Handbook of Cancer Chemotherapy (Lippincott Williams & Wilkins Handbook Series), 8th Ed.

16. Soft-Tissue Sarcomas

Robert S. Benjamin


A. Types of soft-tissue sarcomas

The soft-tissue sarcomas are a group of diseases characterized by neoplastic proliferation of tissue of mesenchymal origin. Thus, they differ from the more common carcinomas, which arise from epithelial tissue. Sarcomas can arise in any area of the body and from any origin; however, they most commonly arise in the soft tissue of the extremities, trunk, retroperitoneum, or head and neck area. There are more than 50 different types of sarcomas, classified according to lines of differentiation toward normal tissue. For example, rhabdomyosarcoma shows evidence of skeletal muscle fibers with cross-striations, liposarcoma shows fat production, and angiosarcoma shows vessel formation, and there are several types within each of these groups. Precise characterization of the types of sarcoma is often impossible, and these tumors are called unclassified sarcomas. All of the primary bone sarcomas may arise in soft tissue, leading to such diagnoses as extraskeletal osteosarcoma, extraskeletal Ewing sarcoma, and extraskeletal chondrosarcoma. A common diagnosis in the recent past was malignant fibrous histiocytoma (MFH). This tumor is characterized by a mixture of spindle (or fibrous) cells and round (or histiocytic) cells arranged in a storiform pattern with frequent areas of pleomorphic appearance and frequent giant cells. There is no evidence of differentiation toward any particular tissue type. Many tumors previously called pleomorphic fibrosarcoma, pleomorphic rhabdomyosarcoma, and so forth were classified as MFH. As immunohistochemistry and molecular diagnostic techniques have improved, many of the tumors previously classified as MFH have been reclassified as pleomorphic something else. Furthermore, there are strong opponents of the term MFH because there is no evidence that the tumors have either fibrous or histiocytic origin, and pleomorphic tumors previously classified as MFH are frequently now referred to as unclassified high-grade pleomorphic sarcomas.

B. Metastases

Metastatic spread of all sarcomas tends to be through the blood rather than through the lymphatic system. The lungs are by far the most frequent site of metastatic disease. Local sites of metastasis by direct invasion are the second most common area of involvement, followed by bone and liver. (Liver metastases are common with intra-abdominal sarcomas, especially gastrointestinal stromal tumors [GISTs]; however, metastases to soft tissue are common with myxoid liposarcomas.) Central nervous system (CNS) metastases are extraordinarily rare except in alveolar soft-part sarcoma.

C. Staging

Staging of sarcomas is complex and demands an expert sarcoma pathologist. Tumors have been staged according to two systems: the American Joint Committee on Cancer (AJCC) staging system and the Musculoskeletal Tumor Society staging system. The new International Union Against Cancer (UICC)/AJCC staging system, with international acceptance, takes portions from each of the older systems and more appropriately identifies patients at increased risk of metastatic disease. Further revisions to this system are still under way, and a final, widely accepted system is still not universally accepted. As current and older publications still refer to the older systems, however, all will be included.

1. The old AJCC staging system

a. Tumor grade. The primary determinant of stage is tumor grade. Grade 1 tumors are stage I; grade 2 tumors are stage II; and grade 3 tumors are stage III. Any tumor with lymph node metastases is automatically stage III. Any tumor with gross invasion of bone, major vessel, or major nerve is stage IV.

b. Stage. Further divisions of stages I to III into A and B are based on tumor size.

bull A: Tumor smaller than 5 cm

bull B: Tumor size 5 cm or larger.

In stage III, lymph node metastases are classified as IIIC. In stage IV, local invasion is called IVA, and IVB represents distant metastases.

2. The Musculoskeletal Tumor Society staging system. The Musculoskeletal Tumor Society stages sarcomas according to grade and compartmental localization. The Roman numeral reflects the tumor grade.

bull Stage I: Low grade

bull Stage II: High grade

bull Stage III: Any-grade tumor with distant metastasis.

The letter reflects compartmental localization. Compartments are defined by fascial planes.

bull Stage A: Intracompartmental (i.e., confined to the same soft-tissue compartment as the initial tumor)

bull Stage B: Extracompartmental (i.e., extending outside of the initial soft-tissue compartment into the adjacent soft-tissue compartment or bone).

A stage IA tumor is a low-grade tumor confined to its initial compartment, a stage IB tumor is a low-grade tumor extending outside the initial compartment, and so forth.

