Introduction

The phosphoinositide 3-kinase p110δ isoform (PI3Kδ) is expressed primarily in hematopoietic cells and is essential in B-cell receptor (BCR) signaling. PI3Kδ is expressed in lymphoid malignancies, including chronic lymphocytic leukemia (CLL) and non-Hodgkin lymphoma (NHL). AMG 319 is an investigational, highly selective, small molecule inhibitor of PI3Kδ that blocks B cell proliferation following BCR stimulation both in vitro and in vivo, inhibits basal AKT phosphorylation, and inhibits proliferation in lymphoid tumor cells. This study evaluated the safety, tolerability, pharmacokinetics (PK), and pharmacodynamics (PD) of AMG 319.

Methods

Patients (pts) with relapsed or refractory CLL or NHL were eligible. Key eligibility criteria included ≥ 18 years old, ECOG ≤ 2, and adequate organ function. AMG 319 was administered once per day (QD; except day 2 for PK) until disease progression or unacceptable toxicity. Doses of 25, 50, 100, 200, 300, and 400 mg were administered in sequential cohorts (3 pts per cohort). Dose-limiting toxicities (DLT) were defined during the first 28 days. Response assessments for CLL and NHL were done using IWCLL 2008 and IWG 2007 criteria, respectively. CT scans for CLL pts were done at baseline and week 8.

Results

28 pts received AMG 319 (6 at 25 mg; 3 at 50 mg, 100 mg, 200 mg, and 300 mg; and 10 at 400 mg). Demographics: median age 68 years, 71% men, ECOG 0/1 93%. 25 pts had CLL, 3 had NHL, including 2 mantle cell and 1 marginal zone. The median number of prior treatment regimens was 4 (range 1-9) for CLL and 7 (range 5-9) for NHL. Central cytogenetic analysis of CLL pts prior to dosing (n=24) revealed 10 with del 17p (42%), 3 with del 11 and not 17p (13%), 1 with trisomy 12 alone (4%), 9 normal (38%), and 1 del 13q alone (4%). Doses up to 400 mg QD were explored without reaching a maximum tolerated dose. There was 1 DLT: grade 3 hemolytic anemia at 25 mg in a CLL pt after 1 dose, considered as possibly related to AMG 319. AMG 319 was well absorbed and exhibited linear PK with mean plasma half-lives of 3.8 to 6.6 hours across dose cohorts. Dose-dependent coverage of BCR-induced pAKT in CLL samples (ex-vivo IgD stimulated) was observed in all samples with an inducible signal (60%); near complete inhibition for 24 hours was seen at 400 mg. 75% of pts had ≥1 treatment-related adverse event (AE, 25% grade ≥3). Grade ≥3 treatment-related AEs (n>1) were colitis 3 (10%), anemia 3 (10%), leukocytosis 2 (7%), infection 2 (7%), and hemolysis 2 (7%). The colitis cases occurred at 400 mg between days 40 and 60; 2 were successfully re-challenged at a lower dose, while the third was diagnosed with C Diff colitis and discontinued. Grade 3 transaminitis was observed in 1 pt (4%) at 50 mg on day 29, which resolved with holding of drug and did not recur with re-challenge. Subset analysis of lymphocytes revealed no changes in % of T, B, or NK cells. Baseline % of T-regulatory cells (as a function of total CD4) was elevated in CLL pts (14.4% ± 7.6%). Elevated T regulatory cells (>10% of CD4+) tended to normalize during treatment (14/19 pts), suggesting immune restoration. Response data are available for 24 CLL pts (pt with DLT removed after 1 dose) and 3 NHL pts. Lymphocyte counts in CLL were highly variable, with some pts experiencing marked and early lymphocytosis (at all doses) and others showing gradual decline. Consistent nodal regression was observed at all doses by CT and physical exam. Early CT imaging (21 pts) at week 8 revealed lymph node (LN) reduction in all pts, and >50% decrease was seen in 7 pts; all 7 were dosed at 400 mg including 4 with del 17p or del 11q. Two pts at week 8 met IWCLL response criteria for partial response. By physical exam, all 20 evaluable pts had >50% LN reduction as a best response, with 15 (75%) having >90%. Response was present in all cytogenetic subtypes. At the 200-400 mg dose levels, 13/15 CLL pts remain on study after a median follow-up of 30 weeks (range 14-65). 1 pt with mantle cell NHL had 46% shrinkage while on therapy but progressed after 26 weeks; the 2 other NHL pts progressed by the first evaluation.

Conclusions

AMG 319 exhibited linear absorption and was tolerated at doses up to 400 mg QD. This degree of inhibition provided complete blockade of ex-vivo stimulated pAKT for 24 hours in CLL. Anti-tumor activity was observed early in CLL pts and included high-risk cytogenetic subgroups. While activity was noted at all dose levels, an apparent dose response was observed, with deeper and faster responses in CLL pts at higher doses.

Disclosures:

Lanasa:Amgen: Consultancy. Glenn:Amgen: Research Funding; Sanofi Aventis: Research Funding. Mato:Amgen: Honoraria. Wong:Amgen: Employment, Equity Ownership. Amore:Amgen: Employment, Equity Ownership. Means:Amgen: Employment, Equity Ownership. Stevens:Amgen: Employment, Equity Ownership. Yan:Amgen: Employment, Equity Ownership. Friberg:Amgen: Employment, Equity Ownership. Goy:Amgen Inc: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution