁎ Corresponding author. ENT Department, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1X8. Tel.: +1 647 241 5303. E-mail address: [email protected] (A. Sharma). 0022-3468/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2008.02.001
Aggressive fibromatosis is a benign tumor arising from connective tissue, the fascial sheaths, and musculoaponeurotic structures of muscle [1,2]. The tumor consists of dense masses of fibroblasts with a poorly defined margin
Pediatric aggressive fibromatosis of the head and neck
1597
and tends to interdigitate with muscle fibers making complete surgical excision difficult. Although aggressive fibromatosis is a nonmetastasizing tumor, it has significant potential for local invasion and recurrence [3,4] and may be fatal because of its size and location [5]. The overall incidence has been reported as 2 to 5 cases per 1 million per year [1,6,7]. Head and neck lesions represent about 12% to 15% of aggressive fibromatosis [8,9] where the tumors appear to infiltrate more widely and rapidly [10]. The incidence of childhood aggressive fibromatosis peaks at about 8 years of age [11] with a range of birth to 19 years. Other series have suggested that several age peaks occur throughout life, with a bimodal age of incidence in childhood and adolescence; an early peak around 4.5 years (0-10 years) and a second peak in 29s (15-35 years) [6]. Aggressive fibromatosis presents as a rapidly growing painless lump in the head and neck region in children. It is often fixed to underlying structures such as muscle and bone. The tumor may cause secondary symptoms such as trismus, airway obstruction, dysphagia, and proptosis depending on its site. Pain is rarely a symptom though can occur as the tumor grows. The tumor may infiltrate around nerves causing pain and dysfunction [12]. The etiology of aggressive fibromatosis is unknown. However, a genetic predisposition [13,14], association with familiar adenomatous polyposis and Gardner's syndrome [6], trauma [15-17] including surgical trauma [18,19], and endocrine factors [20] have all been implicated. Interestingly, the progression of the tumor appears to be hormonally based with regression occurring in females in menarche and menopause [6,21]. Aggressive fibromatosis may involve a deregulation of connective tissue growth. It has been suggested that the process is neoplastic rather than reactive based on Xchromosome analysis and is a monoclonal disorder [22]. Abnormal expression of c-sis and platelet-derived growth factor has been identified, which can act as a mitogen for fibrocytes [23]. It has been suggested that deep fibromatosis have somatic β-catenin or adenomatous polyposis coli (APC) gene mutations [24] leading to increased intranuclear accumulation of β-catenin that may explain the proliferative advantage of these cells [25]. Conversely, the tumor suppressor gene Rb1 has been shown to have decreased expression and may also play a role in tumor progression [26]. The clinical behavior of aggressive fibromatosis is very unpredictable, and the natural history remains unknown. There are documented cases of spontaneous regression without treatment [2,4,27,28], but because of the aggressive nature of the tumor, many clinicians are reluctant to adopt an expectant policy [11,17]. It is accepted that the most successful treatment, defined as least chance of recurrence, is primary surgical excision with a clear margin [17]. Nonsurgical treatment with radiotherapy, chemotherapy, hormonal therapy, and non-
steroidal antiinflammatory drugs (NSAIDs) are adjuvant treatments for residual tumor or recurrence. There is little evidence to support the need for aggressive adjuvant treatment of aggressive fibromatosis [11]. The side effects of treatments are well documented, and particular concern may be raised in the potential iatrogenic morbidity or mortality in a child, who may survive the benign disease for many more years than their adult counterpart. Because of the rarity of the disease, the management of the condition remains unclear and general recommendations for clinical management of residual or recurrent disease are lacking.
1. Method A retrospective chart review was performed to identify all patients with a diagnosis of aggressive fibromatosis of the head and neck region treated at The Hospital for Sick Children, Toronto, Canada, from to January 1, 1987, to December 31, 2006. Patients were identified from an institutional pathology database. A senior pathologist reviewed the original histologic and tumor specimens to confirm the diagnosis of aggressive fibromatosis. The original specimens were reexamined for positivity of the surgical resection margin. Specimens were stained for β-catenin to determine its usefulness as a specific tumor marker for aggressive fibromatosis [24,29,30]. A representative formalin-fixed, paraffin-embedded tumor tissue section form each of the patients was subjected to immunohistochemical staining with commercially available biotinylated mouse monoclonal antibody to β-catenin (1/200 dilution; BD Transduction Laboratories, Lexington, Ky). Sections were deparaffinized and subjected to antigen retrieval procedures. Immunostaining was performed with an automated stainer (Ventanna, Tucson, Ariz). Antibody reaction was detected by the biotinavidin/immunoperoxidase amplification procedures with the appropriate reagents and methods recommended and supplied by the staining equipment supplier (Ventanna, Tucson, Ariz). Nuclear staining/expression of β-catenin was assessed by microscopy and recorded as percent positive-stained nuclei in the areas within the tissue that showed the highest staining intensities. The medical records of identified cases were closely reviewed to follow the progress and outcomes of each individual. The age, mode, and duration of presentation, site, and size of the lesion were all noted. Initial management (surgical or nonsurgical) was noted, as was any subsequent management if any, for residual or recurrent disease. The disease progression was monitored both clinically and radiologically. A literature review was performed using Medline. “Aggressive fibromatosis” and “desmoid tumors” were the keywords used for the search. The review included pediatric case series and series that included both adults and children.
1598
Table 1
Summary of presentation and diagnosis
Case
Sex
Age
Presentation
Pain
Duration of symptoms
Trauma
Associations (eg, Gardner)
Radiology
Diagnosis
1
Male
6 y 7 mo
Right jaw swelling
No
10 d
No
RubensteinTaybi
Intraoral Bx diagnostic
2
Female
1 y 9 mo
Right jaw swelling
No
3wk
No
No
3
Male
3 y 4 mo
Difficulty eating and speech
No
No
No
4
Male
3 y 3 mo
No
No
No
Male
12 y 9 mo
No
1 mo
No
No
6
Female
14 y 8 mo
Yes
No
Female
2 y 5 mo
3 mo, 1 wk, and 1 mo, respectively 3-mo recent facial palsy
No
7
Torticollis and pain, Horner syndrome, and neck mass Right facial swelling, VII N weakness
No
No
8 9
Male Female
1y 5 y 6 mo
Right neck mass Left jaw swelling, malocclusion
No No
few wk 1y
No No
No No
10
Male
1 y 10 mo
Right jaw swelling, increasing trismus
No
6 mo
Yes
No
CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated No CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated
Intraoral Bx diagnostic
5
Obstructive sleep apnea, dysphagia facial swelling Right jaw swelling
2 mo (recurrence at 2 mo) 3-4 mo
CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated MRI demonstrated extent of recurrence
No
Core Bx nondiagnostic, intraoral biopsy diagnostic Initial excision Bx diagnostic
Intraoral Bx diagnostic
Open Bx diagnostic
Transoral Bx diagnostic
Excision Bx Intraoral Bx nondiagnostic
Intraoral Bx diagnostic
A. Sharma et al.
Summary of management and treatment outcome
Case Sex
Age
Region
β-catenin
Primary treatment
Residual disease
Positive margin Negative Positive margin Slightly positive (20%) Positive margin Negative
Adjuvant therapy
Progression after Recurrence Outcome primary treatment (follow-up)
No No
No No
No No
Good (2 y) Good (2 y)
No
No
Yes
Tracheotomy
Yes
N/A
No No Chemotherapy, Yes radiotherapy, hormonal therapy
No N/A
Recurrence (at 2 mo) treated with surgery; good (36 mo) Very poor mouth opening and continues with tracheotomy (14 y) Good (5 y) Morbidity from adjunct treatment; disease now stable (4 y)
No
No
No
Good (4 y)
1 2
Male 6 y 7 mo Right mandible Female 1 y 9 mo Right mandible
Surgery Surgery
3
Male
3 y 4 mo Floor of mouth
Surgery
4
Male
3 y 3 mo Left parapharyngeal Chemotherapy N/A up to skull base
5 6
Male 12 y 9 mo Right mandible Female 14 y 8 mo Left neck from hyoid to thoracic inlet
Surgery Surgery
7
Female 2 y 5 mo Right cheek infratemporal skull base Male 1y Right neck Female 5 y 6 mo Left mandible, skull base, pterygopalitine fossa Male 1 y 10 mo Right mandible, infratemporal fossa to hypopharynx
Surgery
Positive margin Negative Macroscopic Positive (25%) encasing carotid and brachial plexus Macroscopic Positive (30%)
Surgery Surgery
No Negative Positive margin Negative
No No
No No
No No
Good (6 mo) Good (17 y)
Surgery
Positive margin Negative
No
No
No
Good (18 y)
8 9
10
Slightly positive (10%)
Pediatric aggressive fibromatosis of the head and neck
Table 2
1599
1600
A. Sharma et al.
Fig. 1
Macroscopic appearance of tumor.
