Surgeons’ opinions about enhanced recovery after surgery for retroperitoneal sarcoma: A survey

Abstract

BACKGROUND

Enhanced recovery after surgery (ERAS) programs provide recommendations for an optimized management of patients undergoing surgery. An ERAS program tailored on surgery for retroperitoneal sarcomas (RPS) may improve patient outcomes and it has still not been established.

AIM

To determine how an ERAS program tailored to RPS surgery can be agreed upon, structured, and implemented.

METHODS

Twenty-five candidate items from existing ERAS programs, potentially relevant for RPS surgery, were identified via literature review and expert input. These were included in a questionnaire refined through cognitive interviews and pilot testing. Expert sarcoma surgeons rated each item’s relevance and feasibility on a 6-point scale. The survey was recirculated after one year. Intra-observer reproducibility, inter-observer concordance, and agreement with the modal value of the most experienced participants were analyzed.

RESULTS

Thirteen sarcoma surgeons from 6 centers participated in the survey. Although surgeons agreed on several items, their overall concordance was low. After recirculating the survey, the intra-observer reproducibility was low. Interestingly, the median concordance with the reference increased for relevance and decreased for feasibility.

CONCLUSION

Despite interest in ERAS for RPS, surgeon concordance on item relevance and feasibility remains low, underscoring the need for collaborative efforts toward a standardized, consensus-based protocol.

Key Words: Retroperitoneal sarcoma; Enhanced recovery after surgery; Fast-track surgery; Sarcoma surgery; Perioperative care; Surgical outcomes

Core Tip: This study explores the development of an enhanced recovery after surgery (ERAS) program tailored to retroperitoneal sarcoma (RPS) surgery through a structured expert survey. Despite growing interest, surgeon concordance on the relevance and feasibility of proposed ERAS items remains limited, even after a one-year interval. These findings underscore the complexity of translating general ERAS principles into RPS-specific protocols and highlight the urgent need for broader multidisciplinary collaboration to establish a standardized, consensus-based pathway. This work lays the groundwork for future efforts aimed at optimizing perioperative care and improving outcomes in RPS patients.

INTRODUCTION

Enhanced recovery after surgery (ERAS) programs provide multidisciplinary bundles of recommendations to manage surgical patients from preoperative consultation to discharge, with the ultimate goal of improving patient recovery and postoperative outcomes. Following early investigations in colorectal surgery, ERAS programs have been proven effective, safe, and cost-efficient across various surgical procedures[1-3].

Regardless of the surgical subspecialty, the primary objectives of ERAS protocols are to reduce stress response, maintain perioperative homeostasis, and facilitate faster bowel function recovery[4,5]. Preoperatively, comorbidities and nutritional status are evaluated and optimized, smoking and alcohol cessation is encouraged, physical activity is promoted, and preoperative counseling is routinely provided. Before surgery, patients receive thromboembolic, antiemetic, and antimicrobial prophylaxis, overnight fasting is avoided, and a preoperative carbohydrate drink is administered to maintain an anabolic state during and after surgery[6]. Intraoperatively, ERAS guidelines emphasize both a minimally stressful surgical technique and a well-balanced, standardized anesthesiologic approach[7]. Key elements of postoperative care include the resumption of a full oral diet as soon as tolerated and early mobilization. Multimodal pain control strategies and nutritional support, particularly in cases of unsatisfactory oral intake, are implemented[6]. Systematic audit is a cornerstone of ERAS programs, aiming-through extensive data collection-to assess adherence to a multistep protocol involving healthcare providers, patients, and families, and to analyze outcomes[8]. The systematic implementation of ERAS protocols has been shown to shorten hospital stays, reduce healthcare costs, and improve surgical outcomes[3,9]. Notably, recent studies have demonstrated a significant association between ERAS implementation and timely administration of adjuvant chemotherapy in cancer patients, suggesting a potential long-term prognostic benefit[10].

ERAS protocols are now widely accepted and applied worldwide. Ongoing research and implementation continue across an expanding range of surgical subspecialties, with increasing interest in surgical oncology due to a growing understanding of the relationship between improved postoperative recovery, reduced stress, and patient survival[10,11]. However, the application of ERAS in soft tissue sarcoma surgery remains limited to single-center experiences, primarily in superficial truncal and limb sarcomas[12,13]. Evidence is even scarcer for retroperitoneal sarcoma (RPS) surgery[14].

