Since preoperative chemotherapy has been clearly shown to improve outcomes for patients with Ewing sarcoma, rhabdomyosarcoma, and osteosarcoma, practitioners have attempted to extend the use of adjuvant/neoadjuvant chemotherapy to other types of adult soft tissue sarcoma. Given the high risk of distant recurrence and disease-specific death for patients with soft tissue sarcoma tumors larger than 10 cm, these patients should be considered candidates for neoadjuvant chemotherapy as well as investigational therapies. Yet, potential toxicity from cytotoxic chemotherapy is substantial, and there remains little consensus and wide variation regarding the indications for use of chemotherapy in the adjuvant/neoadjuvant setting.
Key points
- •
More effective systemic therapy is a critical unmet need in the combined modality therapy of locally advanced soft tissue sarcoma.
- •
Meta-analyses of adjuvant/neoadjuvant clinical trials of chemotherapy for soft tissue sarcoma show a modest, but statistically significant, improvement in oncologic outcome favoring chemotherapy.
- •
Adjuvant/neoadjuvant chemotherapy for soft tissue sarcoma has not been widely adopted in large part because of the modest benefits and substantial risk of toxicity from intensive anthracycline-based regimens.
- •
Outcomes with adjuvant/neoadjuvant chemotherapy in soft tissue sarcoma vary by histologic subtype, and approaches based on tumor histology and individual patient and tumor factors are indicated.
Introduction
Soft tissue sarcomas (STS) are an uncommon and diverse group of tumors with mesenchymal differentiation, accounting for approximately 1% of US cancer diagnoses annually. As such, clinical behavior of these tumors can vary greatly, and robust prospectively obtained outcomes data are difficult to achieve. The rarity of incidence and diversity of disease biology also make consensus on treatment guidelines challenging, especially for key clinical questions in which the benefit/risk ratio for therapy may be narrow. As a general rule, the primary treatment modality for nonmetastatic STS in all locations and for most histologic types remains wide en bloc surgical resection, frequently in combination with radiotherapy (RT). Although distant disease control and effective systemic therapy remain critical unmet needs in the multimodality management of STS, the role of systemic therapy in the management of primary, nonmetastatic disease remains an area of significant debate.
Overall, local therapy, particularly for extremity and body wall/truncal STS, has proven highly effective. For the past 30 years, function-preserving/limb-sparing surgery combined with RT has successfully replaced amputation and other radical extirpative procedures in 90% to 95% of patients with STS. Long-term local control rates with these combined modality approaches exceed 85% to 90%. However, distant control and overall survival (OS) remain a challenge, particularly for patients with more aggressive disease (typically characterized by tumors larger than 10 cm, high-grade histology, and/or histologic subtypes with high risk of metastasis, such as synovial sarcoma, myxoid/round liposarcoma, and undifferentiated pleomorphic sarcoma). In fact, patients with high-grade tumors (typically American Joint Committee on Cancer [AJCC] stage III) have a risk of distant recurrence and death as high as 50% within 5 years of diagnosis. For these patients, neoadjuvant/adjuvant chemotherapy has frequently been advocated to improve metastasis-free survival and OS. However, the role of chemotherapy in the treatment of patients with STS amenable to complete surgical resection remains a controversial subject. The handful of prospective studies showing survival, local-recurrence, or distant-recurrence benefits to adjuvant/neoadjuvant chemotherapy in localized STS are mitigated by other studies demonstrating no benefit.
In this article, we focus on the role of chemotherapy in the definitive management of STS amenable to surgical resection with curative intent. Although current data are equivocal and the approach to adjuvant/neoadjuvant chemotherapy varies by institution, by specialty, and even by practitioner, there is increasing recognition that outcomes with adjuvant/neoadjuvant chemotherapy in STS vary by histologic subtype and that multimodality treatment approaches based on tumor histology and individual patient and tumor factors are indicated.
