Open access Protocol

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
Mesothelioma and Radical Surgery 2
(MARS 2): protocol for a multicentre
randomised trial comparing (extended)
pleurectomy decortication versus no
(extended) pleurectomy decortication
for patients with malignant
pleural mesothelioma
Eric Lim,1 Liz Darlison,2,3 John Edwards,4 Daisy Elliott,5 D A Fennell,6
Sanjay Popat,7 Robert C Rintoul,8 David Waller,9 Clinton Ali,10 Andrea Bille,11
Liz Fuller,12 Andreea Ionescu,13 Manjusha Keni,14 Alan Kirk,15 Pek Koh,16
Kelvin Lau,9 Talal Mansy,17 Nick A Maskell,18,19 Richard Milton,20
Dakshinamoorthy Muthukumar,21 Tony Pope,22 Amy Roy,23 Riyaz Shah,24
Jonathan Shamash,25 Zacharias Tasigiannopoulos,26 Paul Taylor,27 Sarah Treece,28
Kate Ashton,29 Rosie Harris,29 Katherine Joyce,29 Barbara Warnes ‍ ‍,29
Nicola Mills,5 Elizabeth A Stokes ‍ ‍,30 Chris Rogers,29 On behalf of MARS 2
Trialists

To cite: Lim E, Darlison L,

Edwards J, et al. Mesothelioma
ABSTRACT
and Radical Surgery 2 (MARS Strengths and limitations of this study
2): protocol for a multicentre
Introduction Mesothelioma remains a lethal cancer.
randomised trial comparing To date, systemic therapy with pemetrexed and a ►► (Extended) pleurectomy decortication is currently
(extended) pleurectomy platinum drug remains the only licensed standard offered to patients with mesothelioma on the United
decortication versus no of care. As the median survival for patients with Kingdom National Health Service, but it is unknown
(extended) pleurectomy mesothelioma is 12.1 months, surgery is an important whether it is a clinically beneficial or cost-­effective
decortication for patients consideration to improve survival and/or quality of life. treatment option. MARS 2 is the first randomised
with malignant pleural Currently, only two surgical trials have been performed
mesothelioma. BMJ Open controlled trial to compare this type of surgery with
which found that neither extensive (extra-­pleural no surgery in this patient population.
2020;10:e038892. doi:10.1136/
bmjopen-2020-038892 pneumonectomy) or limited (partial pleurectomy) ►► Surgical quality assurance measures will be imple-
surgery improved survival (although there was some mented to ensure that the intervention will be deliv-
►► Prepublication history for evidence of improved quality of life). Therefore, ered at centres with expertise.
this paper is available online.
clinicians are now looking to evaluate pleurectomy ►► Patients may come with a pre-­conceived perception
To view these files, please visit
the journal online (http://​dx.​doi.​
decortication, the only radical treatment option left. that surgery will be beneficial, which can lead to
org/​10.​1136/​bmjopen-​2020-​ Methods and analysis The MARS 2 study is a crossovers (ie, patients allocated to no surgery may
038892). UK multicentre open parallel group randomised go on to seek surgery elsewhere). The integrated
controlled trial comparing the effectiveness and cost-­ Quintet Recruitment Intervention supports recruit-
Received 27 March 2020 effectiveness of surgery—(extended) pleurectomy ment staff in responding to patient preferences and
Revised 27 May 2020 decortication—versus no surgery for the treatment conveying balanced information.
Accepted 24 June 2020 ►► It is not possible to blind participants or the study
of pleural mesothelioma. The study will test the
hypothesis that surgery and chemotherapy is superior team, but the primary outcome (survival) is objective.
to chemotherapy alone with respect to overall survival. ►► Patient pathways vary at different sites. Some flexi-
© Author(s) (or their Secondary outcomes include health-­related quality bility has been worked into the protocol to allow for
employer(s)) 2020. Re-­use of life, progression-­free survival, measures of safety this.
permitted under CC BY.
(adverse events) and resource use to 2 years. The
Published by BMJ.
QuinteT Recruitment Intervention is integrated into the
For numbered affiliations see on 7 November 2013. We will submit the results for
trial to optimise recruitment.
end of article. publication in a peer-­reviewed journal.
Ethics and dissemination Research ethics approval
Trial registration numbers ISRCTN—ISRCTN44351742
Correspondence to was granted by London – Camberwell St. Giles
and ​ClinicalTrials.​gov—NCT02040272.
Eric Lim; ​e.​lim@r​ bht.​nhs.u​ k Research Ethics Committee (reference 13/LO/1481)

Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892 1

Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
INTRODUCTION high-­quality evidence of clinical efficacy or any evidence
In the UK, approximately 2500 patients are diagnosed on cost-­effectiveness.
each year with pleural mesothelioma,1 a treatment-­
resistant and lethal cancer of the membranes lining the Aims and objectives
outer surface of the lung and the inside of the chest wall MARS 2 is a UK-­wide multicentre RCT which will test the
primarily due to asbestos exposure. Deaths are increasing hypothesis that (extended) pleurectomy decortication
yearly and are estimated to peak this year.2 So far, most and chemotherapy is superior to chemotherapy alone
treatments have proven ineffective. The current standard with respect to overall survival for patients with pleural
of care, consisting of 4 to 6 cycles of platinum and peme- mesothelioma.
trexed chemotherapy, as recommended by the National Specific objectives are to estimate:
Institute for Health and Care Excellence (NICE),3 has A. The difference between groups in overall survival.
been associated with only an additional 3 months of B. The difference between groups with respect to a
survival.4 As the median survival for patients with meso- range of secondary outcomes including HRQoL,
thelioma is 12.1 months,4 surgery to remove as much of progression-­free survival and measures of safety (ad-
the disease as possible remains an important consider- verse health events).
ation to improve survival and/or health-­related quality of C. The cost-­effectiveness of (extended) pleurectomy de-
life (HRQoL).5 cortication compared with no surgery.
Pleurectomy decortication is the most common surgical
procedure for mesothelioma worldwide and is defined METHODS AND ANALYSIS
as parietal and visceral pleurectomy to remove all gross Trial design
tumour without diaphragm or pericardial resection.6 MARS 2 is a multicentre, non-­ blinded parallel two-­
Extended pleurectomy decortication can also be carried group, pragmatic RCT of surgery and chemotherapy
out, when parietal and visceral pleurectomy is undertaken versus chemotherapy alone for suitable patients with
to remove all gross tumour, including the resection of the mesothelioma.
diaphragm and/or pericardium. In the document we use An internal pilot funded by Cancer Research UK (award
the term (extended) pleurectomy decortication to refer ref: C27967/A15895) and coordinated by the Papworth
to either of the two procedures. The other main types of Trials Unit Collaboration demonstrated the feasibility of
surgery for mesothelioma are extra-­pleural pneumonec- recruitment across 14 medical sites and 2 joint medical
tomy, which is defined as en bloc resection of the parietal and surgical sites of excellence, as the target of 50 partic-
and visceral pleura with the ipsilateral lung, pericardium ipants recruited within a 24-­month period was achieved.
and diaphragm (in cases where the pericardium and/or Since the end of the pilot phase in December 2016,
diaphragm are not involved by tumour, these structures an additional eight medical, one surgical, and two joint
may be left intact); and partial pleurectomy, which is medical and surgical sites have been opened for the full
the partial removal of parietal and/or visceral pleura for trial. In addition, the full trial will provide recruiting sites
diagnostic or palliative purposes but leaving gross tumour with the support of an integrated QuinteT Recruitment
behind.6 Intervention (QRI)12–14 to optimise recruitment and
So far, no advantage, in terms of survival, has been retention.
observed with any type of surgery in randomised controlled
trials (RCTs) conducted to date. The MARS feasibility Setting, centre and surgeon eligibility
study (ISRCTN95583524), a trial of extra-­pleural pneu- This study is taking place in National Health Service
monectomy with adjuvant haemothorax irradiation, (NHS) secondary care centres, including teaching and
concluded that surgery was unlikely to offer either an district general hospitals.
improvement to survival or HRQoL and possibly harmed To be eligible as a medical site, the centre must:
patients.7 MesoVATS (ISRCTN34321019) concluded that 1. Be an NHS Trust with access to a multidisciplinary
partial pleurectomy did not improve survival, although it team (MDT) to discuss patients with mesothelioma.
showed that patients in the better prognostic group had 2. Have a track record of treating patients with
improved HRQoL after 6 months.8 mesothelioma.
Suitable patients, both in the UK and internationally, To be eligible as a surgical site, the centre must:
are currently offered pleurectomy decortication as it is 1. Be an NHS Trust with an established mesothelioma
considered to carry less morbidity compared with the MDT.
more extensive extra-­ pleural pneumonectomy but still 2. Have a minimum of two named mesothelioma sur-
achieves complete macroscopic resection which partial geons participating in the trial.
pleurectomy does not.9–11 However, we do not know if All surgeons participating in the full trial must be
(extended) pleurectomy decortication in conjunction accredited by (1) self-­reporting a minimum of five cases
with chemotherapy will improve survival compared with in which they have performed (extended) pleurectomy
the current standard of care (chemotherapy alone). In decortication, (2) observing the procedure being under-
the absence of RCTs, (extended) pleurectomy decor- taken at an established MARS 2 surgical site, (3) having
tication may continue to be offered despite a lack of a surgeon from the pilot phase observe their first MARS

