Available via license: CC BY 4.0
Content may be subject to copyright.
ADD-ASPIRIN: A phase III, double-blind, placebo controlled, randomised
trial assessing the effects of aspirin on disease recurrence and survival
after primary therapy in common non-metastatic solid tumours
Christopher Coyle
a
, Fay H. Cafferty
a
, Samuel Rowley
a
, Mairead MacKenzie
b
, Lindy Berkman
c
, Sudeep Gupta
d
,
C S Pramesh
e
,DuncanGilbert
a,f
, Howard Kynaston
g
, David Cameron
h
, Richard H. Wilson
i
, Alistair Ring
j
,
Ruth E. Langley
a,
⁎, on behalf of the, Add-Aspirin investigators
1
:
a
MRC Clinical Trials Unit, UCL, Aviation House, 125 Kingsway, London WC2B 6NH, UK
b
Independent Cancer Patient Voices, 17 Woodbridge Street, London EC1R 0LL, UK
c
NCRI Consumer Forum, Angel Building, 407 St John Street, London EC1V 4AD, UK
d
Room No. 1109, 11th Floor, Homi Bhabha Block, Tata Memorial Centre/Hospital, Parel, Mumbai 400012, India
e
Department of Surgical Oncology, Tata Memorial Centre, Dr ErnestBorges Marg, Parel, Mumbai 400012, India
f
Sussex Cancer Centre, Royal Sussex County Hospital, Eastern Road, Brighton, Sussex BN2 5BE, UK
g
Room 2F65, Block A2, Cardiff School of Medicine, Heath Park, Cardiff CF14 4XN, UK
h
Edinburgh Cancer Research Centre, University of Edinburgh, Western General Hospital, Crewe Road South, EH4 2XR, UK
i
Centre for Cancer Research and Cell Biology, Queen's University Belfast, 97 Lisburn Road, Belfast BT9 7AE, Northern Ireland, UK
j
The Royal Marsden NHS Foundation Trust, Downs Road, Sutton, Surrey SM2 5PT, UK
abstractarticle info
Article history:
Received 22 June 2016
Received in revised form 10 October 2016
Accepted 17 October 2016
Available online 21 October 2016
Background: There is a considerable body of pre-clinical,epidemiological and randomised data to support the hy-
pothesis that aspirin has the potential to be an effective adjuvant cancer therapy.
Methods: Add-Aspirin is a phase III, multi-centre, double-blind, placebo-controlled randomised trial with four
parallel cohorts. Patients who have undergone potentially curative treatment for breast (n= 3100), colorectal
(n= 2600), gastro-oesophageal (n= 2100) or prostate cancer (n= 2120) are registered into four tumour spe-
cific cohorts. All cohorts recruit in the United Kingdom, with the breast and gastro-oesophageal cohort also
recruiting in India. Eligible participants first undertake an active run-in period where 100 mg aspirin is taken
daily for approximately eight weeks. Participants who are able to adhere and tolerate aspirin then undergo a
double-blind randomisation and are allocated in a 1:1:1 ratio to either 100 mg aspirin, 300 mg aspirin or a
matched placebo to be taken daily for at least five years. Those participants ≥75 years old are only randomised
to 100 mg aspirin or placebo due to increased toxicity risk.
Results:Theprimaryoutcomemeasuresareinvasive disease-free survival for the breast cohor t, disease-free survival for
the colorectal cohort, overall survival for the gastro-oesophageal cohort, and biochemical recurrence-free survival for
the prostate cohort, with a co-primary outcome of overall survival across all cohorts. Secondary outcomes include ad-
herence, toxicity including serious haemorrhage, cardiovascular events and some cohort specificmeasures.
Conclusions: The Add-Aspirin trial investigates whether regular aspirin use after standard therapy prevents recurrence
and prolongs survival in participants with four non-metastatic common solid tumours.
© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
Keywords:
Aspirin
Breast cancer
Colorectal cancer
Gastro-oesophageal cancer
Prostate cancer
Randomised controlled trial
1. Background and introduction
1.1. Rationale
The Add-Aspirin trial includes participants with breast, colorectal,
gastro-oesophageal and prostate tumours which, together, accounts
for approximately one third of all cancer cases and cancer deaths [1].
The selected disease sites are those for which (i) the evidence relating
to potential benefit of aspirin is strongest; (ii) the potential impact is
large (common cancers with large numbers of cases diagnosed at an
early stage, or where outcomes of curative treatment are particularly
poor); and (iii) recruitment is feasible. As a low-cost pharmaceutical,
feasible to administer in both resource poor and rich countries, aspirin
has the potential to significantly impact on cancer outcomes worldwide.
This, combined with other possible health benefits (such as cardiovas-
cular effects), means that aspirin warrants further investigation as an
anti-cancer agent.
Contemporary Clinical Trials 51 (2016) 56–64
⁎Corresponding author.
E-mail address: ruth.langley@ucl.ac.uk (R.E. Langley).
1
The Add-Aspirin investigators are detailed in Appendix A.
http://dx.doi.org/10.1016/j.cct.2016.10.004
1551-7144/© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Contents lists available at ScienceDirect
Contemporary Clinical Trials
journal homepage: www.elsevier.com/locate/conclintrial
1.2. Supporting evidence
There have been well over 100 case-control and cohort studies in-
vestigating the use of aspirin and cancer risk [2]. A meta-analysis of
such studies showed that aspirin use resulted in significant reductions
in the risk of developing cancer, most notably in colorectal (relative
risk (RR) 0.73, 95% confidence-interval (CI) 0.67–0.79), gastric (RR
0.67, CI 0.54–0.83), adenocarcinoma of the oesophagus/cardia (RR
0.67, CI 0.54–0.83), squamous cell carcinoma of the oesophagus (RR
0.64, CI 0.52–0.78), breast (RR 0.90, CI 0.85–0.95), and prostate cancer
(RR 0.90, CI 0.85–0.98) [2]. Observational studies have also shown im-
provements in survival with aspirin use after a diagnosis of breast
[3–5],colorectal[6–11], gastro-oesophageal [12,13] and prostate cancer
[14–16].
