Geriatric oncology screening tools for CGA-based interventions: results from a phase II study of geriatric assessment and management for older adults with cancer

Introduction: Screening tools in geriatric oncology have traditionally been studied for their ability to identify patients who have abnormal domains on a comprehensive geriatric assessment (CGA). However, an alternative outcome of identifying patients who would receive CGA-based interventions could improve selection of patients whose management will be altered by a CGA. The objective of this study was to assess the performance of three geriatric oncology screening tools for their ability to predict for CGA-based interventions.

Materials and methods: G8, Vulnerable Elders Survey (VES-13) and a modified frailty phenotype (mFP) screening tools were collected prospectively for patients enrolled in a phase II trial of geriatric evaluation and management. Interventions were defined as a new clinical diagnosis, change in management of a comorbidity, or referral to an allied health professional. Performance characteristics were calculated for each screening tool based on the outcomes of ≥2 abnormal CGA-domains and ≥1 CGA-based interventions.

Results: Discordance between the outcomes was seen in 31.9% of patients. Using the outcome of ≥2 abnormal CGA-domains, the G8 was most sensitive at 0.73 while VES-13 and mFP were both 1.0 specific. Using the outcome of CGA-based interventions the most sensitive tool was still the G8 at 0.64 and the most specific was the mFP at 0.80.

Discussion: All screening tests' performance characteristics for the G8, VES-13 and mFP were lower for the outcome of CGA-based interventions than for the traditional outcome of abnormal CGA-domains. Significant discordance between the outcomes highlights the difficulty with trying to predict which patients will truly benefit from a CGA.