An Activity Tracker–Guided Physical Activity Program for Patients Undergoing Radiotherapy: Protocol for a Prospective Phase III Trial (OnkoFit I and II Trials)

Background: The positive impact that physical activity has on patients with cancer has been shown in several studies over recent years. However, supervised physical activity programs have several limitations, including costs and availability. Therefore, our study proposes a novel approach for the implementation of a patient-executed, activity tracker–guided exercise program to bridge this gap. Objective: Our trial aims to investigate the impact that an activity tracker–guided, patient-executed exercise program for patients undergoing radiotherapy has on cancer-related fatigue, health-related quality of life, and preoperative health status. Methods: Patients receiving postoperative radiotherapy for breast cancer (OnkoFit I trial) or neoadjuvant, definitive, or postoperative treatment for other types of solid tumors (OnkoFit II trial) will be randomized (1:1:1) into 3-arm studies. Target accrual is 201 patients in each trial (50 patients per year). After providing informed consent, patients will be randomized into a standard care arm (arm A) or 1 of 2 interventional arms (arms B and C). Patients in arms B and C will wear an activity tracker and record their daily step count in a diary. Patients in arm C will receive personalized weekly targets for their physical activity. No further instructions will be given to patients in arm B. The target daily step goals for patients in arm C will be adjusted weekly and will be increased by 10% of the average daily step count of the past week until they reach a maximum of 6000 steps per day. Patients in arm A will not be provided with an activity tracker. The primary end point of the OnkoFit I trial is cancer-related JMIR Res Protoc 2021 | vol. 10 | iss. 9 | e28524 | p. 1 https://www.researchprotocols.org/2021/9/e28524 (page number not for citation purposes) Hauth et al JMIR RESEARCH PROTOCOLS


Introduction
Radiotherapy is a key treatment modality for the curative treatment of various tumor entities. Despite continuous technical improvements in facilitating increasingly precise dose delivery, side effects are inevitable in many cases [1]. Among patients with breast cancer, cancer-related fatigue (CRF) is the most frequent side effect reported during and after postoperative radiotherapy [2]. Among patients with other tumor sites that were treated with preoperative or definitive radiochemotherapy, CRF has also been frequently reported. However, local acute and late side effects of treatment can cause further impairments in health-related quality of life (HRQoL) [1]. In addition, sedentary behavior during preoperative radiotherapy can have a negative impact on the postoperative course of a patient (eg, in terms of recovery and wound complications). The beneficial effect that physical activity has on patients with cancer has been clearly established over recent years and is supported by many clinical studies [3][4][5]. Moreover, many international organizations have updated their recommendations to include exercise as an important part of cancer therapy (ie, exercise before, during, and after cancer therapy) [6][7][8]. Guidelines for the inclusion of exercise in oncologic treatments for patients with cancer have been published recently [9]. In this context, physical activity training can be implemented at different time points during treatment. The goal of prerehabilitation is to improve the fitness of patients prior to undergoing a major medical intervention, such as cancer surgery [10]. Faster recovery, fewer wound complications, and improved HRQoL have been exhibited by patients who have participated in a prerehabilitation program [11][12][13]. In patients with breast cancer, low levels of physical activity are associated with a 22% higher risk of breast cancer mortality [14]. Furthermore, a recent meta-analysis has shown that physical activity during oncological treatments is a highly effective measure for reducing CRF [15]. Additionally, in patients that survive cancer after oncological treatment, a higher level of physical activity is associated with improved HRQoL and, in some studies, even improved cancer-specific survival [14,16]. Although the pathomechanisms leading to CRF remain unclear to date, previous studies have suggested a close relationship between CRF and proinflammatory pathways, including pathways that lead to increases in levels of interleukins [17,18].
Activity trackers, which are also called wearables, have been used by a growing population to record physical activity and biodata, such as pulse and sleep patterns. Although fitness wristbands were initially considered purely as lifestyle products, in recent years, the scientific benefits of these products have increasingly come to the fore [17][18][19]. However, no prospective trials that include activity trackers as part of patients' radiotherapy have been conducted to assess these trackers' efficacy in reducing CRF and improving HRQoL.
In the presented OnkoFit studies, we aim to investigate whether an activity tracker-based fitness program can reduce CRF and improve the quality of life and preoperative health status of patients undergoing radiotherapy.

