Disease and economic burden of surgery in desmoid tumors: a review

ABSTRACT

Introduction

Desmoid tumors (DT) are soft-tissue tumors that infiltrate into surrounding structures with ill-defined margins. Although surgery is a potential treatment option, complete excision with negative margins is not often possible, the postsurgery recurrence rate is high, and surgery can result in disfigurement and/or loss of function.

Areas covered

We conducted a literature review to assess the burden of surgery in patients with DT, focusing on recurrence rates and functional deficits resulting from surgeries. Since economic data related to DT surgery is lacking, reviews of surgery costs in soft-tissue sarcomas and of general costs of amputations were conducted. Risk factors for DT recurrence after surgery are young age (<30 years), tumor location (extremities), tumor size (>5 cm in greatest diameter), positive resection margins, and history of trauma in the area of the primary tumor. Tumors in the extremities have the highest risk of recurrence (30%–90%). Lower rates of recurrences have been reported when radiotherapy was used after surgery (14%–38%).

Expert opinion

Although effective in specific cases, surgery may be associated with poor long-term functional outcomes and higher economic costs. Therefore, it is imperative to find alternative treatments with acceptable efficacy and safety profiles that do not adversely affect functional aspects in patients.

KEYWORDS:

• Aggressive fibromatosis
• costs
• desmoid tumors
• functional outcomes
• quality of life
• recurrence rate
• surgery
• systemic therapy

1. Introduction

Desmoid tumors (DT; or aggressive fibromatosis) are fibroblastic lesions with aggressive, infiltrative, and destructive growth that arise in three major anatomic locations: extra-abdominal (extremities, chest wall, head and neck, or intrathoracic regions), abdominal wall (generally occurs in women during or following pregnancy), and intra-abdominal (either a pelvic or mesenteric location) [1–3]. Approximately 64% of DT cases present as extra-abdominal (including 14% upper extremities and 16% lower extremities), 16% abdominal wall, and 20% intra-abdominal [1]. Most cases of DT are sporadic, whereas some are associated with familial adenomatous polyposis (Gardner syndrome), which are most often intra-abdominal [4,5]. The disease course is often unpredictable, and DT can spontaneously regress, remain stable for a long time, or progress [1,4,6,7].

Surgical excision was the most common treatment option before 2000 and the early 2000s [8,9]. However, because DT are often bulky masses with tentacle-like extensions invading surrounding tissues with poorly defined margins, complete excisions with R0 margins are not often possible, and the rates of recurrences after surgery can be as high as 90%, depending on the location and size of the tumor [6,8–13]. Furthermore, attempting to achieve a negative margin may adversely affect local soft-tissue structures, including muscle, nerves, or vessels, which can lead to greater morbidity [14]. Rates of tumor control of 70% to 80% have been reported with surgery in patients carefully selected with low risk for recurrence or when complete resection with negative margins was possible [15]. There is currently no treatment approved by regulatory agencies for DT. Besides surgery, patients with symptomatic and progressing disease are treated with radiotherapy, ablation procedures, and systemic therapy, including chemotherapy and tyrosine kinase inhibitors [16]. Several promising new agents are being investigated for DT, including gamma secretase inhibitors such as nirogacestat, a novel, oral, small molecule that selectively inhibits γ-secretase noncompetitively [17], and AL102, a selective oral γ-secretase inhibitor [18]; and tegavivint (BC-2059), a small molecule that directly and selectively interferes with the interaction between β-catenin and transducin beta-like protein 1 and transducin beta-like receptor 1 [19,20].

The goals of treatment for DT include long-term tumor control without compromising structure or function as well as a reduction in DT-specific symptom burden (e.g. pain) and its impact on patients’ lives, improvement in functioning with daily activities, and overall quality of life (QOL) [4,7,16,21,22]. In patients with progressive, symptomatic tumors, surgery can lead to significant morbidity, including trauma and functional impairments [7].

We conducted a review of the literature to assess the burden of surgery in patients with DT, focusing on rates of recurrences after surgery and functional deficits resulting from surgery, including limb amputation. Because no studies estimating the economic burden associated with surgery in DT were identified in the literature, we conducted a targeted literature search for the cost of surgery in soft-tissue sarcoma (STS) as well as the general cost of amputation, which is a potential outcome of DT surgeries. We used surgical resections in STS as a proxy for DT because of the similarities in these surgical procedures (i.e. the removal of the entire soft-tissue tumor along with a wide margin, typically performed by surgical oncologists with expertise in STS). Our assumption was based on feedback from several surgical oncology experts in the US and Europe who suggested that STS may be a valid analog to estimate the costs of surgery in DT. These experts noted costs of surgical resections can vary based on the tumor size, location, extent of involvement, and complexity of the case. Similar to STS, surgical resection of DT recurrences may also be needed.

2. Methods

2.1. Data sources and search strategies

To identify studies reporting outcomes of surgical procedures in patients with DT, we searched the PubMed, Embase, and Cochrane Library databases for studies published in English from November 2011 to November 2021. A search strategy, including a combination of Medical Subject Heading (MeSH) terms and key words such as ‘fibromatosis, aggressive,’ ‘desmoid tumor,’ ‘Surgical Procedures, Operative,’ ‘amputation,’ and ‘recurrence’ was developed then adapted to explore Embase and the Cochrane Library (details of each search are included in the Supplemental Appendix). In addition, we reviewed the bibliographies of existing literature reviews; conference abstracts from 2015 to 2022 were searched using Embase and screened for relevance. To supplement the information on disability and functional outcomes after surgery, we conducted desktop searches with no time limits for the topics ‘amputation,’ ‘disability,’ ‘functional outcome,’ and ‘functional status’ in publications on DT.

