Radiation therapy for aggressive fibromatosis (desmoid tumors): Results of a national Patterns of Care Study

Oliver Micke, IJROBP 2005;61: 882

A total of 101 institutions returned the questionnaire: 52.7% reported satisfactory clinical data and experience for inclusion in this analysis. A total accrual rate of 278 patients per year was reported with median number of 2 cases (1–7 cases) per institution. Satisfactory data for a long-term clinical evaluation was reported for 345 patients from 19 different institutions. The applied total doses ranged between 36 and 65 Gy (median, 60 Gy). The local control rate was 81.4% in primary RT for unresectable tumors and 79.6% in postoperative RT. No acute or late radiation toxicities > Grade 2 (RTOG) were observed. No clear dose–response relationship could be established, but there was a tendency toward a lower local control rate in patients with a higher number of operative procedures before RT and patients treated for recurrent aggressive fibromatosis.

Desmoids are rare benign neoplasms of connective tissue that arise from deep muscle fascia, aponeuroses, tendons, and scar tissue . The Anglo-American literature describes them as “aggressive fibromatosis”. The disease has been known since the 19th century: The terms “desmoid” and “desmoidfibrom” were coined by the German physiologist and comparative anatomist Johannes Müller (1801–1858) from Berlin in 1838. The name is derived from the ancient Greek and indicates a bandlike appearance. In 1832 the surgeon John McFarlane from Glasgow was the first to mention the entity in the English-speaking literature. Thus neoplasms in the abdominal walls of women who recently had given birth were described. Later similar tumors were found in other anatomic regions. The term “fibromatosis” was introduced for the first time by Arthur Purdy Stout from New York. He stated that desmoids are masses of proliferated, scarlike tissue, which despite the fact of tissue infiltration rarely recur (which is not true today!) and that do not become true malignant neoplasms.

Desmoids are very rare benign lesions representing less than 3% of all soft tissue tumors and 0.03% of all neoplasms. The incidence is 2–4 cases per million per year. Thus, the estimated frequency of desmoid tumors in the United States ranges from 500–1000 cases per year, and in Germany between 100 and 200 per year.

There is a female predominance (male:female ratio, 1:1.5–2.5). The highest incidence arises among patients in the third or fourth decade of life, but desmoids can occur in infants and the elderly as well.

Although the etiology of desmoids is poorly defined, numerous factors are acknowledged to be strongly associated with their development. Endocrine and physical factors seem to play an important role in the development of the disease . An antecedent history of trauma to the site of the tumor, often surgical in nature, may be elicited in approximately 25% of cases. In very rare cases desmoids can arise after surgery in scar tissue. Desmoid tumors developing around silicon or metal implants have also been described. However, so far no association between hip prostheses and desmoid tumors has been published in the world literature.

Intraabdominal and extraabdominal desmoids are significantly associated with familial adenomatous polyposis of the colon (FAP) and have an even more pronounced association with the subgroup of Gardner syndrome, an autosomal dominant hereditary disease

Clinically, desmoids show a slow, but locally aggressive growth pattern with a strong tendency to reappear after resection Some tumors display low-grade malignancy or spontaneous suspension of growth. Others progress to giant masses, leading to secondary clinical symptoms and complaints. Partial or temporary spontaneous remissions are possible. Some desmoids infiltrate organs or major vessels, leading to fatal outcome; the overall death rate is about 8% . Metastatic disease is always a rarity, but multifocal growth (e.g., on one extremity) instead has been described. In particular, in critical locations the differentiation into benign/malignant is inconsistent, e.g., in the mesenteric root.

Differentiation depending on location (extraabdominal, 70%; intraabdominal, 10%; confinement to the abdominal wall, 20%) can be of prognostic value. Extraabdominal manifestations also seem to have a higher recurrence rate, even after complete (in-sano) resection

The rarity of these tumors coupled with the variability in their clinical course does not allow the demonstration of the efficacy and, even more, the superiority of any specific intervention

The primary therapeutic approach is always radical surgical resection with wide safety margins (2–3 cm). Because of the indistinct tumor borders, however, “true” R0 resections are often very difficult to achieve

There has been a long tradition of using radiotherapy (RT) to achieve local control in primary or postoperative treatment. James Ewing  was the first to propose RT as a treatment modality for the management of desmoids not amenable to surgical resection. In 1928, he described that these lesions respond “slowly but satisfactorily” to RT.

