Validation of POD24 As a Robust Early Clinical Endpoint of Poor Survival in Follicular Lymphoma: Results from the Follicular Lymphoma Analysis of Surrogacy Hypothesis (FLASH) Investigation Using Individual Data from 5,453 Patients on 13 Clinical Trials

Introduction

Follicular lymphoma (FL) is the most common indolent lymphoma with a median survival approaching 20 years. However, there is significant clinical heterogeneity with a subset of patients experiencing transformation, early recurrence or refractory disease. Data from the National LymphoCare Study and others identified that progression of disease within 24 months of diagnosis in patients treated with chemoimmunotherapy (POD24) is associated with poor subsequent overall survival (OS). Using the Follicular Lymphoma Analysis of Surrogacy Hypothesis (FLASH) data, we sought to i) evaluate the association between FL international prognostic index (FLIPI) and other baseline factors on progression-free survival within 24 months of trial enrollment (PFS24) and ii) to validate POD24 as an early clinical endpoint in FL.

Methods

We performed a pooled analysis of 13 randomized clinical trials of patients in both the pre and post rituximab era (total sample size of 5,453, Table). Logistic regression model was used to evaluate the association between FLIPI, gender, and performance status (PS) with PFS24. Cox regression with POD24 as a time-dependent covariate was used to evaluate the association between POD24 and subsequent OS. An additional landmark analysis was conducted to evaluate the association of POD24 on OS for the subset of patients who were alive at 24 months post trial registration. Baseline Beta 2 Microglobulin (B2M) data were available for 9 of the 13 studies (total sample size of 4,361). Additional analyses including B2M along with other baseline factors were conducted on this subset of patients.

Results

29% of patients progressed and 2.5% died without progressive disease within 24 months from trial registration. Patients alive without progression at 24 months were younger and more commonly had favorable PS, limited stage, low FLIPI risk score, normal baseline hemoglobin, and normal baseline B2M. A multivariable logistic regression model indicated that being male (odds ratio (OR) = 1.35; 95%CI = (1.19 - 1.52); p < .01), having PS >= 2 (OR = 1.85; (1.47 - 2.38); p < .01) and high FLIPI risk score (3 - 5) (OR = 2.94; (2.38 - 3.57); p < .01 compared to low risk (0 or 1) were associated with increased risk of progression or death before 24 months. Elevated baseline B2M (>= 3mg/L) (OR = 1.47; (1.25 - 1.75); p < .01) was also associated with increased risk of progression or death before 24 months.

Results of the time-dependent Cox model, adjusted for gender and stratified by PS and FLIPI, indicate that POD24 was associated with poor subsequent OS (hazard ratio (HR) = 5.24; (4.63 - 5.93); p < .01). Results of the landmark analysis confirm the association between POD24 and OS in patients who were alive 24 months after trial registration (Figure). Similar results were observed in the subset of patients with baseline B2M after adjusting for its effect along with other covariates.

Conclusions

This pooled analysis of >5,000 patients with FL included in 13 prospective clinical trials is the largest cohort to date validating early progression as a robust indicator of poor FL survival. We identified male gender, poor PS, high FLIPI score, and elevated baseline B2M as predictors of early death and progression. These provide well-defined clinical factors for building comprehensive prognostic models in the future that incorporate clinical and molecular predictors of POD24. Results on maintenance and response to therapy as factors impacting POD24 will be updated in the presentation. Our results confirm POD24 as an early clinical endpoint of poor survival in FL that should be utilized to identify patients for prospective clinical trials.

View largeDownload slide

View largeDownload slide

Close modal