Background \& Rationale:
Sarcoma is a rare form of bone and soft tissue cancer, making up about 1% of all adult cancers. Management typically involves radiation therapy, chemotherapy, and surgical resection, with or without limb reconstruction. The diagnosis, intensive treatment regimen, and surgery-induced deficits significantly impact patient quality of life and function.
Evaluating outcomes in sarcoma is challenging due to disease heterogeneity, complexity, and variation in surgical strategy. Patient-reported outcomes measures are crucial for evaluating treatment outcomes and facilitating patient-centered care. Traditional validated PROMs, like the Toronto Extremity Salvage Score (TESS), have significant ceiling effects. PROMIS assesses physical, mental, and social well-being using item response theory in a standardized manner for precise and sensitive measures. PROMIS was developed with oncology research considerations and has been validated in both orthopaedic and oncology populations. However, the significance of PROMIS in orthopaedic oncology is not well understood.
Assessing patient reported outcomes is limited by survey completion rates, particularly in the oncology patients who have multiple medical appointments and care teams. To address these challenges, our team has developed ACTIVATION (Activity Capture To Investigate Voluntary ActiviTy In Orthopaedic populatioNs), a smartphone application (app) tailored to musculoskeletal oncology. Unlike general fitness or commercial health applications, ACTIVATION passively captures mobility metrics and remotely administers PROMIS questionnaires. The continuous data collection aligns real-world mobility with clinical recovery timelines, offering a more precise view of patient outcomes.
The overall aim of this study is to validate PROMIS in sarcoma patients undergoing surgery, and to evaluate a novel smartphone app for questionnaire administration and collection of mobility data.
Hypothesis:
PROMIS and mobility metrics will provide a holistic and granular view of recovery trajectories, and the use of a novel smartphone application will improve patient engagement and data collection.
Aims:
1. Compare PROMIS to the current gold standard, the Toronto Extremity Salvage Score (TESS), in sarcoma
2. Determine the minimal clinically important difference (MCID) for PROMIS in sarcoma
3. Assess the feasibility and usability of an novel app for remote PROMIS and TESS administration, and mobility data collection
4. Correlate PROMIS and TESS scores with mobility metrics to define recovery trajectories
5. Identify demographic and socioeconomic factors that influence app engagement
Methods:
Recruitment and Sample Size: The investigator will recruit all eligible subjects identified by musculoskeletal oncology surgeons at the four study sites. The investigators anticipates enrolling approximately 800 sarcoma patients based on current surgical volumes and patient capture rates for collecting quality of life and functional data.
Data Collection:
Patients will be asked to download the ACTIVATION app on their smartphone and complete a demographic intake form. Clinical data will be obtained through the local electronic medical record. The TESS and PROMIS global health, self-efficacy, physical function, upper extremity function, and pain interference questionnaires will completed through the ACTIVATION app. Completion of these forms will be prompted via push notifications pre-operatively and at 6 weeks, 3 months, 6 months, and 12 months post-operative. Additionally, the app will passively collect mobility metrics from the smartphone operating system including step count, distance travelled, flights climbed, gait asymmetry, distinct activity periods (\>10 minutes), and total time active. In patients who have a compatible smart watch, the investigators will also collect physiologic parameters including VO2 max, heart rate, and heart rate variability. At the end of study participation, two additional surveys will be administered. The survey will collect feedback about the overall efficacy of the app and assess participant phone carrying habits.
Statistical Analysis:
Longitudinal changes in PROMIS scores will be analyzed using linear mixed-effects models with time and patient-level covariates (e.g. age, sex, race, socioeconomic status) as fixed effects and patient as a random effect. Logistic regression will identify predictors of PROMIS completion. MCIDs for PROMIS domains will be calculated using an anchor-based method with the Global Rating of Change. The sensitivity of PROMIS will be compared to TESS using correlation coefficients, Bland-Altman plots, and linear regression models.
Mobility data will undergo principal component analysis to derive a composite index capturing the key features of interest. ROC curves will assess whether mobility changes predict meaningful changes in PROMIS scores. Additionally, subgroup analysis will be performed to determine if differences in recovery exist based on tumor location, type of surgery, and adjuvant cancer therapies. Missing data will be evaluated for randomness and addressed using multiple imputation or maximum likelihood estimation as appropriate.
