Number of patients who were randomly assigned to a treatment group, and outcomes. NACT indicates neoadjuvant chemotherapy consisting of doxorubicin, ifosfamide, and etoposide; RHT, regional hyperthermia; STS, soft tissue sarcomas.
A, Median local progression free survival was 5.6 years (95% CI, 2.9-8.7) in the NACT plus RHT group compared with 2.4 years (95% CI, 1.7-4.2) in the NACT-alone group. B, Median disease-free survival was 2.8 years (95% CI, 2.0-4.9) in the NACT plus RHT group compared with 1.5 years (95% CI, 1.1-2.1) in the NACT-alone group. C, Median survival was 15.4 years (95% CI, 6.6 to >17.0 [the upper confidence limit cannot be estimated and represents the lower bound for the value to be expected]) in the NACT plus RHT group compared with 6.2 years (95% CI, 3.2-10.3) in the NACT-alone group. D, Extremity tumor–survival rates at 5 and 10 years were 75.2% and 68.3% in the NACT plus RHT group compared with 60.8% and 59.2% in the NACT-alone group. The absolute difference at 5 years was 14.4% (95% CI, 0%-29.5%) and was 9.1% (95% CI, 0%-24.7%) at 10 years. Nonextremity tumor–survival rates at 5 years and 10 years were 53.5% and 41.3% in the NACT plus RHT group compared with 44.0% and 29.9% in the NACT-alone group. The absolute difference at 5 years was 9.5% (95% CI, 0%-23.8%) and was 11.4% (95% CI, 0%-25.1%) at 10 years. NACT indicates neoadjuvant chemotherapy consisting of doxorubicin, ifosfamide, and etoposide; RHT, regional hyperthermia.
Analyses were univariate and not stratified according to subgroup.
eAppendix. Reasons patients were excluded from the study
eTable 1. Summary of efficacy measures in the Intention-to-Treat population
eTable 2. Distribution of radiotherapy between R0 and R1 resected patients
eTable 3. Effect of radiotherapy in different treatment groups (R0, R1, R0+R1) on local progression-free survival
Customize your JAMA Network experience by selecting one or more topics from the list below.
Issels RD, Lindner LH, Verweij J, et al. Effect of Neoadjuvant Chemotherapy Plus Regional Hyperthermia on Long-term Outcomes Among Patients With Localized High-Risk Soft Tissue SarcomaThe EORTC 62961-ESHO 95 Randomized Clinical Trial. JAMA Oncol. Published online February 15, 2018. doi:10.1001/jamaoncol.2017.4996
Does the previously reported improvement in local progression-free survival with neoadjuvant chemotherapy plus regional hyperthermia translate into improved survival of patients with high-risk soft tissue sarcoma?
In this randomized clinical trial that included 329 eligible patients, survival was significantly improved by adding regional hyperthermia to neoadjuvant chemotherapy with an absolute difference at 5 years of 11.4% and at 10 years of 9.9% compared with neoadjuvant chemotherapy alone.
For patients with localized high-risk soft tissue sarcoma who are candidates for neoadjuvant treatment, adding regional hyperthermia may be warranted.
Patients with soft tissue sarcoma are at risk for local recurrence and distant metastases despite optimal local treatment. Preoperative anthracycline plus ifosfamide chemotherapy improves outcome in common histological subtypes.
To analyze whether the previously reported improvement in local progression-free survival by adding regional hyperthermia to neoadjuvant chemotherapy translates into improved survival.
Design, Setting, and Participants
Open-label, phase 3 randomized clinical trial to evaluate the efficacy and toxic effects of neoadjuvant chemotherapy plus regional hyperthermia. Adult patients (age ≥18 years) with localized soft tissue sarcoma (tumor ≥5 cm, French Federation Nationale des Centers de Lutte Contre le Cancer [FNCLCC] grade 2 or 3, deep) were accrued across 9 centers (6, Germany; 1, Norway; 1, Austria; 1, United States) from July 1997 to November 2006. Follow-up ended December 2014.
After stratification for tumor presentation and site, patients were randomly assigned to either neoadjuvant chemotherapy consisting of doxorubicin, ifosfamide, and etoposide alone, or combined with regional hyperthermia.
