Systematic Assessment of Potential Cardiac Effects of the Novel Histone Deacetylase (HDAC) Inhibitor Romidepsin.

Abstract

Abstract 3709Poster Board III-645 BackgroundRomidepsin is an anti-neoplastic agent that has been identified as a novel pan-HDAC inhibitor with single-agent activity in cutaneous T-cell lymphoma (CTCL) in 2 studies (GPI-04-0001 and NCI 1312). In the GPI study the overall response rate (ORR) was 34% including 6 complete clinical responses (CCRs) in 96 pts with CTCL and the median duration of response (DOR) was 14.9 months (mo). In the NCI study, the ORR was 35% including 4 CCR in 71 pts and the median DOR was 11 mo. NCI 1312 has also reported single-agent activity in patients with peripheral T-cell lymphoma with an ORR of 31% including 4 CRs in the 48 pts analyzed. HDAC inhibitors have variably been reported to have cardiovascular effects, in particular, QTc prolongation. To systemically and rigorously evaluate the potential cardiac effects of romidepsin, a cardiac safety monitoring plan was developed after discussions with the FDA Division of Oncology Drug Products. Subsequently, a romidepsin QTc Clinical/Statisitcal Analysis Plan was developed following the ICH E14 Guidelines, to provide a comprehensive evaluation of the cardiac effects of romidepsin. MethodsAn analysis of romidepsin cardiac safety, including review and analysis of ECG findings, was performed. The primary objective of these analyses was to evaluate the effect of romidepsin on the change from baseline on the corrected QT interval of the ECG using the Fridericia QT correction method (QTcF) during Cycle 1 Day 1 exposure to study drug. A total of 4909 ECGs from 110 patients in 3 clinical studies, GPI study, NCI study, and Study GPI-06-0005 (a Phase 1 PK study) were evaluated. The 110 patients comprised 103 patients with T-cell lymphoma and 7 patients with advanced cancer treated with romidepsin at 14 mg/m2 as a 4-hour infusion on Days 1, 8, and 15 of a 28-day cycle. Using a nonlinear mixed effects model the pharmacodynamics of romidepsin was characterized for exposure-QTc effects. ResultsIn an analysis of ECG data, a mean increase of 5.0 msec (90% CI upper bound 7.7 msec) in QTcF interval was measured following infusion of romidepsin; this value includes a contribution to QTc prolongation by pre-dose antiemetics. Data from the NCI study showed that the QTcF change persisted to 24 hours, with a return to baseline by 48 hours, whereas in the smaller GPI-06-0005 study, QTcF peaked 2 hours post infusion and there was no QTcF effect seen at 24 hours. Categorical analyses showed no patients with a change from baseline >60 msec or an absolute QTcF >480 msec. No cardiac events of Torsade de Pointes were reported. Per the ICH E14 guidance,56 a threshold for regulatory concern for mean QTc prolongation is 5 msec with an upper bound of the 95% CI of 10 msec. Prolongations of >60 msec over baseline or a value >500 msec are also considered cautionary. Treatment-emergent morphological changes on ECG (minor ST-segment depression, T-wave flattening, biphasic T-waves, and T-wave inversion) have been observed during treatment with romidepsin; however the overall frequency of these minor, asymptomatic changes was similar to the frequency of pre-exposure abnormalities on each dosing day, suggesting a high degree of background noise in these data. Furthermore, there was no association between these ECG findings and any clinical, functional, or laboratory findings of cardiac abnormalities.Categorical Analysis of QTc Change on Cycle 1, Day 1CategoryStudyGPI Study N=87 n (%)NCI Study N=41 n (%)GPI-06-0005 N=7 n (%)Total N=135 n (%)QTcF change from baseline1No increase31 (36)10 (24)2 (29)43 (32)1-29 msec increase46 (53)27 (66)4 (57)77 (57)30-60 msec increase2 (2)4 (10)1 (14)7 (5)>60 msec increase0000QTcF not available8 (9)008 (6)QTcF absolute value:>450 msec2 (2)2 (5)04 (3)>480 msec0000>500 msec00001For GPI studies, baseline values are those post co-med, or if missing post co-med, then pre-comed baseline is used. ConclusionsThe findings from the romidepsin studies show that the observed QTc changes were below the threshold levels of concern outlined in ICH E14. Pharmacodynamic modeling showed no evidence of a relationship between the plasma concentration of romidepsin and changes in the QTc interval. Since hypokalemia and hypomagnesemia are associated with ECG abnormalities, supplementation with potassium and magnesium to achieve normal levels prior to romidepsin administration is recommended. Disclosures:Cabell:Gloucester Pharmaceuticals: Consultancy. Whittaker:Gloucester Pharmaceuticals: Research Funding. Nichols:Gloucester Pharmaceuticals: Employment, Equity Ownership. Nix:Gloucester Pharmaceuticals: Employment. Burris:Gloucester Pharmaceutcals: Research Funding.