Feasibility randomised controlled trial of remote symptom chemotherapy toxicity monitoring using the Canadian adapted Advanced Symptom Management System (ASyMS-Can): a study protocol.

INTRODUCTION: Technology is emerging as a solution to develop home-based, proactive 'real-time' symptom monitoring and management in cancer care. The Advanced Symptom Monitoring and Management System-Canada (ASyMS-Can) is a remote phone-based symptom management system that enables real-time remote monitoring of systemic chemotherapy toxicities. METHODS AND ANALYSIS: This study is an open-label, prospective, mixed-method, Phase II, 2-arm parallel group assignment (ASyMS-Can vs usual care) feasibility study in patients with cancer receiving systemic (neo-adjuvant or adjuvant) chemotherapy at Princess Margaret Cancer Centre. A total of 114 patients will be recruited in oncology clinics prior to initiation of chemotherapy. Patients in both arms will complete a demographic and a set of questionnaires at enrolment, mid and end of treatment. Patients in intervention arm will be provided with an encrypted, secure, preprogrammed ASyMS phone for symptom reporting daily for the first 14 days of each chemotherapy treatment cycle up to sixth cycle (16 weeks). Feasibility metrics (recruitment, retention and protocol adherence) and outcomes to assess impact of ASyMS-Can include symptom severity, emotional distress, quality of life and acceptability to patients and clinicians. ETHICS AND DISSEMINATION: The study has received ethical and institutional approvals from the University Health Network. Dissemination will include presentations at national/international conferences, and publications in peer-reviewed journals. TRIAL REGISTRATION NUMBER: NCT03335189. © Author(s) (or their employer(s)) 2020. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

RCT Entities:   Population Interventions Outcomes

In the present randomised controlled clinical trial, an electronic patient-reported outcomes for management of chemotherapy-related toxicities is implemented.

Both qualitative and quantitative research methods were applied to identify any procedural and clinical uncertainties including estimates of recruitment and retention rates, and feasibility and acceptability of the intervention.

Findings may not be generalisable due to a single-centre clinical trial and lack of an attention control group.

Introduction

Treatment toxicities due to systemic chemotherapy are prevalent1 and often under-recognised,2 resulting in high rates of symptom distress, avoidable visits to the emergency department (ED) and hospitalisations.3 4 Empirical evidence has substantiated the role of remote symptom monitoring in ambulatory cancer care, using electronic patient-reported outcomes (ePROs), for symptom management.5 6 Several ePRO systems exist,6–10 and have been demonstrated to improve patient satisfaction, patient–clinician communication, health-related quality of life (HRQoL), overall survival, and reduce acute health service utilisation.5 11–13 These systems facilitate using evidence-based guidelines as part of a comprehensive symptom management approach in cancer care.14 A recent study also showed the cost-effectiveness of an ePRO symptom monitoring programme for patients with advanced solid tumour in Alberta–Canada and recommended implementation of ePRO tools to support management of chemotherapy-related toxicities in Canadian patients with cancer.15 However, few systems use specific features such as decision support for standardising symptom triage10 or provide automated self-care information that may improve the level of patient activation.9 16 Growing evidence suggests that patient activation plays an imperative role for enabling better coping with and managing cancer-related symptoms.17 18

The ASyMS is a mobile phone based system that enables real-time remote monitoring of systemic chemotherapy toxicities, using an ePRO measure aligned with Common Terminology Criteria for Adverse Events.19 ASyMS was developed and tested in Europe20; however, based on best practices for writing actionable statements for self-care advice to support patient activation in symptom self-management7 and ASyMS usability study data in Canadian patients with cancer,18 we enhanced the self-care advice. Additionally, we adapted the risk scoring system and decision-support algorithms within the device to align with Canadian evidence-based protocols for symptom triage21 and for consistency with definitions for fever in provincial guidelines.22

ASyMS-Can supports daily remote symptom reporting and back-end analytics of ePROs to derive risk scores by combining symptom severity with other data that triggers alerts that are sent to clinical trial nurse phones to prompt early intervention in managing cancer patients receiving chemotherapy. ASyMS also provides automated self-care advice to empower patients to take a more active role in self-management; thus has the potential to reduce severity of symptoms. While ASyMS has utility in Europe, evaluation of its feasibility and acceptability in ‘real world’ Canadian ambulatory oncology practices prior to large scale trials are needed. Feasibility studies provide data about whether a study can be done and estimate key parameters to design a larger, definitive trial.23 24 This study is a mixed-method, prospective, Phase II, 2-arm parallel group assignment (ASyMS-Can vs usual care) feasibility trial in patients receiving systemic neo-adjuvant or adjuvant chemotherapy for early stage (Stages I–III) breast, colorectal, and lymphoma (Hodgkin’s), non-Hodgkin’s malignancies.

