National Cancer Institute
Last Modified: September 14, 2012
Estimated new cases and deaths from soft tissue sarcoma in the United States in 2012: 1
Soft tissue sarcomas are malignant tumors that arise in any of the mesodermal tissues of the extremities (50%), trunk and retroperitoneum (40%), or head and neck (10%). The reported international incidence rates range from 1.8 to 5 per 100,000 per year. 2
The risk of sporadic soft tissue sarcomas is increased by prior radiation therapy and, in the case of lymphangiosarcoma, by chronic lymphedema. The chemicals Thorotrast, vinyl chloride, and arsenic are also established carcinogens for hepatic angiosarcomas. 3 4 5
Soft tissue sarcomas may be heterogeneous, so adequate tissue should be obtained via either core-needle or incisional biopsy for microscopic examination to determine histologic type and tumor grade. Careful planning of the initial biopsy is important to avoid compromising subsequent curative resection. Since the selection of treatment is determined by the grade of the tumor, it is essential to have a careful review of the biopsy tissue by a pathologist who is experienced in diagnosing sarcomas. Complete staging and treatment planning by a multidisciplinary team of cancer specialists is required to determine the optimal treatment for patients with this disease.
There is evidence that at least some favorable clinical outcomes may be associated with referral to a specialized sarcoma treatment center. In a population-based consecutive series of 375 soft tissue sarcoma patients in Sweden, local recurrence rates of resected tumors were higher in patients who were not referred to the specialized center: in 35 of 78 (45%) patients not referred; in 24 of 102 (24%) patients referred after initial surgery or incisional biopsy; and in 36 of 195 (18%) patients referred prior to any surgical procedure (P = .0001 for the difference between those never referred vs. those referred prior to any surgical procedure). 6[Level of evidence: 3iDii] However, there were no statistically significant differences in death from sarcoma between the groups of patients.
Factors associated with a poorer prognosis include the following: 8
Although low-grade tumors are usually curable by surgery alone, higher-grade sarcomas (as determined by the mitotic index and by the presence of hemorrhage and necrosis) are associated with higher local-treatment failure rates and increased metastatic potential.
Soft tissue sarcomas are classified histologically according to the soft tissue cell of origin. Additional studies, including electron microscopy, specialized immunohistochemistry, flow cytometry, cytogenetics, and tissue culture studies may allow identification of particular subtypes within the major histologic categories. For example, S100 antigen suggests neural sheath origin, cytokeratin suggests epithelioid or synovial cell origin, and factor VIII-related antigen suggests endothelial origin. Likewise, some subtypes of sarcomas have characteristic genetic markers, but these markers are not generally used in the routine clinical setting (e.g., t(X;18)(p11;q11) in synovial sarcomas and t(12;16)(q13;p11) in myxoid and round-cell sarcomas). 1 2 3
The histologic grade reflects the metastatic potential of these tumors more accurately than the classic cellular classification listed below. Pathologists assign a grade based on the number of mitoses per high-powered field, the presence of necrosis, cellular and nuclear morphology, and the degree of cellularity; discordance among expert pathologists regarding tumor grade, and even histologic subtype, can be substantial. 4
[Note: *It is recognized that dedifferentiated liposarcoma primarily arises in the context of deep atypical lipomatous tumor/well-differentiated liposarcoma, a sarcoma of intermediate malignancy because of the lack of metastatic capacity. **The category of fibrosarcoma can be inclusive of fibrosarcomatous differentiation in dermatofibrosarcoma protuberans. ***Cutaneous angiosarcoma may be difficult to stage using the AJCC system. (Refer to the PDQ® summary on Gastrointestinal stromal tumor [GIST] summary for more information.]
Staging has an important role in determining the most effective treatment for soft tissue sarcoma. Clinical staging involves magnetic resonance imaging (MRI) or computed tomography (CT) of the primary tumor area and a chest CT to look for metastasis to the lung (the most common site of distant spread). An abdominal CT scan is done in the case of retroperitoneal sarcomas because the liver may be the site of initial clinical metastasis for these tumors.
