As clinicians have gained an improved understanding of the biology of soft-tissue sarcoma malignancies, the ability to better distinguish and identify subtypes has extended the hope of targeted treatment options.
Steven Eric Finkelstein, MD, FACRO
As clinicians have gained an improved understanding of the biology of soft-tissue sarcoma (STS) malignancies, the ability to better distinguish and identify STS subtypes has extended the hope that exploiting the molecular characteristics of each subtype will lead to the emergence of targeted treatment options. For patients who experience STS in the extremities, radiation therapy is used to facilitate limb- and function- preserving surgeries while maintaining 5-year local control above 85%. For STS originating from nonextremity sites, in contrast, the rate of 5-year local control is only about 50%. What follows is a review of several studies that have evaluated the role of adjuvant or neoadjuvant chemotherapy for the treatment of localized STS.
In the United States, approximately 15,000 new cases of sarcoma of the bone and soft tissues are diagnosed annually, according to data from the Surveillance, Epidemiology, and End Results database.1 The incidence of sarcoma is within the same order of magnitiude as other malignancies such as myeloma, cervical, glioma, and esophageal, but more frequent that Hodkin lymphoma or testicular cancer. The disease affects all age groups, and 15% are found in children aged 15 years and younger, and 40% in those 55 years and older.1 These tumors encompass a variety of histological subtypes. They appear to arise from the supporting connective tissues, including smooth and skeletal muscles, tendons, fat, fibrous tissue, synovial tissue, vessels, and nerves.2 Approximately 60% of soft-tissue sarcoma (STS) malignancies arise in the extremities, most frequently in the lower extremity.
Despite STS being a relatively uncommon malignancy, research investigating the genetic and molecular characteristics of its subtypes has enhanced our understanding of cancer biology overall, and it has led to the development of targeted therapeutics with applicability to a broad range of other malignancies.
To give several examples, the pioneering use of neoadjuvant chemotherapy was first realized in the treatment of pediatric osteosarcoma3; the first demonstration of a chromosomal translocation manifesting as a pathogenic fusion gene and protein product in solid tumors was4; and the effectiveness of agents like imatinib mesylate (Gleevec) in treating gastrointestinal stromal tumors (GISTs) was first learned during research conducted in sarcoma. Here, we discuss the state of the science for the treatment of sarcoma.
Historically, the standard of therapy for soft-tissue sarcoma affecting the extremities was amputation. Function-preserving limb-conserving resection with negative microscopic margins is now considered the standard of care. The goal of neoadjuvant therapy is to improve both local and—depending on modality—distant control. Risk of recurrence will vary based on histologic subtype, tumor grade, and size. The potential benefit from neoadjuvant therapies, however, must be weighed against the risk of disease progression due to delayed resection during treatment, as well as an increased risk of postoperative complications.Cooperative groups and several institutions of research have been the vanguard of sarcoma research for the past 30 years, particularly in the clinical and translational arena. Notable advancements in the treatment of adult STS, from 1970 to 2000, include improvements in pathologic definition through identification of specific translocations and growing improvements in immunohistochemistry; new imaging modalities, including positron emission tomography (PET) scans; refinements in prognosis and staging; advances in the precision of radiotherapy techniques; surgical advances in functional preservation; and the emergence of doxorubicin and ifosfamide as active drugs, as well as the development of more effective systemic agents.