Extensive surgery is the cornerstone of RPS treatment and often requires concomitant major surgical procedures[15-19]. When performed in referral centers, this surgery is safe, though it remains associated with clinically significant postoperative morbidity[20]. The potential benefits of an enhanced recovery program were explored in a recent study, where an ERAS-based protocol was implemented in 234 sarcoma patients, including 36 RPS cases. The study demonstrated a reduction in postoperative morbidity and hospital length of stay in this specific subgroup[14]. The need for a standardized RPS-tailored ERAS protocol was further highlighted in the latest Transatlantic Australasian RPS Working Group (TARPSWG) consensus paper[21]. However, the heterogeneity of RPS resections and perioperative challenges necessitates a dedicated ERAS framework distinct from other abdominal surgeries.

The aim of this study is to summarize the current evidence on ERAS programs for RPS and to gather expert opinions from sarcoma surgeons, ultimately proposing an enhanced recovery program tailored to RPS surgery.

MATERIALS AND METHODS

Study design

A PubMed search was performed using the keywords “ERAS” and “Enhanced Recovery After Surgery”, in combination with “protocol”, “guidelines”, “retroperitoneal”, and “sarcoma”. A total of 245 potential citations were identified and screened. Of these, 214 manuscripts were excluded as not relevant, either because they addressed unrelated procedures, referred to previous or updated guidelines, or consisted of single-center case series of limited interest (Figure 1). Official ERAS Society guidelines and manuscripts addressing RPS surgery, as well as major abdominal and urological procedures, were included within a timeframe spanning 2012 to 2021. A total of 31 manuscripts were critically reviewed to identify candidate items suitable for an RPS surgery protocol.

Figure 1 PRISMA flow diagram of literature search and review. ERAS: Enhanced recovery after surgery.

Survey

A questionnaire was designed to assess these outcomes, reporting the identified items along with a review of the current evidence and a critical commentary on their potential usefulness and applicability to RPS surgery. The selected items were classified into the following domains: Preoperative, anesthesiology, intraoperative, and postoperative. Relevance and feasibility were rated for each item on a 6-point scale. A commercial internet-based survey application (www.surveymonkey.com) was used to create, distribute, and collect responses. The identity of survey respondents remained anonymous to the authors throughout the study. Prior to distribution, the survey was independently tested by three surgeons to assess readability and usability. All staff surgeons from the sarcoma units collaborating in the European School of Soft Tissue Sarcoma Surgery[22] (Institut Curie, PSL University, Paris, France; Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; Queen Elizabeth University Hospitals, Birmingham, United Kingdom; Antonii van Leeuwenhoek - The Netherlands Cancer Institute, Amsterdam, Netherlands; The Royal Marsden Hospital, London, United Kingdom) were invited to participate in the survey and rate the perceived relevance and feasibility of each item. The survey was recirculated after one year.

The four most experienced surgeons- i.e., those with > 20 years of sarcoma practice, those serving as clinical lead at their center, or both-were identified as the reference group. The intra-observer reproducibility, inter-observer concordance, and comparison with the reference value (defined as the modal value of the reference group) were analyzed. To this end, a three-level classification scale was adopted for each item: Low (score 0 or 1), Moderate (score 2 or 3), and High (score 4 or 5). Due to the ordinal nature of this variable, the weighted kappa statistic (Kw) was used to measure concordance. Specifically, Kw values ≥ 0.70 were considered “quite satisfactory” (if < 0.80) or “satisfactory” (if ≥ 0.80), according to recently suggested interpretation criteria[23]. All statistical analyses were performed using SAS Software (version 9.4; SAS Institute Inc., Cary, NC, United States) with a significance alpha level of 0.05. The statistical analyses of the study were performed and reviewed by biomedical statisticians.

As no patient data were used, this study was deemed exempt from approval by the Institutional Review Board of our institution.

RESULTS

Following a literature review, 25 different items were selected as potentially relevant for an RPS-tailored ERAS program. Of these, 10, 5, 3, and 7 items were categorized under the preoperative, anesthesiology, intraoperative, and postoperative domains, respectively. The full list of items included in the survey, along with the proposed statements and corresponding literature references[7,20,21,24-48], is summarized in Table 1.

Table 1 List of the items proposed, and the statement presented in the survey.

n	Item	Recommendation	Ref.
Preoperative domain
1	Preoperative education	Patients should receive preoperative counselling	[24-33]
2	Centralization	Patients should be centralized to referral center and managed in a multidisciplinary tumor board	[20,21,27,34,35]
3	Prehabilitation	Alcohol consumption and smoking should be avoided for at least 4 weeks before surgery. Physical activity should be encouraged	[7,24-27,29-33]
4	Anemia	Anemia should be identified, investigated, and corrected preoperatively	[7,25,27,31,32]
5	Preoperative nutrition	Routinely preoperative nutritional status evaluation and preoperative oral nutritional support in malnourished patients should be offered	[21,24-32,35]
6	Bowel preparation	Mechanical bowel preparation (MBP) alone should not be used routinely but may be used in selected cases. In patients receiving oral MBP, oral antibiotics should be given	[25,26,29]
7	Preoperative fasting and oral carbohydrate loading (OCL)	Preoperative fasting should not need to exceed 6 hours for solids and 2 hours for liquids. Preoperative OCL should be administered to all non-diabetic patients	[7,24-26,28-33,37,38]
8	Prevention of postoperative nausea and vomiting (PONV)	All patients should be screened preoperatively for PONV risk factors and stratified. A multimodal PONV treatment should be provided according to present guidelines	[7,24-26,28-33]
9	Thromboembolic prophylaxis, skin preparation, and intravenous antibiotic prophylaxis	Patients should receive low molecular weight heparin once daily for 28 days after surgery, additional measures as mechanical thromboprophylaxis by well-fitting compression stockings and/or intermittent pneumatic compression should be prescribed until discharge. Preoperative skin preparation should be done with chlorhexidine-based solution. Preoperative intravenous antibiotic prophylaxis, administrating 3rd generation cephalosporine (plus metronidazole in case of planned colic resection) has to be done 60 minutes prior to surgery	[7,24-26,28-33]
10	Preoperative steroid administration	Administration of methylprednisolone at a dosage of 30 mg/kg 2 hours before surgery in non-diabetic patients is recommended when a concomitant liver resection is planned. The usefulness of routinary preoperative steroid administration before RPS surgery should be investigated	[28]
Anesthesiology domain
11	Preoperative sedative medication	Preoperative routine sedative medication should be avoided. Effective strategies to treat preoperative anxiety should be multimodal, comprising preoperative education. Pre-anesthetic medication with gabapentanoids agents, NSAIDS and acetaminophen to better control postoperative pain are safe and effective	[7,25,26,30-32,40]
12	Intraoperative anesthesia management	To attenuate the surgical stress response, intraoperative maintenance of adequate hemodynamic control, central and peripheral oxygenation, muscle relaxation, depth of anesthesia, and appropriate analgesia is strongly recommended	[7,25,29-33]
13	Perioperative fluid management	The goal of perioperative fluid therapy is to maintain fluid homoeostasis avoiding fluid excess and organ hypoperfusion, and a perioperative near-zero fluid balance is recommended. Goal-directed fluid therapy should be adopted. Inotropes should be considered in patients with poor contractility (CI < 2.5 L/min)	[7,24-33,39]
14	Preventing intraoperative hypothermia	Reliable core temperature monitoring should be undertaken in all patients and methods to actively warm patients to avoid hypothermia should be employed, as by using warming blankest and warmed gases and fluids	[7,25,26,28-32]
15	Perioperative multimodal analgesia	To minimize perioperative opioid use, a multimodal perioperative analgesia should be proposed according to latest guidelines and should include all opioid-sparing techniques as multipharmacological analgesia, thoracic epidural analgesia, tranversus abdominal plane blocks, intravenous lidocaine infusion and patient-controlled analgesia. The long-term impact on postoperative chronic pain should be evaluated	[7,24-31,33,40-42,44,45]
Intraoperative domain
16	Minimally invasive access	Minimally invasive approach should not be employed routinely in RPS surgery	[16-19,35,43]
17	Prevention of delayed gastric emptying (DGE)	Currently affecting about 40% of RPS patients and lacking specific evidence, DGE is a critical area of interest and should be studied. At date, no recommendation about DGE prevention can be made	[38]
18	Abdominal and chest drainage	A tailored drain minimizing policy should be advisable, although specific evidence is lacking. Early removal of drains after 72 hours may be advisable in patients at low risk of pancreatic fistula after pancreatic resection (i.e., amylase content in drain < 5.000 U/L) but the impact of such a policy over the risk of chyle leak and lymphorrhagia is a critical area of interest. In patient underwent diaphragmatic resection, one chest drain should be placed intraoperatively. Chest drain may be removed in the absence of air and chyle leaks with a serous effusion < 450 mL/24 hours	[25-28,33,44,45]
Postoperative domain
19	Nasogastric tube	Postoperative nasogastric tubes should not be used routinely; if inserted during surgery, they should be removed before reversal of anesthesia	[7,24-26,28-32,48]
20	Urinary drainage	Routine transurethral catheterization is recommended for 1-3 days after surgery. The duration should be individualized based on known risk factors for retention: Male gender, epidural analgesia, and pelvic surgery	[25,29,31,33,44,45]
21	Postoperative nutrition	Postoperative nutritional status evaluation and early per os realimentation should be provided in all patients. Nutritional support with both immunonutrition, enteral nutrition and/or parenteral nutrition, should be considered in malnourished patients and follow latest guidelines	[29,36,47]
22	Prevention of postoperative ileus	Moderate level evidence supports use of peripherally acting mu-opioid receptor antagonist as alvimopan, while low grade evidence suggests a possible role for magnesium oxide, bisacodyl, and coffee	[25]
23	Postoperative glycemic control	Insulin treatment is strongly recommended in intensive care unit setting for severe hyperglycemia. Recommendation to treat mild hyperglycemia, and inward patients is weak	[24-26,28,29]
24	Early and scheduled mobilization	Patients should be mobilized actively from the morning of the first postoperative day and encouraged to meet daily targets for mobilization	[7,24-33,44,45,48]
25	Audit	Systematic audit improves compliance and clinical outcomes	[7,24-33]

