Prognostic value of PET-based response assessment in FL: Secondary analysis of the phase III GALLIUM trial

On 8 October 2018, Judith Trotman a member of our Scientific Advisory Board from the Concord Repatriation General Hospital, Sydney, AU, and colleagues, published in The Lancet Oncology the secondary analysis of the GALLIUM phase III trial (NCT01332968). The GALLIUM trial evaluated the effect of first-line obinutuzumab compared to rituximab in follicular lymphoma (FL) patients. The primary analysis of the study was published in 2017 in the New England Journal of Medicine and revealed that obinutuzumab resulted in a longer progression-free survival (PFS) when compared to rituximab in previously-untreated FL patients.

The purpose of this secondary GALLIUM analysis was to assess the prognostic value of positron emission tomography-computed tomography (PET-CT) following first-line chemotherapy with either obinutuzumab or rituximab in the same FL patients. The primary endpoints of this secondary analysis were PET and CT responses at the end of the induction therapy and their association with PFS and overall survival (OS).

Study design (secondary analysis)

• PET scans were obtained from N = 669 patients from the original GALLIUM trial who enrolled after July 2011 (65%). Out of n = 609 patients with baseline PET scans, n = 595 (98%) also had a PET scan at the end of the induction treatment
• Median observation time in the PET population = 43.3 months
• Patients had previously untreated, histologically-confirmed CD20-positive FL (Grade 1−3a), advanced disease stage (III/IV or II with largest tumor diameter ≥ 7 cm), and an Eastern Cooperative Oncology Group (ECOG) score of 0, 1 or 2
• Patients were randomized 1:1 to obinutuzumab (1000 mg on Days 1, 8, and 15 of cycle 1 and Day 1 thereafter) or rituximab (375 mg/m² on Day 1 per cycle), plus chemotherapy of choice with either treatment
• CT scans were performed every four months for the first year and every six months for the second maintenance year
• PET scans were performed at baseline (≤ 35 days prior to randomization) and at the end of treatment induction (median = 19 days after the end of the last chemotherapy cycle and 20 days after the end of the last obinutuzumab cycle) or if early study termination occured
• Scans were assessed blindly by investigators and an independent review committee

Results

• From CT assessment alone, n = 90.6% of patients achieved an overall response in the PET population (95% CI, 88.0–92.8). When PET response assessment was included n = 86.9% (95% CI, 83.9–89.5) achieved an overall response according to the International Harmonization Project (IHP) 2007 response criteria and n = 81.7%; (95% CI, 78.3–84.7), according to the Lugano 2014 criteria
• The number of patients achieving a PET complete response (CR) and complete metabolic response was more than double compared to the patients achieving CT-based CR
• CT-based CR at the end of induction was a significant prognostic marker for PFS between responders and non-responders, but not for OS (2.5-year OS for CR patients = 97.7% [95% CI, 94.0–99.1] vs 93.5% [95% CI, 90.5–95.6] for those not achieving CR; HR = 0.5, [95% CI, 0.3–1.2]; P = 0.124)
• PET-based CR at the end of induction was a significant prognostic marker for both PFS and OS (2.5-year OS for CR patients = 96.9% [95% CI, 94.5–98.2] vs 90.6% [95% CI, 84.6–94.3] for those not achieving CR; HR = 0.4, [95% CI, 0.2–0.9]; P = 0.011)
• Metabolic response at the end of induction was also deemed as a prognostic factor for PFS

The results of this secondary analysis of the GALLIUM phase III trial indicated that PET is more accurate than CT in evaluating the responses of FL patients treated with first-line immunochemotherapy. The authors stated that despite the limitations of the study (retrospective nature and smaller sample size than primary analysis cohort), PET might provide a useful early prognostic marker of PFS and OS and guide response-based FL treatments. 

For a video interview with the lead author Judith Trotman summarizing the key findings of this analysis during the 22^nd European Hematology Association Congress (2017) click here.