GADOLIN: PRIMARY RESULTS FROM A PHASE III STUDY OF OBINUTUZUMAB PLUS BENDAMUSTINE COMPARED WITH BENDAMUSTINE ALONE IN PATIENTS WITH RITUXIMAB-REFRACTORY INDOLENT NONHODGKIN LYMPHOMA

Background
Treatment options are limited and outcomes poor for patients with rituximab-refractory (Rit-ref) indolent non-Hodgkin lymphoma (iNHL). Bendamustine (B) has shown a median progression-free survival (PFS) outcome of 9 months and response duration of 10 months in Phase II trials in Rit-ref iNHL. Obinutuzumab (GA101/Gazyva [G]) is a glycoengineered type II anti-CD20 antibody with activity and an acceptable safety profile in Rit-ref iNHL shown in Phase I/II studies; obinutuzumab is not currently licensed in NHL. Preclinical studies have shown that combining G and B increases their activity; thus, this combination (GB) has potential for improved efficacy in comparison with B alone.

Aims
To evaluate efficacy and safety of GB versus B alone in patients with Rit-ref iNHL.

Methods
GADOLIN (NCT01059630) is a randomized, open-label Phase III study in patients with CD20-positive Rit-ref iNHL. In the control (B) arm, patients received single-agent therapy with B 120mg/m² (days 1 and 2, cycles 1–6); in the test (GB) arm patients received B 90mg/m² (days 1 and 2, cycles 1–6) in combination with G 1000mg (days 1, 8, and 15 of cycle 1 and day 1 of cycles 2–6) for up to six 28-day cycles. All patients gave informed consent. Non-progressing patients in the GB arm received further G monotherapy every 2 months for up to 2 years. The primary endpoint was PFS assessed by an independent radiology facility (IRF), with 80% power to detect a 43% improvement in median PFS.

Results
At a protocol-specified interim analysis, 396 patients were randomized to receive B (n=202 [198 were treated]) or GB (n=194). On February 4, 2015, the IDMC recommended to unblind the study and release the data to the scientific community as the primary endpoint (PFS) had been met. Baseline characteristics were balanced between the treatment arms and follicular lymphoma was the most common iNHL subtype (82.2% B vs 79.9% GB). The median age was 63 yrs and patients had received a median of two prior therapies. The median observation time was 20 months for B and 22 months for GB. IRF-assessed median PFS was 14.9 months for B and not reached (NR) for GB (hazard ratio [HR] 0.55, 95% confidence interval [CI]: 0.4–0.74; p=0.00011) (Figure).

The median investigator-assessed PFS was 14 months for B and 29 months for GB (HR 0.52, 95% CI: 0.39–0.70; p<0.0001). There were no significant differences in IRF-assessed overall response rate (63.0% B vs 69.1% GB) or complete response (12.2% B vs 11.2% GB) at the end of induction, in IRF-assessed best overall response up to 12 months from the start of treatment (76.6% B vs 78.6% GB), or in preliminary overall survival (OS; median OS NR in either arm). The median duration of post-induction G monotherapy was 10.8 months, 74% of patients received at least one dose of G and 25% received all 12 doses. In the treatment period, there were fewer Grade ≥3 adverse events with B than with GB (62.1% B vs 68.0% GB), notably neutropenia (26.3% B vs 33.0% GB) and infusion-related reactions (3.5% B vs 8.8% GB), but more Grade ≥3 thrombocytopenia (16.2% B vs 10.8% GB), anemia (10.1% B vs 7.7% GB) and pneumonia (5.6% B vs 2.6% GB).

Summary
G combined with B (90 mg/m²) followed by G maintenance significantly improved PFS vs B alone (120 mg/m²) in Rit-ref iNHL. The clinically meaningful PFS improvement with GB is the first randomized evidence of benefit for a novel anti-CD20 antibody in Rit-ref iNHL.

Keyword(s): Antibody, CD20, Clinical data, NH

www.eha.org