Desmoid tumors: Systemic therapy

INTRODUCTION — Desmoid tumors (also called aggressive fibromatosis, deep musculoaponeurotic fibromatosis, and fibrosarcoma grade I of the desmoid type) are locally aggressive tumors with no known potential for metastasis or dedifferentiation. However, tumor invasion into vital structures and/or organs can result in substantial morbidity and may be fatal, especially in the case of intra-abdominal desmoids, particularly those arising in patients with familial adenomatous polyposis (FAP). Desmoid tumors are responsible for 9 to 11 percent of deaths among patients with FAP.

Because desmoid tumors lack the ability to metastasize, local control using surgery and radiation has traditionally been the mainstay of therapy for these tumors. However, there is a significant risk of local recurrence, even after complete surgical resection. Increasingly, a multidisciplinary approach to management is being used, particularly among patients with an intra-abdominal desmoid. Multimodality therapy can result in effective disease control in up to 96 percent of such patients with both FAP-associated and sporadic desmoid tumors, even when the disease occurs at a difficult location, such as the intestinal mesentery. On the other hand, desmoid tumors can have an unpredictable clinical course, and an initial period of observation is an acceptable strategy for well-selected patients with an asymptomatic or minimally symptomatic, radiologically nonprogressive tumor, particularly an intra-abdominal/mesenteric tumor arising in a patient with FAP.

This topic review will discuss systemic therapy for desmoid tumors. The epidemiology, risk factors, molecular pathogenesis, clinical presentation, and local treatment options for desmoids are discussed elsewhere.

OBSERVATION VERSUS SYSTEMIC THERAPY — Desmoid tumors can have an unpredictable clinical course, and observation is an acceptable strategy for well-selected patients with an asymptomatic or minimally symptomatic, radiologically nonprogressive tumor [1], particularly an intra-abdominal/mesenteric tumor arising in a patient with familial adenomatous polyposis (FAP), or a collagenous mesenteric desmoid (ie, one that has low contrast uptake on contrast-enhanced magnetic resonance imaging [MRI]) that is not associated with FAP. Spontaneous regressions have been reported. As an example, in a placebo-controlled randomized trial of sorafenib in patients with disease progression or symptomatic desmoid fibromatosis, 20 percent of patients developed disease regression on the placebo arm [2]. The natural history of desmoid tumors, and the available data utilizing initial observation are discussed in detail elsewhere.

GENERAL APPROACH TO SYSTEMIC THERAPY — There are no evidence-based or widely accepted guidelines for management of unresectable desmoids. Systemic therapy is increasingly being integrated into a multidisciplinary approach for selected patients with unresectable desmoids or for management of intra-abdominal desmoids for which local therapy options may cause unacceptable morbidity [3,4]. Early and aggressive systemic therapy in well-selected patients has the potential to prevent complications that might adversely affect survival.

In general, we base our approach to systemic therapy on the urgency of treatment, which takes into account tumor site, symptoms, growth rate, underlying disorders (eg, familial adenomatous polyposis [FAP]), and patient preference.

Indications — Initial observation is an acceptable strategy for most patients with no or minimal symptoms at initial presentation. (See 'Observation versus systemic therapy' above.)

Common indications for systemic therapy are:

●Unresectable desmoids that are symptomatic, or those that are asymptomatic but progressing radiographically.

●Multiple locoregional recurrences despite adequate local therapy.

●Symptomatic, potentially resectable desmoid tumors for which surgery would be unacceptably morbid.

●Initial nonsurgical treatment of a growing or symptomatic intra-abdominal/mesenteric desmoid tumor, particularly one arising in a patient with FAP.

In this setting, because of the infiltrative pattern with encasement of the mesenteric vasculature, microscopically complete (R0) resection is seldom possible without life-altering loss of intestinal function [4]. Furthermore, surgery may provoke the development of more desmoid tumors in patients with FAP [5]. Finally, recurrences tend to be more frequent and aggressive with each surgical intervention, and some studies suggest a decreased response to systemic therapy after multiple surgical procedures [6].

These observations have led many clinicians (including our group), the American Society of Colon and Rectal Surgeons (ASCRS), and a joint task force of the American Society of Clinical Oncology (ASCO)/Society of Surgical Oncology (SSO) to advocate conservative management rather than initial resection for patients with FAP and intra-abdominal desmoid tumors that are large, are slow growing, involve the mesentery, or encase vessels and/or organs [7,8]. These issues are all discussed in greater detail elsewhere.

In the absence of significant symptoms or rapidly progressive disease, conservative management may entail an initial period of observation. However, if treatment is indicated in this setting because of significantly symptomatic or rapidly progressive disease, we prefer to use noncytotoxic systemic therapy initially as recommended in guidelines for the management of desmoid tumors. (See 'Noncytotoxic approaches' below.)

Urgency of treatment — Once the decision is made to pursue systemic therapy, one of the most important factors in deciding the choice of specific approach is the urgency of perceived treatment. We consider low-urgency tumors to be those that are asymptomatic or minimally or moderately symptomatic, slow growing, and in locations at which local growth, if rapid, would not cause irreparable harm. Nonprogressive, collagen-rich tumors that have limited enhancement on magnetic resonance imaging (MRI) with contrast may be considered low-urgency tumors. On the other hand, rapidly growing or severely symptomatic tumors, especially those located in critical locations, are "high urgency." In general, high-urgency tumors are appropriate for more aggressive treatments (such as nirogacestat or cytotoxic chemotherapy) that have a shorter time to response, while low-urgency tumors are more often approached with noncytotoxic approaches (eg, nirogacestat or oral tyrosine kinase inhibitors [TKI]).