To identify recent studies reporting costs of tumor surgeries in people with STS and costs of amputations in general, we searched the PubMed, Embase, and Cochrane Library databases using a search strategy for each topic (see Supplemental Appendix). The search for costs of amputations were limited to the last 4 years (May 2018 through May 2022); the search for studies reporting costs of STS surgeries covered the last 10 years (May 2012 through May 2022) because of the few studies that have reported on this topic. Bibliographies of identified studies were reviewed to identify other relevant studies.

2.2. Inclusion criteria and selection of studies

We reviewed the publications identified in the searches using a priori inclusion and exclusion criteria for each topic. Studies reporting recurrence rates or functional outcomes of patients with DT undergoing surgery were included. Studies of STS that included DT but did not report outcomes by tumor type were excluded. We selected randomized controlled trials, post hoc analyses or retrospective studies, and prospective and longitudinal studies that reported the outcomes of interest. For studies reporting costs of amputations or STS surgeries, only those reporting costs in the United States (US) were included so that values could be compared. Costs reported for a single year were adjusted to 2022 US dollars (USD) using the Medical Care Index of the Consumer Price Index [23].

One reviewer performed the screening of titles and abstracts and the selection of studies. One reviewer extracted data relevant to the inclusion criteria, and a second reviewer performed a quality check.

3. Burden of surgery in desmoid tumors

The search strategy identified 438 unique articles; of these, 293 articles were excluded during level 1 and 145 articles were included for full-text review during level 2. This review included a total of 33 studies covering 4 main topics on the burden of surgery: (1) recurrence after surgery, including the effects of DT tumor location, treatment, surgical margins, and tumor mutations; (2) functional deficits resulting from surgery; (3) surgery impact on QOL; and (4) trends in surgical interventions.

3.1. Recurrence after surgery

Recurrence rates of DT after surgery vary depending on the site of the tumor. Table 1 summarizes the recurrence rates by treatment modality (surgery or surgery plus radiotherapy) reported by studies published in the last 10 years. The studies have been grouped by those reporting recurrence rates for all subtypes of tumors (extra-abdominal, abdominal wall, intra-abdominal) and those reporting recurrence rates for extra-abdominal tumors. In some studies, patients with microscopically incomplete resections (R1) were treated with postoperative radiotherapy, but the reported rate of recurrence is for the overall population only. In studies evaluating recurrences after the primary or subsequent surgery of DT in any location (published within the last 10 years), recurrence rates vary from 15% to 41% for those treated with surgery alone (Table 1). In studies evaluating recurrences in patients with extra-abdominal DT only, rates vary from 30% to 90% with surgery alone and from 14% to 38% with surgery and postoperative radiotherapy (Table 1).

Table 1. Recurrence and progression rates reported in retrospective studies.

				Patients who underwent surgery			Recurrence/progression rates
Reference	Country, study years	Population characteristics	Location of the tumor	Primary	Recurrent	Duration of follow-up, median, mo	Surgery only	Surgery + radiotherapy	Active surveillance^a
All subtypes of tumors (extra-abdominal, abdominal wall, intra-abdominal)
He et al. [24]	China, 1985–2014	N = 114^b; 69% female; median age 33 y	AW: 35% Intra-A: 17% Extra-A: 48%	89	25	72.5	35/114 (31%)
Cates et al. [25]	US, 1983–2011	N = 92^b; 69% female; median age, 40 y	AW: 22% Extra-A: 78%	92	0	38.4	25/92 (27%)
Colombo et al. [26]	France and Italy, 1992–2012	N = 216^b; 63% female; median age, 41 y	Intra-A: 25% Extra-A: 75%	94	0	76 (surgery)	18/94 (19%)		28/70 (40%)
Crago et al. [27]	US, 1982–2011	N = 495; 67% females, median age, 38 y (primary disease); FAP, 3%	AW: 18% Intra-A: 20% Extra-A: 57%	382	113	60	100/439^c (23%)
de Bruyns et al. [28]	Canada, 1990–2013	N = 227; 68% female, median age, 40 y; FAP, 6%	AW: 22% Intra-A: 18% Extra-A: 61%	132		77	29/108 (27%)	4/24 (17%)	33/55 (60%)
Huang et al. [29]	China, 1987–2009	N = 214; 77% female; median age, 33 y	AW: 43% Intra-A: 6% Extra-A: 51%	153	61	55.5	42/214 (20%)
Mueller et al. [30]	Germany, 2000–2015	N = 32^b; 58% female; median age, 36 y	AW: 27% Intra-A: 15% Extra-A: 58%	26	6	65	13/32 (41%)
Nishida et al. [31]	Japan, NR	N = 85^b; 66% female; 53% > 36 y	AW: 15% Extra-A: 85%	73	12	10–42	32/85 (38%)
Peng et al. [32]	US, 1983–2011	N = 211; 68% female; median age, 36 y; FAP, 8%	AW: 21% Intra-A: 23% Extra-A: 56%	179	0	25.7	42/179 (24%)
Turner et al. [33]	Canada, 2004–2015	N = 103^b; 74% female; mean age, 43 y	AW: 30% Intra-A: 17% Extra-A: 53%	53		35	8/53 (15%)		21/50 (42%)
Yang et al. [34]	China, 1999–2019	N = 267; 70% female; 65% > 30 y	Extremity: 38% Non-extremity: 62%	207	60	NR	29/141 (21%)	24/126 (19%)
Extra-abdominal tumors
van Broekhoven et al. [35]	The Netherlands, 1989–2011	N = 132; 64% female; median age, 36 y	Extra-A: 100%	128	0	38	18/128 (14%)
Janssen et al. [11]	SLR and MA 1999–2015	N = 1,295	Extra-A: 100%	1,053	242	25.0–135.0	297/1005 (30%)	79/290 (27%)
Krieg et al. [36]	Switzerland, NR	N = 96; 645% female; mean age, 39 y	Extra-A: 100%	60		100.8	20/44 (46%)	6/16 (38%)	3/15 (20%)
Tsagozis et al. [13]	UK, 1981–2015	N = 174; 56% female; mean age, 35 y	Extra-A: 100%	174		38	44% (primary surgery), ~90% (after surgery for first or second recurrences)

AW = abdominal wall; intra-A = intra-abdominal; extra-A = extra-abdominal; FAP = familial adenomatous polyposis; MA = meta-analysis; N = number of patients; NR = not reported; SLR = systematic literature review; UK = United Kingdom; US = United States.

