A Novel Pan-Deacetylase Inhibitor with Activity in T Cell Lymphoma

Home , Panobinostat, Romidepsin, Vorinostat

Hematology Meeting Reports 2009;3(1):33–38 SESSION II

H.M. Prince Panobinostat (LBH589): M. Bishton a novel pan-deacetylase inhibitor with Peter MacCallum Cancer Centre and University of activity in T cell lymphoma Melbourne, Parkville, Australia

ABSTRACT

Panobinostat is a novel cinnamic hydroxamic acid pan-deacetylase inhibitor able to induce hyperacetylation of lysine residues on both histone and non-histone targets in cancer cells. Panobinostat is currently in clinical development, as both an intravenous and oral formulation, across multiple tumor types, and as proof of principle has resulted in prolonged histone hyperacetylation in tumor cells along with impressive anti-tumor activity in phase I and II trials, in particular T cell lymphomas, Hodgkin lymphoma and myeloid malignancies. Initial concerns regarding cardiac toxicities appear to have been answered by alterations in dose scheduling. Future studies may help unravel molecularly- or cytogenetical- ly-defined patient sub-groups who are most likely to benefit from panobinostat, either as a single agent, or, ultimately more likely, in combination with other therapies.

upregulate pro-apoptotic genes Introduction and the downregulate anti-apoptotic genes,3 activate cell-cycle Histone deacetylase (HDAC) checkpoints, induce cellular dif- enzymes and histone acetylases ferentiation, suppress angiogene- (HAT) control the relative degree sis and enhance host immune sur- of histone acetylation and deacety- veillance.3-5 Acetylation-dependent lation of the chromatin in DNA. In changes also occur in the activities this way, HDACs/HATs regulate the level to which a gene is tran- of important non-histone cytoplas- scribed by controlling the accessi- mic proteins such as p53, Ku70, 5 bility of DNA for transcription.1 α-tubulin and Hsp90. Multiple HDACs, which result in a net cellular pathways are therefore decrease in acetylation and thus affected simultaneously, resulting result in transcriptional silencing, in cancer cell death. Whether pan- have been divided into four gener- DACi, which inhibit both Class I al classes. Class I, located within and II HDACs, are superior to the cell nucleus; Class II and the selective DACi inhibiting class I Class IV HDAC, HDAC 11 which alone, is unknown. Panobinostat is shuttle between the cell cytoplasm a cinnamic hydroxamic acid pan- and nucleus. Class III HDACs deacetylase inhibitor with a consist of the NAD+-dependent marked DAC inhibitor activity at sirtuin family 1 to 7.2 By affecting low nanomolar concentrations HDACs and altering gene expres- both in vitro and in vivo, which has sion, the new anti-cancer drugs the undergone extensive preclinical deacetylase inhibitors (DACi) and clinical scrutiny.

