By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Presented, in part, at the 48th Annual Meeting of ASTRO, Philadelphia, Pennsylvania, November 5–29, 2006.
Myxoid liposarcoma has been reported to be more radiosensitive compared with other soft tissue sarcomas (STS). The authors report the results of multidisciplinary treatment of extremity myxoid liposarcoma compared with a contemporary cohort of other STS subtypes with an emphasis on the role of radiotherapy (RT) in improving local control.
Between 1989 and 2004, 691 patients were identified from a prospective STS database who underwent combined management for localized extremity STS and were followed for a minimum of 12 months or until death. All patients underwent surgery together with pre or postoperative RT, depending on their presenting characteristics and resection margins. Demographics and outcomes were compared between patients with myxoid liposarcoma and other STS subtypes (other-STS).
Of 691 patients, 88 patients had myxoid liposarcoma and 603 had other STS subtypes (other-STS). Median age was 48 and 60 years for the myxoid liposarcoma and other-STS groups, respectively. Median follow-up was 86 and 61 months, respectively. For myxoid liposarcoma and other-STS groups, preoperative RT was used in 57% versus 61% of patients and postoperative RT in 43% versus 39%, respectively. The 5-year local recurrence-free survival was 97.7% for patients with myxoid liposarcoma compared with 89.6% for patients with other-STS tumors (P = .008). High-grade tumors were present in 7% and 59% of myxoid liposarcoma and other-STS patients, respectively (P = .0003). Two myxoid liposarcoma patients with local recurrence had positive resection margins, whereas only 33% of patients in the other-STS group who developed a local recurrence had positive resection margins. No patients with myxoid liposarcoma required amputation as primary management, whereas 8 (1.3%) required amputation as primary management in the other-STS group. Systemic disease control was superior in myxoid liposarcoma compared with other-STS patients, with 5-year overall and metastasis-free survival rates of 93.9% versus 76.4% (P = .0008) and 89.1% versus 66.0% (P = .0001) respectively. Of 12 myxoid liposarcoma patients with distant metastases, 7 appeared in nonpulmonary sites. In comparison, 205 (34%) patients with other-STS tumors developed systemic disease but 78% had pulmonary metastases.
Soft tissue sarcomas (STS) of the extremity are a heterogeneous group of tumors with differing pathologies that clinically behave in a similar fashion. With the exception of certain pediatric and chemotherapy-sensitive histological subtypes, surgery plays a predominant role in the local management of extremity STS.1 Radiotherapy (RT) has been shown to be an important adjunctive treatment and improves local control rates over those seen with conservative surgery alone, particularly when ‘wide’ resection margins cannot be obtained.2, 3 Local control in extremity STS would be expected to be around 90% overall with this approach.2
Myxoid liposarcoma is a common variant of liposarcoma and has a characteristic t(12:16) translocation that is shared with myxoid/round cell liposarcoma (a higher grade variant with more aggressive biology4). Myxoid liposarcoma has a peculiar characteristic of extrapulmonary metastasis5 not commonly seen in other STS subtypes but appears to have a relatively favorable prognosis.6, 7 Although some studies have suggested that myxoid liposarcoma may be relatively responsive to RT more than other types of STS, with high rates of regression and even reports of complete clinical response,8, 9 others have reported high rates of local relapse overall; but, it is unclear that all appropriate patients received combined management with surgery and RT in these studies.10, 11 Pitson et al previously demonstrated an objective response of myxoid liposarcoma to RT in the preoperative setting.12 Thus adjunctive RT might translate into improvement in local control for myxoid liposarcoma patients compared with other adult STS subtypes treated in a similar manner.
The aim of this retrospective study was to compare the outcomes of multidisciplinary treatment of a cohort of myxoid liposarcoma patients with a contemporary group consisting of other STS subtypes (other-STS) treated in a standardized manner with particular emphasis on the role of RT in local control.