Randomised data is available to indirectly substantiate these obser-
vations. A meta-analysis of individual participant data on cancer inci-
dence in randomised trials (designed to investigate the effect of
aspirin on vascular disease) show marked reductions in cancer inci-
dence and cancer mortality associated with regular aspirin use (greater
than three years) in both the short and long-term [17–20]. Similarly,
long-term follow-up from the Women's Health Study, a randomised
placebo-controlled trial designed to assess the effects of aspirin
(100 mg on alternate days) in the primary prevention of cardiovascular
disease and cancer, showed that allocation to aspirin reduced the inci-
dence of colorectal cancer with ten years of follow-up (hazard ratio
(HR) 0.80, CI 0.67–0.97) [21]. A recent meta-analysis of studies examin-
ing aspirin use after a cancer diagnosis has shown a significant reduc-
tion in cancer-specific mortality in colon cancer, but not in breast and
prostate cancer, however significant heterogeneity between studies
was identified [22].
Furthermore, the first randomised trial specifically designed to dem-
onstrate that aspirin can prevent the development of cancer has shown
that 600 mg of aspirin daily for up to four years prevents colorectal and
other cancers associated with Lynch syndrome (a hereditary condition
which predisposes to cancer development) HR 0.45, CI 0.26–0.79 [23].
The potential benefits of aspirin have to be weighed against the risk
of adverse effects. A number of systematic reviews and meta-analyses
have examined the potential risks of adverse events [24,25]. A recent re-
view estimates that, depending on age and sex, regular aspirin use over
a 15-year period would lead to an absolute increase in major bleeding
events of between 0.16% and 0.81%. The authors conclude that prophy-
lactic aspirin use for a minimum of five years ata dose of 75 mg–325 mg
daily has a favourable benefit-harm profile [26]. For individuals treated
for cancer, with a high risk of recurrent disease, the balance could be
even more favourable.
2. Methods
2.1. Aims
The Add-Aspirin trial aims to assess whether regular aspirin use after
standard potentially curative primary therapy can prevent recurrence and
prolong survival in individuals with four common early stage solid tu-
mours. Avoiding recurrent disease, subsequent treatment and the associat-
ed morbidity and mortality in these individuals is an important goal.
Multicentre and international recruitment will allow assessment of the in-
tervention in a range of settings, with the aim of demonstrating that imple-
mentation is both feasible and cost-effective across varying health care
systems and in both the developing and developed world. A secondary
aimistoassessthepotentialoverallhealthbenefits of aspirin for these in-
dividuals including cardiovascular outcomes.
2.2. Overview of design
The Add-Aspirin trial investigates the use of both 100 mg daily and
300 mg daily aspirin compared with matched placebo (double-blind)
in each of four different tumour types, utilising an overarching protocol.
Further details of the rationale for this design are provided in the discus-
sion section of this article. Fig. 1 shows a summary schema for the trial.
2.3. Participants
Participants entering the Add-Aspirin trial have undergone poten-
tially curative treatment (surgery or other radical treatment) for breast,
colorectal, gastro-oesophageal or prostate cancer with standard neoad-
juvant and/or adjuvant therapy if indicated, and may also have partici-
pated in any pre-approved trials and satisfy the eligibility criteria,
summarised in Fig. 2.
2.4. Registration
The Add-Aspirin trial is open to centres in every Cancer Research
Network (CRN) throughout the four devolved nations of the United
Kingdom (UK) and will also recruit participants in India (other coun-
tries may join subsequently). Eligible participants who have provided
consent and meet the timing of entry criteria are registered online
(through the trial website, www.addaspirintrial.org). The timing of
entry window has been designed so that aspirin can be started at the
earliest opportunity to maximise the potential benefits, whilst starting
at a time when it is considered safe to doso and unlikely to compromise
the curative intent of standard primary treatment. Figures describing
the timing of entry criteria for each cohort are available in Appendix B.
2.5. Run-in period
The Add-Aspirin trial incorporates a feasibility phase lasting approx-
imately 2 1/2 years during which recruitment feasibility, treatment ad-
herence and safety will be assessed. During the feasibility phase of the
study, all participants are required to complete an active run-in period
after registrationbut prior to randomisation where they take 100 mg as-
pirin daily (one tablet per day) in an open-label manner for a period of
approximately eight weeks.
2.6. End of run-in period assessment
At the end of the run-in period, the participant's tolerance of aspirin
and adherence to daily treatmentwill be assessed. This approach allows
those individuals who are unlikely to be able to tolerate aspirin, as well
as those whoare unlikely to be able to adhere to the protocol treatment
schedule, to be identified. Adherence will be assessed using a combina-
tion of a participant diary card, used blister packs and patient reported
adherence. Participants will be suitable for randomisation if they have
taken at least 80% of their run-in treatment and have not experienced
any aspirin-related severe toxicity (defined as ≥grade 3 CTCAE v4),
nor any grade of gastrointestinal bleeding, active gastrointestinal ulcer-
ation, new or worsening tinnitus, macular degeneration, intracranial
bleeding or hypersensitivity to aspirin. If the investigator feels that the
reason for inadequate adherenceis temporary (for example, due to tox-
icity resulting from concomitant adjuvant treatment which has subse-
quently finished or a non-recurrent unrelated event), the run-in
period may be extended by four or eight weeks to reassess adherence
and toxicity subject to agreement from the central trial team. Those par-
ticipants identified as suitable for further study participation, and who
remain eligible and are willing to continue in the trial then re-confirm
their consent to participate before being randomised.
2.7. Randomisation
Following assessment at the end of the run-in period, eligible
participants in the UK are randomised by phone and, in India, via the
trial website. Participants undergo a double-blind randomisation.
Randomisation is performed separately within each tumour-specific
57C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
cohort and uses minimization algorithms based on key prognostic fac-
tors (dependent on tumour site), incorporating a random element.
Within each tumour-specific cohort, participants who are below
75 years old are allocated in a 1:1:1 ratio to either 100 mg aspirin,
300 mg aspirin or a matched placebo. Participants who are 75 years
old or over, are only allocated to either 100 mg aspirin or matched pla-
cebo as toxicity is thought to increase with age but the allocation ratio of
2:1 remains so that they have the same chance of receiving active
FOUR PARALLEL
PHASE III COHORTS
BREAST
COHORT
GASTRO-
OESOPHAGEAL
COHORT
PROSTATE
COHORT
300mg
ASPIRIN
FOLLOW-UP
≥ 5 years, including active follow-up largely aligned with standard care, and long term passive follow-up in the UK
Primary Outcome:
DFS
2600 participants
Primary Outcome:
Invasive DFS
3100 participants
Primary Outcome:
OS
2100 participants
Primary Outcome:
Biochemical RFS
2120 participants
100mg
ASPIRIN
PLACEBO
Run-in
(100mg Aspirin
for 8 weeks )
300mg
ASPIRIN
100mg
ASPIRIN
PLACEBO
Run-in
(100mg Aspirin
for 8 weeks )
300mg
ASPIRIN
100mg
ASPIRIN
PLACEBO
Run-in
(100mg Aspirin
for 8 weeks )
300mg
ASPIRIN
100mg
ASPIRIN
PLACEBO
COLORECTAL
COHORT
Run-in
(100mg Aspirin
for 8 weeks )
Fig. 1. Add-Aspirin trial schema. DFS = Disease free survival, OS = overall survival, RFS = recurrence free survival.