The stage is determined by the size of the tumor, the histologic grade, and whether there is spread to lymph nodes or distant sites. Intracompartmental or extracompartmental extension of extremity sarcomas is also important for surgical decision making. For complete staging, a thorough review of all biopsy specimens (including those from the primary tumor, lymph nodes, or other suspicious lesions) is essential. CT scan of the chest is recommended for sarcomas larger than 5 cm (T2) or with moderate to poor differentiation (grades 24). Nodal involvement is rare, occurring in less than 3% of patients with sarcoma. 1
Lymph node involvement in soft tissue sarcomas of adulthood is rare but is somewhat more frequent in some subtypes (e.g., rhabdomyosarcoma, vascular sarcomas, clear cell sarcomas, and epithelioid sarcomas) when they are high grade. 2 Because treatment decisions are predicated on pathology staging, patients should be staged before, and again after, any neoadjuvant therapy. The assessment of tumor grade can be affected in either direction, but more frequently decreased because of differential cellular loss related to the neoadjuvant chemotherapy or radiation. 3 Grade, which is based on cellular differentiation, mitotic rate, and extent of necrosis, should be recorded for all soft tissue sarcomas. A three-grade system (G1G3) is preferred. (See Table 4 below).
The American Joint Committee on Cancer (AJCC) has designated staging by the four criteria of tumor size, nodal status, metastasis, and grade (TNMG). 3 The characteristic molecular markers of some sarcomas are not formally incorporated in the staging system pending further evaluation of their impact on prognosis. Recurrent sarcomas are restaged using the same system as for primary tumors with the specification that the tumor is recurrent.
|TX||Primary tumor cannot be assessed.|
|T0||No evidence of primary tumor.|
|T1||Tumor 5 cm in greatest dimension. (Size should be regarded as a continuous variable, and the measurement should be provided.)|
|T2||Tumor >5 cm in greatest dimension.b|
|NX||Regional lymph nodes cannot be assessed.|
|N0||No regional lymph node metastasis.|
|N1b||Regional lymph node metastasis.|
|M0||No distant metastasis.|
|Stage IA||T1a||N0||M0||G1, GX|
|Stage IB||T2a||N0||M0||G1, GX|
|Stage IIA||T1a||N0||M0||G2, G3|
|Stage III||T2a, T2b||N0||M0||G3|
|Any T||N1||M0||Any G|
|Stage IV||Any T||Any N||M1||Any G|
Neurovascular and bone invasion are indicators of poor prognosis, but they are not incorporated into the formal staging system.
In most cases, a combined modality approach of preoperative radiation therapy (preRX) or postoperative radiation therapy (PORT) is used, rather than the radical surgical procedures, such as amputation, that were used in the past. It may even be possible to use surgery without PORT in selected cases. For example, a case series was reported from a specialized sarcoma treatment referral center in which 74 selected patients with primary extremity and trunk tumors 5 cm or less in size were found to have no histologic involvement of the surgical margins. The patients were observed without radiation therapy, and the estimated local recurrence rate after 10 years was 11%. 1[Level of evidence: 3iiiDiv] The role of chemotherapy is not as well defined as is the role for radiation therapy. Because of the evolving nature of the treatment options for this disease, patients should be offered the option of clinical trials when available. Information about ongoing clinical trials is available from the NCI Web site.
Surgical resection is the mainstay of therapy for soft tissue sarcomas. When feasible, wide-margin functionsparing surgical excision is the cornerstone of effective treatment for extremity tumors. This may be facilitated by soft tissue reconstructive surgery, which generally permits wider margins than those obtained when the surgical plan involves direct closure of the excision site. 2 Cutting into the tumor mass or shelling out the gross tumor along the plane of the pseudocapsule of compressed tumor cells and reactive tissue that often surrounds soft tissue sarcomas are associated with an elevated risk of local recurrence. Even high-grade soft tissue sarcomas of the extremities can usually be effectively treated while preserving the limb with combined-modality treatment consisting of preRX or PORT to reduce local recurrence. (Refer to the Role of Radiation Therapy section of this summary for more information.)