Introduction
Soft tissue sarcomas (STS) are an uncommon and diverse group of tumors with mesenchymal differentiation, accounting for approximately 1% of US cancer diagnoses annually. As such, clinical behavior of these tumors can vary greatly, and robust prospectively obtained outcomes data are difficult to achieve. The rarity of incidence and diversity of disease biology also make consensus on treatment guidelines challenging, especially for key clinical questions in which the benefit/risk ratio for therapy may be narrow. As a general rule, the primary treatment modality for nonmetastatic STS in all locations and for most histologic types remains wide en bloc surgical resection, frequently in combination with radiotherapy (RT). Although distant disease control and effective systemic therapy remain critical unmet needs in the multimodality management of STS, the role of systemic therapy in the management of primary, nonmetastatic disease remains an area of significant debate.
Overall, local therapy, particularly for extremity and body wall/truncal STS, has proven highly effective. For the past 30 years, function-preserving/limb-sparing surgery combined with RT has successfully replaced amputation and other radical extirpative procedures in 90% to 95% of patients with STS. Long-term local control rates with these combined modality approaches exceed 85% to 90%. However, distant control and overall survival (OS) remain a challenge, particularly for patients with more aggressive disease (typically characterized by tumors larger than 10 cm, high-grade histology, and/or histologic subtypes with high risk of metastasis, such as synovial sarcoma, myxoid/round liposarcoma, and undifferentiated pleomorphic sarcoma). In fact, patients with high-grade tumors (typically American Joint Committee on Cancer [AJCC] stage III) have a risk of distant recurrence and death as high as 50% within 5 years of diagnosis. For these patients, neoadjuvant/adjuvant chemotherapy has frequently been advocated to improve metastasis-free survival and OS. However, the role of chemotherapy in the treatment of patients with STS amenable to complete surgical resection remains a controversial subject. The handful of prospective studies showing survival, local-recurrence, or distant-recurrence benefits to adjuvant/neoadjuvant chemotherapy in localized STS are mitigated by other studies demonstrating no benefit.
In this article, we focus on the role of chemotherapy in the definitive management of STS amenable to surgical resection with curative intent. Although current data are equivocal and the approach to adjuvant/neoadjuvant chemotherapy varies by institution, by specialty, and even by practitioner, there is increasing recognition that outcomes with adjuvant/neoadjuvant chemotherapy in STS vary by histologic subtype and that multimodality treatment approaches based on tumor histology and individual patient and tumor factors are indicated.
Treatment approaches
Current National Comprehensive Cancer Network (NCCN) guidelines for the multimodality management of STS, particularly for locally advanced disease amenable to resection with curative intent, clearly endorse management by sarcoma specialists in a tertiary referral center. However, the specific recommendations regarding treatment sequencing and specific modalities (such as chemotherapy and RT) used in a combined modality approach are equivocal. Wide, but function-preserving, surgery in combination with RT remain the backbone of therapy in all possible treatment scenarios in which curative treatment is the goal. Moreover, NCCN guidelines acknowledge that it is acceptable to either include or omit chemotherapy from combined modality treatment, even in cases of locally advanced/stage III disease ( http://www.nccn.org/professionals/physician_gls/PDF/sarcoma.pdf ) despite the substantial risk of distant disease progression and death.
As a result, there is wide variation in the utilization of adjuvant/neoadjuvant chemotherapy in the management of primary STS. For example, an analysis by Wasif and colleagues showed that orthopedic oncologists and physicians with more than 75% of their clinical practice devoted to patients with sarcoma had the greatest preference for chemotherapy in the adjuvant/neoadjuvant management of STS. In contrast, surgical oncologists reported a statistically significant lower predilection for incorporating chemotherapy into the treatment plan for patients with locally advanced STS. Overall, the results of this survey study reinforced the impression that there is no clear consensus to the multimodality approach for localized STS, especially regarding the indications and implementation of adjuvant/neoadjuvant chemotherapy.
In a related analysis, Sherman and colleagues evaluated national practice patterns in the United States using the hospital-based National Cancer Database (NCDB), which has detailed demographic, pathologic, and treatment-related information on cancer-related diagnoses and outcomes. The investigators found that for the time period 2000 to 2009, neoadjuvant therapy with both RT and chemotherapy was increasing, whereas trimodality therapy was decreasing. Factors predictive of receipt of chemotherapy included younger age, synovial histology, high-grade histology, and positive surgical margins. Overall, however, chemotherapy was administered to only 31% of patients with stage III STS and 13.5% of patients with stage II STS. Taken together, these studies underscore the impression that the use of adjuvant/neoadjuvant chemotherapy has failed to gain widespread adoption in the United States in the primary management of nonmetastatic STS.