Study Setting and Participants
The OnkoFit trials are two independent, single-center, randomized prospective phase III trials that have been certified by the working group radiological oncology of the German Cancer Society. After providing informed consent, patients will be randomized to a standard arm (arm A) or 1 of 2 interventional arms (arm B and C) in a 1:1:1 ratio ( Figure 1). All patients will be recruited by the Department of Radiation Oncology at the University Hospital Tuebingen. The OnkoFit I trial is exclusively designed for patients with breast cancer, while the OnkoFit II trial is open to patients undergoing preoperative, definitive, or postoperative radiotherapy for various tumor sites. The separation of the tumor entities in two different trials allows for two different end points. Although the end point of the OnkoFit I trial-fatigue at 3 months after radiotherapy-has been evaluated in a pilot trial that was conducted in our department [2], it was determined that for patients undergoing neoadjuvant radiochemotherapy, this end point might be too early, especially for patients who undergo surgery after neoadjuvant radiochemotherapy.
A summary of key inclusion and exclusion criteria is provided in Textbox 1. We anticipate the recruitment of 50 patients per year in each trial. The course of the study is displayed in Figure  2.  • Easter Cooperative Oncology Group score of 3-4 • Comorbidities with impairments of mobility, such as paraplegia

Treatment Groups and Interventions
After providing informed consent, patients will be included into either the OnkoFit I trial or OnkoFit II trial and randomized into 1 of the 3 arms of the studies ( Figure 1). Both studies aim to include 201 subjects each.

Arm A: Standard of Care
In this arm, patients will be advised to conduct at least 1.5 hours of moderate physical activity or 75 minutes of strenuous physical activity per week. Patients will receive an educational patient brochure (written in the German language) about the rational and potential benefits of physical activity during cancer treatment. Activity trackers will not be provided to the patients ( Figure 3).

Arm B: Activity Tracker Without Weekly Goals
As in arm A, patients in arm B will be advised to conduct 1.5 hours of moderate physical activity or 75 minutes of strenuous physical activity per week and will receive the previously mentioned patient brochure. In addition, patients will be given an activity tracker. They will be instructed to wear it throughout the day and document the daily step count in a patient activity diary, which will be provided to each patient. No instructions will be provided with regard to daily or weekly goals for step counts (Figure 3).

Arm C: Activity Tracker With Weekly Goals
As in the other arms, patients will be counseled about conducting 1.5 hours of moderate physical activity or 75 minutes of strenuous exercise and will receive the patient brochure. The same activity tracker as that in arm B will be given to patients. However, in arm C, patients will receive weekly goals for their average step counts during treatment. The baseline step count will be assessed from the time of informed consent provision until the time of computed tomography simulation (computed tomography scans of patients will be taken prior to treatment to allow for radiation treatment planning). This period usually amounts to approximately 7 days. The patients will document daily step counts in a patient activity diary. At the time of computed tomography simulation, the mean daily step count will be calculated; this will constitute the basis for the upcoming week's goal for step counts. If the average step count is 6000 or higher, the new goal will be to not fall under this value. If the average step count is less than 6000, the new goal will be calculated by increasing the past week's step count average by 10%. If, for instance, the average step count is 3000 steps, the new target will be 3300 steps. The last change to the target step count will take place during the last week of radiotherapy. The highest possible target step count is 6000. This means that an average of 5800 steps in 1 week will result in a new goal of 6000 steps. We chose this 6000-step threshold based on a literature search we conducted, which suggested that a sedentary lifestyle is defined by threshold [16] (Figure 3).
In all arms, blood samples will be taken at baseline, at the end of radiotherapy, and at 6 months after the end of treatment. The activity trackers that will be used in this trial are commercially available accelerometers.

OnkoFit I Trial
For the OnkoFit I trial, the primary end point will be fatigue, which will be scored according to the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F) subscale. The end point will be assessed 3 months after the completion of radiotherapy. The FACIT-F questionnaire, which consists of 40 questions, was validated to assess fatigue in patients with cancer among our own patient cohort in a pilot project, which was conducted prior to the OnkoFit trials [2], as well as in many other studies [20,21]. The secondary end points will include treatment compliance, patient-reported acute and late toxicity, disease-free survival, and overall survival.

