Overview of the initial treatment of metastatic soft tissue sarcoma

INTRODUCTION — Soft tissue sarcomas (STS) are a heterogeneous group of rare tumors that arise from mesenchymal cells, such as muscle, adipose, fibrous, cartilage, nerve, and vascular tissue. STS arise most frequently in the limbs (particularly the lower extremity), followed by the abdominal cavity/retroperitoneum, the trunk/thoracic region, and the head and neck. There are many histologic subtypes of STS with distinct clinical profiles, molecular alterations, treatment responses, and prognoses. (See "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma".)

An overview of the initial treatment of common metastatic STS histologies will be reviewed here. The management of metastatic gastrointestinal stromal tumors and other unique types of STS (as noted below), bone sarcomas (Ewing sarcoma, osteosarcoma, chondrosarcoma), tenosynovial giant cell tumors, and select uncommon sarcoma subtypes are discussed separately.

●Gastrointestinal stromal tumors (see "Tyrosine kinase inhibitor therapy for advanced gastrointestinal stromal tumors")

●Breast sarcoma (see "Breast sarcoma: Treatment")

●Uterine leiomyosarcoma (see "Treatment and prognosis of uterine leiomyosarcoma")

●Head and neck sarcomas (see "Head and neck sarcomas")

●Rhabdomyosarcoma (see "Rhabdomyosarcoma in childhood, adolescence, and adulthood: Treatment")

●Desmoid tumors (see "Desmoid tumors: Systemic therapy")

●Solitary fibrous tumors (see "Solitary fibrous tumor")

●Dermatofibrosarcoma protuberans (see "Dermatofibrosarcoma protuberans: Treatment")

●Kaposi sarcoma (see "Classic Kaposi sarcoma: Clinical features, staging, diagnosis, and treatment" and "AIDS-related Kaposi sarcoma: Staging and treatment")

●Ewing sarcoma (see "Treatment of Ewing sarcoma")

●Osteosarcoma (see "Chemotherapy and radiation therapy in the management of osteosarcoma")

●Chondrosarcoma (see "Chondrosarcoma")

●Tenosynovial giant cell tumors (see "Treatment for tenosynovial giant cell tumor and other benign neoplasms affecting soft tissue and bone", section on 'Tenosynovial giant cell tumor')

●Uncommon sarcoma subtypes (see "Uncommon sarcoma subtypes")

CLINICAL PRESENTATION — In patients with STS, the specific organs that are affected by recurrent and metastatic disease vary by histology. For most soft tissue sarcomas (with the exception of gastrointestinal stromal tumors), the lungs are the primary site of metastatic disease. However, certain STS subtypes can disseminate to the bones, liver, paraspinal soft tissues, retroperitoneum, and regional lymph nodes.

Further details on the clinical presentation of metastatic disease based on STS histology are discussed separately.

●Patterns of disseminated disease (see "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma", section on 'Pattern of spread')

●Patterns of recurrent disease (see "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma", section on 'Pattern of recurrence')

PRETREATMENT EVALUATION — Patients with a confirmed diagnosis of metastatic STS should undergo the following evaluation prior to initiating therapy. (See "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma", section on 'Diagnostic evaluation'.)

When to initiate therapy — The decision to initiate therapy in patients with advanced unresectable metastatic STS is based upon the pace of disease progression and patient symptoms.

●Substantial disease burden – We initiate systemic therapy in patients with substantial disease burden who are symptomatic and/or have rapidly progressive disease. Metastatic STS is, with rare exception, a fatal disease. The goals of systemic therapy are to palliate symptoms, control tumor burden, improve quality of life, and prolong survival. Selection of initial therapy is primarily based upon the specific tumor histology and their sensitivity to certain therapies, patient characteristics, and treatment preferences. Since STS is a rare disease, clinical trial enrollment is encouraged where available. (See 'Prognosis' below.)

●Limited disease burden – Observation may be an appropriate alternative to systemic therapy for patients with select indolent histologies who are asymptomatic with limited (ie, very low) disease burden. Examples of such histologies include well-differentiated abdominal liposarcoma [1] and extraskeletal myxoid chondrosarcoma, which are discussed separately. (See "Uncommon sarcoma subtypes", section on 'Extraskeletal myxoid chondrosarcoma'.)

Select patients with metastatic disease localized to one organ may also be eligible for potentially curative options, such as resection of limited pulmonary metastases. Surgical treatments and other localized therapies (radiation therapy [RT] and ablative therapies) for metastatic STS are discussed separately. (See "Surgical treatment and other localized therapy for metastatic soft tissue sarcoma".)

Imaging studies — For patients who are being treated with initial systemic therapy, we obtain imaging studies to assess burden of metastatic disease and as a baseline to evaluate for subsequent treatment response. (See 'Assessment of treatment response' below.)

The choice of initial imaging study is based on STS histology and is discussed separately. (See "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma", section on 'Evaluation for metastatic disease'.)

Approach to molecular testing

●Sarcoma without a specific histologic diagnosis — For patients with metastatic STS who do not have a specific histologic diagnosis (such as those sarcomas with immunohistochemistry positive for vimentin only, or sarcoma not otherwise specified [NOS]), we obtain diagnostic assays that can identify both mutations and gene fusions/translocations. The choice of test should be guided by a pathologist with expertise in sarcomas. Some sarcoma histologies are associated with specific translocations (table 1) or other molecular alterations, which are mainly used to confirm a histopathologic diagnosis. Occasionally, such testing can also identify a specific sarcoma histology that was not suspected, such as synovial sarcoma or gastrointestinal stromal tumor (GIST). (See "Pathogenetic factors in soft tissue and bone sarcomas".)

In rare cases (<1 percent), a translocation can be found in targetable receptor tyrosine kinase genes such as NTRK, RET, or others. However, in patients with a confirmed STS histology that have a strong concordance with a specific molecular alteration, such as congenital infantile fibrosarcoma and NTRK, molecular profiling is not always necessary for the diagnosis in the appropriate clinical settings. The diagnosis and management of patients with suspected tropomyosin receptor kinase (TRK) fusions in solid tumors is discussed separately. (See "TRK fusion-positive cancers and TRK inhibitor therapy" and 'NTRK gene fusion-positive tumors' below.)

●Role of next-generation sequencing for metastatic STS? — We do not routinely obtain next-generation sequencing (NGS) in all patients with metastatic STS prior to initiating therapy. The yield of actionable mutations identified by NGS is very low in most sarcomas, and potential cost of the test to the patient needs to be considered. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications", section on 'Cost and turnaround time'.)

NGS is increasingly being used for cancer screening and management, but its role in the treatment of metastatic STS is not established [2-4]. The impact of NGS-based targeted therapy on clinical outcomes may also vary widely based on STS subtype [5].For most STS subtypes, the impact of targeted therapy guided by NGS screening is low, based on the clinical experience of the UpToDate authors and clinical trials [6]. Many STS tumors are routinely treated with initial systemic agents that do not specifically target their characteristic translocations or other DNA alterations; examples of such histologies include synovial sarcoma, myxoid/round cell liposarcoma (table 1), desmoid tumors, and well-differentiated or dedifferentiated liposarcoma. (See "Desmoid tumors: Systemic therapy".)