Breast cohort main inclusion criteria
Histologically confirmed invasive breast cancer.
Node positive or node negative with high risk features.
Surgery (R0 resection) with standard. neoadjuvant
and/or adjuvant therapy where indicated.
Known HER2 and ER status.
Gastro-oesophageal cohort main inclusion criteria
Oesophageal or gastric adenocarcinoma or squamous
cell carcinoma.
Previous therapy with curative intent:
Either: Surgery (R0 resection) with standard
neo-adjuvant +/- adjuvant therapy,
or Primary chemoradiotherapy
Proton pump inhibitor is mandated for patients
undergoing partial gastrectomy or oesophagectomy.
Colorectal cohort main inclusion criteria
Stage II or III colon or rectum adenocarcinoma.
(Patients with resected liver metastases are eligible*).
Surgery (R0 resection) with standard neo-adjuvant
/adjuvant therapy where indicated.
CEA ≤1.5 X upper limit of normal.
Prostate cohort main inclusion criteria
Histologically confirmed non-metastatic prostate
adenocarcinoma (T1-3a, N0).
Intermediate or high risk as per D’Amico classification
Previous therapy with curative intent:
Either: Radical radiotherapy,
or Radical prostatectomy (+/- adjuvant radiotherapy)
or Salvage radiotherapy after radical prostatectomy
Common exclusion criteria
No clinical or radiological evidence of residual or distant disease.
(*Patients with colorectal cancer who have undergone resection of liver metastases with clear margins are eligible).
No current, regular use of aspirin, NSAIDs or anti-coagulants.
No pre-disposition to aspirin toxicity (e.g. active ulceration).
For the full eligibility criteria please see the protocol or www.addaspirintrial.org.
Fig. 2. Summary of eligibility criteria.
58 C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
treatment as the other participants. The target randomisation figure is
9920 participants in the United Kingdom (UK) and India combined. As-
suming that approximately 10% of participants will not be randomised
following the run-in (for reasons relating either to toxicity or adher-
ence), it is expected that 11,000 participants will be registered to
begin the run-in period.
2.8. Follow-up
Patientsare followed up at three-monthly intervals initially and then
six-monthly. Adherence to treatment is verbally assessed at every
follow-up visit. In the UK, trial treatment, and active follow-up, con-
tinues for at least five years after randomisation. Long-term passive
follow-up data will be obtained from routinely-collected healthcare da-
tabases for at least ten further years. Indian participants will be actively
followed-up for at least ten years after randomisation. For participants
that are registered but do not go on to be randomised, active participa-
tion in the trial will end at that time. However, passive follow-up will
continue via routinely-collected healthcare datasets where consent for
this has been obtained. The trial assessment schedule for each cohort
is aligned with standard practice where possible to ensure they can be
implemented easily. This is balanced with the need to ensure appropri-
ate monitoring of patients on trial treatment and assessment of out-
come measures. The trial follow-up schedules are available in
Appendix C.
2.9. Toxicity management
Participants that experience any aspirin-related severe toxicity (de-
fined as ≥grade 3 Common Terminology Criteria for Adverse Events
(CTCAE v4)) or any grade of gastrointestinal bleeding, active gastroin-
testinal ulceration,tinnitus, macular degeneration, intracranial bleeding
or hypersensitivity to aspirin are required to permanently discontinue
aspirin immediately.
For those who are asymptomatic, prophylactic measures to reduce
the risk of gastrointestinal toxicity from aspirin (such as proton pump
inhibitor (PPI) prophylaxis and helicobacter pylori eradication) are not
routinely recommended in participants at low risk of gastrointestinal
complications and so are not mandated in the Add-Aspirin trial proto-
col. However, PPI use for the duration of aspirin treatment is recom-
mended for patients who have undergone oesophagectomy or partial
gastrectomy and should also be considered for older patients (≥75
years), or any other participant who might be at increased risk of toxic-
ity. Intracranial bleeding is a rare toxicity of aspirin, and hypertension
can increase the risk. Those with poorly controlled hypertension have
trial treatment withheld until their bloodpressure iscontrolled. Further
guidelines are available in the trial protocol.
Investigators are advised to manage toxicities under the assumption
that the participant is receiving the highest possible dose of the active
product (300 mg aspirin), without the need for unblinding, however
where knowledge of treatment allocation would alter clinical manage-
ment, unblinding is possible. Unblinding can be performed via an
access-controlled system available through the trial website (www.
addaspirintrial.org).
2.10. Sub-studies
The size and diversity of the Add-Aspirin cohort provides opportuni-
ty to address other secondary research questions and evaluate novel
methodology. Aspirin has been proposed to have a number of health
benefits beyond cancer, particularly in older people. To investigate the
overall health benefits of aspirin, functional capacity is assessed using
the Vulnerable Elders Survey (VES-13) [27].This is performed for partic-
ipants that are 65 years old or over at trial registration, and five years
after randomisation and can be administered in person or over the tele-
phone. The hypothesis that aspirin protects against cognitive decline is
assessed using a short version of the Montreal Cognitive Assessment
(the MoCA-blind). The MoCA-blind takes approximately 7 to 10 min
to complete and is administered in all Add-Aspirin trial participants at
registration, then again at one and five years after randomisation. No
training is required to administer the questionnaire, which can be con-
ducted in person, or over the telephone. A methodological sub-study
will compare the quality and completeness of routinely-collected
healthcare data with data collected within the trial, with the aim of
assessing the suitability of passive follow-up data collection for investi-
gating long-term primary and secondary outcome measures within the
trial. This will be an early validation that will determine ongoing use of
routinely collecteddata in the trial. A sub-study is planned to investigate
methods of measuring adherence, including the collection of urine sam-
ples to measure thromboxane B2 (a direct measure of the effects of as-
pirin). Methodological sub-studies to improve site initiation and
recruitment will also be undertaken.