| 33 H.M. Prince and M. Bishton

Preclinical data that panobinostat has potent anti-tumor activity against both hematologic and solid tumors in vivo In vitro . A HH CTCL mouse xenograft model showed complete tumor regression on intra- Panobinostat is a pan-DACi, with low venous panobinostat at 10 mg/kg when com- nanomolar concentration high inhibitory activi- pared to vehicle,9 and a dose-related reduction ty against all Class I, II and IV HDACs. in tumor burden with preservation of bone Panobinostat appears to be at least 10-fold more integrity in a MM xenograft mouse model. effective as a pure DACi when compared with Further experiments confirmed a significant vorinostat, the only currently FDA approved synergistic cytotoxicity between panobinostat HDACi. Marked anti-tumor activity across a and the proteasome inhibitor bortezomib with- broad range of cancer cell lines of hematologic out additional toxicity to normal bone marrow malignancies has been demonstrated, including stromal cells in vitro.10 Potent tumor regression cutaneous T cell lymphoma (CTCL), acute was also reported with panobinostat in colon, myeloid leukemia (AML), chronic myeloid pancreatic and small cell lung cancer (SCLC) leukemia (CML), multiple myeloma (MM), mouse xenograft models.9,11 Hodgkin lymphoma (HL), as well as solid The Eµ-myc B cell lymphoma model has tumors cell lines such as breast, colon, prostate been used to demonstrate that activation of the and pancreas. Moreover, panobinostat was up to intrinsic apoptotic pathway is required for the 100-fold more effective than vorinostat in the apoptotic and therapeutic activity of panobino- HCT116, BT474 and HH cell lines in its ability stat. Conversely, a functional extrinsic apop- to cause inhibition of cancer cell proliferation totic pathway or functional apoptosome is not and cell viability. It should be noted, that required for panobinostat mediated tumor cell although the cytotoxic effects of panobinostat death or therapeutic efficacy (Johnstone R.W were evident over multiple cancer cell lines, et al. submitted for publication). Moreover, there does appear to be differences in sensitivi- knockout murine experiments allow essential ty between haematological and solid tumor cell components of the apoptosome such as apaf-1 lines, with lower concentrations required for or caspase-9, to be deleted, resulting in cytotoxicity seen in AML, CML, HL and CTCL delayed panobinostat-induced lymphoma cells. Despite the marked anti-tumor activity killing in vitro and in vivo, associated with seen with multiple cancer cell lines at nanomo- suppression of a number of biochemical indi- lar concentrations, panobinostat only resulted in cators of apoptosis. Despite this, prolonged apoptosis of normal human cell lines at far exposure to panobinostat did result in tumor greater concentrations, indicating a selective regression, with morphological and biochemi- 6 toxicity for cancer cells. A similar phenomenon cal features of autophagy apparent, suggesting was also seen in human myeloma cell lines and that in the absence of a functional intrinsic myeloma cells from patients with MM resistant apoptosis pathway, panobinostat may initiate to standard chemotherapeutic agents, where an alternative cell death pathway. These pre- panobinostat was able to arrest growth in tumor clinical studies demonstrate that loss of viabil- cells without affecting normal peripheral blood ity, primarily through induction of apoptosis 7,8 mononuclear cells (PBMCs) or granulocytes. via the intrinsic apoptotic pathway but also through additional cell death pathways, pro- In vivo duces therapeutic efficacy in response to Early murine experiments have confirmed panobinostat.

| 34 | Hematology Meeting Reports 2009;3(1) 2006...2009: Now We Know T-Cell Lymphomas Better

Table 1. Phase I/II studies of Panobinostat in haematological malignancies.

Disease Phase No. Responses

Giles et al.12 Advanced leukemia/MDS I 15 HI 1 Sharma et al.14 NHL/Solid tumors I 44 PR 3 (CTCL, PTCL, HRPC) Prince et al.15 NHL/Solid tumors I 19 CR 2 (CTCL), PR 2 (CTCL) Ellis et al.16 CTCL I 10 CR 2, PR 4 Duvic et al.17 CTCL II 95 CR 4, Skin CR 2, PR Ottmann et al.21 AML/MDS/MF/MM/HL/NHL I/II 146 CR 3, CRi 1, PR 17, SD 14 Wolf et al.22 MM II 38 PR 1, SD 4 Siegel et al.23 MM (in combination with Bort) I 18 CR 1, VGPR 3, PR 5

NHL: Non Hodgkin lymphoma; HL: Hodgkin lymphoma; CTCL:Cutaneous T-cell lymphoma; MDS: Myelodysplasia; AML: Acute myeloid leukaemia; MF: Myelofibrosis; MM: Multiple myeloma; Bort: Bortezomib; CR: Complete response; PR: Partial response, SD: Stable disease; HI: haematological impro- vement; VGPR: Very good partial response.