2 Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892

 Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
2 procedure undertaken and (4) having one randomly stratification by recruiting site to ensure that the cohorts
selected MARS 2 operation between procedures 5 and are as balanced as possible.
10 observed by a surgeon from the pilot phase to ensure
fidelity. Trial interventions
Patients from all medical (only) sites are referred to a Patients will be randomised to receive one of the following
trial-­accredited surgical site for CT assessment of eligi- interventions:
bility, further discussion about the study and surgery (if ►► (Extended) pleurectomy decortication and chemotherapy: two
randomised to this group). cycles of platinum and pemetrexed chemotherapy
followed by surgery and then up to four cycles of the
Trial population same chemotherapy.
The target population are patients with a diagnosis of ►► Chemotherapy alone (control intervention): up to six cycles
epithelioid, sarcomatoid or biphasic mesothelioma. of platinum and pemetrexed chemotherapy alone
Patients will be eligible to take part if ALL of the following (current standard of care).
apply: The trial schema is illustrated in figure 1.
►► Adult aged ≥16 years of age. After randomisation, any changes in the choice of
►► Tissue (cytology or histology) confirmed epithelioid, chemotherapy, addition of other agents or entry into ther-
sarcomatoid or biphasic mesothelioma, as reviewed apeutic trials (eg, immunotherapies) will be permitted
by MDT to be of sufficient certainty to recommend for patients with progressive disease. At the time of trial
chemotherapy as treatment. design, there was no national consensus on postoperative
►► Disease confined to one hemithorax based on CT prophylactic radiotherapy, so it was decided that irradia-
assessment. tion to thoracic procedure sites may be undertaken for
►► Disease deemed surgically resectable by a surgeon at a MARS 2 patients. Patients in both groups can also receive
MARS 2 surgical site. further surgery, including thoracic, if it is without radical
►► Deemed fit for surgery by a surgeon at a MARS 2 intent. The aim is to conduct a pragmatic trial while
surgical site. closely monitoring uptake of additional therapies, studies
►► Capacity to provide written informed consent to or surgeries in order to account for them in the trial anal-
participate in the trial. yses, if required.
Patients will not be eligible if they have:
►► Severe shortness of breath (Eastern Cooperative Primary and secondary outcomes
Oncology Group status ≥2, or preoperative FEV1 or The primary outcome is survival, calculated from rando-
TLco less than 20%). misation date (randomisation occurs after the first two
►► Severe heart failure (NYHA III or IV, or ejection frac- cycles of chemotherapy). All participants will be followed
tion less than 30% by echocardiogram). up to the end of the trial (minimum of 2 years after
►► End-­stage kidney failure requiring dialysis. randomisation).
►► Liver failure (eg, encephalopathy and/or coagulation Secondary outcomes have been selected to assess the
abnormalities). efficacy of the two approaches. Secondary outcomes
►► Any other serious concomitant disorder that would are (1) progression-­free survival to the end of the trial
compromise participant safety during surgery. (minimum of 2 years after randomisation); (2) serious
►► Prisoner. adverse health events to 2 years after randomisation; (3)
►► Patient lacks capacity to consent. disease-­specific and generic HRQoL using the following
►► Existing co-­enrolment in another interventional study validated questionnaires—European Organisation for
that aims to improve survival. Research and Treatment of Cancer Quality of Life Ques-
tionnaire (EORTC QLQ-­C30), to assess the HRQoL of
Patient approach, consent and randomisation patients with cancer, and EuroQol EQ-­5D-­5L,15 16 a widely
The local research team at the medical site will take used generic measure of HRQoL (both of these will be
written informed consent from participants. In addi- measured at baseline, pre-­randomisation, and 6 weeks, 6,
tion to the main study, the team may also seek consent 12, 18 and 24 months post-­randomisation); and (4) health-
for audio-­ recording of consultations and participation care resource use to the end of the study: chemotherapy
in interviews, for QRI purposes. Participants will then cycles and initial surgical admission (for chemotherapy
receive two cycles of chemotherapy (standard care) and plus surgery group), and further resources measured at
have a further CT scan to confirm eligibility (ie, disease 6 weeks post-­randomisation then every 6 months, with a
still resectable) before being randomised, using a secure final follow-­up at the end of the study if not followed up
web-­ based randomisation system (Sealed Envelope in the previous 4 months.
https://​sealedenvelope.​com).
Participants will be randomised in a 1:1 ratio. Mini- Data collection
misation (with a random component) will be applied The schedule for data collection for the study is shown
for selected baseline variables (age, performance status in table 1. Data will be collected onto purpose-­designed
and cell type) that influence survival, in addition to case report forms (CRFs) and participant-­ completed

Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892 3

Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
Figure 1 Trial schema showing the recruitment pathway for the MARS 2 study.

questionnaires and entered onto a bespoke database for their HRQoL. Missing outcome data will be minimised,
data cleaning and analysis. Access to the database will be as survival and progression-­ free survival data can be
via a secure password-­protected web interface hosted on obtained from hospital records. Losses to follow-­up will
an NHS server. Data about adverse events will be collected be minimised by maintaining regular contact with partic-
and reported in accordance with sponsor’s and regula- ipants (by telephone and post) to complete follow-­up
tory requirements. questionnaires. Non-­adherence to randomised allocation
will be documented. Bias in the reported results will be
Risk of bias
minimised by having pre-­specified outcomes in the trial
Participants and clinical personnel cannot be blinded
protocol and a pre-­specified analysis plan.
to allocation due to the nature of the study interven-
tion. However, standard local protocols will be followed
in terms of patient care. The patient information leaflet Sample size
and conversations with MARS 2 site staff will describe and The total sample size has been set at 328 participants (164
balance the potential benefits and risks of both having per group). The patients randomised in the pilot trial will
and not having surgery. Therefore, this approach will contribute to the total sample size. The study will have
reduce participant’s expectations that one or other treat- 80% power to detect a HR of 0.7 at 5% statistical signifi-
ment protocol will lead to a more favourable result. cance (two-­sided), modelled on a published assumption
In addition, the study’s primary outcome is an objec- of a median survival time of 16.8 months in patients with
tive measure (survival), and clear definitions of each mesothelioma who were fit enough to receive surgery, but
secondary outcome measure will be provided to trial did not have it17 and allowing for 10% cross-­over from
personnel. The HRQoL follow-­ up questionnaires may the medical to surgery groups (as noted in previous trials
be more at risk of bias than other measures, but patients such as MARS).7 Cross-­over will be minimised through
will not have had this surgery previously and as such instruction (ie, recruit only patients who have equipoise
should not have any expectation regarding its effect on from the outset) and education.

4 Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892

 Table 1 Data collection
Pre-­randomisation Post-­randomisation
Follow-­up
Every 6
End of months
2 cycles of chemotherapy Surgical Up to 4 cycles of 6 12 18 24 until end of
Screening Baseline chemotherapy cycle 2 admission* chemotherapy 6 weeks months months months months trial†
Screening log X X                
CT scan X‡     X                
Informed consent   X                    
Demography, medical   X                    

Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892

history, blood test
results
Lung function tests X X§                    
HRQoL   X   X     X X X X X  
Chemotherapy     X     X            
treatment given‡
Surgery and in hospital         X              
postoperative data¶
Adverse events         X   X X X X X  X
Patient-­reported             X X X X X X
resource and health
service use

*Patients allocated to surgery and/or receiving surgery only.

