The pharmacological treatment of aggressive fibromatosis: a
systematic review
Abstract
Background:
Despite the use of surgery and radiotherapy, 20–35% of patients with aggressive fibromatosis (AF) will have local recurrence. The purpose of this review was to collect and analyze all available information regarding the role of non-cytotoxic and cytotoxic chemotherapy in AF that has been accumulated over the past few decades.
Patients and methods:
A systematic review of published clinical trials, studies and case series was carried out using the Medline Express Databases and the Cochrane Collaboration Database from 1970 to October 2000.
Results:
Most studies published in the literature are in the form of successful case reports and single-arm series with small patient numbers. Most commonly used agents include hormonal agents, non-steroidal anti-inflammatory drugs (NSAIDs), interferons and cytotoxics. The literature data support the use of hormonal agents. Several questions, however, remain unresolved, such as which is the most suitable endocrine manipulation and what is the optimal dose and duration of treatment. NSAIDs and interferons have demonstrated activity against AF either alone or in combination with hormone therapy or chemotherapy but the precise mechanism of action is still unknown. Finally, there is growing evidence in the literature that chemotherapy is effective against AF with almost one in two patients being likely to respond.
Conclusions:
The evidence in the literature supports the opinion that both non-cytotoxic and cytotoxic chemotherapies are effective against AF. However, the lack of sufficient patient numbers and randomized trials compromises the validity of the reported results and mandates further investigation with properly designed prospective studies including larger patient numbers, with main end points to include not only tumor response rate and survival but also quality-of-life issues.
Key words
Received 25 March 2002 ; revised 16 August 2002; accepted 16 September 2002
Introduction
The term aggressive fibromatosis (AF) describes a broad group of benign fibrous tissue proliferations of similar microscopic appearance that are intermediate in their biological behavior between benign fibrous tissues and fibrosarcomas [1]. AF (also known as desmoid tumors, from the Greek word desmos for ‘band of tendons’) is a relatively rare lesion representing <3% of all soft tissue tumors with a reported annual incidence of 0.2–0.5 per 100 000 population [2–4]. The most frequent sites of involvement by these fibromatoses are the torso and the extremities. An important association of AF is reported in patients with familial adenomatous polyposis (FAP) (Gardner’s syndrome). Approximately 10% of patients with Gardner’s syndrome will develop mesenteric fibromatosis, which encases the mesenteric vasculature [5]. AF is characterized by low mitotic activity and a strong infiltrative growth pattern along tissue planes, with an ability to invade adjacent tissues [6, 7].
Surgery and radiotherapy are the principal modalities of therapy and their aim is curative, especially for extra-abdominal, localized, small-volume disease. Local control rates after surgery can vary considerably depending on surgical margin status. Response rates of 72% and 41% have been reported for tumor-free and tumor-positive margins, respectively. The addition of postoperative radiotherapy in an adjunctive fashion yields superior results. Local control rates of 94% and 75% were reported for tumor-free and tumor-positive margins, respectively [8]. Radiotherapy without an attempt at resection may also achieve good results. Sherman et al. [9], using megavoltage radiotherapy with a median dose of 50 Gy, and McCollough et al. [10] achieved local control rates of 75% and 83%, respectively. It is noteworthy that spontaneous regressions have also been reported, a fact that supports a wait-and-see policy after resection without wide margins [11].
Local control is often achieved at a considerable cost, due to significant treatment-related morbidity from the high doses of external radiation therapy (>50 Gy) or the disfiguring surgical procedures that are often required. In addition, despite the use of surgery and radiation therapy 20–36% of patients will show local recurrence [8–18]. Recurrent tumors can progress locally and occasionally lead to death; however, it is an uncommon outcome. Thus although classified as benign, recurrent AF can result in death due to local destruction in a small but important fraction of patients.
The potential morbidity of surgery and radiation therapy and the high local recurrence rates have lead investigators to assess the role of non-cytotoxic and cytotoxic chemotherapy in settings in which surgery and radiation therapy are either not possible or unsuccessful.
A number of issues should be considered in clinical trials assessing the role of non-cytotoxic and cytotoxic chemotherapy in AF. These include: criteria and duration of response; the importance of disease stabilization; measuring and dating progressions; documentation of progressive disease before starting treatment; heterogeneity of patients; heterogeneity of previous treatment; and the timing and indications for the investigated treatment. All of these issues are important shortcomings hampering the validity of the results reported.