^aPatients who progressed under active surveillance.

^bPatients with familial adenomatous polyposis were excluded.

^cRecurrence information is not available for all patients undergoing surgery.

3.1.1. Effect of DT location on recurrence rate

Some studies [13,24,35,37] have indicated that the location of the DT is a major prognostic factor for recurrence. Desmoid tumors in the abdominal wall, intra-abdominal cavity, breast, or lower limb have been reported to be associated with a higher event-free survival—defined as local relapse after surgery or progressive disease per Response Evaluation Criteria in Solid Tumors (RECIST) during nonsurgical approach, change in treatment strategy, or disease-related death—than DT in the chest wall, head and neck, or upper limb, which were associated with lower rates of event-free survival (2-year event-free survival: 66% vs. 41%) [37]. In a study of 114 patients with sporadic DT by He et al. [24], patients with recurrent disease more commonly presented with extra-abdominal tumors compared with those presenting with primary disease (80% vs. 39%, respectively; P = 0.001). In patients (n = 174) with primary resections of extra-abdominal tumors [13], tumors in the extremities had a higher risk of local recurrence compared with tumors located in the trunk or pelvis (P < 0.001). Attempted resection of recurrent disease in the extremities was associated with approximately 90% disease recurrence after each procedure, despite the quality of the surgical margins being equivalent to primary resections. Recurrence was observed after a median of 17 months following the primary resection and after a median of 12 months following the second surgery. Adjuvant treatments (radiotherapy or chemotherapy) had no significant effect on the local control rate of recurrent disease: odds ratio (OR) = 0.693 (95% confidence interval [CI], 0.370–1.299) and OR = 0.969 (95% CI, 0.402–2.338), respectively [13].

A study by van Broekhoven et al. [26] analyzed the risk of recurrence in 128 patients with regard to tumor location for extra-abdominal DT. The risk of recurrence increased for lesions located on the extremities compared with tumors on the trunk (OR = 6.69; 95% CI, 1.42–31.54). The youngest age group of patients had the highest risk of local recurrence at 5 years (1–28 years, 34%; 29–35 years, 10%; 36–44 years, 14%; 45–80 years, 15%), but this was not statistically significant.

3.1.2. Effect of treatment on recurrence rate

Seinen et al. [10] conducted a systematic review of studies published from 1999 to 2017 to compare the local control rate for surgery, surgery plus radiotherapy, radiotherapy alone, and active surveillance. Local control was defined as no recurrence or no progression of disease. A total of 37 studies were included for analysis, representing 2,780 patients (1,670 underwent surgery, 815 underwent surgery plus adjuvant radiotherapy, 155 underwent radiotherapy alone, and 140 were under active surveillance alone). The median follow-up was 63 months (range, 16–150). The local control rates were 75% for surgery alone, 78% for surgery plus radiotherapy, 85% for radiotherapy alone, and 78% for active surveillance. Local recurrence was more frequent after surgery with positive margins than after surgery with negative margins. Adjuvant radiotherapy after positive margins did not improve the local control rate compared with surgery alone (P = 0.549). When local control was compared between the radiotherapy and active surveillance groups, no significant difference was observed irrespective of surgical margins (P = 0.355). Patients with recurrent disease had fewer local recurrences after being treated with adjuvant radiotherapy compared with surgery alone (P < 0.001). Moreover, patients who were on active surveillance had a better local control rate than patients treated with surgery alone (P = 0.001). In the observation group, stabilization of the tumor was seen in a median of 14 months (range, 12–35). The median time to local recurrence for all treatment groups (surgery, radiotherapy, and surgery plus radiotherapy) was 17 months (range, 11–52).

Radiotherapy showed no significant effect on recurrence-free survival or time to recurrence (TTR) in a cohort of 267 patients with R0 resections (70% female; 38% with tumors in extremity and 62% with non-extremity tumors; 78% of patients with primary tumors; 65% older than 30 years) (Table 1), except for delaying recurrences in those 30 years or younger (TTR = 35 months with surgery plus radiotherapy, TTR = 11 months with surgery alone). Twenty percent of patients experienced recurrence [34].

Another study by Krieg et al. [36] compared recurrence rates in 96 patients with extra-abdominal DT (mean age: 39 years; 64% female; 47% of patients had tumors in the extremities) (Table 1) when the initial treatments were surgery (n = 44), surgery plus radiotherapy (n = 16), active surveillance (n = 15), radiotherapy only (n = 9), or systemic therapy (n = 12). At a median follow-up time of 8.4 years, recurrence rates were lower for the active surveillance (20%) and radiotherapy-only groups (0%) than for the other treatment groups; surgery alone had the highest recurrence rate (46%), followed by surgery plus radiotherapy (38%) and systemic therapy (42%).