Biological questions in sarcoma chemotherapy
It is important to remember that the routine use of adjuvant/neoadjuvant chemotherapy has been unequivocally accepted into the combined modality therapy for certain types of sarcoma. For example, in pediatric bone and STSs, there is clearly an established role for adjuvant/neoadjuvant chemotherapy, and this approach has been shown to improve outcomes. In the early 1980s, the standard treatment of osteosarcoma was wide surgical resection, including amputation. However, more than 80% of patients with osteosarcoma developed metastatic disease, typically within 6 months of surgical intervention, in the setting of surgical monotherapy. The short disease-free interval routinely observed in these patients suggested that occult micro-metastasis was present at diagnosis, and the introduction of adjuvant/neoadjuvant chemotherapy for patients with osteosarcoma improved outcomes dramatically.
In fact, over the past 30 years, the 5-year survival of patients with malignant bone sarcoma has improved from approximately 16% without chemotherapy to approximately 70% with multiagent chemotherapy. These dramatic (and clearly unequivocal) results have provided proof-in-principle that chemotherapy can be effective for primary bone sarcomas, reinforcing the concept that subclinical metastatic disease is present at diagnosis, which can be effectively eradicated by systemic cytotoxic chemotherapy. Similarly, multiagent chemotherapy is a routine component of primary therapy for small round blue sarcomas, such as rhabdomyosarcoma and Ewing sarcoma. In fact, some practitioners consider chemotherapy to be primary therapy in these chemosensitive subtypes with surgical resection/local therapy serving as adjuvant therapy. Irrespective of these considerations, it is clear that chemotherapy has been established as a critical component of combined modality therapy in these histologies, and the role of chemotherapy has been endorsed by major oncology organizations, such as the NCCN and European Society for Medical Oncology.
However, it has also been observed that survival outcomes are significantly worse when adult patients (using variable age cutoffs between 18 and 40) are diagnosed with sarcoma histologies typically diagnosed in pediatric patients. For example, using surveillance, epidemiology, and end results (SEER) data, Mirabello and colleagues observed that 5-year OS rates for patients with osteosarcoma were 62% for patients younger than 24 years, 59% for patients 25 to 59 years, and 24% for patients 60 years and older. Similarly, a single-institution analysis by Koohbanani and colleagues demonstrated significant differences in survival of patients with Ewing sarcoma by age. Patients younger than 18 experienced a 5-year OS of 61%, whereas the 5-year OS for patients older than 40 was 38%. Several hypotheses have been offered to explain these age-related differences with chemotherapy in typically chemosensitive sarcomas. Some have hypothesized that these differences in oncologic outcome among pediatric and adult patients with similar bone and STS histologies are related to intrinsic differences in tumor biology between pediatric and adult patients with similar sarcoma histologies. On the other hand, others have attributed these differences in outcome between pediatric and adult patients to their respective ability to tolerate maximal cytotoxic chemotherapy, with greater dose intensity of chemotherapy in younger patients correlating with better outcomes.
Collins and colleagues performed a pooled analysis of individual data from multiple prospective neoadjuvant chemotherapy trials involving nearly 5000 patients with osteosarcoma. Consistent with prior studies assessing the impact of age on outcome, the investigators observed that younger patients had significantly better OS than older patients, including adults. However, using a landmark analysis analytical technique that incorporated chemotherapy-induced tumor necrosis into the model (an established biomarker of survival in patients with osteosarcoma receiving neoadjuvant chemotherapy), the investigators found that age was no longer a statistically significant predictor of OS. The investigators therefore concluded that the receipt of high-dose chemotherapy and the manner of its metabolism by the host was the key driver of outcome rather than differences in tumor biology among young and older patients. Despite these results, it is important to remember that STS subtypes in adult patients are different from those diagnosed in children, and although chemotherapy may be critical to outcomes in osteosarcoma and Ewing sarcoma, it may not be in adult STS histologies, such as liposarcoma, leiomyosarcoma, and high-grade undifferentiated pleomorphic sarcoma, in which tumor biology and intrinsic chemosensitivity may be different.