Evidence quality is reported according to the GRADE system. RPS: Retroperitoneal sarcomas; NSAIDS: Non-steroidal anti-inflammatory drugs.

Fifteen expert sarcoma surgeons from the ESSTSS were invited to participate in this survey, and 13 (87%) from six different centers responded. An overview of the results from the two survey rounds is presented in Figure 2. The heatmap uses a three-color scale to depict the rating assigned by observers to each item. Although a trend towards similar ratings was observed for a limited number of items, the overall concordance of surgeons' responses was low. The overall intra-observer reproducibility- i.e., the ability of a sarcoma surgeon to provide consistent responses one year apart-was low, with a median Kw value of 0.411 (range: 0.028-0.762). The intra-observer reproducibility was higher among the reference group. Moreover, the median intra-observer reproducibility was greater for relevance than for feasibility, with Kw values of 0.436 (range: 0-0.770) and 0.380 (range: -0.048-0.754), respectively. We also assessed the distribution of intra-observer reproducibility by comparing the different domains in terms of relevance and feasibility. As depicted in Figure 3, the lowest reproducibility was observed in the anesthesiology domain, while the highest was found in the intraoperative domain.

Figure 2 Heatmap representing an overview on response to the survey. On the vertical axis, the different observer over the two rounds of the survey; on the horizontal axis, the items proposed (R, relevance; F, feasibility). Color scale: Red, rating Low (score 0-1); yellow, rating Moderate (score 2-3); green, rating High (score 4-5); grey, Not Available (bimodal distribution in reference modal value; no rating given in observer axis).

Figure 3 Distribution of the intra-observer reproducibility over the whole survey, considering feasibility and relevance, respectively, and according to the four different domains (preoperative, anesthesiology, intraoperative, and postoperative). Each box indicates the 25^th and 75^th percentiles. The horizontal line in the box indicates the median, and the whiskers indicate the extremes. The dotted line indicates the weighted kappa value of 0.70- i.e., a “quite satisfactory” level of reproducibility.

We further explored the concordance between the observers and the reference group regarding both the relevance and feasibility of each item across the two survey rounds. The median Kw value for relevance and feasibility in the two rounds was 0.227 (range: -0.154-0.881) vs 0.426 (range: 0-0.737), and 0.292 (range: -0.243-0.647) vs 0.265 (range: -0.143-1.00), respectively (Table 2).

Table 2 Inter-observer concordance with the reference group (median Kw) in the two rounds of the survey.

