Desmoid tumors (also called desmoid fibromatosis) are benign fibrous growths that occur rarely in the general population (5 to 6 per 1 million per year) but frequently in one of the familial cancer predisposition conditions known as familial adenomatous polyposis (FAP) or Gardner syndrome, affecting between 3.6% and 20% of patients. About 2% of all desmoids arise in patients with FAP. Desmoids may occur in any musculoaponeurotic tissue structures of the body, although they tend to be in extremities and spinal areas in the general population and in the abdomen in FAP.

Desmoid tumors are not considered sarcomas; however, they have been classified as aggressive fibromatosis, and the optimal treatment plan usually requires multidisciplinary teams only found in sarcoma centers. Surgeons with expertise in sarcomas should guide the approach to extra-abdomnial lesions, while intra-abdominal desmoids are best considered by teams including sarcoma surgeons together with GI surgeons, oncologists and radiation therapists all with desmoids experience and expertise. 

Desmoids do not metastasize but may infiltrate adjacent structures, extend along fascial planes, attach to and erode bones, and engulf and compress blood vessels, nerves, ureters, and other hollow organs of the abdomen. Severe and even fatal clinical problems are sometimes caused by these tumors, especially if mesenteric vessels or other abdominal organs are obstructed. Other complications include bowel perforation, fistulization, bleeding and ureteral obstruction. Desmoid tumors usually enlarge very gradually and sometimes stop growing altogether. Only about one third of abdominal desmoids cause pain, although the most common symptom is abdominal pain. Intra-abdominal desmoids sometimes become massive, occupying much of the abdominal cavity and encasing many segments of viscera.

There are a number of issues involved in the management of desmoid tumors, and the complexities of recognition and management have led to a designated session devoted to fibromatosis at the next CTOS meeting in London (November, 2008). In addition to fibromatosis there are 3 other designated sessions; Neurofibromatosis Type 1: MPNST and other sarcomas; Gynecological sarcomas; and finally, Sarcomas: impact of age. The goal of these sessions, as it arose from discussions of the CTOS Board of Directors led by Ian Judson, is to highlight the multidisciplinary nature of sarcoma care. In addition to the usual sarcoma team, desmoid tumor care optimally involves gastroenterologists and geneticists.

Desmoid tumors may be the first manifestation of FAP in some patients and families. Furthermore, some families with APC mutations exhibit desmoids as their only disease manifestation. A distinctive histopathology has been demonstrated in desmoid tumors of young FAP patients, whose first manifestation of the condition was often this lesion. Desmoids in FAP are monoclonal growths of hyperproliferative fibroblastic cells. Excess cell growth appears to occur on the basis of APC gene inactivation and accumulation of ß-catenin in the cells. The relationship of desmoids in FAP to sporadic cases is uncertain in view of the distinctive early histopathology and the observation that APC gene mutations are uncommon in sporadic desmoids.

The relative risk for desmoids in FAP patients compared with the general population is 825. The peak incidence of desmoids in FAP is between 28 and 31 years, although they may occur at any age. Independent predictors for desmoids in FAP include an APC mutation 3' of codon 1444, family history of desmoids, female gender, and the presence of osteomas. Desmoids most commonly occur following surgery after an average of 4.6 years. These growths are one of the most common causes of mortality in FAP patients who have had prophylactic colectomy, with a 10% to 50% mortality rate from the tumor in those with desmoids. The 10-year survival rate is 63%.

CT and magnetic resonance imaging are equally effective in following desmoid tumors. Treatment is undertaken for symptoms, cosmetic issues, functional disruption or imminent risk to adjacent structures. Extra-abdominal and abdominal wall desmoids are best treated by surgery with adequate resection of margins. Fewer than half of tumors so treated recur, and further therapy with radiation and repeat excision is usually successful. Surgery is difficult, and sometimes impossible, in cases of intra-abdominal desmoids but remains as an important option in selected cases.