Clinical trials evaluating outcomes
Based on the results of chemotherapy in osteosarcoma and other pediatric sarcoma treatment paradigms, proponents of chemotherapy have advocated for a similar treatment approach in adult STS given the risk of distant metastasis and poor survival in the subgroup of patients with locally advanced disease. The risk of poor OS is particularly relevant for patients with high-grade primary STS larger than 5 cm in maximal dimension (AJCC stage IIb and III). These patients carry a risk of distant recurrence and death that approaches 50% at 5 years. This cohort of patients has traditionally been the subject of much controversy regarding the risks and benefits of adjuvant/neoadjuvant chemotherapy.
Approximately 20 randomized trials and 2 meta-analyses have been performed to evaluate the role of chemotherapy in patients undergoing surgery and RT with curative intent, and there are data both for and against the routine use of perioperative chemotherapy for patients with STS. Ultimately, despite the importance of this research question in the STS community, there remain fundamental unresolved questions regarding the role of chemotherapy in the management of primary disease, primarily because the rarity and diversity of these tumors, as well as differences in chemotherapy regimens and dosing, hinder an adequate patient accrual for definitive randomized trials.
Doxorubicin Monotherapy Trials
Studies analyzing the utility of adjuvant chemotherapy for adult STS were first performed in 1970s. In fact, from the 1970s to the 1990s, 14 randomized trials were performed comparing surgical resection (±RT) to surgical resection (±RT) plus chemotherapy. These studies primarily used doxorubicin as monotherapy in the adjuvant setting. Ultimately, these studies set the stage for subsequent decades of controversy, as there was a minority of studies showing survival, local-recurrence, and distant-recurrence benefits to adjuvant chemotherapy that were offset by other studies demonstrating no benefit. Further confounding the results of these studies were methodological and other flaws that left question marks over the validity of these data. For example, the use of doxorubicin monotherapy is considered a limitation by chemotherapy proponents, especially in the United States, as the combination of doxorubicin-ifosfamide is associated with greater response rates and is widely considered a more active regimen. In addition, these studies were confounded by a lack of standardization of surgical and RT techniques at the time, a lack of widespread adoption of cross-sectional imaging for accurate staging assessment, and a lack of uniform pathologic diagnostic criteria, among others. All of these issues raise legitimate questions regarding the generalizability and external validity of the study results.
Given these concerns, the role of adjuvant chemotherapy in STS was considered an important clinical problem that warranted further evaluation in a meta-analysis. Previous published results were equivocal, but these clinical trials were small and therefore potentially underpowered to detect moderate, but clinical meaningful, treatment effects. This equipoise lead to the Sarcoma Meta-Analysis Collaboration (SMAC), which was the first meta-analysis of pooled data involving 1568 patients with localized, resectable STS. Overall, the investigators of this meta-analysis observed statistically significant reductions in local recurrence-free survival (LRFS), distant recurrence-free survival (DRFS), and recurrence-free survival (RFS). These results corresponded to an absolute benefit at 10 years of 6% in LRFS, 10% in DRFS, and 10% in RFS. There was a 4% absolute benefit in OS with chemotherapy, but importantly, this result was not statistically significant (hazard ratio [HR] 0.89, 95% confidence interval [CI] 0.76–1.03). Skeptics of chemotherapy frequently point to this overall negative result for the primary endpoint of OS as the essential finding. Conversely, advocates of chemotherapy emphasize that a significant improvement in OS was observed in patients with extremity and truncal STS with a 7% absolute benefit at 10 years ( P = .029).