Superior local control with the addition of adjuvant radiation therapy (RT) after limb-sparing surgery in STS was demonstrated in two randomized controlled trials. The first was reported by Yang and colleagues, from the National Cancer Institute (NCI) of Canada, in which the investigators assessed the impact of postoperative external-beam radiation therapy on local recurrence (LR), overall survival (OS), and quality of life after limb-sparing resection of extremity sarcomas.5 Patients with high-grade STS received concurrent doxorubicin and cyclophosphamide with external beam radiation therapy (EBRT). For patients with high-grade tumors, 10-year local control (LC) was significantly higher for those treated with RT versus those who were not (100% vs 78%), but there was no difference in 10-year distant metastasis-free survival (DMFS) or overall survival (OS). For patients with low-grade tumors, LC was similarly statistically higher for those treated with RT, and again, there was no difference in DMFS or OS.5 In the second, Pisters and colleagues conducted a trial in which patients with high- and low-grade STS who had a complete resection were randomized while in surgery to receive RT in the form of an Iridium-192 brachytherapy implant (42 to 45 Gray [Gy]) over the next 4 to 6 days, or no RT. For patients with high-grade tumors, 5-year LC favored the RT arm (89% vs 66%, respectively), but there was no OS difference. For patients with low-grade tumors, neither LC nor OS were significantly impacted by RT.6
Interestingly, the NCI Canada trial randomized patients with extremity STS to preoperative RT (50 Gy in 25 fractions [fx] 16 to 20 Gy boost treatment for positive surgical margins) versus postoperative RT (66 Gy in 33 fx). With respect to radiation treatment field design, the initial treatment field used 5-cm proximal and distal margins beyond the tumor (preoperative RT) or tissues at risk (postoperative RT). Additionally, the boost was designed with 2-cm proximal and distal margins. The primary endpoint of this study was the rate of major wound complications within 120 days of surgery. The trial closed after accruing 190 of the planned 266 patients because of a significantly higher rate of wound complications with preoperative RT (35%) versus postoperative RT (17%), with the highest rates of complications in the upper leg (45%). In this study, function at six-weeks was better with postoperative RT.7 At a median follow-up of 6.9 years, there was no significant difference in LC (93% for preoperative RT vs 92% for postoperative RT), recurrence-free survival (RFS; 58% vs 59%), or OS (73% vs 67%). With respect to significant adverse predictors for outcome, positive surgical margins significant for LC; large tumor size for and high grade for RFS and OS.8 With respect to long-term toxicity, postoperative RT was associated with worse fibrosis and joint stiffness compared with preoperative RT.
Given that the efficacies of preoperative and postoperative RT are deemed equivalent, decisions regarding which approach to use are driven by the difference in toxicity profiles. Preoperative RT is associated with a higher rate of wound complications, which are typically reversible, whereas postoperative RT is associated with higher rates of limb edema, fibrosis, and stiffness, which are typically not reversible.
The role of adjuvant or neoadjuvant chemotherapy in localized STS has been investigated in several studies. Specifically, the National Cancer Institute performed a trial that randomized 43 patients with high-grade extremity STS to amputation at the joint proximal to the tumor versus limb-sparing resection and postoperative RT. Randomization favored limb sparing (2:1). The RT dose was 45 to 50 Gy, plus a boost dose to 60 to 70 Gy. RT was delivered with concurrent doxorubicin, cyclophosphamide, and high-dose methotrexate. With respect to margin status, 4 of 27 patients in the RT group had positive margins. There was increased LR with limb-sparing and adjuvant RT compared with amputation (20% vs 0%, respectively), but the impressive limb preservation rate established limited surgery and RT as the new standard of care.10 There were no significant differences in 5-year DFS (78% for the amputation group vs 71% for the limb-sparing and RT group) or in OS (88% vs 83%, respectively). In a separate trial, Rosenberg and colleagues reported in the same publication, 65 patients with extremity STS were randomized to receive surgery alone (either limb-sparing or amputation) versus surgery and adjuvant doxorubicin, cyclophosphamide, and high-dose methotrexate. Compared with patients who did not receive chemotherapy, those who received adjuvant chemotherapy enjoyed both improved DFS (92% vs 60%, P = .0008) and OS (95% vs 74%, P = .004).10