Studies that examined the management of intraabdominal aggressive fibromatosis were ruled out.
2. Results For a 20-year period, a total of 10 cases (6 male and 4 female) of aggressive fibromatosis of the head and neck were identified from a total of 340 cases of fibromatosis of the whole body, of which 95 were extraabdominal; that is, 10% of the extraabdominal fibromatosis occur within the head and neck region in our series. None were associated with Gardner's syndrome. The presentation and diagnosis of the individual cases are summarized in Table 1. The age at presentation ranged from 12 months to 14 years with a median age of 3 years. The duration of symptoms ranged from 1 week to a year with simple swelling as the most common symptom. Half of the cases presented as jaw swelling, 2 of which slowly presented for many months as trismus and malocclusion (cases 9 and 10) before obvious external swelling. The other cases presented with local swelling (case 8) and/or secondary symptoms to the swelling, for example, dysphagia (case 3) and obstructive sleep apnea (case 4). Only one case presented with pain and later developed a Horner syndrome (case 6); this tumor was subsequently found to have significant neurovascular involvement. Another case (case 7) presented with facial swelling and later a facial nerve weakness; the tumor was found to be involving the skull base. In both of these cases the neurologic signs were a recent onset phenomena at presentation, with initial symptoms present for several months. Only one patient (case 10) had any suggestion of trauma. None of the patients had a history of Gardner's or familial adenomatous polyposis, though one patient did have Rubinstein-Taybi syndrome (case 1). Nearly every patient had radiologic imaging as part of the diagnostic workup. The CT scan was used in most cases to demonstrate the tumor size and extent; one patient (case 3) had a magnetic resonance imaging (MRI) scan to diagnose the extent of recurrent disease before further management.
Radiology was not diagnostic of the tumor disease itself. One patient required no preoperative imaging because of the superficial clinical presentation (case 8). Tumor diagnosis was with open biopsy, either intraorally or externally depending in the site of the tumor. Case 2 had a core biopsy initially performed; this was nondiagnostic as a low-grade fibrohistocytic lesion before an open biopsy that confirmed the diagnosis of aggressive fibromatosis. After a pathologic diagnosis of aggressive fibromatosis, the management and treatment outcomes are summarized in Table 2. In total, 8 patients were treated with surgery alone, and all had good long-term outcomes. Seven of these patients had pathologic microscopic extension of the tumor to the resection margin. In 2 patients (cases 6 and 7), gross macroscopic tumor was left unresected. Macroscopic surgical resection was not possible in case 6 because of involvement of the carotid artery and brachial plexus. A large amount of the tumor (two thirds) was left unresected. The patient received adjuvant chemoradiotherapy for the residual tumor. The initial chemotherapy vinblastine and methotrexate was tolerated but with minor side effect of an extravasation burn. This course of chemotherapy appeared to be ineffective in controlling the residual disease progression. A second course of chemotherapy with vincristine and actinomycin was not tolerated because of side effects and stopped after only a few doses. Radiotherapy was commenced at the same time of the second course of chemotherapy and the tumor appeared to have a slight response. However, the radiotherapy did not cause significant tumor regression, rather a cessation of progressive disease. Within this patient, the tumor remained quiescent until 3 years postsurgery; at this time, it again appeared to enlarge, diagnosed on follow-up MRI scan. Tamoxifen and sulindac were started as therapy, initially inducing regression, and to date, the disease appears stable on this medication. In case 7, macroscopic surgical clearance was not possible because of involvement with the skull base. The patient was
Fig. 2
Microscopic appearance of tumor.
Pediatric aggressive fibromatosis of the head and neck
1601
closely observed radiologically and no adjuvant therapy given. The disease was seen to not progress and appeared to “burn out” over several years. At 1-year postoperatively, the residual contrast-enhancing tumor at the skull base was seen on CT scan; for the next 2 years, the follow-up scans revealed no contrast enhancement. Case 3 illustrates that tumors recurring aggressively, that is, within 2 months, can still be treated with surgical excision. Despite positive resection margins the child is recurrence-free 3 years postoperatively. One patient's guardian (case 4) opted for chemotherapy as a primary treatment. The initial course of chemotherapy with vincristine and actinomycin failed to prevent early tumor progression. The patient required emergency surgical intervention in the form of a tracheotomy. A second course of chemotherapy with vinblastine and methotrexate did appear to prevent tumor progression but in the long term did not appear to induce tumor regression. Chemotherapy appeared to be unsuccessful as a primary treatment in this case. All patients were followed up with either CT or MRI scan on a regular basis—the exception was case 8, who was discharged at 6 months postoperatively. The pathologic finding demonstrated complete resection of the tumor with a clear resection margin. Clinically, the original tumor was also very superficial, and follow-up was under the patient's local physician. As well as regular radiologic follow-up, cases 9 and 10 underwent significant reconstructive surgery after their original surgical procedure. During these procedures, the original tumor site was examined clinically and also rebiopsied, with no evidence of recurrent disease, both at around 2 years original postsurgery. Case 7 also, during a routine tonsillectomy 3 years postoperatively, had a biopsy taken of scar tissue in the tonsillar fossa, which was pathologically negative for recurrence of fibromatosis.
patients with extension of disease to the tumor margin had any clinical or radiologic evidence of tumor progression or recurrence. In the 2 patients with gross macroscopic residual disease; one patient underwent regression, that is, 50% reduction; the other significant progression despite adjuvant chemotherapy.
3. Pathologic results 3.1. Typical appearance of tumors The macroscopic appearance of fibromatosis is often indistinguishable from scar tissue, having a white fibrous appearance (Fig. 1). The tumor often interdigitates with muscle fibers, making surgical excision difficult and often requiring excision of muscle with the specimen. Microscopically, the tumor showed intersecting broad sweeping bands of spindle cells within a collagenous stroma. These spindle cells are uniform in shape and size and had cytological appearance of fibroblasts (Fig. 2).
3.2. Tumor resection margin and residual disease Only one case had a negative pathologic tumor margin (case 8) of the 9 patients treated surgically. None of the
3.3. Local invasion and metastasis In all patients where enlarged lymph nodes were taken as specimens with the tumor, pathologically these lymph nodes had negative result for fibromatosis. Where local salivary glands were taken, there was also no microscopic disease invasion of the salivary gland tissue.
3.4. Nuclear b-catenin expression By immunohistochemistry, 6 (60%) of 10 patients had negative results for nuclear β-catenin staining, and 2 cases showed 25% to 30% positive nuclear expression and a further two showed less than 25% positive nuclear staining, which is considered by some to be a negative result [29]. However, by following the criteria of at least focal nuclear staining in every studied sample of fibromatosis [24], 4 (40%) of 10 patients were reported as positive in this study.
4. Discussion 4.1. Incidence and etiology of pediatric fibromatosis The age of incidence of pediatric fibromatosis has been reported with an average age around 8 years old [11]. Our experience is that possibly 2 groups of incidence may occur; most pediatric cases occur before 6 years of age and an older group later in childhood, which is in keeping with previously described models [6]. It has been suggested that aggressive fibromatosis of the head and neck region occurs in a younger age group than other sites [16,31]. It is interesting that in our series, the case of progressive disease occurred in an older female child; in the literature, it has previously been suggested that the disease in adolescents is more aggressive or more likely to progress in young female adults [6]. Within our series, we have a male preponderance, in keeping with recent literature reviews [11,32,33], though other nonpediatric studies have found a female preponderance [8,34], possibly skewed by the fertile female group [6], whereas other studies suggest no sex preponderance in pediatric aggressive fibromatosis [17]. None of the patients with our series had any association with Gardner's syndrome, an autosomal disease characterized by the presence of colonic polyposis, osteomas, and a multitude of soft tissue tumors as described in the literature
1602 [6,17,35,36]. This is similar to the experience of similar extraabdominal pediatric aggressive fibromatosis series [15,32], suggesting less relevance of this association within this group. One of the children in our series (case 1) had Rubinstein-Taybi syndrome [37], which is a genetic multisystem disorder characterized by facial abnormalities, broad thumbs and great toes, and mental retardation. Rubinstein-Taybi syndrome is associated with soft tissue tumors [38]. However, no previous association of fibromatosis and Rubenstein-Taybi syndrome has previously been reported in the literature. It is difficult from a single case to say whether a rare disease can be associated with a rare syndrome, though the syndrome's link with soft tissue tumors may suggest a true association of the disease. Trauma has been suggested as a possible etiologic factor [15-17], but only 1 of 10 in our series (case 10) had possible trauma to the area; indeed, it has been suggested that trauma may not be so relevant [17].