Only one small, single-institution randomized trial has directly compared amputation to limb-sparing surgery for soft tissue sarcomas of the extremities. 3 In a 2:1 randomization ratio, 27 patients with high-grade extremity sarcomas were assigned to a wide excision plus PORT (45 Gy50 Gy to the wide local excision area, and a total of 60 Gy70 Gy to the tumor bed over 67 weeks), and 16 were assigned to amputation at or above the joint proximal to the tumor. Both groups received adjuvant chemotherapy (i.e., doxorubicin, cyclophosphamide, and high-dose methotrexate). At 63 months, with a median follow-up of 56 months, there were four local recurrences in the 27 patients who underwent limb-sparing surgery and no recurrences in the 16 patients who underwent amputation P2 = .12. Overall survival (OS) rates were not statistically significantly different (actuarial 5-year survival rate, 83% vs. 88%, P2 = .99). 3[Level of evidence: 1iiA]
Local control of high-grade soft tissue sarcomas of the trunk and the head and neck can be achieved with surgery in combination with radiation therapy. 4 It may be possible to use surgery without PORT in selected cases. For example, a case series was reported from a specialized sarcoma treatment referral center in which 74 selected patients with primary extremity and trunk tumors 5 cm or less in size were found to have no histologic involvement of the surgical margins. 1 They were observed without radiation therapy, and the estimated local recurrence rate after 10 years was 11%. 1[Level of evidence: 3iiiDiv] The role of chemotherapy is not as well defined as is the role of radiation therapy. Because of the evolving nature of the treatment options for this disease, patients should be offered the option of clinical trials when available.
Effective treatment of retroperitoneal sarcomas requires removal of all gross disease while sparing adjacent viscera not invaded by tumor. The prognosis for patients with high-grade retroperitoneal sarcomas is less favorable than for patients with tumors at other sites, partly because of the difficulty in completely resecting these tumors and the dose-limiting toxicity of high-dose radiation therapy on visceral organs. 5 6 7 8
In the setting of distant metastasis, surgery may be associated with long-term disease-free survival in patients with pulmonary metastasis and optimal underlying disease biology (i.e., patients with a limited number of metastases and slow nodule growth) who have undergone or are undergoing complete resection of the primary tumor. 9 10 11 It is not clear to what degree the favorable outcomes are attributable to the efficacy of surgery or the careful selection of patients based on factors that are associated with less-virulent disease.
Radiation plays an important role in limb-sparing therapy. Pre- and postoperative external-beam radiation therapies (EBRT), as well as brachytherapy, have been shown to decrease the risk of local recurrence. They have not been shown to increase OS but are used to avoid amputation for all but the most locally advanced tumors or for limbs seriously compromised by vascular disease, where acceptable functional preservation is not possible. In the case of EBRT, irradiation of the entire limb circumference is avoided to preserve vascular and nerve structures that are critical to function and preservation of the limb.
PORT has been tested in a single-institution randomized trial of 141 patients with extremity sarcomas who were treated with limb-sparing surgery. Patients with high-grade tumors (n = 91) also received adjuvant chemotherapy (i.e., five 28-day cycles of doxorubicin and cyclophosphamide). All patients were randomly assigned to receive radiation (45 Gy to a wide field, plus a tumor-bed boost of 18 Gy over 67 weeks), concurrent with chemotherapy in the case of high-grade tumors versus no radiation. 12 At up to 12 years of follow-up, there was one local recurrence in the 70 patients randomly assigned to receive radiation versus 17 recurrences in the 71 control patients (P = .0001), with similar reduction in risk of local recurrence for both high- and low-grade tumors. However, there was no difference in OS between the radiation and control groups. 12[Level of evidence: 1iiDiii] Global quality of life was similar in the two groups, but the radiation therapy group had substantially worse functional deficits resulting from reduced strength and joint motion as well as increased edema.