A rapid recurrence of a desmoid tumor postresection is an indication for systemic therapy, but whether such patients fall into the low-urgency or high-urgency category depends on the clinical setting and the factors outlined above.

Choice of agent — Systemic therapy may be broadly classified into cytotoxic chemotherapy and less toxic therapies, such as TKIs, γ-secretase inhibitors, nonsteroidal anti-inflammatory drugs (NSAIDs), and hormonal therapy. There are no evidence-based or consensus-based guidelines as to the appropriate choice of agent for systemic therapy, and there are limited randomized trials to aid in the decision-making process. Consensus-based guidelines for treatment of desmoid tumors from the National Comprehensive Cancer Network [9] suggest systemic therapy as an option for unresectable desmoid tumors and those for which surgery would be unacceptably morbid, but provide limited guidance as to choice of agent.

We base our choice of initial therapy on the urgency of the clinical situation, tumor location, clinical scenario, and patient preference. The efficacy and toxicity of these various approaches are discussed in detail below. (See 'Efficacy and toxicity of different treatments' below.)

●Patients who are severely symptomatic – Aggressive systemic therapeutic options, such as cytotoxic chemotherapy, should be reserved for patients with an impending threat to either life or function because of tumor location or rapid radiographic progression, and for those who are severely symptomatic. Cytotoxic chemotherapy is also preferred for patients with mesenteric involvement or when there is a concern about absorption. Nirogacestat is a reasonable first-line alternative for selected patients if cytotoxic chemotherapy is not feasible or if patient preference is to avoid chemotherapy. (See 'Tyrosine kinase inhibitors' below and 'Cytotoxic chemotherapy' below.)

In general, if cytotoxic therapy is chosen, pegylated liposomal doxorubicin, doxorubicin, combinations of doxorubicin plus dacarbazine, or methotrexate plus vinblastine are all reasonable options. Some institutions also use methotrexate alone, although there is almost no published experience utilizing this approach.

●Patients who are minimally or moderately symptomatic – Other patients who qualify for systemic therapy should be managed with less toxic approaches, such as nirogacestat, TKIs, NSAIDs, or hormonal therapy. This includes the management of patients with FAP who have intra-abdominal/mesenteric desmoid tumors that are large, are slow growing, involve the mesentery, or encase vessels or organs. However, in this setting, if the mesenteric tumor is impairing function of the gastrointestinal tract where oral absorption of nirogacestat, TKIs, hormonal therapy, and NSAIDs is unpredictable, or if there is concern for perforation risk with TKIs, cytotoxic chemotherapy may be preferred.

In general, if noncytotoxic therapy is chosen, and limiting therapy-related toxicity is an important goal (in addition to obtaining a response), tamoxifen, with or without an NSAID, may be used if the patient is minimally symptomatic or the tumor is slow growing. However, this recommendation is not based on placebo-controlled trials. Observation is a reasonable alternative in this population, as approximately 20 percent of patients on the placebo arm of other desmoid trials have shown tumor regression.

In patients with moderately symptomatic desmoids, initial therapy with nirogacestat is preferred over a TKI or hormone therapy, with or without an NSAID, due to the higher response rates (see 'Nirogacestat' below). If a TKI is chosen for initial treatment (such as for patients concerned about ovarian dysfunction), we suggest sorafenib, pazopanib, or sunitinib rather than imatinib.

Response evaluation during systemic therapy — Desmoid tumors are typically slow growing. Response to systemic therapy can be slow to manifest and may take six to eight months or more [10,11]. In some situations, the response may be delayed past the point of completion of the treatment regimen [6]. Patients who exhibit a delayed response may even experience an initial increase in tumor size followed by an objective decrease in size [10]. (See 'Duration of therapy' below.)

At our institution, patients with desmoid tumors who are receiving systemic therapy are imaged every two to three cycles (or every two to three months for continuously dosed agents or weekly treatment regimens). The persistence of disease that is stable or minimally increased in size should not be considered a treatment failure during initial treatment. Symptomatic or functional improvement should be regarded as a reason to continue therapy when changes in the dimensions of the tumor are minimal or absent. From an imaging standpoint, contrast-enhanced MRIs are the preferred approach for following these tumors [12]. A decrease in contrast enhancement over time is an indication of treatment effect [13].

EFFICACY AND TOXICITY OF DIFFERENT TREATMENTS

Noncytotoxic approaches

Nirogacestat — Nirogacestat, an oral gamma secretase inhibitor that targets the Notch pathway, is an effective treatment for locally aggressive or symptomatic desmoid tumors. Although nirogacestat has been shown to cause ovarian dysfunction, this resolves upon discontinuation for the majority of patients. However, we do not have long-term data about the duration and ultimate frequency of resumption of menstrual cycles after discontinuation of nirogacestat. We do not consider desire for future childbearing to be a contraindication to use of this agent; however, some patients who desire future childbearing may elect for an alternative strategy such as a TKI, or may elect to pursue egg harvesting. (See 'Choice of agent' above.)