Patients and methods
Data available on the topic AF in the English language literature were obtained, reviewed and analyzed. Published trials, studies and series were identified using Medline Express databases (National Library of Medicine, Washington DC, USA) from 1970 to October 2000 and the Cochrane Collaboration database. The keywords ‘aggressive fibromatosis’, ‘desmoid tumors’, ‘hormonal therapy’ and ‘chemotherapy’ were used. Cross-referencing, using the references of all identified studies, helped complement the computer-aided searches. Only cases that had adequate clinical information were included in the review. Basic clinical parameters that were required included patients’ age and sex, a history of FAP, tumor location (extra- versus intra-abdominal), tumor status (primary versus recurrent), previous therapy, type of therapy, response and duration of response.
Results
Non-cytotoxic agents
Hormonal agents. A number of observations regarding the natural history of desmoid tumors suggest that they may be hormonally responsive. Desmoid tumors are more frequent in women. About 80% of these tumors occur in women [4]. Reitamo et al. [2] observed that the speed of growth of desmoid tumors is higher during pregnancy, in premenopausal compared with postmenopausal women, and in females relative to males. In fertile women the growth rate is double that observed in men, suggesting that estrogens stimulate the growth of AF. The theory of hormone dependency has been supported also by preclinical and in vitrostudies. Lipschutz [18] described the formation of fibrous tumors histologically similar to desmoids in the abdominal organs, anterior abdominal wall, and thorax of guinea-pigs following prolonged estrogen administration. They also showed that it was possible to prevent these tumors by the administration of testosterone, progesterone and desoxycorticosterone [18, 19]. Estrogen receptors (ERs) have been demonstrated in desmoids of FAP patients, although the receptor levels were low [20, 21]. The presence of antiestrogen binding sites (AEBS) distinct from ERs probably plays an important role in modulating or mediating the action of antiestrogens. Furthermore, the presence of high levels of AEBS in desmoid tumors found to be ER-negative might explain responses to tamoxifen when the overall incidence and concentration of ERs is low in these tumors [22, 25].
Because of their rarity, there are no randomized studies conclusively proving that AF responds to endocrine manipulation. However, several single-arm trials and case reports document both stabilization and regression of desmoid tumors to hormones alone or in combination with other non-cytotoxic agents (Tables 1 and 2). A number of hormonal agents have been tested and found to be effective in AF. Such agents include tamoxifen and toremifene, progesterone, medroxyprogesterone acetate, prednizolone, testolactone and gosereline [26–44].
Table 1.
Antiestrogen therapy in patients with aggressive fibromatosis: single-arm trials and case reports
Table 2.
Other hormonal therapy in patients with aggressive fibromatosis: single-arm trials and case reports
One of the most commonly used antiestrogens in AF is tamoxifen, with many citations in the literature [26–32, 35–37]. Most of these reports, however, include single-case reports that have demonstrated either some sort of response or disease stabilization during therapy. Since larger series of patients are not available it is not possible to reach any firm conclusions regarding the effectiveness of tamoxifen against AF.
Triphenylethylenes other than tamoxifen also appear to be effective against desmoids as first- and second-line therapy. In a study by Brooks and co-workers [33], 20 patients with desmoid tumors of various sites were treated with toremifene at an average dose of 200 mg/day. With toremifene as first-line therapy (12 patients) the overall response rate was 50% (one complete and five partial responses) while an additional five patients (42%) had stabilization of their disease. Toremifene as second-line therapy (six patients) gave a response rate of 33% (two partial responses). One patient had disease stabilization.
Based on Lipschutz’s observations that the development of estrogen-induced fibrous tumors in guinea-pigs was prevented by the simultaneous administration of progesterone [18, 19], Lanari reported a series of 11 patients with mediastinal and retroperitoneal AF treated with progesterone at a dose of 100 mg/day [38]. There was a complete response in six patients (55%). Significant data, however, were lacking from this study, such as duration of response, the age and sex of the patients and whether treatment was first- or second-line. Furthermore, these results have not been confirmed by other investigators.
Another hormonal compound frequently tested was testolactone. Testolactone is not androgenic, estrogenic, progestational or antiprogestational in the usual sense but acts by irreversibly inhibiting aromatase, a rate-controlling enzyme responsible for the conversion of testosterone to estradiol and androstenedione to estrone [43]. The precise mechanism of action is unknown, although it is believed that testolactone acts through inhibition of cyclic AMP synthesis, which is known to influence growth and proliferation of fibroblasts [44]. The largest reported study on the activity of testolactone is that of Waddell and Kirsch in a series of 17 patients. Testolactone administered as a single agent gave an overall response rate of 40%, while combination with an anti-inflammatory non-steroidal drug (NSAID) such as sulindac or indomethacin yielded a response rate of 70% [41]. Most of the data available in the literature are on testolactone, which is regarded as a first-generation steroidal and irreversible inactivator of aromatase. Whether the second- and third-generation steroidal aromatase inactivators, such as formestane and exemestane, will yield superior results against AF remains to be seen in future trials.