†If not within previous 4 months.
‡Previous CT scan to be used (not to be done again specifically for the trial protocol).
§Only one assessment of lung function is needed, so if this has been done prior to screening, there is no need for another test at baseline.
¶Including resource and health service use.
Open access

5
BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
The relative difference of 30% (HR 0.7) was regarded will be used to ensure similar topic areas are covered
as the minimally important difference for patients and across interviews, while still providing the scope for
clinicians to choose surgery given the risks of the proce- participants to raise issues of pertinence to them.
dure. The figure was chosen by the trial’s patient and ►► Audio-­recording of consultations between healthcare
public involvement (PPI) group. The possibility that staff and potentially eligible patients across centres to
survival could be worse with surgery was also discussed, understand the recruitment process at each centre
and a relative difference of 30% was also regarded as an and to identify and investigate the challenges to
appropriate difference to indicate harm, therefore a two-­ recruitment. The QRI researcher will listen to and
tailed test for superiority was agreed. qualitatively analyse the appointments, documenting
instances such as unclear, insufficient or imbalanced
Patient and public involvement information provision and unintentional transferring
Patient and public representatives were involved from of clinician treatment preferences to patients.
inception and advised on the trial design of MARS 2, the ►► Observation of Trial Management Group (TMG)
identification of the choice of the primary outcome and and investigator meetings to gain an overview of trial
defined the minimally important difference in relative conduct and overarching challenges (logistical issues,
survival. etc).
The study team have continuing engagement with the An account of the anonymised findings from all
Royal Brompton Hospital Cancer Consortia PPI group, the data will be fed back to the Chief Investigator and
which consists of patients and carers who have under- TMG. The data will be used by the QRI team to provide
gone surgery for lung cancer and mesothelioma, to supportive and confidential individual and group feed-
advise on patient-­orientated questions that arise from the back to recruiters to help them to communicate equi-
trial conduct. One patient from the PPI group, a meso- poise, balance treatment options and explain to patients
thelioma survivor, has agreed to sit on the Trial Steering the benefits and purposes of trial participation, while
Committee. The PPI group will also be involved in the optimising informed consent.
dissemination of study results.
Statistical analyses
Integrated QRI
The data will be analysed for randomised patients
Recruitment to RCTs can be challenging,18 particularly
according to intention to treat and follow Consolidated
for surgical trials.19 An integrated QRI will therefore
Standards of Reporting Trials (CONSORT) guidelines.
be employed during the main study phase to optimise
Analyses will be adjusted for site and for design factors
recruitment and retention. The aim of the QRI is to
understand the recruitment process and how it operates included in the cohort minimisation (eg, age, perfor-
in clinical centres, so that sources of recruitment difficul- mance status and cell type).
ties can be identified, and suggestions made to change Survival time and progression-­free survival time from
aspects of design, conduct, organisation or training. randomisation will be compared using survival methods,
A multi-­faceted, flexible approach will be used to inves- allowing for censoring of any participant who is either
tigate site-­specific or wider recruitment obstacles. These alive or lost to follow-­up at the end of the follow-­up period.
will comprise the following: Patient-­reported outcome scores (HRQoL EQ-­ 5D-­
5L
►► Mapping of eligibility and recruitment pathways to and QLQ-­C30) will be compared using a mixed regres-
collate basic data about the levels of eligibility and sion model and adjusted for baseline measures where
recruitment, and identify points at which patients opt appropriate. Changes in treatment effect with time will
in or out of the trial. be assessed by adding a treatment × time interaction to
►► In-­depth, semi-­structured interviews with a purposive the model and comparing models using a likelihood ratio
sample of staff members involved with aspects of trial test. Deaths will be accounted for by modelling survival
design/management and recruitment across centres, and HRQoL jointly. Model fit will be assessed using
and patients eligible for recruitment to the trial. Inter- standard methods and alternative models and/or trans-
views will explore participants’ perspectives of the formations will be explored if appropriate. Treatment
trial, views on the presentation of study information, differences and 95% CIs will be reported.
understanding of trial processes (eg, randomisation), Missing data on patient questionnaires will be dealt
and reasons underlying decisions to accept or decline with according to the scoring manuals. Multiple imputa-
the trial. In addition, interviews with staff and other tion methods will be used if greater than 5% of cases have
individuals involved in the trial will explore perspec- missing data, otherwise complete case analysis will be
tives on the trial design and protocol, views about undertaken. Compliance rates will be reported, including
the evidence on which the trial is based, perceptions the number of participants who have withdrawn from the
of uncertainty/equipoise for themselves and their study, have been lost to follow-­up or died. Causes of death
colleagues, methods for identifying eligible patients, for trial participants will be recorded.
views on eligibility, and examples of actual recruit- Frequencies of adverse events will be described. The
ment successes and difficulties. Interview topic guides proportion of participants experiencing one or more

