Ann Surg Oncol (2020) 27:3672–3682

https://doi.org/10.1245/s10434-020-08872-1

CONTINUING EDUCATION– SARCOMA

The Landmark Series: Multimodality Treatment of Extremity

Sarcoma
Christina L. Roland, MD, MS1, Winan van Houdt, MD2, and Alessandro Gronchi, MD3
1
Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; 2Department of
Surgical Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands; 3Sarcoma Service, Department of
Surgery, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy

ABSTRACT Multimodality treatment of primary soft SURGERY AND RADIATION THERAPY

tissue sarcoma has evolved over the last 50 years, includ-
ing seminal studies in amputation versus limb-sparing Amputation Versus Limb-Sparing Resection
surgery, incorporation of radiation therapy (XRT), and the for Localized Extremity Soft Tissue Sarcoma
continuing controversy over the utilization and efficacy of
systemic chemotherapy. We review the landmark studies in Purpose and Rationale The infiltrative nature of STS
the multimodality management of primary extremity and results in local recurrence of 30–50% when simple excision
trunk soft tissue sarcoma. alone is performed for high-grade tumors. These high
recurrence rates led to the use of amputation or radical
compartment resections for primary treatment of STS,
Multimodality treatment of soft tissue sarcoma (STS), which resulted in a reduction of local recurrence to \ 5%,
consisting of [ 70 histologic subtypes,1,2 is the standard of although with significant long-term morbidity. However, it
care for patients with localized disease. Seminal works, was noted that, following complete resection via
including randomized studies evaluating limb-preservation, amputation, up to 40% of patients developed distant
radiation sequencing, and systemic chemotherapy in the metastatic disease primarily to the lungs, leading to the
localized setting, have helped to inform the current rec- introduction of wide local resection followed by radiation
ommendations. In this Landmark article, we review therapy (XRT) as an alternative option to obtain local
seminal studies in surgery, radiation therapy, and systemic control. This randomized trial was designed to compare
chemotherapy to highlight the critical studies informing local control following amputation versus limb-sparing
practice to date. surgery (LSS) ? XRT in patients with localized extremity
STS.3

Study Design and Endpoints Patients with extremity STS

treated at the National Cancer Institute (NCI) between
1975 and 1981 were randomized 2:1 to limb-sparing
surgery ? XRT versus amputation. Patients were stratified
based on primary tumor histology, tumor grade, time from
diagnosis, and location (proximal or distal in the
extremity). Amputation was performed at or above the
joint and included all muscle groups. Patients in the LSS
Ó Society of Surgical Oncology 2020 group underwent wide resection to include all gross tumor
First Received: 27 May 2020 with several centimeters of surrounding normal tissue.
Accepted: 4 July 2020; Clips were placed in the resection bed to guide
Published Online: 27 July 2020 postoperative radiation therapy at a dose of
A. Gronchi, MD 6000–7000 rads (Gy). Of note, the study also included a
e-mail: [email protected]
The Landmark Series: Multimodality Treatment of Extremity Sarcoma 3673

nested randomized trial of chemotherapy versus no Results: Primary and Secondary Endpoints Actuarial
chemotherapy. estimates of freedom from local recurrence for the entire
cohort at 60 months were 82% and 69% for the BRT and
Results: Primary and Secondary Endpoints Sixteen no BRT groups, respectively (p = 0.04). However, when
patients were randomized to amputation and 27 to low-grade and high-grade tumors were analyzed
LSS ? XRT. Although patients that underwent LSS were separately, patients with high-grade tumors were
more likely to develop local recurrence (0% vs. 15% significantly less likely to develop local recurrences in
p = 0.06), there was no difference in 5-year disease-free the BRT arm, whereas there was no impact of BRT on
survival (78% amputation vs. 71% limb-sparing, p = 0.75) local control in patients with low-grade tumors. Similar to
or overall survival (88% vs. 83%, respectively; p = 0.99). previous studies evaluating XRT for local control, there
were no differences in distant recurrence between the
Conclusions Limb-sparing surgery ? XRT is an treatment arms, with 5-year disease-specific survival rates
effective treatment for patients with soft tissue sarcomas of 84% in the BRT arm compared with 81% in the no BRT
of the extremities. arm.

Commentary Based on this important trial, limb-sparing Conclusions Adjuvant BRT improves local control after
surgery is the standard of care for patients with primary complete resection of soft tissue sarcomas, though the
extremity STS. Although amputation is associated with benefit appears to be limited to patients with high-grade
better local control, this does not translate into differences tumors.
in long-term survival, which is dictated by the development
of distant metastatic disease. Although primary amputation Commentary The study by Pisters et al. from MSKCC
is still necessary in \ 5% of patients with primary STS, the demonstrated the benefit of adjuvant BRT in patients with
43 patients in this trial who were willing to undergo localized extremity and trunk STS undergoing limb-
randomization to amputation or limb preservation should sparing surgery. Although BRT had theoretical
be really commended as they contributed to the advantages of decreasing treatment volume possibly
preservation of limbs of countless subsequent patients, leading to reduced long-term toxicity, the complexity of
which should not be understated. catheter placement, the required inpatient hospitalization
during XRT, and the development of sophisticated XRT
Adjuvant Brachytherapy in Soft Tissue Sarcoma modalities, such as intensity-modulated radiation therapy
(IMRT), have significantly reduced the use of BRT in
Purpose and Rationale Noting the toxicity associated current practice. However, this trial is significant for its
with postoperative XRT in the NCI trial, the brachytherapy ability to perform a randomized trial in the treatment of
(BRT) technique of XRT delivery was investigated by the patients with a rare disease.
group at Memorial Sloan-Kettering Cancer Center
(MSKCC) as a way to provide adjuvant XRT with Benefit of Adjuvant Radiation Therapy in the Treatment
smaller total treatment volumes.4 of Soft Tissue Sarcomas of the Extremity