Based on initial data from phase I and II trials [14,15], nirogacestat was evaluated in a randomized, double-blind, placebo-controlled phase III trial (DeFi). In this study, 142 patients with objectively progressive desmoid tumors were randomly assigned to either nirogacestat at 150 mg orally twice a day in 28-day cycles or placebo until disease progression [16]. At median follow-up of 19 months, nirogacestat improved PFS (median not reached versus 15 months; HR 0.29, 95% CI 0.15-0.55), ORR (41 versus 8 percent), and complete responses rates (7 versus 0 percent) compared with placebo. Nirogacestat also improved tumor-related symptom burden, pain, physical/role functioning, and health-related quality of life.

Grade ≥3 toxicities for nirogacestat included diarrhea (16 percent), maculopapular rash (6 percent), stomatitis (4 percent), fatigue and hypophosphatemia (3 percent each), and nausea (1 percent).

Ovarian dysfunction occurred in 75 percent of females of child-bearing potential. Most patients (74 percent) with ovarian dysfunction had resolution based on hormonal levels and/or symptoms while receiving nirogacestat or after stopping for any reason. Nineteen percent of patients continued to have ovarian dysfunction while receiving treatment. There was no follow-up in 7 percent of patients who had discontinued nirogacestat for any reason.

The United States Food and Drug Administration approved nirogacestat for patients with progressing desmoids who require systemic treatment [17].

Tyrosine kinase inhibitors — Although comparative studies are lacking, if a tyrosine kinase inhibitor (TKI) is chosen for initial treatment or after progression on nirogacestat, we suggest sorafenib, pazopanib, or sunitinib rather than imatinib, given the greater degree of activity seen with these broader-spectrum, albeit more toxic, TKIs. However, imatinib is a reasonable option for patients with low-urgency tumors for whom hormonal approaches, nonsteroidal antiinflammatory drugs (NSAIDs), and other TKIs have failed or are not tolerated.

Sorafenib — Sorafenib is a multitargeted TKI that inhibits the vascular endothelial growth factor receptor (VEGFR) tyrosine kinase, among others. Activity against desmoid tumors was initially suggested in a review of 26 patients who received sorafenib 400 mg daily, 11 as first-line systemic therapy [18]. Twenty-three had documented radiographic disease progression prior to enrollment, and two had achieved maximal benefit from chemotherapy. There were six partial responders (25 percent), and 17 had stable disease.

Benefit was confirmed in a subsequent randomized phase III trial comparing sorafenib (400 mg daily) with placebo in 87 patients with unresectable progressive or symptomatic desmoid tumors [2]. At a median follow-up of 27.2 months, there was an objective response in 16 of 49 patients treated with sorafenib (33 percent) compared with 7 of 35 placebo-treated patients (20 percent). The two-year progression-free survival rate was significantly higher for sorafenib (81 versus 36 percent), and some responses were durable. Sorafenib was generally well tolerated; the most frequently reported adverse events were grade 1 or 2 rash (73 percent), fatigue (67 percent), hypertension (55 percent), and diarrhea (51 percent) [2].

Pazopanib — Pazopanib is approved by the US Food and Drug Administration (FDA) for use in certain advanced soft tissue sarcomas that have progressed on prior chemotherapy. Our approach is to use pazopanib for those with desmoid tumors who have moderate to rapidly progressive disease and who have received multiple prior lines of chemotherapy or are not candidates for chemotherapy due to possible toxicity. It may also be an appropriate off-label option for patients with more indolent disease who have not received chemotherapy but have experienced progression on other lines of therapy (eg, cyclooxygenase 2 [COX-2] inhibitors, hormonal therapy). (See "Second and later lines of therapy for metastatic soft tissue sarcoma", section on 'Pazopanib'.)

In patients with advanced desmoid tumors, pazopanib has demonstrated both improved efficacy and an improved toxicity profile relative to chemotherapy [19-22]. In an open-label phase II trial (DESMOPAZ), 72 patients with progressive desmoid tumors were randomly assigned to pazopanib at 800 mg daily or to chemotherapy with intravenous methotrexate (30 mg/m²) plus vinblastine (5 mg/m²) given weekly for six months, then every two weeks for another six months [22]. Treatment was administered until progression (with crossover allowed), until onset of unacceptable toxicity, or to a maximum of one year.

After a median follow-up of approximately 23 months, pazopanib increased the proportion of patients who had not progressed at six months relative to chemotherapy (84 versus 45 percent), with a nonsignificant trend towards improved response rates (37 versus 25 percent with partial response; 59 versus 50 percent with stable disease). Grade ≥3 treatment-related adverse events included hypertension and diarrhea with pazopanib, and neutropenia and transaminitis with chemotherapy. Rates of dose modification due to toxicity were similar in both treatment arms (73 versus 77 percent). Eight percent of patients discontinued pazopanib for toxicity versus 23 percent of those receiving chemotherapy. Approximately three-quarters of patients required dose modification for toxicity in the pazopanib arm, suggesting that reduced doses of pazopanib are still clinically effective. Further studies are necessary to confirm this finding.