6 Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892

 Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
serious adverse events in the follow-­ up period will be study analyses, unless the participant expresses a wish for
compared using a generalised linear model. their data to be excluded. Withdrawing patients will be
Two subgroup analyses are planned: (1) comparing asked if they would continue in follow-­up and complete
primary and secondary outcomes by the experience level the requisite questionnaires. Participants who choose to
of the surgical site; (2) comparing the primary outcome withdraw from the study will be treated according to their
by type of mesothelioma (epithelioid, sarcomatoid or hospitals’ standard procedures.
biphasic). An exploratory analysis investigating the effect The findings will be disseminated by usual academic
of surgeon (surgical group only) will be performed for channels, that is, presentation at international meetings
the primary outcome. and peer-­ reviewed publications. A full report for the
No interim analyses are planned. The primary analysis funder will be written on completion of the study and a
will take place when follow-­up is complete for all recruited lay summary of the results provided to patients.
participants.
Major changes to protocol
Economic evaluation Since the first study protocol was approved by the
The economic evaluation will compare the costs and Research and Ethics Committee (the current version is
effects of surgery versus no surgery, following estab- V.6.0, 10 April 2019), the following changes have been
lished guidelines as set out by NICE.20 The within-­trial made:
cost-­
effectiveness analysis will be undertaken from an ►► Qualitative assessment substudy added, as part of the
NHS and personal social services perspective, with a pilot phase only.
time horizon from time of consent to 24 months post-­ ►► The EuroQol EQ-­5D-­5L was added.
randomisation. The primary outcome measure for the ►► Updates to transition from pilot phase to main
economic evaluation will be quality-­ adjusted life years study, including addition of the integrated QRI and
(QALYs), estimated using the EuroQol EQ-­5D-­5L at each economic evaluation, and removal of the collection
follow-­up timepoint.15 16 Resource use data collection will of blood and tissue samples, and one of the disease-­
be integrated into the trial CRFs for chemotherapy cycles specific questionnaires—the EORTC QLQ LC-13.
and surgery (if applicable, this will include details of the ►► Length of follow-­up extended from 2 years until the
surgical procedure, length of stay in hospital by level of end of the study for all participants to ensure that the
care, and postoperative complications) and be collected study has 80% power.
at each follow-­up timepoint. ►► Video-­ recording aspect of the surgical quality assur-
Unit costs will be sought to value resource use data, ance removed as this was deemed impractical by
and the total costs per participant calculated. Responses sites, and it was agreed that it was unnecessary by
to the EQ-­5D-­5L will be assigned valuations according to the Data Safety and Monitoring Committee and the
NICE guidance at the time of analysis,21 and combined Trial Steering Committee, acknowledging the other
with survival to calculate QALYs gained per participant. surgical quality assurance measures that are in place.
Missing resource use and EQ-­5D-­5L data will be handled
using multiple imputation methods.22 From the average Study progress
costs and QALYs gained in each trial group, the incre- Recruitment started in May 2015 and 308 patients have
mental cost-­effectiveness ratio will be derived, producing been randomised so far (correct on 25 May 2020). A total
an incremental cost per QALY gained of surgery of 66 patients from the pilot study are included in this
compared with no surgery. Sensitivity analyses will assess figure. Recruitment will continue until 30 September
the impact of varying key parameters on baseline cost-­ 2020.
effectiveness results. Results will be expressed in terms of The full protocol is available online (https://www.​jour-
a cost-­effectiveness acceptability curve, which indicates nalslibrary.​nihr.​ac.​uk/​programmes/​hta/​1518831/).
the likelihood that surgery is cost-­effective for different
Author affiliations
levels of willingness to pay for health gain. 1
Academic Division of Thoracic Surgery, Royal Brompton and Harefield NHS Trust,
London, UK
Ethics and dissemination 2
The Glenfield Hospital, University Hospitals of Leicester NHS Trust, Leicester, UK
The study intervention is already routinely used in the 3
Mesothelioma UK, Glenfield Hospital, Leicester, UK
4
NHS. The pilot study was managed by Papworth Trials Cardiothoracic Centre, Sheffield Teaching Hospitals NHS Foundation Trust,
Unit Collaboration and the main trial is managed by Sheffield, UK
5
Population Health Sciences, University of Bristol, Bristol, UK
the Bristol Trials Centre Clinical Trials and Evaluation 6
Cancer Research UK Centre Leicester, University Hospitals of Leicester NHS Trust,
Unit and sponsored by Royal Brompton & Harefield Leicester, UK
NHS Foundation Trust. Each participant has the right to 7
Department of Medicine, Royal Marsden NHS Foundation Trust, London, UK
8
withdraw at any time. In addition, the investigator may Department of Thoracic Oncology, Papworth Hospital NHS Foundation Trust,
withdraw the participant from their allocated treatment Cambridge, UK
9
Department of Thoracic Surgery, Barts Health NHS Trust, London, UK
group if a clinical reason for not performing the surgical 10
The Beatson West of Scotland Cancer Centre, Gartnavel General Hospital,
intervention is discovered. If a participant wishes to with- Glasgow, UK
draw, any data already collected will be included in the 11
Thoracic Surgery, Guy’s and Saint Thomas’ Hospitals NHS Trust, London, UK

Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892 7

 Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
12
Respiratory Medicine, South Tyneside NHS Foundation Trust, South Shields, UK Katrin Sainudeen, Helen Morgan; Peterborough City Hospital, North West Anglia
13
Lung Cancer, Aneurin Bevan University Health Board, Newport, UK NHS Foundation Trust—medical site (opened 16/05/2016): Abigail Hollingdale,
14
Oncology, Derby Teaching Hospitals NHS Foundation Trust, Derby, UK Chloe Eddings, Holly Warman; Participating Sites Members: main trial only, Barts
15
Department of Thoracic Surgery, Golden Jubilee National Hospital, Clydebank, UK Health NHS Trust—medical and surgical site (opened 05/06/2017): Jeremy
16
Department of Oncology, New Cross Hospital, Wolverhampton, UK Steele, Jo Hargrave, Resmi Jayachandran, Pratistha Panday, The Beatson West of
17
Oncology, South Tees Hospitals NHS Foundation Trust, Middlesbrough, UK Scotland Cancer Centre; Greater Glasgow Health Board—medical site (opened
18
Academic Respiratory Unit, School of Clinical Sciences, University of Bristol, 14/07/2017): Austin McInnes; Golden Jubilee National Hospital—surgical site
Bristol, UK (opened 14/07/2017): Rocco Bilancia, Julie Buckley, Elizabeth Boyd; North Bristol
19
Respiratory Research Unit, North Bristol NHS Trust, Westbury on Trym, UK NHS Trust—medical site (opened 28/02/2018): Natalie Zahan-­Evans, Anna Morley;
20
Thoracic Surgery Department, Leeds Teaching Hospitals NHS Trust, Leeds, UK Norfolk and Norwich University Hospitals NHS Foundation Trust—medical site
21
Oncology, Colchester Hospital University NHS Foundation Trust, Colchester, UK (opened 12/06/2018): Adela Dann, Eleanor Mishra, Pinelopi Gkogkou; University
22 Hospitals Plymouth NHS Trust—medical site (opened 16/07/2018): Irene Harvey,
Clatterbridge Cancer Centre, Clatterbridge Cancer Centre NHS Foundation Trust,
Hilary Congdon; Barking, Havering and Redbridge University Hospitals NHS
Bebington, UK
23 Trust—medical site (opened 24/07/2018): Alison Ray; Guy’s and St. Thomas’
Plymouth Oncology Centre, University Hospitals Plymouth NHS Trust, Plymouth, UK
24 NHS Foundation Trust—medical and surgical site (opened 10/08/2018): Jehan
Kent Oncology Centre, Maidstone and Tunbridge Wells NHS Trust, Maidstone, UK
25 Mansi, Amy Quinn; Oxford University Hospitals NHS Foundation Trust—medical
Department of Medical Oncology, Barts Health NHS Trust, London, UK
26 site (opened 19/10/2018): Najib M Rahman, Jack Seymour, Hannah Ball, Meenali
The Oncology Care Team, Norfolk and Norwich University Hospital NHS Trust, Chitnis; Maidstone and Tunbridge Wells NHS Trust—medical site (opened
Norwich, UK 19/10/2018): Eirini Petroyannou, Kimberley Snoad, Monica Tavares Barbosa;
27
Medical Oncology, The Christie NHS Foundation Trust, Manchester, UK University Hospitals Birmingham NHS Foundation Trust—medical site (opened
28
Heamatology and Oncology Unit, North West Anglia NHS Foundation Trust, 03/01/2019): Gary Middleton, Philip Earwaker, Haider Abbas, Parminder Sohal;
Peterborough, UK Independent Trial Steering Committee members: Marcus Flather, Paul Beckett and
29
Bristol Trials Centre (CTEU), University of Bristol, Bristol, UK Carol Tan have declared the following competing interest: Ethicon endostaplers—
30
Health Economics Research Centre, Nuffield Department of Population Health, consultancies; Fergus Gleeson, Fergus MacBeth, Mavis Nye, Harvey Pass and
University of Oxford, Oxford, UK Pauline Leonard have declared the following competing interests: Teva Amgen, Tom
Treasure. Unless otherwise stated above, committee members have declared no
Acknowledgements The MARS 2 trial is sponsored by The Royal Brompton and competing interests, Independent Data Monitoring and Safety Committee members:
Harefield NHS Foundation Trust. The sponsor will be responsible for the oversight Linda Sharples, Valerie Rusch, Mark Britton, Robin Rudd, Joseph S Friedberg, Peter
of the MARS 2 study and to ensure that the trial is managed appropriately. We Goldstraw. Unless otherwise stated above, committee members have declared no
want to thank the large teams at each hospital who work on the MARS 2 study competing interests.
(representatives from each are listed below). We would also like to thank Dr Fiona Contributors EL: Study design, preparation and drafting of protocol and
McDonald from the Royal Marsden Hospital, Dr Nagmi Qureshi from Papworth manuscript, Chief Investigator for the trial. LD: Study design, preparation of
Hospital and Professor Simon Padley from the Royal Brompton Hospital for their protocol and review of manuscript. JE: Study design, preparation of protocol
radiotherapy advisory roles. Thank you also to Professor Andrew Nicholson for his and review of manuscript, Principal Investigator and acquired data for the
histopathology advisory role for MARS 2. study. DE: Design of integrated qualitative study, preparation of study protocol,
Collaborators MARS 2 Trialists: Project management team members: Athanasia review of manuscript. DAF: Study design, preparation of protocol and review
Gravani, Holly McKeon, Wendy Underwood, Rachel Brophy, Nicola Farrar; Papworth of manuscript, Principal Investigator and acquired data for the study. SP: Study