3. The new AJCC staging system. The stage is determined by tumor grade, tumor size, and tumor location relative to the muscular fascia. Tere are now four tumor grades.

bull Grade 1: Well differentiated

bull Grade 2: Moderately differentiated

bull Grade 3: Poorly differentiated

bull Grade 4: Undifferentiated.

Tumor size is now divided at less than or equal to 5 cm or more than 5 cm (in the old AJCC system, it was less than 5 cm or more than or equal to 5 cm).

bull T1: ≤5 cm

bull T2: >5 cm.

Tumor status is subdivided by location relative to the muscular fascia.

bull Ta: Superficial to the muscular fascia

bull Tb: Deep to the muscular fascia.

The AJCC stage grouping is as follows.


The new staging system divides patients according to necessary therapy.

bull Stage I patients are adequately treated by surgery alone.

bull Stage II patients require adjuvant radiation therapy.

bull Stage III patients require adjuvant chemotherapy.

bull Stage IV patients are managed primarily with chemotherapy, with or without other modalities.

D. Evaluation

Patients are evaluated and followed according to the plan in Table 16.1.



E. Primary treatment

1. Surgery and radicrtherapy. Treatment of the primary tamor involves surgery with or without radiation therapy. If radiation therapy is not used, surgery must be radical. Although this may often involve amputation or complete excision of the involved muscle group from origin to insertion, more and more frequently, wide local resection is performed, with or without adjuvant radiation, depending on stage and extent of negative margins.

2. Adjuvant chemotherapy. The role of adjuvant chemotherapy remains controversial, with both positive and negative results reported. A meta-analysis of individual patient data indicated a highly significant decrease in the risk of disease recurrence (either local or distant) and death in patients treated with adjuvant chemotherapy; thus, some investigators believe that adjuvant therapy is clearly indicated for patients whose histologic type, grade, or location is known to convey a poor prognosis. The meta-analysis confirms a survival benefit for patients with primary sarcomas of the extremities as well as increased local or distant disease-free interval for all patients treated with doxorubicin (Adriamycin)-based adjuvant chemotherapy. The Italian Cooperative Study Group using epirubicin and ifosfamide for patients with current stage III disease also demonstrated overall survival and disease-free survival advantage for patients treated with chemotherapy; however, the statistical significance of the survival advantage was lost with further follow-up. An updated meta-analysis, including studies combining ifosfamide with anthracyclines, shows a survival advantage, but the largest, modern study from the European Organization for the Research and Treatment of Cancer is negative; thus the controversies continue.

F. Prognosis

Prognosis is related to stage, with a 5-year survival rate of 99% for new AJCC/UICC stage I, 82% for stage II, and 52% for stage III. Corresponding rates of disease-free survival at 5 years are 78% for stage I, 64% for stage II, and 36% for stage III. Long-term results are still worse. Te survival rate for stage IV disease is less than 10%; however, a definite fraction of patients in this category can be cured. Most patients with stage IV disease, if left untreated, die within 6 to 12 months; however, there is great variation in actual survival, and patients may go on with slowly progressive disease for many years.

G. Treatment response

Response to treatment is measured in the standard fashion for solid tumors with the addition of tumor necrosis, both radiologically and pathologically, but there are increasing examples where good responses are missed by standard criteria, and newer approaches to computed tomography (CT) and magnetic resonance imaging (MRI) evaluation and use of positron emission tomography is becoming more frequent.

1. Complete remission. This implies complete disappearance of all signs and symptoms of disease.

2. Partial remission. Standard Response Evaluation Criteria in Solid Tumors (RECIST) (see Chapter 3Section IV.D.1) are generally employed. Tis requires a 30% or greater decrease in measurable disease, calculated by comparing the sum of the longest diameters of all lesions before and after therapy. When disease can be followed objectively by MRI or CT, marked tumor necrosis attributable to chemotherapy demonstrated by imaging or pathology is at least the equivalent of a partial response by RECIST criteria. For GIST, Choi criteria—a 10% decrease in size or a 15% decrease in tumor density—are more predictive of time to progression or disease-specific survival than RECIST response, and Choi criteria or minor modifications thereof are under study in other sarcomas.

