Desmoid tumours (DT), also known as fibromatosis (aggressive, deep or desmoid-type), are a rare and unusual soft tissue neoplasm. Desmoid tumours result from monoclonal proliferation of myofibroblastic tissue which tends to infiltrate and recur locally, but never metastasise.^1,2 Despite their classification as a benign neoplasm, their capacity for local invasion may cause significant morbidity and even death. Therefore, appropriate and timeous treatment is essential. McFarlane first described the condition in 1832³ and the term ‘desmoids’ (from the Greek ‘desmos’, meaning band- or tendon-like) was coined by Müller in 1838.⁴ Desmoid tumours account for 0.03% of all neoplasms and 3% of soft tissue tumours^5,6 and have an estimated incidence of 2.4–4.3 per million people per year in the general adult population.¹ They commonly originate from deep musculo-aponeurotic structures, but also develop intra-abdominally.²

Desmoid tumours can occur sporadically (around 85%⁷) or in association with familial adenomatous polyposis (FAP),¹ the latter combination being termed Gardener’s syndrome.⁸ Pregnancy is an associated condition (either during or following a pregnancy), suggesting high oestrogen states as contributory.⁵ The association with antecedent trauma or previous surgery⁹ may implicate a dysregulated wound healing process in the pathogenesis of this condition.⁸

Treatment of DT is complicated by the heterogeneity of the condition with regard to natural history, location and symptomatology. Surgery aims to completely excise the tumour with limited functional or cosmetic morbidity, and is generally indicated for symptomatic or progressive DT.¹⁰ However, recurrence post-surgery is common and is higher in patients with macroscopically positive margins.⁷ Radiotherapy (RT) can be used as a definitive treatment modality with results that compare favourably to surgery.¹¹ Also, when used in combination with surgery, RT appears to decrease local recurrence rates in patients with incomplete surgical excision, particularly following surgery for recurrent tumours.⁷ Systemic therapies, including cytotoxic therapies, hormonal therapies, anti-inflammatory agents and biologicals, are also occasionally used.¹² Recently, practice guidelines in many countries have shifted to more expectant management of DT because of increasing evidence that a significant percentage of these tumours may regress or remain stable without any intervention.^10,13,14

There are limited published data on this condition in low- and middle-income countries. As part of a review of local treatment protocols, the study was conducted to assess the demographics, clinical characteristics, treatment modalities and outcomes of adult patients who were diagnosed with DT, over a 13-year period. This study aims to describe the demographic and clinical characteristics, management strategies, local recurrence and outcomes for all patients treated with DT over this period.

Ethical clearance for the study was granted by the Human Research Ethics Committee of the Faculty of Health Sciences at the University of Cape Town (HREC REF: 679/2017).

The demographic profile of study patients with DT is similar to other published studies with the majority being female with a median age in the fourth decade and the majority being sporadic DT.^17,18,19,20 Sporadic DT affects β-catenin production with the mutation being in the CTNNB1 gene, whereas in FAP-related DT, the mutation is in the adenomatous polyposis coli (APC) gene.¹² Positive β-catenin immunohistochemistry in those tested (38/70 – 54%) in this study was much higher (95%) than the reported rate (67% – 80%) in other studies.²¹ The average size of desmoids in the current study was 9.0 cm (range: 2.0 cm – 29.0 cm), which is larger than other similar reports, where the mean was between 6.3 cm and 7.7 cm.^2,20 Larger size at presentation is significant as size greater than 7 cm has been shown to be a poor prognostic factor for progression-free survival.¹³

In terms of the location, the majority (93%) of patients had extra-abdominal DT which is similar to a report from a study involving 426 patients by Salas et al.,¹³ which showed 87% of DT to be extra-abdominal. Interestingly, only one of the six confirmed FAP-associated DT cases was intra-abdominal (four were located in the abdominal wall and one in the neck region). This preceding finding is unusual as the majority of reported DT associated with FAP are intra-abdominal, followed by the anterior abdominal wall. Pregnancy (previous or current) was noted in close to 25% patients. Similar to what is reported in the literature,²² over two-thirds of pregnancy-related DTs occurred in the anterior abdominal wall and this group had a good outcome with a local recurrence below 5%.

