Comprehensive Geriatric Assessment (CGA) in Elderly Patients with Hematological Neoplasia: CGA and Not Age Predicts Mortality - Results from in-GHO^®-Registry

Background: The internet based registry of IN-GHO^α prospectively collects data of elderly cancer patients (pts) ³70 yrs. We report data from 450 pts with hematological neoplasia (HE) out of 1580 cancer patients registered currently. To gain more insight into feasibility of treatment in elderly patients, results of a comprehensive geriatric assessment (CGA), physicians’ and pts’ judgement of fitness, and outcome measures were studied.

Methods: Besides demographic data, activities of daily living (ADL), instrumental activities of daily living (IADL), Karnofsky-Performance-Statuts (KPS), co-morbidity (Charlson score), medication (number of drugs not related the HE treatment), mobility (Timed-Up&Go-Test), cognition (MMSE), and depression (SCID screening) results were collected. Age, results of CGA, physicians’ rating (fit vs. compromised vs. frail) and pts’ self rated fitness (lickert scale for 1= very fit to 6 very unfit) for treatment were analysed for their association with death. Follow up assessment was scheduled 8–12 weeks and 5–7 months after inclusion.

Results: 450 patients, mean age 77.7 years (SD 5.7; range 69–97), 43.3%male, were included. Main diagnoses were aggressive lymphoma n = 135 (30%), chronic lymphocytic leukemia n = 66 (14.7), multiple myeloma n = 54, acute myeloid leukemia n = 27 (6.0%), and others. Physicians rating of patientś fitness for treatment: fit 54.4%, compromised 39.1%, frail 5.5%. Patients rating of patientś fitness for treatment: 1 = 19.6%, 2 = 31.8%, 3 = 25.1%, 4 = 13.6%, 5 = 3.8%, 6 = 1.1%. During follow up (median 178 days), 98 pts (21.8%) patients died: mean age (77.5 vs. 78.5; p=0.219), mean BMI (25.6 vs. 25.3; p=0.228), mean Charlson-score (2.0 vs. 2.3; p=0.28) and mean number of drugs (4.0 vs. 4.2; p=0.14) were not associated with death. However significantly associated with death during follow up were mean ADL score (94.0 vs. 87.4; p=0.047), mean IADL score (7.1 vs. 6.3; p<0.001), mean KPS score (83 vs. 74; p=0.001), mean MMSE score (26 vs. 24; p=0.03), categorical Timed-Up&Go-Test (< 10 sec 18% vs. > 20 sec 38%; p<0.001), categorical SCID screening (no depression 18% vs. depression 31%; p<0.001), physicianś rating of fitness (fit 18.8%, compromised 21.6%, frail 52.0%; p=0.001) and patientś rating of fitness (lickert scale 2.3 vs. 3.4; p<0.001).

Conclusion: Patients characteristics as assessed by CGA and not age are associated with death in elderly patients with hematological malignancies. Results of CGA should be validated in treatment algorithm for this population.