Study Design and Endpoints Between 1982 and 1992, Purpose and Rationale Following the replacement of
164 patients with localized extremity and trunk STS were amputation with limb-sparing surgery followed by high-
randomized intraoperatively after complete surgical dose XRT as the standard of care for localized extremity
resection to BRT versus no BRT. Randomization was sarcoma, it was noted that high-dose XRT to the extremity
stratified by tumor grade, tumor size, tumor location was costly and resulted in significant functional
(superficial or deep), proximal or distal extremity site, and impairment of the limb and, in some cases, complications
primary versus locally recurrent disease. BRT catheters leading to amputation. Therefore, a randomized study of
were placed intraoperatively in the tumor bed, with a target surgery with or without postoperative XRT was designed to
region to be irradiated 2 cm circumferentially and catheters evaluate local recurrence, disease-free survival, and overall
implanted approximately 1 cm apart and fixed with sutures. survival in patients with low-grade and high-grade
Postoperative localization films were obtained, dosimetry extremity soft tissue sarcoma.5,6
was performed, and BRT to a dose of 42–45 Gy was
delivered over 4–6 days, starting no sooner than Study Design and Endpoints Between 1983 and 1991, 91
postoperative day 6. patients with high-grade extremity sarcoma and 50 patients
with low-grade STS were randomized to receive
postoperative XRT versus no XRT. All patients with
3674 C. L. Roland et al.

high-grade sarcoma received postoperative chemotherapy Preoperative Versus Postoperative Radiation Therapy
(doxorubicin and cyclophosphamide) given concurrently in the Treatment of Extremity and Trunk STS
with radiation. Concurrent prospective quality-of-life and
functional studies were performed to evaluate the effects of Purpose and Rationale Although the aim of combined
XRT. modality XRT and surgery is to obtain maximum control
while preserving function, differences were noted in
Results: Primary and Secondary Endpoints In the initial complications (short term and long term) between
publication by Yang et al.5 at a median follow-up of preoperative and postoperative XRT. Preoperative XRT
9.6 years, there were no local recurrences in patients with required a smaller field with lower doses but was thought to
high-grade sarcoma treated with XRT compared with nine have higher wound complications, whereas postoperative
local recurrences in patients randomized to not receive radiation was thought to have higher long-term toxicity.
XRT, with an estimated actuarial local failure rate at This multicenter randomized trial by the Canadian
10 years of 22% in the no XRT arm. Similarly, patients Sarcoma Group and NCI Canada was designed to
with low-grade sarcoma treated with adjuvant XRT were evaluate whether preoperative versus postoperative XRT
significantly less likely to recur than those treated with affected the rate of wound complications.11,12
surgery alone (p = 0.016). In a long-term follow-up study,6
at a median of 17.9 years, the incidence of local recurrence Study Design and Endpoints One-hundred ninety patients
was 25% in the surgery alone group compared with 1.4% in were randomized to preoperative (50 Gy in 25 fractions)
patients treated with postoperative XRT (p = 0.0001). versus postoperative (66 Gy in 33 fractions) XRT,
However, patients treated with XRT were more likely to stratified by maximum tumor dimension. The primary
have wound complications, clinically significant edema, endpoint was rate of wound complications within 120 days
and functional limb deficits long term. of surgery, defined as: secondary operation, invasive
wound management, readmission for wound care, or
Conclusions Postoperative XRT is highly effective in persistent deep packing. Secondary endpoints included
preventing local recurrences. However, there is significant metastatic failure, acute toxicity, and function/general
long-term toxicity that may be permanent. health utilizing the Musculoskeletal Tumor Society rating
scale, the Toronto extremity salvage score (TESS), and the
Commentary This trial (confirmed with the subsequent short form-36.
long-term follow-up) established the inclusion of XRT for
optimal local control in the multimodality treatment of Results: Primary and Secondary Endpoints At a follow-
patients undergoing limb-sparing surgery for extremity up of 3.3 years, 35% patients undergoing preoperative
STS. However, it was noted that there was significant XRT developed a wound complication compared with 17%
short-term and long-term toxicity associated with XRT, in the postoperative group (p = 0.01). Patients with larger
including edema and reduced limb function. In addition, tumors and those with tumors located in the lower
the overall risk of local-only recurrence was relatively low extremity were more likely to develop wound
(\ 25% for low- and high-grade tumors), suggesting that complications. However, 68% patients in the
patients with very low risk of local recurrence could be postoperative group had acute toxic skin effects
considered for omission of XRT if the risks outweigh the compared with 36% in the preoperative group
benefits. A longitudinal study by Pisters and the MD (p \ 0.0001). Although patients undergoing postoperative
Anderson Cancer Center (MDACC) sarcoma group XRT had greater function at 6 weeks after surgery, there
demonstrated a local recurrence risk of \ 10% for tumors was no difference between treatment groups at later time
\ 5 cm resected with negative margins,7 supporting the points. Further, in a follow-up study,12 patients in the
consideration of surgery alone for patients with small, postoperative arm had greater fibrosis, joint stiffness, and
superficial STS resected with widely negative margins. In edema, with lower function scores on all measures
addition, there is growing evidence that, in experienced compared with those in the preoperative arm. There was
centers, limb-sparing surgery alone can result in excellent no difference in local recurrence or distant failure between
local control regardless of the malignancy grade, size, or the groups.
depth of the tumor.8,9 However, histologic subtype should
be factored into the decision, highlighting the importance Conclusions Preoperative XRT is associated with greater
of regionalization of sarcoma care to high-volume centers, risk of wound complications than postoperative XRT, but
where the highest surgical skills and knowledge of the patients treated with postoperative XRT had greater long-
disease are available.10 term fibrosis, joint stiffness, and edema related to radiation
field size. There is no difference in local or distant
The Landmark Series: Multimodality Treatment of Extremity Sarcoma 3675

