Phase 1 Study of a Hepcidin Antagonist, LY2787106, in Cancer-Associated Anemia

BACKGROUND:  Hepcidin is a 25-amino-acid peptide that binds to ferroportin and causes internalization and degradation of the hepcidin-ferroportin complex, leading to decreased iron absorption and reduced iron export from reticuloendothelial cells. High hepcidin levels are found in cancer patients (pts) and implicated in anemia pathogenesis. It is hypothesized that neutralizing hepcidin with a monoclonal antibody (mAb) will prevent internalization of ferroportin, restore iron efflux from cells, and allow transferrin‑mediated iron transport to the bone marrow to support erythropoiesis. LY2787106 is a fully humanized mAb with high affinity for human hepcidin. This phase 1 study evaluated the safety, PK/PD, and efficacy (effects on serum iron panel [iron, ferritin, transferrin saturation (TSAT), soluble transferrin receptor, reticulocyte hemoglobin], reticulocytosis, and hemoglobin [Hb]) of LY2787106 in cancer pts with anemia.

METHODS:  A 3+3 design was used in the dose-escalation phase to determine the recommended phase 2 dose of LY2787106. Eligible pts were ≥18 years old with previously treated metastatic/incurable nonmyeloid cancer, Hb <11 g/dL, serum hepcidin ≥5 ng/mL, and ECOG performance status ≤2. The primary objective of Part A was to assess the safety of LY2787106 over a range of doses (0.3‑10 mg/kg IV every 3 weeks [Q3W]); the primary objective of Part B was to assess mean change in Hb from baseline to end of Cycle 4 following LY2787106 treatment (10 mg/kg IV weekly) with (Cohort B1) or without (Cohort B2) oral iron supplementation.

RESULTS:  A total of 33 pts were enrolled (Part A, n=19; Part B, B1, n =7; B2, n=7). Overall, most pts had multiple myeloma (n=14 [42.4%]). Mean age was 65 years (range, 44‑88 years). There were 20 females and 13 males. Patients had received a median of 4.5 prior oncology treatments. Mean (SD) Hb level at baseline was 9.2 (0.95) g/dL. The median number of cycles given and completed was 3.  No DLTs were seen in Part A, so the MTD was not reached. Data from Part A showed that LY2787106 was well tolerated when given Q3W, had a short half-life (~7 days), and transiently increased serum iron. This supported more frequent weekly dosing in Part B. Treatment‑emergent AEs related to LY2787106 occurred in 4 pts (12.1%):  increased creatine phosphokinase (grade 2), cardiac failure (grade 3), QT prolongation (grade 1), and neutropenia (grade 2) (1 pt each).  Eight pts generated antibodies to LY2787106, but none had hypersensitivity reactions. The LY2787106 PK profile was consistent with reported PK characteristics of a mAb; the profile included a small clearance of ~0.032 L/h associated with limited volume of distribution, resulting in a terminal half‑life of ~7 days. LY2787106 PK was independent of time and dose and exposure linear in the 0.3‑10 mg/kg dose range. PD data indicated that LY2787106 administration led to a mean maximum 3.48‑fold increase (90% CI, 2.5‑4.8) from baseline in serum iron at the highest dose (10 mg/kg) in Parts A and B. Serum iron increase was maximal ~24‑48 hours post dose, but returned to baseline within a week after dosing. The increase in serum iron paralleled a transient increase in TSAT at dose levels of 1, 3, and 10 mg/kg. This translated to a transient increase in mean ratio of reticulocyte count relative to baseline at 2 weeks post dose in pts who received 10 mg/kg Q3W, but Hb levels remained relatively unchanged. In Part B, a ≥0.5 g/dL increase in Hb from baseline, albeit transient, was seen in 4 pts at Week 12 (2 pts each cohort). No consistent increase in reticulocytes or changes in iron profiles were seen in these pts. Factors that may have confounded the limited change in Hb and reticulocyte profiles included the myelosuppressive effects of chemotherapy concomitantly given during the study, comorbidities, and study‑related blood draws. As expected, LY2787106 administration led to increased hepcidin concentrations secondary to hepcidin binding/neutralization by LY2787106 and hepcidin release in response to iron increase.

CONCLUSIONS:  This is the first clinical trial of a fully humanized mAb against hepcidin. LY2787106 was well tolerated in cancer pts with anemia. Targeting the hepcidin-ferroportin pathway by neutralizing hepcidin resulted in transient iron mobilization and reticulocyte count relative to baseline, thus supporting the role of hepcidin in iron regulation and anemia pathogenesis.