While the routine use of NGS is not established for leiomyosarcoma, in some cases the information, such as the presence of a somatic alteration in breast cancer susceptibility gene 2 (BRCA2) may have implications for treatment and/or clinical trials. This is discussed separately. (See "Treatment and prognosis of uterine leiomyosarcoma", section on 'Role of molecular diagnostic testing and somatic tumor profiling'.)

Several clinical trials are investigating the use of multigene panel sequencing to identify targeted therapies in adult and pediatric patients with refractory cancers, including soft tissue sarcomas. These trials include the National Cancer Institute Molecular Analysis for Therapy Choice (MATCH) [3,4], and the American Society for Clinical Oncology Targeted Agent and Profiling Utilization Registry (TAPUR) [7]. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications", section on 'Cancer screening and management'.)

INITIAL THERAPY — Patients with metastatic STS should be referred to sarcoma centers of excellence for treatment. Enrollment in clinical trials is encouraged, where available. For patients who are not candidates for or decline clinical trials, we offer the following treatment approach.

Histologies where anthracyclines are preferred — For patients with any of the metastatic STS subtypes (ie, histologies) listed below, we suggest anthracycline-based chemotherapy rather than other systemic regimens, based largely on observational data suggesting response to these agents. (See 'Doxorubicin' below.)

●Synovial sarcoma, which is also sensitive to regimens that contain ifosfamide [8-11]

●Liposarcoma (myxoid/round cell and pleomorphic subtypes [11,12])

●Undifferentiated/unclassified soft tissue sarcoma (eg, sarcoma not otherwise specified [NOS]), which includes the following subtypes [11]:

•Undifferentiated pleomorphic sarcoma (UPS)

•Myxofibrosarcoma

Some histologies are treated with anthracycline-based therapy but have a lower response rate, such as:

●Dedifferentiated liposarcoma

●Malignant peripheral nerve sheath tumor (MPNST) [13]

●Low-grade fibromyxoid sarcoma [14,15]

●Extraskeletal osteosarcoma (see 'Extraskeletal osteosarcoma' below)

Combination anthracycline-based therapy — For patients with any of the metastatic STS histologies where anthracyclines are preferred (see 'Histologies where anthracyclines are preferred' above) and symptomatic or rapidly progressive disease, we suggest initial therapy with a combination anthracycline-based regimen rather than a single-agent anthracycline. Patients who are candidates for this approach include those with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 (table 2) who are eligible for anthracyclines (no clinical heart failure and, generally, a lifetime cumulative doxorubicin dose less than 450 mg/m² for prior treatment of localized STS). (See "Risk and prevention of anthracycline cardiotoxicity".)

In clinical trials of patients with treatment-naïve metastatic STS, combination anthracycline-based chemotherapy improved objective response rates but not overall survival, compared with single-agent doxorubicin.

Selection of combination anthracycline-based therapy may be based upon the tumor histology, patient characteristics (age, comorbidities), toxicity profiles, and clinician or institutional preference. The most frequently used regimens include (see "Treatment protocols for soft tissue and bone sarcoma"):

●Doxorubicin and ifosfamide with mesna (AIM) [16-19] (table 3 and table 4 and table 5)

●Doxorubicin and dacarbazine (AD) [20-24]

●Epirubicin and ifosfamide with mesna [25]

Many initial clinical trials on combination anthracycline-based therapy enrolled patients with different STS subtypes (including leiomyosarcoma, liposarcoma, synovial sarcoma, UPS, malignant peripheral nerve sheath tumors, and undifferentiated sarcoma NOS), but few presented outcomes according to histology. Contemporary studies that enrolled patients and assessed outcomes based on STS histology have suggested sensitivity for certain agents. As examples:

●Synovial sarcoma – For most patients with metastatic synovial sarcoma, we suggest initial therapy with doxorubicin with or without ifosfamide rather than other combination anthracycline-based regimens. Synovial sarcoma is an anthracycline-responsive histology that is also relatively sensitivity to ifosfamide [8-11,26-28]. For patients with synovial sarcoma, observational studies and clinical trials suggest that doxorubicin plus ifosfamide results in the best objective response rates (26 to 40 percent) compared with other anthracycline-based regimens, but is not associated with better overall survival (OS) [10,26-29].

●Malignant peripheral nerve sheath tumors – For most patients with metastatic malignant peripheral nerve sheath tumors (MPNST), we suggest initial therapy with doxorubicin plus ifosfamide rather than other combination anthracycline-based regimens. Doxorubicin plus ifosfamide is modestly effective in MPNST, which typically presents with rapidly progressive disease and usually requires aggressive intervention. Ifosfamide plus etoposide is also an appropriate option that avoids the potential toxicities of anthracyclines [13,30]. Single-agent doxorubicin is an alternative for those who cannot tolerate combination therapy, such as older adults or those with significant comorbidities. (See 'Single-agent anthracyclines' below.)

In an observational study of 115 patients with unresectable or metastatic MPNST, initial therapy with doxorubicin plus ifosfamide was associated with a better progression-free survival (PFS) compared with doxorubicin plus dacarbazine (median PFS 10 versus 4 months, hazard ratio [HR] 0.30, 95% CI 0.09-0.96) [13]. Doxorubicin plus ifosfamide also had the highest treatment response among all the regimens evaluated (67 percent versus 28 percent for doxorubicin plus dacarbazine; 25 percent for high-dose ifosfamide; 24 percent for etoposide and ifosfamide; and 0 percent for doxorubicin plus cisplatin).

●Borderline resectable disease – AIM is used in younger patients with high-grade, borderline resectable lesions or pulmonary metastases in whom a rapid treatment response is needed [21].

In patients with treatment-naïve metastatic STS, randomized trials generally demonstrated higher objective response rates for combination anthracycline-based regimens compared with single-agent doxorubicin. Objective response rates are between 26 to 34 percent for AIM [16,17,27]; 24 to 30 percent for doxorubicin plus dacarbazine [20,22,23,31]; and 59 percent for epirubicin and ifosfamide [25]. In contrast, objective response rates for single-agent doxorubicin range between 10 to 25 percent [16,17,22,27,32-34]). (See 'Doxorubicin' below.)

However, most treatment responses from combination anthracycline-based therapy are not durable (with median PFS ranging between four to eight months) and complete responses were seen in less than 10 percent of patients. Combination therapy also does not improve overall survival and is associated with greater toxicity than single-agent doxorubicin. Similarly, the addition of olaratumab, a platelet-derived growth factor receptor alpha inhibitor, to single-agent doxorubicin did not improve survival and this agent was withdrawn from the market [34-36]

There is no role for high dose doxorubicin and ifosfamide in combination with hematopoietic growth factor support [18,25,32,37-41] or autologous hematopoietic stem cells [42-46], as this approach has not demonstrated an OS benefit compared with standard dosing regimens [47].