Clinical studies involving panobinostat doses of panobinostat. For this schedule the maximum tolerated dose (MTD) was 20 mg/m2 with three dose-limiting toxicities In a two-arm, dose-escalation phase IA/II (DLTs) reported: fatigue, hyperglycemia (both study in patients with AML, acute lymphocyt- Grade 3) and thrombocytopenia (Grade 4). ic leukemia (ALL) or myelodysplastic syn- Despite intensive ECG monitoring, effects on drome (MDS), panobinostat was administered the QTcF interval appeared mild. One patient intravenously daily for seven days on a 21-day known to have had complete left bundle cycle. This dosing schedule was based on the branch block on enrolment had a QTcF inter- presumption that leukemic cells may need pro- val ≥500 milliseconds, and two further longed exposure to the drug for disease con- patients having transient asymptomatic trol. Fifteen patients were treated, and despite increases in QTcF (>60 milliseconds). Three a reduction in blasts in peripheral blood in patients had a partial response (PR), including seven patients, dose limiting cardiac toxicity one patient with CTCL, one with peripheral T was noted early, with Grade 3 QTcF prolonga- cell lymphoma (PTCL), and one heavily pre- tion in four patients, resulting in premature dis- treated patient with nodal hormone refractory 12 continuation of the study. In view of this, sub- prostatic cancer. Dose-dependent increases in sequent studies have utilized an intermittent histone H3 acetylation were reported in dosing schedule with minimal cardiac effects PBMCs seven days after both the first and observed to date.13 second doses of panobinostat at all dose lev- A further phase I study, examined a dose- els, suggesting sustained hyperacetylation of escalating, intravenous schedule of 10-20 tumor cells had been achieved as intended. mg/m2 given weekly for three weeks out of However, a high rate of thrombocytopenia four, was evaluated in 44 patients with resulted in several delays and dose reductions. advanced solid tumors or NHL.14 This sched- This has prompted investigation of a schedule ule was chosen primarily on the basis that a administering panobinostat on weekly for two sustained pharmacodynamic effect would be weeks in a three week cycle, with the intention present, whilst avoiding progressive QTcF of reducing the incidence and depth of throm- prolongation seen with consecutive daily bocytopenia with only a moderate reduction in