recurrence related to the timing of XRT. Choice of XRT evaluate the efficacy and toxicity of a modified mesna,
regimen for patients with STS should take into account the doxorubicin, ifosfamide, and dacarbazine (MAID) regimen
timing of surgery and radiotherapy, as well as the size and in a multiinstitutional setting (RTOG 9514).15,16
anatomical site of the tumor.
Study Design and Endpoints One-hundred ninety patients
Commentary The study by O’Sullivan et al. with high-grade primary or locally recurrent STS [ 8 cm
demonstrated equivalent oncologic outcomes between were included. Patients were to receive three cycles of the
preoperative and postoperative XRT. However, modified MAID chemotherapy interdigitated with 44 Gy of
preoperative XRT is associated with worse short-term preoperative XRT. Surgery was planned for 80 days after
outcomes, but postoperative XRT is associated with worse the initiation of the first cycle of chemotherapy.
long-term functional outcomes. The interpretation of these Postoperatively, three additional cycles of modified
data varies significantly by institution, whereby some MAID were given between 21 and 35 days following
sarcoma centers almost exclusively perform preoperative surgery. Endpoints included wound complications, clinical
XRT, whereas others almost exclusively perform response assessed by the response evaluation criteria in
postoperative XRT. The decision for treatment solid tumors (RECIST) criteria, and the logic response
sequencing is dependent upon institutional and patient according to the percentage of viable tumor and
factors. As a matter of fact, postoperative radiotherapy chemotherapy and radiation-associated toxicity.
(RT) is delivered at higher doses (usually 60–66 Gy in
30–33 fractions) and to a larger field size. As a result, long- Results: Primary and Secondary Endpoints Sixty-six
term edema and fibrosis are more frequent, but wound patients from 31 institutions were enrolled between 1997
complications are less frequent, and tumor tissue is better and 2000. Patients treated on this protocol were high risk,
assessable in terms of malignancy grade and other possible with a median tumor size of 15 cm and 80% grade 3
biological factors. On the contrary, preoperative RT can be tumors. Only 59% of patients received all six cycles of
delivered at lower doses (usually 50 Gy in 25 fractions— chemotherapy, mainly related to treatment toxicity. There
attempts to reduce further the dose in sensitive histologic were three deaths (5%) associated with treatment, two of
subtypes are presently underway) and to smaller fields size. whom developed acute myelogenous leukemia secondary
As a result, long-term edema and fibrosis are less frequent, to chemotherapy. Of all patients, 84% experienced grade 4
and the limb function is better. However, the wound toxicity, and 97% experienced grade 3 or higher toxicity.
complication rate is high (over 30%), although in recent At a median follow-up of 7.7 years in surviving patients,
years, thanks to the introduction of intensity modulated RT 22.2% of patients developed local failure, and distant
able to spare more normal tissue, the wound complication metastases were identified in 28.1% of patients.
rate has been reduced. In addition, one should consider the
expected quality of surgical margins, as there is growing Conclusions Although the toxicity is significant, the
evidence that preoperative XRT can offset the negative long-term outcomes were potentially better than might be
prognostic impact of a microscopically positive margin expected in such high-risk STS patients.
when the positive margin is planned in advance on a
critical structure.13,14 Commentary Combination chemoradiation is used in a
variety of solid tumors with improved efficacy over either
Combination Chemoradiation in the Treatment modality alone. RTOG 9514 utilized an aggressive
of Extremity and Trunk STS preoperative and postoperative chemotherapy regimen in
combination with preoperative XRT, in which up to 40% of
Purpose and Rationale Multimodality therapy including patients were not able to complete all cycles of therapy due
margin-negative surgical resection with adjuvant XRT to toxicity. The survival outcomes were favorable for this
results in local control in 80–95% of patients. However, high-risk group, but the significant toxicity must be
patients with high-grade tumors over 5 cm are at balanced with efficacy. It is worthy to mention that there
significant risk for distant failure, which increases with are some other experiences combining XRT with
tumor size. As the role of adjuvant chemotherapy remains anthracycline ? ifosfamide, showing a more favorable
controversial (see below), a pilot study by Spiro et al. was toxicity profile.17 In addition, these combinations seem of
performed evaluating an aggressive chemotherapy regimen particular benefit when a margin over a critical structure is
interdigitated with radiotherapy with impressive 5-year expected to be close/positive and function preservation is
local control, disease-free survival, and overall survival the goal.18 New combinations are currently being tested,
rates. Based on these data, the Radiation Therapy with histology-specific drugs (i.e., trabectedin and XRT in
Oncology Group (RTOG) initiated a phase II trial to
3676 C. L. Roland et al.