2.11. Translational objectives
The Add-Aspirin trial incorporates a sample repository where a
baseline blood sample and tumour sample are stored for future transla-
tional projects. The sample repository is jointly hosted by two institu-
tions in the UK, Tayside Tissue Bank and the Wales Cancer Bank. In
India, a baseline blood sample and tumour sample from selected sites
will be stored at the Tata Memorial Centre biobank. A number of studies
are expected to be initiated whilst the trial is ongoing (subject to
funding), including studies to identify groups that will benefitmost
from aspirin use (for example, investigation of the role of tumour
PIK3CA mutation status), and to investigate the mechanisms underlying
the anti-cancer effects of aspirin, particularly effects on platelet function
and the pro-thrombotic tumour microenvironment.
3. Results
3.1. Outcomes
Tumour site-specific primary analyses will take place 5–6 years after
recruitment of the last participant for that cohort, with the exact timing
based on the observed numbers of events. Primary and secondary out-
come measures are available in Table 1. Overall survival is a secondary
outcome measure in all cohorts except the gastro-oesophageal cohort,
where it is the primary outcome. Overall survival will also be assessed
as a co-primary outcome measure in all participants after 15 years.
The longer follow-up and large sample size associated with this analysis
will enable any long-term benefits of aspirin to be realised, including
those unrelated to theprimary cancer, for example the potential for pre-
vention of deaths related to vascular events and second malignancies.
Consideration of rates of serious toxicity(and particularlyserious haem-
orrhage), as well as other secondary health outcomes, alongside the ef-
ficacy results will be particularly important in these analyses in order to
provide an holistic assessment of the potential risks and benefits associ-
ated with different doses.
3.2. Statistical considerations
Primary analyses will compare outcomes for participants allocated to
aspirin (100 mg and 300 mg arms combined) and participants allocated
to placebo, regardless of the treatment received (i.e. intention-to-treat).
The primary analyses will include both those participants b75 years
who underwent the full randomisation and those ≥75 years who
underwent randomisation between 100 mg aspirin or placebo only, but
the dose effects of aspirin will be investigated only on those randomised
between the two doses.
If an overall effect of aspirin vs. placebo is observed in the primary
treatment comparison for one or more cohorts, a further analysis will
be performed to investigate differences in efficacy according to aspirin
59C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
dose. This analysis will be performed only in the cohorts that show a
positive result for aspirin vs. placebo and will be stratified by cohort.
By making these analyses conditional on a benefit of aspirin being ob-
served in the primary analysis, the likelihood of a false-positive result
is reduced. The rationale for combining the data across cohorts is to
maximise power, as we anticipate that any difference between doses
of aspirin will be smaller than thedifference between aspirin and place-
bo. The trial has also established collaborative links with research
groups running other aspirin cancer trials internationally with a view
to future meta-analyses.
3.3. Sample size breast cohort
Based on data from recent trials, we expect that five-year invasive
disease-free survival (IDFS) in the control group will be approximately
80% [29–32]. 717 IDFS events will be required to achieve 90% power
to detect a 4% (HR = 0.78) improvement in this rate. Assuming that
the cohort takes 3 1/2 years to recruit, with analysis six years later, we
anticipate that 3100 participants will be required to observe this num-
ber of events.
3.4. Sample size colorectal cohort
Based on data from recent trials, we expect that five-year disease-
free survival (DFS) in this cohort will be approximately 70% [33,34].
899 DFS events will be required to achieve 90% power to detect a 5%
(HR = 0.80) improvement in this rate. Assuming that the cohort takes
3 1/2 years to recruit, with analysis six years later, we anticipate that
2600 participants will be required to observe this number of events.
3.5. Sample size gastro-oesophageal cohort
Based on data from recent trials, we expect thatfive-year overall sur-
vival in this cohort will be approximately 45% [35–39]. 1120 deaths will
be required to achieve 80% power to detect a 6% (HR = 0.84)
improvement in this rate. Assuming that thecohort takes six years to re-
cruit, with analysis five years later, we anticipate that 2100 participants
will be required to observe this number of events.
3.6. Sample size prostate cohort
The radical prostatectomy and radical radiotherapy groups are
powered to assess effects separately. In the radical prostatectomy
group, we anticipate that biochemical recurrence-free survival (bRFS)
at five years will be approximately 75% [40].For the radical radiotherapy
group, five year bRFS is estimated to be approximately 65% [41].To
achieve 90% power to detect an 8% improvement in these rates, 673
bRFS events will be required. Assuming that the cohort takes five
years to recruit, with analysis five years later, we anticipate that 2120
participants will be required to observe this number of events.
Sample size calculations for all cohorts are based on a two-sided 5%
significance level and account for a degree of loss to follow-up and
slower recruitment in the early stages of the trial. Target registrations
have been inflated to allow a 10% dropout after the run-in period.
3.7. Ethical considerations
The trial will be conducted in compliance with the approved proto-
col, the Declaration of Helsinki 2008, the principles of Good Clinical
Practice (GCP), Commission Directive 2005/28/EC with the implemen-
tation in national legislation in the UK by Statutory Instrument 2004/
1031 and subsequent amendments, the UK Data Protection Act (DPA
number: Z5886415), and the National Health Service (NHS) Research
Governance Framework for Health and Social Care (RGF). International
centres will comply with the principles of GCP as laid down by the ICH
topic E6 (Note for Guidance on GCP) and applicable national regula-
tions. The Add-Aspirin trial is registered with the International Standard
Randomised Controlled Trial Number ISRCTN74358648, and has also
been submitted for registration with the Clinical trials Registry of India
(REF/2016/06/011465). The Add-Aspirin trial was approved by the
South Central –Oxford C research ethics committee and is part of the
UK National Cancer Research Network (NCRN) portfolio. In India, the
trial has been approved by the Directorate of the National Cancer Grid
(
NCG),andispartoftheNCGtrialsportfolio.UniversityCollege
London (UCL) and the Tata Memorial Centre (TMC) are co-sponsors of
the trial and have delegated responsibility for the overall management
of the Add-Aspirin trial to the MRC CTU at UCL and Tata Memorial Cen-
tre CTU for India.
4. Discussion
The Clinical Trials Authorisation for the Add-Aspirin trial was
granted by the Medicines and Healthcare products Regulatory Agency
(MHRA) on 25th November 2014, and ethics approval was given on
4th June 2014. The first participant was recruited on 8th October
2015. At the time of writing on 1st May 2016, the Add-Aspirin trial is
recruiting across 120 centres in the UK, and is recruiting ahead of its
overall projected targets, with the breast cohort the fastest recruiter.
The last of the four cohorts is now expected to complete recruitment
in 2021. Current recruitment figures are available at www.
addaspirintrial.org.