Hematology Meeting Reports 2009;3(1) | 35 | H.M. Prince and M. Bishton dose intensity. diarrhoea, thrombocytopenia and fatigue. A phase I dose-escalation trial of oral Notably, despite intensive ECG monitoring, panobinostat given on three days of each week only two patients have had a QTcF >480 ms, over a 28-day cycle on weekly for advanced and four had a QTcF increase of >60 ms from solid tumors or NHL enrolled 10 CTCL their baseline.17 Comparison of the results of patients, with two achieving CR, four PR, two panobinostat with other HDACi in advanced SD, and two patients PD. Two patients actual- clinical development for this disease, such as ly achieved CR several weeks after treatment vorinostat and romidepsin, is problematic cessation. The MTD was 20 mg, with three given the lack of consensus on the definition of DLTs seen in total; diarrhea, transient throm- response and disease progression in CTCL. bocytopenia and fatigue. The incidence of The different endpoints in studies used for Grade ≥2 QTF prolongation throughout was cutaneous manifestations of CTCL, such as the only 6.3%.15 Pharmacodynamic analysis of mSWAT and the modified Physicians Global histones from normal PBMCs reveal Assessment [PGA] tool) means meaningful increased acetylation at doses of 20 mg and comparison is near impossible given that the above, and for patients receiving doses of 20 interpretation of key efficacy data, including mg and 30mg, hyperacetylation was seen progression-free survival varies widely.18-20 beyond 72 hours in 50% of the patients fol- An ongoing two-arm phase IA/II study, has lowing their last dose. Given the accessibility enrolled 77 patients thus far, receiving either of tumor cells, correlative skin biopsies were weekly MWF or alternate weekly MWF treat- taken from six patients at several timepoints ment, with doses ranging from 20 mg - 80 mg. within the first 24 hours of drug administra- The arms are further sub-divided, allowing tion. Gene expression profiling of these biop- enrolling of patients with acute leukaemias and sies demonstrated that of genes whose expres- myeloid neoplasms, as well as enrolling sion as shown to be altered, the majority were patients with myeloma, HL and NHL. At doses down-regulated, while 23 genes were com- ≥20 mg, increased histone acetylation in both monly up- or down-regulated in all patients PBMCs and bone marrow core biopsies was tested. These genes had wide ranging func- seen. No responses were observed in patients tions important in cancer cell growth, includ- treated at doses <40 mg, nor at any dose level ing cell cycling, proliferation, angiogenesis in those treated alternate weekly. However, and immune regulation.16 anti-leukemic activity was evident in seven of Based on the results of the above study, a 12 evaluable patients at dose levels ≥40 mg, phase II study of oral panobinostat MWF is with two CR, one CRi and two patients with enrolling patients with refractory CTCL. The prolonged SD. Moreover, of the other three modified Severity Weighted Assessment Tool responders, one patient achieved CR several (mSWAT) was used to monitor skin response weeks after discontinuation of treatment, and CT to monitor systemic disease. Currently another had peripheral blood count recovery ninety five patients have been treated, and the following the end of treatment and one patient 62 patients who had previously received had SD for 1.5 treatment cycles and 10% blasts bexarotene, there have been 11 confirmed skin in bone marrow 10 months after the end of responses, including two complete skin treatment. Clinical activity in AML appears to responses. In the bexarotene naïve group four be both dose- and schedule-dependent, out of 33 patients have confirmed skin and CT although whether responses are purely related CR’s. Most common adverse events include to the duration and intensity of acetylation is

| 36 | Hematology Meeting Reports 2009;3(1) 2006...2009: Now We Know T-Cell Lymphomas Better not yet known. related rather than related to the pharmacoki- Impressive interim results were also seen in netic profile of the drug and include constitu- this study in HL, where response was assessed tional symptoms, particularly fatigue, gastroin- by both 18FDG-PET (PET) and CT. A total of testinal symptoms and myelosuppression.17,21 12 out of 18 patients demonstrated a metabol- Fatigue and transient thrombocytopenia has ic PR, and eight an anatomical PR. been dose-limiting in several studies but gen- Furthermore, seven of nine patients had reso- erally is rapidly reversible on drug cessation lution of their constitutional symptoms, while and subsequent dose reductions. In general, two patients have been on therapy for more neutropenia and anemia are less common than 13 months. On the basis of these results among panobinostat-treated patients than an international multi-centre trial for relapsed thrombocytopenia. Gastrointestinal AEs, or refractory HL has been initiated, and is including nausea, diarrhea and vomiting are already accruing at a starting dose of 40 mg also common but are generally of Grade 1/2 three times per week.21 severity and manageable in most patients. Results from phase II study of oral panobi- Initial concerns surrounding clinically signifi- nostat in patients with MM have also been cant QTc interval prolongation, an apparent reported. Of 38 patients enrolled, only one PR class effect of the DACi, appear to have been was seen, with a further four patients having answered with the use of alternative drug SD. This response rate was somewhat lower schedules.13 than expected given the pre-clinical data avail- able, and may relate to the comparatively low dose of 20 mg three weekly.22 The combination Conclusion and future prospects of oral panobinostat (10 or 20 mg three weekly continuously) with the bortezomib in Early results of phase I and II studies of patients with relapsed myeloma is under inves- panobinostat suggest efficacy against a range tigation. Thus far, nine clinical responses of hematologic malignancies, with particular among 18 evaluable patients have been seen, promise shown in CTCL, AML and HL. including one CR, three very good PR and five Despite recent progress in these diseases, PR). Moreover, three of the PR occurred in large numbers of patients remain in need of patients who had previously failed to respond novel and effective treatments for relapsed to bortezomib.23 The multitude of recent novel and refractory disease. Early combination biological agents approved in myeloma studies have already begun, and given the ensures that future studies of panobinostat in acceptable toxicity profile and synergy with combination with bortezomib, thalidomide, multiple anti-cancer agents, it seems likely lenalidomide as well as traditional chemother- phase II followed by phase III combination apeutic agents such as melphalan and steroids studies will follow in patients with poor prog- are guaranteed. nosis disease in the up-front setting. Other areas of potential interest for panobinostat Adverse effects include maintenance therapy, its use as a potential radio-sensitizing agent, combination Both formulations of panobinostat appear to with other immunomodulatory drugs, and a be generally well tolerated with a good safety greater knowledge of the influence of the bone profile. Similar to reports of other DACi, the marrow and lymph node microenvironment most common toxicities appear to be dose on drug response.