Despite the 10% improvement in RFS at 10 years with chemotherapy in this much cited meta-analysis, the lack of clear statistically significant improvements in OS undermined any conclusions that adjuvant chemotherapy was routinely indicated in STS, especially given the potential toxicity of doxorubicin-based regimens. Moreover, the statistical power of this meta-analysis was weakened by important methodological flaws, such as missing patient data, absence of central pathologic review of cases, and heterogeneity of the study population, including 5% low-grade tumors, 18% with tumor size smaller than 5 cm, and nonsarcoma diagnoses in approximately 10% of cases. These factors are cited as evidence undermining the case for adjuvant/neoadjuvant chemotherapy in primary STS. Finally, and significantly, a 2011 study of meta-analyses by Roseman and colleagues observed that possible conflicts of interest, including industry/pharmaceutical sponsor funding, were rarely disclosed in the research methods of meta-analyses. In their analysis, Roseman and colleagues provided evidence that these conflicts of interest are significant factors impacting the inclusion of studies, synthesis methods, and investigator assumptions about patient status and censoring strategies that affect the results of meta-analyses. In the SMAC meta-analysis, the authors of the meta-analysis were many of the same investigators from the individual randomized clinical trials evaluating chemotherapy for STS. In addition, the SMAC meta-analysis did not evaluate for the risk of bias across individual studies (as recommended by the PRISMA criteria) nor did they disclose any external sources of funding.
Combination Doxorubicin-Ifosfamide Trials
Following the adoption of ifosfamide for the treatment of sarcoma in the 1990s, subsequent studies of adjuvant and neoadjuvant therapy of primary STS focused on the combination of ifosfamide and an anthracycline. These have included both prospective, randomized trials and retrospective cohort analyses with more complete data and more homogeneity of the study groups. Although these studies have shown improved outcomes for patients with specific sarcoma histologies receiving chemotherapy, the results with doxorubicin/ifosfamide combination regimens have also been equivocal, and the risk/benefit ratio of adjuvant/neoadjuvant chemotherapy in these patients remains an unresolved question.
Studies have generally shown that regimens using doxorubicin and ifosfamide result in the greatest response rates (∼25%). Frustaci and colleagues assessed high-dose epirubicin and ifosfamide in a randomized phase II trial of 104 patients with STS undergoing radical surgery with or without RT. The investigators focused on extremity and truncal primary sites (given the SMAC results). Patients with both primary and locally recurrent disease were included, but patients with metastatic disease were excluded. The trial was further complicated by a stratification schema based on both primary versus locally recurrent disease as well as tumor size larger or smaller than 10 cm in maximal tumor dimension. Eight tumor histologies were included, but they were all high grade.
In this study, it was notable that a large majority of patients (42/46 or 91%) were able to complete all 5 cycles of chemotherapy, but aggressive supportive therapy, including hydration, mesna, and granulocyte colony-stimulating factor, was administered and most patients were younger than 55 (with no patients >65). Importantly, the rate of distant recurrence was markedly better in the chemotherapy arm at 2 years (28% recurrence with chemotherapy vs 45% with observation, P = .04). Similarly, OS at 4 years was improved in the chemotherapy group (69% chemotherapy vs 50% observation, P = .03). Although these results suggested important benefits to combination chemotherapy in the adjuvant setting, the patient stratification process in a small cohort created imbalances in the treatment groups, which may have contributed to differences in patient outcome. Furthermore, the results were also likely contingent on careful patient selection, which may not translate to other treatment scenarios.
A subsequent analysis of the same patient population with longer follow-up demonstrated that the benefits of adjuvant chemotherapy were not durable. With a median follow-up of 90 months, OS remained better in the chemotherapy cohort compared with the observation cohort (59% vs 43%), but this difference was no longer statistically significant ( P = .07). Although this result could be interpreted as type II error from an underpowered trial of a rare disease in which it is difficult to detect moderate but clinically meaningful treatment effects, another valid interpretation of the updated analysis of Frustaci and colleagues is that early benefits from chemotherapy may be lost with longer follow-up because chemotherapy given at the time of recurrence may be as beneficial to patient outcome as chemotherapy given in the adjuvant setting. This ongoing reservation regarding the benefits of early versus late chemotherapy was articulated in a retrospective analysis of the MD Anderson Cancer Center and Memorial Sloan Kettering Cancer Center combined experience with anthracycline-based perioperative chemotherapy for patients with large, deep, and high-grade extremity STS. This study demonstrated a time-varying effect associated with chemotherapy. During the first year, the HR of disease-specific death for patients treated with chemotherapy compared with those who were not was 0.37. However, following the first year, the HR of disease-specific death was 1.36, favoring the patients treated without chemotherapy. Although the findings in this study are potentially confounded by the effects of selection bias, an important implication of this article is that studies showing benefit to chemotherapy in patients with primary STS should be interpreted with caution because early benefits may be counteracted by later detrimental effects.