4.2. Diagnosis of fibromatosis, the role of nuclear b-catenin expression Pathologic diagnosis of the condition of aggressive fibromatosis in children still appears to be simple histologic diagnosis. Nuclear β-catenin does not appear to be an accurate marker in the diagnosis of aggressive fibromatosis in children. This is probably related to pediatric fibromatosis that does not appear to have as strong an association with familial adenomatous polyposis (FAP) or Gardner's syndrome as its adult form. Nuclear β-catenin accumulation/expression is produced by the APC gene mutation associated with FAP and may therefore be much less relevant in diagnosis of pediatric fibromatosis as found in our series. It is also of note that the pathologic series identifying β-catenin as a tumor marker for aggressive fibromatosis included only 2 pediatric or adolescent patients [24,30]. The pathophysiology of pediatric aggressive fibromatosis becomes less clear than in the adult form, with the lesser association with the APC gene or the activation of the wnt signaling pathway or somatic mutation that affects phosphorylation sites within exon 3 of β-catenin, causing protein stabilization that resulted in nuclear β-catenin accumulation/expression as seen in adult colonic cancer [39]. Although we did not perform sequence analyses of exon 3 to rule out somatic mutation as a cause of nuclear β-catenin accumulation/expression, in a study of palmar fibromatosis in adults and one infantile digital fibromatosis that showed nuclear βcatenin expression, the exon 3 region were found to be nonmutated [40]. Thus, it appears that somatic mutation in exon 3 of β-catenin is uncommon in fibromatosis [24,40]. It may be that the pediatric form is a similar but distinct pathologic condition, and literature based on
A. Sharma et al. treatment of adult series of aggressive fibromatosis becomes much less relevant.
4.3. Relevance of tumor margin and management outcome It is suggested that surgical resection with a wide clearance margin, free of macroscopic, and microscopic disease is the treatment of choice [31]; and the use of adjuvant therapies is recommended when the tumor is unresectable because of vital structures or surgery has failed to obtain surgical clearance or [11,31]. The goal of complete macroscopic and microscopic surgical clearance is difficult, as in the authors experience; these tumors tend to interdigitate with the muscle fibers [41] and may be difficult to distinguish from scar tissue in recurrent disease. Microscopic residual disease is a significant risk factor for recurrence of tumor [17]. Most recurrences occur within 2 years [33,36] to 3 years and nearly all within 6 years of presentation [31]. The primary consideration for treatment is regarded as prevention of recurrence [42], and the evidence in the literature appears to be based on treatment failure as recurrence of tumor [1,8,11,17,31,43-45] and not necessarily the progression of the disease. Aggressive fibromatosis has no features of anaplasia and does not seem to show classical neoplastic progression; if kept under control, there might be a high probability that eventually the tumor will stop growing and regress [17]. Aggressive fibromatosis in the head and neck is said to have a higher incidence of recurrent disease [8], although this may reflect that excision is more technically difficult in this region [41] and more likely to leave residual disease. It has been suggested also that residual disease after surgical resection is a significant risk factor of recurrent disease, with positive resection margins after surgery indicating high risk for disease recurrence [11,17]. This does not appear to be our experience. Aggressive fibromatosis can be a locally destructive disease, invading local structures, particularly relevant in the head and neck region that contains many vital structures. The risk of morbidity of pediatric aggressive fibromatosis therefore is higher within this region. In a review of pediatric series containing 114 cases, there were only 2 deaths from invasion both within the head and neck [17]. Aggressive management with adjuvant therapy is therefore advocated in the presence of any residual disease by some authors [17]. However, this panacea of management does not take into account that residual disease may indeed not progress but remain quiescent and indeed in time regress, as demonstrated in case 7 of our series of patients. In aggressive fibromatosis of the limbs, it has been shown that aggressive management surgically to achieve negative margins may not be necessary and may result in unnecessary morbidity [46]. This may also
Pediatric aggressive fibromatosis of the head and neck
1603
be the case for adjuvant therapies such as chemotherapy or radiotherapy that are not without their own iatrogenic morbidity, of especial consideration in the child. Chemotherapy, radiotherapy, and drug therapies including hormonal and NSAIDs have all been suggested as alternatives to surgery or as adjuvant therapy to primary surgery in the presence of microscopic residual disease or extension to the margins. In adults, the standard approach for microscopic residual disease is adjuvant radiotherapy to achieve a high local control rate [11,17,45,47,48]. A dose of 50 to 60 Gy is thought to be efficacious [45]. However, doses higher than 50 Gy required for tumor control may lead to skin ulceration, poor wound healing, and local tissue necrosis [32,33]; in pediatric patients in particular, inherent risks for growth problems and long-term risk of secondary malignancy [11,45] are of concern. Radiotherapy appears to be less effective treatment of aggressive fibromatosis in children [36]. Treatment failure to control disease progress from primary radiotherapy and then requiring salvage curative therapy by surgery has been reported [49]. Chemotherapy may be a reasonable alternative to radiotherapy in a growing child [32], though there is an inherent risk of secondary malignancies, fertility problems, cardiotoxicity [11,50], and neutropenia [51]. Chemotherapy has been considered for delay of progression of aggressive fibromatosis until children are considered old enough for radiotherapy [43] or surgery [32]. There is no strong evidence that chemotherapy agents are efficacious, and possible role may be with a rapid growing tumor to obtain local control [17]. It still remains unclear from the literature as to which cytotoxic chemotherapeutic agents are most effective or indeed what combination [50], and this is indeed an area for future research [11,45]. Vinblastine and methotrexate in low does has been shown to control progression of pediatric aggressive fibromatosis [32,52]. It has been suggested recently that vinblastine and methotrexate may induce tumor regression [51]; however, this study did not contain a control group and may simply reflect the natural progression within the population. In our series, it was only partially successful with case 4 and unsuccessful in case 6. Although successful treatment with vincristine and actinomycin has been reported [5], it was also not the case in our limited experience. Concerns over potential morbidity of cytotoxic agents have led to interest in noncytotoxic drugs such as tamoxifen and NSAIDs. Regression of desmoid tumors has been described at menarche and menopause [6,21] suggesting a hormonal influence. Tamoxifen is an antiestrogen and has been found to have an inhibitory effect on desmoid tumors, and estrogen receptors have been found in aggressive fibromatosis tumors [11]. Nonsteroidal antiinflammatory drugs are thought to stimulate the immune response, thus impairing the proliferation of tumor cells [42] and may well have a role. Tamoxifen and sulindac (NSAID) have been used with success in patients with FAP [53], though FAP is not associated with our patients in this series. Vitamin D has
also been suggested as a treatment by its metabolites inhibiting proliferation of fibroblasts [54]. Although the role of adjuvant therapies remains uncertain, surgery can be considered as a primary treatment of recurrent disease in aggressive fibromatosis case 3 or in management of progressive disease case 4. It has been shown that the recurrence rate after a second or third resection is equal to or lower than the recurrence rate after the primary surgery [6,46]. It has been suggested that if residual aggressive fibromatosis may be kept under control, that is, treating progression, there may be a high probability that eventually it will stop growing and even regress [55], as demonstrated in case 7. Within our series, the presence of a positive tumor margin does not appear to be a predictor of disease progression with none of the macroscopically cleared disease but positive resection margin tumors undergoing disease progress.
4.4. Disease monitoring and follow-up We recommend that in the presence of residual disease, in most of our cases, the patient may simply be observed using MRI or CT scan. Our experience and review of the literature suggests that more aggressive disease can reoccur or actively progress within the first few months. Therefore, we suggest a routine scan 3 monthly for first 6 months, then 6 monthly until 2 years, then yearly review until the disease is stable. Should progression of residual tumor or tumor recurrence occur, indicating active disease, surgical resection may still be considered although adjuvant chemotherapy, radiotherapy, or hormonal therapy should also be considered within the context of the morbidity of the disease versus the morbidity of the treatment.
References [1] Greenberg HM, et al. Radiation therapy for the treatment of benign locally aggressive fibromatosis. Int J Radiat Oncol Biol Phys 1980; 6(10):1412. [2] Enzinger FM, Shiraki M. Musculo-aponeurotic fibromatosis of shoulder girdle (extra abdominal desmoid)—analysis of 30 cases followed up for more years. Cancer 1967;20(7):1131-40. [3] Fletcher CDM, Krishnan K, Merlens F. World Health Organization: tumours of soft tissue and bone. Lyon: WHO; 2002. [4] Weiss SW, Goldblum JR. Enzinger and Weiss's soft tissue tumours. 4th ed. Mosby: St Louis; 2001. [5] Stein R. Chemotherapeutic response in fibromatosis of neck. J Pediatr 1977;90(3):482-3. [6] Reitamo JJ, Schenin TM, Hayry P. The desmoid syndrome: new aspects in the cause, pathogenesis and the treatment of the desmoid tumor. Am J Surg 1986;151:230-7. [7] Caglar K, et al. Effective treatment of multifocal aggressive fibromatosis with low-dose chemotherapy. Turk J Pediatr 2006;48(4): 365-8. [8] Masson JK, Soule EH. Desmoid tumors of the head and neck. Am J Surg 1966;112:615-22. [9] West CB, Shagets FW, Mansfield MJ. Nonsurgical treatment of aggressive fibromatosis in the head and neck. Otolaryngol Head Neck Surg 1989;101(3):338-43.