3.1.3. Effect of surgical margins on recurrence rate

The effect of surgical margins on rates of DT recurrences was analyzed in a meta-analysis of 16 studies that were published from 1999 through 2015 [11]. A total of 1,295 patients with extra-abdominal DT were treated with either surgery alone or surgery plus adjuvant radiotherapy and followed up for 25 to 135 months. Negative margins (R0) as compared with positive margins led to significantly lower recurrence rates after complete resection for both primary and recurrent tumors. For patients who were treated with only surgical resection, those with microscopically positive resection margins experienced a local recurrence risk that was nearly twice as high as that for patients with R0 margins (risk ratio [RR], 1.78; 95% CI, 1.40–2.26). Adjuvant radiotherapy after surgery with negative margins had no noticeable impact on recurrence. In contrast, adjuvant radiotherapy after incomplete surgical resection reduced recurrence rates both in patients with primary tumors (RR, 1.54; 95% CI, 1.05–2.27) and in those with recurrent DT (RR, 1.60; 95% CI,1.12–2.28). Due to the infiltrative growth pattern of DT, it is difficult to achieve microscopically negative surgical margins [38].

3.1.4. Effect of mutations on the recurrence rate

Two meta-analyses by Timbergen et al. [39] and Guo et al. [40] concluded that primary sporadic DT harboring a CTNNB1 S45F mutation have a higher risk of recurrence after surgery than DT with other mutations, such as T41A, S45P, or wild-type DT. The OR for recurrence of DT with S45F mutation was 4.76 versus wild type (P = 0.004), 4.35 versus T41A (P = 0.03), and 5.35 versus other mutations (P = 0.04) [40]. This association seems to be mediated by tumor size [39]. However, in a recent large prospective study, the S45F mutation was not significantly associated with progression or relapse [41].

3.1.5. Other risk factors for recurrence

In addition to the factors discussed above, the following risk factors have been significantly associated with higher rates of recurrence:

• Age: Patients younger than 30 years at the time of presentation [10,24]. Peng et al. [32] estimated that for each 5-year increase in age, the hazard ratio (HR) incrementally decreased (HR, 0.91; 95% CI, 0.82–0.98). Yang et al. [34] estimated a HR of 2.145 for recurrence rates in patients 30 years or younger (P = 0.007).

• Tumor size: Tumors with diameters>5 cm [24]. Yang et al. [34] estimated a HR of 2.72 for recurrence of tumors with a diameter>5 cm (P = 0.002) [42].

• Others: History of trauma in the area of the primary tumor has been associated with a HR of 2.88 (95% CI, 1.20–6.90) for recurrence [32].

A prognostic nomogram (including tumor size, tumor site, patient’s age) to estimate the risk of local recurrence after surgery has been published [27].

3.2. Functional deficits resulting from surgery

Management of extensive recurrent limb DT is challenging because of the failure of current nonsurgical treatments (radiotherapy, systemic chemotherapy, or endocrine treatment) in controlling the disease. In fact, repeated resections can result in mutilating surgery or even amputation [43]. However, there is no study in the literature examining the rate of amputations resulting from surgical treatment of DT or the functional outcomes after surgery and amputation due to DT. A few case reports [44,45] provide information on specific patients undergoing amputation, which left the patients severely disabled. For example, Sakamaki et al. [45] reported on a patient who underwent amputation of the right arm after experiencing complete paralysis resulting from 2 surgical procedures for a recurrent chest wall DT involving the brachial plexus.

Two studies by Catton et al. [46] and Gaposchkin et al. [47] described the functional outcomes of patients with DT who underwent surgery and other treatments (Table 2). Catton et al. [46] described the functional outcome of 24 patients with DT involving a limb or limb girdle. Five patients underwent amputation: 1 amputation was performed as treatment for the primary disease before referral to a specialized center; the other 4 were performed after initial management at the specialized center as treatment of a painful recurrence. At a median follow-up of 86 months, 11 of 24 patients (46%) had major functional limitations, with a functional grade of 2 or less (functional grades range from 0 [limb amputated] to 4 [no functional limitations]; see Table 2). In patients referred for recurrences (n = 12), 5 (42%) had a functional grade of 2 or less at presentation, and this proportion increased to 8 of 12 (66%) after treatment. In patients without a history of recurrence (n = 12), only 1 (8%) had a functional grade of 2 or less at presentation, and this proportion increased to 3 of 12 (25%) after treatment.

Table 2. Functional outcomes in patients with DT in their extremities after tumor resections.

Reference	Population	Symptoms/functional deficits	At presentation, n (%)	After treatment, n (%)
Catton et al. [46]	Patients with DT in limb or limb girdle (N = 24)	Limb amputated (grade 0)	1 (4)	5 (21)
		Functional deficit severe enough to make amputation an acceptable therapy of choice (grade 1)	1 (4)	0 (0)
		Resection of major muscle group or nerve. Major functional limitations requiring an orthopedic appliance. Chronic pain requiring regular narcotic analgesia (grade 2)	4 (17)	6 (25)
		Mild functional limitations but able to perform routine activities (grade 3)	7 (29)	3 (13)
		No functional limitations. Able to participate in prolonged activity (grade 4)	9 (38)	8 (33)
		Not recorded	2 (8)	2 (8)
Gaposchkin et al. [47]	Patients with DT of the brachial plexus (N = 15)	Limb amputated	0 (0)	1 (7)
		Upper extremity or neck pain	4 (27)
		Chronic pain syndrome		9 (60)
		Upper extremity paresthesia/sensory loss	5 (33)	7 (47)
		Upper extremity weakness	2 (13)	7 (47)
		Upper extremity decreased movement	6 (40)
		Upper extremity significant functional loss		6 (40)
		Upper extremity neurologically intact	9 (60)	3 (20)
		Painless mass	8 (53)
		Muscle spasm	2 (13)

DT = desmoid tumors; N = number of patients.

