Abstract 1833P
Background
Grip strength (GS) and the Short Physical Performance Battery (SPPB) are both objective non-invasive tests used in Geriatric Assessment (GA) for older cancer patients and they add great value to the GA. Studies have shown that abnormalities in GS & SPPB are associated with morbidity, mortality. We explored the value of GS, SPPB or both in predicting chemotherapy completion (dose reduction, therapy delay) using univariate and multivariate logistic regression.
Methods
We retrospectively reviewed all cancer patients ≥ 65 years, seen in the geriatric oncology clinic at our academic cancer centre for pre-therapy assessment (chemotherapy or chemoradiation only). GS and SPPB were assessed using standardized protocols. We reviewed electronic charts at the end of therapy, looking for completion/modification in doses and cycles.
Results
85 patients with mean age 78y, 51% female were recruited; the most common cancer was lymphoma (30%). The majority received therapy for curative intent (54%). Mean GS for males and females were 17.3kg and 29.7kg, respectively, with abnormal scores in females (34.1%) and males (31.7%). Mean SPPB was 8.1, abnormal in 55.3%. Abnormal GS and/or abnormal SPPB in 67.1%. Dose reduction was required in 62.4% in at least one cycle of therapy and 30% needed treatment delay of at least 7 days at least once. In multivariable logistic regression the only boardline significant factor to predict dose reduction was high comorbidity (odds ratios (OR) 4.13, 95% confidence interval (CI) 0.96-17.6, P=0.055) but GS, SPPB or both, age, gender, BMI and other geriatric domains were not predictive. However, gender (OR 3.91, 95 %CI, 1.24-12.34, P=0.019), abnormal GS and/or SPPB (OR 7.35, 95% CI, 1.72-31.37, P= 0.007) and one of the eight geriatric domains; social supports (OR 4.04, 95% CI, 1.22-13.38, P= 0.022) were factors in predicting therapy delay.
Conclusions
GS and SPPB are simple tests which can be used to predict chemotherapy delay but not subsequent dose reduction in older cancer patients. Validation studies with larger samples in multiple centres should be done.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.