Anti-inflammatory agents. The use of NSAIDs against AF was based on the surprising observation of total regression of a single recurrent desmoid tumor of the sternum in a patient taking indomethacin for radiation-induced pericarditis [45]. Furthermore, there is evidence that endogenous prostaglandin synthesis plays a role in neoplastic growth and that prostaglandin inhibitors can control the growth of experimental tumors [46, 47]. Subsequently, a variety of NSAIDs such as indomethacin and sulindac (a long-acting analog of indomethacin) were used to treat patients with desmoid tumors and were associated with partial and complete responses in several non-randomized retrospective studies (Table 3). NSAIDs have been tested either alone [23, 41, 48–50] or in combination with hormonal agents such as tamoxifen and testolactone [36, 37, 41]. Tsukada and co-workers assessed the efficacy of sulindac in 14 patients with a history of FAP and recurrent abdominal desmoid tumors [48]. The overall response rate was 57% and the majority of responders experienced a delayed response with a mean time of 24 months. Such delays in response have also been observed in patients treated with radiation therapy [10, 51] and with systemic chemotherapy [52–54]. Besides indomethacin and sulindac other NSAIDs such as colchicine have been tested against AF with success [50]. The response of AF to NSAIDs is an interesting phenomenon, the precise mechanism of which is still not clear and remains to be investigated.
Table 3.
Anti-inflammatory therapy in aggressive fibromatosis: case reports
Biological agents. There is theoretical background to support the use of interferons (IFNs) in AF. In vitro observations and animal experimental data suggest an antiproliferative effect of both IFN-α and retinoic acid on fibroblasts [55, 56]. Furthermore, IFN-γ has been shown to inhibit fibroblast collagen synthesis [57]. The clinical experience with the use of interferons in AF, however, is limited in comparison with endocrine manipulation or cytotoxic chemotherapy. Most published studies utilized IFN-α [58–61], while anecdotal cases report on the efficacy of IFN-γ [42,58]. From a total of nine patients with recurrent and measurable AF reported in the literature there were four responses (two with complete response) and the remaining five patients had stabilization of their disease lasting from 6 to 53 months (Table 4). Apart from mild toxicity related to IFN the treatment was well tolerated in all patients. There is also a suggestion that IFN-α can be used as adjuvant treatment for patients with completely resected AF in order to prevent further disease recurrence [61]. However, the number of patients analyzed so far is small and there is no comparison arm, therefore no conclusive statements can be made regarding the adjuvant role of interferons in AF. Whether IFN can be used as a non-cytotoxic alternative in patients in whom endocrine therapy fails remains to be addressed in a properly designed study.
Table 4.
Interferons in patients with aggressive fibromatosis
Cytotoxic chemotherapy
Single-arm studies utilizing cytotoxic chemotherapy in AF (Table 5) were reported as early as 1982, although case reports on the successful use of chemotherapy in AF date back to 1977 [62–78]. The earliest reports involved pediatric patients suffering from AF of the head and neck in whom physicians were reluctant to use radiation therapy because of the risk of epiphyseal destruction and the subsequent deformation of the irradiated area [62, 63]. Subsequent reports showed the efficacy of chemotherapy on AF involving various sites such as mesentery in combination with FAP, abdominal wall, extra-abdominal and extremity desmoid tumors. Assessment of response during the early trials was more difficult to ascertain due to the retrospective nature of the studies and the lack of modern imaging techniques at the time. Subjective improvement of symptoms and direct visualization of the tumor by physical examination or at laparotomy were serving as the principal assessment modalities. However, with the advent of computed tomography scanning and magnetic resonance imaging, objective response assessment has become more feasible in recent years.
Table 5.