OnkoFit II Trial
For the OnkoFit II trial, the primary end point will be the overall HRQoL, which will be assessed by the Functional Assessment of Cancer Therapy-General sum score. This end point will be measured 6 months after the end of treatment in order to account for possible surgical interventions and possible recovery thereafter. In addition to the secondary end points listed for the OnkoFit I trial, we will assess the frequency of unplanned inpatient treatments for patients planned for ambulatory treatments and the frequency of postoperative complications (which will be assessed by using the Clavien-Dindo classification system) [22] in patients receiving preoperative radiotherapy.
Treatment-related toxicities will be scored by using the German translation of the Patient-Reported Outcomes Common Terminology Criteria for Adverse Events (PRO-CTCAE), which were developed by the National Cancer Institute. Briefly, the assessment of the severity and frequency of symptoms will be based on a 5-tier scale that ranges from "none" to "very severe" or from "never" to "almost always." Depending on the treated site (breast cancer: 8 questions; head and neck: 27 questions; thoracic and abdominal lesions: 21/24 questions; brain cancer: 20 questions; sarcoma: 14 questions), specific sets of questions were created. The patient-reported outcome measurement questionnaires for patients with head and neck cancer as well as patients with thoracic and abdominal tumor lesions have been used in our department [1]. The selection of PRO-CTCAE items was based on our long-term experience with the most frequent side effects that occur during and after radiotherapy for treated regions. Depression will be assessed with the Patient Health Questionnaire-8 [23,24]. In order to assess baseline fitness and sports activities, a questionnaire was developed in cooperation with the Department of Sports Medicine at the University Tuebingen. The questionnaire contains questions concerning the individual fitness histories and baseline activity levels of individual patients in the form of patient-reported outcome measurement-related questions. The questionnaire includes 18 questions and is based on the previous work of the Department of Sports Medicine at the University Tuebingen and other groups [25,26].

Translational Subproject
To further expand our knowledge on the possible pathomechanisms behind CRF and the potential effects that exercise has on the course and severity of CRF, blood samples will be taken from patients before radiotherapy, at the end of radiotherapy, and at 6 months after radiotherapy for the evaluation of inflammatory markers, such as neutrophil counts and interleukin-6 and C-reactive protein levels.

Follow-up
Based on patients' preferences, patients can keep the activity tracker after the completion of treatment. Independent of this decision, patients will be seen for follow-up at 3, 6, and 12 months after the completion of radiotherapy. Thereafter, follow-ups will take place yearly for up to 5 years.

Sample Size and Statistical Considerations
Previous studies were able to demonstrate a medium to high effect size with regard to the influence that physical activity has on the fatigue and HRQoL of patients with cancer. Based on a medium effect size (Cohen f=0. 25) with an α error of 5% and a power of 80%, a one-way independent sample analysis of variance was conducted. Per the results of this analysis, 201 patients (67 patients per arm) will be recruited for each trial [27]. A dropout rate of 20% was assumed for the calculation of the number of patients. Randomization will be carried out via block randomization with variable block lengths.

Data Safety and Confidentiality
It will not be possible to record the position of patients in terms of their geographical location via the activity tracker. The clinical and personal data of patients will not be stored on the device. The assignment of the fitness trackers to the patients will be pseudonymized. The storage of fitness data and the linking of patient-related data will only be carried out within the hospital's information technology system. Therefore, these data will be subject to the hospital's data protection and data security regulations. For the purpose of publication, all data will be presented anonymously.

Results
Both studies have been approved by the Institutional Review Board of University Clinic Tuebingen (OnkoFit I trial reference number: 201/2020BO2; OnkoFit II trial reference number: 202/2020B02) and are registered on ClinicalTrials.gov (OnkoFit I trial number: NCT04506476; OnkoFit II trial number: NCT04517019). The recruitment of patients started on August 1, 2020. The results will be published in a peer-reviewed journal upon the completion of the trial. As of June 2021, we have recruited 49 patients in the OnkoFit I trial and 12 patients in the OnkoFit II trial. Per our recruitment plan, 50 patients are expected to be recruited in both OnkoFit I and OnkoFit II trials per year. The reasons for the more rapid recruitment in the OnkoFit I trial than in the OnkoFit II trial are manifold. One reason is the prioritization of the OnkoFit I trial over the OnkoFit II trial during the COVID-19 pandemic. Another reason is the decline of the number of patients who are eligible for the OnkoFit II trial during the pandemic. However, we expect that the recruitment rate in the OnkoFit II will accelerate soon.