In a study of 15 patients with DT of the brachial plexus who underwent tumor resection, 40% experienced functional loss of the involved limb [47]. Seven patients received a single resection, whereas 8 patients underwent multiple resections for recurrent, symptomatic disease. Thirteen patients received postoperative radiotherapy. At presentation, 9 patients (60%) were neurologically intact (nerves of the brachial plexus were unaffected), most often complaining of a painless mass (Table 2); but after a mean follow-up of 65 months, only 3 patients (20%) remained neurologically intact after they were treated surgically for their tumors. Lasting neurologic deficits included paresthesia and sensory loss (n = 7, 47%), motor deficits (n = 7, 47%), and chronic pain syndromes (n = 9, 60%). Furthermore, 6 patients (40%) reported significant functional loss of the upper extremities—defined as loss of useful function due to motor deficit, debilitating pain syndrome, or both. One patient had forequarter amputation. At the end of treatment, only 3 of 9 patients (33%) with no neurologic deficits preresection remained deficit free.

Some studies compared the outcomes and side effects experienced by patients undergoing surgery or systemic therapy. A study by Lewis et al. [14] reported on 22 patients with DT in the extremities that were unresectable without amputation; 7 patients underwent amputation and 15 did not. Patients undergoing amputation had painful, functionless, or infected extremities resulting from recurrent DT or local complications of treatment. Patients who did not undergo amputation were followed for 25–92 months; they received active surveillance (n = 6), systemic treatment with tamoxifen and/or nonsteroidal anti-inflammatories (n = 4), cytotoxic chemotherapy (n = 3), or chemotherapy with either tamoxifen or nonsteroidal anti-inflammatory drugs (n = 2). In the 15 patients who did not undergo amputation, tumor progression was insignificant or did not occur; in the 3 patients who underwent active surveillance, some tumor regression occurred based on clinical and radiological data. A study by Sparber-Sauer et al. [48] found that of 90 children and adolescents with DT, 89 were treated with either low-dose chemotherapy (n = 35) or surgery (n = 54). Severe late effects (cardiomyopathy) due to chemotherapy were seen in 1% of treated pediatric patients. In contrast, 20% of pediatric patients who underwent surgery experienced functional deficiencies ranging from limited mobility to scoliosis, and deformation of extremities in affected joints. The authors suggest that side effects of chemotherapy may be negligible compared with late effects of surgical resection. In summary, procedures preserving function and structure should be the objective of DT treatment.

3.3. Impact of surgery on QOL

A patient’s wellbeing may be influenced by the treatment strategy. Patients with DT undergoing surgery have a high likelihood of receiving additional surgeries [22]. Recurrent surgeries in patients with DT may negatively impact their QOL. Analysis of data from the Danish National medical databases shows that the average number of surgical procedures in patients with DT who were initially treated surgically was higher than that of those who were initially treated nonsurgically (1.6 vs. 1.2 procedures per individual). For comparison, the number of surgical procedures in a matched cohort of healthy individuals was 0.7 per individual. Patients who were treated surgically used more medications (i.e. nonsteroidal anti-inflammatory drugs, steroids, and opioids) than those who were not treated surgically [22].

A retrospective study found that patient-reported outcome scores as measured with the Patient-Reported Outcomes Measurement Information System (PROMIS) were worse among patients with DT who underwent 2 or more surgical interventions and among those treated with surgery and radiotherapy at any time (not necessarily to treat the same tumor[s]) as compared with patients who did not [49]. This indicates that a more aggressive local treatment strategy may be associated with poorer long-term functional outcomes [49]. Mean PROMIS function scores were 39 for2 or more resections, 51 for 1 resection, and 47 for 0 resections (P = 0.025). A score of 50 on PROMIS instruments corresponds to the normative mean for the general population, with a standard deviation of 10; higher scores indicate better functioning. For both patients with primary and recurrent tumors, event-free survival was not improved for patients treated with local modalities compared with those treated without local modalities [49].

3.4. Trends in surgical interventions

Given the high rate of recurrence and adverse functional outcomes reported after surgery in DT and the realization that, in some cases, better tumor control is obtained with active surveillance or radiotherapy alone, the use of surgery as initial treatment has decreased in the last 10 years (Figure 1) [22,37,50,51]. Penel et al. [37] reported that surgery as initial treatment in France has consistently decreased between 2010 and 2015 from 55% to 42% (P = 0.031). Similar findings were reported in Japan (2006–2012) using data from 530 patients in the Bone and Soft Tissue Tumor Registry database: in 2006, almost equal numbers of cases were treated with (n = 17) or without surgery (n = 18); by 2012, one third (n = 34) of cases were treated with surgery while two thirds (n = 73) were treated without[50].

Figure 1. Changes in the percentage of cases treated with surgery as initial treatment.

AW = abdominal wall; DT = desmoid tumors; Extra-A = extra-abdominal.

^a Source: Penel et al. [37]

^b Source: Nishida et al. [50]

^c Source: Sobczuk et al. [51]

^d Source: Anneberg et al. [22]

In Poland, before 2006, surgery was preferred in 62% of cases and an active surveillance approach was used in 14% of cases. From 2006 through 2010, rates for surgery and active surveillance were 75% and 8%; from 2011 through 2015, surgery and surveillance rates were 62% and 33%; and from 2016 through 2018, rates of surgery and surveillance were 30% and 67%, respectively [51].

A recent Danish study reported that rates of surgery as initial DT treatment fell from 75% between 2009 and 2014 to 32% between 2015 and 2018 [22]. The change from surgery to active surveillance was observed more often in patients with abdominal wall DT (91% in 2009–2014 vs. 36% in 2015–2018, or 55% change) than in patients with extra-abdominal DT (59% in 2009–2014 to between 30% and 36% in 2015–2018, or 29% change).

