Symptom severity reported by older adults with cancer and its impact on survival prognostication

Abstract

Introduction

Symptom burden is a crucial factor for survival prognostication, but older adults may report fewer and less severe symptoms than younger patients. We investigated the impact of age on symptom severity and its prognostic implications in the prognostic model for advanced cancer (PRO-MAC).

Materials and Methods

Eight hundred forty subjects with incurable cancer, who were reviewed by palliative medicine at a tertiary university hospital, were categorized into four groups based on age: < 65 (young), 65–74 (young-old), 75–84 (old-old), and ≥ 85 (oldest-old) for comparison. The study compared individual symptoms reported on the Edmonton Symptom Assessment System-revised (ESAS-r) and analyzed the significance of each prognostic variable in the PRO-MAC model using multivariate Cox proportional hazards regression analysis for all age categories. We generated the log-rank test for each age group to evaluate the effectiveness of PRO-MAC across age groups.

Rsults

The older adults had significantly lower unadjusted mean scores in pain, tiredness, anxiety, wellbeing, and ESAS-r summative score compared to young adults. There was a higher proportion of older adults with no symptoms. The ESAS-r summative score remained a significant prognostic factor only in the young and young-old groups. More participants in the older age groups had dementia, which may impact reporting of symptom severity. Despite these findings, the PRO-MAC model remains applicable across all age groups as a multi-domain 90-day survival prognostication tool.

Discussion

Symptom burden, including pain, tiredness, anxiety, wellbeing, as well as summative ESASr scores, are lower in older adults with cancer. This could affect the prognostication of survival using symptom burden. However, a multi-dimensional prognostic model like PRO-MAC remains effective for older adults with advanced cancer. Further research is needed to explore tools that can help older adults express the severity of their symptoms and incorporate this into survival prognostication.