Abstract
Haematological neoplasias in elderly represent a major challenge, as median age at diagnosis is 70+ years in frequent and clinically relevant diseases such as acute myeloid leukaemia (AML), multiple myeloma or chronic lymphocytic leukaemia (CLL). Thus, algorithms for individualised treatment of elderly are urgently needed. The G8 screening tool was developed to separate fit older cancer patients able to receive standard treatment from those that should undergo a geriatric assessment to guide tailoring of therapy. In this study the discriminative power and prognostic relevance of G8 in a well-defined cohort of elderly patients with a haematological malignancy was analysed.
Between September 2009 and May 2013, in 108 consecutive patients aged ≥67 years diagnosed with blood cancer at the Department of Internal Medicine V (Haematology and Oncology), Innsbruck Medical University a multi-dimensional geriatric assessment (MGA) was performed. The most frequent entities included were AML (29%), myelodysplastic syndromes (23%) and aggressive or indolent Non-Hodgkin lymphomas (29 and 12%, respectively). MGA assessed the dimensions (instrumental) activities of daily living (ADL/iADL), cognition, mood, nutritional status, mobility, polypharmacy and social support. The G8-scoring was administered in parallel, using a cut-off ≤14. The discriminative power of the G8 for detecting the presence of an impaired geriatric assessment (defined as impairments in ≥2 geriatric domains) was calculated from a 2x2 table. For comparisons between patients with a normal G8 score and those with an impaired G8, the chi-square test and risk analysis were used for nominal and ordinal variables, as well as for continuous variables with a non-normal distribution; for continuous variables with a normal distribution, the Student t-test was used. The impact of variables on mortality was evaluated by uni- and multivariate Cox-regression analyses.
Prevalance of geriatric conditions in haematological malignancies was high. Applying a cut-off ≥2 impaired domains, 70% of patients were scored as being impaired. Domains most often involved were iADL (iADL ≤6: 45%), polypharmacy (≥5 medications: 65%) and malnutrition (BMI ≤18kg/m2 and/or weight loss ≥1kg in last 3 months: 45%). An impaired G8 was observed in 61% of cases and revealed a significant correlation with MGA (cc=-0.624; p <0.001). However, the G8 lacked discriminative power for impairments on MGA, with a sensitivity of 69%, a specificity of 79%, a negative predictive value of 50% and a positive predictive value of 89%. One-year mortality was correlated with diagnosis of AML (hazard ratio (HR) 1.24, 95%-confidence interval (CI) 1.06-1.44; p=0.01)), impaired mobility (timed up and go test ≥12 seconds) (3.20 (1.26-8.14); p=0.02) and risk of malnutrition (1.31 (1.07-1.37); p=0.01) in uni- and in multivariate analyses. Remarkably, G8 was an independent predictor of mortality within the first year after inclusion, displaying a HR of 3.93 (95%-CI 1.67-9.22; p=0.002).
This is the first report on the clinical and prognostic relevance of G8 in a well-defined cohort of elderly patients with a haematological malignancy. Clinical relevant restrictions were identified in a substantial proportion of patients. Due to the high prevalence of geriatric impairments, the G8 lacked discriminative power. However, G8 appears to be a powerful prognosticator and might thus represent a useful tool in treatment decisions. This novel finding certainly deserves further exploration.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