3. Stable disease or improvement. Lesser degrees of tumor shrinkage are categorized by some physicians as stable disease and by others as improvement or minor response. Stable disease implies a smaller than 20% increase in disease for at least 8 weeks. There is increasing recognition that stable disease or improvement that persists for at least 4 months is at least as meaningful for ultimate patient benefit as partial response. For all response categories, no new disease must appear during response. Many investigators consider absence ofprogressive disease by RECIST, especially if maintained for more than 24 weeks, to be as good as response.

4. Progression. New disease in any area or a 20% or more increase in measurable disease constitutes progressive disease.

5. Survival. All patients whose disease responds objectively to chemotherapy survive longer than do patients with progressive disease, and the degree of prolongation of survival is directly proportional to the degree of antitumor response that can be measured.


A. General considerations and aims of therapy

Although there are numerous types of soft-tissue sarcomas, there are few differences among them regarding responsiveness to a standard soft-tissue sarcoma regimen. GISTs and alveolar soft-part sarcomas and, to a lesser extent, solitary fibrous tumors (also called hemangiopericytomas), clear cell sarcomas, and epithelioid sarcomas respond less frequently to standard regimens than do the other soft-tissue sarcomas. GISTs, in particular, should not be treated with doxorubicin- and ifosfamide-based chemotherapy. GISTs are usually characterized by mutated c-Kit and have a high response rate with prolonged remissions after treatment with imatinib at 400 mg daily. Patients who do not respond or who relapse after initial therapy may respond to higher doses up to 800 mg in divided doses daily or to sunitinib. Patients with GIST and exon 9 mutations respond more frequently and for longer to 800 mg of imatinib than to 400 mg. There is increasing suggestion that alveolar soft-part sarcomas should be treated with vascular endothelial growth factor inhibitors, and that solitary fibrous tumors are best treated with a combination of antiangiogenic agents and chemotherapy (e.g., bevacizumab-temozolomide). Angiosarcomas, especially those primary in the skin, can respond to paclitaxel, while other sarcomas do not. Two tumors—Ewing sarcoma and rhabdomyosarcoma—particularly in children, are responsive to dactinomycin, vincristine, or etoposide. The other tumors are not. The goal of therapy for patients with advanced disease is primarily palliative, although a small fraction (about 20%) of patients who achieve complete remission are, in fact, cured. The first aim, therefore, is to achieve complete remission. Several investigators, including the author, have shown that the prognosis is the same whether complete remission is obtained by chemotherapy alone or by chemotherapy with adjuvant surgery, that is, surgical removal of all residual disease. Short of complete remission, partial remission causes some palliation, with relief of symptoms and prolongation of survival by about 1 year. Any degree of improvement or stabilization of previously advancing disease likewise increases survival.

B. Effective drugs

The most important chemotherapeutic agent is doxorubicin, which forms the backbone of all combination chemotherapy regimens. Ifosfamide, an analog of cyclophosphamide that has documented activity even in patients who are refractory to combinations containing cyclophosphamide, is usually included in frontline chemotherapy combinations. It is always given together with the uroprotective agent mesna to prevent hemorrhagic cystitis. Dacarbazine, a marginal agent by itself, adds significantly to doxorubicin in prolonging remission duration and survival as well as increasing the response rate. Cyclophosphamide adds marginally, if at all, but is included in some effective regimens. Gemcitabine is an active agent, more so when combined with docetaxel, but with exceptions of uterine leiomyosarcoma and some angiosarcomas, these agents are usually reserved for second-line therapy. Trabectedin, approved in most countries outside of the United States, is also an active agent.

The key to effective sarcoma chemotherapy is the steep dose–response curve for doxorubicin. At a dose of 45 mg/m2, the response rate is lower than 20% compared with a 37% response rate at a dose of 75 mg/m2. A similar dose–response relationship exists for ifosfamide and for combination chemotherapy, and the regimens with the best reported results are those using the highest doses.

C. Primary chemotherapy regimen (adjuvant or advanced)

The most effective primary chemotherapy regimens include doxorubicin and ifosfamide (high-dose AI) or doxorubicin and dacarbazine (ADIC), with or without the addition of cyclophosphamide (CyADIC) or ifosfamide and mesna (MAID). The CyADIC regimen is a modification of the standard CyVADIC regimen, which includes vincristine. Because analysis has shown that vincristine makes no significant contribution and produces neurotoxicity, its addition at a dose of 2 mg maximum or 1.4 mg/m2 weekly for 6 weeks and then once every 3 to 4 weeks is recommended only for treatment of rhabdomyosarcoma and Ewing sarcoma.