Trials Unit Collaboration (pilot study): Victoria Hughes, Jane Elliott, Claire Matthews, design, preparation of protocol and review of manuscript, Principal Investigator
Philip Noyes; Participating Sites Members: pilot study and main trial, University and acquired data for the study. RCR: Study design, preparation of protocol and
Hospitals of Leicester NHS Trust—medical and surgical site (opened 22/04/2015): review of manuscript, Principal Investigator and acquired data for the study. DW:
Apostolos Nakas, Louise Nelson, Sheffield Teaching Hospitals NHS Foundation Study design, preparation of protocol and review of manuscript, acquired data
Trust—medical and surgical site (opened 13/05/2015): Sara Tenconi, Laura Socci, for the study. CA, AB, LF, AI, MK, AK, PK, KL, TM, NAM, RM, DM TP, AR, RS, JS, ZT,
Hilary Wood, Helena Hanratty, Helena Stanley; South Tyneside and Sunderland NHS PT, ST: Review of manuscript, Principal Investigator and acquired data for the
Foundation Trust—medical site (opened 19/06/2015): Judith Moore; Papworth study. KA: Study design, preparation and drafting of protocol and manuscript,
Hospital NHS Foundation Trust—medical site (opened 01/07/2015): Robert Rintoul, oversaw study conduct and acquisition of data. RH: Statistical analysis plan,
Suzanne Miller, Amy Gladwell, Jenny Castedo, Amanda Stone; Colchester Hospital review of manuscript. KJ: Preparation and drafting of manuscript, oversaw
University NHS Foundation Trust—medical site (opened 03/11/2015): Charlotte study conduct and acquisition of data. BW: Drafting of manuscript, oversaw
Ingle, Hayley Hewer; South Tees Hospitals NHS Foundation Trust—medical site study conduct and acquisition of data. NM: Conduct of integrated qualitative
(opened 16/11/2015): Louise Li, Eleanor Aynsley, Andrea Watson, Charlotte Jacobs; study, preparation of study protocol, review of manuscript. EAS: Design of health
The Clatterbridge Cancer Centre NHS Foundation Trust—medical site (opened economic component, preparation of study protocol, review of manuscript. CR:
25/11/2015): Alison Hassall, Masuma Begum; University Hospitals of Derby and Study design, sample size and statistical analysis plan, drafting of protocol,
Burton NHS Foundation Trust—medical site (opened 26/11/2015): Christopher review of manuscript.
Worth, Ellie Piggott, Elizabeth Nadin; Leeds Teaching Hospitals NHS Trust—medical Funding This research is funded by the National Institute for Health Research
site (opened 02/12/2015): Victoria Ashford-­Turner, Matthew Callister, Manchester (NIHR) Health Technology Assessment (HTA) Programme (project number
University NHS Foundation Trust—medical site (opened 21/12/2015): Yvonne 15/188/31). The pilot study was funded by Cancer Research UK and Mesothelioma
Summers, Raffaele Califano, Laura Cove-­Smith, Matt Evison, Maria Blinston, Sara UK has contributed towards patient travel expenses. This study was designed and
Waplington, Amal Ismail, Rachel Chant, Asmita Desai, Juliette Novasio, Marie delivered in collaboration with the Bristol Trials Centre Clinical Trials and Evaluation
Kirwan; The Royal Wolverhampton NHS Trust—medical site (opened 04/01/2016): Unit (CTEU), a UK Clinical Research Collaboration (UKCRC) registered clinical trials
Ian Morgan, Victoria Lake, Nichola Harris; Royal Gwent Hospital, Aneurin Bevan unit which, as part of the Bristol Trials Centre (BTC), is in receipt of NIHR Clinical
University Health Board—medical site (opened 08/02/2016): Simon Hodge; The Trials Unit (CTU) support funding. DE was supported by the NIHR Biomedical
Royal Marsden NHS Foundation Trust—medical site (opened 08/04/2016), Chelsea Research Centre at University Hospitals Bristol NHS Foundation Trust and the
sub-­site: Nadia Yousaf, Nadza Tokaca, Adam Januszewski, Avani Athauda, Anisha University of Bristol. SP acknowledges NHS funding to the Royal Marsden Hospital/
Ramessur, Emily Grist, Niamh Colman, Michael Flynn, Joan Joyce, Sarah Vaughan, Institute of Cancer Research NIHR Biomedical Research Centre.
Maria Piga, Derya Sahin, Agnieszka Yongue, Emma Turay, Sutton sub-­site: Mary
Disclaimer The views and opinions expressed therein are those of the authors
O’Brien, Jaishree Bhosle, Rajiv Kumar, Charlotte Milner-­Watts, Jessica Brown,
and do not necessarily reflect those of the HTA programme, NIHR, NHS or the
David Walder, Alexandros Georgiou, Xiaorong Wu, Naila Kaudeer, Kroopa Joshi,
Department of Health and Social Care. The funder and sponsor approve any
Michael Davidson, Shelize Khakoo, Bee Ayite, Kathryn Priest, James Dobbyn,
amendments to the study but have no direct involvement in study design;
Vasanthi Prathapan, Deborah McCrimmon, Natalie Ash, Alison Norton, Bianca
collection, management, analysis and interpretation of data; writing of the report;
Peet, Libby Hennessy, Rosemary Johnson, Laura White; Kingston sub-­site: Edward
and the decision to submit this report for publication.
Armstrong, Maria Coakley, Scott Shepherd, Narda Chaabouni, Katherina Sreter,
Vasileios Angelis, Mariko Morishita, Jose Roca, Mary Jane de los Reyes Lauigan, Competing interests None declared.

