Four frailty classification systems show good discrimination for elderly patients with cancer

The first frailty classification put forth by the International Society of Geriatric Oncology, or SIOG1, showed the best discrimination of several classification systems used on older patients with a wide array of cancers, according to results of a study.

All four classifications included in the study had strong prognostic performance.

“Older patients with cancer raise therapeutic challenges, because they constitute a heterogeneous population with various combinations of comorbidities, disabilities and geriatric syndromes that contribute to frailty,” Emilie Ferrat, MD, PhD, of the Université Paris-Est Créteil, and colleagues wrote.

“However,” they wrote, “there is no consensus about the best means of measuring frailty.”

Ferrat and colleagues compared frailty classifications for 763 elderly patients (mean age, 80 years; 52.4% men) with cancer among four commonly used systems: SIOG1, SIOG2, Balducci and a latent class typology. Researchers reviewed 1-year mortality and 6-month unscheduled hospital admissions.

All four classifications demonstrated good discrimination for 1-year mortality (C-index ≥ 0.70), with SIOG1 showing the best discrimination.

All classifications also showed good discrimination for 6-month unscheduled admissions (C-index ≥ 0.70).

However, agreement among classifications varied widely for sorting patients into two (fit vs. vulnerable, frail and fit or vulnerable vs. frail) or three (fit, vulnerable or frail) categories, ranging from very poor (κ ≤  0.2) to good (0.6 < κ  ≤  0.8). The best agreement occurred between SIOG1 and the latent class typology, and between SIOG1 and Balducci.

“Despite poor to moderate agreement among the four frailty classifications of older patients with cancer, performance in predicting 1-year overall mortality and 6-month unscheduled admissions was consistently good when evaluated in a large cohort of in- and outpatients with untreated cancer at various sites,” the researchers wrote. “The observed variations in agreement and performance across tumor sites suggest means of optimizing performance and better characterizing frailty. Studies of clinical impact are needed to determine whether classifications deserve to be integrated into the cancer treatment decision-making process.” – by Andy Polhamus

Disclosure: Ferrat reports no relevant disclosures. Please see the study for a full list of all other authors’ relevant financial disclosures.