Main Outcomes and Measures
The primary end point was local progression-free survival. Secondary end points included treatment safety and survival, with survival defined from date of randomization to death due to disease or treatment. Patients lost to follow-up were censored at the date of their last follow-up.
A total of 341 patients were randomized, and 329 (median [range] age, 51 [18-70] years; 147 women, 182 men) were eligible for the intention-to-treat analysis. By December 2014, 220 patients (67%; 95% CI, 62%-72%) had experienced disease relapse, and 188 (57%; 95% CI, 52%-62%) had died. Median follow-up was 11.3 years. Compared with neoadjuvant chemotherapy alone, adding regional hyperthermia improved local progression-free survival (hazard ratio [HR], 0.65; 95% CI, 0.49-0.86; P = .002). Patients randomized to chemotherapy plus hyperthermia had prolonged survival rates compared with those randomized to neoadjuvant chemotherapy alone (HR, 0.73; 95% CI, 0.54-0.98; P = .04) with 5-year survival of 62.7% (95% CI, 55.2%-70.1%) vs 51.3% (95% CI, 43.7%-59.0%), respectively, and 10-year survival of 52.6% (95% CI, 44.7%-60.6%) vs 42.7% (95% CI, 35.0%-50.4%).
Conclusions and Relevance
Among patients with localized high-risk soft tissue sarcoma the addition of regional hyperthermia to neoadjuvant chemotherapy resulted in increased survival, as well as local progression-free survival. For patients who are candidates for neoadjuvant treatment, adding regional hyperthermia may be warranted.
clinicaltrials.gov Identifier: NCT00003052
Soft tissue sarcoma accounts for less than 1% of all malignancies. According to the American Cancer Society, about 12 000 new cases per year are diagnosed in the United States, and more than 4900 people die of these tumors annually.1 Tumor size, grade, and location are the predominant prognostic factors used to define patients at high risk for local recurrence or early dissemination.2 To account for prognostic differences, site-specific nomograms have been developed for both extremity and retroperitoneal tumors.3,4 For localized tumors, surgery combined with preoperative or postoperative radiotherapy is considered the backbone of care. Regarding perioperative chemotherapy, current clinical practice guidelines recommend it as an option in patients deemed high risk.5
Heat exposure (40 °C to 43 °C) of cancer cells in preclinical studies, and hyperthermia regionally applied to patients in early randomized clinical studies, have shown synergistic activity with ionizing radiation and chemotherapy.6 For the combination of hyperthermia with chemotherapy, the study group at Munich7 was the first to demonstrate the safety and efficacy of regional hyperthermia (RHT) in patients with high-risk sarcoma. As a consequence, this study—the EORTC 62961-ESHO 95—was designed as the first randomized study that we know of to compare RHT added to neoadjuvant chemotherapy with neoadjuvant chemotherapy alone in patients undergoing surgery followed by radiotherapy whenever possible. Results for the primary end point of local progression-free survival and second end points including tumor response, survival outcome, and adverse effects accompanying therapy have been published previously.8 Herein, we present the final, long-term results with a cutoff date of December 2014.
The study details have been reported previously.8 Briefly, eligible patients were ages 18 to 70 years and had histologically proven soft tissue sarcoma with the following risk criteria: tumor diameter 5 cm or larger, French Federation Nationale des Centers de Lutte Contre le Cancer (FNCLCC) grade 2 or 3, deep to the fascia, and no evidence of distant metastases. In patients who had undergone an attempt of prior surgical resection with the result of marginal margins (tumor-free margins less than 1 cm) random allocation to treatment was allowed within 8 weeks of surgery.
The trial protocol is available in Supplement 1.
EORTC 62961-ESHO 95 was a multicenter, open-label, parallel group study8 with centralized randomization to either an experimental treatment group (neoadjuvant chemotherapy plus RHT) or a control group (neoadjuvant chemotherapy alone), with a similar follow-up schedule, stratified according to site and presentation of tumor.
The trial was initiated by the European Society of Hyperthermia Oncology (ESHO), with trial coordination carried out by the Klinikum der Universität München, Munich, Germany in collaboration with the European Organisation for Research and Treatment of Cancer (EORTC) Soft-Tissue Bone Sarcoma Group (STBSG). The participating university centers were in Germany (6), Norway (1), Austria (1), and the United States (1). The study protocol was approved by the EORTC in May 1997 and by review boards of each study site. Written informed consent for all patients was obtained. External pathological review was performed by one of us (S.D.) on behalf of EORTC.
