Early PK-Analysis Predicts Molecular Response In Patients With Early Chronic Phase Chronic Myelogenous Leukemia (CML-CP) Treated With Frontline Nilotinib

Introduction Tyrosine kinase inhibitor (TKI) trough plasma level testing and pharmacodynamic evaluation (i.e. quantification of phosphorylated BCR-ABL downstream targets) may help to optimize patient-tailored dosing in CML-CP. For 2nd generation TKIs the clinical value of TKI plasma level and protein phosphorylation quantification remains unclear. The ENEST1st study ([NCT01061177][1]) is focused on examining the role of first-line nilotinib therapy in CML-CP. This ENEST1st substudy addresses the value of plasma-level testing and protein single cell phosphorylation pharmacodynamics as predictors of response and long-term outcomes. Methods Nilotinib levels were centrally quantified by means of mass spectrometry (HPLC-MS/MS) in 54 patients at 3 hours after first drug intake at day (d) 1 (Cmax) and at d7, d28 as well as at months (mo) 3, 6, 12 prior to morning drug intake (Cmin=trough level). Single cell quantification of intracellular protein phosphorylation was performed by flow cytometry (phosphoflow) in the myelocyte cell population after full blood fixation/cell permeabilization/erythrocyte lysis and analyzed 3 hours after first drug intake at d1 and on d7 and d28. Primary endpoint of the study was MR4 (BCR-ABL <0.01% IS) at 18 months. Results Our initial analysis focuses on correlation of drug levels with molecular response by correlating PK data to early molecular response, i.e. ≤10% BCR-ABL (IS) reduction at 3 mo as well as 0.1% IS (MMR), MR4 and MR5. At 3, 6, 12 and 18 mo, 24%, 56%, 74% and 77 % of the here investigated patients achieved MMR; 2%, 22%, 34% and 44% achieved MR4 and MR5 was detected in 0%, 5%, 15% and 14% of patients at the respective time points. Only 1 patient at 3mo had >IS 10% BCR-ABL. On d1 at 3 hours after intake of the first tablet the median peak plasma levels reached 408 ng/ml increasing to a median trough level of 834 ng/ml at d7, 888 ng/ml at d28, 919 ng/ml at 3mo, 1122 ng/ml at 6m and 1003 ng/ml at 12mo. Initial Cmax-levels at d1 significantly correlated with steady state levels thereafter (e.g. with trough levels at 12mo: r=0.56, p=0.004). Three mo drug levels were significantly negatively associated with lower BCR-ABL mRNA burden at 12m and 24m (r=-0.44, p=0.011 for m12 and r=-0.35, p=0.014 for m24). When median steady state drug levels were compared in different response categories (yes/no: MR4, MR5, MMR, >IS 10% at 3mo), significantly higher nilotinib trough levels at day 90 and mean trough level (day 28-180) were seen in patients achieving the primary endpoint of the study (MR4 at 18mo). Similarly, 3mo and mean through levels were also significantly higher in patients achieving MR5 at 12 or 18mo. As only one patient did not reach <10% at 3mo, we cannot evaluate if early PK data are able to predict this endpoint. The probability of achieving MR5 at 12mo and 24mo was significantly higher in the top third of the day 90 PK level (cumulative incidence of MR5 in the respective PK groups at 12mo: 5.6% vs. 5.6% vs. 44.4% and at 24mo: 5.6% vs. 22% vs. 44.4%). Cumulative incidence of MR4 was faster in the higher PK group (6mo: 6% vs. 18% vs. 42%; 12mo: 22% vs. 29% vs. 57% and at 24mo: 46% vs. 67% vs. 58%), which however did not reach statistical significance. Phosphoflow of myelocytes revealed that compared to baseline, phosphorylation of almost all investigated proteins decreases over time, including significant reduction of phospho-Gab2(Y452), Abl(Y245), STAT5(Y694), Erk1/2(T202/Y204), and p38(T108/Y182) at d28. Correlation of plasma levels with changes in phosphorylation status revealed a negative correlation of increasing drug levels with decreasing phospho-p38 and phospho-STAT5 status. A more detailed signaling analysis of cellular subsets related to molecular response criteria are in progress. Conclusions Single cell myelocyte phosphoprotein detection allowed monitoring of the BCR-ABL signalling network. Inter-individual plasma levels show a relatively wide variability and Cmax levels upon first drug exposure correlate with later trough levels pointing out to a potential individual pharmacogenetic background regulating drug availability. PK-based optimizing of nilotinib-dosing may help to further improve response depth (i.e. reaching MR4 or MR5), which is assumed to be a prerequisite for treatment discontinuation. Disclosures: Wolf: Novartis: Honoraria, Membership on an entity’s Board of Directors or advisory committees, Research Funding. Mustjoki: Novartis: Honoraria, Research Funding. Greil: Novartis: Honoraria, Research Funding. Hjorth-Hansen: Pfizer, BMS: Honoraria, Travel expenses Other. Verhoef: Novartis: Research Funding. Mark: Novartis: Employment. Haenig: Novartis: Employment. Jurjonas: Novartis: Employment. Giles: Novartis: Consultancy, Research Funding. Hochhaus: Novartis: Research Funding. Porkka: Novartis: Honoraria, Research Funding. Ossenkoppele: Novartis: Consultancy, Research Funding; BMS: Consultancy. Gjertsen: Novartis: Honoraria, Research Funding. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01061177&atom=%2Fbloodjournal%2F122%2F21%2F1485.atom