4.1. Challenges and methodological solutions
There are a number of practical and operational challenges present-
ed by a large adjuvant trial of a generic and repurposed intervention.
These include the need for cost efficiencies due to a lack of industry fi-
nancial support for a trial of a generic pharmaceutical; ensuring suffi-
cient long-term adherence in a largely asymptomatic population; and
the potential for control arm contamination due to over the counter
Table 1
Outcome measures.
Cohort Primary outcome measures
Breast cancer Invasive disease-free survival (IDFS) [28]
Colorectal cancer Disease-free survival (DFS)
Gastro-oesophageal
cancer
Overall survival
Prostate cancer Biochemical recurrence-free survival (bRFS)
All cohorts combined Overall survival
Cohort Secondary outcome measures
All cohorts Overall survival (except for gastro-oesophageal cohort)
Adherence
Toxicity
Serious haemorrhage CTCAE (v4) grade 3 or greater
Serious vascular events
Thrombotic events
Diabetes and associated complications
Second malignancies
Age-related macular degeneration
Cognitive assessment (using the MOCA-blind
questionnaire)
Dementia
Comorbidities (using the Charlson Index)
Obesity (using the Body Mass Index)
Functional capacity (using the VES-13 questionnaire)
Breast Breast cancer-specific survival
Bone metastases-free survival
Invasive disease-free survival-ductal carcinoma insitu
(IDFS-DCIS)
Colorectal Colorectal cancer-specific survival
Gastro-oesophageal Disease-free survival
Prostate Prostate cancer-specific survival
Time to initiation of salvage treatment
Bone metastases-free survival
60 C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
(OTC) availability of aspirin. These have been addressed using a number
of contemporary methodological approaches.
4.2. Overarching protocol
A platform trial design was chosen because there is evidence that as-
pirin is potentially effectivein multiple tumours. Investigating the use of
both 100 mg daily and 300 mg daily aspirin across cohorts addresses
uncertainty surrounding the optimal aspirin dose required to achieve
anti-cancer effects, potentially saving many years of research time.
Keeping all four cohorts within a single protocol ensures that the man-
agement o f each cohort is a s comparable as possible (with the excep tion
of some site specific procedures). This allows a combined analysis of
overall survival as a co-primary outcome measure and cross cohort sec-
ondary analyses of toxicity, cardiovascular and other health benefits,
thus increasing the overall potential impact of the trial. In addition,
this design provides the capacity to add further tumour sites and also
provides a potential platform for evaluation of other repurposed agents.
An overarching protocol provides economies of scale both centrally and
site level, includingsite set-up, regulatory approval, central staffing, co-
ordination, oversight and data management. The resulting cost efficien-
cies improve the financial viability of the trial given the lack of industry
support, and provide value for money for our charitable and govern-
mental funders.
4.3. Antecedent aspirin use
A proportion of individuals who are otherwise eligible for the trial
will already be taking aspirin regularly. It is conceivable that pre-
existing aspirin use could alter tumour biology [42], and that those al-
ready taking aspirin might be randomised to placebo which would be
unethical. Consequently a decision was made to exclude current or pre-
vious regular aspirin users.
4.4. Over the counter aspirin
Unlike most other cancer trials, the intervention in the Add-Aspirin
can be purchased without a prescription. There is a risk that some po-
tential participants might opt to not enter the trial and purchase aspirin
independently. OTC availability of aspirin also leads to the potential for
control arm contamination and an increased risk of toxicity if they are
randomised to 300 mg aspirin. To combat OTC aspirin use, site staff
are trained to re-enforce the key message that as yet, there is no clear
evidence from a randomised trial that adjuvant aspirin use improves
survival, and equipoise is emphasised in all information provided to po-
tential participants. For those already registered, site staff are trained to
regularly ask about and discourage OTC aspirin use. Randomisation to
two doses of aspirin or placebo also leads to a 2:1 chance of receiving
a potentially active agent.
4.5. Timing of entry considerations
The timing of trial registration has been aligned across all four co-
horts to make the treatment of each as similar as possible, however a
number of adjustments have been necessary to account for the variety
of treatment modalities and pathways, and differences in patient char-
acteristics between cohorts. The timing of entry around adjuvant che-
motherapy requires particular consideration. The risk of developing
dyspepsia during adjuvant chemotherapy varies according to the regi-
men used, the need for dexamethasone as a supportive therapy, and
the incidence of risk factors in that group. Dyspepsia is a common occur-
rence during adjuvant chemotherapy for breast cancer and as such, pa-
tients can only register once chemotherapy is complete. In the
colorectal and gastro-oesophageal cohorts dyspepsia developing as a
consequence of adjuvant chemotherapy is less common, and registra-
tion is permitted when six weeks of adjuvant chemotherapy has been
administered (without developing dyspepsia and subject to acceptable
platelet counts).
4.6. Run-in period
The run-inperiod was implementedafter funders' concern about the
risk of poor adherence. Adherencewill be assessed during the run-in pe-
riod using three different methods to allow a more accurate assessment.
These include a participant diary card, return of used blister packs and
participant interview at an end of run-in assessment. The run-in period
provides an early opportunity to assess feasibility, in terms of early tox-
icity, recruitment and patient acceptability, and provides a population
for the randomised phase who are more likely to tolerate and adhere
to the trial treatment for the duration of the trial [43]. This strategy
has also been used successfully in other aspirin trials [44,45].Wedo
not believe that eight weeks of aspirin will reduce the effect between
the aspirin and placebo comparison in the trial as data have consistently
shown that long-term treatment (a minimum of 2 years [1], and up to
5–10 years [2,3]) with aspirin is required for the anti-cancer effects of
aspirin to become identifiable [1–3].The use of the run-in period is
being monitored carefully and will be reviewed on completion of an ini-
tial feasibility phase. Other methods of encouraging adherence include
the use of blister packs labelled with days of the week, provision of
diary cards, participant newsletters and promotion of the trial website
which includes updates and reaction to stories related to aspirin in the
news media.
4.7. Drug supply
Aspirin is a generic drug, Bayer AG donated all doses of the blinded
active intervention and matched placebo, but not the packaging, label-
ling, blinding or distribution, which therefore represented amajor oper-
ational and funding challenge. These challenges have been met by
outsourcing some of these processes and development of an in-house
drug supply management system to track stock levels at sites and auto-
matically trigger re-orders based on projected demand. The systemalso
includes an unblinding capability.