By giving doxorubicin and dacarbazine by continuous 72- or 96-hour infusion, with the two drugs mixed in the same infusion pump, nausea and vomiting are markedly reduced, and the chemotherapy can be continued until a cumulative doxorubicin dose of 600 to 800 mg/m2 is reached, with less cardiac toxicity than with standard doxorubicin administration and a cumulative dose of 450 mg/m2.

1. The high-dose AI regimen is as follows:

bull Doxorubicin by continuous 72-hour infusion at 75 mg/m2 IV (25 mg/m2/day for 3 days), and

bull Ifosfamide 2.5 g/m2 intravenously (IV) over 2 to 3 hours daily for 4 days.

bull Vincristine 2 mg total dose is added on day 1 for small-cell tumors such as rhabdomyosarcoma and Ewing sarcoma (high-dose AI with vincristine).

bull Mesna 500 mg/m2 is mixed with the first ifosfamide dose, and 1500 mg/m2 is given as a continuous infusion over 24 hours for 4 days in 2 L of alkaline fluid.

bull Filgrastim (granulocyte colony-stimulating factor) 5 (μg/kg subcutaneously (SC) is given on days 5 to 15 or until granulocyte recovery to 1500/(μL. Alternatively, pegfilgrastim at a dose of 6 mg is given on day 5.

bull Repeat cycle every 3 weeks.

2. The continuous-infusion CyADIC regimen is as follows:

bull Cyclophosphamide 600 mg/m2 IV on day 1, and

bull Doxorubicin, by continuous 96-hour infusion at 60 mg/m2 IV (15 mg/m2/day for 4 days), and

bull Dacarbazine by continuous 96-hour infusion at 1000 mg/m2 IV (250 mg/m2/day for 4 days) mixed in the same bag or pump as the doxorubicin. Doses should be divided into four consecutive 24-hour infusions.

3. The continuous-infusion ADIC regimen is as follows:

bull Doxorubicin by continuous 96-hour infusion at 90 mg/m2 IV (22.5 mg/m2/day for 4 days), and

bull Dacarbazine by continuous 96-hour infusion at 900 mg/m2 IV (225 mg/m2/day for 4 days) mixed in the same bag or pump as the doxorubicin. Doses should be divided into four consecutive 24-hour infusions.

bull Repeat cycle every 3 to 4 weeks.

4. The MAID regimen is as follows:

bull Mesna by continuous 96-hour infusion at 8000 mg/m2 IV (2000 mg/m2/day for 4 days).

bull Doxorubicin by continuous 72-hour infusion at 60 mg/m2 IV (20 mg/m2/day for 3 days).

bull Ifosfamide by continuous 72-hour infusion at 6000 mg/m2 IV (2000 mg/m2/day for 3 days). Doses should be divided into three consecutive 24-hour infusions. (Some investigators prefer to infuse ifosfamide over 2 hours rather than 24 hours because of higher single-agent activity with the shorter infusions.

bull Dacarbazine by continuous 72-hour infusion at 900 mg/m2 IV (300 mg/m2/day for 3 days) mixed in the same bag or pump as the doxorubicin. Doses should be divided into three consecutive 24-hour infusions.

bull Repeat cycle every 3 to 4 weeks.

5. Dose modification. Doses of doxorubicin, cyclophosphamide, ifosfamide, and mesna should be increased by 25% and may be decreased by 20% for each course of therapy to achieve a lowest absolute granulocyte count of about 500/μL if growth factors are not used. The maximum doxorubicin dose is limited to 600 to 800 mg/m2, depending on the duration (48 to 96 hours) of infusion, at which point therapy should be discontinued unless cardiac biopsy specimens indicate that it is safe to continue. With Ewing sarcoma and rhabdomyosarcoma, therapy may be continued, and dactinomycin 2 mg/m2 in a single dose or 0.5 mg/m2 daily for 5 days may be substituted for the doxorubicin, with continuation of the regimen for a total of 18 months.

6. An alternative regimen for children with rhabdomyosarcoma is an alternating regimen, using ifosfamide and etoposide alternating with the so-called vincristine, Adriamycin, and cyclophosphamide (VAdriaC) regimen.

bull Vincristine 1.5 mg/m2 is given weekly × 3 for the first two cycles of VAdriaC and then on day 1 only.

bull Doxorubicin is given at a dose of 60 to 75 mg/m2 as a 48-hour continuous infusion, and

bull Cyclophosphamide 600 mg/m2 is given daily for 2 days (with mesna).