These findings are also consistent with the two key DT and STS guidelines [7,16] that recommend active surveillance with continuous monitoring for asymptomatic DT that do not cause functional limitations. If there is ongoing tumor progression, increased symptoms, or higher risk of disease morbidity, the guidelines recommend treatment options depending on the anatomic tumor location and potential morbidity.

4. Costs of STS surgeries and amputations

A total of 4 US studies (Table 3) reporting the cost of tumor surgeries in people with STS were identified in the literature. Cost of surgery ranged from $23,080 in 2022 USD in Medicare beneficiaries [52] (measured by mean actual payments including costs of index surgeries, associated hospitalizations, and services provided within 30 days of discharge for unplanned excisions [including first and subsequent resections] of truncal or extremity STS) to $55,801 in 2008–2013 USD (measured by mean direct hospital costs including surgical supplies, services, room and care, and pharmacy-related costs for resections of spinal primary tumors [chordomas, chondrosarcomas, giant-cell tumors, nerve sheath tumors, schwannomas, neurofibromas, and hemangiomas]) [54]. Lower costs were reported by a study evaluating Medicare payments for STS planned or unplanned excisions [52], whereas the remaining studies reporting costs from single-center databases reported higher hospital costs or charges.

Table 3. Costs of STS surgeries in the US.

Reference	Population	Type of surgery/source	Type of costs/charges	Costs/charges per patient
Bateni et al. [52]	Medicare patients (N = 3,913, age≥66 y) with truncal or extremity STS in 1992–2011	Planned and unplanned excisions (including first and subsequent resections)/SEER-Medicare database	Costs for first excisions or re-excisions of STS tumors include costs of surgeries, hospitalizations, and services within 30 days after discharge; adjusted to 2022 USD	Planned excisions, total mean actual payments: $36,138 Unplanned excisions, total mean actual payments: $23,080
Mundinger et al. [53]	Patients (N = 13; average age, 55 y) undergoing STS resections in 1998–2004	Limb-sparing upper-extremity STS resection and functional reconstruction/single-center database	Total charges include routine, operating room, pharmacy, tests, supplies, therapy, and miscellaneous charges; adjusted to 2022 USD	Proximal upper extremity, total mean charges: $29,658 Distal upper extremity, total mean charges: $41,598
Lau et al. [54]	Adults (N = 181; age≥18 y) who underwent resection of spinal tumors (primary or metastatic) in 2008–2013	Surgery for primary tumors (chordomas, chondrosarcomas, giant-cell tumors, nerve sheath tumors, schwannomas, neurofibromas, and hemangiomas)/single-center database	Direct hospital costs for patient’s admission included surgical supplies, services, room and care, and pharmacy from 2008–2013	Mean direct hospital costs for surgery of primary tumors: $55,801
Alamanda et al. [55]	Patients (N = 304, age≥18 y) who underwent surgical resection of STS in 2006–2008	Primary excisions and re-excisions of extremity STS/single-center database	Billed charges include provider, technical (supplies, operative staff, imaging), and miscellaneous charges from 2006–2008	Total median charges, primary excisions: $40,231 Total median charges, secondary excisions: $44,770

N = number of patients; SEER = Surveillance, Epidemiology, and End Results; STS = soft-tissue sarcoma; USD = United States dollars.

Six US studies reporting costs of amputations, mostly lower-extremity amputations (LEAs), were identified in the literature (Table 4). Five of these studies [56–59,61] reported costs of nontraumatic amputations resulting from comorbidity complications (i.e. diabetes, peripheral artery disease [PAD], or coronary artery disease [CAD]); one study [60] reported costs of traumatic amputations resulting from injuries. The costs evaluated varied from inpatient acute care costs to total medical costs for an amputation and to total healthcare costs per year. One systematic review [61] reported an estimate for total projected lifetime healthcare costs after lower-extremity amputations (requiring rehabilitation) calculated as the product of the number of expected life-years and an estimate of future annual healthcare costs, to which was added an estimate of future costs associated with purchasing and maintaining prosthetic devices ($878,927 in 2019; $953,274 in 2022). Acute inpatient care costs calculated by adjusting the inpatient charges by the hospital’s cost-to-charge ratio for a LEA varied from $23,467 to $33,028 in 2012 ($30,573 to $43,030 in 2022 USD) in patients with diabetes. Costs per patient per year (in 2022 USD) for nontraumatic amputations varies from $70,944 in patients with CAD [58] to $112,751 in patients with PAD [59]. Cost depends on the type of amputation; for example, in veterans with diabetes, total direct medical cost in 2022 USD for a toe amputation was estimated at $54,046, whereas costs of $92,588 and $107,819 were estimated for amputations below and above the knee, respectively [56].

Table 4. Costs of amputations in the US.