4.8. Co-enrolment
Since aspirin is intended to be given followingor alongside standard
primary therapy, rather than replacing any element of current treat-
ment, it will be appropriate to include participants who have already
taken part in trials of primary treatments wherever possible. This will
allow assessment of the efficacy of aspirin in participants who have re-
ceived both current and potentially future standard of care treatments.
Including participants from other treatment trials will help to ensure
the future relevance of both trials, is important for recruitment feasibil-
ity, and maximises the opportunities for patients to participate in trials.
We have found that the acceptability of co-enrolment to researchers
varies by tumour group. This may be due to variation in the amount of
trial activity or even differences in patient group demographics. Our ap-
proach has been to consider co-enrolment on a trial-by-trial basis,
discussing this with the relevant trial teams, with a careful assessment
of any conflicts in eligibility criteria, scheduling and potential impact
on safety and the results of either trial. Where concerns exist, orreassur-
ance is required, statistical modelling is conducted to assess the poten-
tial impact on trial results (there is often limited overlap leading to
negligible impact). Co-enrolment has been agreed with 16 trials to-
date and is planned with other trials currently in development.
4.9. Recruitment in India
Since aspirin is easily available worldwide, demonstrating its imple-
mentation in different resource settings will increase the global impact
of the results. Recruitment in India is also important to ensure adequate
61C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
recruitment in the gastro-oesophageal cohort and also allows thedevel-
opment of new international collaborations. Academic multi-centre tri-
als are rare in India and the set-up of Add-Aspirin has helped with
development of a research infrastructure for this and future trials. Re-
cent changes in clinical regulations in India have delayed opening and
the trial is anticipated to open in India later in 2016.
4.10. Conclusions
Aspirin is a low-cost, generic drug that is easily available worldwide.
Consequently, if aspirin is shown to be beneficial as an adjuvant treatment,
even with a modest effect, it would change practice globally. Compared
with many new agents or complex regimens, the intervention could be im-
plemented quickly and on a broad scale, including in lower resource set-
tings, with the potential to have a huge impact on the global cancer burden.
Funding
The Add-Aspirin trial is jointly funded by Cancer Research UK
(CRUK) (grant number C471/A15015) and the National Institute Health
Research (NIHR) Health Technology Assessment Programme (HTA)
(project number 12/01/38). Bayer Pharmaceuticals AG has agreed to
provide the Investigational Medicinal Products (IMPs). In India, the Sir
Dorabji Tata Trust provides funding, and CIPLA Ltd. is providing supplies
of aspirin 100 mg for the run-in period. The Add-Aspirin translational
sample collection is funded by CRUK (C471/A19252). The trial is coordi-
nated and supported by the Medical Research Council Clinical Trials
Unit at University College London (MRC CTU at UCL).
Disclosures
Ruth E. Langley has received financial support through grants from
Cancer Research UK and the National Institute for Health Research
(NIHR) Health Technology Assessment (HTA) as Chief Investigator of
the Add-Aspirin trial, has received compensation from Bayer and Aspi-
rin Foundation for service on scientific advisory boards, and has re-
ceived a supply of aspirin and placebo from Bayer Pharmaceuticals AG
for the Add-Aspirin trial. Richard Wilson has received a supply of aspirin
and placebo from Bayer Pharmaceuticals AG for the FOCUS4-B trial in
metastatic colorectal cancer and for which he is the co-Chief
Investigator.
Acknowledgements
We are very grateful to our funders in the UK (CRUK and NIHR
HTA) and India (Sir Dorabji Tata Trust). We would like to thank
Bayer AG for supply of aspirin and placebo, and to CIPLA Ltd. in
India for providing supplies of aspirin 100 mg for the run-in period
in India. We would like to thank the NIHR, devolved nation R&D
funders, and the Cancer Research Networks in the UK and India
for trial delivery. We would also like to show our gratitude to
our patient representatives in both countries; to the participating
investigators and research teams at sites, and to all the trial
participants. We would like to acknowledge the contribution of
Dr. Geoffrey Venning towards initiating this project. His foresight,
patience and wisdom are appreciated.
Appendix A. The Add-Aspirin trial investigators
Gastro-oesophageal Cohort Trial Management Group
Professor Ruth Langley Chief Investigator and Lead Investigator, Gastro-oesophageal Cohort (UK) London, UK
Professor C S Pramesh Lead Investigator –India, Lead Investigator, Gastro-oesophageal Cohort (India) Mumbai, India
Dr Richard Hubner Medical Oncologist –Gastro-oesophageal Cohort Manchester, UK
Professor Janusz Jankowski Gastroenterologist –Gastro-oesophageal Cohort Warwick, UK
Mr Tim Underwood Surgeon –Gastro-oesophageal Cohort Southampton, UK
Professor Anne Thomas Medical Oncologist –Gastro-oesophageal Cohort Leicester, UK
Verity Henson Clinical Trials Officer Bristol, UK
Professor John Bridgewater Medical Oncologist –Gastro-oesophageal Cohort London, UK
Breast Cohort Trial Management Group
Dr Alistair Ring Lead Investigator –Breast Cohort (UK) London, UK
Professor David Cameron Medical Oncologist –Breast Cohort. Lead Investigator, Translational Research Edinburgh, UK
Professor Sudeep Gupta Lead Investigator –Breast Cohort (India) Mumbai, India
Colorectal Cohort Trial Management Group
Professor Richard Wilson Lead Investigator –Colorectal Cohort Belfast, UK
Dr Tim Iveson Medical Oncologist –Colorectal Cohort Southampton, UK
Professor Robert Steele Surgeon –Colorectal Cohort Dundee, UK