Single-agent anthracyclines — A single-agent anthracycline is a reasonable alternative in patients with any of the metastatic STS histologies where anthracyclines are preferred (see 'Histologies where anthracyclines are preferred' above) but cannot tolerate combination anthracycline-based therapy or have indolent disease that does not require a rapid reduction in tumor burden. Patients who may not be able to tolerate such combination therapy include older adults, those with worsened ECOG performance status (table 2) or those with significant comorbidities. Options include doxorubicin, pegylated liposomal doxorubicin (PLD), and epirubicin.

Doxorubicin — For most patients who are receiving a single-agent anthracycline, we offer conventional doxorubicin, which is one of the most active chemotherapy agents in metastatic STS. PLD is another appropriate alternative. (See 'Pegylated liposomal doxorubicin' below.)

Doxorubicin is administered at 75 mg/m² on day 1 of a 21-day cycle, usually up to a maximal cumulative dose of 450 mg/m². In select cases, higher doses of doxorubicin may be delivered with close monitoring of cardiac function and administration of dexrazoxane. Alternatively, some studies suggest that dexrazoxane may be administered at the start of doxorubicin therapy for metastatic STS [48]. (See "Risk and prevention of anthracycline cardiotoxicity".)

For patients who are unable to tolerate the standard initial dose, we reduce the doxorubicin dose by either 10 percent (67.5 mg/m²) or 20 percent (60 mg/mg²). In metastatic STS, optimal activity with doxorubicin occurs at ≥60 mg/m² per cycle. Doxorubicin has a narrow therapeutic index: Doses lower than 60 mg/m² are associated with inferior antitumor activity [20], and doses higher than 75 mg/m² per cycle do not demonstrate increased efficacy and result in increased toxicity [27].

Toxicities from doxorubicin include myelosuppression, mucositis, alopecia, nausea and vomiting, and both acute and chronic cardiotoxicity. (See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity" and "Risk and prevention of anthracycline cardiotoxicity".)

Single-agent doxorubicin was the first chemotherapy used to effectively treated metastatic STS [49], and it remains one of the most active agents in this disease, with extensive data supporting it use. In randomized trials, doxorubicin (at a dosing between 70 to 80 mg/m² per cycle) has objective response rates between 10 to 25 percent (most of which are partial responses); median PFS (PFS) ranges between 4 and 11 months [16,17,20,22,27,32-34,50-52].

Doxorubicin demonstrated similar PFS and OS and was less toxic compared with gemcitabine plus docetaxel in a randomized phase III trial (GeDDiS) conducted in Europe [53]. In this study, 257 previously untreated patients were randomly assigned to treatment every three weeks with either doxorubicin at 75 mg/m² on day 1 or gemcitabine at 675 mg/m² on days 1 and 8 and docetaxel at 75 mg/m² on day 8.

At median follow-up of 22 months, PFS and OS were similar between the two treatment arms (median PFS 23 versus 24 weeks for doxorubicin versus gemcitabine plus docetaxel, HR 1.28, 95% CI 0.99-1.65; median OS 76 versus 67 weeks, HR 1.14, 95% CI 0.83-1.57). Objective response rates were also similar (19 versus 20 percent). Survival outcomes were similar among STS subtypes, including uterine and non-uterine leiomyosarcoma, synovial sarcoma, and pleomorphic sarcoma. Compared with patients receiving doxorubicin, those receiving gemcitabine and docetaxel were less likely to adhere to therapy, have more dose delays and reductions, or stop treatment early due to toxicity. However, one limitation is that the doses of gemcitabine plus docetaxel used in this study were lower than those used in other studies that showed a benefit for this combination [54]. (See 'Gemcitabine plus docetaxel' below.)

Pegylated liposomal doxorubicin — PLD is an option for patients at higher risk of cardiotoxicity with conventional doxorubicin or those who prioritize a reduced risk of cardiotoxicity. PLD is similarly effective to conventional doxorubicin but has a different and more tolerable toxicity profile. PLD can also be safely administered over extended periods of time, which is attributed to the liposomal encapsulation [50]. (See "Risk and prevention of anthracycline cardiotoxicity", section on 'Risk factors'.)

PLD is administered at 50 mg/m² on day 1 of a 28-day cycle (table 6). Alternatively, some UpToDate experts administer PLD at an initial dose of 40 mg/m² on day 1 of a 28-day cycle to reduce the risk and severity of skin toxicity. Objective response rates range between 10 and 50 percent [50,55-58], and are durable with some patients experiencing disease control over [50,55-58]. (See 'Assessment of treatment response' below.)

PLD causes less myelosuppression, febrile neutropenia, alopecia, and cardiotoxicity compared with conventional doxorubicin [50]. However, other toxicities more commonly seen with this agent include acute hypersensitivity reactions, palmar-plantar erythrodysesthesia (hand-foot syndrome), and esophagitis. (See "Cutaneous adverse effects of conventional chemotherapy agents" and "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity" and "Risk and prevention of anthracycline cardiotoxicity".)

Based on data from initial early phase clinical trials [55,56], a randomized phase II trial was conducted by the European Organisation for Research and Treatment of Cancer (EORTC) in 94 patients with advanced STS, of which approximately one-third had leiomyosarcoma [50]. In this study, patients were randomly assigned to either PLD at 50 mg/m² every four weeks or conventional doxorubicin (75 mg/m² every three weeks). PLD and doxorubicin had similarly low objective response rates (10 versus 9 percent; 14 versus 12 percent when gastrointestinal stromal tumor [GIST] tumors were excluded). The low objective response rates were attributed to the high proportion of patients with GIST, which are not responsive to chemotherapy. Stable disease was seen in 16 of 50 patients (32 percent) treated with PLD and 18 of 44 patients (41 percent) with doxorubicin.

Compared with doxorubicin, PLD had lower grade ≥3 rates of neutropenia (6 versus 77 percent), thrombocytopenia (0 versus 2 percent), febrile neutropenia (2 versus 16 percent), and alopecia (6 versus 86 percent), but higher rates of palmar-plantar erythrodysesthesia (20 versus 0 percent) [50].

Epirubicin — Epirubicin is a less-preferred option for single-agent anthracycline therapy in metastatic STS in the United States. In randomized trials, epirubicin has equivalent activity compared with doxorubicin, but data are conflicting as to which agent has more overall toxicity [52,59]. Both agents have a similar risk for cardiotoxicity. (See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity" and "Risk and prevention of anthracycline cardiotoxicity".)

●In a randomized trial, 334 patients with treatment-naïve metastatic STS received either doxorubicin (75 mg/m² on day 1) or epirubicin on one of two different schedules (50 mg/m² daily for three days or 150 mg/m² on day 1, both over a 21 day cycle) [59]. Compared with doxorubicin, both epirubicin schedules had similar objective response rates, time to progression, and OS, but higher rates of myelotoxicity. Grade ≥3 cardiotoxicity rates were similar among the treatment arms (1, 2, and 0 percent, for doxorubicin, epirubicin at 50 mg/m², and epirubicin at 150 mg/m², respectively).