The majority of patients underwent surgery as their primary treatment with the aim of achieving clear surgical margins. The rates of R0 and R1 surgical resections rate in our cohort are comparable to published studies despite relatively late presentation and large tumour size. However, the clinical relevance of achieving clear surgical margins and its impact on local recurrence is not clearly proven and is the subject of conflicting reports in the literature.⁷ In a series by Gronchi et al.²³, there was no significant difference in disease-free survival in those with microscopically negative or positive surgical margins, although there was a trend towards significance in patients with microscopically positive margins after repeat surgery for local recurrence. In a systematic review in 2017 which included 16 studies and 1295 patients,⁷ microscopic margins did seem to be an important factor with an almost twofold increase in risk of recurrence for patients treated with surgery alone and positive microscopic margin. In our series, there was no local recurrence detected in those with negative surgical margins; however, the sample was too small to prove statistical significance. There were three local recurrences in patients with microscopically positive margins and two local recurrences in patients with macroscopically positive margins.

Another subject of contention in the management of DT is the role of adjuvant RT following surgery with involved margins, with some series showing a local control benefit^11,24,25,26 and others showing no clear benefit.^23,27 In our series, the effect of adjuvant RT cannot be determined because of the small sample size, heterogeneous treatment regimens and lack of adequate follow-up.

Factors associated with recurrence noted in published literature include age < 37 years, tumour size > 7 cm in diameter and extra-abdominal location.¹³ The average age of patients with recurrence in our series was 25 years compared to the overall mean age of 37 years. Neo-adjuvant RT has been used with promising results in some centres,²⁸ but in our series, only two patients received neoadjuvant RT with one proceeding to an R0 resection with no recurrence and the other with no response to RT.

Radiotherapy as definitive treatment is an acceptable alternative treatment to surgery, with local control rates as high as 90.9% at 3 years, including 13.6% complete responses, 36.4% partial responses and 40.9% stable cases being reported.²⁹ Our results showed local control rates well over 90% in patients treated primarily with RT, with over 77% of cases attaining either complete or partial response. The average follow-up was close to 5 years, and it is important to note that the effects of RT can be slow and ongoing even beyond 3 years.²⁹ This response was measured according to the RECIST criteria despite its limitations in assessing the slow response of some tumours to RT.¹⁶

Systemic therapy, previously employed only in situations where surgery was not an option (e.g. intra-abdominal FAP-associated DT), is becoming a more commonly used option in the management of DT.¹² It consists of non-cytotoxic therapy and cytotoxic therapy. The non-cytotoxic therapies include hormonal agents (e.g. tamoxifen), anti-inflammatory agents (NSAIDS) or biologicals (imatinib, sorafenib).^12,30 The cytotoxic therapies include chemotherapy agents such as doxorubicin, vinblastine and methotrexate.¹ Other newer local treatments include local ablative therapies using thermal or chemical means (e.g. isolated limb perfusion with tumour necrosis factor alpha),³¹ particularly in those poorly suited to surgery.¹⁰ The use of systemic therapy in our setting was limited to only a few patients and in heterogeneous treatment settings. Only doxorubicin had a clinically significant impact with a good response in one of the patients managed with this agent.

Active surveillance for 1–2 years for DT is a management strategy that has been adopted by many guidelines in recent years.^{7,10,14,32,33} This strategy has developed because of reports that up to 15% of DT regress spontaneously and a significant number remain stable with a progression-free survival of up to 50% at 5 years.^34,35,36 These findings have brought into question traditional therapies, primarily surgery, as the mainstay of treatment, particularly in cases where surgical excision results in significant morbidity.¹³ The aim of active surveillance is to determine which DTs are aggressive and will progress and which are indolent, slow-growing or may regress. Unfortunately, to date, there are no reliable biological markers to distinguish these two groups although genetic mutations in the CTNNB1 gene are being investigated.³⁷ Because of the time frame of our study, none of the patients in our study underwent an active surveillance strategy, although it is clearly a preferable option to reduce patient morbidity and also to limit unnecessary procedures in our resource-limited setting. When considering the safe implementation of this strategy in our local context, the issues of late clinical presentation, delay in referral pathways, larger tumour size and poor follow-up will need to be taken into consideration.

This study has many obvious weaknesses, including being retrospective, small sample size, poor follow-up, heterogeneous treatment regimens and missing or incomplete patient records. This impacts the external validity of the study findings. Associations between clinical characteristics and outcomes could not be explored further because of the small sample size.

This retrospective review of patients affected by DT demonstrates the rarity of the condition, the unpredictable natural history and the variety of treatment options available. While many of our findings mirror previously published studies, the mean size of DT in this series was greater, possibly because of later presentation or delayed referral. The majority of patients in this series underwent surgical management and a subset of patients were treated with adjuvant or definitive RT. Systemic treatments played a minor role. While surgical and RT treatment outcomes in this series were acceptable, strong conclusions cannot be drawn because of small numbers and inadequate follow-up. Newer treatment approaches emphasising active surveillance may need to be incorporated into our management protocols but with an awareness of the specific clinical context and through an individualised multidisciplinary decision-making process.