To limit acute toxicity with preRX, smaller fields and lower doses are generally given than is the case with PORT. PreRX has been directly compared with PORT for extremity soft tissue sarcomas in a multicenter randomized trial. 13 14 15 Designed to include 266 patients, the trial was stopped early after 190 patients had been accrued because of an increase in wound complications in the preRX group. The scheduled radiation in the preRX group was a wide field of 50 Gy in 2 Gy fractions (first phase of the trial) with an additional 16 Gy to 20 Gy to the tumor bed and a 2-cm margin (second phase of the trial) only if tumor cells were found at the surgical margins.
Patients in the PORT group were scheduled to receive radiation during both phases of the trial. The wound-complication rates were 35% versus 17% in the preRX and PORT groups, respectively (P = .01). In addition, limb function at 6 weeks after surgery was worse in the preRX group (P = .01). 13 At 5 years, the two groups had similar local control rates (93% vs. 92%) and OS (73% vs. 67%, P = .48). 14 Of the 129 patients evaluated for limb function at 21 to 27 months after surgery (n = 73 for preRX and n = 56 for PORT), limb function was similar in both groups, but there was a statistical trend for less fibrosis in the preRX group (P = .07). 15
Brachytherapy has also been investigated as an adjuvant therapy for soft tissue sarcomas. Although it has possible advantages of convenience and less radiation to normal surrounding tissue relative to EBRT, the two treatment strategies have not been directly compared in terms of efficacy or morbidity. However, adjuvant brachytherapy has been compared to surgery without radiation.
In a single-institution trial, 164 patients with sarcomas of the extremity or superficial trunk were randomly assigned during surgery, if all gross tumor could be excised, to receive an iridium-192 implant (delivering 42 Gy45 Gy over 46 days; 78 patients) or to a control arm of no radiation (86 patients). 16 17 Some of the patients with high-grade tumors received adjuvant doxorubicin-based chemotherapy if they were thought to be at a high risk for metastasis (34 patients in each study arm). With a median follow-up of 76 months, the 5-year actuarial local recurrence rates were 18% and 31% in the brachytherapy and control arms, respectively (P = .04). This difference was limited to patients with high-grade tumors. There was no discernible difference in sarcoma-specific survival rates between the brachytherapy and control arms (84% and 81%, respectively; P = .65), and there was no difference in the high tumor-grade group. 16[Level of evidence: 1iiDiii] The rates of clinically important wound complications (e.g., need for operative revision or repeated seroma drainage, wound separation, large hematomas, or purulent infection) were 24% and 14% in the radiation and control arms, respectively (P = .13); wound reoperation rates were 10% and 0%, respectively (P = .006). 17
Intensity-modulated radiation therapy (IMRT) has been used to deliver preRX or PORT to patients with extremity soft tissue sarcomas in an effort to spare the femur, joints, and selected other normal tissues from the full prescription dose and to maintain local control while potentially reducing radiation therapy-related morbidity. Initial single-institution reports suggest that high rates of local control with some reduction in morbidity are possible with this technique. 18 19
In some tumors of the extremities or trunk, surgery alone can be performed without the use of radiation. Evidence for this approach is limited to single-institution, relatively small, case series 1 20 21 or analysis of outcomes in the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) tumor registry. 22 However, these comparisons suffer from low statistical power and differential evaluability rates that could have introduced bias. 1 Patient selection factors may vary among surgeons. In general, this approach is considered in patients with low-grade tumors of the extremity or superficial trunk that are less than or equal to 5 cm in diameter (T1) and have microscopically negative surgical margins; long-term local tumor control is about 90% in such patients. 23
On occasion, surgical excision cannot be performed in the initial management of soft tissue sarcomas because the morbidity would be unacceptable or nearby critical organs make complete resection impossible. In such circumstances, radiation has been used as the primary therapy. 24 However, this must be considered a treatment of last resort. Experience is limited to retrospective case series from single centers. 24[Level of evidence 3iiiDiv]
The role of adjuvant chemotherapy is not completely clear. The investigation of its use falls into two categories or generationspre- and postifosfamide regimens. In discussions with a patient, any potential benefits should be considered in the context of the short- and long-term toxicities of the chemotherapy.