METHODS AND MATERIALS
The research ethics board approved this study. From our prospectively collected institutional database, 691 patients who had surgical treatment and RT at our center, between 1989 and 2004, for localized extremity STS were identified. Patients were excluded if: chemotherapy was a part of their treatment (n = 75), they had metastatic disease at diagnosis, they had either dermatofibrosarcoma protuberans or well-differentiated liposarcoma. Patients who underwent initial surgery elsewhere before referral, usually in the form of an unplanned excision,13, 14 and were referred for definitive management, were included if they were treated with further conservative surgery and RT at our center. In this setting, the ‘management policy’ evolved over time but currently, patients would generally have preoperative RT and re-excision unless it was anticipated that excision with wide margins could be achieved and, thus, RT was not likely to be needed. All patients had their pathology reviewed to confirm the diagnosis and subtype of STS. The grade of STS was assigned according to a 3-tier system. Myxoid liposarcoma patients were assigned a minimum of grade 2, and if there was evidence of 5% or more round cell component within the specimen, then they were assigned grade 3. Staging included full history and examination together with cross sectional imaging of the extremity (usually magnetic resonance imaging [MRI]). Thoracic computed tomography (CT) scans were obtained and other imaging was done at the discretion of the clinician. Patients with locally recurrent disease and no distant metastases were included as long as the intent of management was conservative surgery and RT.
Patients were treated with limb conserving surgical methods whenever possible. RT was used as an adjunct whenever surgical resection could not be achieved with wide negative margins (usually defined as a minimum of 2 cm). Resections were categorized as complete excision with microscopic negative margins (R0), excision with microscopic positive margins (R1) or excision with gross positive margins (R2). External beam RT was delivered in the preoperative setting (50 Gy) or postoperatively (60-66 Gy). In some cases, in the setting of positive margins after preoperative RT and surgery, a postoperative boost of up to 16 Gy was delivered. The timing of RT evolved over the period under study as postoperative RT was more commonly delivered up to 1994. From 1994-1997, a national randomized clinical trial (NCIC SR2) of pre versus postoperative RT was conducted and influenced the timing of RT for patients who participated in this trial. Subsequent to this, preoperative RT became more common. After treatment, patients were followed on a standard schedule every 3 months to 6 months for the first 3 years (high-grade patients were seen every 3 months), 6 monthly until Year 5, and then annually until 10 years. At each follow-up visit, physical examination and chest x-ray were performed, and other imaging was done at the discretion of the physician. Minimum follow-up was 12 months.
Local recurrence-free survival, metastasis-free survival, and overall survival were estimated using the Kaplan-Meier method. Survival time was calculated from the date of definitive surgery for the primary tumor. Differences in survival were compared using the log-rank test. A p-value of <.05 was considered to be statistically significant.
Of 691 patients, 88 had myxoid liposarcoma and 603 had other-STS pathology. Patient and tumor characteristics are shown in Table 1. Median age of the 2 groups of patients was 48 years (myxoid liposarcoma) and 60 years (other-STS), respectively. Overall, 17% (118 of 691) of tumors were superficial and 83% (573 of 691) were deep, with similar proportions in both groups. Myxoid liposarcoma patients had a round cell component in 7% of cases. Patients in the other-STS group had grade 3 of 3 tumors in 59% of cases. Median tumor size was 8 cm and 7.5 cm for the myxoid liposarcoma and other-STS groups, respectively.
Table 1. Patient and Treatment Characteristics
Myxoid Liposarcoma n = 88
All Other STS n = 603
STS indicates soft tissue sarcoma.
For grade comparison, grades 1 and 2 were combined into low grade and compared with grade 3 tumors (high grade).
Of patients in the myxoid liposarcoma group, 34% (30 of 88) had prior surgery elsewhere compared with 40% (241 of 603) of the other-STS group. Macroscopic complete resection was achieved in over 99% of the patients in both groups with positive microscopic margins (R1) in 19% and 23% of the myxoid liposarcoma and other-STS groups, respectively. Amputation was not required for any patients in the myxoid liposarcoma group but was necessary for 8 (1.3%) patients in the other-STS group.
In the myxoid liposarcoma group, RT was delivered preoperatively, preoperatively with postoperative boost, and postoperatively in 47%, 10%, and 43%, respectively. The corresponding RT treatments in the other-STS group were 52.5%, 8.5%, and 39%, respectively.
Median follow-up was 86 months for the myxoid liposarcoma group and 61 months for the other-STS group. The main oncological outcomes are shown in Table 2.
Table 2. Oncologic Outcomes
Myxoid Liposarcoma, n = 88
All Other STS, n = 603
STS indicates soft tissue sarcoma; DOD, dead of disease; LR, local recurrence.
Alive, no evidence of disease
Alive with evidence of disease
Deceased (nonsarcoma related)
5-Year overall survival
Preoperative + postoperative boost
5-Year LR-free survival
5-Year metastasis-free survival
Overall survival at 5 years (Fig. 1) was 93.9% for the myxoid liposarcoma group and 76.4% for the other-STS group (p < .01). Not surprisingly, metastatic disease developed more frequently in patients who were in the other-STS group, in keeping with the larger proportion of patients with high-grade disease in that group. Metastasis-free survival at 5 years (Fig. 2) was 89.1% and 66.0% for the myxoid liposarcoma and other-STS groups, respectively (P < .01).
