Study reference

Study characteristics

Patient characteristics 

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Bonvalot, 2020

Type of study: RCT (open label phase 3 trial)

Setting and country:

31 research institutions, hospitals, and cancer centres in Europe (France, Italy, UK, the Netherlands, Norway, Poland, Belgium, Denmark, Sweden, Spain, and Germany, in order of the number of inclusions), Canada, and the USA

Funding and conflicts of interest:

Role of the funding source:

EORTC had a role in the study design, data collection, data analysis, data interpretation, and writing of the report. Data were collected by investigators and associated site personnel, analysed by a statistician (SL) working in EORTC headquarters, and interpreted by members of the steering committee. Raw data are available from SL. The corresponding author had the final responsibility for the decision to submit for publication and had full access to all the data.

Declaration of interests:

SB reports personal fees and non-financial support from Nanobiotix and PharmaMar, and non-financial support from Pfizer, outside the submitted work. AG reports personal fees from Novartis, Pfizer, Bayer, Lilly Oncology, SpringWorks, and Nanobiotix, and grants and personal fees from PharmaMar, all outside the submitted work. CLP reports personal fees from AstraZeneca, Amgen, Nanobiotix, Roche, Medscape, PrimeOncology, and Lilly, outside the submitted work. PR reports personal fees from Novartis, Merck Sharp & Dohme, Bristol-Myers Squibb, Roche, Pfizer, Blueprint Medicines, Pierre Fabre, and Sanofi, outside the submitted work. PC reports personal fees from AbbVie and AstraZeneca, outside the submitted work. AM reports grants from National Health Service (NHS) funding to the National Institute for Health Research Biomedical Research Centre for Cancer at The Royal Marsden Hospital and The Institute of Cancer Research, during the conduct of the study. JYB reports grants from European Clinical Trials in Rare Sarcomas (EUROSARC), Lyon Integrative Cancer Research Program (LYRICAN), the European Network for Rare Adult Solid Cancers (EURACAN), NetSarc+, and Intersarc, during the conduct of the study. APDT reports personal fees from Roche, PharmaMar, and Bayer, outside the submitted work. All other authors declare no competing interests.

Inclusion criteria:

Eligible patients were aged 18 years or older with histologically documented, centrally reviewed, localised, primary soft tissue sarcoma of the retroperitoneal or intraperitoneal spaces of the pelvis. The tumour had to be unifocal; non-metastatic; not previously treated, not extending through the sciatic notch or across the diaphragm; and not originating from bone structure, abdominal, or gynecological viscera; and both operable and suitable for radiotherapy as per evaluation by an institutional multidisciplinary tumour board. A contrast-enhanced chest, abdomen, and pelvis CT scan or MRI scan was required within 28 days before randomisation, with radiologically measurable disease (as per Response Evaluation Criteria in Solid Tumors [RECIST] version 1.1). Patients were required to have a WHO performance status of 2 or lower; an American Society of Anesthesiologist (ASA) score of 2 or lower; and an absence of history of bowel obstruction, mesenteric ischaemia, or severe chronic inflammatory bowel disease. In addition, patients had to have norm al function (calculated creatinine clearance ≥50 mL/min and functional contralateral kidney), normal bone marrow and hepatic function (white blood cell count ≥2.5 × 10⁹ cells per L, platelet count ≥80 × 10⁹ cells per L, and total bilirubin <2 times upper limit of normal); cardiac function less than or equal to New York Heart Association class II; normal 12 lead electrocardiogram; a negative pregnancy test within 3 weeks before the first day of study treatment; adequate birth control measures; no relevant previous abdominal or pelvic radiation; no co-existing malignancy within the last 5 years, except for adequately treated basal cell carcinoma of the skin or carcinoma in the cervix; and no psychological, familial, sociological, or geographical conditions that could interfere with compliance with the study protocol.

Exclusion criteria:

Patients were ineligible if a macroscopically incomplete (R2) surgery was anticipated on the prerandomisation CT scan and if the tumour was one of the following histological subtypes: gastrointestinal stromal tumour, rhabdomyosarcoma, primitive neuroectodermal tumour or other small round blue cell sarcoma, osteosarcoma, chondrosarcoma, aggressive fibromatosis, or sarcomatoid or metastatic carcinoma.

N total at baseline:

Intervention: 133

Control: 133

Important prognostic factors:

age median (IQR)

I: 61 (52-68)

C: 61 (53-67)

Sex:

I: 53% M

C: 50% M

WHO performance status:

I: 0: 83%, 1: 17%, 2: <1%

C: 0: 75%, 1: 25%, 2: 0%

Groups comparable at baseline.

Intervention group: preoperative radiotherapy

followed by en-bloc curative-intent surgery:

Multivisceral en-bloc curative-intent surgery was done within 4–8 weeks from the end of radiotherapy in the radiotherapy plus surgery group.

In the radiotherapy plus surgery group, preoperative radiotherapy was delivered via a 3D conformal radiotherapy (3DCRT) or intensity modulated radiotherapy (IMRT) technique (including tomotherapy) done according to EORTC quality assurance in radiotherapy (as detailed in the protocol).