Aims

The overall aim of this research project is to improve the management of cancer treatment-related toxicities by early identification of symptoms and prompting intervention using a mobile phone-based technology. The aims of this feasibility study are as follows:

To assess recruitment/retention rates and adherence to the intervention. Rates, reasons and factors associated with attrition, in both study arms, will be examined. Recruitment rate aim is 7–8 patients/month, based on other studies.25 A composite score for adherence rate (ratio of completed ePRO/ratio of daily ePRO reports that should be completed), number of alerts ((amber or red) per cycle) will be calculated.

To evaluate acceptability of ASyMS-Can and explore the views and experiences of the intervention in a sample of cancer patients and clinicians.

To estimate whether, compared with the control group, the ASyMS-Can intervention impacts on outcomes:

Reduced symptom severity measured by Memorial Symptom Assessment Scale (MSAS). The MSAS is a valid and reliable instrument26 and includes 24 items for report on whether a symptom (eg, pain, lack of energy and shortness of breath) occurred during the previous week, as well as any distress it may have caused.

Reduced psychological distress measured by Depression, Anxiety and Stress Scale (DASS21) at three time points (baseline, midpoint of treatment cycles and end of treatment cycles). DASS21 contains 21 items for self-reporting for measuring a range of symptoms common to both depression and anxiety. It has high internal consistency (Cronbach’s alpha scores of >0.7) and significantly correlates with other measures.27

Improved self-efficacy for coping measured by Cancer Behavior Inventory (CBI-B) at baseline, midpoint of treatment cycle, and end of treatment. CBI-B is a valid and reliable instrument with 12-items designed to assess coping self-efficacy of cancer patients and takes approximately 2 min to complete.28

Improved HRQoL measured by EuroQual-5D-5L (EQ-5D-5L) at three time points (baseline, midpoint of treatment cycles and end of treatment cycles). EQ-5D-5L a simple and relatively quick instrument for patient completion which has been validated in a diverse patient population including patient groups with chronic conditions.29

In addition, the following outcomes will be assessed:

Health service utilisation will be measured by a self-report questionnaire and include information such as days in hospital, ED visits, urgent care use, unscheduled clinic visits and will be collected at baseline, midpoint and end of treatment. It will be assessed as a binary variable and for comparison in odds between groups at midpoint and end of treatment.

Participant’s satisfaction with using the ASyMS-Can will be measured by The Post-Study System Usability Questionnaire (PSSUQ) at the end of intervention. The PSSUQ is a research instrument that consists of 16 items designed to assess users’ perceived satisfaction with computer systems or applications.30

Methods

Study design

This study is an open-label, prospective, mixed-method, Phase II feasibility study with parallel randomisation (1:1) of participants into the mobile remote monitoring intervention arm or a usual care arm (RCT (randomised controlled trial)). The protocol is written in compliance with the Standard Protocol Items: Recommendations for Interventional Trials and its checklist.31 32 On completion of the study, clinicians and patients in the intervention arm are invited to participate in interviews/focus groups based on a qualitative descriptive study design33 to gain a more in-depth understanding of acceptability of the intervention, outcomes measures, and implementation barriers to use of the ASyMS-Can mobile device. Figure 1 outlines the flow of the study.

Figure 1

Flow diagram of study design. ASyMS, Advanced Symptom Monitoring and Management System; ePRo, electronic patient-reported outcome; MSAS, Memorial Symptom Assessment Scale.

Setting

The study is undertaken at the Princess Margaret Cancer Centre (PM), which is an academic research centre and teaching hospital, affiliated with the University of Toronto Faculty of Medicine as part of the University Health Network (UHN), Toronto, ON, Canada. The Institutional Review Board Approval was obtained from the Research Ethics Board (REB) of UHN to conduct the study.

Adults patients (≥18 years old) diagnosed with early stage breast, colorectal and lymphoma.

Scheduled to receive a minimum of two cycles of systemic chemotherapy in 2-weekly, 3-weekly or 4-weekly cycles (ie, administered at repeated cycles of 14, 21 or 28 days, respectively).

Ability to use or be trained in use of a mobile phone for symptom reporting and able to complete questionnaires in English.

Enrolled/receiving an investigational treatment.

Scheduled to receive concurrent radiotherapy or a weekly chemotherapy protocol.