1604 [10] Wilkins SA, et al. Aggressive fibromatosis of fibromatosis of the head and neck. Am J Surg 1975;130(4):412-5. [11] Buitendijk S, et al. Pediatric aggressive fibromatosis—a retrospective analysis of 13 patients and review of the literature. Cancer 2005;104(5):1090-9. [12] Allen PW. The fibromatosis: a clinicopathological classification based on 140 cases. Am J Surg Pathol 1977:255-60. [13] Tejpar S, Nollet F, Li C. Predominance of beta-catenin mutations and beta-catenin dysregulation in sporadic aggressive fibromatosis (desmoid tumor). Oncogene 1999;18:6615-20. [14] Skubitz KM, Skubitz APN. Gene expression in aggressive fibromatosis. J Lab Clin Med 2004;143(2):89-98. [15] Rao BN, et al. Challenge in the treatment of children fibromatosis. Arch Surg 1987;122:1987. [16] Ayala AG, et al. Desmoid fibromatosis—a clinicopathological study of 25 children. Semin Diagn Pathol 1986;3(2):138-50. [17] Faulkner LB, et al. Pediatric desmoid tumor: restrospective analysis of 63 cases. J Clin Oncol 1995;13(11):2813-8. [18] Penick RM. Desmoid tumors developing in operative scars. Int Surg Dig 1937;23:323-9. [19] McKinnon JG, Neifield JP, Kay S. Management of desmoid tumors. Surg Gyecol Obstet 1989;169:104-6. [20] Janinis J, et al. The pharmacological treatment of aggressive fibromatosis: a systematic review. Ann Oncol 2003;14(2):181-90. [21] Dahn I, Jonsson N, Lundh G. Desmoid tumors: a series of 33 cases. Acta Chir Scand 1963(126):305-14. [22] Alman BA, et al. Aggressive fibromatosis (desmoid tumor) is a monoclonal disorder. Am J Surg Pathol 1996:194-200. [23] Alman BA, et al. Aggressive fibromatosis. J Pediatr Orthop 1992;12:1-10. [24] Bhattacharya B, et al. Nuclear beta-catenin expression distinguishes deep fibromatosis from other benign and malignant fibroblastic and myofibroblastic lesions. Am J Surg Pathol 2005;29(5):653-9. [25] Alman BA, et al. Increased beta catenin and somatic APC mutations in sporadic aggressive fibromatosis. Am J Pathol 1997;151:329-34. [26] Muller E, et al. Molecular genetic and immunohistochemical analysis of the tumor suppressor genes Rb and p53 in palmar and aggressive fibromatosis. Am J Surg Pathol 1996:194-200. [27] Jenkins N, Freeman LS, McKibbin B. Spontaneous regression of a desmoid tumour. J Bone Joint Surg 1986;68(5):780-1. [28] Dormans JP, Spiegel DA, Mayer J. Fibromatosis in childhood: the desmoid/fibromatosis complex. Blood Cancer 2001;37:126-31. [29] Ng TL, et al. Nuclear beta-catenin in mesenchymal tumors. Mod Pathol 2005;18(1):68-74. [30] Hauben EI, et al. Desmoplastic fibroma of bone: an immunohistochemical study including b-catenin expression and mutational analysis for b-catenin. Hum Pathol 2005;36(9):1025-30. [31] Spiegel DA, et al. Aggressive fibromatosis in infancy to adolescence. J Pediatr Orthop 1999;19(6):776. [32] Skapek SX, et al. Combination chemotherapy using vinblastine and methotrexate for the treatment of progressive desmoid tumor in children. J Clin Oncol 1998;16(9):3021-7.
A. Sharma et al. [33] Perez-Cruet MJ, et al. Aggressive fibromatosis involving the cranial base in children. Neurosurgery 1998;43(5):1096-102. [34] Janahi WM, et al. Infantile fibromatosis. J Laryngol Otol 1999;113(3): 246-9. [35] Carr RJ, et al. Infantile fibromatosis with involvement of the mandible. Br J Oral Maxillofac Surg 1992;30(4):257-62. [36] Merchant TE, et al. Long-term results with radiation therapy for pediatric desmoid tumors. Int J Radiat Oncol Biol Phys 2000;47(5): 1267-71. [37] Rakheja D, et al. Immunohistochemical expression of beta-catenin in solitary fibrous tumors. Arch Pathol Lab Med 2005;129(6):776-9. [38] Iyer NG, Özdag H, Caldas C. p300/CBP and cancer. Oncogene 2004;23(24):4225. [39] Segditsas S, Tomlinson I. Colorectal cancer and genetic alterations in the Wnt pathway. Oncogene 2006;25(57):7531-7. [40] Montgomery E, et al. Superficial fibromatosis are genetically distinct from deep fibromatosis. Mod Pathol 2001;14(7):695-701. [41] Saunders KW, et al. Aggressive fibromatosis of the parapharyngeal space: two cases and treatment recommendations. Ear Nose Throat J 2004:262-8. [42] Yousry ES. Fibromatosis of the head and neck. J Laryngol Otol 1992;106:459-62. [43] Tostevin PMJ, Wyatt M, Hosni A. Six cases of fibromatosis of the head and neck in children. Int J Pediatr Otorhinolaryngol 2000;53(3): 235-44. [44] Siegel NS, Bradford CR. Fibromatosis of the head and neck: a challenging lesion. Otolaryngol Head Neck Surg 2000;123(3):269-75. [45] Spiegel DA, et al. Aggressive fibromatosis from infancy to adolescence. J Pediatr Orthop 1999;19(6):776-84. [46] Lewis JJ, et al. The enigma of desmoid tumors. Ann Surg 1999;229(6): 866-72. [47] 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(3):882-91. [48] Shin KH, et al. The role of radiotherapy in the treatment of aggressive fibromatosis. Yonsei Med J 1999;40(5):439-43. [49] Watzinger F, et al. Aggressive fibromatosis of the mandible: a case report. Int J Oral Maxillofac Surg 2005;34:211-3. [50] Goepfert H, et al. Chemotherapy of locally aggressive head and neck tumors in the pediatric age group—desmoid fibromatosis and nasopharyngeal angiofibroma. Am J Surg 1982;144(4):437-44. [51] Skapek SX, et al. Vinblastine and methotrexate for desmoid fibromatosis in children: results of a Pediatric Oncology Group phase II trial. J Clin Oncol 2007;25(5):501-6. [52] Reich S. Low-dose chemotherapy with vinblastine and methotrexate in childhood desmoid tumors. J Clin Oncol 1999;17(3):1086. [53] Hansmann A, et al. High-dose tamoxifen and sulindac as first-line treatment for desmoid tumors. Cancer 2004;100:612-20. [54] Ferah Y, et al. Possible therapeutic role of vitamin D-3 in aggressive fibromatosis. Jpn J Clin Oncol 2004;34(8):472-5. [55] Rock MG, et al. Extra-abdominal desmoid tumors. J Bone Joint Surg 1984:1369-74.
www.elsevier.com/locate/jpedsurg
Review
Pediatric aggressive fibromatosis of the head and neck: a 20-year retrospective review Alok Sharma a,⁎, Bo-Yee Ngan b , George K.B. Sándor c , Paolo Campisi a , Vito Forte a a
Department of Otolaryngology Surgery, Hospital for Sick Children, Toronto, Ontario, Canada Department of Pathology, Hospital for Sick Children, Toronto, Ontario, Canada c Department of Oral Surgery, Hospital for Sick Children, Toronto, Ontario, Canada b
Received 16 October 2007; revised 1 February 2008; accepted 1 February 2008
Key words: Pediatric; Fibromatosis; Head and neck; Diagnosis; Surgical management; Treatment; APC; β-catenin
Abstract Aggressive fibromatosis in children is a rare, benign condition that is locally infiltrative and destructive. It often presents as a rapidly growing, painless lump in the head and neck region. To date, only small series and case reports have been reported, and the management of the condition remains unclear. Recently, nuclear β-catenin expression has been suggested as a tumor-specific marker for aggressive fibromatosis (desmoid). Aim: The aims of the study were to review our experience of the presentation, management, and treatment outcome of pediatric aggressive fibromatosis in the head and neck and to identify the presence of the desmoid tumor marker β-catenin within this population. Method: The study was conducted as a retrospective case review of children diagnosed with aggressive fibromatosis in the head and neck for a period of 20 years and a review of the literature. Pathologic review of the original tumor specimens was undertaken for evidence of positive tumor margins and presence of nuclear β-catenin expression. Results: A total of 10 patients (6 males, 4 females) were identified. The age at presentation ranged from 12 months to 14 years. In total, 8 patients were treated with surgery alone. This included 7 patients with extension of the tumor to the resection margin; all had good long-term outcomes with no disease progression. Two patients received chemoradiotherapy, one as primary treatment, and the other as adjuvant treatment after gross incomplete resection. Both resulted in poor outcomes requiring further treatments.Within our series of pediatric fibromatosis, only 4 cases (40%) had positive results for any nuclear β-catenin expression, and 6 (60%) of 10 patients had negative results for β-catenin. Conclusion: Our experience is that total gross resection and preservation of form and function is of higher priority than achieving a negative resection margin. Pediatric fibromatosis though aggressive is still a benign condition, and careful thought should be taken before considering adjuvant chemoradiotherapy. Nuclear β-catenin expression should not be considered a specific tumor marker for pediatric aggressive fibromatosis of the head and neck. Pediatric aggressive fibromatosis in this region may be a distinct subtype of desmoid tumor from its adult form. © 2008 Elsevier Inc. All rights reserved.