The primary objective was local progression-free survival. Among secondary end points, tumor response to induction therapy, disease-free survival, and survival were included. Tumor response was based on investigator assessment by imaging using World Health Organization (WHO) criteria for patients with measurable disease at baseline. According to the STBSG recommendation at the time of the study, a blinded review of responses was performed by board members of the STBSG. Survival was defined as the time to death due to sarcoma or its treatment with survivors being censored at the time of last follow-up. Deaths from other causes were not considered events and censored at the time of death. Patients alive without recurrence were censored on the date of last follow-up. Adverse events related to chemotherapy were graded according to Common Toxicity Criteria (CTC) of the National Cancer Institute. Toxic effects related to hyperthermia were scored according to protocol guidelines.
Patients were randomized in a 1:1 ratio to both treatment arms. Block randomization was performed centrally at the EORTC data center with stratification according to site (extremity vs nonextremity) and presentation of tumor (primary vs recurrent vs prior surgery).
Patients were to receive either 4 cycles of chemotherapy alone (neoadjuvant chemotherapy consisting of doxorubicin, ifosfamide, and etoposide [NACT] alone) or chemotherapy combined with RHT every 3 weeks as induction therapy followed by evaluation of tumor response. Tumor assessments included abdominal computed tomography or magnetic resonance imaging and chest radiography. Local treatment consisted of definitive surgery within 4 to 6 weeks of induction therapy, including re-resections for patients with initial inadequate surgery. For external beam radiation therapy, the dose was administered to 50.0 to 60.0 Gy (to convert Gy to rad, multiply by 100), with daily fractions of 1.8 to 2.2 Gy, and a boost up to 66.0 Gy. Within 6 weeks of local therapy, patients were to receive another 4 cycles of their allocated treatment for postinduction therapy. Patients with previous surgery had to receive the complete induction and postinduction therapy. NACT consisted of doxorubicin (50 mg/m2 over a 60-minute period on day 1), ifosfamide (1500 mg/m2 on days 1 to 4), and etoposide (125 mg/m2 on days 1 and 4). Treatment continued unless progressive disease, unacceptable toxic effects, or withdrawal from the study occurred. Regional hyperthermia (42 °C for a 60-minute period) was given concurrently with ifosfamide on day 1 and day 4 of each cycle during both induction and postinduction therapy. Quality of hyperthermia was ensured by European Society for Hyperthermic Oncology guidelines.8,9
The accrual goal of 334 eligible patients was based on a statistical power of 80% to detect, on a 5% significance level, an improvement in local progression-free survival (median, 86 months for NACT plus RHT vs 43 months for NACT alone). An accrual period of 6 years and a follow-up time of 9 years were set. As defined in the study protocol, the final analysis required 146 distal failures. The analysis was undertaken using SAS version 9.2 (SAS Institute Inc). Survival of patients was estimated according to the Kaplan-Meier method, providing medians with 95% CIs and survival differences at specific time points. Number-needed-to-treat analysis was performed by standard procedure. Comparisons between the groups of stratified patients were performed using the log-rank test. The stratified proportional hazard Cox model was used for multivariate analysis. The subgroup effects were represented by a forest plot using the Cochrane Review Manager software version 5.3 (Cochrane Community). All P values are 2-sided and of exploratory nature except for the primary analysis. Results were considered significant at P ≤ .05. The survival-type analyses presented were based on the intention-to-treat population, which includes all eligible patients in the study who started their allocated treatment.
Between July 1997 and November 2006, a total of 341 patients were enrolled and underwent randomization over a 9-year period. Of these, 169 patients were assigned to the NACT plus RHT group and 172 to NACT-alone group. A total of 162 patients from the NACT plus RHT group and 167 patients from the NACT-alone group were eligible for the intention-to-treat analyses of survival end points. Seven patients of the NACT plus RHT group were excluded (6 withdrew consent and 1 had metastatic disease), and 5 patients of the NACT-alone group were excluded (4 withdrew of consent and 1 had metastatic disease) (eAppendix in Supplement 2). The major baseline characteristics of eligible patients were well-balanced across study groups (Table 1). The number of patients who received study treatment and outcomes of surgery is summarized in Figure 1. For local therapy, nearly all patients underwent surgery. About two-thirds of patients in both treatment arms underwent postoperative external beam radiotherapy; the mean (SD) doses were 53.2 (8.9) Gy vs 52.7 (9.6) Gy.