myxoid liposarcoma),19 different classes of drugs (tyrosine Results

kinase inhibitors and XRT in unselected soft tissue
sarcoma),20–22 immunotherapy and XRT in Primary Publication-1997 A total of 1568 patients
undifferentiated pleomorphic sarcoma or dedifferentiated were identified from 14 trials with doxorubicin-based
liposarcoma (NCT03092323),23 or devices (nanoparticles adjuvant chemotherapy.25 Median follow-up time was
and XRT in unselected soft tissue sarcoma).24 As we 9.4 years. Sarcomas of all sizes and locations were inclu-
continue to evaluate novel treatment regimens, these ded, and a minority of tumors were low grade. Out of these
studies serve as basis for combination multimodality trials, six were performed with doxorubicin monotherapy,
therapy. while the other trials used different combinations along
with doxorubicin. For overall survival, the hazard ratio
SYSTEMIC THERAPY (HR) for the chemotherapy arm was 0.89, but this did not
reach significance (p = 0.12). HRs for local recurrence-free
Sarcoma Metaanalysis Collaboration (SMAC) survival, distant recurrence-free survival, and overall
recurrence-free survival were 0.73, 0.70, and 0.75,
Purpose and Rationale Several randomized clinical trials respectively, all of which were significant (p \ 0.05). The
were conducted before 1997 to examine the efficacy of absolute risk reduction for distant metastases was 10%
adjuvant doxorubicin-based chemotherapy after complete (5–15%) in the chemotherapy group compared with pla-
resection of STS. However, individually, these trials failed cebo. None of the analyzed characteristics was significantly
to show a significant benefit of adjuvant chemotherapy, correlated with a beneficial effect of chemotherapy, how-
although there were some trends towards improved ever sarcoma in the extremity seemed to benefit more.
outcomes. Therefore, the Sarcoma Meta-analysis
Collaboration was initiated to assess whether a SMAC Update-2008 A total of 18 trials representing
metaanalysis of the individual data from these trials 1953 patients were included.26 All four new trials com-
would lead to more definitive evidence for the use of bined doxorubicin with ifosfamide. For overall survival,
adjuvant chemotherapy in localized STS.25,26 the odds radio (OR) for adjuvant doxorubicin-based
chemotherapy without ifosfamide was 0.84, which was not
Study Design and Endpoints A comprehensive literature statistically significant. However, the OR for overall sur-
search was performed to identify randomized controlled vival for doxorubicin ? ifosfamide was 0.56 (0.36–0.85)
trials (RCTs) of adjuvant chemotherapy after resection of in favor of chemotherapy. There were also benefits for
localized STS. The updated data from individual patients local recurrence (0.66) and distant recurrence (0.61) in
from all available randomized trials was collected, after favor of chemotherapy which were significant. The abso-
which a quantitative metaanalysis was performed. The lute risk reduction in terms of overall survival was 11%
primary endpoints were set at overall survival, recurrence- (3–19%) in the doxorubicin ? ifosfamide group when
free survival, and local and distant recurrence-free survival. compared with placebo.
Also, a potential beneficial effect of chemotherapy was
analyzed for several patient and tumor characteristics such Conclusions SMAC demonstrated that adjuvant
as age, sex, disease status at randomization, disease site, doxorubicin-based chemotherapy significantly improves
histology, grade, tumor size, extent of resection, and use of the time to local and distant recurrence. The addition of
radiotherapy. ifosfamide to doxorubicin-based chemotherapy is
An update was performed in 2008, in which four addi- associated with significant, but modest, improvements in
tional RCTs were identified.26 In the newer trials, overall survival.
doxorubicin was combined with ifosfamide, while in the 14
trials analyzed in the original SMAC, only 1 out of 14 Commentary Based on the original and updated SMAC,
combined doxorubicin with ifosfamide. Therefore, the adjuvant doxorubicin-based chemotherapy is associated
objective was to update the 1997 metaanalysis to assess the with improved local recurrence, distant recurrence, and
efficacy of doxorubicin-based chemotherapy both with and overall survival in combination with ifosfamide. When
without ifosfamide, as well as to assess differences considering adjuvant treatment for sarcoma, the
between doxorubicin regimens without ifosfamide and combination of doxorubicin and ifosfamide should be
doxorubicin with ifosfamide. considered standard of care. However, only a small
minority of patients will benefit from adjuvant
chemotherapy, and therefore balancing the benefits
against toxicity and carefully selecting patients remains
essential.
The Landmark Series: Multimodality Treatment of Extremity Sarcoma 3677

Adjuvant Doxorubicin and Ifosfamide Versus Conclusions This study showed no benefit of adjuvant
Observation for Intermediate- and High-Grade chemotherapy with doxorubicin and ifosfamide after
Sarcoma (EORTC 62931) resection of STS in relapse-free survival or overall
survival.
Purpose and Rationale This study started before SMAC
was published and aimed to address the unanswered Commentary This trial evaluating the effects of adjuvant
question of whether adjuvant doxorubicin and ifosfamide doxorubicin and ifosfamide failed to show a benefit in
would improve survival in intermediate- and high-grade overall survival. However, 44 (25%) patients in the
sarcoma. A previous European Organisation for Research chemotherapy arm did not receive chemotherapy or
and Treatment of Cancer (EORTC) study27 had shown no started too late to be included but were included in the
benefit of adjuvant chemotherapy on overall survival of intention-to-treat analysis. The data suggest that future
both low- and high-grade sarcoma of all histologies but did studies should focus on patients with larger, grade III, and
show a benefit on relapse-free survival. However, the extremity sarcomas, as the forest plots show a
chemotherapy regimen in this study was suboptimal, and nonsignificant advantage for adjuvant chemotherapy in
therefore EORTC 62931 was initiated to investigate the tumors with these characteristics. Interestingly, this study
effects of doxorubicin and ifosfamide on overall survival.28 included tumors of all sizes and all locations, as well as
both intermediate- and high-grade tumors. Also, the
Study Design and Endpoints This multicenter ifosfamide dose was relatively low, with 5 mg/m2, while
randomized trial from the EORTC was performed in 12 currently the most common dose is 9–12 mg/m2. All of this
European countries and Canada in 36 different sarcoma might have influenced the outcome of this study, and this
centers. Patients with completely resected intermediate- or study teaches us that a potential benefit for (neo-) adjuvant
high-grade (grade 2 or 3) primary or recurrent STSs at any chemotherapy will most likely only be found in high-risk
site and with any size were included. Patients were tumors. It is worthy to note that patients included in the
randomized between five cycles of adjuvant doxorubicin, study were affected by a disease with a median risk of
ifosfamide (5 g/m2), and lenograstim versus no adjuvant death in the 30% range. While this risk cannot be
chemotherapy. Patients in both groups received considered low, especially on a patient’s perspective, the
radiotherapy after a resection with margins \ 1 cm and proportional (not only the absolute) risk reduction of the
for recurrent sarcoma, but radiotherapy was started in the administration of chemotherapy looks lower when the
treatment group after finalizing adjuvant chemotherapy. baseline risk is lower.29 This limits the potential of
The primary endpoint was overall survival, and analyses detecting the efficacy of chemotherapy when the
were done by intention to treat. population selected in the trial is marked by an
intermediate risk. This study was recently revisited, and
Results: Primary and Secondary Endpoints Between patients were stratified by the predicted overall survival
1995 and 2003, 351 patients were randomly assigned to using a validated nomogram.30 When a cut-off of predicted
either adjuvant chemotherapy or no chemotherapy. risk of death of 40% was used, patients with a predicted
Adjuvant radiotherapy was equally given in both groups. overall survival inferior to 60% had a significant benefit in
Out of 163 patients starting chemotherapy, 130 patients disease-free survival and overall survival by the
(80%) completed all five cycles, and chemotherapy was in administration of adjuvant chemotherapy.31 Tools like
general well tolerated. There were no deaths due to toxic Sarculator to help select patients for future studies will
effects. In the control group, five patients did receive have to be considered.
chemotherapy due to patient’s choice, while in the
treatment group, 44 patients did not receive Three Cycles Preoperative Chemotherapy Versus Three
chemotherapy or received it too late and were not Cycles Preoperative Followed by Two Cycles
eligible for analysis. Neither relapse-free nor overall Postoperative Chemotherapy for High-Risk Extremity
survival differed significantly between both groups. The and Trunk Soft-Tissue Sarcoma (ETSTS)
5-year overall survival rate was 66.5% in the chemotherapy
group versus 67.8% in the control group. However, when Purpose and Rationale A previous Italian randomized
analyzing the effects of adjuvant chemotherapy for study in high-risk sarcoma patients had demonstrated an
different prognostic factors, higher grade, larger size, and improved overall survival after five cycles of adjuvant
limb sarcoma were more likely to benefit from adjuvant chemotherapy with full-dose epirubicin and ifosfamide
chemotherapy. (included in SMAC).32 After this trial, there were
suggestions that only the first three cycles of
chemotherapy would be sufficient for distant-disease
3678 C. L. Roland et al.

