Correlation between Real-World Progression-Free Survival (rwPFS) and Overall Survival (OS) in Patients with Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma (CLL/SLL) Who Relapsed or Are Refractory, Intolerant, Resistant to Treatment with a Bruton's Tyrosine Kinase Inhibitor (BTKi)

Introduction: Given the long maturation time of OS, PFS is often used in oncology and hematology as a surrogate for OS. However, the association between the two endpoints varies across cancer types. In CLL clinical trial populations, PFS and OS were shown to be strongly correlated in patients receiving at least two lines of therapy (LOT). The present study used real-world individual patient data to evaluate the strength of correlation between real-world PFS (rwPFS) and OS in CLL/SLL patients who relapsed or were resistant/refractory/intolerant to treatment with a covalent BTKi.

Methods: A retrospective chart review was conducted of adult CLL/SLL patients at Dana-Farber Cancer Institute who were treated with at least one covalent BTKi and relapsed or were resistant/refractory/intolerant to the treatment and have discontinued BTKi. Patients were included if they had records available from 6 months before to 12 months after initiation of the first observed (index) BTKi, and were excluded if they were treated with a non-hormonal antineoplastic agent for other primary malignancies in the period between CLL/SLL diagnosis and receipt of their first BTKi. Demographic and clinical variables, oncologic and medical histories, treatments, and clinical outcomes were abstracted from patients' medical charts. Time to event endpoints were described using Kaplan-Meier curves. Time to progression (TTP) was defined as time from index to progression. OS was defined as time from index to death; patients were censored at the last visit if they did not die. rwPFS was defined as the time to progression or death; patients who did not progress or die were censored at index BTKi treatment discontinuation. Fleischer's method was used to estimate the individual-level Pearson correlation coefficient between rwPFS and OS among patients treated with BTKi in any line of treatment. Confidence intervals (CIs) were calculated by bootstrapping (1000 replicates).

Results: A total of 104 patients who initiated treatment with the index BTKi from 07/2011 to 09/2019 and relapsed or were resistant/refractory/intolerant to the treatment were included in the analysis. Median (interquartile range [IQR]) age at BTKi initiation was 66.7 (60.3-73.0) years, median (IQR) time from index BTKi initiation to last clinical visit or death was 5.7 (3.3-7.0) years. Among patients with a known index BTKi LOT (N=102 [98%]), 28% received the index BTKi in first line, 32% in second line, 23% in third line, 12% in fourth line, and 6% in or after the fifth line. An assessment date for response to the index BTKi and for whether a patient had progressed was available for 95 patients (91%) ( Table 1). Among these patients, progression after index BTKi treatment was observed for 48 (51%) patients. Median (95% CI) time from BTKi initiation to progression was 3.7 (2.6, 4.1) years. More than one-third of patients (35%) died and median (95% CI) OS was 8.9 (6.7, not reached) years. 7 patients died without progressing and cause of death was available for 5 of these patients: 3 died because of an infection, 1 because of a secondary malignancy, and 1 because of other unspecified causes. 58% patients progressed or died and median (95% CI) rwPFS was 3.2 (2.6, 3.9) years. The correlation (95% CI) between rwPFS and OS was 0.79 (0.63, 0.94).

Conclusions: This analysis found a strong, positive, and significant correlation between rwPFS and OS in patients with CLL/SLL who received treatment with a covalent BTKi and relapsed on or were resistant/refractory/intolerant to the treatment. These findings reinforce the hypothesis that rwPFS may be a useful early endpoint for evaluating the effectiveness of BTKi treatment outside of clinical trials among the patient population of CLL/SLL patients who relapsed or were refractory/resistant/intolerant to BTKi, at time points when OS is not mature. However, further analysis of the change in rwPFS that predicts an improvement in OS (surrogate threshold effect) is required to establish rwPFS as a valid surrogate for OS.