The database was closed in December 2014, when 220 disease relapses including 149 distant events had occurred in 329 patients: 101 disease relapses (62%; 95% CI, 55%-69%) in the NACT plus RHT group and 119 disease relapses (71%; 95% CI, 64%-78%) in the NACT-alone group with no failures in 109 patients (61 in the NACT plus RHT group [38%; 95% CI, 31%-45%] vs 48 in the NACT-alone group [29%; 95% CI, 22%-36%]). The median (interquartile range) follow-up duration was 11.3 (9.2-14.7) years.
The relative hazard for local progression or death between patients receiving NACT plus RHT or NACT alone was 0.65 (95% CI, 0.49-0.86; P = .002) with a median duration of 67.3 months vs 29.2 months (Figure 2A). The addition of RHT prolonged the median disease-free survival from 17.4 months to 33.3 months (HR for local or distant failure or death, 0.71; 95% CI, 0.55-0.93; P = .01; Figure 2B).
By December 2014, 188 patients (57%; 95% CI, 52%-62%) had died, and 141 patients were still alive (75 in the NACT plus RHT group and 66 in the NACT-alone group). One-hundred seventy four patients had died due to disease or treatment (77 in the NACT plus RHT group and 97 in the NACT-alone group); 5 deaths (3.1%) were attributable to treatment in the NACT plus RHT treatment group, and 2 deaths (1.2%) to treatment in the NACT-alone group. Fourteen patients had died from other causes (4, myocardial infarction; 7, second malignancy; 1, drug abuse; and 2, other reasons), of which 10 (6.2%) occurred in the NACT plus RHT group and 4 (2.4%) in the NACT-alone group.
Survival between the study groups was significantly improved in the NACT plus RHT group, with a median duration of 15.4 years compared with 6.2 years in the NACT-alone group (HR 0.73; 95% CI, 0.54-0.98; P = .04; Figure 2C). Survival rates at 5-years and 10 years were 62.7% (95% CI, 55.2%-70.1%) and 52.6% (95% CI, 44.7%-60.6%), respectively, in the NACT plus RHT group, and 51.3% (95% CI, 43.7%-59.0%) and 42.7% (95% CI, 35.0%-50.4%), respectively, in the NACT-alone group. The number of patients needed to treat to achieve the survival benefit at 5 years and 10 years were 8.8 and 10.1, respectively. By post hoc analyses, in patients with extremity tumors survival rates at 5 years and 10 years in favor of RHT were 75.2% vs 60.8% (absolute difference, 14.4%; 95% CI, 0.0%-29.5%), and 68.3% vs 59.2% (absolute difference, 9.1%; 95% CI, 0%-24.7%), respectively. In patients with nonextremity survival rates at 5 years and 10 years in favor of RHT were 53.5% vs 44% (absolute difference, 9.5%; 95% CI, 0%-23.8%) and 41.3% vs 29.9% (absolute difference, 11.4%; 95% CI 0%-25.1%), respectively (Figure 2D). The summary of treatment outcomes is provided in eTable 1 in Supplement 2.
A consistently higher survival was observed with the NACT plus RHT treatment across all subgroup factors (age, site, disease status, definitive/re-resection, R0, R1, R2, amputation, prior surgery, no resection, radiotherapy, size, grade, and histologic subtype), with no major treatment and subgroup interaction (Figure 3). The univariate and multivariate analyses showed that beside treatment, grade and tumor size remain the dominant prognostic factors in terms of survival (Table 2).
Considering the effect of further salvage treatment, the survival from local progression to the time of death (HR, 1.02; 95% CI, 0.69-1.52; P = .90) or from distant metastasis to the time of death (HR, 1.06; 95% CI, 0.74-1.50; P = .77), comparing both treatment groups, showed no statistical difference.
That we know of, EORTC 62961-ESHO 95 was the first phase 3 randomized trial in soft tissue sarcoma research that investigated the effects of neoadjuvant chemotherapy combined with RHT.