8 Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892

 Open access

BMJ Open: first published as 10.1136/bmjopen-2020-038892 on 1 September 2020. Downloaded from http://bmjopen.bmj.com/ on October 1, 2023 by guest. Protected by copyright.
Patient and public involvement Patients and/or the public were involved in the pleurodesis in patients with malignant pleural mesothelioma
design, conduct, reporting and dissemination plans of this research. Refer to the (MesoVATS): an open-­label, randomised, controlled trial. Lancet
Methods section for further details. 2014;384:1118–27.
9 Cao C, Tian DH, Pataky KA, et al. Systematic review of pleurectomy
Patient consent for publication Not required. in the treatment of malignant pleural mesothelioma. Lung Cancer
2013;81:319–27.
Provenance and peer review Not commissioned; externally peer reviewed.
10 Lang-­Lazdunski L, Bille A, Lal R, et al. Pleurectomy/decortication
Open access This is an open access article distributed in accordance with the is superior to extrapleural pneumonectomy in the multimodality
Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits management of patients with malignant pleural mesothelioma. J
others to copy, redistribute, remix, transform and build upon this work for any Thorac Oncol 2012;7:737–43.
purpose, provided the original work is properly cited, a link to the licence is given, 11 Teh E, Fiorentino F, Tan C, et al. A systematic review of lung-­
sparing extirpative surgery for pleural mesothelioma. J R Soc Med
and indication of whether changes were made. See: https://​creativecommons.​org/​
2011;104:69–80.
licenses/​by/​4.0​ /. 12 Donovan JL, Rooshenas L, Jepson M, et al. Optimising recruitment
and informed consent in randomised controlled trials: the
ORCID iDs development and implementation of the quintet recruitment
Barbara Warnes http://​orcid.​org/0​ 000-​0002-​1326-​0448 intervention (QRI). Trials 2016;17:283.
Elizabeth A Stokes http://o​ rcid.​org/0​ 000-​0002-​4179-​1369 13 Rooshenas L, Paramasivan S, Jepson M, et al. Intensive triangulation
of qualitative research and quantitative data to improve recruitment
to randomized trials: the quintet approach. Qual Health Res
2019;29:672–9.
REFERENCES 14 Rooshenas L, Scott LJ, Blazeby JM, et al. The quintet recruitment
1 Health and Safety Executive. Asbestos-­related disease statistics in intervention supported five randomized trials to recruit to target: a
Great Britain, 2019, 2019. https://www.​hse.​gov.​uk/​statistics/​causdis/​ mixed-­methods evaluation. J Clin Epidemiol 2019;106:108–20.
asbestos-​related-​disease.​pdf 15 EuroQol Group. EuroQol—a new facility for the measurement of
2 Peto J, Hodgson JT, Matthews FE, et al. Continuing increase in health-­related quality of life. Health Policy 1990;16:199–208.
mesothelioma mortality in Britain. Lancet 1995;345:535–9. 16 Herdman M, Gudex C, Lloyd A, et al. Development and preliminary
3 National Institute of Health and Care Excellence. Pemetrexed for the testing of the new five-­level version of EQ-­5D (EQ-­5D-­5L). Qual Life
treatment of malignant pleural mesothelioma. Technology appraisal Res 2011;20:1727–36.
guidance [TA135], 2008. 17 Utley M, Fiorentino F, Treasure T. Obtaining an upper estimate of
4 Vogelzang NJ, Rusthoven JJ, Symanowski J, et al. Phase III study the survival benefit associated with surgery for mesothelioma. Eur J
of pemetrexed in combination with cisplatin versus cisplatin alone Cardiothorac Surg 2010;38:241–4.
in patients with malignant pleural mesothelioma. J Clin Oncol 18 Briel M, Olu KK, von Elm E, et al. A systematic review of discontinued
2003;21:2636–44. trials suggested that most reasons for recruitment failure were
5 Royal College of Physicians. National lung cancer audit pleural preventable. J Clin Epidemiol 2016;80:8–15.
mesothelioma report 2016 (for the audit period 2014). London: Royal 19 Blencowe NS, Cook JA, Pinkney T, et al. Delivering successful
College of Physicians, 2016. randomized controlled trials in surgery: methods to optimize
6 Rice D, Rusch V, Pass H, et al. Recommendations for uniform collaboration and study design. Clin Trials 2017;14:211–8.
definitions of surgical techniques for malignant pleural mesothelioma: 20 National Institute for Health and Care Excellence. Guide to the
a consensus report of the International Association for the Study of methods of technology appraisal 2013. London: National Institute for
Lung Cancer International Staging Committee and the International Health and Care Excellence, 2013.
Mesothelioma Interest Group. J Thorac Oncol 2011;6:1304–12. 21 National Institute for Health and Care Excellence. National Institute
7 Treasure T, Lang-­Lazdunski L, Waller D, et al. Extra-­pleural for Health and Care Excellence. Position statement on use of the
pneumonectomy versus no extra-­pleural pneumonectomy for EQ-­5D-­5L valuation set for England. Available: https://www.​nice.​org.​
patients with malignant pleural mesothelioma: clinical outcomes of uk/​about/​what-​we-​do/​our-​programmes/​nice-​guidance/​technology-​
the mesothelioma and radical surgery (MARS) randomised feasibility appraisal-​guidance/​eq-​5d-​5l [Accessed 18 Dec 2019].
study. Lancet Oncol 2011;12:763–72. 22 White IR, Royston P, Wood AM. Multiple imputation using
8 Rintoul RC, Ritchie AJ, Edwards JG, et al. Efficacy and cost of chained equations: issues and guidance for practice. Stat Med
video-­assisted thoracoscopic partial pleurectomy versus talc 2011;30:377–99.

Lim E, et al. BMJ Open 2020;10:e038892. doi:10.1136/bmjopen-2020-038892 9