Sunitinib — Sunitinib, a multitargeted TKI that inhibits the VEGFR tyrosine kinase, among others, is an available option for patients with moderately symptomatic or faster-growing tumors. For patients with minimally symptomatic, slowly progressive desmoids, initial therapy with sunitinib may also be a preferred option over hormone therapy, with or without an NSAID, due to higher and more durable responses seen in a randomized phase II trial [23]. In this study, 32 patients with progressive, recurrent, or symptomatic desmoid tumors were randomly assigned to either sunitinib at an oral dose of 52 mg daily or a combination of tamoxifen and meloxicam [23]. In preliminary results, at median follow-up of 27 months, compared with tamoxifen and meloxicam, sunitinib improved objective response rates (75 versus 0 percent) and two-year progression-free survival (81 versus 36 percent). Although toxicities were more frequently reported in those patients receiving sunitinib compared with tamoxifen plus meloxicam, all responded to dose reductions in sunitinib to 37.5 mg.

Other tyrosine kinase inhibitors (imatinib) — Compared with other TKIs, we consider imatinib to be an inferior agent for desmoid tumors because of a lower response rate, but it may have a role in patients with low-urgency tumors for whom hormonal approaches, NSAIDs, and other TKIs have failed or are not tolerated. For most patients, imatinib is less toxic than chemotherapy or other TKIs. In some settings, the lower potential response rate with imatinib might be counterbalanced by toxicity concerns when considering the choice of imatinib versus chemotherapy, especially in younger patients with this diagnosis.

The efficacy of imatinib for systemic treatment of advanced desmoid tumors can be illustrated by the following reports:

●In a large multi-institutional phase II trial of 51 patients (45 evaluable), patients were dosed based on body surface area, with patients ≥1.5 m² receiving 300 mg twice daily, patients 1 to 1.49 m² receiving 200 mg twice daily, and patients <1 m² receiving 100 mg twice daily [24]. Forty-three (84 percent) reached the primary endpoint defining clinical benefit (a complete or partial response within 16 weeks or stable disease lasting at least 16 weeks). Three patients had an objective partial response. The progression-free survival rate at three years was 58 percent, and five patients remained progression free after ≥4 years of treatment.

●Benefit for imatinib was also shown in a French Sarcoma Group trial in which 40 patients failing local treatment (and with documented evidence of progressive disease) received imatinib 400 mg per day (increased to 800 mg per day in the event of progression) [25]. In the 35 evaluable patients at three months, there was one complete and three partial responses (objective response rate 11 percent), 28 patients (80 percent) had stable disease, and three patients progressed. The nonprogression rates at 6, 9, and 12 months were 80, 69, and 67 percent, respectively. Among the patients who developed progressive disease while receiving 400 mg daily, dose escalation resulted in stabilization of disease for a median time of 12 months (range 2 to 30 months), but eventually, 8 of 10 patients progressed.

●A similar phase II trial from the German Interdisciplinary Sarcoma Group (GISG) enrolled 38 patients with unresectable desmoid tumors treated with an 800 mg oral daily dose of imatinib [26]. The nonprogression rates at 6, 9, 12, 18, and 24 months were 65, 65, 59, 53, and 45 percent, respectively. The overall response rate was 19 percent.

The mechanism whereby imatinib provides benefit remains unclear as tissue analysis has not uniformly demonstrated activating mutations in the genes that encode receptor tyrosine kinases such as KIT or platelet-derived growth factor receptor (PDGFR)-alpha [27]. However, at least some data support the presence of PDGFR-alpha and PDGFR-beta activation that appears to be sustained by an autocrine/paracrine loop [28].

Hormonal therapy — Historically, hormonal therapy has been used in the treatment of desmoid tumors, but the true contribution of hormonal agents over the spontaneous regression rate of 20 percent is unclear due to the lack of placebo-controlled trials. We reserve hormonal therapy for low-urgency tumors that are no more than minimally symptomatic and slowly growing and strong patient preference to limit therapy-related toxicity.

Alternatively, initial therapy with TKIs is preferred over hormone therapy, with or without an NSAID, even for patients with minimally symptomatic, slowly progressive desmoids, due to higher and more durable response rates. (See 'Sunitinib' above.)

A potential role for female hormones in the pathogenesis of desmoid tumors was suggested by the finding that desmoids are more common in females than in males and by the occurrence of abdominal wall desmoids during pregnancy and their resolution with the cessation of hormonal influences. These observations prompted the investigation of antiestrogenic therapies.

The efficacy of hormone therapy for management of unresectable, symptomatic desmoid tumors was first described in 1983 [29]. Tamoxifen was used to treat a patient with Gardner syndrome who had metastatic adenocarcinoma of the colon and multiple, large, painful desmoid lesions; within a week, the patient's pain had disappeared, and subsequently, the size of the desmoid tumors decreased. Since this case report, there have been several such case reports and small studies that have confirmed the activity of hormone therapy against desmoid tumors.

The exact mechanism of action of hormonal agents such as tamoxifen remains largely unknown. Desmoid tumors are known to express estrogen receptors [30], particularly estrogen receptor beta [31], and this may account for the clinical activity of tamoxifen observed in this disease. (See "Mechanisms of action of selective estrogen receptor modulators and down-regulators", section on 'Estrogen receptor-beta: A second ER isoform'.)