A study from Massachusetts investigated neoadjuvant chemoradiation (CRT) for large STS included 48 patients with >8 cm STS of the extremities. Patients underwent intergitated sequential CRT as follows: mesna/doxorubicin/ifosfamide/dacarbazine (MAID) followed by RT (22 Gy in 11 fx), followed by MAID, followed by RT (22 Gy in 11 fx), followed by MAID. The researchers from Harvard University found that if surgical margins were positive, patients received an additional 16 Gy boost postoperatively. Five-year LC was 92%, DFS was 86%, and OS was 44%. Compared with historical controls, there was a significant decrease in distant metastatses. and significant increases in DFS and OS. There was a 29% wound complication rate and 2% treatment-related deaths.11
The findings of this study led to a multi-institutional cooperative group follow up trial, Radiation Therapy Oncology Group (RTOG) 9514. This was a phase II trial of 64 patients with >8 cm grade 2 or 3 STS of the extremity or torso. Investigators found that 44% of patients had malignant fibrous histiocytoma, 13% had leiomyosarcoma, and 88% had STS of the extremity. Patients underwent interdigitating chemoradiation, which was defined as 1 cycle of MAID, followed by RT (22 Gy in 11 fx), followed by a second cycle of MAID, a second course of RT (22 Gy in 11 fx), and then a third cycle of MAID. Patients then underwent resection followed by fourth and fifth cycles of adjuvant MAID, preceded by a 14 Gy postoperative boost in the setting of positive surgical margins. The researchers reported that 91% of R0 resections were achieved, and 59% of patients received the full chemotherapy course. The 3-year locoregional failure was 18% (if amputation was considered a failure; 10%, if not). The 3-year DFS was 57%, distant DFS was 64%, OS was 75%, and there was a 92% amputation-free rate. There was a 5% rate of treatment-related deaths (including 2 secondary acute myeloid leukemias), and 84% of patients had grade 4 toxicity (mostly hematologic). The authors concluded that the regimen was effective, but the substantial toxicity encountered has precluded widespread adoption of this regimen.12 Of note, the RTOG 9514 trial used a more intense version of MAID (employing a higher dose of cyclophosphamide) than was used in the preceding single institution study; it also employed very generous radiation fields, with 9 cm proximal and distal margins on tumors that were all >8 cm. Both of these factors were likely contributors to the observed toxicity.
The European Organization for Research and Treatment of Cancer (EORTC) undertook 2 important trials to assess the role of neoadjuvant chemotherapy with unresolved results. EORTC/Soft Tissue and Bone Sarcoma Group 62871 was a randomized phase II trial that enrolled 134 patients with STS >8 cm or grade 2 or 3. Patients were randomized to surgery alone or to surgery plus neoadjuvant doxorubicin/ifosfamide. Postoperative RT was given for marginal surgery, positive surgical margins, or LR. There were no differences in 5-year DFS (52% vs 56%) or OS (64% vs 65%), but the study was closed early due to poor accrual, and thus it was not sufficiently powered to detect a difference.13 In addition, EORTC 62961 randomized 341 patients with >5 cm, grade 2 or 3, deep and extracompartmental STS to receive neoadjuvant etoposide/ifosfamide/doxorubicin (EIA) versus neoadjuvant EIA plus hyperthermia. With respect to this study, hyperthermia resulted in improved median (3.8 yrs vs 2 yrs, P = .044) and median DFS (2.6 yrs vs 1.4 yrs, P =.040).14
The Sarcoma Meta-Analysis Collaboration published an analysis in 2008 that included 18 randomized trials with 1953 patients treated with (primarily) adjuvant or neoadjuvant chemotherapy.15 Findings included statistically significant absolute reductions of 4% for LR and 9% for distant recurrence, and an absolute improvement of 6% for survival attributable to chemotherapy. However, the fact that this update did not include the largest negative trial renders the results less conclusive.16Perhaps the most dramatic advance for the treatment of sarcomas over the past 2 decades comes from basic research in GISTs, which are the most common mesenchymal tumors in the gastrointestinal tract. Historically, these tumors had been treated by surgery alone, were refractory to systemic therapy, and were associated with a poor prognosis. In 1998, Hirota and colleagues reported that molecular analyses on GISTs identified constitutively active mutations in c-kit, which encodes a tyrosine kinase receptor.17 The tyrosine kinase inhibitor imatinib mesylate (Gleevec), which was developed to inhibit BCR-ABL in chronic myelogenous leukemia, is also a potent inhibitor of KIT.18 In a preclinical cell culture model, imatinib mesylate showed activity against GIST.19 Therefore, Demetri and colleagues performed a multi-institutional clinical trial using imatinib mesylate in patients with metastatic disease and observed an objective response in more than half of the patients.20 A randomized phase III double-blind, placebo-controlled trial of adjuvant imatinib mesylate after surgery, versus surgery alone, for GISTs that were at least 3 cm in size and positive for KIT protein expression showed improved RFS at 1 year (98% vs 83%, P <.0001).21