After 3 weeks,

bull Ifosfamide is given at a dose of 1800 mg/m2 daily for 5 days (with mesna), and

bull Etoposide is given at a dose of 100 mg/m2 daily for 5 days. Chemotherapy cycles are alternated every 3 weeks for 39 weeks.

7. Vincristine, dactinomycin, and cyclophosphamide (VAC). A less-intensive, older, but still effective regimen for children with good-prognosis rhabdomyosarcoma is the so-called pulse VAC regimen. Dactinomycin is given at a total dose of 2 to 2.5 mg/m2 by divided daily injection over 5 to 7 days (e.g., 0.5 mg/m2 daily for 5 days) repeated every 3 months for a total of five courses. Cyclophosphamide pulses of 275 to 330 mg/m2 daily for 7 days are begun at the same time but are given every 6 weeks with vincristine 2 mg/m2 on days 1 and 8 of each cyclophosphamide cycle. Cyclophosphamide cycles are terminated prematurely if the white blood cell counts fall below 1500/(μL. Chemotherapy continues for 2 years. (The necessity of the 2-year duration of the chemotherapy program is not certain.)

D. Secondary chemotherapy

Secondary chemotherapy for patients with sarcoma is relatively unrewarding, with response rates lower than 10% for almost all conventional drugs or regimens tested. The best commercially available drug is ifosfamide, which, if not used in primary treatment, produces a response in about 20% of patients. High-dose ifosfamide (12 g/m2 or higher) may produce responses in patients resistant to lower doses in combination. Gemcitabine in the author's experience has a response rate of18% and has become the standard drug for salvage therapy. Recent data indicates that the combination of gemcitabine and docetaxel (the Gem-Tax regimen) improves response rate, time to progression, and survival in a randomized comparison with gemcitabine alone; thus the Gem-Tax regimen should be considered as the secondary treatment of choice for most soft-tissue sarcomas. Exceptions are myxoid liposarcoma and synovial sarcoma, where the regimen has minimal activity. In contrast, at least for myxoid liposarcoma, trabectedin is highly active and should be considered the secondary treatment of choice. The Gem-Tax regimen is as follows:

bull Gemcitabine 900 mg/m2 over 90 minutes on days 1 and 8.

bull Docetaxel 100 mg/m2 on day 8 only.

bull Filgrastim (granulocyte colony-stimulating factor) 5 µg/kg SC is given on days 9 to 15 or until granulocyte recovery to 1500/μL.

Alternatively, pegfilgrastim at a dose of 6 mg is given on day 9.

The duration of gemcitabine infusion is critical, as it can only be converted to its active metabolite, gemcitabine triphosphate, at a rate of 10 mg/m2/min. Doses are reduced 25% to 675 mg/m2 and 75 mg/m2, respectively, for patients with extensive prior therapy or pelvic radiation. Dexamethasone 8 mg by mouth twice a day should be given for 3 days starting 1 day prior to docetaxel.

Trabectedin has a response rate of no more than 10% to 20% overall but can cause prolonged stability in previously progressing patients, dramatic tumor regressions in a few, and frequent responses in those with myxoid liposarcoma. For patients with this subtype, RECIST responses are seen in about 40% of patients, Choi responses in about 80% of patients, and progressive disease in just over 10%.

Trabectedin is administered as follows:

bull Trabectedin 1.5 mg/m2 over 24 hours IV day 1 only. Premedicate on day 0 with dexamethasone 8 mg by mouth.

bull Pegfilgrastim 6 mg SC on day 3.

Patients who do not respond to doxorubicin, ifosfamide, Gem-Tax, or trabectedin should be entered in a phase II study of a new agent to see if some activity can be established because other reasonably good alternatives do not exist.

E. Complications of chemotherapy

Side effects of sarcoma chemotherapy can be classified into three categories: life threatening, potentially dangerous, and unpleasant.

