Integration of an App-Based Cognitive Evaluation Program into Radiosurgery Practice: Outcomes and Patient Survey Analysis from a Prospective Observational Study.

Abstract

Cognitive decline is a significant consequence of stereotactic radiosurgery (SRS) in patients with brain metastases. This is often underrecognized and understudied outside of a formal clinical trial setting as traditional methods are often logistically difficult and need specialized personnel and additional time. To address these challenges, we implemented a prospective study (NCT05504681), incorporating an innovative app-based solution to monitor patient assessments over time, collect patient data easily, and be suitable for cross-cultural use in multiple languages for monitoring post-SRS cognitive decline. Patients undergoing SRS from December 2021 to October 2022 were enrolled in this study and completed feedback surveys. The assessments consisted of learning and memory (Hopkins verbal learning test-revised [HVLT-R]), attention and processing speed (Digit symbols modalities test [SDMT]), verbal fluency (Controlled oral word association test [COWAT]) and executive function (Trail making test [TMT]). Baseline and 3-month follow-up testing were conducted in conjunction with routine imaging and clinical assessments. Neurocognitive deterioration (ND) was defined as a decrease of ≥2 standard errors of the mean in any of the tests, without evidence of new intracranial disease. Any differences between the baseline and follow-up assessments were confirmed using the reliable change index. A total of 30 patients with median age of 68 (range: 47-87) were enrolled. The median KPS was 90 and 60% were female. 43% of participants had a high school or equivalent level of education. 19 (63%) patients were Hispanic and 43% tested in Spanish. The median number of treated lesions was 2 (1-13) and 7 (23%) patients were treated to ≥5 lesions. The median mean left and right hippocampal doses were 0.3 Gy (0-3 Gy) and 0.2 Gy (0-5.2 Gy), respectively. Overall, 50% of patients met criteria for ND at 3 months. 20% of patients showed ND in HVLT-R-immediate recall (IMM), 23% in HVLT-R-delayed recall (DR), 13% in HVLT-R-recognition (Rec), 3% in COWAT and 20% in TMT. None of the patients had ND in SDMT. The mean relative decline was 28% for HVLT-R- IMM, 34% for HVLT-R-DR, 6% for HVLT-R- Rec, 25% for COWAT, and 21% for TMT. There was no significant association between the number of treated lesions and ND. Based on the feedback results, the majority of patients found the system easy to use (94%) and relevant to their care (73%). Most patients reported that app-based evaluation improved discussions with clinicians (77%), made them feel more in control of their care (79%), and they would recommend the system to other patients (87%). Our study showed a high incidence of ND using an app-based tool in a patient who underwent SRS and from diverse education and language backgrounds. Patient survey results demonstrated that the tool was easy to use and relevant to the patient's care, especially due to improvements in discussions with clinicians and a sense of being in control of their own care.