Chemotherapy in patients with aggressive fibromatosis: single-arm studies
There is limited information on single-agent activity of chemotherapy against AF. Seiter and Kemeny [76] reported on the use of single-agent doxorubicin in a case report. There have been no single-arm trials utilizing single-agent therapy in the literature. On the other hand, combination chemotherapy regimens were employed more frequently in these patients and most information regarding the effectiveness of cytotoxic chemotherapy is mainly derived from them [52–54, 63–75]. Overall response rates to combination chemotherapy in single-arm studies range between 17% and 100% with a median response rate of 50%. Most commonly used regimens included doxorubicin-based chemotherapy (doxorubicin with dacarbazine or doxorubicin with cyclophosphamide and vincristine) [52, 63, 67–69, 71, 76], actinomycin-D-based chemotherapy [52, 64–66] or a combination of methotrexate with a vinca alkaloid (vinblastine or vinorelbine) [53, 70, 72, 73]. Doxorubicin-based chemotherapy is effective with responses similar to or even higher than those achieved in soft tissue sarcomas. The dose of doxorubicin ranged between 60 and 90 mg/m2/cycle. This combination, however, had considerable acute and late toxicity, mainly in the form of cardiotoxicity. To avoid this, some investigators discontinued doxorubicin or substituted it with carboplatin once a cumulative dose of 400–500 mg/m2 was reached [77]. A reasonable alternative to doxorubicin-based chemotherapy is the VAC (vincristine, actinomycin-D, cyclophosphamide) regimen with comparable responses. Main concerns with VAC chemotherapy are those of sterility and carcinogenesis, especially when treatment is administered to younger patient populations for protracted periods. The combination of methotrexate with vinblastine administered on a weekly basis was proposed as a less toxic alternative to doxorubicin-based and VAC chemotherapy. This is especially true for pediatric patients and young patients of child-bearing potential in whom treatment would have to be administered for prolonged periods [53]. Weiss and Lackman [53] and Skapec [54] reported separately that the combination was effective in both adult and pediatric patients with minimal side-effects. Another report, however, by Azarrelli et al. demonstrated that weekly treatment was not feasible due to myelotoxicity and hepatotoxicity and the mean interval between cycles was found to be 15 days [70]. In a recent study the substitution of vinorelbine for vinblastine resulted in significantly less neurotoxicity without a compromise in response rates [72].
A number of reports suggest that the response of AF to chemotherapy may be slow [52–54]. This may be due to the unique pathological features of this tumor, which is characterized by the presence of abundant collagen tissue, scanty malignant cells, especially in the center of the tumor, and rare mitoses. A number of investigators observed no evidence of response in some patients for several months under cytotoxic chemotherapy. Pilz et al. recommends that chemotherapy should not be discontinued before the 20th week of treatment [52], while Skapec suggests that treatment should be continued for prolonged periods from 12 to 18 months even if there is no radiographic evidence of response [54]. Azzarelli et al. found that any response to treatment was not evident before 8–12 weekly cycles of treatment [70]. Similar delays in response have been shown in the case of radiation therapy, where response of desmoid tumors may be delayed by as much as 8–27 months [10, 57].
Regional chemotherapy. Regional chemotherapy in the form of isolated limb perfusion (ILP) is another alternative to systemic therapy in patients with limb desmoids. This is a particularly attractive method if one considers that AF rarely metastasizes. The first report on ILP in AF was published in 1989 by Klaase et al. [75]. In this study, ILP with melphalan and doxorubicin obtained three complete responses in seven patients for an overall response rate of 43% and duration of response ranging between 9 and 11 years. One patient with stable disease after ILP had been in a stable condition for 6 years at the time of the report. ILP with melphalan and tumor necrosis factor was recently attempted in six patients with recurrent limb desmoid tumors [74]. The overall response rate was 83% with 33% (two of six) complete responses and 50% (three of six) partial responses. The limb salvage rate was 100%. This method is primarily useful in patients with locally advanced or recurrent tumors not amenable to conservative resections. The high complete response rates achieved by this method mandates further confirmation in well-designed studies.
Discussion
An endocrine approach to the management of AF is a rational one. However, the true role of endocrine therapy still remains unknown. There does not appear to be a correlation between response to tamoxifen and ER status. On the other hand a significant number of AF tumors exhibit AEBS. Whether levels of AEBS in AF can be predictive for a clinical response to tamoxifen or toremifene is an issue that has not been investigated yet. Further issues yet to be addressed include which is the most appropriate endocrine manipulation, at which dose and for how long. Newer hormonal compounds such as the second and third generation steroidal aromatase inactivators may yield superior results against AF, but this remains to be seen in future trials.
Other non-cytotoxic agents that appear to be active against AF either alone or in combination with cytotoxic chemotherapy include the NSAIDs and interferons. NSAIDs have been tested either alone or in combination with hormonal agents. The response of AF to NSAIDs is an interesting phenomenon, the precise mechanism of which is still not clear. The use of IFN as a non-cytotoxic alternative for patients in whom endocrine therapy fails is an issue that remains to be addressed in a properly designed study.