Several prospective, randomized trials were unable to determine conclusively whether doxorubicin-based adjuvant chemotherapy benefits adults with resectable soft tissue sarcomas. The majority of these studies accrued small numbers of patients and did not demonstrate a metastasis-free survival or an OS benefit for adjuvant chemotherapy. 4 A small study of adjuvant chemotherapy showed a positive effect on both disease-free survival (DFS) and OS in patients treated with postoperative chemotherapy. 25 There was wide interstudy variability among the reported trials, including differences in therapeutic regimens, drug doses, sample size, tumor site, and histologic grade.
A quantitative meta-analysis of updated data from 1,568 individual patients in 14 trials of doxorubicin-based adjuvant therapy showed an absolute benefit from adjuvant therapy of 6% for local relapse-free interval (95% confidence interval [CI], 1%10%), 10% for distant relapse-free interval (95% CI, 5%15%), and 10% for recurrence-free survival (95% CI, 5%15%). A statistically significant OS benefit at 10 years was not detected: absolute difference 4% (95% CI, -1%+9%). 26 27[Level of evidence: 1iiDii] However, only a small proportion of patients in this meta-analysis were treated with ifosfamide, an agent with demonstrated activity against soft tissue sarcoma. In addition, a subset analysis suggested that patients with sarcomas of the extremities may have benefited from adjuvant chemotherapy (hazard ratio [HR] for death, 0.8, P = .029), but there was no clear evidence that patients with extremity sarcomas had outcomes that were statistically significantly different from the outcomes of patients with tumors at other sites (P = .58). 27
Subsequent chemotherapy trials were performed using anthracycline and ifosfamide combinations in patients who primarily had extremity or truncal soft tissue sarcomas. The data are conflicting, and the issue is still not settled. In a small feasibility study, 59 patients with high-risk soft tissue sarcomas, 58 of whom had an extremity or trunk as the primary site, underwent primary resection plus PORT and were randomly assigned to observation versus a dose-dense regimen of six 14-day courses of ifosfamide, dacarbazine (DTIC), and doxorubicin (IFADIC regimen) with granulocyte colony-stimulating factor (G-CSF) bone marrow support and mesna uroprotection. 28 There were no statistically significant differences in OS or relapse-free survival (RFS), but the study was severely underpowered.
In a second trial performed by the Italian National Council for Research, high-risk patients were treated with local therapy (i.e., wide resection plus preRX or PORT, or amputation as clinically necessary) and were then randomly assigned to observation versus five 21-day cycles of 4-epidoxorubicin (epirubicin) plus ifosfamide (with mesna and G-CSF). 25 29 Based on power calculations, the planned study size was 190 patients, but the trial was stopped after 104 patients had been entered because an interim analysis revealed a statistically significant (P = .001) difference in DFS favoring the chemotherapy arm. By the time of the initial peer-reviewed report of the study, the DFS still favored the chemotherapy group (median DFS of 48 months vs. 16 months), but the P value had risen to .04. 25
Although there was no difference in metastasis-free survival at the time of the report, there was an improvement in median OS (75 months vs. 46 months, P = .03). However, at the follow-up report (at a median of 89.6 months in a range of 56119 months), OS differences were no longer statistically significant (58.5% vs. 43.1% [P = .07]). The DFS difference had also lost statistical significance (47.2% vs. 16.0% [P = .09). 29 In summary, the trial was underpowered because it was stopped early, and the early promising results that led to stopping the trial diminished as the trial matured.
In a third, underpowered, single-center trial, 88 patients with high-risk soft tissue sarcomas (64 of whom had extremity or truncal primary tumors) underwent surgery (with or without radiation) and were then randomly assigned to receive four 21-day cycles of chemotherapy (epirubicin [n = 26] or epirubicin plus ifosfamide [n = 19]) versus no adjuvant chemotherapy (n = 43). 30 The trial was closed prematurely because of a slow accrual rate. After a median follow-up of 94 months, the 5-year DFS in the chemotherapy and control arms was 69% versus 44%, respectively (P = .01); the 5-year OS rates were 72% versus 47% (P = .06). All of the benefit associated with chemotherapy appeared restricted to the 19 patients who received epirubicin plus ifosfamide.