●Similar results were seen in a separate phase II/III trial of 210 patients with locally advanced or metastatic STS [52]. In this study, patients were randomly assigned to either doxorubicin or epirubicin at 75 mg/m² on day 1 of a 21-day cycle. Compared with doxorubicin, epirubicin had similar response rates (18 versus 25 percent), median duration of response (77 versus 45 weeks), time to progression (12 versus 15 weeks), and OS (median 48 versus 41 weeks). However, compared with epirubicin, doxorubicin had higher rates of myelotoxicity, alopecia, nausea, and vomiting.

Ineligible for anthracyclines — For fit patients with a histology where anthracyclines are preferred (see 'Histologies where anthracyclines are preferred' above) who are ineligible for or decline anthracyclines, we use a gemcitabine-based combination such as gemcitabine plus docetaxel (table 7), gemcitabine plus vinorelbine, or gemcitabine plus dacarbazine.

Single-agent gemcitabine is an appropriate alternative for patients with decreased performance status, older adults, or those with comorbidities who cannot tolerate combination chemotherapy. (See 'Single-agent gemcitabine' below.)

Gemcitabine plus docetaxel — Gemcitabine plus docetaxel (table 7) is one available combination regimen for patients with metastatic STS who are ineligible for or decline anthracycline-based chemotherapy. In a randomized phase II trial, the addition of docetaxel to gemcitabine improved PFS, OS, and objective response rates [54]. Lower doses (particularly of docetaxel) may be needed for patients with prior radiation therapy (RT) to fields encompassing large amounts of bone marrow. Gemcitabine plus docetaxel is also particularly effective in leiomyosarcoma. (See 'Leiomyosarcoma' below.)

Based on initial data from a phase II trial [60], gemcitabine plus docetaxel was subsequently evaluated in a randomized, open-label phase II trial of 119 patients with treatment-naïve metastatic STS [54]. Patients were randomly assigned to three-week cycles of either gemcitabine (900 mg/m² on days 1 and 8; 675 mg/m² if prior pelvic RT) plus docetaxel (100 mg/m² on day 8; 75 mg/m² if prior pelvic RT) or gemcitabine alone (1200 mg/m² on days 1 and 8; 900 mg/m² in patients with prior pelvic RT). All patients received gemcitabine via fixed-dose-rate infusion (10 mg/m²/min) and hematopoietic growth factor support.

In this study, the addition of gemcitabine to docetaxel improved tumor response (ie, complete or objective responses or stable disease for at least 24 weeks; 32 versus 27 percent) and objective response rates (16 versus 8 percent). Compared with gemcitabine alone, the combination also improved PFS (six versus three months) and overall survival (18 versus 12 months). The histologies that derived the most benefit included leiomyosarcoma and undifferentiated/unclassified STS (see 'Leiomyosarcoma' below). However, toxicity rates were higher for the combination, and more patients receiving gemcitabine plus docetaxel discontinued treatment due to toxicity.

Gemcitabine plus vinorelbine — Gemcitabine plus vinorelbine is another available combination regimen for patients with metastatic STS who are ineligible for or decline anthracycline-based chemotherapy. In a single-arm, open-label phase II trial, 40 patients with advanced STS were treated with gemcitabine plus vinorelbine. Patients could have previously received ≤1 prior systemic regimen for advanced disease, but approximately 63 percent had not previously received treatment for advanced disease. The clinical benefit rate for gemcitabine plus vinorelbine (complete or partial response, or stable disease lasting more than four months) was 25 percent, including a complete response lasting more than one year in one patient with UPS [61]. Gemcitabine plus vinorelbine was initial therapy for advanced disease in all five patients with an objective (one complete and four partial) response. The most common grade ≥3 toxicities were hematologic and included neutropenia (15 percent), although febrile neutropenia was uncommon (<5 percent).

Gemcitabine plus dacarbazine — Gemcitabine plus dacarbazine is an effective, well tolerated combination regimen for patients with metastatic STS who are ineligible for or decline anthracycline-based chemotherapy. The use of gemcitabine plus dacarbazine as initial therapy is extrapolated from data in patients with previously treated disease. In a randomized phase II trial of 113 patients with previously treated soft tissue sarcoma (prior therapy with anthracyclines and ifosfamide or contraindication for their use), the addition of gemcitabine to dacarbazine improved PFS and OS (median PFS 4 versus 2 months, HR 0.58, 95% CI 0.39-0.86); median OS 17 versus 8 months, HR 0.56, 95% CI 0.36-0.90) [62]. The combination also improved objective response plus stable disease rates versus dacarbazine alone (49 versus 25 percent). Outcomes were similar for all histologies, including leiomyosarcoma, liposarcoma, UPS, synovial sarcoma, and miscellaneous sarcoma. Gemcitabine plus dacarbazine was well-tolerated, with the patients rarely stopped treatment due to toxicity. The most common toxicities including granulocytopenia (39 percent), febrile neutropenia (9 percent), asthenia (7 percent), vomiting, and stomatitis (2 percent each).

Single-agent gemcitabine — Single-agent gemcitabine is an appropriate alternative for patients who cannot tolerate gemcitabine-based combination therapy (eg, due to age, poor performance status, or comorbidities). There are conflicting data for the efficacy of single-agent gemcitabine [63-70]. Some studies suggest that gemcitabine at a fixed-dose-rate infusion (mg/m²/min) can induce partial responses [64] or prolonged periods of stable disease (mostly in leiomyosarcomas) [69]. However, other studies suggest minimal activity for gemcitabine monotherapy, both in previously treated and untreated patients [66-68].

Pazopanib — Pazopanib is an option for older adults (age ≥60 years) with select histologies (eg, leiomyosarcoma, UPS, and angiosarcomas) who either cannot tolerate/wish to avoid anthracyclines or prefer oral therapy. However, data are limited for this approach [71,72] and other treatments with more supporting evidence are available for these histologies, which are discussed separately. (See 'Leiomyosarcoma' below and 'Histologies where anthracyclines are preferred' above and 'Angiosarcoma' below.)

We also do not offer pazopanib to patients with liposarcoma, as several phase II trials suggest no benefit for pazopanib in this population as initial or subsequent therapy [71,73]. For older adults with liposarcoma who are willing and able to tolerate anthracyclines, single-agent doxorubicin or PLD remain our preferred initial treatments. (See 'Single-agent anthracyclines' above.)

Data from one randomized phase II trial of older adults with advanced STS suggest that initial therapy with pazopanib has non-inferior PFS and less myelotoxicity compared with doxorubicin [71]. Given the limited evidence, clinicians who offer pazopanib as initial therapy in this setting should provide a risk-benefit discussion about the unique toxicity profiles of each agent. For example, pazopanib can be administered orally versus doxorubicin which is administered intravenously. Pazopanib also has lower rates of myelosuppression, alopecia, and mucosal inflammation, but higher rates of common toxicities seen with antiangiogenic agents (eg, hypertension, hypothyroidism, and diarrhea). Pazopanib may be offered initially at a lower dose (400 to 600 mg orally daily) and uptitrated as tolerated to reduce such toxicities in older patients. (See "Cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Hypertension' and "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents".)

