Simplified Geriatric Assessment Validated as OS Predictor in Elderly Patients With DLBCL

A simplified geriatric assessment offers a validated objective tool to assess fitness status and should be considered the new Elderly Prognostic Index (EPI) standard to predict overall survival (OS) in older patients with diffuse large B-cell lymphoma (DLBCL), according to findings from the Elderly Project study (NCT02364050) published in the Journal of Clinical Oncology.

The tool incorporates activities of daily living, instrumental activities of daily living, Cumulative Illness Rating Scale for Geriatrics, and age (≥80 or <80) to refine and improve the original geriatric assessment tool adopted by the Fondazione Italiana Linfomi (FIL) that categorized patients into 3 categories of fitness: fit, unfit, and frail.

Utilizing the simplified tool that also categorized patients into the same fitness categories as the original tool, the 3-year median OS was 65% (95% CI, 62%-68%) in the overall patient population (n = 1163); however, the median OS varied significantly based on fitness. The hazard ratio (HR) was 1.98 for unfit vs fit patients (95% CI, 1.55-2.54; P < .001), 3.27 for frail vs fit patients (95% CI, 2.52-4.22; P < .001), and 1.65 for frail vs unfit patients (95% CI, 1.26-2.14; P < .001). This translated to 3-year OS rates of 87% (95% CI, 81%-91%), 69% (95% CI, 63%-73%), and 42% (95% CI, 36%-49%) for fit, unfit, and frail patients, respectively.

“The development and validation of the EPI [Elderly Prognostic Index] on an independent series of patients provides clinicians with a unique tool to better account for the complexity of each older patient with DLBCL,” lead study author Francesco Merli, MD, managing director of the Department of Oncology and Advanced Technologies at Azienda Osperdaliera Santa Maria Nuova di Reggio Emilia in Italy, and co-authors wrote.

DLBCL is the most frequently occurring lymphoma subtype. Moreover, about 70% of patients with DLBCL are over the age of 65 years.

The integration of rituximab (Rituxan) to standard anthracycline-containing regimens has significantly improved survival for patients with DLBCL; however, older patients with DLBCL have poorer outcomes compared with younger patients. Moreover, age-related comorbidities limit available treatment options, which are associated with increased risk of treatment-related toxicity. Notably, despite this disparity, elderly patients are largely underrepresented in clinical trials, so physician preference plays a significant role in treatment selection.

The integration of geriatric assessment has provided a useful tool to gauge multiple geriatric conditions and subsequently tailor therapies in terms of patient goals, intensity, tolerance, and prognosis.

“FIL has been using the same [original geriatric assessment] in its studies on older patients with DLBCL for several years,” the authors wrote.

The Elderly Project was a prospective, multicenter, observational study aimed at evaluating how simplified geriatric assessment affected OS in older patients with newly diagnosed DLBCL.

The 3-step study prospectively validated the original geriatric assessment, refined the new, simplified geriatric assessment to account for confounding factors, and developed and validated a prognostic model that integrated patient- and lymphoma-related factors.

The study evaluated 1163 patients from 36 hematology/oncology centers of the FIL in Italy.

Patients had to have histologic confirmed DLBCL, be 65 years of age or older, provide informed consent, and have original geriatric assessment results available. Patients who had a diagnosis other than DLBCL, including grade 3b follicular lymphoma or high-grade lymphoma were not eligible for evaluation.

Patients were classified as fit, unfit, or frail per the simplified geriatric assessment criteria and the results of a Web-based calculator.

Overall, patients were a median age of 76 years (range, 65-94; P < .001), and 32% were over the age of 80 years (P < .001). Half of the patients (50%) were male (P = .032), 67% had stage III or IV disease (P = .217), 29% had more than 1 extranodal site (P = .795), and 20% had an ECOG performance status greater than 1 (P < .001). Most patients’ (55%) lactase dehydrogenase was above the upper limit of normal (P = .180), and 56% of patients had an International Prognostic Index (IPI) score between 3 and 5 (P = .004).

Notably, the protocol for simplified geriatric assessment did not include any recommendations for patient management, and treatment selection was left to the provider’s discretion and did not need to be informed by the results of the assessment.

Of the overall population, 54% (n = 652) were classified as fit at baseline, 28% (n = 334) were unfit, and 18% (n = 221) were frail. Additionally, 63% of patients previously received full-dose chemotherapy, 24% received prior reduced-dose chemotherapy, and 13% received prior palliative care. The full-dose chemotherapy was administered to 86%, 48%, and 16% of fit, unfit, and frail patients, respectively.

“The prognostic score was obtained [by] giving a weight to each variable according to its relative importance, derived from the z-Wald values found in the Cox PH [proportional-hazards] model. The weights were obtained [by] rounding the ratio, and the score was the sum of [the] weights. We thus obtained a score ranging from 0 to 8 that showed a good correlation with OS,” wrote the authors.

At a median follow-up of 30 months (range, 1-59), 354 deaths were reported from causes included lymphoma progression (n = 243), treatment-related toxicity (n = 74), secondary malignancy (n = 10), and unknown (n = 27).

Fitness level, IPI score, and hemoglobin level were found to be the most significant prognostic factors in building the EPI score. The EPI model was developed based on a 1065-patient cohort with a 3-year OS rate of 66% (95% CI, 62%-69%); the cohort excluded patients with incomplete data.

In addition to OS, the prognostic role for the EPI model was validated for progression-free survival and disease-free survival, the authors wrote.

External validation of 456 cases with a median follow-up of 30 months was conducted in Italy (n = 172), Australia (n = 204), and Brazil (n = 80). Complete data were available in 78% of cases. A total of 117 patients died on study, and the 3-year OS rate was 61% (95% CI, 55%-67%).

In the validation cohort, 22% were classified as low risk, 46% were intermediate risk, and 32% were high risk. The 3-year OS rates were 85% (95% CI, 71%-93%), 65% (95% CI, 55%-73%), and 44% (95% CI, 34%-54%), respectively. The HR was 2.14 in the intermediate-risk group vs the low-risk group (95% CI, 1.08-4.24; P = .029) and 2.18 in the high-risk group vs the intermediate-risk group (95% CI, 1.49-3.21; P < .001).

Notably, the simplified geriatric assessment takes less than 10 minutes to complete by the oncologist/hematologist during a patient’s regularly scheduled visit, the authors wrote.

“The time has come to consider the fitness status of older patients with DLBCL to better identify their treatment goals. The EPI is the first prognostic index to integrate fitness evaluation into prognosis in older patients with DLBCL, which contributes to improving patient assessment,” concluded the authors.

Reference

Merli F, Luminari S, Tucci A, et al. Simplified geriatric assessment in older patients with diffuse large B-cell lymphoma: the prospective Elderly Project of the Fondazione Italiana Linfomi. J Clin Oncol. Published online February 12, 2021. doi:10.1200/JCO.20.02465