The role of chemotherapy in the treatment of desmoid tumors has not been adequately studied for several reasons: (i) They are relatively rare tumors that often respond to other treatment modalities such as surgery and local radiotherapy. These tumors, however, despite aggressive local therapy, tend to recur locally and frequently surgical removal or re-irradiation is no longer a reasonable option. Other medical treatment modalities such as hormonal manipulation or NSAIDs appear to be effective in controlling or even reversing the growth process of recurrent AF but they are effective only in a small percentage of patients and the effect is not long-lasting. (ii) There is a general misconception that chemotherapy will not be effective in this low-grade tumor, which is largely acellular, and (iii) many investigators are reluctant to propose cytotoxic chemotherapy because of concerns regarding the toxicity and its impact on the patients quality of life. With the recent advances in supportive care, however, especially after the introduction of the newer antibiotics, antiemetics, and growth factors, this concern should become less of an issue. Furthermore, the emotional and functional well-being in the long run should take precedence over short-term toxicities.
There are certain situations where chemotherapy should be considered as a reasonable option in AF such as: (i) In those situations in which aggressive surgical resections in the form of limb-sacrificing surgical procedures, hemimandibulectomies, hemipelvectomies, or chest wall resections could result in severe disfigurement [13, 78]. The psychological impact of these procedures is even more compounded when applied to younger patients. (ii) In patients with large mesenteric or retroperitoneal tumors encasing vital structures such as vessels, nerves or ureters. In these situations surgery or radical radiation therapy is either not feasible or excessively morbid. (iii) In patients with tumors recurring after use of other non-cytotoxic treatment such as hormonal agents or NSAIDs.
Most information on the use of cytotoxic chemotherapy is primarily based on combination chemotherapy regimens. There is substantial evidence in the literature that chemotherapy is effective against AF. Almost one in two patients with AF is likely to respond to cytotoxic chemotherapy. This is quite a satisfactory response since chemotherapy is often considered as a last resort in the management of AF, and the patients selected often have advanced inoperable or recurrent disease after combined therapeutic modalities such as surgery and radiation, presenting with severe complications. Although there is no general consensus, the data suggest that chemotherapy should be continued for a prolonged period (at least 1 year) unless there is evidence of disease progression or severe treatment-related toxicity, even if there is no radiographic evidence of response. These results should be interpreted with some caution, however, for two reasons. First, most trials have not been designed in a prospective fashion and secondly the numbers of participating patients are small. Nevertheless, the finding of considerable tumor regression or even complete disappearance with a long-lasting effect in several studies is encouraging. Still, most conclusions on the role of non-cytotoxic and cytotoxic chemotherapy in AF are logical, based primarily on incomplete evidence in the form of low-powered phase II studies or case reports, the majority of which are supportive of the role of various agents against AF. High-level evidence in the form of randomized clinical trials or meta-analyses to support the role of non-cytotoxic and cytotoxic chemotherapy in AF is lacking.
The identification of CD117 expression in AF has raised the question as to whether there is a role for imatinib in this disease. Currently there is insufficient evidence to support its use but hopefully its role will be explored in well-designed clinical trials.
Systemic treatments should be considered in patients with AF for whom local treatment approaches have failed. Documentation of disease progression and/or assessable symptoms before the initiation of systemic chemotherapy is essential. All patients should be included in clinical trials whenever possible. In the absence of an appropriate clinical trial, initiation of treatment in a stepwise manner may be appropriate. The first logical step of systemic treatment should include an NSAID such as sulindac or indomethacin. If this fails and providing that the tumor’s growth rate is not rapid the second step should be with tamoxifen. If despite this AF continues to progress, cytotoxic chemotherapy should commence, the combination of methotrexate and vinblastine being the regimen with best results at lowest morbidity. The duration of any of these treatments should be based on tolerance and evidence of continued clinical benefit.
It should be stressed, however, that AF is a variable disease with several different clinical entities existing within the same label (abdominal, extra-abdominal, extremity AF, etc.). In addition, several other factors, such as the high rate of spontaneous disease stabilization may confound analysis of chemotherapy effectiveness. The primary focus of this review was to present a comprehensive analysis of activity of different agents in AF. The final decision of how to incorporate the above information in everyday clinical practice should be a rational one based on patient characteristics and the tumor’s biological properties.
Footnotes
- ↵+ Correspondence to: Dr J. Janinis, 17 Kapodistriou Street, Pefki 15121, Athens, Greece. Tel: +30-10-6126582; Fax: +30-10-6126582; E-mail: janinis@internet.gr
References
- 1.↵
- 2.↵
- 3.
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.
- 13.↵
- 14.
- 15.
- 16.
- 17.
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
Nenhum comentário:
Postar um comentário