Overall, 5-year local relapse-free survival was 97.7% for the myxoid liposarcoma group and 89.6% for other-STS patients.
When patients with locally recurrent disease were excluded, local recurrence-free rates were 98% for the myxoid liposarcoma group and 91% for the other-STS group (P = .02) at 5 years.
Of patients with local recurrence, both in the other-STS group had positive margins (and both patients had grade 2 disease), whereas 33% (20 of 60) in the other-STS group had positive margins.
Overall, 12 (13.6%) myxoid liposarcoma developed metastatic disease and 2 had multiple sites involved including mediastinum, spine, lung, and bone. Of 10 myxoid liposarcoma patients who developed a solitary site of metastatic disease, 4 had pulmonary metastases, 2 intra-abdominal disease, 3 a soft tissue metastasis, and 1 a spinal bone metastasis. Subsequently, 6 of these patients developed further metastases. In contrast, of those in the other-STS group who developed metastases, 78% (160 of 205) had pulmonary metastases.
Patients with extremity STS, managed at our center with combined surgery and RT, had excellent local control rates (∼90%) that compare favorably with other large series reported in the literature.15 In a recently published series of 279 patients undergoing primary treatment for STS, the local recurrence rate was reported to be 16%.16
Although the myxoid liposarcoma and other-STS groups were comparable, the overall 5-year local disease control for patients with myxoid liposarcoma treated with surgery and RT was exceptional. We hypothesized that the radiosensitivity of myxoid liposarcoma would facilitate improved local control in patients treated with combined surgery and RT. The results of this study support our theory as myxoid liposarcoma patients treated with (neo)adjuvant RT had 98% 5-year local control compared with 90% for other-STS patients.
Reports of patients with well-differentiated liposarcoma, which generally requires surgery alone without RT, and pleomorphic liposarcoma, which may have a higher rate of local recurrence, even when treated with adjuvant RT, have been included together with myxoid liposaroma.17, 18 Myxoid liposarcoma may be distinct from other types of STS, and although some historical series reported high rates of local recurrence (up to 33%) in patients treated with surgery alone,6, 19 other small series have noted good local control with the addition of RT.17, 20, 21 In 1 such study of 10 patients, even in the presence of marginal or intralesional excisions without further surgery, the addition of RT resulted in excellent local control with no local relapses at mean follow-up of 58 months, suggesting that the radiosensitivity of the tumor improved local control.22 Another more recent series examining combined modality therapy reported 100% local control in 9 patients with positive margins, 99% local control in 91 patients with negative margins, and only 87% local control in 25 patients where the margin status was unknown.23 In our series, we did not have any patients with gross positive margins (R2 resections) in the other-STS group, but there were 7 (1.2%) patients within the other-STS group who had R2 resections. We were not able to distinguish patients with narrow but microscopically negative margins of excision compared with those with a more widely negative excision as all were coded as R0. Of myxoid liposarcoma patients in our study, 17 had positive margins, of whom only 2 had local recurrence (compared with 20 of 60 of other-STS patients). Although the local control in myxoid liposarcoma patients is excellent, we believe our data confirm the need for adequate local surgery resulting in negative margins whenever possible because only those myxoid liposarcoma patients with positive resection margins recurred locally.
Although the sensitivity of myxoid liposarcoma to RT was first reported more than 70 years ago and has been described by others,8, 24, 25 only recently has objective evidence of this response been reported. Pitson et al compared patients with myxoid liposarcoma to a group with malignant fibrous histiocytoma who were both treated with preoperative RT.12 Using MRI, before and after RT, a proportional reduction of 59% in median tumor volume was measured in myxoid liposarcoma patients. The mechanism by which this volume reduction occurs is unknown but could be due to a particular cellular susceptibility or loss of tumor stroma. The TLS-CHOP (FUS-DDIT3) translocation, which is commonly present in myxoid liposarcoma, and the resulting fusion protein play a central role in formation of these tumors. These mutant fusion proteins involving CHOP fail to allow growth arrest in response to radiation and may, thereby, lead to the radiosensitivity typical of myxoid liposarcoma tumor cells. It must be noted that while the presence of this translocation may aid the diagnosis of myxoid liposarcoma, it has not been shown, at least thus far, to have prognostic significance.26 This tendency to respond to RT is not only important when planning combined treatment, but it appears to increase the likelihood of local control for myxoid liposarcoma compared with other STS subtypes.