Hematology Meeting Reports 2009;3(1) | 37 | H.M. Prince and M. Bishton

Prince HM. Clinically relevant QTc prolongation is not References associated with current dose schedules of LBH589 (panobinostat). J Clin Oncol 2008;26:332-3; discussion 333-4. 1. Redner RL, Wang J, Liu JM. Chromatin remodeling and 14. Sharma S, Vogelzang NJ, Beck J, Patnaik A, Mita M, leukemia: new therapeutic paradigms. Blood Dugan M, et al. Phase I pharmacokinetic and pharmaco- 1999;94:417-28. dynamic study of once-weekly i.v. panobinostat 2. Glozak MA, Seto E. Histone deacetylases and cancer. (LBH589). European Cancer Conference (ECCO 14), Oncogene 2007;26:5420-32. 23-27 September 2007, Barcelona, Spain. Poster presen- 3. Bolden JE, Peart MJ, Johnstone RW. Anticancer activi- tation. ties of histone deacetylase inhibitors. Nat Rev Drug 15. Prince HM, George D, Patnaik A, Mita M, Dugan M, Discov 2006;5:769-84. Butterfoss D, et al. Phase I study of oral LBH589, a novel 4. Bishton M, Kenealy M, Johnstone R, Rasheed W, Prince deacetylase (DAC) inhibitor in advanced solid tumors HM. Epigenetic targets in hematological malignancies: and non-hodgkin's lymphoma. J Clin Oncol (Meeting combination therapies with HDACis and demethylating Abstracts) 2007;25(18_suppl):3500. agents. Expert Rev Anticancer Ther 2007;7:1439-49. 16. Ellis L, Pan Y, Smyth GK, George DJ, McCormack C, 5. Bhalla KN. Epigenetic and chromatin modifiers as tar- Williams-Truax R, et al. Histone deacetylase inhibitor geted therapy of hematologic malignancies. J Clin Oncol 2005;23:3971-93. panobinostat induces clinical responses with associated 6. Shao W, Growney JD, Feng Y, Wang P, Yan-Neale Y, alterations in gene expression profiles in cutaneous T-cell O’Connor G, et al. Potent anticancer activity of the pan- lymphoma. Clin Cancer Res 2008;14:4500-10. deacetylase inhibitor panobinostat (LBH589) as a single 17. Duvic M, Becker JC, Dalle S, Vanaclocha F, Bernengo agent in in vitro and in vivo tumor models. 99th MG, Lebbe C, et al. Phase II Trial of Oral Panobinostat American Association Cancer Research Annual Meeting, (LBH589) in Patients with Refractory Cutaneous T-Cell 12-16 April, 2008, San Diego, California. Poster presen- Lymphoma (CTCL). ASH Annual Meeting Abstracts tation. (Abstract 6244). 2008a 2008;112:1005. 7. Maiso P, Carvajal-Vergara X, Ocio EM, Lopez-Perez R, 18. Mann BS, Johnson JR, He K, Sridhara R, Abraham S, Mateo G, Gutierrez N, et al. The histone deacetylase Booth BP, et al. Vorinostat for treatment of cutaneous inhibitor LBH589 is a potent antimyeloma agent that manifestations of advanced primary cutaneous T-cell overcomes drug resistance. Cancer Res 2006;66:5781-9. lymphoma. Clin Cancer Res 2007;13:2318-22. 