control, as the dose intensity of the latter two cycles had to complete the postoperative treatment cycles. However, a
dropped significantly.32 Therefore, the Italian and Spanish per protocol analysis confirmed the noninferiority of three
Sarcoma Groups performed an RCT comparing three cycles against five cycles. When the goal of preoperative
cycles to five cycles of chemotherapy, also shifting treatment is downsizing the tumor, more cycles can be
towards a neoadjuvant or combined neoadjuvant/adjuvant considered in individual cases after a good response.
approach. The hypothesis was that three cycles was However, there is no proven benefit on overall survival of
noninferior to five cycles of standard chemotherapy, administrating more than three cycles even in responding
leading to an improved cost–benefit ratio towards three patients as of yet. The fact that responders have a better
cycles.33 outcome than nonresponders suggest that treating patients
with neoadjuvant chemotherapy can provide insight into
Study Design and Endpoints This randomized controlled prognosis and tumor biology in individual patients. Finding
trial was conducted in 13 Italian and Spanish sarcoma good early markers predicting response is however
centers. Patients with a high-grade sarcoma [ 5 cm, essential to further personalize neoadjuvant chemotherapy.
deeply seated, and high grade [grade 3 according to the
Fédération Nationale des Centres de Lutte Contre le Cancer Histotype-Specific Neoadjuvant Chemotherapy Versus
(FNCLCC)] were randomized into two groups: patients in Standard Chemotherapy of Extremity and Trunk STS
arm A would receive three preoperative cycles of
epirubicin and ifosfamide, while patients in arm B were Purpose and Rationale As our understanding of the
to receive the same three preoperative cycles followed by importance of sarcoma histology has evolved, it was
two postoperative cycles (three vs. five cycles). There was recognized that previous trials were designed with the same
no preference for neoadjuvant or adjuvant radiotherapy, but standard chemotherapy regimens for all the different
this was left to the discretion of the individual centers. The histological sarcoma subtypes, which might not
primary endpoint was overall survival, and the trial was set necessarily reflect the differences in inherent tumor
up as a noninferiority study. biology between the sarcoma subtypes. Therefore, an
international, open-label randomized multicenter trial
Results: Primary and Secondary Endpoints Between comparing three cycles of standard chemotherapy with
2002 and 2007, 328 patients were randomized between three cycles of histology-tailored chemotherapy based on
the two groups. Ten-year overall survival was 61% for the histological sarcoma subtypes, as determined by expert
entire group, 64% in arm A and 59% in arm B. Three opinions, was performed. The hypothesis was that
cycles of chemotherapy was not inferior to five cycles neoadjuvant histotype-specific chemotherapy would result
according to the intention-to-treat analysis. In arm A, 14% in improved disease-free survival compared with standard
of all patients did not complete the chemotherapy cycles. In chemotherapy.34,35
arm B, 8% of all patients did not complete the first three
cycles, and 25% of the patients did not complete the Study Design and Endpoints This phase 3 randomized
postoperative chemotherapy cycles. Best responders were controlled trial was conducted in 4 European countries at
undifferentiated pleomorphic sarcoma (UPS), and worst 32 centers. Patients with localized STS [ 5 cm, deeply
responders were leiomyosarcoma (LMS). Response to seeded, and high grade (grade 3 FNCLCC) were included
therapy was confirmed to be associated with improved in one of the following five histological subtypes: myxoid
overall survival. liposarcoma (MLPS), undifferentiated pleomorphic
sarcoma (UPS), leiomyosarcoma (LMS), synovial
Conclusions After 10 years of follow-up, this study sarcoma (SS), and malignant peripheral nerve sheath
shows no difference in overall or recurrence-free survival tumors (MPNST). Patients were to receive either three
between three cycles of neoadjuvant epirubicin and cycles of epirubicin/ifosfamide or the following regimens
ifosfamide and three neoadjuvant cycles followed by two per subtype: trabectedin for MLPS, gemcitabine and
adjuvant cycles. docetaxel for UPS, gemcitabine and dacarbazine for
LMS, high-dose ifosfamide for SS, and etoposide with
Commentary When considering preoperative or ifosfamide for MPNST. The primary endpoint was disease-
perioperative chemotherapy for high-risk sarcoma, this free survival, and safety was a secondary endpoint.
study confirms that three cycles of preoperative
chemotherapy was noninferior to five cycles of Results: Primary and Secondary Endpoints A total of
perioperative chemotherapy. Of note, 25% of patients 287 patients were included between 2011 and 2016. The
scheduled to get postoperative chemotherapy were not able patients were diagnosed with MLPS (n = 64), UPS
(n = 97), LMS (n = 28), SS (n = 70), and MPNST
The Landmark Series: Multimodality Treatment of Extremity Sarcoma 3679