In yet another underpowered trial, 137 patients with high-risk soft tissue sarcomas (93% with extremity or truncal primary tumors) who met the eligibility criteria were randomly assigned to undergo surgical resection (with or without radiation) or to receive three preoperative 21-day cycles of doxorubicin plus ifosfamide. 31 This multicenter European Organization for Research and Treatment of Cancer trial (EORTC-62874) was closed because of slow accrual and results that were not promising enough to continue. With a median follow-up of 7.3 years, the 5-year DFS in the surgery alone and chemotherapy plus surgery arms was 52% and 56%, respectively (P = .35); and OS was 64% and 65%, respectively (P = .22).
These last four trials have been combined with the 14 first-generation trials in a trial-level meta-analysis. 32 Of the 18 randomized trials of patients with resectable soft tissue sarcomas, five trials used a combination of doxorubicin (5090 mg/m2 per cycle) plus ifosfamide (15005000 mg/m2 per cycle). The remaining 13 trials used doxorubicin (5070 mg/m2 per cycle) alone or with other drugs. The absolute risk reduction in local recurrence rates associated with any chemotherapy added to local therapy was 4 percentage points (95% CI, 0%7%), and it was 5 percentage points (95% CI, 1%12%) when ifosfamide was combined with doxorubicin. The absolute reduction in overall mortality was 6 percentage points with any chemotherapy (95% CI, 2%11%; [i.e., a reduction from 46%40%]), 11 percentage points for doxorubicin plus ifosfamide (95% CI, 3%19%; [i.e., a reduction from 41%30%]), and 5 percentage points for doxorubicin without ifosfamide. 32[Level of evidence: 1iiA]
An additional multicenter randomized trial (EORTC-62931 [NCT00002641]), the largest trial reported to date using adjuvant doxorubicin (75 mg/m2) plus ifosfamide (5000 mg/m2), was subsequently published in abstract form and was not included in the above meta-analysis. 33 The results differed from those reported in the meta-analysis. 32 After local therapy, 351 patients were randomly assigned to five 21-day cycles of adjuvant therapy versus observation. The trial was stopped for futility because the 5-year RFS was 52% in both arms. OS was 64% in the chemotherapy arm versus 69% in the observation arm. In a subsequent abstract, the EORTC investigators reported a combined analysis of this trial and their previous trial (EORTC-62771) 34 of adjuvant cyclophosphamide plus doxorubicin plus DTIC (CYVADIC), representing the two largest trials of adjuvant therapy for adult soft tissue sarcoma in the literature. 35 The combined analysis showed no improvement in either RFS or OS associated with adjuvant chemotherapy. 35[Level of evidence: 1iiA]
In summary, the impact of adjuvant chemotherapy on survival is not clear but is likely to be small in absolute magnitude. Therefore, in discussions with a patient, any potential benefits should be considered in the context of the short- and long-term toxicities of the chemotherapy.
The use of regional hyperthermia to enhance the local effects of systemic chemotherapy in the neoadjuvant and adjuvant setting is under investigation. In a multicenter phase III trial, 341 patients with high-risk (tumor 5 cm, grade 23, and deep to fascia) soft tissue sarcomas (149 extremity tumors and 192 nonextremity tumors) were randomly allocated to receive four 21-day cycles of chemotherapy (etoposide 125 mg/m2 on days 1 and 4; ifosfamide 1500 mg/m2 on days 14; doxorubicin 50 mg/m2 on day 1) with or without regional hyperthermia both before and after local therapy. 36 Approximately 11% of the patients were being treated for recurrent tumors. The regional hyperthermia was designed to produce tumor temperatures of 42C for 60 minutes and was given on days 1 and 4 of each chemotherapy cycle. After the first four cycles of chemotherapy, definitive surgical excision of the tumor was performed, if possible, followed by radiation therapy, if indicated (i.e., a 52.7 Gy median dose delivered), and then the last four cycles of chemotherapy plus or minus hyperthermia. Three of the nine treatment centers with particular expertise in hyperthermia treated 91% of the patients in the trial.