The main result was that with a median follow-up of more than 11 years neoadjuvant chemotherapy combined with RHT lead to a 27% improvement in survival, with a statistically significant absolute 11.4% improvement in the 5-year survival rate (62.7% vs 51.3%) and a 9.9% improvement in the 10-year survival rate (52.6% vs 42.7%), compared with neoadjuvant chemotherapy alone. The treatment effect was robust and consistent among all prespecified risk factors and stratification criteria. Owing to the fact that our study comprises a 20-year data set that included an older age group between 41 to 70 years that represented more than 70% of the patients, there was an increasing risk of death from natural causes unrelated to sarcoma. Therefore, the survival benefit has been analyzed as death of disease or its treatment so to be not confounded by the occurrence of disease-unrelated deaths.10 In extremity and nonextremity tumors, the hazard for death or its treatment was equally pronounced, but the study was not powered for these subgroups. Because of the larger subgroup of nonextremity tumors, the survival effect is most likely driven by downsizing and prevention of early progression of these tumors because local failure is the leading cause of death in patients with abdominal and/or retroperitoneal tumors.4 The positive impact of RHT of completely resected tumors in this subgroup has been previously reported.11
A puzzling observation in the study was the delayed divergence of the survival curves after treatment completion (Figure 2C). The same observation was made recently in 2 other randomized studies12,13 of soft tissue sarcoma testing eribulin as second-line therapy and olaratumab as first-line therapy. The delayed improvement of survival was discussed to be related to effects of further salvage therapies which seemed not to be the case in our study. Similar to these multitargeting agents, RHT also affects different targets encompassing DNA repair, microenvironment, and immunity.14-16 Our results fit to the early action–late benefit model of immunotherapy trials, where the therapeutic effects are exerted prior to the curve divergence. The survival curves will not separate until the time when corresponding control patients (who did not receive RHT) experience disease relapse and die.17
The multidisciplinary approach included the best possible local treatment. Surgery as the backbone of care was performed in almost all patients. Postoperative external beam radiotherapy was equally limited in one-third of patients owing to the risk of functional restrictions or adjacent organs at risk. The number of patients who received radiotherapy with R0 or R1 resected tumors were well-balanced. For local progression-free survival, radiotherapy after R0 resection had no effect, whereas after R1 resection the positive effect seen in both treatment arms was comparable (eTable 2 and eTable 3 in Supplement 2). Today, more advanced techniques involving image-guided radiotherapy may improve both tolerance and effectiveness.18,19 Results of using preoperative or postoperative external beam radiotherapy in the neoadjuvant setting from nonrandomized studies in extremity tumors, as well as results expected from the recently completed randomized STRASS trial, should be the basis for future trials with the addition of RHT.20-22 Noncompliance and the rate of early dropouts were higher than expected from our previous experience in a phase 2 study.23 However, the number of patients with progressive disease or death prior to postinduction chemotherapy was higher in the NACT-alone group, thereby reducing the number of candidates for postinduction chemotherapy (OR, 3.4; 95% CI, 1.5-7.9; P = .003).
There are only a few trials in the neoadjuvant setting, and some with a similar parallel group design of chemotherapy. A small, phase 2 trial randomized 134 patients with heterogeneous risk criteria to doxorubicin (50 mg/m2) plus ifosfamide (5 g/m2) given for 5 cycles or to local treatment. The study was stopped owing to low accrual and no evidence that neoadjuvant chemotherapy improved survival.24
Using isolated limb perfusion under hyperthermic conditions as induction therapy in 231 patients who were all candidates for functional or anatomic amputation, the limb salvage rate was 81%, but 5-year overall survival was only 42% and poorest in patients with large tumors (P = .01) and with leiomyosarcoma (P = .001).25
The benefit of preoperative systemic chemotherapy in high-risk patients is supported by the results of the Italian Sarcoma Intergroup (ISG) and the Spanish Sarcoma-Intergroup trial. Designed as a noninferiority trial, 328 patients were randomized to 3 cycles of preoperative epirubicin (120 mg/m2) plus ifosfamide (9 g/m2) chemotherapy with or without 2 further cycles postoperatively.26 The 5-year overall survival rate was 70% in both treatment arms, and these results were similar to the published results of the Italian Sarcoma-Intergroup adjuvant trials, which demonstrated improved overall survival rates at 5 years of 66% and 70%, respectively, while the 5-year survival rates of the control arms were significantly lower (46% and 47%, respectively).27,28 A recent update confirmed the noninferiority of the preoperative 3 cycles with a 10-year overall survival of 61% (95% CI, 56%-67%) for the entire group of patients.29 The results apply predominantly to extremity tumors. With this restriction, the results of the EORTC 62961-ESHO 95 control arm for patients with extremity tumors who were treated with the 3-drug NACT regimen alone as comparator showed a 10-year survival rate of 59%, which was similar to the results of the Intergroup trial (61%).29 In addition, the 5-year survival rate after neoadjuvant chemotherapy alone was also much better than the 5-year overall survival rates after local treatment (surgery plus radiation) in the Italian adjuvant trials.29 Taking together, survival of patients with high-risk extremity tumors who were treated in our control arm without RHT was almost identical to those receiving short, full-dose preoperative chemotherapy, and was further improved adding RHT by almost 10%.