⁎ Corresponding author. ENT Department, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1X8. Tel.: +1 647 241 5303. E-mail address: [email protected] (A. Sharma). 0022-3468/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2008.02.001
Aggressive fibromatosis is a benign tumor arising from connective tissue, the fascial sheaths, and musculoaponeurotic structures of muscle [1,2]. The tumor consists of dense masses of fibroblasts with a poorly defined margin
Pediatric aggressive fibromatosis of the head and neck
1597
and tends to interdigitate with muscle fibers making complete surgical excision difficult. Although aggressive fibromatosis is a nonmetastasizing tumor, it has significant potential for local invasion and recurrence [3,4] and may be fatal because of its size and location [5]. The overall incidence has been reported as 2 to 5 cases per 1 million per year [1,6,7]. Head and neck lesions represent about 12% to 15% of aggressive fibromatosis [8,9] where the tumors appear to infiltrate more widely and rapidly [10]. The incidence of childhood aggressive fibromatosis peaks at about 8 years of age [11] with a range of birth to 19 years. Other series have suggested that several age peaks occur throughout life, with a bimodal age of incidence in childhood and adolescence; an early peak around 4.5 years (0-10 years) and a second peak in 29s (15-35 years) [6]. Aggressive fibromatosis presents as a rapidly growing painless lump in the head and neck region in children. It is often fixed to underlying structures such as muscle and bone. The tumor may cause secondary symptoms such as trismus, airway obstruction, dysphagia, and proptosis depending on its site. Pain is rarely a symptom though can occur as the tumor grows. The tumor may infiltrate around nerves causing pain and dysfunction [12]. The etiology of aggressive fibromatosis is unknown. However, a genetic predisposition [13,14], association with familiar adenomatous polyposis and Gardner's syndrome [6], trauma [15-17] including surgical trauma [18,19], and endocrine factors [20] have all been implicated. Interestingly, the progression of the tumor appears to be hormonally based with regression occurring in females in menarche and menopause [6,21]. Aggressive fibromatosis may involve a deregulation of connective tissue growth. It has been suggested that the process is neoplastic rather than reactive based on Xchromosome analysis and is a monoclonal disorder [22]. Abnormal expression of c-sis and platelet-derived growth factor has been identified, which can act as a mitogen for fibrocytes [23]. It has been suggested that deep fibromatosis have somatic β-catenin or adenomatous polyposis coli (APC) gene mutations [24] leading to increased intranuclear accumulation of β-catenin that may explain the proliferative advantage of these cells [25]. Conversely, the tumor suppressor gene Rb1 has been shown to have decreased expression and may also play a role in tumor progression [26]. The clinical behavior of aggressive fibromatosis is very unpredictable, and the natural history remains unknown. There are documented cases of spontaneous regression without treatment [2,4,27,28], but because of the aggressive nature of the tumor, many clinicians are reluctant to adopt an expectant policy [11,17]. It is accepted that the most successful treatment, defined as least chance of recurrence, is primary surgical excision with a clear margin [17]. Nonsurgical treatment with radiotherapy, chemotherapy, hormonal therapy, and non-
steroidal antiinflammatory drugs (NSAIDs) are adjuvant treatments for residual tumor or recurrence. There is little evidence to support the need for aggressive adjuvant treatment of aggressive fibromatosis [11]. The side effects of treatments are well documented, and particular concern may be raised in the potential iatrogenic morbidity or mortality in a child, who may survive the benign disease for many more years than their adult counterpart. Because of the rarity of the disease, the management of the condition remains unclear and general recommendations for clinical management of residual or recurrent disease are lacking.
1. Method A retrospective chart review was performed to identify all patients with a diagnosis of aggressive fibromatosis of the head and neck region treated at The Hospital for Sick Children, Toronto, Canada, from to January 1, 1987, to December 31, 2006. Patients were identified from an institutional pathology database. A senior pathologist reviewed the original histologic and tumor specimens to confirm the diagnosis of aggressive fibromatosis. The original specimens were reexamined for positivity of the surgical resection margin. Specimens were stained for β-catenin to determine its usefulness as a specific tumor marker for aggressive fibromatosis [24,29,30]. A representative formalin-fixed, paraffin-embedded tumor tissue section form each of the patients was subjected to immunohistochemical staining with commercially available biotinylated mouse monoclonal antibody to β-catenin (1/200 dilution; BD Transduction Laboratories, Lexington, Ky). Sections were deparaffinized and subjected to antigen retrieval procedures. Immunostaining was performed with an automated stainer (Ventanna, Tucson, Ariz). Antibody reaction was detected by the biotinavidin/immunoperoxidase amplification procedures with the appropriate reagents and methods recommended and supplied by the staining equipment supplier (Ventanna, Tucson, Ariz). Nuclear staining/expression of β-catenin was assessed by microscopy and recorded as percent positive-stained nuclei in the areas within the tissue that showed the highest staining intensities. The medical records of identified cases were closely reviewed to follow the progress and outcomes of each individual. The age, mode, and duration of presentation, site, and size of the lesion were all noted. Initial management (surgical or nonsurgical) was noted, as was any subsequent management if any, for residual or recurrent disease. The disease progression was monitored both clinically and radiologically. A literature review was performed using Medline. “Aggressive fibromatosis” and “desmoid tumors” were the keywords used for the search. The review included pediatric case series and series that included both adults and children.
1598
Table 1
Summary of presentation and diagnosis
Case
Sex
Age
Presentation
Pain
Duration of symptoms
Trauma
Associations (eg, Gardner)
Radiology
Diagnosis
1
Male
6 y 7 mo
Right jaw swelling
No
10 d
No
RubensteinTaybi
Intraoral Bx diagnostic
2
Female
1 y 9 mo
Right jaw swelling
No
3wk
No
No
3
Male
3 y 4 mo
Difficulty eating and speech
No
No
No
4
Male
3 y 3 mo
No
No
No
Male
12 y 9 mo
No
1 mo
No
No
6
Female
14 y 8 mo
Yes
No
Female
2 y 5 mo
3 mo, 1 wk, and 1 mo, respectively 3-mo recent facial palsy
No
7
Torticollis and pain, Horner syndrome, and neck mass Right facial swelling, VII N weakness
No
No
8 9
Male Female
1y 5 y 6 mo
Right neck mass Left jaw swelling, malocclusion
No No
few wk 1y
No No
No No
10
Male
1 y 10 mo
Right jaw swelling, increasing trismus
No
6 mo
Yes
No
CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated No CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated
Intraoral Bx diagnostic
5
Obstructive sleep apnea, dysphagia facial swelling Right jaw swelling
2 mo (recurrence at 2 mo) 3-4 mo
CT-nondiagnostic tumor extent demonstrated CT-nondiagnostic tumor extent demonstrated MRI demonstrated extent of recurrence
No
Core Bx nondiagnostic, intraoral biopsy diagnostic Initial excision Bx diagnostic