In an open-label phase II trial (EPAZ), 120 older adults (age ≥60 years) with advanced, unresectable STS (including those with leiomyosarcoma, UPS, liposarcoma, and angiosarcomas) without previous exposure to systemic therapy were randomly assigned to either pazopanib or doxorubicin for up to six cycles [71]. A majority of patients were fit without significant comorbidities; approximately one-third had impairments in activities of daily living, and importantly only 16 percent had two or more comorbidities [74]. Compared with doxorubicin, pazopanib demonstrated non-inferior PFS (median 4.4 versus 5.3 months, HR 1.00, 95% CI 0.65-1.53) and similar overall survival (OS; median 12 versus 14 months, HR 1.08, 95% CI 0.68-1.72). Objective response rates were similar between the two agents (12 versus 15 percent). In subgroup analyses, although data are limited by small numbers of patients, pazopanib was more effective than doxorubicin among those younger than age 71, but less effective among those with liposarcoma and those with ECOG performance status (table 2) of 2.

Overall toxicity rates were similar between the two groups (91 versus 94 percent) and were consistent with the unique toxicity profiles of each agent. For example, compared with doxorubicin, rates of myelosuppression were lower with pazopanib, including grade 4 neutropenia (0 versus 56 percent) and febrile neutropenia (0 versus 10 percent). Pazopanib also demonstrated lower rates of alopecia (3 versus 54 percent), stomatitis (4 versus 19 percent), and mucosal inflammation (12 versus 24 percent), but higher rates of hypertension (37 versus 8 percent), hypothyroidism (14 versus 0 percent), diarrhea (43 versus 14 percent), and general deterioration (14 versus 3 percent).

Histologies where anthracyclines and non-anthracycline options are equally appropriate — There are some STS histologies where either anthracyclines or non-anthracycline agents are equally appropriate for initial therapy.

Leiomyosarcoma — For patients with symptomatic or rapidly progressive metastatic leiomyosarcoma and a good ECOG performance status (table 2), we offer initial therapy with either anthracycline-based chemotherapy or gemcitabine plus docetaxel.

Anthracycline-based regimens — For patients with metastatic leiomyosarcoma who select anthracyclines, we suggest initial therapy with either doxorubicin plus dacarbazine or doxorubicin plus trabectedin rather than other anthracycline-based regimens. Single-agent doxorubicin is an appropriate alternative for patients who are unable to tolerate combination anthracycline-based therapy. (See 'Doxorubicin' above.)

●Doxorubicin plus dacarbazine – Doxorubicin plus dacarbazine is one preferred regimen in leiomyosarcoma because it appears to be more active than other anthracycline-based regimens. In addition, ifosfamide is less active in leiomyosarcoma [26].

In patients with treatment-naïve leiomyosarcoma, observational data suggest improved efficacy and survival with doxorubicin plus dacarbazine. In a retrospective study, 303 patients with advanced leiomyosarcoma were treated with either doxorubicin plus dacarbazine (39 percent); doxorubicin plus ifosfamide (23 percent); or doxorubicin (38 percent) [24]. Doxorubicin plus dacarbazine had the highest objective response rates (31 percent) compared with the other regimens (20 percent for doxorubicin plus ifosfamide and 26 percent for doxorubicin, respectively). Doxorubicin plus dacarbazine was associated with higher PFS and OS compared with doxorubicin alone (median PFS 9 versus 5 months, HR 0.72, 95% CI 0.52-0.99; median OS 37 versus 30 months, HR 0.66, 95% CI 043-0.99). Median OS for doxorubicin plus dacarbazine versus doxorubicin plus ifosfamide was 37 and 22 months respectively, but the difference was not statistically significant (HR 0.65, 95% CI 0.40-1.06).

●Doxorubicin plus trabectedin – For patients with treatment-naïve advanced leiomyosarcoma, the addition of trabectedin to doxorubicin improved PFS but increased toxicity in a phase III trial [75]. Doxorubicin plus trabectedin may be an appropriate option for younger patients with leiomyosarcoma and good performance status with symptomatic or rapidly progressive disease who require an immediate treatment response. However, it is unclear if the clinical benefit seen is from the combination of these two agents or the use of maintenance therapy with trabectedin.

Based on initial phase II trials [76,77], a phase III trial (LMS-04) evaluated the combination of doxorubicin plus trabectedin in 150 patients with treatment naïve unresectable or metastatic leiomyosarcoma, including 83 patients (55 percent) with soft tissue disease and 67 patients (45 percent) with uterine disease [75]. In this study, patients were randomly assigned to either doxorubicin (60 mg/m²) plus trabectedin (1.1 mg/m²) once every three weeks for up to six cycles, followed by maintenance trabectedin, or doxorubicin alone (75 mg/m²) once every three weeks for up six cycles. The median ages of patients treated with the combination versus doxorubicin alone were 59 and 64 years old, respectively.

In the entire study population, at median follow-up of approximately 39 months, the addition of trabectedin to doxorubicin improved PFS (median 12 versus 6 months, HR 0.41, 95% CI 0.29-0.58) and objective response rates (36 versus 13 percent). Toxicity was higher for the combination compared to doxorubicin alone, including neutropenia (80 versus 30 percent), anemia (31 versus 15 percent), thrombocytopenia (47 versus 0 percent), and febrile neutropenia (28 versus 9 percent). Among the 83 patients with soft tissue leiomyosarcoma, the addition of trabectedin to doxorubicin also improved objective response rates (37 versus 12 percent) but did not demonstrate a statistically significant difference in PFS between the two treatment arms (HR 0.82, 95% CI 0.58-1.16).

Further details on the efficacy of doxorubicin plus trabectedin in patients with uterine leiomyosarcoma are discussed separately. (See "Treatment and prognosis of uterine leiomyosarcoma", section on 'In combination with trabectedin'.)

Gemcitabine plus docetaxel — Gemcitabine plus docetaxel is also an appropriate option for patients with treatment-naïve metastatic leiomyosarcoma, particularly those who are ineligible for or decline anthracyclines [35,54,60,78].

In a randomized phase II trial, the addition of docetaxel to gemcitabine improved progression-free and overall survival in 122 patients with treatment-naive metastatic STS [54]. Leiomyosarcoma had the greatest impact on overall survival, as patients with leiomyosarcoma had longer OS than those with other histologies (median OS 58 versus 16 months). In addition, among the 29 patients with leiomyosarcoma who received gemcitabine plus docetaxel, objective responses were seen in five patients (17 percent). Further data for this study are discussed separately. (See 'Gemcitabine plus docetaxel' above.)