Any distant metastasis or receiving ONLY hormonal therapy, oral chemotherapy, targeted agents and monoclonal antibody/PD-1 (programmed cell death)/PDL-1 (programmed cell death receptor ligand) inhibitors.

Cognitive impairment assessed by the treatment team which may impede completion of study measures and unable to perform self-care (ECOG (Eastern Cooperative Oncology Group) ≥3).

Unable to provide written informed consent.

Sample size

A one-sided type I error rate of 0.10 and type II error rate of 0.15 (power of 85%) are reasonable statistical parameters for a feasibility study.34 A sample size of 88 subjects (44 per group) is required to test the difference of 0.5 SD change35 at midpoint of cycles using a two-sample t-test. Accounting for a 30% drop out rate,25 a total sample of 114 patients will need to be recruited. This sample size will allow us to estimate a recruitment rate of 70% to within a 95% CI of ±10% and to calculate an effect size for the main trial.36 Feasibility study sample sizes range from 24 to 50 for estimation of variance in outcomes with precision in feasibility studies (low SE >0.1).37 38

Patient recruitment/randomisation

Patients are screened through the Pathways Healthcare Scheduling, Electronic Patient Record (EPR), as well as outpatient lists. Eligible patients are recruited prior to initiating treatment in medical oncology consultation appointments. All patients will be informed that they will be randomly allocated to either the ‘mobile phone arm’ or ‘usual care arm.’ Consented participants are randomised in equal numbers to one of two groups with stratification by cancer type to ensure balance in groups. Randomisation will be done centrally by the PM Biostatistics department using a SAS computerised randomisation process in permutated blocks (random blocks of varying size). Those who decline to participate will be asked for verbal consent to collect basic information (eg, age, education and type of cancer) and reasons for refusal. Patients who consent to study participation will be given the baseline questionnaire to complete prior to randomisation. Research staff who collected the data and clinicians are not blinded to group assignment given the nature of the intervention.

Intervention

Participants allocated to the intervention group will be provided with the encrypted, secure, pre-programmed ASyMS-Can phone and instructed of its use; how to report their symptomatology on a daily basis using the ASyMS symptom questionnaire (Chemotherapy Toxicity Self-Assessment Questionnaire (CTAQ).19 The CTAQ assesses ten chemotherapy-related symptoms with an additional option to report up to six further symptoms. Data reported by the participants will be sent to a secure, encrypted PM clinical central server hosting the risk-alerting algorithms and clinical symptom platform. When patients send their symptoms data, they immediately receive evidence-based self-care advice for the specific symptoms reported.

In addition, patients can access a self-care library and symptom graphs (detailing trends in individual symptoms experienced) through the ASyMS-Can patient phones. If the incoming symptom reports are of clinical concern, the server software will generate two levels of alerts (amber and red) that will be sent to the designated nurse, who will receive alerts on a dedicated ASyMS-Can nurse handset (mobile phone). The nurse will view the patient’s symptom reports on a secure web page, and contact the patient directly at home by telephone, guided through a decision-support algorithm on the web-based platform to systematise the triage based on the COSTARS guideline,39 facilitating the initiation of ‘real-time’ clinical interventions. An ‘amber alert’, which requires response within 4 hours, indicates that the symptom(s) are not severe or life-threatening but early intervention might prevent further symptom progression. The second level of the triage alert, ‘red alert’, will be sent to the nurse for severe symptoms and will require response within 30 min of receipt of the alert (figure 2).

Figure 2

ASyMS-Can monitoring system. ASyMS-Can, Advanced Symptom Monitoring and Management System—Canada.

Data collection

Socio-demographic and clinical/disease characteristics will be collected at baseline. Participants from both arms will complete self-report measures to assess the participants’ physical and psychological symptoms, quality of life and healthcare utilisation (table 1). Patient outcomes will be measured at baseline, 2 weeks after completion of cycle 3, and 2 weeks after completion of last cycle (up to a maximum of six cycles) to determine suitability and timing of the primary endpoint for the main trial. All participants will also be contacted over the phone to complete the MSAS about 8 days after each treatment cycle. This timeframe was selected based on the typical pattern of symptom distress from our previous PRO study38 and other studies that have estimated appropriate time-points for capturing ‘true symptom burden’ during treatment in mixed cancers. Data collection measurements and time points are shown in table 1.