Reference	Population	Source	Type of costs/charges	Costs per patient
Franklin et al. [56]	Veterans with diabetes and LEA	VHA clinic users with diabetes in 2010	Diabetes-related LEA expenditures due to inpatient medical, inpatient surgical, and outpatient care were estimated using data from the VHA system; cost adjusted to 2022 USD	Total mean direct medical costs for toe amputation: $54,046 Total mean direct medical costs for amputation below the knee: $92,588 Total mean direct medical costs for amputation above the knee: $107,819
Labovitz et al. [57]	Patients discharged with T2D and Charcot foot undergoing LEA	Discharge files from the California Office of Statewide Health Planning and Development (OSHPD) from 2009 to 2012	Acute care costs were calculated by adjusting the inpatient charges by the hospital’s cost-to-charge ratio. This estimated cost was then adjusted to 2022 USD	Inpatient acute care mean costs, minor LEA: $30,573 for Charcot diagnosis at admission to $43,030 for delayed Charcot foot diagnosis Inpatient acute care mean costs, major LEA: $31,502 for Charcot diagnosis at admission to $41,585 for delayed Charcot foot diagnosis
Berger et al. [58]	Patients with CAD or PAD undergoing nontraumatic amputations	Optum Integrated Database from 2009 to 2016	Costs were assessed over the first year of follow-up and include inpatient, outpatient, and prescription drug costs; adjusted to 2022 USD	Total mean healthcare costs for the first year after amputation in CAD patients: $70,944 Total mean healthcare costs for the first year after amputations in PAD patients: $82,095
Desai et al. [59]	Patients (age≥50 y) with lower-extremity PAD undergoing major nontraumatic LEA	Health insurance claims using the Optum Clinformatics Data Mart database (January 2014 to June 2019)	All claims associated with a primary diagnosis code or procedure code for the event during an inpatient or emergency department visit and up to 1 y after the event; adjusted to 2022 USD	Major nontraumatic LEA, mean cost per patient-year: $112,751
Gomez et al. [60]	California workers with injuries resulting in amputations	Claims data reported to the California Department of Industrial Relations Workers’ Compensation Information System from 2007 to 2018	Medical costs were calculated as the sum of medical, pharmaceutical, and durable medical equipment dollars paid over the duration of the closed claims during 2007–2018	Median costs for amputations: Above elbow: $14,855 Below elbow: $30,825 Partial hand: $7,785 Above knee: $18,390 Below knee: $20,901 Partial foot: $21,469 Multiple: $46,637
Lo et al. [61]	Patients with diabetes or PAD undergoing LEA	Systematic review of studies published from 2013 to 2019	Medicare claims adjusted to 2022 USD	Mean costs of LEA procedure due to PAD: $21,916 Total projected lifetime healthcare costs after LEA: $953,274

CAD = coronary artery disease; LEA = lower-extremity amputation; major LEA = amputation at or proximal to the ankle; minor LEA = amputation of the foot or partial foot; PAD = peripheral artery disease; T2D = type 2 diabetes; US = United States; USD = United States dollars; VHA = Veterans Health Administration.

Note: costs reported for a period of time and not for a single year were not adjusted to 2022 USD because of the lack of information on how the costs were estimated for the whole period in the original source.

5. Summary and conclusions

Key treatment goals of DT include progression-free survival and objective response as well as patient-relevant endpoints such as reduction in symptom burden and increased functioning (i.e. the ability to perform everyday activities) that, when taken together, may lead to improvements in overall health-related QOL [21,22]. Studies [22,37,50,51] reported a decrease in surgical procedures of DT of up to 50% in the last 10 years. Recurrence rates of DT after surgery vary depending on the tumor location, resection margins, type of mutations, tumor size, patient age, and history of trauma in the area of the primary tumor [10,24,32,34]. Because the recurrence rate can be as high as 90% [8–11,13], patients with DT should be carefully selected for surgery to avoid repeated surgical procedures and complications that may result in poor functional outcomes. In addition to loss of function, unnecessary resections and amputations increase the economic burden of DT. Because there is no study in the literature evaluating the economic impact of surgery and amputations in patients with DT, we conducted two additional reviews to identify studies reporting costs of tumor surgeries in patients with STS and general costs of amputations. We believe we can use the costs of these procedures as proxies for the economic burden of patients undergoing surgeries and amputations due to DT. Our assumption that STS is a valid proxy to estimate the costs of surgery in DT is based on feedback from several surgical oncology experts in the US and Europe. Costs of surgical resections can vary based on factors including tumor size, tumor location, extent of involvement, and complexity of the the case. Common goals of surgery for STS and DT include minimizing local recurrence and perioperative morbidity and maximizing function. Although the studies reporting costs of surgery in patients with STS present different populations and different types of surgeries, it seems reasonable to estimate that a resection of a STS tumor would likely cost more than $20,000 in 2022 USD (e.g. the mean total charges including routine surgery services, operating room, pharmacy, radiology, laboratory, medical supply, physical therapy, and miscellaneous charges for limb-sparing upper-extremity STS resection and functional reconstruction at a high-volume referral center can range from $29,658 to $41,598); and it could even cost more than $50,000 in 2022 USD (e.g. the mean direct hospital costs including surgical supplies, services, room and care, and pharmacy for resections of spinal tumors can cost approximately $55,801 in 2008–2013 USD) depending on the location and type of tumor. Likewise, costs of DT surgery would follow similar cost ranges. Regarding costs of amputations, most studies identified in our review report these costs in patients with other comorbidities (diabetes, PAD, CAD); one study [60] provided costs or amputations resulting from work injuries for 2007–2018 with no data for specific years. Nevertheless, these studies inform that a LEA could cost as much as $112,751 per patient-year in 2022 USD [59] and have lifetime healthcare projected costs of approximately 1 million 2022 USD. These figures suggest that amputations in patients with DT would have a similar impact.

Although we conducted a comprehensive literature search to identify studies reporting recurrence rates during the last 10 years and those reporting functional outcomes after surgery, it is possible that we missed relevant studies. We limited our search to studies published in English and may have missed non-English studies that are relevant to our research question. Besides the lack of data to quantify the economic impact of surgeries and amputations in patients with DT, we identified only 2 studies reporting functional outcomes in patients with DT undergoing surgery or amputations. This is a gap in the literature that needs to be addressed, as more studies describing the functional outcomes of patients undergoing surgeries for DT and the frequency of amputations are needed to evaluate the benefit of these procedures. Nevertheless, the few studies reporting functional outcomes after surgery show that patients with DT who underwent multiple surgical procedures experienced functional deficits and limitations, with many having affected nerves, particularly in the extremities, that resulted in loss of function and debilitating pain, clearly affecting their QOL.

