Gait speed and recommended treatment intensity among older adults with blood cancers.

Abstract

12041 Background: Gait speed identifies frailty and predicts survival among older adults with hematologic malignancies (Liu, Blood, 2019). It is not known if gait speed correlates with the intensity of oncologists’ recommended treatment in this population. Methods: From 2/2015-11/2019, patients ≥75 years presenting for an initial hematologic malignancy consultation at the Dana-Farber Cancer Institute were approached for a screening frailty assessment including a 4-meter gait speed test, reported as <0.4, 0.4-0.6, 0.6-0.8, or >0.8 meters/second (m/s). Faster gait speed is associated with less frailty and predicts better survival. Gait speed was not reported to the oncologist. Treatment recommendations were categorized into standard, reduced, or no therapy based on NCCN guidelines, as applicable. Gait/treatment intensity “mismatches” were characterized as patients with lowest quartile gait speed recommended standard intensity and highest quartile not recommended standard intensity. Multivariable regression was performed to assess if gait speed predicted treatment intensity (controlling for age, sex, ECOG performance status [PS], and disease type). Results: Of 786 patients enrolled, 408 required active treatment where NCCN guidelines vary by fitness. Mismatches were seen in 26.7% of patients (Table: column percentages with 95% CI, mismatches starred): 10 (21.3%) with lowest quartile gait speed recommended standard intensity and 99 (55.0%) with highest quartile recommended reduced or no therapy. In multivariable analysis, PS was predictive of no therapy as compared to standard intensity (all p<0.02) and age was predictive of reduced as compared to standard intensity (p<0.01); gait speed was not reliably predictive in either case. Conclusions: In this large cohort of older adults with hematologic malignancies, gait/treatment intensity mismatches occurred in over one-quarter of patients. Oncologists’ recommendations were predicted by age and PS but not gait speed. Given that gait speed is a strong predictor of survival in this population, oncologists should integrate it to minimize over- and under-treatment when making treatment recommendations. [Table: see text]