Alternatively, it has been postulated that the latency of action of hormonal agents (and other therapies, such as radiation or chemotherapy) may be reflective of the deprivation of a growth signal or cytokine [32].

Efficacy of tamoxifen and other selective estrogen receptor modulators — Tamoxifen, a selective estrogen receptor modulator (SERM), is the most commonly used agent for systemic therapy of desmoid tumors. Most of the data on tamoxifen are derived from isolated case reports [6,33,34] and small series [35]. In general, objective responses occur in 15 to 20 percent of patients, with an additional 25 to 30 percent achieving symptomatic improvement and disease stabilization; thus, the clinical benefit rate is approximately 50 percent.

In one of the largest reports investigating benefit from tamoxifen or another SERM, toremifene, clinical benefit (stable disease and partial or complete responses) was observed in 65 percent of 20 patients with progressive desmoid tumors (table 1) [36].

In a second series, 11 of 44 patients with sporadic desmoids who were treated with toremifene had a partial response, and 29 others had stable disease [37]. Benefit was observed both in the first-line setting and after failure of tamoxifen.

Limited data are also available on raloxifene. In one report, 13 patients with familial adenomatous polyposis (FAP) and a progressive intra-abdominal desmoid tumor received raloxifene (120 mg daily); all 13 patients responded, eight with a complete response [38].

Other hormonal therapies — Other therapies that have been used for hormonal manipulation of desmoid tumors include megestrol (one patient treated with 500 mg daily had a partial response that was sustained for 10 months) [39], progesterone (table 1) [40], goserelin (which in combination with tamoxifen, resulted in a partial response for 17 months in one patient) [41], and testolactone (an aromatase inhibitor (table 1)) [42].

Antiinflammatory agents — We reserve the use of NSAIDs for low-urgency tumors that are no more than minimally symptomatic. Whether this approach is preferred over initial treatment with a hormonal agent is uncertain; there are no comparative trials. Furthermore, it is difficult to know if NSAIDs have real antitumor activity given the spontaneous improvement in desmoids that has been borne out in observational series and in the placebo arm of the sorafenib phase III trial. (See 'Sorafenib' above.)

The first report of an NSAID being associated with regression of a desmoid tumor came in 1980, and since then, several small studies have indicated the efficacy of NSAIDs in desmoid tumors (table 2) [10,42-45]. In one of the largest of these series, 12 of 14 patients (86 percent) treated at the Cleveland Clinic with sulindac (median dose 300 mg per day) had clinical benefit (defined as an objective response or stable disease) [10].

An important point is that these response data largely come from small subsets of the larger populations reported in these papers and should therefore be interpreted with caution. Durability of response was also not reported in these studies. Given the lack of randomized studies, the benefit of NSAIDs over any other form of therapy remains unclear.

Combined hormone plus antiinflammatory therapy — Combinations of NSAIDs with hormonal therapy have been investigated, but the benefit of combined therapy over tamoxifen alone remains unclear (table 3) [6,10,42,46,47]. The largest trial that examined this concept separately investigated sporadic (n = 8) and FAP-associated (n = 17) desmoids [6]. Patients received sulindac (300 mg daily) combined with either tamoxifen (doses ranging from 20 to 120 mg daily) or raloxifene. Results were as follows:

●Eleven patients (65 percent) with FAP-associated desmoid tumors had clinical benefit (stable disease or an objective response). There was a lower rate of response in patients with FAP who had had prior surgery for a desmoid tumor than in those who had not undergone surgery, but the number of patients in this subset was small (n = 4, with one partial response and three with progressive disease).

●Of the eight patients with sporadic desmoid tumors, five (63 percent) had stable disease, one (13 percent) had a complete response, and one (13 percent) remained disease free on adjuvant therapy with raloxifene and sulindac after surgical resection. There was only one patient with progressive disease as the best response. Thus, at least 65 percent of patients with FAP-associated desmoids and 75 percent of patients with sporadic desmoid tumors derived some clinical benefit from combination therapy in this study.

Cytotoxic chemotherapy — Chemotherapy is primarily used in the following clinical situations, in which the benefits of therapy clearly outweigh the risks (see 'Choice of agent' above):

●An unresectable tumor that continues to be progressive and symptomatic despite prior noncytotoxic systemic therapy (TKIs, hormonal therapy, NSAIDs).

●A mesenteric tumor with impaired function of the gastrointestinal tract as evidenced by weight loss and malnutrition, where oral absorption of TKIs, nirogacestat, hormonal therapy, and NSAIDs is unpredictable, or for which there is concern as to the risk of perforation with TKIs.

●Patients with an impending threat to either life or function because of tumor location or rapid radiographic progression, and those who are severely symptomatic with unresectable disease.

The choice of chemotherapy regimen should be based on the clinical situation and the ability of the patient to tolerate the adverse effects associated with each regimen. Although response rates are higher with doxorubicin-based combination chemotherapy (eg, doxorubicin plus dacarbazine) as compared with anthracycline monotherapy or weekly low-dose methotrexate plus a vinca alkaloid, toxicity is also more significant, and it is not clear that progression-free survival is better. For most patients, we suggest single-agent pegylated liposomal doxorubicin (PLD) rather than a combination approach, given the differences in toxicity.