8. Catley L, Weisberg E, Kiziltepe T, Tai YT, Hideshima T, 19. Olsen EA, Kim YH, Kuzel TM, Pacheco TR, Foss FM, Neri P, et al. Aggresome induction by proteasome Parker S, et al. Phase IIb multicenter trial of vorinostat in inhibitor bortezomib and alpha-tubulin hyperacetylation patients with persistent, progressive, or treatment refrac- by tubulin deacetylase (TDAC) inhibitor LBH589 are tory cutaneous T-cell lymphoma. J Clin Oncol 2007;25: synergistic in myeloma cells. Blood 2006;108:3441-9. 3109-15. 9. Shao W, Growney JD, Feng Y, O'Connor G, Pu M, Yao 20. Bates S, Piekarz R, Wright J, Frye R, Figg WD, Sr., Allen Y-M, et al. Efficacy of Panobinostat (LBH589) in CTCL SL, et al. Final clinical results of a Phase 2 NCI multicen- Cell Lines and a Murine Xenograft Model: Defining ter study of romidepsin in recurrent cutaneous T-cell Molecular Pathways of Panobinostat Activity in CTCL. lymphoma (molecular analyses included). ASH Annual ASH Annual Meeting Abstracts 2007;110:1375. Meeting Abstracts 2008;112:1568. 10. Growney JD, Atadja P, Shao W, Wang Y, Pu M, Firestone 21. Ottmann OG, Spencer A, Prince HM, Bhalla KN, Fischer B, et al. Efficacy of Panobinostat (LBH589) in Multiple T, Liu A, et al. Phase IA/II Study of Oral Panobinostat Myeloma Cell Lines and In Vivo Mouse Model: Tumor- (LBH589), a Novel Pan- Deacetylase Inhibitor (DACi) Specific Cytotoxicity and Protection of Bone Integrity in demonstrating Efficacy in Patients with Advanced Multiple Myeloma. ASH Annual Meeting Abstracts Hematologic Malignancies. ASH Annual Meeting 2007;110:1510. Abstracts 2008;112:958. 11. Atadja P, Shao W, Wang Y, Growney JD, Cheng J, Kowal 22. Wolf JL, Siegel D, Matous J, Lonial S, Goldschmidt H, C, et al. Potent anticancer activity of panobinostat Schmitt S, et al. A Phase II Study of Oral Panobinostat (LBH589) in small-cell lung cancer in both in vitro and (LBH589) in Adult Patients with Advanced Refractory in vivo tumor models. Ann oncol 2008; viii50. Abstract Multiple Myeloma. ASH Annual Meeting Abstracts 86. 2008;112:2774. 12. Giles F, Fischer T, Cortes J, Garcia-Manero G, Beck J, 23. Siegel Dd, Sezer O, San Miguel JF, Mateos M-V, Prosser Ravandi F, et al. A phase I study of intravenous LBH589, I, Cavo M, et al. A Phase IB, Multicenter, Open-Label, a novel cinnamic hydroxamic acid analogue histone Dose-Escalation Study of Oral Panobinostat (LBH589) deacetylase inhibitor, in patients with refractory hemato- and I.V. Bortezomib in Patients with Relapsed Multiple logic malignancies. Clin Cancer Res 2006;12:4628-35. Myeloma. ASH Annual Meeting Abstracts 2008;112: 13. Zhang L, Lebwohl D, Masson E, Laird G, Cooper MR, 2781.

| 38 | Hematology Meeting Reports 2009;3(1)