Table 1

Schedule of data collection measurement

Assessment/measurement instrument	Prior to/on enrolment	Baseline	Mid cycle(8 days after administration of each chemo cycle)over the phone	Mid treatment(2 weeks after third administration of chemo cycle)	End treatment(2 weeks after last administration of chemo cycle)	After completion of the study(within 7–10 days)
Eligibility checklist	X
Informed consent	X
Registration (ASyMS system)	X
Demographic variables	X
Clinical characteristics	X
Symptom severity:Memorial Symptom Assessment Scale (MSAS)		X	X	X	X	X
Depression Anxiety Stress Scale (DASS21)		X		X	X
Self-Efficacy for Coping (CBI-B)		X		X	X
HRQoL (EQ-5D-5L)		X		X	X
Health Care Utilisation (self-report questionnaire)		X			X
Acceptance and patients’ satisfaction of device (PSSUQ)					X
Patients’ experience—interview						X
Clinicians’ experience—focus group						X

ASyMS, Advanced Symptom Monitoring and Management System; CBI-B, Cancer Behavior Inventory; EQ-5D-5L, EuroQual-5D-5L; HRQoL, health-related quality of life; PSSUQ, Post-Study System Usability Questionnaire.

Patients in the intervention group will be asked to use the ASyMS-Can mobile phone to report their symptoms for the first 14 days of each treatment cycle until end of the final cycle of treatment.

Training/education

Both control and intervention participants will receive the standard preparatory training (usual treatment) of how to manage their chemotherapy symptoms by their clinic nurses and/or pharmacists. All participants will receive patient education materials and the standard toxicity monitoring care at PM clinics.

Intervention arm

Patients in the intervention arm will receive a training session to learn how to use the ASyMS-Can mobile phone to report their symptoms. Patients will be instructed to complete symptom reports at least once a day in the morning before 12:00 and whenever they feel unwell. Further training will be provided if a patient experiences any problems using the phone. The training will be provided by the research coordinator. In addition, a manual with instructions on how to use the phone will be given to each patient. Patients will also be instructed that nurses will not receive alerts during the evening hours or night (after 17:00), on weekends/holidays, or in case of system failure. They will be instructed in these cases to follow clinic team advice as standard care will apply.

Control arm

Patients in control arm will be informed about data collection procedures and how to complete the questionnaires at different time points.

Nurses training

Clinical trials nurses, designated for trial management in each cancer site, will receive training on computers as to how to use the device and platform with additional one-to-one training by the research coordinator to respond to the patients alerts during business hours.

A standard operating procedure for clinical trials nurses to respond, handle alerts, communicate, and collaborate with the clinical and research team is developed. On receipt of an alert, the nurse reviews the symptom report, clinical and demographic information, as the first step in response on the ASyMS-Can web-based system. In addition, trial nurses can also review the patients EPR and then phones the patient to further assess, using evidence-based telephone triage protocols39 and determines the appropriate disposition (urgent vs non-urgent) and directs the patient accordingly. In the case of amber alerts, the nurse will make their own assessment using information provided in ASySM-Can and can elect to close the alert using their professional judgement and without running through the initial alert handling protocols. However, in the case of any red alerts the nurse will handle the alert and inform the most responsible physician to ensure all appropriate actions are taken. All actions and interventions will be documented in the ASyMS-Can system and Clinical Research Record in the patient’s chart. Therefore, all clinicians have access to these documentations through EPR.

Qualitative interviews and focus groups

At study completion (within 7–10 days), all participants in the intervention group will be invited to participate in an audio-taped semi-structured interview. We aim to have a total of 12–16 patient participants, as it is proposed that saturation most often occurs around 11–15 participants in homogeneous groups.40 41 Also, all clinicians exposed to patients who used the mobile device will be invited to provide feedback to increase our understanding of the feasibility and care processes; implementation barriers in work flow impeding timely response to alerts; evaluate acceptability for use in routine practice and explore contexts and mechanisms in action, as well as implications of the ASyMS-Can in clinical practice. The leading sites members in PM who are involved with the ASyMS-Can intervention will be contacted to make initial contact to potential clinician participants. The leading sites members will inform their staff about the study and give them the research team’s contact information to contact them if they are interested to participate in focus group sessions. Those who are interested will then receive the appropriate consent form and will be provided with as much as time needed to consider participation prior to participation in the focus group. In each focus group interview, the researcher will use an interview guide to facilitate discussion among the participants and allow them to freely share their common thoughts/experiences/concerns.42 Clinicians’ focus groups (5–8 healthcare professionals per group) will last approximately 45 min, while the patients’ one-on-one audio-taped semi-structured interviews will last about 30–45 min.

Treatment duration

Participants remain on the study until end of chemotherapy treatment cycles (up to six cycles) or discontinuation of cancer treatment, voluntary withdrawal or death.

