The use of G8 and VES13 as screening and prognostic tools in older Indian patients with cancer.

Abstract

e24024 Background: The ‘G8’ and ‘VES13’ are quick and easy-to-use screening tools, developed and validated in Western patients for predicting abnormalities in the subsequent geriatric assessment (GA). These tools predict functional decline and survival in older patients with cancer, but their applicability in our older Indian patients with cancer is not known. We performed this study in an attempt to validate the use of these screening tools in our patient population. Methods: An observational study with a retrospective and prospective cohort of 308 patients, aged 60 years and above, presenting to the Geriatric Oncology clinic at Tata Memorial Hospital, Mumbai, between June 2018 - November 2020. Patients either planned for or recently started on systemic therapy were enrolled and underwent the G8 & VES13 screening tools followed by a GA. The primary objective was to determine the appropriateness of the use of G8/VES13 screening tools to detect an abnormal GA. Our secondary objectives were to determine the optimal G8 cut-off value (using the AU-ROC curves) in our patient population, correlation between abnormal G8/VES13 scores and OS and to assess the utility of combining the G8 and VES13 scores (i.e. abnormal score on either of the two screening tools) to predict for an abnormal GA and poorer OS. Results: The abnormal G8 cut-off score appropriate for our population was 12. This revised cut-off score was compared with the international standard (Abnormal G8 < / = 14). With abnormal G8 cut-offs at < / = 14 the sensitivity, specificity & overall accuracy was 84%, 18% & 80% respectively; the corresponding values were 57%, 88% & 58% with cut-off < 12. An abnormal G8 ( < / = 14) and VES13 score correlated with poor ECOG PS (PS 2/3, p = 0.0001 [G8], p = 0.00001 [VES13]) and CARG high risk scores (p = 0.006 [G8], p = 0.005 [VES13]) however it did not correlate with an abnormal GA (p = 0.736 [G8], p = 0.195 [VES13). The median OS in patients with abnormal ( < / = 14) vs normal G8 scores was 13m vs 18m respectively (HR 0.777). Abnormal G8 cut-off scores < 12 correlated with a poor ECOG PS (p < 0.00001), CARG high risk scores (p = 0.006) and also with an abnormal GA (p < 0.001). The median OS in patients with abnormal ( < 12) vs normal G8 scores was 11m vs 17m (HR 1.658). Abnormal G8 ( < 12) + VES13 scores also correlated with abnormal GA (p 0.0001) and predicted for worse survival outcomes (Median OS 13m vs 17mHR 1.641). Abnormal VES13 scores also predicted for shorter survival (Median OS 10m vs 17m, HR 1.097). Conclusions: An abnormal G8 cut-off score < 12 is appropriate in our older Indian patients with cancer as compared to the internationally validated cut-off of < / = 14. This revised G8 cut-off score ( < 12) predicts for the presence of non-oncological vulnerability, poorer OS and translated to a 35% reduction in the number of patients undergoing a full GA. Combined with VES13, the G8 can be used as a screening tool which may help in optimal resource utilization especially in busy Indian centers.