Association between Geriatric 8 frailty, guideline treatment, treatment adherence, and overall survival in older patients with cancer (PROGNOSIS-G8).

Abstract

1623 Background: Frailty is frequent among older adults with cancer and may affect oncologic treatment tolerance. Frailty screening, with tools such as the Geriatric 8 (G8), is recommended to help guide clinical decision-making. While the G8 has been strongly associated with survival, its relationship with treatment adherence remains less clear. This study aimed to evaluate the association between G8-identified frailty and treatment outcomes in a large cohort of older adults with diverse cancer types. Methods: This single-center prospective cohort included adults, age ≥70 years, with solid cancers who underwent G8 screening at their initial oncology consultation. Treatment-related outcomes included one-year overall survival, first-line oncologic treatment adherence within 9 months, and whether patients were offered guideline treatment. Guideline treatment was defined as regimens consistent with recommendations from national guidelines for first-line oncologic treatment, allowing add-on protocol treatment, while less-than-guideline treatment referred to regimens not among first choices, often deemed inferior. Adherence to the doctor-patient selected treatment plan was defined as the absence of discontinuations, dose reductions after treatment initiation, or un-administered treatments (i.e., excluding delays). Data on demographics, comorbidity, cancer diagnosis, treatment, and survival were extracted from medical records. Associations between G8 frailty (≤14/17 points) and outcomes were analyzed using multivariate logistic regression and Cox proportional hazards regression, adjusting (adj.) for confounders. Results: Among the 1,398 patients screened, 65% were frail. Frailty doubled the risk of death at one year (adj. HR 2.0, 95% CI 1.7-2.4, p < 0.001). Frail patients who adhered to less-than-guideline treatment had a 69% lower mortality risk compared to frail patients unable to adhere to guideline treatment (adj. HR 0.31, 95% CI 0.21-0.47, p < 0.001). Non-frail patients were more likely to adhere to treatment (adj. OR 2.38, 95% CI 1.49-3.81, p < 0.001) and were more often offered guideline treatment (adj. OR 1.98, 95% CI 1.28-3.06, p = 0.002) compared to frail patients. Lastly, when receiving guideline treatment, non-frail patients had significantly better adherence than frail patients (adj. OR 3.08, 95% CI 1.72-5.52, p < 0.001). Conclusions: G8 frailty screening effectively identifies older adults at a higher risk of treatment non-adherence and mortality, facilitating tailored treatment approaches. Our findings suggest that frail patients may benefit from initial less-intensive treatments with potential escalation to improve adherence and survival. Implementing G8 screening in routine practice addresses the unique challenges associated with frailty, ensuring more effective, equitable care for at-risk older adults.