Initial therapy with gemcitabine plus docetaxel for metastatic uterine leiomyosarcoma is discussed separately. (See "Treatment and prognosis of uterine leiomyosarcoma", section on 'Docetaxel and gemcitabine'.)

Extraskeletal osteosarcoma — Extraskeletal osteosarcoma, a malignant neoplasm that produces osteoid, bone, and chondroid material without direct attachment to bone or periosteum, can arise within the soft tissue. For patients with extraskeletal osteosarcoma, there is no consensus on whether to treat this diagnosis as a soft tissue sarcoma or as osteosarcoma of bone [79]. Treatment options include doxorubicin with or without cisplatin [80] and doxorubicin with or without ifosfamide.

Epithelioid sarcoma — For patients with ES and indolent disease, tazemetostat is a reasonable alternative to an anthracycline-based regimen as initial therapy, and it has regulatory approval in this setting. (See "Uncommon sarcoma subtypes", section on 'Tazemetostat'.)

For those with more aggressive disease, we prefer initial treatment with chemotherapy, such as doxorubicin plus ifosfamide. Further details on management of epithelioid sarcoma are discussed separately. (See "Uncommon sarcoma subtypes", section on 'Treatment of advanced/metastatic disease'.)

Histologies where non-anthracyclines are preferred — For some metastatic STS histologies, non-anthracycline agents are the preferred initial therapy.

Angiosarcoma — For most patients with metastatic angiosarcoma, options for initial therapy include taxanes and anthracycline-based regimens. The choice of initial therapy is influenced by the location of disease as well as clinician and patient preference. Our approach is as follows:

●Cutaneous/head and neck angiosarcomas – For most patients with cutaneous or head and neck angiosarcoma, we suggest single-agent paclitaxel (table 8) rather than anthracyclines or other systemic agents, as this approach is effective, well tolerated, and avoids the potential toxicities of anthracyclines.

●Visceral angiosarcomas – For patients with angiosarcomas that involve the viscera (eg, cardiac angiosarcoma), both anthracycline-based therapy and taxanes are equally appropriate options for initial therapy. There are limited data to guide treatment selection for visceral angiosarcomas since these regimens have not been directly compared in randomized trials. Therefore, clinical practice is variable among UpToDate contributors.

Some contributors prefer initial therapy with anthracycline-based regimens, as angiosarcomas are anthracycline-responsive tumors. Data from a phase II trial also suggested that anthracyclines are more effective in metastatic soft tissue sarcoma, including angiosarcomas, when used prior to taxanes [81]. For those with significant visceral involvement (such as cardiac sarcomas) and rapidly progressive disease, combination regimens such as AIM (table 3 and table 4 and table 5) can be used to achieve a rapid treatment response [82]. Single-agent doxorubicin and PLD are options for patients with more indolent disease or those who are unable to tolerate combination anthracycline therapy [57,83,84].

Alternatively, other contributors prefer initial therapy with paclitaxel, which is also an active, well-tolerated agent in angiosarcoma. For those with significant visceral involvement and rapidly progressive disease, they favor combining gemcitabine with a taxane, such as gemcitabine plus docetaxel (table 7) or gemcitabine plus paclitaxel, to achieve a more rapid treatment response. This approach also avoids the associated toxicities of anthracyclines, which are reserved for use upon disease progression. (See "Second and later lines of therapy for metastatic soft tissue sarcoma", section on 'Second-line therapy'.)

Angiosarcomas are responsive to initial therapy with anthracycline-based regimens based on observational studies and prospective clinical trials [57,83-86]. As an example, a pooled analysis of 11 prospective randomized and non-randomized clinical trials evaluated the efficacy of various initial anthracycline-based chemotherapy regimens in 2665 patients with advanced STS of various histologies, including 108 patients with advanced or metastatic angiosarcoma [84]. At median follow-up of 4.2 years, among those with angiosarcoma, the objective response rate for initial anthracycline-based therapy was 25 percent. Median PFS and OS were 5 and 10 months, respectively, which was similar to the survival outcomes seen with other STS histologies. When the various anthracycline-based regimens used to treat angiosarcomas were compared, doxorubicin plus ifosfamide was associated with improved PFS and OS compared with single-agent anthracyclines.

Angiosarcomas are also uniquely sensitive to taxanes, although they have not been directly compared to anthracycline-based therapy in randomized trials. Most observational studies and prospective clinical trials also suggest that taxanes are especially effective in cutaneous and head and neck angiosarcomas [57,83,87,88]. Paclitaxel and docetaxel can achieve high clinical response rates (approximately 50 to 70 percent or greater) that are also somewhat durable (median PFS of 4 to 8 months), with some responses lasting more than three years [88-91].

Weekly paclitaxel was evaluated in a single-arm phase II trial (ANGIOTAX) of 30 patients with unresectable angiosarcoma [88]. Most (22 patients) had primary tumors involved the soft tissue, skin, and scalp; the remainder (8 patients) had visceral primaries (involving liver, kidney, small bowel, bone and penis). At median follow-up of 8 months, median PFS and OS were 4 and 8 months, respectively. Among chemotherapy-naïve patients, the PFS was 71 percent. Among 22 evaluable patients, non-progression (complete or partial responses, or stable disease) at four months was seen in 10 patients (45 percent), including objective responses in four patients (19 percent). In addition, three patients with breast angiosarcoma were able to undergo curative intent resection after chemotherapy, with two patients demonstrating a complete pathologic response. The outcomes for taxanes in this study among patients with chemotherapy-refractory disease are discussed separately. (See "Second and later lines of therapy for metastatic soft tissue sarcoma", section on 'Second-line therapy'.)

Single-agent gemcitabine [70] and pazopanib have been evaluated as initial therapy for angiosarcoma, but they are less preferred as high-quality data are limited and other more effective therapies are available. (See 'Pazopanib' above.)

Further details on the management of angiosarcomas involving the head and neck and breast are discussed separately. (See "Head and neck sarcomas", section on 'Angiosarcoma' and "Breast sarcoma: Treatment", section on 'Angiosarcoma'.)

PEComa — For patients with advanced unresectable or metastatic malignant neoplasm with perivascular epithelioid cell differentiation (PEComa), we suggest initial therapy with nab-sirolimus over other systemic agents. Nab-sirolimus is the only drug that is approved by the US Food and Drug Administration (FDA) for this disease. For patients living in areas where nab-sirolimus is not approved or unavailable, other mechanistic (mammalian) target of rapamycin (mTOR) inhibitors (sirolimus, everolimus, and temsirolimus) are reasonable alternatives.

The management (including sirolimus and supportive care measures) of other diseases in the PEComa family of tumors, such as lymphangioleiomyomatosis (LAM) and renal angiomyolipoma, are discussed separately. (See "Sporadic lymphangioleiomyomatosis: Treatment and prognosis" and "Renal angiomyolipomas (AMLs): Management".)

