Safety and Efficacy of Obinutuzumab Plus Bendamustine in Previously Untreated Patients with Chronic Lymphocytic Leukemia: Subgroup Analysis of the Green Study

Background

The glycoengineered type II anti-CD20 monoclonal antibody obinutuzumab (GA101; GAZYVA/GAZYVARO; G) combined with chlorambucil (Clb) has superior efficacy to Clb monotherapy and to rituximab plus Clb with an acceptable safety profile in patients with chronic lymphocytic leukemia (CLL), as shown in the CLL11 study (Goede V, et al. NEJM 2014). GREEN is an ongoing, non-randomized, multi-cohort phase IIIb study (NCT01905943) investigating the safety (primary objective) and efficacy of G alone or in combination with chemotherapy in patients with previously untreated or relapsed/refractory CLL and assessing various strategies (cohorts 1-3) for reducing the rate of infusion-related reactions (IRRs) during the first infusion of G (Bosch F, et al. Blood 2014). We report safety and efficacy data from a subgroup of previously untreated patients in cohort 1 who received G-bendamustine (G-B).

Methods

Subjects were aged ≥18 years withdocumented CLL (except one case of SLL), an ECOG performance status of 0-2, and adequate hematologic function. Non-fit patients were those with a CrCl of <70 mL/min, and/or a CIRS score of >6. Fit patients comprised all others.

Treatment was six 28-day cycles of G-B, where G was administered IV on D1/D2 of C1 (split dose: 25mg D1/975mg D2), and 1000mg on D8 and D15 of C1 and D1 C2-6. B was administered ≥30 minutes after G on D1 and D2 of each cycle at 90mg/m² IV, or at 70mg/m² in non-fit patients at the investigator's discretion.

Safety endpoints included incidence, type and severity of AEs. Efficacy endpoints included ORR (investigator-assessed) and minimal residual disease (MRD) measured 3 months post-treatment. ORR was strictly assessed per International Workshop Group criteria (iwCLL 2008). Patients with missing response assessment components had their responses downgraded mandatorily. MRD negativity was defined as <1x10^-4 malignant B cells in peripheral blood or bone marrow aspirate, measured in a central laboratory by 4-color flow cytometry. The population comprised all patients from cohort 1 of GREEN who received at least a partial dose of both G and B, and was based on a data cut-off of 26 March 2015.

Results

With a planned overall sample size of 950 patients in GREEN, the G-B subgroup in cohort 1 comprised 158 patients (157 CLL, 1 SLL; 74 fit, 84 non-fit). Median age was 67.6 years, 15.8% of patients had a CIRS score of >6, and 44.9% had a CrCl of <70 mL/min; 31.6% of patients had Binet stage A disease, 38% Binet B, and 30.4% Binet C. 7.0% of patients' disease displayed 17p deletion, 16.5% 11q deletion, and 58.2% unmutated IGHV. 91.1% of patients receiving B and 93.0% of those receiving G took ≥90% of the recommended total dose.

The safety profile of G-B was as expected. 50% of patients developed grade 3-5 neutropenia and 12.7% developed a grade 3-5 infection. Other common grade 3-5 AEs included thrombocytopenia (12.7%) and tumor lysis syndrome (TLS; 10.1%). The most common serious AEs were neutropenia (10.8%), pyrexia (7.6%), febrile neutropenia (7.0%), and TLS (5.1%). There were nine deaths - one due to progression, and eight due to AEs (considered related to study drug by the investigator: 1 infection, 1 sudden death, 1 acute hepatic failure, and 1 febrile neutropenia combined with TLS; considered unrelated: 2 infections and 2 secondary neoplasms). IRRs occurred in 55.7% of patients (15.2% grade 3-5, none fatal). Overall, 26 patients (16.5%) prematurely discontinued treatment due to ≥1 adverse event.

The ORR was 78.5% (124/158). The rate of CR (including incomplete CR [CRi]) was 32.3% (51/158), PR 46.2% (73/158), SD 10.8% (17/158), and PD 0.6% (1/158); 10.1% (16/158) of patients had missing data. Response rates were similar in non-fit (34.5% CR/CRi, 41.7% PR, 10.7% SD, and 1.2% PD) and fit (29.7% CR/CRi, 51.4% PR, 10.8% SD, and 0% PD) patients.

In an intent-to-treat analysis including all missing (not taken or evaluable) MRD results, MRD negativity was 58.9% (93/158, including 56 missing) in blood and 27.8% (44/158, including 95 missing) in bone marrow. With a median observation time of 11.2 months, PFS data were immature and the median was not reached.

Conclusions

Treatment with G-B in previously-untreated CLL patients is generally well tolerated and the observed toxicities are manageable and not unexpected. G-B achieves a promising rate of CRs and a high rate of MRD-negative remissions, and may offer a new treatment option for fit and non-fit patients with CLL.