Many of our patients were treated with preoperative (neoadjuvant) RT, which has been shown to result in the same level of local control as postoperative (adjuvant) RT in STS patients.27 An advantage of preoperative RT over postoperative RT may be to limit late morbidity, particularly in patients with large tumors as the smaller treatment volumes and lower doses delivered in preoperative RT may result in better functional outcome with less fibrosis28 as well as reduction in risk of bone fracture.29 Although the use of preoperative RT in STS patients has not been universally adopted, its use in myxoid liposarcoma patients may be preferable for several reasons. First, preoperative RT would be expected to cause a large reduction in tumor volume, which may, in turn, allow less morbid surgery, resulting in improved functional outcomes without compromising local control. Second, preoperative RT facilitates surgical sparing of critical structures adjacent to the tumor margin. Even in the setting of “planned” positive margins after preoperative RT, where the tumor does not shrink away from a vital structure after preoperative RT, for example, after epineural dissection of a major motor nerve, local control rates do not appear to be significantly reduced.30 This may be particularly relevant in the setting of myxoid liposarcoma as such tumors frequently arise in intermuscular locations and may involve neurovascular bundles.31 An example of the type of response that may be seen with preoperative RT in myxoid liposarcoma is shown in Figure 3.
Patients with myxoid liposarcoma generally have a good prognosis and have the potential to gain maximum benefit from strategies that minimize the long-term consequences of RT. With modern RT techniques, such as intensity-modulated radiotherapy, further benefit might be derived from reducing the normal tissue volume that receives high doses while maintaining optimal local control. This may be advantageous in the preoperative setting, but it may be more applicable in the postoperative setting since limiting the volume of normal tissue exposed to the higher doses of RT used with the aforementioned highly conformal RT techniques might be expected to reduce the severity of late toxicity in these patients.32
As with all retrospective data, patient selection significantly influences outcome, as is the case in our 2 patient cohorts. One might speculate that including patients with locally recurrent disease may reduce the apparent local control rate for other-STS patients, especially because this group had a larger proportion of patients with locally recurrent disease at initial presentation. However, even after excluding these patients, the local control rate was ∼ 90% for the other-STS group. For patients who underwent ‘unplanned’ excision of a sarcoma elsewhere with positive margins before referral, our policy is for further re-excision as these patients are at significant risk of local recurrence even with the addition of RT. Our group and others have reported high rates (59%) of residual disease following ‘unplanned’ excisions, which provides evidence to support this treatment algorithm.13, 14, 33, 34 We do not believe that this should be a significant confounding factor as the proportion of patients referred after ‘unplanned’ excisions and their subsequent management was similar in both patient groups.
The unusual metastatic disease pattern in myxoid liposarcoma has previously been documented with predilection for intra-abdominal/retroperitoneal sites6, 7, 35, 36 in addition to osseous metastases most commonly involving the spine.5 In our series, metastatic disease developed in 14% of patients, and over half of these were in nonpulmonary sites. Although the pattern of distant metastasis is not unusual for this disease, the incidence was relatively low in our myxoid liposarcoma patients. This may be a reflection of the small number of patients with a round cell component (which appears to confer greater metastatic potential21) in our cohort, but it could also be related to the high rate of local disease control. One investigation of patients with liposarcoma reported a 22% incidence of distant metastasis in myxoid liposarcoma compared with over 40% in pleomorphic liposarcoma.18 Given the relatively atypical metastatic pattern in myxoid liposarcoma compared with other sarcomas, it has been recommended that patients with myxoid liposarcoma should have routine staging and follow-up to include abdominopelvic CT scans and imaging of the spine.5, 23
In summary, we found excellent rates of local control for STS patients treated with surgery and RT. Patients with myxoid liposarcoma treated at our center had significantly lower rates of local and distant failure compared with other STS subtypes. Improved local control in myxoid liposarcoma might well be due to its radiosensitivity, while a lower rate of metastasis may be a reflection of the relatively low numbers of patients with high-grade disease in the myxoid liposarcoma cohort as well as the high rate of local control. It has been recognized that particular subtypes of STS may have specific characteristics; myxoid liposarcoma is 1 such subtype, and the data presented here suggest that its radio-responsiveness translates into excellent local control when RT is appropriately applied.