As a general rule, benign and slow-growing neoplastic processes are generally regarded as unresponsive to cytotoxic chemotherapy [32]. However, this rule does not apply to desmoid tumors; chemotherapy is effective against these slowly growing, nonmetastasizing tumors.

Historically, given the availability of several less toxic options for upfront use, cytotoxic chemotherapy has typically been used late in the course of treatment for desmoid tumors, after all other therapeutic options have been exhausted. However, clinical practice has shifted away from using cytotoxic chemotherapy predominantly as a "last resort" treatment in patients with advanced disease toward earlier use of a short course of treatment with active single agents such as PLD. Cytotoxic chemotherapy is associated with good response rates (ranging from 17 to 100 percent in various studies, median 50 percent) [33]. Patients receiving chemotherapy can derive functional and symptomatic benefits, such as a decrease in pain, as early as four weeks into treatment [32,48], and such symptomatic improvement may long precede any decrease in tumor size.

Anthracycline monotherapy — Anthracyclines are the most commonly reported agents used to treat desmoid tumors, including doxorubicin [49,50] and PLD [4,50-53]. Monotherapy appears to be very active in controlling desmoid tumors, with response rates that exceed 75 percent in some series (table 4). There are no comparative trials of these agents, but in general, PLD is less toxic than unencapsulated doxorubicin and is preferred by many clinicians. To minimize toxicity in what are generally moderately symptomatic patients, we initiate therapy at 40 rather than 50 mg/m² every four weeks. PLD is very active, perhaps with as high a response rate as TKI therapy. There are no randomized comparative trials to demonstrate this; however, the responses observed with PLD are more compelling than those seen with hormonal therapy or methotrexate.

The optimal duration of treatment is not established and is discussed in detail below. (See 'Duration of therapy' below.)

Combination chemotherapy — In well-selected patients with rapidly progressive desmoid tumors for which other local and systemic therapy options have failed, combination chemotherapy can effectively achieve disease control. Doxorubicin-based combinations are most often used, and associated response rates range from 40 to 100 percent. Non-doxorubicin-based regimens use either combinations of methotrexate and vinblastine on a weekly schedule or other combinations of vinca alkaloids plus an alkylating agent.

Doxorubicin-containing combinations — Doxorubicin-containing regimens have a higher response rate than non-anthracycline-containing regimens against desmoid tumors, but they are also more toxic [54-60]. Doxorubicin plus dacarbazine represents the most commonly used cytotoxic chemotherapy combination for desmoid tumors at some centers, in particular when a response is needed (high-risk tumors). Phase II trials with this combination are lacking, but results from the largest retrospective studies are outlined in the table (table 5). Many patients developed radiologic responses that were sustained. In some cases, the response was sufficient to allow subsequent surgical resection.

In the largest published series, 11 patients received doxorubicin (60 to 90 mg/m²) plus dacarbazine (750 to 1000 mg/m² infused continuously over a 72-hour period) for a median of five cycles [55]. Six of nine evaluable patients had an objective response (two complete, four partial). One complete responder died, two were lost to follow-up, and the remainder remained alive and progression free with follow-up ranging from 28 to 235 months postdiagnosis. Cardiac toxicity, myelosuppression, mucositis, and nausea are the most common side effects experienced with this regimen. A lower dose regimen (doxorubicin 50 mg/m² plus dacarbazine 600 to 700 mg per cycle) has been reported to provide disease control with a more favorable side effect profile, although experience with this regimen is limited to a few patients [61].

The optimal duration of chemotherapy is not known and is addressed below. (See 'Duration of therapy' below.)

Methotrexate with or without a vinca alkaloid — Several studies, including two phase II trials (one in adults and one in children), have demonstrated the activity of methotrexate plus a vinca alkaloid, with disease control rates (tumor shrinkage of stable disease) that range from 70 to 100 percent (table 6) [11,48,53,62-68]. Randomized controlled trials are unavailable, and most studies used the combination of methotrexate and a vinca alkaloid, although there is no evidence that combined therapy is better than either methotrexate or a vinca alkaloid alone. As a result, some clinicians favor methotrexate alone, without a vinca alkaloid, although there is almost no published experience utilizing this approach.

The two prospective phase II trials and one large retrospective study are described in detail:

●In the adult phase II trial, methotrexate was administered at a dose of 30 mg/m², and vinblastine was administered at a dose of 6 mg/m² (total dose rounded to 50 mg of methotrexate and 10 mg of vinblastine); both were given intravenously weekly for a median interval of one year [62]. Of the 30 patients, 21 (70 percent) had durable radiologic changes (including stable disease and partial and complete responses that did not progress on or off chemotherapy with a median follow-up of 62 months). Nine patients progressed at a median of 10 months after stopping therapy. Neutropenia was the predominant toxicity.

●In the pediatric phase II trial, a lower rate of durable responses (30 percent) was seen, and they lasted a median of 43 months [63]. In this trial, 38 percent of patients progressed while receiving chemotherapy (compared with none in the adult trial), the majority of whom progressed at or before the first magnetic resonance imaging (MRI), scheduled at 12 weeks. Responses to treatment for desmoid tumors may take several months to evolve, and this slow response time may be responsible for the greater proportion of patients with "early progression" in the pediatric trial than in the adult trial.