Therefore, these results reinforced the significance of the additional benefit by RHT because they were not confounded by an insufficient efficacy of the chemotherapy regimen. The survival benefit was also observed in patients with less favorable, abdominal-retroperitoneal tumors, and was even more pronounced in grade 2 tumors, due to yet unknown mechanisms. This observation is surprising because only high-grade tumors are supposed to be chemosensitive as supported by the retrospective analysis of the French Sarcoma Group30 showing that grade 2 tumors did not benefit from adjuvant chemotherapy in contrast to grade 3 tumors. Because distinct histotypes were thought to be more sensitive to specific cytotoxic drugs, the most recent ISG trial randomized 287 patients to their standard of preoperative epirubicin plus ifosfamide chemotherapy, or to 1 of 5 histologically tailored chemotherapy regimens.31 The study was stopped because the experimental arm showed a significantly lower relapse-free and overall survival. In EORTC 62961-ESHO 95, improved survival by RHT was seen in L-sarcoma, as well as in all other high-grade histological subtypes.
That we know of, EORTC 62961-ESHO 95 is still the first randomized trial to be carried out and completed comparing systemic chemotherapy with or without RHT in a high-risk patient population. As such, we should not exclude the potential therapeutic benefits RHT may also have in solid tumors other than soft tissue sarcoma. To test this further, a multicenter, randomized phase 3 trial in resected pancreatic cancer is ongoing (NCT01077427). We have also been conditioned to discount observational studies, and practice changes are only made based on results from randomized trials.32 However, in the rare subset of pediatric, malignant nontesticular germ-cell tumors, a phase 2 study adding RHT to salvage chemotherapy has demonstrated outcome benefits almost similar to first-line treatment.33 Therefore, there is an urgent need to raise more interest in this treatment modality by oncologists in dedicated centers.
The EORTC 62961-ESHO 95 trial showed a significant improvement in survival in patients receiving neoadjuvant chemotherapy combined with regional hyperthermia. However, the study design was not powered enough to show the statistical evidence for all subgroups (eg, extremity vs abdominal and/or retroperitoneal sarcomas). For patients who were treated in combination with regional hyperthermia, completion of induction therapy was significant for survival, however only two-thirds of these patients received post-induction therapy. Therefore, the required number of post-induction therapy cycles, for the overall survival benefit, remained open.
The EORTC 62961-ESHO 95 study provides robust evidence that among patients with localized high-risk soft tissue sarcoma, the use of RHT added to neoadjuvant chemotherapy resulted in increased survival, as well as local progression-free survival. For patients who are candidates for neoadjuvant treatment, adding RHT may be warranted.
Corresponding Author: Rolf D. Issels, MD, PhD, Department of Medicine III, University Hospital, LMU Munich, Marchioninistr 15, 81377 Munich, Germany (firstname.lastname@example.org).
Accepted for Publication: November 2, 2017.
Correction: This article was corrected on April 12, 2018, for errors in the author affiliations for Drs Verweij and Gronchi; this article was previously corrected on March 8, 2018, for a typographical error in the label for panel C in Figure 2.
Published Online: February 15, 2018. doi:10.1001/jamaoncol.2017.4996
Open Access: This article is published under the JN-OA license and is free to read on the day of publication.
Author Contributions: Dr Issels had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Issels, Verweij.