Observer	Relevance (Kw)		Feasibility (Kw)
	Round 1	Round 2	Round 1	Round 2
11	0.5681	0.7001	0.2941	0.3531
2	-0.154	0.386	0.185	0.350
31	0.2191	0.4621	0.3331	0.1881
41	0.6151	0.7281	0.6471	1.0001
51	0.4091	0.7371	0.3641	0.6021
6	0.227	0.438	0.333	0.143
7	-0.099	0.088	-0.154	-0.083
8	0.881	0.361	-0.059	0.265
9	0.144	0.612	0.292	0.306
10	0.298	0.426	0.391	0.138
11	-0.008	0.337	-0.243	-0.143
12	-0.094	0.000	0.037	0.091
13	0.298	0.353	0.286	0.352
Median Kw (range)	0.227 (-0.154; 0.881)	0.426 (0; 0.737)	0.292 (-0.243; 0.647)	0.265 (-0.143; 1.00)

¹The surgeons constituting the reference group.

Kw: The weighted kappa statistic.

At the individual item level, Figure 4 illustrates the observed concordance for each item across the investigated domains. Specifically, the perceived relevance of items 1, 2, 4, 9, 13, 14, 21, and 24, as well as the perceived feasibility of items 2, 9, 14, 21, and 24, were identified as those with the highest agreement both intra-observer and in comparison with the reference group across the two survey rounds (i.e., > 10 out of 13). In practice, substantial agreement was limited to the following eight items across the preoperative, anesthesiology, intraoperative, and postoperative domains: (1) Preoperative patient education (item 1, relevant); (2) Centralization and multidisciplinary management (item 2, relevant and feasible); (3) Anemia screening and treatment (item 4, relevant); (4) Thromboembolic prophylaxis, skin preparation, and intravenous antibiotic prophylaxis (item 9, relevant and feasible); (5) Perioperative fluid management (item 13, relevant); (6) Prevention of intraoperative hypothermia (item 14, relevant and feasible); (7) Postoperative nutrition (item 21, relevant and feasible); and (8) Early and scheduled mobilization (item 24, relevant and feasible).

Figure 4 Item-level bar plot. Each bar reports the observed concordance in terms of absolute frequency. The higher is the bar, the stronger is the observed agreement. A: Intra-observer concordance evaluation between the first and second round; B: Concordance evaluation vs reference in the first round; C: Concordance evaluation vs reference in the second round.

DISCUSSION

Our study found that agreement among expert sarcoma surgeons regarding the relevance and feasibility of selected ERAS items for an RPS-tailored program was overall low, both in terms of intra-observer reproducibility (Kw = 0.411) and concordance with the reference group of the most experienced surgeons (Kw ranging from 0.227 to 0.426 over one year). Notably, the reference group demonstrated a higher intra-observer reproducibility. The present study provides an overview of current evidence supporting the development of an ERAS program for RPS surgery and establishes benchmarks for a structured collaborative effort to achieve international consensus.

ERAS programs have demonstrated efficacy in improving surgical outcomes across various procedures. The development of an enhanced recovery program tailored to multivisceral resections for RPS is anticipated to yield similar benefits and has been advocated by sarcoma experts in a recent consensus statement[21]. Moreover, an ERAS program for RPS could facilitate the design and execution of studies assessing its effectiveness at specialized sarcoma centers in a standardized and reproducible manner.

Historically, RPS surgery has adopted perioperative practices from other abdominal specialties, including prolonged preoperative fasting, bowel preparation, and delayed mobilization. However, the morbidity profile of RPS resections-often involving multivisceral procedures-calls for a tailored approach. Collaborative efforts have analyzed the morbidity associated with multivisceral surgery for RPSs[49], and recent single-center ERAS initiatives have explored potential interventions to improve surgical outcomes in RPS patients, demonstrating promising reductions in complications and hospital stay[35,50].

Although the survey identified consensus on the potential benefits of an RPS-tailored ERAS program, our analysis revealed low concordance among experts regarding which items should be prioritized for inclusion, which may reflect institutional variability in ERAS adoption. In addition, we observed limited intra-observer reproducibility, with the lowest concordance in the anesthesiology domain. The low intra-observer agreement (Kw = 0.41) may reflect the continuous evolution of RPS surgery, a rapidly advancing field given that the first international guidelines were published in 2015[15] and updated in 2021[21]. Moreover, it may mirror the dynamic nature of ERAS integration into RPS practice, where evolving evidence and institutional protocols influence perspectives. Future surveys with shorter intervals and standardized educational components could improve consistency. The modest Kw values indicate that surgeons’ perceptions are inconsistent over time, likely reflecting the lack of established standards in RPS. This underscores the need for repeated assessments alongside protocol roll-out to track consensus maturation.