Trial Implications
The benefit of physical activity both during and after various kinds of cancer treatments has been shown in several randomized trials. Physical activity not only improves HRQoL but also has the capability to reduce the severity of negative treatment effects. With regard to patients undergoing radiotherapy, the feasibility and positive impact of combining radiotherapy with physical activity has been observed, and further clinical studies are currently being conducted (Exercise Therapy in Radiation Therapy [EXERT] trial; NCT03905356) [28]. However, measured variables as well as definitions of end points greatly vary between different studies on using activity trackers in cancer care, and there is a great need for randomized trials that clarify optimal time points as well as strategies [24]. The rationale behind the OnkoFit trials is to test a physical activity program that requires only few resources and is easy to implement even in the workflow of departments that treat a very large numbers of patients (≥1000 patients per year). Even though there is no doubt that a one-to-one supervised physical activity program that is conducted over several weeks would be desirable, this is often not feasible due to costs, limitations in geographical reach, and a lack of qualified personal [29]. Moreover, in a qualitative interview study conducted by Hardcastle et al [30], survivors of breast cancer expressed that they would favor home-based programs for exercise. The high acceptance of activity trackers among patients with cancer has been reported previously [31][32][33]. In our pilot study, which investigated the feasibility of continuous activity monitoring, 19 of 23 patients regularly used a commercially available activity tracker during radiotherapy. In the same study, we observed very plausible results regarding changes in physical activity during the course of treatment [33]. These results are also supported by a study conducted by Ohri et al [34], who showed a correlation between decreasing step counts and a forthcoming need for the inpatient treatment of patients with lung cancer during radiotherapy.
We see several advantages in an activity tracker-based training program. First, most activity tracker devices are easy to use and can be intuitively used by most older patients without any previous experience with such devices. Second, these devices provide real-time feedback to patients, which results in self-awareness and motivation. Third, the quantitative measures of physical activity levels can be remotely shared with caregivers, thereby providing an objective view on a patient's constitution and whether activity goals have been met [17]. The OnkoFit I and II trials have been designed as 3-arm trials. By introducing an arm in which patients receive an activity tracker but no predefined goals, we hope to study whether patients in the interventional arm, which provides weekly targets, actually have a higher level of physical activity and whether any potential effects that are observed at the end of the study are associated with the intervention.
Since acute and long-term side effects vary widely between patients with breast cancer undergoing radiotherapy and patients receiving preoperative or definitive radiotherapy, different end points were defined for the OnkoFit I and OnkoFit II trials. Additionally, the time points for assessment vary. In the OnkoFit II trial, the end point will be assessed 6 months after the end of radiotherapy and therefore later than in the OnkoFit I trial (the assessment will be conducted at 3 months), since patients receiving preoperative radiotherapy may still be in reconvalescence from surgery 3 months after the end of radiotherapy.
Patients undergoing radiochemotherapy in a neoadjuvant setting are at risk of increased side effects during and after concomitant surgery if they experience (severe) toxicity after the completion of radiation treatment. The concept of prehabilitation, which is defined as a rehabilitation program that is initiated before treatment, has gained more and more recognition in the field of oncology in recent years [35]. In two studies on colorectal patients, a prehabilitation program markedly improved functional recovery after resection, thereby underscoring the program's potentially large benefit for patients [36,37]. In the context of neoadjuvant treatment, a study is currently underway for evaluating the effect that a concomitant prehabilitation program (conducted during chemotherapy) has on treatment outcomes and morbidity in patients with ovary cancer [38]. Similarly, another group has evaluated the effect that a prehabilitation program has on functional outcomes, particularly swallowing and the quality of life among patients undergoing radiochemotherapy for head and neck cancer [39]. Available data have been recently summarized by Squires and colleagues [40], with an emphasis on cardiovascular health after cancer therapy. In the context of the OnkoFit II trial, we aim to investigate whether the proposed fitness tracker-based exercise program can function as a prehabilitation program for patients undergoing neoadjuvant radiotherapy and thus decrease the number of postoperative complications.
One of the limitations of the proposed study is the lack of a supervised preintervention and postintervention fitness test (eg, ergometry), which would provide a more objective perspective on the fitness levels of patients than the fitness questionnaire-the one that will be used in our trials-alone. Furthermore, the trials were designed as monocenter trials that will be conducted at a single institution, which might limit the generalizability of the data.

Conclusion
The randomized OnkoFit I and II trials will investigate a unique approach to conducting a patient-executed fitness program during and after radiation therapy. In these trials, activity trackers will be used to improve HRQoL; reduce the severity of treatment side effects, including CRF; and improve the preoperative health status of patients. The findings of these trials will help to further our knowledge on combining exercise therapy with radiation treatment that focuses on the patient point of view.