Malignant PEComa is a rare epithelioid malignancy that typically arises in the gastrointestinal tract, retroperitoneum, uterus, or somatic soft tissues and intimately related to blood vessel walls. These tumors may exhibit an aggressive clinical course with local recurrences and distant metastases, most commonly in the lung, and respond poorly to systemic chemotherapy (such as anthracycline- or gemcitabine-based regimens) and antiangiogenic agents [92]. Malignant PEComas belong to the PEComa family of tumors which also includes more benign tumors such as LAM and renal angiomyolipoma. This family of mesenchymal neoplasms shares certain pathologic features including myomelanocytic differentiation, involvement of the perivascular epithelioid cell [93-95], and inherited (in those with tuberous sclerosis complex) or sporadic mutations in the TSC1 or TSC2 genes which dysregulate the mTOR signaling pathway [96-101]. (See "Sporadic lymphangioleiomyomatosis: Epidemiology and pathogenesis" and "Renal angiomyolipomas (AMLs): Epidemiology, pathogenesis, clinical manifestations, and diagnosis" and "Tuberous sclerosis complex: Clinical features".)

In an open-label phase II trial (AMPECT), nab-sirolimus, an mTOR inhibitor, was evaluated in 31 patients with locally advanced unresectable or metastatic malignant PEComa [102]. Nab-sirolimus was administered intravenously at 100 mg/m² on days 1 and 8 of a 21-day cycle until disease progression or unacceptable toxicity. At median follow-up of approximately two years, overall responses were seen in 12 patients (39 percent), including one complete responder. Among the 12 patients with treatment response, a majority (58 percent) had durable responses lasting two years or more. Patients with a TSC2 mutation were more likely to respond to therapy, with objective responses seen in eight of nine patients with TSC2-positive tumors (89 percent).

Treatment was well tolerated overall. Grade 1 or 2 interstitial lung disease/non-infectious pneumonitis occurred in 18 percent of patients. Grade ≥3 toxicities included stomatitis (18 percent), anemia, infections (12 percent each), hyperglycemia (9 percent), dehydration (6 percent), diarrhea, vomiting, constipation, fatigue, edema, musculoskeletal pain, hypertension, hemorrhage, hypertriglyceridemia, thrombocytopenia, and insomnia (3 percent each).

Based on these data, the FDA approved nab-sirolimus for adult patients with locally advanced unresectable or metastatic malignant PEComa [103]. Although nab-sirolimus is the first agent to be specifically approved for this disease, it may not be approved or available in all countries. In this situation, other mTOR inhibitors (sirolimus, everolimus, and temsirolimus) are reasonable alternatives to nab-sirolimus, since they also have shown activity in malignant PEComa [96,104-106].

Other histologies — The management of other metastatic histologies is discussed separately.

●Alveolar soft part sarcoma (see "Uncommon sarcoma subtypes", section on 'Alveolar soft part sarcoma')

●Clear cell sarcoma (see "Uncommon sarcoma subtypes", section on 'Clear cell sarcoma')

●Extraskeletal myxoid chondrosarcoma (see "Uncommon sarcoma subtypes", section on 'Extraskeletal myxoid chondrosarcoma')

●Inflammatory myofibroblastic tumor (see "Uncommon sarcoma subtypes", section on 'Inflammatory myofibroblastic tumor')

●Tenosynovial giant cell tumor (see "Treatment for tenosynovial giant cell tumor and other benign neoplasms affecting soft tissue and bone", section on 'Tenosynovial giant cell tumor')

●Dermatofibrosarcoma protuberans (see "Dermatofibrosarcoma protuberans: Treatment", section on 'Treatment of locally advanced, recurrent, and metastatic disease')

●Desmoid tumors (see "Desmoid tumors: Systemic therapy", section on 'Noncytotoxic approaches')

●Solitary fibrous tumor (see "Solitary fibrous tumor")

SPECIAL CONSIDERATIONS

NTRK gene fusion-positive tumors — For patients with tumors expressing the NTRK gene fusion, we offer initial targeted therapy with either or larotrectinib or entrectinib. Between these two agents, we typically prefer larotrectinib because it has more durable responses and a better toxicity profile than entrectinib (which is associated with cardiac toxicity and skeletal fractures). Dosing recommendations and toxicity for larotrectinib and entrectinib are discussed separately. (See "TRK fusion-positive cancers and TRK inhibitor therapy", section on 'Dosing' and "TRK fusion-positive cancers and TRK inhibitor therapy", section on 'Side effects'.)

Fewer than 1 percent of unselected STS have gene fusions involving one of the neurotrophic tyrosine receptor kinase (NTRK) genes [107-110]. Infantile fibrosarcoma, a very rare NTRK-translocation sarcoma nearly exclusively occurring in children under age 2, is one of the index tumors evaluated using these agents. Inflammatory myofibroblastic tumors also express NTRK gene fusions at intermediate frequencies [111]. NTRK gene fusions are otherwise very rare in other sarcoma subtypes. (See "TRK fusion-positive cancers and TRK inhibitor therapy" and "Uncommon sarcoma subtypes", section on 'Inflammatory myofibroblastic tumor'.)

Two highly selective NTRK inhibitors, larotrectinib and entrectinib, are approved by the FDA for use in adults and children with solid tumors with an NTRK gene fusion without a known acquired resistance mutation [112,113]. Patients who are candidates for NTRK inhibitors have metastatic disease, have disease that is unresectable or likely to result in severe postoperative morbidity with no satisfactory alternative treatments or have progressive disease following initial treatment. Confirmation of an NTRK molecular alteration is critical prior to the use of larotrectinib or entrectinib.

●Larotrectinib – The efficacy of larotrectinib was shown in a combined analysis of three phase I and II trials of 159 adult and pediatric patients with various NTRK fusion-positive malignancies; among these patients, 69 had STS [110,114]. Objective responses were seen in 27 of 28 patients with infantile fibrosarcoma (96 percent), including three patients with complete responses; all four patients with gastrointestinal stromal tumors (100 percent); and 29 of 31 patients with other STS histologies (81 percent) [110]. The longest duration of response was seen in one patient with sarcoma (44 months). Although patients were eligible for treatment based on the presence of an NTRK gene fusion, a limitation of this study was that there was no central histology review of this patient population.

●Entrectinib – The efficacy of entrectinib was demonstrated in a pooled analysis of three multicenter, single-arm, open-label trials (ALKA, STARTRK-1, and STARTRK-2) conducted in 54 patients with relapsed advanced or metastatic solid tumors harboring an NTRK gene fusion [113,115]. Among the 13 patients with sarcoma, the overall response rate was 46 percent, and the duration of response ranged between 3 and 15 months.

Of note, sarcomas with a NTRK-translocation are also sensitive to standard chemotherapy agents to some degree as well. For example, in congenital fibrosarcoma, activity with chemotherapy was seen prior to the development of NTRK-directed therapy [111].

ASSESSMENT OF TREATMENT RESPONSE — The appropriate method to determine treatment response in metastatic STS is evolving. Important indicators of clinical benefit include improvements in progression-free survival (PFS) and overall survival (OS) as well as stable or lack of progressive disease (ie, clinical tumor response or non-progression rate).