The median duration of therapy for all patients was 40 weeks (range 10 to 52 weeks). The best responses were a complete response in one patient, a partial response in four (objective response rate 19 percent), a minor response in three, and stable disease in 10. Treatment was complicated by neutropenia in 22 patients, which was grade 4 in five. In contrast, only two patients developed grade 1 sensory neuropathy.

●These results were confirmed in a 75-patient retrospective study that examined the use of methotrexate plus vinorelbine (57 percent of patients), methotrexate plus vinblastine (40 percent), or vinorelbine alone (3 percent) in patients with progressive desmoid fibromatosis [67]. The best responses were a complete response in 1 percent, a partial response in 47 percent, and stable disease in 51 percent; the median duration of chemotherapy administration was 14 months. Eight patients discontinued therapy due to toxicity rather than disease progression. Median progression-free survival was 75 months (136 months in patients with a response). Retreatment at the time of disease progression off chemotherapy was attempted in 11 patients, with a partial response in four patients, stable disease in six patients, and progressive disease in one patient.

One setting in which methotrexate plus vinorelbine might be preferred is for patients who develop neurotoxicity during therapy with vinblastine [68].

There are no data comparing methotrexate plus a vinca alkaloid with PLD, doxorubicin alone, or combination doxorubicin plus dacarbazine. In general, personal experience and preference dictate the choice. One setting in which a methotrexate/vinca alkaloid combination might be chosen over other chemotherapy regimens is if there are infectious complications from persistent neutropenia caused by other systemic therapy strategies.

DURATION OF THERAPY — For both cytotoxic and noncytotoxic agents, the duration of therapy should be based on toxicity and response. As noted above, responses are typically delayed in desmoid tumors, and this complicates the use of radiographic imaging during therapy to guide the optimal duration of therapy. (See 'Response evaluation during systemic therapy' above.)

There are no data addressing the optimal duration of treatment, and there is no consensus on this issue. Although noncytotoxic agents are often continued indefinitely in the absence of toxicity, treatment breaks can be incorporated into the treatment plan, particularly if there are side effects. For cytotoxic therapy, dose-limiting toxicities tend to develop over time, and the duration of therapy will depend on the regimen:

●Doxorubicin-based regimens are limited by cumulative cardiac toxicity and myelosuppression. In responding patients, we generally limit treatment to six cycles of therapy, with close monitoring for disease response well beyond discontinuation of treatment. Alternatively, pegylated liposomal doxorubicin is active in desmoid tumors with a lower risk of cardiac toxicity. (See 'Doxorubicin-containing combinations' above.)

●The weekly methotrexate and vinblastine combination is less toxic and is typically administered for one year.

For patients who have stable disease, it is unusual to see tumor shrinkage if no such benefit has been observed in the first six months of treatment. If a response is needed, it is reasonable to switch to other agents.

TREATMENT OF RECURRENT/RELAPSED DISEASE — Patients who develop a tumor recurrence or disease progression after effective systemic therapy has been discontinued can be retreated with systemic therapy (either the same or an alternative regimen) or evaluated for surgical resection or other ablative therapies. (See "Desmoid tumors: Epidemiology, molecular pathogenesis, clinical presentation, diagnosis, and local therapy", section on 'Management of recurrent disease'.)

Patients with a short progression-free interval (eg, progressive disease within three to six months) or toxicity associated with their previous systemic regimen may be offered an alternative regimen.

Patients with a longer progression-free interval (eg, one year or longer) and minimal toxicity on their prior systemic therapy may be retreated with the same regimen. Retreatment may produce responses that are similar to the original response [11].

We avoid radiation therapy (RT) in most patients with recurrent or relapsed disease. For younger patients, we do not use RT due to the potential long-term toxicities (such as tissue fibrosis and radiation-associated cancers) and the availability of other treatment options. Older adults (age >65 years) may be cautiously evaluated for RT if they progress on at least two different systemic therapies, have no other alternative local therapies (such as surgical resection, radiofrequency ablation, or microwave ablation), and are informed of the potential long-term toxicities of RT. (See "Radiation therapy techniques in cancer treatment", section on 'Radiation side effects'.)

INVESTIGATIONAL AGENTS

Tegavivint — Tegavivint is a TBL1 inhibitor that selectively degrades nuclear beta-catenin. This agent targets the Wnt signaling pathway, which is involved in the molecular pathogenesis of desmoid tumors. In a phase I trial of unresectable desmoid tumors, tegavivint demonstrated an ORR of 25 percent and was well tolerated [69]. (See "Desmoid tumors: Epidemiology, molecular pathogenesis, clinical presentation, diagnosis, and local therapy", section on 'Wnt/beta-catenin pathway and its role in desmoid tumors'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Soft tissue sarcoma".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Soft tissue sarcoma (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical presentation – Desmoid tumors (also called aggressive fibromatosis) are benign, slowly growing fibroblastic neoplasms with no metastatic potential but a propensity for local recurrence, even after complete surgical resection. Despite being histologically benign, they are locally infiltrative and can cause death through invasion and destruction of adjacent vital structures and organs. (See 'Introduction' above.)