Dr Daniel Swinson Medical Oncologist –Colorectal Cohort Leeds, UK
Ms Farhat Din Surgeon –Colorectal Cohort Edinburgh, UK
Prostate Cohort Trial Management Group
Professor Howard Kynaston Lead Investigator –Prostate Cohort Cardiff, UK
Dr Duncan Gilbert Clinical Oncologist –Prostate Cohort Brighton, UK
Mr Paul Cathcart Surgeon –Prostate Cohort London, UK
Cross-Study Collaborators
Professor Mahesh Parmar Director, MRC CTU London, UK
Professor Peter Rothwell Clinical Neurologist Oxford, UK
Professor Carlo Patrono Pharmacologist Rome, Italy
Professor Sir John Burn Clinical Geneticist Newcastle, UK
Dr David Adlam Cardiologist Leicester, UK
Dr Michael Peake Clinical Lead, National Cancer Intelligence Network London, UK
Participant Representatives
Lindy Berkman Participant Representative, NCRI Consumer Forum UK
Mairead MacKenzie Participant Representative, Independent Cancer Patient Voices UK
Vandana Gupta Participant Representative, VCare India
Arnold Goldman Participant Representative UK
62 C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
MRC CTU at UCL
Marta Campos Trial Manager MRC CTU, UK
Anna Thomason Trial Manager MRC CTU, UK
Ben Sydes Data Manager MRC CTU, UK
Shabinah Ali Data Manager MRC CTU, UK
Alex Robbins Data Manager MRC CTU, UK
Sam Rowley Statistician MRC CTU, UK
Dr Fay Cafferty Project Leader/Senior Statistician MRC CTU, UK
Dr Chris Coyle Trial Physician and Trial Clinical Research Fellow MRC CTU, UK
Appendix B. Figures describing cohort specific timing of entry criteria
Appendix C. Trial schedules
Assessments Prior to registration Prior to randomisation
(end of run-in period)
Months since randomisation
3 6 9 121824303642485460
All cohorts
Main assessments Registration assessments
a
✓
End of run-in assessment
b
✓
Follow-up assessments
c
✓✓✓✓✓✓✓✓✓✓✓✓
Intermittent assessments VES-13 questionnaire
(65 ≥years at registration)
✓ ✓
Cognitive assessment ✓✓✓
Blood tests FBC, LFT, U&E & eGFR ✓ ✓ ✓✓✓✓✓✓
C-Reactive Protein (CRP) ✓✓✓
Fasting lipid profile ✓
Other tests Tumour and blood sample ✓
Breast cohort
Imaging Mammography ✓ ✓✓✓✓✓
Colorectal cohort
Imaging and procedures CT (chest/abdomen/pelvis) ✓✓✓✓
Colonoscopy ✓✓✓✓
Blood tests CEA test ✓ ✓ ✓✓✓✓✓✓✓✓✓✓✓✓
(continued on next page)
(continued)
63C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
(continued)
Assessments Prior to registration Prior to randomisation
(end of run-in period)
Months since randomisation
3 6 9 121824303642485460
Prostate cohort
Blood tests PSA ✓ ✓ ✓✓✓✓✓✓✓✓✓✓✓✓
Gastro-oesophageal cohort
No cohort specific investigations
a
Registration assessments include: eligibility, co-enrolment (if applicable), blood pressure, height, weight, blood results, concomitantmedication and comorbidities.
b
End of run-in assessments include: symptoms and toxicity, adherence, blood pressure.
c
Follow-up assessments include: symptoms and toxicity, adherence, blood pressure, weight, concomitant medication.
References
[1] Cancer Research UK. CancerStats - Cancer Statistics for the UK. In.
[2] C. Bosetti, V. Rosato, S. Gallus, et al., Aspirin and cancer risk: a quantitative review to
2011, Ann. Oncol. 23 (2012) 1403–1415.
[3] S. Zhong, X. Zhang, L. Chen, et al., Association between aspirin use and mortality in
breast cancer patients: a meta-analysis of observational studies, Breast Cancer Res.
Treat. 150 (2015) 199–207.
[4] M.D. Holmes, W.Y. Chen, L. Li, et al., Aspirin intake and survival after breast cancer, J.
Clin. Oncol. 28 (2010) 1467–1472.
[5] D.M. Fraser, F.M. Sullivan, A.M. Thompson, C. McCowan, Aspirin use and survival
after the diagnosis of breast cancer: a population-based cohort study, Br. J. Cancer
111 (2014) 623–627.
[6] A.T. Chan, S. Ogino, C.S. Fuchs, Aspirin use and survival after diagnosis of colorectal
cancer, JAMA 302 (2009) 649–658.
[7] E. Bastiaannet, K. Sampieri, O.M. Dekkers, et al., Use of aspirin postdiagnosis im-
proves survival for colon cancer patients, Br. J. Cancer 106 (2012) 1564–1570.
[8] C. McCowan, A.J. Munro, P.T. Donnan, R.J. Steele, Use of aspirin post-diagnosis in a
cohort of patients with colorectal cancer and its association with all-cause and colo-
rectal cancer specific mortality, Eur. J. Cancer 49 (2013) 1049–1057.
[9] A.J. Walker, M.J. Gr ainge, T.R. Card, Aspirin and other non-steroidal anti-
inflammatory drug use and colorectal cancer survival: a cohort study, Br. J. Cancer
107 (2012) 1602–1607.
[10] S. Bains, M. Mahic, M. Cvancarova, et al., Impact of aspirin as secondary prevention
in an unselected cohort of 25,644 patients with colorectal cancer: a population-
based study, J. Clin. Oncol. 33 (2015) 33 (2015).
[11] K. Ng, J.A. Meyerhardt, A.T. Chan, et al., Aspirin and COX-2 inhibitor use in patients
with stage III colon cancer, J. Natl. Cancer Inst. 107 (2015) 345.
[12] J.-F. Liu, G.G. Jamieson,T.-C. Wu, et al., A preliminary study on the postoperative sur-
vival of patients given aspirin after resection for squamous cell carcinoma of the
esophagus or adenocarcinoma of the cardia, Ann. Surg. Oncol. 16 (2009)
1397–1402.
[13] J. van Staalduinen, M. Frouws, M. Reimers, et al., The effect of aspirin andnonsteroi-
dal anti-inflammatory drug use after diagnosis on survival of oesophageal cancer
patients, Br. J. Cancer 114 (2016) 1053–1059.
[14] N.G. Zaorsky, M.K.Buyyounouski, T. Li,E.M. Horwitz, Aspirin and statinnonuse asso-
ciated with early biochemical failure after prostate radiation therapy, Int. J. Radiat.
Oncol. Biol. Phys. 84 (2012) e13–e17.
[15] K.S. Choe, J.E. Cowan, J.M. Chan, et al., Aspirin use and the risk of prostate cancer
mortality in men treat ed with prostatectomy or radiotherapy, J . Clin. Oncol. 30
(2012) 3540–3544.
[16] E.J. Jacobs, C.C. Newton, V.L. Stevens, et al., Daily aspirin use and prostate cancer-
specific mortality in a large cohort of men with nonmetastatic prostate cancer, J.
Clin. Oncol. 32 (2014) 3716–3722.
[17] P.M. Rothwell, M. Wilson, C.E. Elwin, et al., Long-term effect of aspirin on colorectal
cancer incidence and mortality: 20-year follow-up of five randomised trials, Lancet
376 (2010) 1741–1750.