In a subsequent randomized trial, Petrioli and colleagues evaluated 88 patients with STS, of whom 45 were randomized to surgical resection plus chemotherapy (58% epirubicin alone and 42% epirubicin/ifosfamide) compared with 43 patients who were randomized to surgical resection alone. Despite the locally advanced disease present in this patient population, only 50% received RT. This trial was open from 1985 until 1996, but closed prematurely because of poor patient accrual. An analysis of the accrued patients revealed that OS was significantly improved in the chemotherapy group (5-year OS 72% with chemotherapy vs 47% with observation). However, despite the 25% absolute improvement in OS in the chemotherapy group, this difference approached, but did not reach, statistical significance ( P = .06) because of the sample size. As with the studies by Frustaci and colleagues, the results of the study by Petrioli and colleagues provided evidence both for and against adjuvant/neoadjuvant chemotherapy depending on one’s interpretation of the data.
Updated Meta-Analysis
After the publication of the 1997 SMAC meta-analysis, adjuvant chemotherapy was not routinely adopted. In addition, new systemic agents, specifically ifosfamide, were in use. Finally, the results of additional randomized controlled trials were available for analysis. Given this background, investigators at the University of Waterloo conducted an updated meta-analysis of the impact of adjuvant/neoadjuvant chemotherapy on outcomes in STS.
Overall, the results were similar to the prior meta-analysis. Statistically significant improvements in LRFS, DRFS, and RFS were observed in the chemotherapy group, corresponding to an absolute benefit at 10 years of 5% in LRFS, 9% in DRFS, and 10% in RFS. In this analysis, there was a 6% absolute benefit in OS with chemotherapy, which was statistically significant (HR 0.77, 95% CI 0.64–0.93). Although this updated meta-analysis confirmed the modest efficacy of adjuvant/neoadjuvant chemotherapy for STS and demonstrated incrementally superior effects with the addition of ifosfamide, the study was criticized for failing to include an important European Organisation for the Research and Treatment of Cancer (EORTC) trial of adjuvant chemotherapy that demonstrated no benefit in the chemotherapy arm.
Neoadjuvant Treatment
Although historically adjuvant chemotherapy has received greater evaluation in clinical trials and therefore has greater evidence on which to base treatment decisions, the results of survey studies demonstrate that neoadjuvant chemotherapy is typically favored in the management of locally advanced STS, perhaps because clinical response and downstaging can be followed prospectively and patients with rapidly progressive disease can be spared operation.
The only randomized trial of neoadjuvant chemotherapy for STS was conducted by the EORTC. This study enrolled 150 patients (analyzing 134) in a phase II trial, comparing 3 cycles of neoadjuvant doxorubicin (50 mg/m 2 per cycle) plus ifosfamide (5 g/m 2 per cycle) with surgical monotherapy. With more than 7 years of follow-up, 5-year OS was 65% with chemotherapy compared with 64% without chemotherapy ( P = .22). Although the phase II design of this study may be viewed as a sign of inadequate statistical power, many consider the results of this study to provide additional evidence against the routine use of adjuvant/neoadjuvant chemotherapy for STS. However, it is also important to acknowledge another frequent criticism of this study, namely that the doses of chemotherapy were low and considered suboptimal to provide a benefit, at least according to US standards.
Related posts:
Contemporary Management and Controversies of Sarcoma
Refinements in Sarcoma Classification in the Current 2013 World Health Organization Classification of Tumours of Soft Tissue and Bone
Sarcomas 2016
Sarcomas of the Breast with a Spotlight on Angiosarcoma and Cystosarcoma Phyllodes
Management of Sarcoma Metastases to the Lung
Myxofibrosarcoma
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