●Asymptomatic or minimally symptomatic disease – Desmoid tumors can have an unpredictable clinical course, and observation is an acceptable strategy for well-selected patients with an asymptomatic or minimally symptomatic, radiologically nonprogressive tumor, particularly an intra-abdominal/mesenteric desmoid arising in a patient with familial adenomatous polyposis (FAP), or a collagenous mesenteric desmoid (ie, one that has low contrast uptake on contrast-enhanced magnetic resonance imaging [MRI]) that is not associated with FAP. (See 'Observation versus systemic therapy' above.)

●Indications for systemic therapy – Common indications for systemic therapy are as follows (see 'Indications' above):

•Unresectable desmoids that are symptomatic or asymptomatic but are progressing radiographically.

•Multiple locoregional recurrences despite adequate local therapy.

•Symptomatic desmoid tumors for which surgery would be unacceptably morbid.

•Initial nonsurgical treatment of a growing or symptomatic intra-abdominal/mesenteric desmoid tumor, particularly one arising in a patient with FAP.

●Options for systemic therapy – Systemic therapy may be broadly classified into cytotoxic chemotherapy and less toxic therapies, such as nirogacestat, tyrosine kinase inhibitors (TKIs), hormonal therapy, and nonsteroidal antiinflammatory drugs (NSAIDs). The choice of therapy depends on the urgency of the clinical situation, tumor site, symptomatology, clinical scenario, and patient and clinician preference. (See 'Urgency of treatment' above.)

●Initial treatment approach – Our general approach to initial systemic therapy is  summarized below:

•Life-threatening or severely symptomatic disease – For patients with an impending threat to either life or function because of tumor location or rapid radiographic progression, and for those who are severely symptomatic (ie, high urgency for treatment), we suggest cytotoxic chemotherapy (Grade 2C). Nirogacestat is a reasonable first-line alternative for selected patients if cytotoxic chemotherapy is not feasible or if patient preference is for starting with nirogacestat. (See 'Urgency of treatment' above.)

-In general, if cytotoxic therapy is chosen, pegylated liposomal doxorubicin, doxorubicin, combinations of doxorubicin plus dacarbazine, or methotrexate plus vinblastine are all reasonable options. Some institutions also use methotrexate alone, although there is almost no published experience utilizing this approach. (See 'Cytotoxic chemotherapy' above.)

•Other patients – For other patients, we offer the following:

-Asymptomatic patients with stable disease are observed, with treatment initiation for new symptoms or progression.

-For patients who are minimally symptomatic and/or have slowly progressive disease, we suggest tamoxifen (Grade 2C), to minimize side effects. (See 'Hormonal therapy' above and 'Tyrosine kinase inhibitors' above.)

-For patients who are moderately symptomatic or have more rapidly progressive disease, we suggest nirogacestat (Grade 2C), although TKIs, NSAIDs, or hormonal therapy are reasonable alternatives. This includes the management of patients with FAP who have intra-abdominal desmoid tumors that are large, are slow growing, involve the mesentery, or encase vessels or organs. However, in this setting, if the mesenteric tumor is impairing function of the gastrointestinal tract where oral absorption of treatment is unpredictable, cytotoxic chemotherapy may be a preferred approach. If patients are concerned regarding the risks for ovarian dysfunction on nirogacestat, a TKI may be used (eg, sorafenib, pazopanib, or sunitinib), or patients may be referred for egg harvesting. (See 'Indications' above.)

●Evaluating treatment response – Responses in patients treated with noncytotoxic agents can take months to manifest, and therapy should be continued until progression, especially when there is symptomatic improvement and treatment is well tolerated. If there are treatment-related side effects, treatment breaks can be incorporated into the treatment plan. (See 'Duration of therapy' above and 'Response evaluation during systemic therapy' above.)

●Duration of therapy – There are no data addressing the optimal duration of cytotoxic treatment, and there is no consensus on this issue. Dose-limiting toxicities tend to develop over time, and the duration of therapy will depend on the regimen:

•Doxorubicin-based regimens are limited by cumulative cardiac toxicity and myelosuppression. In responding patients, we generally limit treatment to six cycles of therapy, with close monitoring for disease response well beyond discontinuation of treatment. (See 'Anthracycline monotherapy' above and 'Doxorubicin-containing combinations' above.)

•The weekly methotrexate and vinblastine combination is less toxic and is typically administered for one year. (See 'Methotrexate with or without a vinca alkaloid' above.)

●Recurrent/relapsed disease – Patients who develop a tumor recurrence or disease progression after an effective therapy has been discontinued can be retreated with systemic therapy (either the same or an alternative regimen), or evaluated for surgical resection or other ablative therapies. (See 'Treatment of recurrent/relapsed disease' above.)

●Prior hormonal therapy – For patients in whom initial hormonal approaches have failed, we suggest nirogacestat rather than cytotoxic therapy, as long as there is no urgency to induce a response (Grade 2C). Alternative options include TKI therapy (sorafenib, pazopanib, or sunitinib) or an NSAID alone or the combined use of hormone therapy plus an NSAID. (See 'Antiinflammatory agents' above and 'Tyrosine kinase inhibitors' above and 'Combined hormone plus antiinflammatory therapy' above.)