TABLE 1 Landmark surgery and radiation trials

Year Authors RCT arms or study group Major finding Impact on practice

19823 Rosenberg Amputation (n = 16) versus limb- No difference in disease-free survival. Limb-sparing surgery standard of care
et al. sparing resection ? postoperative Increased local recurrence in limb-
XRT (n = 27) sparing group (0 vs. 15%, p = 0.06)
19964 Pisters Adjuvant brachytherapy (n = 78) Decreased 5-year local recurrence in Brachytherapy improved local control
et al. versus no brachytherapy (n = 86) brachytherapy group (89%) after complete resection of high-
in patients with low- and high- compared with no brachytherapy grade STS
grade ETSTS (66%) in high-grade tumors
No difference in local recurrence for
low-grade tumors
19985 Yang et al. Postoperative adjuvant radiotherapy Decreased 10-year local recurrence Adjuvant radiotherapy standard of care
2010 6
Beane et al. (n = 70) versus surgery alone with radiotherapy (0 vs. 22%, for STS
(n = 71) p = 0.003)
200211 O’Sullivan Preoperative 50 Gy versus No difference in local recurrence Preoperative or postoperative XRT
200512 et al. postoperative 66 Gy external Increased acute wound complications acceptable as adjuvant therapy for
Davis et al. beam radiotherapy in preop XRT group (35% vs. 17%). localized STS. Increased acute
Increased late toxicity on toxicity with preop XRT, increased
postoperative XRT group (48.2% vs. late toxicity with postop XRT
31.5%)
200615 Kraybill Three cycles mesna, doxorubicin, 97% grade 3 toxicity at 7.7 years Combination chemoradiation feasible,
201016 et al. ifosfamide, and follow-up, 28% distant metastasis, but with high toxicity. Better
RTOG dacarbazine ? XRT favorable compared with historic toxocity profile in recent experiences
9514 preoperatively ? three cycles controls with doxorubicine-ifosfamide
postoperative chemotherapy. without dacarbazine combined to
XRT17

(n = 27). At the third futility analysis after a median of high-risk sarcomas is shifting from a set of fixed
follow-up of 12.3 months, patients treated with histotype- parameters to a more flexible prediction model for high risk
tailored chemotherapy had significant reduced disease-free with the developments of new predictor tools for outcome
survival and overall survival (4-year DFS and OS 38% and (Sarculator app), and patients in this trial treated with
64%, respectively) compared with those treated with standard chemotherapy had overall survival outcomes
standard chemotherapy (4-year DFS and OS 62% and better than the control arm of the first Italian Sarcoma
89%, respectively), and therefore the study was closed. Group (ISG) study32 and the control arm of the high-risk
Due to toxicity, 6% of patients in the standard arm and 4% subgroup analysis of the EORTC 62931 study.31 Also, the
in the histotype-specific arm interrupted treatment. On the opinions and experience with other more histotype-specific
final analysis, with a median follow-up of 52 months, the drugs are developing further, potentially leading to other
reduction in disease-free survival lost its significance (5- preferred systemic treatment regimens per subtype.
year DFS 47% against 55%, p = 0.323), while the
difference in overall survival stood (5-year OS 66% FUTURE DIRECTIONS FOR MULTIMODALITY
against 76%, p = 0.018). THERAPY FOR SOFT-TISSUE SARCOMA

Conclusions There was no benefit of histotype-tailored Margin-negative surgical resection remains the corner-
chemotherapy over standard chemotherapy for localized stone of the management of localized STS. The use of
ETSTS belonging to the five histologic subtypes included radiation therapy, especially in the preoperative setting, in
in the study. addition to surgery has led to increases in limb-sparing
procedures and low risk of local recurrence. Trials evalu-
Commentary Although histologic subtype is one (if not ating short-course XRT, combination of XRT with
the most) important determinant of outcomes for patients immunotherapeutics (i.e., immune checkpoint blockade
with STS, the use of histotype-tailored chemotherapy was and intralesional therapy), and other novel agents are
not superior over standard chemotherapy for localized ongoing, with the aim to provide improved systemic con-
ETSTS with currently available agents in the five trol in these high-risk patients. The use of systemic
histologic subtypes included in the study. However, care cytotoxic chemotherapy should be considered in high-risk
should be taken interpreting these data, since the definition patients, but histotype-specific approvals for metastatic
 3680