Because intra-abdominal desmoids often involve the mesentery or encase vessels or organs, medical therapies are often first attempted. The most consistent initial therapies for abdominal desmoids appear to be the NSAID sulindac and the antiestrogen tamoxifen. Both give a better than 50% response after 3 to 6 months of therapy for FAP. The response may be even better with combination of the two. Combination chemotherapy is effective and can be offered for unresectable cases that fail to respond to this therapy. A combination of low-dose methotrexate and vinblastine demonstrated stabilization of tumor growth or shrinkage in 60% of patients with sporadic desmoids at various stages. Tyrosine kinase inhibition has also shown response. Radiation therapy has also shown success in a number of patients, even those with intra-abdominal tumors. A multi-disciplinary approach is highly recommended for difficult desmoids, especially those that are intra-abdominal.

The diagnosis of desmoid tumor, or aggressive fibromatosis, should alert members of the sarcoma team to the potential underlying diagnosis of FAP. This recognition is imperative to the recruitment of gastroenterologists and geneticists who can provide a more focused approach to polyposis screening and genetic testing, which could benefit the extended family as well as the patient.

Sporadic aggressive fibromatosis has a high tendency for local recurrence, even after apparently adequate resection. In institutional retrospective studies, local failure rates range between 25-60% at 5 years. This wide range shows the great variability of accrual, treatments and follow-up in such a rare disease which has never been investigated in a controlled randomized study. Traditionally, patients undergo standard surgery approaches with the primary goal always being complete resection with negative margins, as indicated for sarcomas.

However, differences in the aggressive nature of these tumours exist, and surgical indications change over time. During the last 10 years, the following concepts evolved from individual series. In 1998, the standard treatment was primary resection with negative margins when feasible, and exclusive radiotherapy was proposed in situations where surgery would result in major functional or cosmetic defects.¹ After that, authors² advocated function-sparing operations, and concluded that attempting to achieve the goal of negative margins could result in unnecessary morbidity. Thus, function-sparing surgery became a "reasonable" choice when feasible without leaving macroscopic residual disease. Medical observation alone was first proposed for recurrent but stable lesions.³ After that, an initial period of expectant observation was proposed in case of irresectable primary tumour.⁴

In 2007, neoadjuvant treatments were introduced since they could be associated with improved patient outcome.⁵ Some tumours respond to chemotherapy⁶ or other systemic treatments, thereby avoiding the functional consequences of surgery. More recently, Imatinib has been tested on advanced aggressive fibromatosis.⁷

Presently, authors address the question whether surgery and other aggressive treatments should systematically be part of first-line treatment.⁸ In this last study, a subset of patients with extra abdominal primary fibromatosis was managed with a systematic, non-aggressive protocol, and growth arrest occurred in 2/3 of non-operated patients. This "non-aggressive" protocol was applied on a larger multi-institutional series of patients⁹ and its implementation avoided aggressive surgery or radiotherapy on the majority of primary tumours. Discrepancies between the results of the impact of the quality of surgery suggest that additional factors may influence the natural history of these tumours. Among these factors, intrinsic biological characteristics of tumour cells and the host microenvironment could account for highly diverse outcomes. The subgroup of progressive or recurrent tumours might have the same aggressive biological characteristics, whatever their treatment.

Molecular determinants of fibromatosis recurrence or progression remain obscure. Recently,¹⁰ the prevalence of the mutations of the gene encoding beta-catenin was evaluated in a large cohort of sporadic desmoids (180 patients). Mutations were observed in 85% of the specimens. Four mutations in two codons of exon 3 were identified: T41A (41%), S45F (36%), and S45P (4%), AA 45 (3%). In another retrospective study,¹¹ five-year recurrence-free survival was significantly poorer in S45F-mutated desmoids (23%, P < 0.0001) versus either T41A (57%) or non-mutated tumors (65%). Nuclear beta-catenin expression was observed in 98% of specimens and intensity was inversely correlated with incidence of desmoid recurrence (P < 0.01).

Clearly, we need prospective studies to understand the molecular mechanisms behind desmoid tumorigenesis and progression, and to confirm whether some mutations are at particular risk for recurrence or progression. The final objective is to use molecular findings of tumours to individualize the selection of management protocols. This application of personalized medicine in fibromatosis could lead to the right treatment at the right time for the optimal outcome.