Radiotherapy was started within 8 weeks of randomisation in the same centre as surgery. The prescribed dose was 50.4 Gy in 28 once-daily fractions of 1.8 Gy, with five fractions per week during 5.5 weeks.

Control group: en-bloc curative-intent surgery alone:

Multivisceral en-bloc curative-intent surgery was done within 4 weeks of randomisation in the surgery alone group.

Length of follow-up:

Follow-up scans in both groups were planned at 24 weeks after randomisation and every 12 weeks subsequently during the first year, and then every 6 months until recurrence or death.

Loss-to-follow-up:

Intervention:

14 (10%)

Reasons: 8 had radiotherapy, but did not have surgery (1 withdrew consent, 3 had distant metastasis, 1 did not meet operability criteria, 1 had problem with anaesthesia, 2 died before surgery), 4 did not have radiotherapy but had surgery (3 patients refused radiotherapy, 1 radiotherapy planning not acceptable), 2 did not have radiotherapy and did not have surgery (1 withdrew consent for the study,

1 non-eligible tumour identified by central

pathology review)

Control:

5 (4%)

Reasons: 5 patients did not have surgery (2 distant metastasis, 1 did not meet operability criteria, 1 had problem with anaesthesia, 1 died before surgery)

Incomplete outcome data:

Intervention:

7 (6%)

Reasons not described.

Control:

4 (3%)

Reasons not described.

Overall survival

At 3 years % (95%CI)

I: 84.0% (76.3–89.4)

C: 84.6% (76.5–90.1)

At 5 years

I: 76.7% (66.9–84.0)

C: 79.4% (69.1–86.5)

and in the radiotherapy

plus surgery group

overall survival was 84.0% (76.3–89.4) at 3 years and 76.7% (66.9–84.0) at 5 years.

Median overall survival

was not reached in either group (95% CI not reached to not reached in both groups; HR 1.16, 95% CI 0.65–2.05.

Progression-free survival

Median abdominal recurrence-free survival:

I: 4.5 years (95% CI 3.9 to not estimable)

C: 5.0 years (3.4 to not estimable)

Hazard ratio: 1.01, 95% CI 0.71–1.44

Local recurrence

Not reported.

Quality of life

QLQ-C30 questionnaires at baseline, year 1, and year 5. Compliance was low and data were too sparse to allow any meaningful estimation of treatment differences, thus, results will not be reported.

Safety

Adverse events:

Common Terminology Criteria for Adverse Events (CTCAE) version 4.018.

Serious adverse events (not further specified)

I: 30 (24%) of 127

C: 13 (10%) of 128

RR = 2.33 (95%CI 1.27 to 4.25)

Mortality:

I: 1/127

C: 0/128

Author’s conclusion: This trial is negative, with similar abdominal recurrence-free survival and overall survival in both groups at 3 years of follow-up.

As a consequence, preoperative radiotherapy cannot be considered as the standard of care for retroperitoneal

sarcoma.

Yang, 1998  (long term follow-up: Beane, 2014)

Type of study: RCT

Setting and country:

Setting not reported, Bethesda, USA

Funding and conflicts of interest:

Not reported.

Inclusion criteria:

Patients with extremity soft tissue tumors of high or low grade: the study included both high and low grade tumors, data is only extracted for the subgroup of patients with low-grade tumors.

Exclusion criteria:

Patients with evidence of metastatic disease, a history of a second malignancy, or contraindications to receiving doxorubicin, cyclophosphamide, or XRT were excluded.

N total at baseline:

Intervention: 26

Control: 24

Important prognostic factors:

Age ± SD:

I: not reported

C: not reported

Sex:

I: 58% M

C: 71% M

Groups comparable at baseline: Demographic

characteristics in these two patient populations were evenly distributed.

Intervention group: surgery and adjuvant radiotherapy (XRT).

Surgery: Patients who presented with recent excision of their primary tumors were widely re-excised at the NCI, unless clear documentation was available to confirm the adequacy of the previous surgery. As a minimum, surgery had to result in the removal of all gross disease. In patients with a prior operation, definitive surgery was planned to entirely encompass the previous surgery, including all biopsy and drain sites. Wherever possible, a margin of 1 to 2 cm of normal tissue or an uninvolved fascial boundary was maintained around the tumor specimen. This  standard was compromised only if a limited positive (or close) surgical margin would spare the patient the disabilities resulting from resection of major nerves, vessels, or weight-bearing bone. Resections included periosteum or vessel adventitium in continuity where necessary. Patients with gross residual tumor or multiple, widely positive margins following maximum LSS were offered amputation and not included in the study.

Adjuvant XRT: randomized within 4 months of definitive resection.

Radiation consisted of 4,500 cGy to a wide field followed by an 1,800 cGy boost to the tumor bed (as defined by perimeter surgical clips). Care was taken to avoid circumferential limb irradiation and unnecessary irradiation of joints and tissues not at risk, through the use of filters, compensatory wedges, and electrons. One hundred eighty cGy fractions were given 5 days a week for a total of 6 to 7 weeks of therapy. Therapy was delayed for marked cutaneous reactions or wound complications.