Acquisition, analysis, or interpretation of data: Issels, Lindner, Verweij, Wessalowski, Reichardt, Wust, Ghadjar, Hohenberger, Salat, Vujaskovic, Daugaard, Mella, Mansmann, Dürr, Knoesel, Abdel-Rahman, Schmidt, Hiddemann, Jauch.
Drafting of the manuscript: Issels, Lindner, Verweij, Wessalowski.
Critical revision of the manuscript for important intellectual content: Issels, Lindner, Verweij, Wessalowski, Reichardt, Wust, Ghadjar, Hohenberger, Angele, Salat, Vujaskovic, Daugaard, Mella, Mansmann, Dürr, Knoesel, Abdel-Rahman, Schmidt, Hiddemann, Jauch, Gronchi.
Statistical analysis: Mansmann, Schmidt.
Obtained funding: Issels.
Administrative, technical, or material support: Issels, Wessalowski, Reichardt, Hohenberger, Vujaskovic, Mella, Dürr, Knoesel, Abdel-Rahman.
Study supervision: Issels, Lindner, Verweij, Wessalowski, Reichardt, Hohenberger, Salat, Vujaskovic, Hiddemann, Gronchi.
Conflict of Interest Disclosures: Dr Issels has received travel support and honoraria from Pyrexar, PharmaMar, Medtherm and Bayer. Dr Lindner has received research and travel support from Dr Sennewald Medizintechnik; travel support from PharmaMar; and honoraria from Novartis, Lilly, Eisai, EL Medconsult. Mr Abdel-Rahman has received travel support and honoraria from Pyrexar and Medtherm. No other conflicts are reported.
Funding/Support: This work was supported by the European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group (EORTC-STBSG) and promoted by research grants from the Deutsche Krebshilfe and Helmholtz Association, as well as the National Institutes of Health (grant No. NIH P01 CA42745).
Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Group Information: The EORTC 62961-ESHO 95 Randomized Clinical Trial group members are Rolf D. Issels, MD, PhD; Lars H. Lindner, MD; Jaap Verweij, MD; Peter Reichardt, MD; Peter Hohenberger, MD; Soeren Daugaard, MD; Alessandro Gronchi, MD; and the European Society for Hyperthermic Oncology (ESHO) group members are, Rüdiger Wessalowski, MD; Peter Wust, MD; Pirus Ghadjar MD; Olav Mella, MD; Zeljko Vujaskovic, MD; and Sultan Abdel-Rahman, PhSc.
Additional Contributions: In addition, the following investigators supported the EORTC-ESHO multicentre trial: Austria: F. Ploner, who died July 22, 2015, and R. Windhager. Germany: H. J. Sauer, J. Gellermann, R. Wilkowski, M. Kuhlencordt, B. Weber, and F. Hörterer. Norway: O. Dahl. United States: M. Dewhirst and J. O’Regan (Bingham Mayne and Smith Ltd), who provided editorial assistance with the final draft of the manuscript and coordinated the submission. We are also indebted to Wolfgang Wilmanns, MD, PhD, who died December 24, 2003, for his continuous support of “Hyperthermia” at the campus Klinikum Grosshadern, LMU. We also thank the many investigators in medical oncology, surgical oncology, and radiation oncology who worked with the principal investigator at each participating institution as part of the combined-modality treatment team for the patients. We are indebted to all the patients who took part in the study and to their families. The statistical analysis was performed independently from the sponsor at the Institute of Medical Information Sciences, Biometrics, and Epidemiology (LMU) (Ulrich Mansmann, PhD, and Michael Schmidt, MD). We thank Christina Klose and Rüdiger Paul Laubender for supporting data management.
Additional Information: The principal investigator, Rolf D. Issels, MD PhD, and Jaap Verweij, MD, designed and conducted the study in close collaboration with the EORTC-STBSG. Blinded interim and safety reports were given at the annual EORTC-STBSG meetings. The Independent Data Monitoring Committee (IDMC) of the EORTC reviewed the data in June 2004 and advised the trial to be continued; EORTC 62961-ESHO 95 was an investigator-initiated trial; the legal sponsor was the ESHO.
Meeting Presentations: There are 2 analyses of the full data set, the first with a median follow-up of 34 months published in Lancet Oncology (2010;11:561-570). The second was presented in part at European Cancer Organisation/European Society For Medical Oncology meeting in Vienna, Austria; September 26, 2015 (late-breaking abstract, oral presentation).