TABLE 2 Landmark chemotherapy trials and metaanalysis

Year Authors RCT arms or study group Major finding Impact on practice

199725 Sarcoma metaanalysis 1568 patients from 14 trials of 10% (5–15%) absolute risk reduction for distant in Adjuvant doxorubicin should be considered for
collaboration doxorubicin-based adjuvant chemotherapy group compared with placebo patients with localized ETSTS
chemotherapy
200826 Sarcoma metaanalysis Added four trials to SMAC-1998; 1953 11% (3–19%) absolute risk reduction for survival in Combination doxorubicin ? ifosfamide
collaboration— patients from 18 trials doxorubicin ? ifosfamide compared with placebo potentially more efficacious than doxorubicin
update alone
201228 Woll et al. Adjuvant doxorubicin ? ifosfamide (five No difference in overall survival in the overall Adjuvant chemotherapy should be considered in
EORTC 62931 cycles) versus observation group. A trend towards better RFS in large high-grade high-risk ETSTS
STS of the extremity in the 10% range, mirrored by a
subsequent unplanned analysis31 stratifying patient by the
predicted risk of death with a 20% absolute reduction of
the risk of death in high-risk ETSTS patients (predicted
OS \ 60%)30
201633 Gronchi et al. Preop epirubicin ? ifosfamide: three No difference in survival. 25% in pre/post group did not Consideration of short-course preop chemo
ISG-STS 0101 cycles preop versus three cycles complete postop chemo
preop ? two cycles postop (i.e., three
vs. five)
201734 Gronchi et al. Three cycles preoperative standard Reduced disease-free survival in histiotype-tailored No benefit of histiotype-tailored chemotherapy
202035 ISG-STS 1001 chemotherapy chemotherapy group (38%) versus standard over standard chemotherapy for localized
(epirubicin ? ifosfamide) versus chemotherapy (62%) ETSTS
histiotype-tailored chemotherapy
C. L. Roland et al.
The Landmark Series: Multimodality Treatment of Extremity Sarcoma 3681

disease and new agents, including the incorporation of body wall: Radiation Therapy Oncology Group Trial 9514. J Clin
immunotherapy and other targeted agents, hold promise for Oncol. 2006;24(4):619–625.
16. Kraybill WG, Harris J, Spiro IJ, et al. Long-term results of a
moving the field further and minimizing toxicity (Tables 1, phase 2 study of neoadjuvant chemotherapy and radiotherapy in
2). the management of high-risk, high-grade, soft tissue sarcomas of
the extremities and body wall: Radiation Therapy Oncology
DISCLOSURES C. L. Roland, W. van Houdt and A. Gronchi have Group Trial 9514. Cancer. 2010;116(19):4613–4621.
nothing to disclose. 17. Palassini E, Ferrari S, Verderio P, et al. Feasibility of preopera-
tive chemotherapy with or without radiation therapy in localized
soft tissue sarcomas of limbs and superficial trunk in the Italian
REFERENCES
Sarcoma Group/Grupo Español de Investigación en Sarcomas
randomized clinical trial: three versus five cycles of full-dose
1. IARC WHO Classification of Tumours-Soft Tissue and Bone epirubicin plus ifosfamide. J Clin Oncol.
Tumours. Vol 3. 5th ed. Lyon: IARC Press; 2020. 2015;33(31):3628–3634.
2. von Mehren M, Kane JM, III, Benjamin RS, et al. NCCN Clinical 18. Gronchi A, Verderio P, De Paoli A, et al. Quality of surgery and
Practice Guidelines in Oncology: Soft Tissue Sarcoma. In: 2019. neoadjuvant combined therapy in the ISG-GEIS trial on soft
3. Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft- tissue sarcomas of limbs and trunk wall. Ann Oncol.
tissue sarcomas of the extremities: prospective randomized 2013;24(3):817–823.
evaluations of (1) limb-sparing surgery plus radiation therapy 19. Gronchi A, Hindi N, Cruz J, et al. Trabectedin and RAdiotherapy
compared with amputation and (2) the role of adjuvant in Soft Tissue Sarcoma (TRASTS): Results of a phase I study in
chemotherapy. Ann Surg. 1982;196(3):305–315. myxoid liposarcoma from Spanish (GEIS), Italian (ISG), French
4. Pisters PW, Harrison LB, Leung DH, Woodruff JM, Casper ES, (FSG) Sarcoma Groups. EClinicalMedicine. 2019;9:35–43.
Brennan MF. Long-term results of a prospective randomized trial 20. Jakob J, Simeonova A, Kasper B, et al. Combined sunitinib and
of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol. radiation therapy for preoperative treatment of soft tissue sar-
1996;14(3):859–868. coma: results of a phase I trial of the German interdisciplinary
5. Yang JC, Chang AE, Baker AR, et al. Randomized prospective sarcoma group (GISG-03). Radiat Oncol. 2016;11:77.
study of the benefit of adjuvant radiation therapy in the treatment 21. Haas RL, Gelderblom H, Sleijfer S, et al. A phase I study on the
of soft tissue sarcomas of the extremity. J Clin Oncol. combination of neoadjuvant radiotherapy plus pazopanib in
1998;16(1):197–203. patients with locally advanced soft tissue sarcoma of the
6. Beane JD, Yang JC, White D, Steinberg SM, Rosenberg SA, extremities. Acta Oncol. 2015;54(8):1195–1201.
Rudloff U. Efficacy of adjuvant radiation therapy in the treatment 22. Canter RJ, Borys D, Olusanya A, et al. Phase I trial of neoad-
of soft tissue sarcoma of the extremity: 20-year follow-up of a juvant conformal radiotherapy plus sorafenib for patients with
randomized prospective trial. Ann Surg Oncol. locally advanced soft tissue sarcoma of the extremity. Ann Surg
2014;21(8):2484–2489. Oncol. 2014;21(5):1616–1623.
7. Pisters PW, Pollock RE, Lewis VO, et al. Long-term results of 23. Keung EZ, Lazar AJ, Torres KE, et al. Phase II study of
prospective trial of surgery alone with selective use of radiation neoadjuvant checkpoint blockade in patients with surgically
for patients with T1 extremity and trunk soft tissue sarcomas. Ann resectable undifferentiated pleomorphic sarcoma and dediffer-
Surg. 2007;246(4):675–681; discussion 681–672. entiated liposarcoma. BMC Cancer. 2018;18(1):913.
8. Fiore M, Ford S, Callegaro D, et al. Adequate local control in 24. Bonvalot S, Rutkowski PL, Thariat J, et al. NBTXR3, a first-in-
high-risk soft tissue sarcoma of the extremity treated with surgery class radioenhancer hafnium oxide nanoparticle, plus radiother-
alone at a reference centre: should radiotherapy still be a stan- apy versus radiotherapy alone in patients with locally advanced
dard? Ann Surg Oncol. 2018;25(6):1536–1543. soft-tissue sarcoma (Act.In.Sarc): a multicentre, phase 2–3, ran-
9. Gundle KR, Kafchinski L, Gupta S, et al. Analysis of margin domised, controlled trial. Lancet Oncol. 2019;20(8):1148–1159.
classification systems for assessing the risk of local recurrence 25. Adjuvant chemotherapy for localised resectable soft-tissue sar-
after soft tissue sarcoma resection. J Clin Oncol. coma of adults: meta-analysis of individual data. Lancet.
2018;36(7):704–709. 1997;350(9092):1647–1654.
10. Blay J-Y, Honore C, Stoeckle E, et al. Surgery in reference 26. Pervaiz N, Colterjohn N, Farrokhyar F, Tozer R, Figueredo A,
centers improves survival of sarcoma patients: a nationwide Ghert M. A systematic meta-analysis of randomized controlled
study. Ann Oncol. 2019;30(7):1143–1153. trials of adjuvant chemotherapy for localized resectable soft-tis-
11. O’Sullivan B, Davis AM, Turcotte R, et al. Preoperative versus sue sarcoma. Cancer. 2008;113(3):573–581.
postoperative radiotherapy in soft-tissue sarcoma of the limbs: a 27. Bramwell V, Rouesse J, Steward W, et al. Adjuvant CYVADIC
randomised trial. Lancet. 2002;359(9325):2235–2241. chemotherapy for adult soft tissue sarcoma–reduced local recur-
12. Davis AM, O’Sullivan B, Turcotte R, et al. Late radiation mor- rence but no improvement in survival: a study of the European
bidity following randomization to preoperative versus Organization for Research and Treatment of Cancer Soft Tissue
postoperative radiotherapy in extremity soft tissue sarcoma. Ra- and Bone Sarcoma Group. J Clin Oncol. 1994;12(6):1137–1149.
diother Oncol. 2005;75(1):48–53. 28. Woll PJ, Reichardt P, Le Cesne A, et al. Adjuvant chemotherapy
13. Dagan R, Indelicato DJ, McGee L, et al. The significance of a with doxorubicin, ifosfamide, and lenograstim for resected soft-
marginal excision after preoperative radiation therapy for soft tissue sarcoma (EORTC 62931): a multicentre randomised con-
tissue sarcoma of the extremity. Cancer. trolled trial. Lancet Oncol. 2012;13(10):1045–1054.
2012;118(12):3199–3207. 29. Callegaro D, Miceli R, Bonvalot S, et al. Impact of perioperative
14. O’Donnell PW, Griffin AM, Eward WC, et al. The effect of the chemotherapy and radiotherapy in patients with primary
setting of a positive surgical margin in soft tissue sarcoma. extremity soft tissue sarcoma: retrospective analysis across major
Cancer. 2014;120(18):2866–2875. histological subtypes and major reference centres. Eur J Cancer
15. Kraybill WG, Harris J, Spiro IJ, et al. Phase II study of neoad- (Oxford, England: 1990). 2018;105:19–27.
juvant chemotherapy and radiation therapy in the management of 30. Callegaro D, Miceli R, Bonvalot S, et al. Development and
high-risk, high-grade, soft tissue sarcomas of the extremities and external validation of two nomograms to predict overall survival
3682 C. L. Roland et al.