Control group: surgery only.

Surgery: similar to intervention group, see description.

Length of follow-up:

Yang, 1998: All patients were followed up by clinical assessment and chest radiograph every 2 to 3 months for 2 years, 3 to 4 months for 2 more years, and 6 to 12 months at 4 years and beyond.

Beane 2014: a 20-year follow-up (update).

Loss-to-follow-up:

Yang, 1998: There was one protocol violation in which a patient refused XRT after randomization. She is included in all analyses according to randomization.

Beane, 2014:

Since the original publication 55 patients have died (39 %), 19 (13 %) have been lost to follow-up, and 76 (48 %) confirmed living. Of the patients confirmed living, 54 (71 %) completed telephone interviews. A total of 22 patients (29 %) did not complete the questionnaire because they were unwilling to participate or were unable to be contacted by telephone and thus excluded.

Overall survival

There have been four deaths from metastatic disease among patients with low-grade tumors (two in each treatment arm), with only one of these patients having a local recurrence.

Overall survival (Beane 2014): proportion surviving reported graphically until 30 years after randomization, P2 = 0.14.

Progression-free survival

Not reported.

Local recurrence

Local recurrence- free survival is reported graphically for a follow-up period of 12 years.

With a median follow-up of 9.9 years (range 1.4 to 12.4 years), eight patients randomized to not receive XRT have locally recurred, and one treated with XRT has locally recurred.

Quality of life

Functional Living Index-Cancer (FLIC) (154 = best score):

Baseline: 114/112

6 months: 118/125

12 months: 129/127

24 months: 125/130

36 months: 131/127

Independence in activities of daily living, modified Erdman scale: No scores reported, only described as no significant differences between patients in the two treatment arms.

Safety

Beane, 2014: wound complications, number of patients that required wound care or subsequent major surgical interventions:

I: 8/30 patients (27%; 95% CI 12 to 46%)

C: 5/24 patients (20%; 95% CI 7 to 42%) .

Author’s conclusion: In this prospective randomized trial, adjuvant postoperative external-beam radiotherapy was shown to result in a statistically significant reduction in LRs in patients with either high-grade or low-grade extremity tumors. Overall survival and nonlocal recurrences were nearly identical for patients receiving or not receiving radiation. (…) With different strategies yielding similar overall survival rates, recommendations for the use of XRT may rest primarily on quality-of-life issues and individual patient risk factors for LR. Although this study had too few local failures to identify risk factors for LR (other than lack of XRT), previous studies have suggested that previous recurrence and surgical margins have the greatest impact on local recurrence.

Author’s conclusion Beane, 2014:

In summary, the initial results of this study demonstrated

that adjuvant EBRT for extremity STS improves local control without a statistically significant improvement in overall survival. Although it is possible an OS benefit exists but was not detected due to limited power, this has remained true on long-term follow-up. Our recommendation has been that adjuvant EBRT be reserved for those with significant risk of local recurrence to avoid multiple surgeries and limb loss from such preventable recurrences. In our study some late limb-loss events occurred in patients who had undergone EBRT, and we maintain that its use for patients at low risk of recurrence should be selective.

Study reference

(first author, publication year)

Was the allocation sequence adequately generated?

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure

LOW

Some concerns

HIGH

Bonvalot, 2020

Definitely yes

Reason: Patients were randomly assigned (1:1) centrally, at the headquarters of the European Organisation for Research and Treatment of Cancer (EORTC), using an interactive web response system, to receive either en-bloc curative-intent surgery alone or preoperative radiotherapy followed by en-bloc curative-intent surgery. Randomisation was stratified by hospital and WHO performance status (0–1 vs 2) using a minimisation algorithm, and was not balanced by histological subtype.

Definitely yes

Reason: Central assignment at the headquarters.

Definitely no

Reason: open-label study. No masking of treatment assignments was possible because of the differences in treatment.

Probably yes

Reason: Loss to follow-up was relatively infrequent and similar across groups. All randomly assigned patients were included in the intention-to-treat analysis.

Probably yes

Reason: Relevant outcomes were reported. Quality of life data was not reported, but it was explained why (Compliance was low and data were too sparse to allow any meaningful estimation of treatment differences, thus, results will not be reported.)

Probably yes

Reason: No other problems noted.

LOW

Yang, 1998 (Beane, 2014)

Definitely yes

Reason:

A fixed block randomization with stratification for primary versus recurrent tumors, grade 1 versus aggressive benign lesions and positive versus negative surgical margins was used.

No information

about allocation concealment.

Definitely no

Reason: blinding not possible due to the type of intervention treatment (radiotherapy)

Probably yes

Reason: infrequent: There was one protocol violation

in which a patient refused XRT after randomization. She

is included in all analyses according to randomization.

Probably yes

Reason: all relevant outcomes from methods section are reported (no protocol available)

Probably yes

Reason: No other problems noted, but no information on funding and possible conflicts of interest.

Some concerns