Retrospective analysis of surrogate endpoints for outcomes in patients with FL post-ASCT

On 26^th October 2017, a retrospective analysis on surrogate endpoints in follicular lymphoma (FL) conducted by Ana Jiménez-Ubieto, from the University Hospital 12 de Octubre, Madrid, Spain, and colleagues, was published in Cancer Medicine. The authors aimed to evaluate surrogate end points for overall survival (OS) in patients with FL after autologous stem cell transplantation (ASCT).

Determination of OS requires long observation time periods, so the authors wanted to determine the validity of endpoints with earlier observation times. They analyzed the value of two surrogate endpoints for OS; progression-free survival at 24 months (PFS24) and complete response at 30 months (CR30) post-ASCT.

Study Highlights

• Patients included in the study were selected from the GELTAMO registry database in Spain who had non-transformed FL and received an ASCT between 1989 and 2007
• 626 chemo-sensitive patients were included in the analysis
• PFS24 analysis contained two groups:
  • Patients with disease progression within 2 years from ASCT
  • Patients alive without progression at 2 years
• CR30 analysis contained two groups:
  • Patients not in CR at 30 months post-ASCT
  • Patients in CR, who had not relapsed at 30 months, post-ASCT

Key Findings

• Median follow-up = 12.2 years from ASCT
• PFS24 analysis = 153 (24%) patients progressed within 24 months post-ASCT
  • Early progression was associated with shorter OS in all patients
  • HR 6.8, 95% CI, 5.2–9.0 (P = 0.00001)
• CR30 analysis = 183 (31%) progressed/died within 30 months post-ASCT
  • The absence of CR30 was associated with shorter OS
  • HR 7.8, 95% CI, 6.0–10.3 (P < 0.00001)
  • Including patients with prior rituximab treatment HR 8.2, 95% CI, 4.5–15.1 (P < 0.00001)
• A subgroup analysis revealed characteristics associated with poor outcome: older age, male (P = 0.002), Eastern Cooperative Oncology Group (ECOG) performance status >1 (P = 0.007), bone marrow infiltration (P = 0.0005), and high lactate dehydrogenase level (P = 0.04), or more than four nodal areas affected (P = 0.04)

The authors suggested that PFS24 and CR30 could be valid primary and surrogate end points for OS for FL patients following ASCT. The authors noted limitations of their study such as its retrospective design and that the histologic results were not centrally reviewed. The authors stressed that while PFS24 or CR30 could not be recommended as superior endpoints, since they are associated with poor patient outcomes they could help to identify high-risk patients earlier on so that they can be offered alternative treatment.

Abstract

Overall survival (OS) is the gold-standard end point for studies evaluating autologous stem cell transplantation (ASCT) in follicular lymphoma (FL), but assessment may be elusive due to the lengthy disease course. We analyzed the validity of two earlier end points, proposed in the setting of first-line chemo-/immunotherapy, as surrogates for OS-progression-free survival (PFS) status at 24 months (PFS24) and complete response at 30 months (CR30) post-ASCT. We also have investigated the clinical features of patients with early progression after ASCT. Data were available for 626 chemosensitive FL patients who received ASCT between 1989 and 2007. Median follow-up was 12.2 years from ASCT. In the PFS24 analysis, 153 (24%) patients progressed within 24 months and 447 were alive and progression-free at 24 months post-ASCT (26 who died without disease progressions within 24 months were excluded). Early progression was associated with shorter OS (hazard ratio [HR], 6.8; P = 0.00001). In the subgroup of patients who received an ASCT in the setting or relapse after being exposed to rituximab, the HR was 11.3 (95% CI, 3.9-30.2; P < 0.00001). In the CR30 analysis, 183 of 596 (31%) response-evaluable patients progressed/died with 30 months post-ASCT. The absence of CR30 was associated with shorter OS (HR, 7.8; P < 0.00001), including in patients with prior rituximab (HR, 8.2). PFS24 and CR30 post-ASCT are associated with poor outcomes and should be primary end points. Further research is needed to identify this population to be offered alternative treatments.