RTOG 0132/American College of Radiology Imaging Network 6665 was among the prospective studies to investigate the effectiveness of imatinib mesylate on GIST and to contribute data to the management guidelines for GIST. Different from other studies, this multi-institutional prospective phase II study was designed to assess the clinical outcomes and tolerability of imatinib (600 mg/day) given for 8 to 12 weeks either (i) as preoperative therapy prior to a planned resection of intermediate- to high-risk primary GIST, or (ii) as a cytoreductive agent prior to planned resection of metastatic and/or recurrent GIST. All patients planned to continue postoperative imatinib (600 mg/day) therapy for an extended postoperative period of 2 years. The short-term analysis of the patients enrolled into this clinical trial showed encouraging results of outcomes with a low rate of toxicity.22 The long-term analysis of results from this study suggests that a high percentage of patients developed progressive disease after discontinuation of 2-year maintenance imatinib therapy following surgery.23 Based on these and other studies, imatinib mesylate is now a standard treatment for many patients with GIST.Finkelstein and colleagues have explored combining radiation therapy and immunotherapy for sarcoma.24 His group determined the effect of combination of intratumoral administration of dendritic cells (DCs) and fractionated EBRT on tumor-specific immune responses in patients with STS. Targeted accrual was achieved with 17 patients with large (>5 cm) high-grade STS enrolled in the phase I/II study. Indeed, patients were treated in the neoadjuvant, preoperative setting with 50.4 Gy of EBRT, delivered in 28 fx, 5 days per week, combined with intratumoral injection of 107 DCs followed by complete surgical resection. DCs were injected on the second, third, and fourth Fridays of the treatment cycle to maximize time before re-exposure to radiation field. Detailed clinical outcome evaluation and immunological assessments were performed. With respect to outcomes, the treatment was found to be well tolerated. Of note, no patient had a tumor-specific immune response before combined EBRT/DC therapy; following therapy, 9 patients (52.9%) developed tumor-specific immune responses, which lasted from range of 11 to 42 weeks. With respect to clinical outcomes, 12 of 17 patients (70.6%) were progression-free after 1 year. Treatment caused a dramatic accumulation of thymus (T) cells in the tumor. The presence of CD4+ T cells in the tumor was positively correlated with tumor-specific immune responses that developed following combined therapy. Accumulation of myeloid-derived suppressor cells but not regulatory T cells was negatively correlated with the development of tumor-specific immune responses. Experiments with 111Indium labeled DCs demonstrated that these antigen-presenting cells need at least 48 hours to start migrating from the tumor site. Thus, these data suggest combination of intratumoral DC administration with EBRT was safe, and it resulted in induction of antitumor immune responses. These results are promising and have generated further testing to assess clinical efficacy more broadly.24In conclusion, conducting meaningful trials in sarcoma has been increasingly challenging due to funding constraints coupled with both the rarity of the disease and the large variety of sarcoma histologic subtypes. As such, the study of local management with surgery and/or RT is additionally challenging given the variety of anatomic sites where sarcomas occur. Indeed, each anatomic site is associated with specific and varied potential toxicities related to resection and RT. Despite these factors, the aforementioned work continues to advace the state of science for this important area. We must continue to be vigalent in a such low-incidence disease such as sarcoma as breakthrough work will pave the foundations for continued work in rare disease states.Steven Eric Finkelstein is Co-Chair, NRG Immunotherapy Committee and Dian Wang, MD, is Professor of Radiation Oncology and Urology, Department of Radiation Oncology, Rush University Medical CenterThe authors report no conflicts of interest.Steven Eric Finkelstein, MD, FACRO e-mail: firstname.lastname@example.org.