Single Agent JNJ-56022473 in MDS and AML Patients FAILING HYPOMETHYLATING AGENT BASED THERAPY

Myelodysplastic syndromes (MDS) comprise a heterogeneous group of hematologic malignancies of the pluripotent hematopoietic stem cells characterized by clonal hematopoiesis and progressive bone marrow failure. Clinically, patients suffer from fatigue and are at high risk of infections or bleeding. Also, MDS patients are endangered by progression to secondary acute myeloid leukemia (AML), a disorder of early hematopoietic progenitor cells characterised by the clonal expansion of myeloid blasts in the peripheral blood, bone marrow, and/or other tissues.

Up to now, allogeneic transplantation is considered the only curative option for both MDS and AML patients. However, as MDS and AML mainly manifest in the 7th and 8th decade of life, only a limited number of patients tolerating bone marrow transplantation and are therefore eligible for this procedure. In fact, comorbid conditions are common among the elderly such as heart disease, renal insufficiency and vascular disease thus influencing the ability to withstand exhausting transplantation procedures.

Hence in recent years, novel therapeutic strategies have been developed and hypomethylating agents (HMAs) azacitidine and decitabine have emerged as new therapeutic options. In 2009, azacitidine was the first HMA in use having received approval for treatment of MDS and AML patients with 20-30% blasts. In 2012, also the sister compound decitabine got approval for treatment of AML patients ≥ 65 years of age and in 2015, azacitidine's approval was extended to AML patients ≥ 65 years of age that are considered unfit for allogeneic transplantation with > 30% blasts.

However, despite the efficacy and reported activity, only 50% of patients respond to the treatment with HMAs and the majority of them do relapse. Also the outcome of HMA- refractory patients is dismal with a median survival of 5 months after failure and hence MDS and AML patients with HMA failure represent a significant unmet medical need due to the poor prognosis and very limited treatment options.

In this context, a specific targeting of the leukemic stem cell (LSC) seems a promising option to selectively combat the leukemic cells. In fact, CD123 (also known as interleukin-3-Rα) is expressed in different hematological malignancies, among others in MDS and AML. Interestingly, CD123 is expressed by the majority of AML blasts and it is also overexpressed by CD34+ CD38- LSCs compared with normal hematopoietic stem cells and other normal tissues. These features make CD123 an attractive target for anti-leukemia therapy. Recent data suggests that treatment with HMA can induce expression of CD123 on AML tumors opening up the potential that HMA failure patients may have even higher expression levels of CD123.

Indeed, JNJ-56022473 is a promising compound that has been engineered with regard to this strategy. JNJ-56022473 is a humanized monoclonal antibody that targets the alpha chain of the interleukin-3-receptor and is optimized at the Fcγ region for enhanced activation of antibody- dependent cell-mediated cytotoxicity (ADCC) via natural killer cells.

Recently, JNJ-56022473 has been evaluated in a phase I study with AML patients at high risk for early relapse. There was a clear dose dependency on depletion of peripheral CD123 cell populations such as pDC and basophile with complete and sustained depletion seen in pts at doses ≥3 mg/kg Given the encouraging results of this phase I trial we now propose a phase II trial to evaluate JNJ-56022473 in AML patients as well as in MDS patients that have failed or are refractory to HMA- treatment.

The clinical trial will be accompanied by a translational research program which aims at elucidating specific immune responses mediated by JNJ-56022473, in particular by descriptive and functional analyses. In this context bone marrow and peripheral blood will be collected during the study as well as before and after therapy.

Constitutive overexpression of the α-subunit of the interleukin-3 receptor i.e. IL-3Rα, is a hallmark of the early AML progenitor cell population (i.e. leukemic stem cells) as well as leukemic blasts and IL-3Rα is also overexpressed in myelodysplastic syndromes. In fact, CD123 is unique to the IL-3 receptor and is responsible for the high specificity and low affinity binding of IL-3. Upon ligand binding, the α-subunit CD123 associates with the βc-subunit (CD131) to form a high-affinity receptor complex initiating an intracellular signaling cascade through phosphorylation and activation of at least 3 distinct signal transduction pathways: the JAK/STAT-kinase, MAP-kinase-, and PI3-kinase-pathways. Under physiological conditions, activation of the IL-3 receptor induces proliferative, anti-apoptotic, and differentiating signals.

JNJ-56022473 (CSL362) is a second generation antibody developed from the murine 7G3 anti- CD123 monoclonal antibody (mAb) in a stepwise process of humanization, affinity maturation, and Fc engineering. To enhance the cytotoxicity of the first-generation antibody CSL360, the proprietary Xencor (Xmab®) technology was applied and two amino acid mutations (S239D and I332E) were introduced into the Fc region. These amino acid substitutions led to improved binding to CD16 (FcγRIIIa) on natural killer (NK) cells and significantly enhanced the ability to induce ADCC.

The JNJ-56022473 drug product that will be used in this study is produced in a different cell line with a different process compared with CSL362, the drug product used in the phase I study CSLCT-AML-11-73 described below. Comparability testing of JNJ-56022473 and CSL362 has shown the antibodies to behave similarly in biophysical, biochemical, and preclinical assessments. The primary structure, charge heterogeneity, size heterogeneity, purity, higher order structure, and process impurities were found to be highly similar between CSL362 and JNJ56022473. Pharmacodynamic (PD) studies in cynomolgus monkeys showed similar levels of biological activity between CSL362 and JNJ-56022473 (ie, basophil and pDC depletion; NK cell number and activity). Additionally, the pharmacokinetic (PK) profiles of JNJ-56022473 following 30 and 100 mg/kg IV dosing of cynomolgus monkeys are highly similar to those of CSL362. Following the first weekly intravenous dose, mean Cmax and AUC0-167h of JNJ-56022473 are 91% and 94% of those observed for CSL362, respectively.

This is an open-label, single-arm, multicenter, phase II study of JNJ-56022473 in subjects suffering from MDS or AML. The trial will involve participating sites from Germany and France. Up to a maximum of 43 subjects will be enrolled in the study.

JNJ-56022473 will be given intravenously to all subjects at a dose of 9 mg/kg once every 14 days for an initial treatment period of 3 months (6 infusions). Responders will then receive up to 20 additional infusions whereas for non-responders the initial treatment will be followed by a up to 9 months observation period without further JNJ-56022473 treatment. Thus, the individual study duration for a subject will be approx. 1 year. A follow up visit will be performed after 3 months after last study drug administration for all patients to track pregnancy status according to IB.

For patients who still benefit from the treatment and are graded as responders after the time period defined above, treatment will continue as long as medically indicated. End of study as a whole will be 16 months after LPFV (maximum of 26 treatment cycles plus follow up of three months). Data obtained after the end of the study (EoS) for patients who still responding will be reported under separate cover.