and occurrence of distant metastases in adults after surgical 34. Gronchi A, Ferrari S, Quagliuolo V, et al. Histotype-tailored
resection of localised soft-tissue sarcomas of the extremities: a neoadjuvant chemotherapy versus standard chemotherapy in
retrospective analysis. Lancet Oncol. 2016;17(5):671–680. patients with high-risk soft-tissue sarcomas (ISG-STS 1001): an
31. Pasquali S, Pizzamiglio S, Touati N, et al. The impact of international, open-label, randomised, controlled, phase 3, mul-
chemotherapy on survival of patients with extremity and trunk ticentre trial. Lancet Oncol. 2017;18(6):812–822.
wall soft tissue sarcoma: revisiting the results of the EORTC- 35. Gronchi A, Palmerini E, Quagliuolo V, et al. Neoadjuvant
STBSG 62931 randomised trial. Eur J Cancer (Oxford, England: chemotherapy in high-risk soft tissue sarcomas: final results of a
1990). 2019;109:51–60. randomized trial from Italian (ISG), Spanish (GEIS), French
32. Frustaci S, Gherlinzoni F, De Paoli A, et al. Adjuvant (FSG), and Polish (PSG) Sarcoma Groups. J Clin Oncol. 2020;
chemotherapy for adult soft tissue sarcomas of the extremities 38(19):2178–2186.
and girdles: results of the Italian randomized cooperative trial. J
Clin Oncol. 2001;19(5):1238–1247.
33. Gronchi A, Stacchiotti S, Verderio P, et al. Short, full-dose Publisher’s Note Springer Nature remains neutral with regard to
adjuvant chemotherapy (CT) in high-risk adult soft tissue sar- jurisdictional claims in published maps and institutional affiliations.
comas (STS): long-term follow-up of a randomized clinical trial
from the Italian Sarcoma Group and the Spanish Sarcoma Group.
Ann Oncol. 2016;27(12):2283–2288.