[18] P.M. Rothwell, F.G. Fowkes, J.F. Belch, et al., Effect of daily aspirin on long-term risk
of death due to cancer: analysis of individual patient data from randomised trials,
Lancet 377 (2011) 31–41.
[19] P.M. Rothwell, J.F.Price, F.G. Fowkes,et al., Short-term effects of dailyaspirin on can-
cer incidence, mortality, and non-vascular death: analysis of the time course of risks
and benefits in 51 randomised controlled trials, Lancet 379 (2012) 1602–1612.
[20] P.M. Rothwell, M. Wilson, J.F. Price, et al., Effect of daily aspirin on risk ofcancer me-
tastasis: a study of incident cancers during randomised controlled trials, Lancet 379
(2012) 1591–1601.
[21] N.R. Cook, I.M. Lee, S.M. Zhang, et al., Alternate-day, low-dose aspirin and cancer
risk: long-term observational follow-up of a randomized trial, Ann. Intern. Med.
159 (2013) 77–85.
[22] P.C. Elwood, G. Morgan, J.E. Pickering, et al., Aspirin in the treatment of cancer: re-
ductions in metast atic spread and in mortality: a systemati c review and meta-
analyses of published studies, PLoS One 11 (2016), e0152402.
[23] J. Burn, A.M. Gerdes, F. Macrae, et al., Long-term effect of aspirin on cancer risk in
carriers of hereditary colorectal cancer: an analysis from the CAPP2 randomised
controlled trial, Lancet 378 (2012) 2081–2087.
[24] C. Baigent, L. Blackwell, R. Collins, et al., Aspirin in the primary and secondary pre-
vention of vascular disease: collaborative meta-analysis of individual participant
data from randomised trials, Lancet 373 (2009) 1849–1860.
[25] E.P. Whitlock, B.U. Burda, S.B. Williams, et al., Bleeding risks with aspirin use for pri-
mary prevention in adults: a systematic review for the U.S. Preventive Services Task
Force, Ann. Intern. Med. 164 (12) (2016) 826–835.
[26] J. Cuzick, M.A. Thorat, C. Bosetti, et al., Estimates of benefits and harms of prophylac-
tic use of aspirin in the general population, Ann. Oncol. (2014).
[27] S.E.M. Saliba, L.A. Rubenstein, D.H. Solomon, et al., The vulnerable elders survey
(VES-13): a tool for identifying vulnerable elders in the community, J. Am. Geriatr.
Soc. 49 (2001) 1691–1699.
[28] C.A. Hudis, W.E. Barlow, J.P. Costantino, et al., Proposal for standardized definitions
for efficacy end points in adjuvant breast cancer trials: the STEEP system, J. Clin.
Oncol. 25 (2007) 2127–2132.
[29] P. Ellis, P. Barrett-Lee, L. Johnson, et al., Sequential docetaxel as adjuvant chemother-
apy for early breast cancer (TACT): an open-label, phase III, randomised controlled
trial, Lancet 373 (2009) 1681–1692.
[30] P. Francis, J. Crown, A. Di Leo, et al., Adjuvant chemotherapy with sequential or con-
current anthracycline and docetaxel: Breast International Group 02 98 randomized
trial, J. Natl. Cancer Inst. 100 (2008) 121–133.
[31] M. Martin, M.A. S egui, A. Anton, e t al., Adjuvant docetaxel for high-risk, nod e-
negative breast cancer, N. Engl. J. Med. 363 (2010) 2200–2210.
[32] R. Badwe, R. Hawaldar, V. Parmar, et al., Single-injection depot progesterone before
surgery and survival in women with operable breast cancer: a randomized con-
trolled trial, J. Clin. Oncol. 29 (2011) 2845–2851.
[33] T. Andre, C. Boni, M. Navarro, et al., Improved overall survival with oxaliplatin, fluo-
rouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the
MOSAIC trial, J. Clin. Oncol. 27 (2009) 3109–3116.
[34] Quasar Collaborative Group, Adjuvant chemotherapy versus observation in patients
with colorectal cancer: a randomised study, Lancet 370 (2007) 2020–2029.
[35] D.P. Kelsen, K.A. Winter, L.L. Gunderson, et al., Long-term results of RTOG trial 8911
(USA Intergroup 113): a random assignment trial comparison of chemotherapy
followed by surgery compared with surgery alone for esophageal cancer, J. Clin.
Oncol. 25 (2007) 3719–3725.
[36] W.H. Allum, S.P. Stenning, J. Bancewicz, et al., Long-term results of a randomized
trial of surgery with or without preoperative chemotherapy in esophageal cancer,
J. Clin. Oncol. 27 (2009) 5062–5067.
[37] J.S. Macdonald, S.R. Smalley, J. Benedetti, et al., Chemoradiotherapy after surgery
compared with surgery alone for adenocarcinoma of the stomach or gastroesopha-
geal junction, N. Engl. J. Med. 345 (2001) 725–730.
[38] D. Cunningham, W.H.Allum, S.P. Stenning, et al., Perioperative chemotherapyversus
surgery alone for resectable gastroesophageal cancer, N. Engl. J. Med. 355 (2006)
11–20.
[39] O. Groene, D. Cromwell, R. Hardwick, et al., National oesophago-gastric cancer audit,
The Royal College of Surgeons of England, 2012.
[40] M. Bolla, H. van Poppel, L. Collette, et al., Postoperative radiotherapy after radical
prostatectomy : a randomised con trolled trial (EORTC trial 22911 ), Lancet 366
(2005) 572–578.
[41] D.P. Dearnaley, M.R. Sydes, J.D. Graham, et al., Escalated-dose versus standard-dose
conformal radiotherapy in prostate cancer: first results from the MR C RT01
randomised controlled trial, Lancet Oncol. 8 (2007) 475–487.
[42] B. Kim, S.J. Park, S.P. Hong, et al., Theeffect of prediagnosticaspirin use on the prog-
nosis of stage III colorectal cancer, Int. J. Clin. Exp. Med. 8 (2015) 13435–13445.
[43] J.M. Lang, The use of a run-in to enhance compliance, Stat. Med. 9 (1990) 87–93
(discussion 93–85).
[44] J.A. Baron, B.F. Cole, R.S. Sandler, et al.,A randomized trial of aspirin to prevent colo-
rectal adenomas, N. Engl. J. Med. 348 (2003) 891–899.
[45] J.M. Lang, J.E. Buring, B. Rosner, et al., Estimating the effect of the run-in on the
power of the physicians' health study, Stat. Med. 10 (1991) 1585–1593.
64 C. Coyle et al. / Contemporary Clinical Trials 51 (2016) 56–64