The median duration of follow-up was 34 months. Local progression occurred in 56 patients in the hyperthermia group and 76 patients in the control group. The relative HR for local progression or death was 0.58 (95% CI, 0.410.84), with an absolute difference at 2 years of 15% (76% vs. 61%; 95% CI of the difference 626). The decreased risk of local progression or death was seen in both extremity and nonextremity tumors. However, hyperthermia had no effect on distant failure rates nor was there a statistically significant effect on OS (HR, .88, 95% CI, 0.641.21; P = .43). 36[Level of evidence: 1iiDiii] There was a higher rate of grade 3 to 4 leucopenia in the hyperthermia group: 77.6% versus 63.5% (P = .005). Since a large proportion of the patients were treated at centers with special expertise, there is no certainty that the finding can be generalized to apply to other settings.
Isolated limb perfusion is under investigation as a means to deliver high doses of chemotherapy and permit limb salvage in unresectable primary or recurrent extremity soft tissue sarcomas that would otherwise require amputation, in the opinion of the surgeon. 37 38 Common drugs used in the procedure are TNF-alpha, melphalan, and interferon-gamma. Experience is limited to case series with response rates and reported avoidance of amputation as the outcome. 37 38[Level of evidence 3iiiDiv] The technique requires specialized expertise to avoid severe local and systemic toxicity including systemic effects of TNF-alpha. The technique has not been directly compared to standard approaches with combined systemic and local therapy.
Doxorubicin is a mainstay of systemic therapy in the management of locally advanced and metastatic soft tissue sarcoma. Pegylated liposomal encapsulated doxorubicin is a formulation of doxorubicin designed to prolong the half-life of circulating doxorubicin and slow the release of active drug. 39 The changed pharmacokinetics result in less myelosuppression and possibly less cardiotoxic effects, but there is a substantial incidence of hypersensitivity-like reactions and hand-foot syndrome. Its clinical activity relative to unencapsulated doxorubicin is not clear. 39[Level of evidence 3iiiDiv] Other drugs that are thought to have clinical activity as single agents are ifosfamide, epirubicin, gemcitabine, and paclitaxel. 40 41 42 43[Level of Evidence 3iiiDiv] Their clinical activity relative to single-agent doxorubicin is not clear, and they are not known to have superior activity.
There is controversy about the clinical benefit of adding other drugs to doxorubicin as a single agent. A systematic evidence review and meta-analysis conducted by the Cochrane Collaboration summarized the eight randomized trials reported from 1976 to 1995. 44 No additional randomized trials had been reported or were known to be in progress between 1995 and the 2002 literature search. Single-agent doxorubicin had been compared with a variety of doxorubicin-containing combinations that included vincristine, vindesine, cyclophosphamide, streptozotocin, mitomycin-C, cisplatin, and/or ifosfamide. Combination regimens consistently caused more nausea and hematologic toxicity. However, the better response rates associated with combination therapy were marginal and depended on the statistical model used (fixed effects model ORresp = 1.29; 95% CI, 1.031.60, P = .03; random effects model ORresp = 1.26; 95% CI, 0.961.67, P = .10) There was no statistically significant difference in the 1- (ORmortality = 0.87; 95% CI, 0.731.05, P = .14) or 2-year mortality rates (ORmortality = 0.84; 95% CI, 0.671.06, P = .13).
These results were very similar even when the analyses were restricted to the four trials that used DTIC and/or ifosfamide as part of the combination regimen with doxorubicin agents that were postulated to have greater activity than the others tested. A subsequent meta-analysis of all three published randomized trials of chemotherapy regimens that contained ifosfamide versus those that did not came to similar conclusions: tumor response rates were better when the regimen included ifosfamide (RRresp = 1.52; 95% CI, 1.112.08), but mortality at 1 year was not (RRmortality = 0.98; 95% CI, 0.851.13). 45[Level of evidence: 1iiDiv]. Therefore, response rate was a poor surrogate for OS. Quality-of-life outcomes were not reported in any of the above-mentioned randomized trials, but toxicity was worse when agents were added to doxorubicin.