Practical Geriatric Assessment in Older Adults With Non-Small Cell Lung Cancer Undergoing Stereotactic Body Radiation Therapy

Comprehensive geriatric assessment (CGA) is one proven mechanism for delineating baseline care needs and improving outcomes in older adults with lung cancer. This type of geriatrician-led assessment, which captures functional ability, health, and socio-environmental situation, can be used to identify vulnerable older adults for whom tailored interventions might optimize care. However, CGA can be resource-intensive to perform, and may not practically possible in all settings, particularly given national shortages in geriatricians. At BWH/DFCI, retrospective work among patients with stage I-II NSCLC suggests fewer than 5% of patients receive CGA, despite 76% meeting national guidelines for this type of evaluation. These findings underscore the practical challenges of assessing geriatric needs, even in high-resource settings. To address these barriers, abridged instruments capturing the key domains of the CGA which can be completed by any provider within 10-25 minutes have recently been developed, including the practical geriatric assessment (PGA) tool. This screening tool, which is now recommended by American Society of Clinical Oncology (ASCO), the International Society for Geriatric Oncology (SIOG), and the Cancer \& Aging Research Group (CARG), has the potential capacity to delineate relevant baseline features in geriatric populations without creating undue provider burden. However, the feasibility of implementing the PGA and its acceptability to patients remain unclear. Further, relationships between the PGA and salient outcomes, including subsequent patterns of recommended care delivery, remain underexplored. To improve outcomes among high-risk subgroups interfacing with radiation oncology, including older adults with NSCLC undergoing SBRT, further interrogation of these practical factors is needed. Finally, the investigators will use facial photographs and audio-visual data from the PGA to develop and evaluate artificial intelligence algorithm(s) to identify vulnerable patients who might benefit from additional supportive care services; namely, FaceAge, a validated deep learning model capable of estimating biological age from still facial images.