1. Life-threatening complications of chemotherapy are infection or bleeding. Trombocytopenia lower than 20,000/μL occurs with this type of chemotherapy when growth factors are used to maintain dose intensity, but bleeding is rare and can be minimized by transfusing platelets at 10,000/μL. Approximately 20% to 40% of patients have documented or suspected infection related to drug-induced neutropenia at some time during their treatment course. These infections are rarely fatal if treated promptly with broad-spectrum, bactericidal antibiotics at the onset of the febrile neutropenia episode.

2. Potentially dangerous side effects of chemotherapy include the following.

a. Mucositis, which occurs in fewer than 25% of patients, may interfere with oral intake or may act as a source of infection.

b. Granulocytopenia predisposes the patient to infection but, because of its brevity, rarely causes infection.

c. Cardiac damage from doxorubicin rarely causes clinical problems at the doses recommended, with usually reversible congestive heart failure occurring in fewer than 5% of patients.

d. Renal insufficiency is a rare complication of ifosfamide. Fanconi syndrome, particularly manifested by a significant loss of bicarbonate, is a dose-related complication of ifosfamide, occurring in 10% to 30% of patients at standard ifosfamide doses and in close to 100% with high-dose regimens, the morbidity of which can be minimized by the routine use of alkaline infusions and correction of electrolyte levels with intravenous or oral replacement therapy. Only rarely does the nephrotoxicity progress to renal failure, often precipitated by dehydration or administration of minimally nephrotoxic drugs such as nonsteroidal anti-inflammatory drugs (NSAIDs). Patients treated with ifosfamide should be instructed to avoid NSAIDs, even years after chemotherapy!

e. CNS toxicity of ifosfamide is rarely a serious complication. Patients frequently demonstrate minor confusion, disorientation, or difficulty with fine movements. Somnolence and coma are rarely seen in patients without hypoalbuminemia and/or acidosis.

f. Hemorrhagic cystitis, arare complication of cyclophosphamide therapy, used to be the dose-limiting toxicity of ifosfamide. It can be prevented in most cases by administration of another agent, mesna, before and after each ifosfamide dose, allowing higher doses of ifosfamide to be used.

g. Pulmonary toxicity, manifested by increasing dyspnea, is seen in less than 10% of patients treated with the Gem-Tax combination, but occurs with about twice the frequency of that seen with gemcitabine alone. Careful attention to the possible occurrence of this problem and prompt treatment with high doses of corticosteroids can be life-saving.

h. Hepatic toxicity is the limiting toxicity of trabectedin. If the drug is withheld when alkaline phosphatase is elevated, serious hepatitis is rarely a problem, and the incidence of this effect is less than 5% when dexamethasone premedication is utilized.

3. Unpleasant but rarely serious problems include nausea and vomiting (primarily from dacarbazine, ifosfamide, and docetaxel) and alopecia (from doxorubicin, cyclophosphamide, ifosfamide, and docetaxel). Gemcitabine, and to a greater extent, the Gem-Tax combination, can cause profound fatigue. Gemcitabine can also cause drug fever and a rash (often confused for cellulitis) that can respond to corticosteroids.

F. Special precautions

1. Ifosfamide. Patients must be kept well hydrated with an alkaline pH to prevent CNS toxicity and minimize nephrotoxicity. Sodium bicarbonate or sodium acetate should be added to IV fluids at an initial concentration of 100 to 150 mEq/L, and fluid administration should be adjusted to produce a urine output of at least 2 L/day and to maintain the serum bicarbonate concentration at 25 mEq/L or higher. Other electrolytes should be adjusted as needed on a daily basis. Serum albumin should be kept within normal limits.

2. Doxorubicin. Avoid extravasation. Continuous infusions must (and short infusions should) be administered through a central venous catheter. Attention to cumulative dose administered (varying according to the schedule of administration) is critical to minimize the risk of cardiac toxicity.

3. Trabectedin. Avoid extravasation. Continuous infusions must (and short infusions should) be administered through a central venous catheter.

Selected Readings

Antman KH, Crowley J, Balcerzak SP, et al. An intergroup phase III randomized study of doxorubicin and dacarbazine with or without ifosfamide and mesna in advanced soft tissue and bone sarcomas. J Clin Oncol. 1993;11:1276–1285.

Antman KH, Montella D, Rosenbaum C, Schwen M. Phase II trial of ifosfamide with mesna in previously treated metastatic sarcoma. Cancer Treat Rep. 1985;69:499–504.

Benjamin RS, Legha SS, Patel RS, Nicaise C. Single agent ifosfamide studies in sarcomas of soft tissue and bone: the M.D. Anderson experience. Cancer Chemother Pharmacol. 1993;31:S174–S179.

Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med. 2002;347:472–480.

Elias A, Ryan L, Sulkes A, et al. Response to mesna, doxorubicin, ifosfamide, and dacarbazine in 108 patients with metastatic or unresectable sarcoma and no prior chemotherapy. J Clin Oncol. 1989;7:1208–1216.

Fata F, O’Rielly E, Ilson D, et al. Paclitaxel in the treatment of patients with angiosarcoma of the scalp or face. Cancer. 1999;86:2034–2037.

Frustaci S, Gherlinzoni F, De Paoli A, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian Randomized Cooperative Trial. J Clin Oncol.2001;19:1238–1247.

Grosso F, Jones RL, Demetri GD, et al. Efficacy of trabectedin (ecteinascidin-743) in advanced pretreated myxoid liposarcomas: a retrospective study. Lancet Oncol. 2007;8:595–602.

Harrison L, Franzese F, Gaynor J, Brennan MF. Long-term results of a prospective randomized trial of adjuvant brachytherapy in the management of completely resected soft tissue sarcomas of the extremity and superficial trunk. Int J Radiat Oncol Biol Phys. 1993;27:259–265.

Joensuu H, Roberts PJ, Sarlomo-Rikala M, et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med. 2001;344:1052–1056.

Le Cesne A, Blay JY, Judson I, et al. Phase II study of ET-743 in advanced soft tissue sarcomas: a European Organisation for the Research and Treatment of Cancer (EORTC) soft tissue and bone sarcoma group trial. J Clin Oncol.2005;23:576–584.

Lindberg RD, Martin RG, Romsdahl MM, et al. Conservative surgery and radiation therapy for soft tissue sarcomas. In: Martin RG, Ayala AG, eds. Management of primary bone and soft tissue tumors.Chicago: Year Book; 1977:289–298.

Maki RG, Wathen JK, Patel SR et al. An adaptively randomized phase 2 study of gemcitabine and docetaxel vs. gemcitabine alone in patients with metastatic soft-tissue sarcomas. J Clin Oncol. 2007;25:2755–2763.

Patel SR, Benjamin RS. Sarcomas: part I and II. Hematol Oncol Clin North Am. 1995;9:513–942.

Patel SR, Gandhi V Jenkins J, et al. Phase II clinical investigation of gemcitabine in advanced soft tissue sarcomas and window evaluation of dose-rate on gemcitabine triphosphate accumulation. J Clin Oncol.2001;19:3483–3489.

Patel SR, Vadhan-Raj S, Burgess MA, et al. Results of two consecutive trials of dose-intensive chemotherapy with doxorubicin and ifosfamide in patients with soft-tissue sarcomas. Am J Clin Oncol.1998;21:317–321.

Patel SR, Vadhan-Raj S, Papadopoulos N, et al. High-dose ifosfamide in bone and soft-tissue sarcomas: results of phase II and pilot studies—dose response and schedule dependence. J Clin Oncol.1997;15:2378–2384.

Pisters P, Leung D, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1,041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol. 1996;14:1679–1689.

Soft tissue sarcoma. In: Greene FL, Page DL, Fleming ID, et al. (eds.) AJCC cancer staging manual (6th ed.). NewYork, NY: Springer-Verlag; 2002:193–197.

Sarcoma Meta-analysis Collaboration. Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta-analysis of individual data. Lancet. 1997; 350:1647–1654.

Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst. 2000;92:205–216.

van Oosterom AT, Judson I, Verweij J, et al. Safety and efficacy of imatinib (STI571) in metastatic gastrointestinal stromal tumours: a phase I study. Lancet. 2001; 358:1421–1423.

Verweij J, Casali PG, ZalcbergJ, et al. Progression-free survival in gastrointestinal stromal tumours with high-dose imatinib: randomised trial. Lancet. 2004;364:1127–1134.

Wunder J, Healey J, Davis A, Brennan MF. A comparison of staging systems for localized extremity soft tissue sarcoma. Cancer. 2000;88:2721–2730.

Zalupski MM, Ryan J, Hussein M, et al. Defining the role of adjuvant chemotherapy for patients with soft tissue sarcoma of the extremities. In: Salmon SE, ed. Adjuvant therapy of cancer VII. Philadelphia: Lippincott; 1993:385–392.