Currently, there is no treatment specifically approved for DT. Treatment guidelines [7,16] recommend an initial period of active surveillance for the management of DT. Nonsurgical treatment approaches to DT avoid surgery-related morbidity, recurrences, and potential limb amputation with its negative consequences in physical function, health-related QOL, and economic costs. Although locoregional and systemic therapy are used for the management of progressive, symptomatic DT, it is imperative to find alternative treatment options that are safe and effective.

6. Expert opinion

Desmoids tumors (DT) are rare, locally aggressive, and infiltrative soft-tissue tumors that can cause debilitating pain, deformity, and even life-threatening organ damage. There is no treatment for DT that is approved by the US Food and Drug Administration; however, there is a stepwise approach recommended for DT management, from active surveillance to systemic therapy (e.g. targeted therapies and chemotherapy) or surgery. Given the limited mortality associated with DT, treatment goals should not solely focus on clinical markers, such as progression-free survival, but also consider patient-relevant endpoints such as reduction in DT-specific symptom burden (e.g. pain) and its impact on patients’ lives, improvement in functioning with daily activities, and overall QOL. Surgical resection for DT may result in significant patient burden in terms of high rates of recurrence, disfigurement, and loss of function, particularly in DT of the limbs. The recurrence rates of DT after surgery vary depending on the location of the tumor, resection margins, tumor size, patient’s age, and history of trauma in the area of the primary tumor. In addition to loss of function, unnecessary resections and amputations increase the economic burden of DT. Removing DT located in the abdominal wall or extremities may cause large anatomic defects; this often requires restoration of these defects with additional costly and potentially morbid procedures.

In most tumor locations, an aggressive surgical approach is neither the standard of care nor the recommended first-line treatment option for patients with DT. It is important for surgeons to avoid unnecessary surgery in patients with DT, considering the high risk of recurrence and poor long-term functional outcomes. While surgery is no longer recommended by guidelines as the primary treatment modality for DT, it remains an option for select patients with resectable tumors.

Our findings highlight the importance of a multidisciplinary approach to evaluating and managing DT to support a more accurate patient assessment, timely initiation of appropriate treatment, and individualized management of symptoms. This can help optimize patient management and lead to improved treatment goals of clinical effectiveness (tumor control) and improvements in patient symptom burden, functioning with daily activities, and overall QOL.

Future considerations to advance the field of DT should address the following:

• Further elucidate the role of negative surgical margins and their correlation with recurrence

• Quantify the economic impact of surgeries and amputations in patients

• Generate additional data on the epidemiology, humanistic burden, and economic burden of DT

• Identify other risk factors associated with DT progression and recurrence

• Identify alternative treatments with acceptable safety profiles that improve the patient’s symptom burden, functioning, and QOL

• Explore the optimal sequencing of medical treatments based on genetic, molecular, or clinical factors

• Incorporate more advanced imaging criteria for assessing DT response in clinical trials

• Identify the role of adjuvant and/or neoadjuvant therapies in the management of DT

In the future, we expect the field to evolve away from surgery for most patients with DT to a management strategy that includes more effective and targeted systemic oral and locoregional therapies that will improve the patient’s disease-specific symptom severity and burden, functioning, and overall QOL.

Article highlights

• Desmoid tumors (DT) are soft-tissue tumors with aggressive, infiltrative, and destructive growth, and complete excisions with R0 margins are not often possible, with recurrence rates after surgery as high as 90%. Surgery can result in disfigurement and/or loss of function.

• Surgical excision was the most common treatment option before 2000 and the early 2000s. Recent key guidelines recommend active surveillance with continuous monitoring.

• This study assessed the burden of surgery in patients with DT, including rates of recurrences postsurgery and functional deficits resulting from surgery. We also assessed the economic burden associated with surgery in DT.

• The rate of surgery has decreased in the past 10 years, and recurrence rates of DT postsurgery depend on the location of the tumor, resection margins, type of mutation, tumor size, patient age, and history of trauma in the area of the primary tumor. To avoid repeat surgeries, patients should be carefully selected to avoid complications resulting in poor functional outcomes and unnecessary resections and amputations that may contribute to the increased economic burden of this disease. Using surgeries in patients with soft-tissue sarcomas as a proxy due to a lack of published data for DT, resection of a tumor can cost more than $50,000 in 2022 US dollars, depending on type and location of tumor, and the cost of amputations would be even higher.

• Repeat surgery may negatively impact a patient’s quality of life. Therefore, a stepwise approach is recommended for the management of DT, from active surveillance to systemic therapy.

Declaration of interest

T. Bell, S. Zhou, B. Tumminello, and A. B. Oton are employees of SpringWorks Therapeutics, Inc., Stamford, CT, U.S.A and have an equity or financial interest in SpringWorks Therapeutics, Inc. S. Khan is an employee of RTI Health Solutions, Research Triangle Park, NC, U.S.A. M. M. Fernandez is a former employee of RTI Health Solutions. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors thank John Forbes, Medical Editor at RTI Health Solutions, for his contributions of editorial revisions of this manuscript, and Uchenna Iloeje and Jennifer Han from SpringWorks Therapeutics, Inc., for reviewing this manuscript.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14737167.2023.2203915

Additional information

Funding

SpringWorks Therapeutics, Inc., Stamford, CT, USA, provided funding to RTI Health Solutions for conducting this study and was involved in reviewing the manuscript