Obesity Prevention in Children and Young People Treated for Acute Lymphoblastic Leukaemia with ALLTogether

The international randomised controlled trial registry shows that no lifestyle intervention (diet, physical activity with integrated counselling) has been performed in the UK. The IDEAL trial performed in the USA demonstrated that diet with energy intake restriction and physical activity was feasible and significantly improved Minimal Residual Disease (MRD) in children diagnosed with Acute lymphoblastic leukaemia (CYP_ALL) and treated with National Cancer Institute/Rome high-risk B-ALL protocol during induction. ALLTogether is similar and will for the first time provide an excellent platform to prevent obesity in CYP_ALL.

Obesity associated relapse is attributed to adipocyte-mediated chemo-resistance, which occurs in CYP_ALL with higher fat mass and leads to more persistent MRD. MRD is a marker of leukaemic cells and the strongest predictor of prognosis including both remission and relapse. Therefore, obesity is identified as a prognostic factor for relapse (measured by MRD) and it is associated with increased morbidity and mortality, and poorer wellbeing.

Only two dietary feasibility interventions have investigated CYP_ALL differing in the type (diet alone vs. diet and physical activity) and the length of intervention (1 - 6 months). Both showed improvements in diet quality, high participation (>70%) and 56 - 82% adherence. Indeed, the IDEAL trial also showed improvements in MRD and body composition, making it suitable to base BREVARY's design. Consensus exists regarding using patient-centre interventions in clinical settings; like the Individualised Macronutrient Meal-Equivalent Menu (Menus). This approach reduces food associated stress, anxiety and financial costs, while empowering patients and their families/carers, thus improving adherence. It has been shown to be successful in adults with breast cancer and provides a validated approach to CYP_ALL.

A systematic review showed that individualised physical activity combining fitness and strength mitigated complications in CYP_ALL. However, it was difficult to draw firm conclusions because evidence were limited by sample sizes, adherence was seldom reported and study designs were heterogenous differing in length (1 - 30 months), exercise type and intensity and assessed outcomes (e.g. cardiorespiratory fitness, strength and wellbeing). Six trials are underway including all cancers with interventions being either individualised ("as able" and play-based exercise for children aged <5 years) or employing generic exercise prescriptions with intensity "as able". They are investigating fitness, strength and wellbeing; however, none have incorporated dietary assessment or intervention.

1. Research proposal

Aim

Investigate the feasibility of conducting a randomised controlled trial (RCT) to prevent obesity in CYP_ALL and treated with ALLTogether.

The hypothesis is that BREVARY will reduce obesity rates, improve clinical outcomes by reducing side-effects and relapse and improving CYP_ALL and their families/carers' quality of life. BREVARY will establish if a multi-site RCT is feasible and acceptable to families and clinical teams, examine recruitment and data collection strategies, and assess candidate outcomes.

BREVARY comprises three workstreams (WS):

WS1: Implementing a RCT of BREVARY for 3 months involving newly diagnosed CYP_ALL treated with ALLTogether in SW England NHS-sites and Hospital Infantil Universitario Nino Jesus Madrid WS2: Process evaluation of BREVARY RCT examining feasibility and maintenance of intervention (3 months intervention, maintenance at 6 months) WS3: Assessment of feasibility of progression to a definitive trial, including setting trial parameters

Objectives

• Assess recruitment, including acceptability of randomisation to stakeholders, across ethnically, geographically and socially diverse communities (WS1)
• Assess interventions acceptability and retention (WS1/WS2)
• Assess feasibility and appropriateness of quantitative and qualitative outcomes (WS1)
• Assess RCT resource implications of interventions and potential service use savings (WS1/WS3)
• Assess fidelity and adherence to protocol within interventions and control groups (WS1/WS2)
• Assess feasibility of progression to a definitive trial, set out main parameters and estimate both sample size and cost of a RCT (WS3)

Project plan

WS1: Implementing BREVARY

Study design BREVARY is phase 2 of Medical Research Council framework for complex interventions. Development and identification of the complex intervention (phase 1) was performed based on the IDEAL pilot-trial. BREVARY was designed in consultation with NIHR Nutrition and Cancer and PPI, reviewed by NIHR RSS and is supported by ALLTogether sponsors (UK and Spain).