Notably, median concordance values were higher for relevance than for feasibility. However, overall observer concordance with the reference group increased over one year. While agreement on relevance improved, feasibility scores declined, suggesting that while an RPS-tailored ERAS program is considered valuable, its implementation remains a significant challenge, even among experts. This underscores the need for a broader effort, including other healthcare professionals and stakeholders, to establish a systematic and effective program.

Substantial agreement on relevance was limited to a small set of items that are widely accepted by surgeons and included in most existing ERAS programs. Notably, none of these items are specific to RPS surgery (e.g., thromboembolic prophylaxis, early mobilization), reflecting a paucity of evidence for tailored interventions. Further studies testing RPS-focused strategies-such as drainage management, nutritional support and perioperative steroid administration-are needed to bridge this gap. Interestingly, the greatest disagreement concerned the feasibility of items requiring collaboration with other healthcare professionals (nurses, case managers, anesthesiologists, hematologists), underscoring the necessity of multidisciplinary discussion.

We acknowledge certain limitations in this study. Primarily, the small number of participants in this exploratory survey may serve as a selection bias. While our survey captured expertise from high-volume European centers, the small sample size and pre-existing network affiliations may limit generalizability. However, the observed variability even within this homogeneous group suggests widespread uncertainty, warranting broader international consensus. Moreover, only surgeons were surveyed, which partially contradicts the multidisciplinary nature of the ERAS approach and precludes insights from other ERAS-critical disciplines. This “surgeon-centric” design, while pragmatic for an initial feasibility assessment, underscores the need for broader stakeholder engagement. Future Delphi-based consensus-building should integrate perspectives from anesthesia, nursing, and rehabilitation teams to address implementation barriers, particularly for items rated as relevant but less feasible (e.g., nutritional support, pain management). Finally, the absence of patient-reported outcomes limits assessment of ERAS acceptability.

Nonetheless, we aimed to leverage a well-established group of European experts with a history of collaboration over the past years[18] as part of the European School of Soft Tissue Sarcoma Surgery[22], to evaluate the current landscape and stimulate interest in this topic.

Despite its limitations, the core set of ERAS items identified through this survey may serve as a foundation for broader consensus-building efforts and the development of a comprehensive RPS-tailored program. While high-consensus items were identified, we refrained from proposing a formal protocol due to limited agreement on most components. A Delphi process involving a larger, multidisciplinary panel is needed to draft actionable guidelines. Prospective evaluation of the agreed-upon items will pave the way for protocol standardization, with pilot studies at high-volume centers prioritized to address feasibility concerns. Ultimately, this work aims to inform an international Delphi consensus initiative under the auspices of TARPSWG, formalizing a standardized ERAS protocol for prospective validation

CONCLUSION

The integration of a tailored ERAS program into RPS surgery has the potential to improve short- and long-term patient outcomes. Several ERAS items may be incorporated into a dedicated RPS protocol designed for multivisceral surgery. However, beyond a small number of agreed-upon items, expert concordance remains low. While the need for an RPS-specific ERAS protocol is widely recognized, a structured, collaborative, and multidisciplinary effort is essential to achieve consensus on its key components and optimize recovery pathways for this complex patient population.

ACKNOWLEDGEMENTS

The present work is part of the European School of Soft Tissue Sarcoma Surgery (ESSTSS) Fellowship run by the European Society of Surgical Oncology (ESSO) in which LI took part. The Authors acknowledge the sarcoma experts involved in the ESSO-ESSTSS which participated in the survey (in alphabetical order): Sylvie Bonvalot (a), Dario Callegaro (b), Chiara Colombo (b), Samuel J. Ford (c), Stefano Radaelli (b), Yvonne Schrage (d), Myles Smith (e), Dirk Stauss (e), Dimitri Tzanis (a) and Winan J. van Houdt (d).

Institut Curie, PSL University, Paris, France; Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; Queen Elizabeth University Hospitals, Birmingham, United Kingdom; Antonii van Leeuwenhoek-The Netherlands Cancer Institute, Amsterdam, Netherlands; The Royal Marsden Hospital, London, United Kingdom.