(P 132) SARCOMA PATEINTS' VIEWS ON USING A RISK-PREDICTION MODEL TO INFORM TREATMENT DECISIONS.

Objective: Risk prediction models (RPM) are statistical tools which can predict the likelihood of a certain outcome or event for an individual based on their (clinical) characteristics. RPM can support (shared) decision-making by providing patients and clinicians with insights into the diagnosis or prognosis of the disease. The number of RPM in medicine increased exponentially over the last decade, and also in the field of soft-tissue sarcoma (STS). Despite efforts to develop, validate and update RPM, it is unknown how sarcoma patients perceive and accept the use of RPM in their decision-making process. This study therefore evaluates the views of STS patients on using RPM to inform clinical decision-making.

Methods: A convergent parallel mixed-methods design is used, combining the results from a survey among 60 STS patients and semi-structured interviews among 10-15 STS patients treated in three tertiary sarcoma referral centers in the Netherlands. In these centers, the PERsonalised SARcoma Care (PERSARC) RPM was used by sarcoma specialists in multidisciplinary tumour boards to guide treatment advice and in patients consultations, where the oncological/orthopedic surgeon informs the patient about his/her diagnosis and discusses benefits and harms of all relevant treatment options. Specifically for the latter reason, the PERSARC RPM was implemented in a new patient-friendly app called VALUE-PERSARC.

Results: Preliminary results based on 6 interviews with STS patients and 28 completed questionnaires showed that 89% of the patients perceived the PERSARC, embedded in the VALUE-PERSARC app as useful and the majority (88%) would use VALUE-PERSARC again. Only one-third (32%) reported that the RPM aided in clinical decision-making. The most mentioned reasons for STS patients to promote VALUE-PERSARC was that it helped in understanding the impact of different treatment options (25%) and provided insights into personalized risk estimates (21%). Interviews revealed that the RPM clarified the information provided by the clinician, but not per se to support (shared) decision-making often because only one treatment option was discussed with the PERSARC RPM.

Conclusion: Patients perceived the use of VALUE-PERSARC as helpful to support information provision but it did not facilitate (shared) decision-making, mostly because patients did not feel there was a choice in treatment. Our findings highlights the importance of studies investigating the influence of RPM in daily clinical practice, to evaluate if it worked as intended to actually support (shared) decision-making. Future research should explore how to best involve patients in their decision-making process when using a RPM.