Feasibility three-arm non-blinded parallel RCT with integrated quantitative and qualitative measures. Quantitative measures will be performed at baseline, end of induction (4 - 6 weeks) and during consolidation treatment (3 months weeks) and a follow up at 6 months. Qualitative measures will be performed at the end of BREVARY in a sample subgroup. The three-arms include combined diet and physical activity, diet only and standard care.

Recruitment Recruitment will be conducted by researchers. Inclusion criteria

• CYP aged between 2 - 21 years.
• Newly diagnosed and relapse CYP_ALL treated in University Hospital Bristol and Weston NHS Foundation Trust and Hospital Infantil Universitario Nino Jesus, Madrid
• Within two weeks of being recruited to ALLTogether.
• Treated with curative intent. Exclusion criteria
• CYP not partaking in ALLTogether.
• CYP treated with palliative intent.
• Those excluded by NHS-staff for clinical reasons. BREVARY aims to recruit 10 - 15 in each arm and each international centre (n= 60 - 90 in SW England and Madrid; 50% recruitment rate, which is below previous studies and the IDEAL trial to test feasibility and to calculate the variability of reliable estimates of continuous variables.

Eligible participants will be randomised using computer-generated restricted (nutritional status, age and sex) randomisation procedure in 1:1:1 ratio.

BREVARY Intervention (table 1)

Twenty % energy deficit and equicaloric intervention will be applied to those classified as overweight/obese and healthy weight respectively by diet (- 20%).

Researchers will deliver weekly one-to-one supervised sessions during induction and fortnightly during consolidation (total 8 sessions) online/home or at routine appointments. Physical activity intervention will include two additional sessions supervised non-consecutive sessions as tolerated. Both designed following PPI preferences.

Dietary intervention Menus based on Eatwell guide in the UK and Food and Agriculture Organisation in Spain will be used. Menus consists of prescribing 7 interchangeable meal options matched in energy and macronutrients requirements to reduce participants meal planning burden. It accounts for side-effects, psycho-social factors and religious, cultural, personal (and PPI) preferences. Nutritics® will be used to create Menus and estimate nutrients.

CYP_ALL referred to Dietetics will receive the intervention post-Dietetic discharged to avoid interference with nutritional support.

Physical activity intervention Individualised plans will be prescribed based on fitness and clinical condition. The programme will start at enrolment unless contraindicated. Sessions will include a combination of aerobic, balance and resistance exercise based on individual fitness and strength, age and clinical history as well as individual preferences and available equipment.

Standard care CYP_ALL will receive usual care (described in "The problem"). Counselling sessions for CYP_ALL and families/carers (all groups) Counselling sessions were requested by PPI and are recommended for motivation and adherence in long-term clinical interventions.

Two online/home (or at routine appointments) one-to-one sessions will be delivered. Sessions will be designed by all co-applicants, reviewed by our clinical psychologist and delivered by the researchers with PPI CoA support and focusing on:

(i) Diet and physical activity benefits (ii) Behaviour and autonomy (iii) Difficulties encountered and how to address them (iv) Motivation Data collection (all groups)

• Eligible CYP_ALL
• Eligible CYP_ALL approached to participate
• Families declining intervention before randomisation, after randomisation and after the first set of measurements (before follow-ups)
• Families allocated to interventions dropping out between second and final follow-up
• Families allocated to standard care dropping out between second and final follow-up

Health Electronic Records (HER) will be accessed to collect demographics (CYP_ALL sex, ethnicity and age), socio-economics (parents' relationship status, siblings, employment status, education level, income and housing status), clinical (diagnosis, treatments, side-effects, hospital admissions), routine biomarkers (MRD, immunoglobulins, full blood count, total cholesterol, high density lipoprotein and low density lipoprotein and glucose) and other biomarkers (leptin and adiponectin) and stool samples (microbiome). Participants' postcodes will be linked to the Indices of Multiple Deprivation.

Based on the IDEAL trial, feasibility and acceptability and PPI feedback, the following outcome measures will be taken at baseline, 4 - 6 weeks, 3 months and a follow up at 6 months (to assess intervention sustainability - no intervention from 3 - 6 months).