Intraoral Bx diagnostic
Open Bx diagnostic
Transoral Bx diagnostic
Excision Bx Intraoral Bx nondiagnostic
Intraoral Bx diagnostic
A. Sharma et al.
Summary of management and treatment outcome
Case Sex
Age
Region
β-catenin
Primary treatment
Residual disease
Positive margin Negative Positive margin Slightly positive (20%) Positive margin Negative
Adjuvant therapy
Progression after Recurrence Outcome primary treatment (follow-up)
No No
No No
No No
Good (2 y) Good (2 y)
No
No
Yes
Tracheotomy
Yes
N/A
No No Chemotherapy, Yes radiotherapy, hormonal therapy
No N/A
Recurrence (at 2 mo) treated with surgery; good (36 mo) Very poor mouth opening and continues with tracheotomy (14 y) Good (5 y) Morbidity from adjunct treatment; disease now stable (4 y)
No
No
No
Good (4 y)
1 2
Male 6 y 7 mo Right mandible Female 1 y 9 mo Right mandible
Surgery Surgery
3
Male
3 y 4 mo Floor of mouth
Surgery
4
Male
3 y 3 mo Left parapharyngeal Chemotherapy N/A up to skull base
5 6
Male 12 y 9 mo Right mandible Female 14 y 8 mo Left neck from hyoid to thoracic inlet
Surgery Surgery
7
Female 2 y 5 mo Right cheek infratemporal skull base Male 1y Right neck Female 5 y 6 mo Left mandible, skull base, pterygopalitine fossa Male 1 y 10 mo Right mandible, infratemporal fossa to hypopharynx
Surgery
Positive margin Negative Macroscopic Positive (25%) encasing carotid and brachial plexus Macroscopic Positive (30%)
Surgery Surgery
No Negative Positive margin Negative
No No
No No
No No
Good (6 mo) Good (17 y)
Surgery
Positive margin Negative
No
No
No
Good (18 y)
8 9
10
Slightly positive (10%)
Pediatric aggressive fibromatosis of the head and neck
Table 2
1599
1600
A. Sharma et al.
Fig. 1
Macroscopic appearance of tumor.
Studies that examined the management of intraabdominal aggressive fibromatosis were ruled out.
2. Results For a 20-year period, a total of 10 cases (6 male and 4 female) of aggressive fibromatosis of the head and neck were identified from a total of 340 cases of fibromatosis of the whole body, of which 95 were extraabdominal; that is, 10% of the extraabdominal fibromatosis occur within the head and neck region in our series. None were associated with Gardner's syndrome. The presentation and diagnosis of the individual cases are summarized in Table 1. The age at presentation ranged from 12 months to 14 years with a median age of 3 years. The duration of symptoms ranged from 1 week to a year with simple swelling as the most common symptom. Half of the cases presented as jaw swelling, 2 of which slowly presented for many months as trismus and malocclusion (cases 9 and 10) before obvious external swelling. The other cases presented with local swelling (case 8) and/or secondary symptoms to the swelling, for example, dysphagia (case 3) and obstructive sleep apnea (case 4). Only one case presented with pain and later developed a Horner syndrome (case 6); this tumor was subsequently found to have significant neurovascular involvement. Another case (case 7) presented with facial swelling and later a facial nerve weakness; the tumor was found to be involving the skull base. In both of these cases the neurologic signs were a recent onset phenomena at presentation, with initial symptoms present for several months. Only one patient (case 10) had any suggestion of trauma. None of the patients had a history of Gardner's or familial adenomatous polyposis, though one patient did have Rubinstein-Taybi syndrome (case 1). Nearly every patient had radiologic imaging as part of the diagnostic workup. The CT scan was used in most cases to demonstrate the tumor size and extent; one patient (case 3) had a magnetic resonance imaging (MRI) scan to diagnose the extent of recurrent disease before further management.
Radiology was not diagnostic of the tumor disease itself. One patient required no preoperative imaging because of the superficial clinical presentation (case 8). Tumor diagnosis was with open biopsy, either intraorally or externally depending in the site of the tumor. Case 2 had a core biopsy initially performed; this was nondiagnostic as a low-grade fibrohistocytic lesion before an open biopsy that confirmed the diagnosis of aggressive fibromatosis. After a pathologic diagnosis of aggressive fibromatosis, the management and treatment outcomes are summarized in Table 2. In total, 8 patients were treated with surgery alone, and all had good long-term outcomes. Seven of these patients had pathologic microscopic extension of the tumor to the resection margin. In 2 patients (cases 6 and 7), gross macroscopic tumor was left unresected. Macroscopic surgical resection was not possible in case 6 because of involvement of the carotid artery and brachial plexus. A large amount of the tumor (two thirds) was left unresected. The patient received adjuvant chemoradiotherapy for the residual tumor. The initial chemotherapy vinblastine and methotrexate was tolerated but with minor side effect of an extravasation burn. This course of chemotherapy appeared to be ineffective in controlling the residual disease progression. A second course of chemotherapy with vincristine and actinomycin was not tolerated because of side effects and stopped after only a few doses. Radiotherapy was commenced at the same time of the second course of chemotherapy and the tumor appeared to have a slight response. However, the radiotherapy did not cause significant tumor regression, rather a cessation of progressive disease. Within this patient, the tumor remained quiescent until 3 years postsurgery; at this time, it again appeared to enlarge, diagnosed on follow-up MRI scan. Tamoxifen and sulindac were started as therapy, initially inducing regression, and to date, the disease appears stable on this medication. In case 7, macroscopic surgical clearance was not possible because of involvement with the skull base. The patient was
Fig. 2
Microscopic appearance of tumor.
Pediatric aggressive fibromatosis of the head and neck
1601
closely observed radiologically and no adjuvant therapy given. The disease was seen to not progress and appeared to “burn out” over several years. At 1-year postoperatively, the residual contrast-enhancing tumor at the skull base was seen on CT scan; for the next 2 years, the follow-up scans revealed no contrast enhancement. Case 3 illustrates that tumors recurring aggressively, that is, within 2 months, can still be treated with surgical excision. Despite positive resection margins the child is recurrence-free 3 years postoperatively. One patient's guardian (case 4) opted for chemotherapy as a primary treatment. The initial course of chemotherapy with vincristine and actinomycin failed to prevent early tumor progression. The patient required emergency surgical intervention in the form of a tracheotomy. A second course of chemotherapy with vinblastine and methotrexate did appear to prevent tumor progression but in the long term did not appear to induce tumor regression. Chemotherapy appeared to be unsuccessful as a primary treatment in this case. All patients were followed up with either CT or MRI scan on a regular basis—the exception was case 8, who was discharged at 6 months postoperatively. The pathologic finding demonstrated complete resection of the tumor with a clear resection margin. Clinically, the original tumor was also very superficial, and follow-up was under the patient's local physician. As well as regular radiologic follow-up, cases 9 and 10 underwent significant reconstructive surgery after their original surgical procedure. During these procedures, the original tumor site was examined clinically and also rebiopsied, with no evidence of recurrent disease, both at around 2 years original postsurgery. Case 7 also, during a routine tonsillectomy 3 years postoperatively, had a biopsy taken of scar tissue in the tonsillar fossa, which was pathologically negative for recurrence of fibromatosis.
patients with extension of disease to the tumor margin had any clinical or radiologic evidence of tumor progression or recurrence. In the 2 patients with gross macroscopic residual disease; one patient underwent regression, that is, 50% reduction; the other significant progression despite adjuvant chemotherapy.
3. Pathologic results 3.1. Typical appearance of tumors The macroscopic appearance of fibromatosis is often indistinguishable from scar tissue, having a white fibrous appearance (Fig. 1). The tumor often interdigitates with muscle fibers, making surgical excision difficult and often requiring excision of muscle with the specimen. Microscopically, the tumor showed intersecting broad sweeping bands of spindle cells within a collagenous stroma. These spindle cells are uniform in shape and size and had cytological appearance of fibroblasts (Fig. 2).
3.2. Tumor resection margin and residual disease Only one case had a negative pathologic tumor margin (case 8) of the 9 patients treated surgically. None of the
3.3. Local invasion and metastasis In all patients where enlarged lymph nodes were taken as specimens with the tumor, pathologically these lymph nodes had negative result for fibromatosis. Where local salivary glands were taken, there was also no microscopic disease invasion of the salivary gland tissue.
3.4. Nuclear b-catenin expression By immunohistochemistry, 6 (60%) of 10 patients had negative results for nuclear β-catenin staining, and 2 cases showed 25% to 30% positive nuclear expression and a further two showed less than 25% positive nuclear staining, which is considered by some to be a negative result [29]. However, by following the criteria of at least focal nuclear staining in every studied sample of fibromatosis [24], 4 (40%) of 10 patients were reported as positive in this study.
