Background: Elderly cancer patients may have important benefits from innovative treatments. However, they are often barred from clinical trials because of highly selective eligibility criteria, or due to biased and subjective physician standpoints including reluctance to invite elderly patients and fear of excessive toxicity. Indeed, geriatric assessment has been increasingly recognized as predictive and prognostic instrument to detect frailty in older adults with cancer. In this perspective, the G8 score is a simple and reproducible instrument to identify elderly patients who should undergo full geriatric evaluation. The aim of our study was to evaluate the impact of frailty assessment by the G8 screening tool on the outcome of onco-hematological patients. Methods: Between January 2017 and December 2020 the G8 screening tool was administered to patients, aged >65 years, referred to our center for solid and hematological malignancies. G8 score was assessed at the time of first access. The primary endpoint was overall survival. Multivariate analysis was performed according to G8 score, age, tumor type, stage and treatment. Results: In the observation period, 430 patients were screened for frailty by G8; median age was 77 years (65-92); of these, 331 (77%) had a G8 score <14. Pts with solid tumors were 310 (72%), 175 (57%) of whom had metastatic diseases; 227 (73%) had a G8 score <14. Pts with hematological malignancies were 120 (28%), 100 (83%) of whom had a G8 score <14. Systemic therapy was administered to 336 patients (78%). At a median follow up of 7.2 months (range 1 to 52) 101 pts (24%) were dead. Median overall survival (mOS) was 27 months (1-52+).Patients with solid tumors, classified as frail by a G8 score <14 had a 3-fold risk of death compared with those with G8 > 14 (OR 3.26, CI 95 1.5-7.2, p = 0.003). Conversely, this increased risk was not observed in hematological malignancies (OR 1.4, CI 95 0.4-4.6, p = 0.57). By multivariate analysis, G8 score was associated with a worse prognosis only in patients with solid tumors. Conclusions: Our analysis suggest that elderly frail patients with solid tumors have a significantly increased risk of death as compared to elderly fit patients. Conversely, no impact of frailty, as assessed by a G8 score < 14, was evident in elderly patients with hematological malignancies.