Phase I Study of the T-Cell Receptor-like Antibody, Hu8F4 in Patients with Relapsed and Refractory Myeloid Malignancies

Abstract

Background: Relapse after conventional chemotherapy for newly diagnosed AML is responsible for the majority of treatment failure. Relapsed (R/R) disease is frequently refractory to salvage attempts with further chemotherapy. The effectiveness of stem cell transplant and associated graft vs. leukemia effect implies an important role for immune-based therapy in producing long lasting remissions. Existing approaches using immunotherapy have failed to establish a suitable surface target or treatment paradigm that is effective in myeloid malignancies. Hu8F4 is a humanized T-cell receptor-like monoclonal antibody that binds to the conformational epitope of PR1 bound to HLA-A2, which is highly, differentially expressed on the surface of AML compared to normal progenitors. Methods: This is a first-in-human, Phase I dose escalation trial of Hu8F4 in patients (pts) with myeloid malignancies. Pts with R/R AML, MDS, CMML, and myeloid blast phase of CML with adequate organ function and PS ≤ 2 were eligible. Pts were treated on 7 escalating dose levels, ranging from 0.01 mg/kg to 10 mg/kg IV on D1 & 15. Initial dose levels required 1 pt per dose (0.01, 0.03, 0.1, 0.3, 1), followed by 3 pts per dose (3, 10). Results: Fifteen pts with R/R AML have been enrolled, with a median age of 62 years (range, 23-77 years), including 7 females (47%). Pts had received a median of 4 prior therapies; 5 pts (33%) had a PS of 2. At enrollment, the median WBC was 2 (0.1 - 18.4), median BM blasts were 21% (0 - 76); 9 pts (60%) had complex karyotype and 4 (27%) had a TP53 mutation. All pts had > 98% surface expression of PR1 on the myeloid blasts. Hu8F4 C max ranged up to 160,000 ng/mL with t 1/2 of 48 hours and clearance of 2.61 hr*ng/mL at the highest dose. Weak anti-drug antibodies were observed after week 4 in 2 of 3 pts treated at 3 mg/kg. With a median follow up of 5 months (1.1 - 9.9), pts have received a median of 1 (1-4) cycle of therapy. No DLTs were observed at the 3 mg/kg cohort.; 4 additional pts who were on concomitant tacrolimus were enrolled at this level to assess safety with concurrent use, and no DLTs were observed. Three pts were enrolled at the 10 mg/kg level, and 1 experienced dose limiting thrombocytopenia, triggering expansion to 6 pts. Two pts had a decline in BM blasts and 4 had stable disease. There was rapid decline in peripheral blasts immediately after infusion of Hu8F4 on D1 and 15 with associated elevation in serum LDH in some pts and a rise in normal granulocytes, consistent with drug-mediated anti-leukemia effects. The pharmacokinetic parameters and transient blast reduction indicated a possible sink effect mediated by high levels of circulating blasts. SAEs documented on study were mostly disease-related and included infections, cytopenias, hemoptysis, pneumonia, and GI bleeding. Treatment related AEs were temporally related to the drug infusion, including hypotension (Grade 2: N=2), and rigors (Grade 2: N=10; Grade 1: N=1). Infusion reactions were observed at all dose levels, more common at 3 and 10 mg/kg, but were transient, and managed with steroids and antihistamines. All pts proceeded with their next dose without further issues. Neither cytokine release syndrome nor neurologic toxicities were observed. Correlative studies support antibody dependent cellular cytotoxicity and phagocytosis as important mechanisms of anti-AML activity, which was more effective at lower dose levels. Conclusion: Hu8F4, a first-in-class TCR mimic antibody, was well tolerated with expansion planned at more frequent dose intervals. On-target peripheral blast reduction temporally related to infusion suggests biological activity. Real-time pharmacokinetic data indicate a possible sink effect that may be overcome by a more frequent dosing strategy that is planned.