In the next decades his observations were confirmed by several clinical studies . However, the role of RT in the management of desmoids is still not clearly defined today . Therefore, the analysis of larger collectives followed by the implementation of treatment guidelines and suitable tools for quality assurance is crucial for the further promotion of this treatment.

Outcome analysis in the context of PCS is an important tool to set up a national benchmark on treatment outcome that should be expected from RT in a specific disorder. So far this national PCS has collected the largest number of cases (345 patients from 34 institutions) ever reported on the use of RT for aggressive fibromatosis.

Our clinical data showed a long-term local control of 79.6% in postoperative RT and a slightly higher local control rate of more than 80% in unresectable tumors.

A review of recent literature on the management of aggressive fibromatosis revealed a wide range of treatment outcomes and heterogeneous treatment strategies. After surgery alone, local recurrence rates ranged between 20% and 80% .  Microscopic residual disease does not necessarily impair local control rates. One may well speculate that the recurrence rate depends highly on inherent characteristics of the disease, which might be more or less aggressive on its own, so that it might recur or not rather independently of the fine quality of surgery). It is recommended that high-risk surgical procedures to achieve negative margins be restricted to very special circumstances

Postoperative irradiation can reduce the recurrence rate significantly. Walther described a reduction from 80% to 10% with doses of at least 50 Gy. Other reports observed relapse rates of 5–40%, depending on patient selection

Obviously, aggressive fibromatosis is not a radioresistant disease. Even radiotherapy alone is able to achieve a high local control rate, comparable to the results of postoperative irradiation. Leibel reported a local control rate of 68% during a median follow-up of 8 years for 19 patients, who received doses between 50 Gy and 55 Gy. 

In the recent well-conducted meta-analysis of Nuyttens comprising 22 studies, a local control rate of 78% was shown, which is very similar to that achieved in our PCS. To achieve high local control rates with radiotherapy prompts the question whether this treatment approach can replace large and mutilating surgical interventions.

Other systemic therapeutic approaches such as hormonal therapies, chemotherapy, or immunotherapy are still lacking systematic evaluation. There have been several attempts of systemic therapy in primary and adjuvant settings at a high cost level and diverse response rates. A few reports deal with systemic hormonal therapy in aggressive fibromatosis, where the most commonly used antiestrogen is tamoxifen. Most reports, however, are single-case reports that have demonstrated either some response or disease stabilization during therapy. As large series are not available, no firm conclusion is possible regarding the effectiveness of tamoxifen. Nonsteroidal anti-inflammatory agents have an uncertain therapeutic impact and the precise mechanism remains to be investigated. The role of systemic chemotherapy is not yet clearly defined, including the use of alkylating agents with inconsistent results. No convincing results exist from conventional antineoplastic drugs, possibly owing to the low mitotic index exhibited by aggressive fibromatosis, although antisarcoma therapy primarily presented some promising results in small series Weiss and Lackman  found little effectiveness of vincristine and methotrexate. Generally, systemic treatment approaches should be reserved for selected cases

All participating institutions observed a low rate (27.8% < Grade II) of radiogenic side effects, which were mostly mild to moderate. Similarly, the complications reported in the literature were moderate (edema, pain paresthesias, stiffness, and weakness), caused no major disability, and were sometimes required to avoid an amputation. So far, in this PCS no radiation-associated malignancies were reported with a median follow-up of 43 months and a maximum of 306 months, while sporadic reports on secondary malignancies exist

Overall, the majority of German radiation oncologists in this national survey judged RT to be a worthwhile and useful treatment indication for aggressive fibromatosis. So far, this study comprises the largest database of cases reported for RT in aggressive fibromatosis. Radiotherapy achieves a high local control rate in the postoperative setting as well as in primary treatment of unresectable tumors. Despite some variations of the national RT praxis, the study presents the largest number of aggressive fibromatosis patients effectively treated by external beam RT.