In contrast, using objective response rates alone (a decrease in size of measurable lesions) to determine treatment response is a less favored approach. Therapeutic agents that are associated with low objective antitumor response may still slow tumor progression and prolong survival. Systemic treatment can successfully kill massive amounts of tumor, but hyalinized acellular tissue may remain. This can lead to a false-negative assessment or underestimate of the true treatment response on imaging studies. This disconnect between tumoral objective response rates and treatment efficacy is particularly evident in studies of immune checkpoint inhibitors, molecularly targeted therapies, and drugs such as trabectedin.

Further details on assessing treatment response with these agents are discussed separately. (See "Principles of cancer immunotherapy", section on 'Immunotherapy response criteria' and "Tyrosine kinase inhibitor therapy for advanced gastrointestinal stromal tumors", section on 'Assessing response to therapy' and "Second and later lines of therapy for metastatic soft tissue sarcoma", section on 'Trabectedin (LMS)'.)

LATER LINES OF THERAPY — The treatment of metastatic soft tissue sarcoma that progress on initial therapy is discussed separately. (See "Second and later lines of therapy for metastatic soft tissue sarcoma".)

PROGNOSIS — Metastatic STS typically has a poor prognosis, with a five-year overall survival (OS) between 15 and 25 percent and median survival between 12 to 24 months, although the prognosis varies based on histology. [116-118]. Survival outcomes have also improved over time with introduction of more effective and well-tolerated systemic agents [117,118]. Other prognostic factors that are associated with improved outcomes include female sex, younger age, good performance status, lack of bone or liver metastases, low-grade histology, and longer disease-free interval between initial diagnosis and metastatic disease [117,119,120].

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

●When to initiate therapy – For patients with metastatic soft tissue sarcoma (STS), we initiate systemic therapy in those with substantial disease burden who are symptomatic and/or have rapidly progressive disease. Metastatic STS is, with rare exception, a fatal disease and treatment is palliative. (See 'When to initiate therapy' above.)

●Histologies in which anthracyclines are preferred

•Histologies included – For patients with any of the metastatic STS subtypes (ie, histologies) listed below, we suggest anthracycline-based therapy rather than other systemic regimens (Grade 2C), based largely on observational data suggesting responses to these agents. (See 'Histologies where anthracyclines are preferred' above and 'Doxorubicin' above.)

-Synovial sarcoma

-Liposarcoma (myxoid/round cell and pleomorphic subtypes)

-Undifferentiated sarcomas not otherwise specified (NOS), which include undifferentiated pleomorphic sarcoma (UPS) and myxofibrosarcoma

Histologies that are treated with anthracycline-based therapy but have a lower response rate include:

-Dedifferentiated liposarcoma

-Malignant peripheral nerve sheath tumor (MPNST)

-Low-grade fibromyxoid sarcoma

-Extraskeletal osteosarcoma (see 'Extraskeletal osteosarcoma' above)

•Selection of therapy – For patients with the above histologies and symptomatic or rapidly progressive disease, we suggest initial therapy with a combination anthracycline-based regimen rather than a single-agent anthracycline (Grade 2C), as this approach improved objective response rates. Appropriate candidates have Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 (table 2) and are eligible for anthracyclines. Patients who cannot tolerate combination therapy are treated with single-agent anthracyclines. (See 'Combination anthracycline-based therapy' above and 'Single-agent anthracyclines' above.)

For most patients with the above histologies with indolent disease that does not require a rapid reduction in tumor burden, we suggest single-agent anthracyclines rather than combination therapy (Grade 2C). However, observation may be an appropriate alternative for select indolent histologies who are asymptomatic with limited (very low) disease burden (ie, well-differentiated liposarcoma). (See 'Single-agent anthracyclines' above and 'When to initiate therapy' above.)

•Ineligible for anthracyclines – For fit patients who are ineligible for or decline anthracycline-based therapy, we use a gemcitabine-based combination regimen such as gemcitabine plus docetaxel (table 7), gemcitabine plus vinorelbine, or gemcitabine plus dacarbazine. Single-agent gemcitabine is an alternative for patients who cannot tolerate gemcitabine-based combination therapy. (See 'Ineligible for anthracyclines' above and 'Single-agent gemcitabine' above.)

●Histologies in which anthracyclines or non-anthracycline options are equally appropriate

•Leiomyosarcoma – For patients with symptomatic or rapidly progressive metastatic leiomyosarcoma and a good ECOG performance status (table 2), we offer initial therapy with either anthracycline-based chemotherapy or gemcitabine plus docetaxel (table 7). (See 'Leiomyosarcoma' above.)

For patients who select anthracycline-based chemotherapy, we suggest initial therapy with either doxorubicin plus dacarbazine or doxorubicin plus trabectedin rather than other anthracycline-based regimens (Grade 2C). Single-agent doxorubicin is an appropriate alternative for those who are unable to tolerate combination anthracycline-based therapy. (See 'Anthracycline-based regimens' above.)

•Epithelioid sarcoma – For patients with metastatic epithelioid sarcoma, treatment options include anthracycline-based regimens or tazemetostat. (See "Uncommon sarcoma subtypes", section on 'Epithelioid sarcoma'.)

●Histologies in which non-anthracycline regimens are preferred – Patients with anthracycline-resistant STS histologies are treated based on histology rather than one single therapeutic approach. (See 'Histologies where non-anthracyclines are preferred' above.)

•Angiosarcoma – For most patients with metastatic cutaneous or head and neck angiosarcoma, we suggest initial therapy with single-agent paclitaxel (table 8) rather than anthracyclines or other systemic agents (Grade 2C), as this approach is effective, well tolerated, and avoids the potential toxicities of anthracyclines. For those with metastatic visceral angiosarcoma, both anthracycline-based regimens and taxanes are equally appropriate options for initial therapy. (See 'Angiosarcoma' above.)

•PEComa – For patients with advanced unresectable or metastatic malignant neoplasm with perivascular epithelioid cell differentiation (PEComa), we suggest initial therapy with nab-sirolimus over other systemic agents (Grade 2C), which is also the only drug approved by the US Food and Drug Administration (FDA) for this disease. For patients living in areas where nab-sirolimus is not approved or unavailable, other mechanistic (mammalian) target of rapamycin (mTOR) inhibitors (sirolimus, everolimus, and temsirolimus) are reasonable alternatives. (See 'PEComa' above.)

•Other histologies – The management of other metastatic STS histologies is discussed separately. (See 'Other histologies' above.)

●NTRK fusion-positive tumors – For patients with tumors expressing the NTRK gene fusion, we offer initial targeted therapy with either or larotrectinib or entrectinib. We typically prefer larotrectinib over entrectinib because it has more durable responses and a better toxicity profile. (See 'NTRK gene fusion-positive tumors' above and "TRK fusion-positive cancers and TRK inhibitor therapy".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges George Demetri, MD, who contributed to earlier versions of this topic review.