• Weight, height and body composition (Multi-frequency Bio-electrical Impedance)
• Dietary intake using a multiple-pass 24 hour recall
• Seven days physical activity levels at each follow up using GENEActiv accelerometers, strength by hand-held hydraulic dynamometer and fitness by 15-foot walk (excluding children aged <5y).
• CYP_ALL and their families/carers' QoL will be assessed at baseline and at completion using two validated questionnaires PedQL-Cancer and Parents PedQL-Cancer.

We will determine primary and secondary candidate outcome measures for future trial based on performance of selected measures.

Candidate primary outcome measures for definitive trial

• Body mass index Z-score (≤ 19 years)
• Muscle and fat mass percentage
• MRD
• Quality of Life: PedQL-Cancer and Parents PedQL-Cancer Candidate secondary outcome measures for definitive trial
• Biomarkers and Microbiome
• Food group indicators and nutrient intake
• Physical activity levels
• Hand strength (percentiles 5 ≤ 18 years)
• CPET - Fitness excluding children aged < 5 years
• Clinical outcomes, number and type of treatment side-effects and intervention side-effects
• Hospital admissions
• Changes in treatment regime (ALLTogether)
• Microbiome: composition and diversity of microbiome

Analysis CONSORT flow diagram will be used to report participant flow and descriptive statistics for demographics, clinical and socio-economic data with overall data and stratified by arm. Appropriate distribution or frequencies will be used (count/percentages, mean/SD and median/IQR). IBM_SPSS statistics® will be used for analysis and a Statistical Analysis Plan will be signed-off prior to any analysis taking place.

WS2: Process evaluation The process evaluation assesses how implementation of BREVARY is achieved, influenced by contextual factors including randomisation and trial design, and quality of implementation and acceptability. The following stakeholders will be included: Parents/carers whose CYP_ALL partake in BREVARY, CYP_ALL able to understand the questions and communicate an answer and researchers plus NHS-staff directly involved in implementing BREVARY. RE-AIM (http://www.re-aim.org/) based on reach, effectiveness, adoption, implementation and maintenance will be used.

Semi-structured one-to-one online (or at routine appointment) interviews will be conducted to a subgroup (n = 15) with equal representation from each arm to explore enjoyment:

• Best and worst things about partaking
• Overall satisfaction
• Likelihood to recommend BREVARY
• Reasons for dropping out Facilitator records will be completed by the researchers using structured case-records to collect RE-AIM

Analysis Quantitative survey data will be analysed using descriptive statistics. Free text responses will be analysed using thematic analysis and audio recordings will be transcribed, anonymised and data analysed using thematic analysis in Nvivo®.

Fidelity: Intervention delivery (usage statistics), receipt (usage statistics and interviews) and enactment (interviews, questionnaires and measurements) as well as repeated engagement (usage statistics and counselling sessions) will be conducted separately. Findings will be integrated following a triangulation protocol to assess whether data agree (convergence), complement one another (complementary), or contradict each other (dissonance).

WS3: Progression criteria and definitive RCT design

Results from WS1 and WS2 will be reviewed by all co-applicants. A trial will be regarded feasible if the following criteria are met:

Progression criteria and definitive RCT design

• Recruitment rate ≥ 50%
• ≥ 70% of recruited families complete the trial in one of the intervention groups
• ≥ 70% of families, researchers and clinical staff agree that the intervention is acceptable (RE-AIM)
• Recruited participants complete ≥ 60% of the supervised sessions (RE-AIM)
• ≥ 70% of families agree that the intervention is enjoyable

Health economics Within-trial economic and resource use analysis based on complex intervention protocols will be used. This reduces participants' burden and meets PPI feedback.

The within-trial analysis will investigate the cost-effectiveness of each arm. The primary within-trial analysis used in the full trial will be a cost-utility-analysis (CUA), which estimates the incremental cost per quality adjusted life (QALY) of each arm. QALY will be generated via measurement of utility values using PedsQL-cancer and Parents PedQL-Cancer instrument mapped to the EQ-5D.

The resource use analysis will be estimated using HER. Data (e.g. admissions, medication) will be collected by researchers. A cost-consequences framework will be used to describe resources used and benefits gained from the intervention without developing a full cost-effectiveness estimate.