4. Discussion 4.1. Incidence and etiology of pediatric fibromatosis The age of incidence of pediatric fibromatosis has been reported with an average age around 8 years old [11]. Our experience is that possibly 2 groups of incidence may occur; most pediatric cases occur before 6 years of age and an older group later in childhood, which is in keeping with previously described models [6]. It has been suggested that aggressive fibromatosis of the head and neck region occurs in a younger age group than other sites [16,31]. It is interesting that in our series, the case of progressive disease occurred in an older female child; in the literature, it has previously been suggested that the disease in adolescents is more aggressive or more likely to progress in young female adults [6]. Within our series, we have a male preponderance, in keeping with recent literature reviews [11,32,33], though other nonpediatric studies have found a female preponderance [8,34], possibly skewed by the fertile female group [6], whereas other studies suggest no sex preponderance in pediatric aggressive fibromatosis [17]. None of the patients with our series had any association with Gardner's syndrome, an autosomal disease characterized by the presence of colonic polyposis, osteomas, and a multitude of soft tissue tumors as described in the literature
1602 [6,17,35,36]. This is similar to the experience of similar extraabdominal pediatric aggressive fibromatosis series [15,32], suggesting less relevance of this association within this group. One of the children in our series (case 1) had Rubinstein-Taybi syndrome [37], which is a genetic multisystem disorder characterized by facial abnormalities, broad thumbs and great toes, and mental retardation. Rubinstein-Taybi syndrome is associated with soft tissue tumors [38]. However, no previous association of fibromatosis and Rubenstein-Taybi syndrome has previously been reported in the literature. It is difficult from a single case to say whether a rare disease can be associated with a rare syndrome, though the syndrome's link with soft tissue tumors may suggest a true association of the disease. Trauma has been suggested as a possible etiologic factor [15-17], but only 1 of 10 in our series (case 10) had possible trauma to the area; indeed, it has been suggested that trauma may not be so relevant [17].
4.2. Diagnosis of fibromatosis, the role of nuclear b-catenin expression Pathologic diagnosis of the condition of aggressive fibromatosis in children still appears to be simple histologic diagnosis. Nuclear β-catenin does not appear to be an accurate marker in the diagnosis of aggressive fibromatosis in children. This is probably related to pediatric fibromatosis that does not appear to have as strong an association with familial adenomatous polyposis (FAP) or Gardner's syndrome as its adult form. Nuclear β-catenin accumulation/expression is produced by the APC gene mutation associated with FAP and may therefore be much less relevant in diagnosis of pediatric fibromatosis as found in our series. It is also of note that the pathologic series identifying β-catenin as a tumor marker for aggressive fibromatosis included only 2 pediatric or adolescent patients [24,30]. The pathophysiology of pediatric aggressive fibromatosis becomes less clear than in the adult form, with the lesser association with the APC gene or the activation of the wnt signaling pathway or somatic mutation that affects phosphorylation sites within exon 3 of β-catenin, causing protein stabilization that resulted in nuclear β-catenin accumulation/expression as seen in adult colonic cancer [39]. Although we did not perform sequence analyses of exon 3 to rule out somatic mutation as a cause of nuclear β-catenin accumulation/expression, in a study of palmar fibromatosis in adults and one infantile digital fibromatosis that showed nuclear βcatenin expression, the exon 3 region were found to be nonmutated [40]. Thus, it appears that somatic mutation in exon 3 of β-catenin is uncommon in fibromatosis [24,40]. It may be that the pediatric form is a similar but distinct pathologic condition, and literature based on
A. Sharma et al. treatment of adult series of aggressive fibromatosis becomes much less relevant.
4.3. Relevance of tumor margin and management outcome It is suggested that surgical resection with a wide clearance margin, free of macroscopic, and microscopic disease is the treatment of choice [31]; and the use of adjuvant therapies is recommended when the tumor is unresectable because of vital structures or surgery has failed to obtain surgical clearance or [11,31]. The goal of complete macroscopic and microscopic surgical clearance is difficult, as in the authors experience; these tumors tend to interdigitate with the muscle fibers [41] and may be difficult to distinguish from scar tissue in recurrent disease. Microscopic residual disease is a significant risk factor for recurrence of tumor [17]. Most recurrences occur within 2 years [33,36] to 3 years and nearly all within 6 years of presentation [31]. The primary consideration for treatment is regarded as prevention of recurrence [42], and the evidence in the literature appears to be based on treatment failure as recurrence of tumor [1,8,11,17,31,43-45] and not necessarily the progression of the disease. Aggressive fibromatosis has no features of anaplasia and does not seem to show classical neoplastic progression; if kept under control, there might be a high probability that eventually the tumor will stop growing and regress [17]. Aggressive fibromatosis in the head and neck is said to have a higher incidence of recurrent disease [8], although this may reflect that excision is more technically difficult in this region [41] and more likely to leave residual disease. It has been suggested also that residual disease after surgical resection is a significant risk factor of recurrent disease, with positive resection margins after surgery indicating high risk for disease recurrence [11,17]. This does not appear to be our experience. Aggressive fibromatosis can be a locally destructive disease, invading local structures, particularly relevant in the head and neck region that contains many vital structures. The risk of morbidity of pediatric aggressive fibromatosis therefore is higher within this region. In a review of pediatric series containing 114 cases, there were only 2 deaths from invasion both within the head and neck [17]. Aggressive management with adjuvant therapy is therefore advocated in the presence of any residual disease by some authors [17]. However, this panacea of management does not take into account that residual disease may indeed not progress but remain quiescent and indeed in time regress, as demonstrated in case 7 of our series of patients. In aggressive fibromatosis of the limbs, it has been shown that aggressive management surgically to achieve negative margins may not be necessary and may result in unnecessary morbidity [46]. This may also
Pediatric aggressive fibromatosis of the head and neck
1603
be the case for adjuvant therapies such as chemotherapy or radiotherapy that are not without their own iatrogenic morbidity, of especial consideration in the child. Chemotherapy, radiotherapy, and drug therapies including hormonal and NSAIDs have all been suggested as alternatives to surgery or as adjuvant therapy to primary surgery in the presence of microscopic residual disease or extension to the margins. In adults, the standard approach for microscopic residual disease is adjuvant radiotherapy to achieve a high local control rate [11,17,45,47,48]. A dose of 50 to 60 Gy is thought to be efficacious [45]. However, doses higher than 50 Gy required for tumor control may lead to skin ulceration, poor wound healing, and local tissue necrosis [32,33]; in pediatric patients in particular, inherent risks for growth problems and long-term risk of secondary malignancy [11,45] are of concern. Radiotherapy appears to be less effective treatment of aggressive fibromatosis in children [36]. Treatment failure to control disease progress from primary radiotherapy and then requiring salvage curative therapy by surgery has been reported [49]. Chemotherapy may be a reasonable alternative to radiotherapy in a growing child [32], though there is an inherent risk of secondary malignancies, fertility problems, cardiotoxicity [11,50], and neutropenia [51]. Chemotherapy has been considered for delay of progression of aggressive fibromatosis until children are considered old enough for radiotherapy [43] or surgery [32]. There is no strong evidence that chemotherapy agents are efficacious, and possible role may be with a rapid growing tumor to obtain local control [17]. It still remains unclear from the literature as to which cytotoxic chemotherapeutic agents are most effective or indeed what combination [50], and this is indeed an area for future research [11,45]. Vinblastine and methotrexate in low does has been shown to control progression of pediatric aggressive fibromatosis [32,52]. It has been suggested recently that vinblastine and methotrexate may induce tumor regression [51]; however, this study did not contain a control group and may simply reflect the natural progression within the population. In our series, it was only partially successful with case 4 and unsuccessful in case 6. Although successful treatment with vincristine and actinomycin has been reported [5], it was also not the case in our limited experience. Concerns over potential morbidity of cytotoxic agents have led to interest in noncytotoxic drugs such as tamoxifen and NSAIDs. Regression of desmoid tumors has been described at menarche and menopause [6,21] suggesting a hormonal influence. Tamoxifen is an antiestrogen and has been found to have an inhibitory effect on desmoid tumors, and estrogen receptors have been found in aggressive fibromatosis tumors [11]. Nonsteroidal antiinflammatory drugs are thought to stimulate the immune response, thus impairing the proliferation of tumor cells [42] and may well have a role. Tamoxifen and sulindac (NSAID) have been used with success in patients with FAP [53], though FAP is not associated with our patients in this series. Vitamin D has
also been suggested as a treatment by its metabolites inhibiting proliferation of fibroblasts [54]. Although the role of adjuvant therapies remains uncertain, surgery can be considered as a primary treatment of recurrent disease in aggressive fibromatosis case 3 or in management of progressive disease case 4. It has been shown that the recurrence rate after a second or third resection is equal to or lower than the recurrence rate after the primary surgery [6,46]. It has been suggested that if residual aggressive fibromatosis may be kept under control, that is, treating progression, there may be a high probability that eventually it will stop growing and even regress [55], as demonstrated in case 7. Within our series, the presence of a positive tumor margin does not appear to be a predictor of disease progression with none of the macroscopically cleared disease but positive resection margin tumors undergoing disease progress.