Patient and public involvement

We work with patient advocacy groups at UHN. A patient with cancer from this group was involved to read and complete the study questionnaires. The valuable input we received from the patient helped in selecting appropriate measurement tools for this study in terms of using plain language and the time needed to complete them. We aim to elicit participant experiences, thoughts, feelings and satisfaction with the ASyMS-Can by interviewing them after completion of the intervention. Patients are not involved in the recruitment to this study.

Data analysis plan

Data will be coded and entered separately into an SPSS-12 database. Descriptive statistics will be used to examine baseline equivalence between groups and for calculating adherence, recruitment, retention rates and assessing differential attrition. Univariate and multivariate regression methods appropriate to the data will be used to assess relationships between independent variables and dependent outcome variables. We will examine between and within group effects using generalised linear mixed models on MSAS symptom severity over time controlling for baseline severity scores and covariates (age and sex) to examine the plausible estimates of the effect based on confidence intervals to inform decision-making for proceeding to the main trial. A repeated measure analysis and a midpoint comparison using t-tests or ANOVA (analysis of variance) will be conducted for other outcomes (QoL, self-efficacy and distress) based on data type.

T-tests will be used to compare QoL outcomes symptom severity between arms at the midpoint and

All taped interviews and focus groups will be transcribed verbatim and with data from observations entered into a qualitative data management software programme (NVivo V.8) to facilitate coding, sorting and refining of subcategories and themes. Qualitative interview data will be an inductive thematic content analysis approach based on Graneheim and Lundman.43

Ethics and dissemination

All related study documents reviewed by UHN REB and ethical approval was obtained. The study is explained to all potential participants and clarifies that they are under no obligation to participate, and there will be no negative consequences if they do not agree. If they agree to participate, they are then told they may decline any question and/or withdraw at any stage of the study. Traditional dissemination methods (ie, publications in peer-reviewed journals and conference presentations) will be used to disseminate findings of this study.

Trial management and risks to safety

The intervention does not replace normal clinic contact but enhances usual care. Any changes to the conduct of the study or to the protocol will be amended and approved by the REB before implementation, unless required to eliminate an immediate hazard to participants.

Should an error occur in the transmission of patient side-effect data, from the patient phone to the system server/website, the software application will advise the patient of the failure of data transmission, and based on alerting algorithms, provide detailed instructions on who and when to call. The server containing the system will be monitored daily for performance and any faults addressed by technicians. Back-up systems are in place for power failures and patients will be advised of system failures, and who and when to call.

A Trial Management Committee (TMC) will be responsible for trial oversight including regular assessment of study conduct, recording and monitoring of risks to safety and adverse events, review accumulating study data related to the safety and efficacy of the study intervention and ensure continued scientific validity and merit of the study. This will provide a level of protection for the participants and the integrity of the trial. If concerns are raised about the conduct or participation of an independent member, then these concerns will be discussed in the TMC meetings.

Discussion

The technology in ASyMS was developed based on systematic reviews of cancer symptom problems and through extensive engagement with clinicians and patients and has been shown to support symptom management in the UK health system.20 44 However, its effects on patient outcomes is uncertain and is being tested in a large multi-site trial in European countries.45

Based on our systematic review of cancer self-management8 and ASyMS usability study data from patients and clinicians in Ontario,46 we have modified some of the content in ASyMS for customisation to the Canadian cancer system and the intervention approach. The intervention emphasises the use of a mobile device and its actionable self-care advice to build patients knowledge, skills and confidence in managing health and the effects of cancer treatment and a systematic, structured clinician response to alerts using the Canadian evidence-based toxicity triage decision support tools, which have been tested in chemotherapy clinics throughout Canada.47

The study will determine the feasibility of recruitment, retention, compliance and implementation barriers in response to alerts and acceptability for use in ‘real-world’ ambulatory oncology practices in Ontario. Also, the findings of this study will provide plausible estimates of effect of the ASyMS-Can intervention on reducing symptom burden, health services utilisation, and improving patient self-efficacy and acceptability thereby providing the necessary information to design methods and procedures and to inform decisions on proceeding to the full Phase III trial if the results of the study are favourable. Despite a lack of consensus on estimating preliminary effects of an intervention on outcomes in feasibility and pilot studies due to the small samples and possible imprecision, a preliminary estimate of effects alongside clinical considerations is often used to inform decision making regarding the plausible effects that are important for deciding on whether to proceed to the full main trial.48 Finally, we will explore the influence of variability in treatment cycles and intensity of treatment regimes in the sample on estimates of effect to inform methods and procedures for the future trial.