4.4. Disease monitoring and follow-up We recommend that in the presence of residual disease, in most of our cases, the patient may simply be observed using MRI or CT scan. Our experience and review of the literature suggests that more aggressive disease can reoccur or actively progress within the first few months. Therefore, we suggest a routine scan 3 monthly for first 6 months, then 6 monthly until 2 years, then yearly review until the disease is stable. Should progression of residual tumor or tumor recurrence occur, indicating active disease, surgical resection may still be considered although adjuvant chemotherapy, radiotherapy, or hormonal therapy should also be considered within the context of the morbidity of the disease versus the morbidity of the treatment.
References [1] Greenberg HM, et al. Radiation therapy for the treatment of benign locally aggressive fibromatosis. Int J Radiat Oncol Biol Phys 1980; 6(10):1412. [2] Enzinger FM, Shiraki M. Musculo-aponeurotic fibromatosis of shoulder girdle (extra abdominal desmoid)—analysis of 30 cases followed up for more years. Cancer 1967;20(7):1131-40. [3] Fletcher CDM, Krishnan K, Merlens F. World Health Organization: tumours of soft tissue and bone. Lyon: WHO; 2002. [4] Weiss SW, Goldblum JR. Enzinger and Weiss's soft tissue tumours. 4th ed. Mosby: St Louis; 2001. [5] Stein R. Chemotherapeutic response in fibromatosis of neck. J Pediatr 1977;90(3):482-3. [6] Reitamo JJ, Schenin TM, Hayry P. The desmoid syndrome: new aspects in the cause, pathogenesis and the treatment of the desmoid tumor. Am J Surg 1986;151:230-7. [7] Caglar K, et al. Effective treatment of multifocal aggressive fibromatosis with low-dose chemotherapy. Turk J Pediatr 2006;48(4): 365-8. [8] Masson JK, Soule EH. Desmoid tumors of the head and neck. Am J Surg 1966;112:615-22. [9] West CB, Shagets FW, Mansfield MJ. Nonsurgical treatment of aggressive fibromatosis in the head and neck. Otolaryngol Head Neck Surg 1989;101(3):338-43.
1604 [10] Wilkins SA, et al. Aggressive fibromatosis of fibromatosis of the head and neck. Am J Surg 1975;130(4):412-5. [11] Buitendijk S, et al. Pediatric aggressive fibromatosis—a retrospective analysis of 13 patients and review of the literature. Cancer 2005;104(5):1090-9. [12] Allen PW. The fibromatosis: a clinicopathological classification based on 140 cases. Am J Surg Pathol 1977:255-60. [13] Tejpar S, Nollet F, Li C. Predominance of beta-catenin mutations and beta-catenin dysregulation in sporadic aggressive fibromatosis (desmoid tumor). Oncogene 1999;18:6615-20. [14] Skubitz KM, Skubitz APN. Gene expression in aggressive fibromatosis. J Lab Clin Med 2004;143(2):89-98. [15] Rao BN, et al. Challenge in the treatment of children fibromatosis. Arch Surg 1987;122:1987. [16] Ayala AG, et al. Desmoid fibromatosis—a clinicopathological study of 25 children. Semin Diagn Pathol 1986;3(2):138-50. [17] Faulkner LB, et al. Pediatric desmoid tumor: restrospective analysis of 63 cases. J Clin Oncol 1995;13(11):2813-8. [18] Penick RM. Desmoid tumors developing in operative scars. Int Surg Dig 1937;23:323-9. [19] McKinnon JG, Neifield JP, Kay S. Management of desmoid tumors. Surg Gyecol Obstet 1989;169:104-6. [20] Janinis J, et al. The pharmacological treatment of aggressive fibromatosis: a systematic review. Ann Oncol 2003;14(2):181-90. [21] Dahn I, Jonsson N, Lundh G. Desmoid tumors: a series of 33 cases. Acta Chir Scand 1963(126):305-14. [22] Alman BA, et al. Aggressive fibromatosis (desmoid tumor) is a monoclonal disorder. Am J Surg Pathol 1996:194-200. [23] Alman BA, et al. Aggressive fibromatosis. J Pediatr Orthop 1992;12:1-10. [24] Bhattacharya B, et al. Nuclear beta-catenin expression distinguishes deep fibromatosis from other benign and malignant fibroblastic and myofibroblastic lesions. Am J Surg Pathol 2005;29(5):653-9. [25] Alman BA, et al. Increased beta catenin and somatic APC mutations in sporadic aggressive fibromatosis. Am J Pathol 1997;151:329-34. [26] Muller E, et al. Molecular genetic and immunohistochemical analysis of the tumor suppressor genes Rb and p53 in palmar and aggressive fibromatosis. Am J Surg Pathol 1996:194-200. [27] Jenkins N, Freeman LS, McKibbin B. Spontaneous regression of a desmoid tumour. J Bone Joint Surg 1986;68(5):780-1. [28] Dormans JP, Spiegel DA, Mayer J. Fibromatosis in childhood: the desmoid/fibromatosis complex. Blood Cancer 2001;37:126-31. [29] Ng TL, et al. Nuclear beta-catenin in mesenchymal tumors. Mod Pathol 2005;18(1):68-74. [30] Hauben EI, et al. Desmoplastic fibroma of bone: an immunohistochemical study including b-catenin expression and mutational analysis for b-catenin. Hum Pathol 2005;36(9):1025-30. [31] Spiegel DA, et al. Aggressive fibromatosis in infancy to adolescence. J Pediatr Orthop 1999;19(6):776. [32] Skapek SX, et al. Combination chemotherapy using vinblastine and methotrexate for the treatment of progressive desmoid tumor in children. J Clin Oncol 1998;16(9):3021-7.
A. Sharma et al. [33] Perez-Cruet MJ, et al. Aggressive fibromatosis involving the cranial base in children. Neurosurgery 1998;43(5):1096-102. [34] Janahi WM, et al. Infantile fibromatosis. J Laryngol Otol 1999;113(3): 246-9. [35] Carr RJ, et al. Infantile fibromatosis with involvement of the mandible. Br J Oral Maxillofac Surg 1992;30(4):257-62. [36] Merchant TE, et al. Long-term results with radiation therapy for pediatric desmoid tumors. Int J Radiat Oncol Biol Phys 2000;47(5): 1267-71. [37] Rakheja D, et al. Immunohistochemical expression of beta-catenin in solitary fibrous tumors. Arch Pathol Lab Med 2005;129(6):776-9. [38] Iyer NG, Özdag H, Caldas C. p300/CBP and cancer. Oncogene 2004;23(24):4225. [39] Segditsas S, Tomlinson I. Colorectal cancer and genetic alterations in the Wnt pathway. Oncogene 2006;25(57):7531-7. [40] Montgomery E, et al. Superficial fibromatosis are genetically distinct from deep fibromatosis. Mod Pathol 2001;14(7):695-701. [41] Saunders KW, et al. Aggressive fibromatosis of the parapharyngeal space: two cases and treatment recommendations. Ear Nose Throat J 2004:262-8. [42] Yousry ES. Fibromatosis of the head and neck. J Laryngol Otol 1992;106:459-62. [43] Tostevin PMJ, Wyatt M, Hosni A. Six cases of fibromatosis of the head and neck in children. Int J Pediatr Otorhinolaryngol 2000;53(3): 235-44. [44] Siegel NS, Bradford CR. Fibromatosis of the head and neck: a challenging lesion. Otolaryngol Head Neck Surg 2000;123(3):269-75. [45] Spiegel DA, et al. Aggressive fibromatosis from infancy to adolescence. J Pediatr Orthop 1999;19(6):776-84. [46] Lewis JJ, et al. The enigma of desmoid tumors. Ann Surg 1999;229(6): 866-72. [47] 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(3):882-91. [48] Shin KH, et al. The role of radiotherapy in the treatment of aggressive fibromatosis. Yonsei Med J 1999;40(5):439-43. [49] Watzinger F, et al. Aggressive fibromatosis of the mandible: a case report. Int J Oral Maxillofac Surg 2005;34:211-3. [50] Goepfert H, et al. Chemotherapy of locally aggressive head and neck tumors in the pediatric age group—desmoid fibromatosis and nasopharyngeal angiofibroma. Am J Surg 1982;144(4):437-44. [51] Skapek SX, et al. Vinblastine and methotrexate for desmoid fibromatosis in children: results of a Pediatric Oncology Group phase II trial. J Clin Oncol 2007;25(5):501-6. [52] Reich S. Low-dose chemotherapy with vinblastine and methotrexate in childhood desmoid tumors. J Clin Oncol 1999;17(3):1086. [53] Hansmann A, et al. High-dose tamoxifen and sulindac as first-line treatment for desmoid tumors. Cancer 2004;100:612-20. [54] Ferah Y, et al. Possible therapeutic role of vitamin D-3 in aggressive fibromatosis. Jpn J Clin Oncol 2004;34(8):472-5. [55] Rock MG, et al. Extra-abdominal desmoid tumors. J Bone Joint Surg 1984:1369-74.