Novel Targets And Strategies in Glioblastomas

Patrick Y. Wen, M.D.

Center For Neuro-Oncology Dana Farber/Brigham and Women’s Cancer Center

Division of Neuro-Oncology, Department of Neurology Brigham and Women’s Hospital

Harvard Medical School DISCLOSURES

• RhStResearch Support • Adv isory Boar d – Amgen – Merck – Astra Zeneca – Novartis – Boehringer Ingelheim – Vascular Biogenic – Esai – NeOnc Inc – EliiExelixis • Speaker – Genentech/Roche – Merck – Geron – Genentech/Roche – Medimmune – Merck – NtiNovartis – Sanofi-Aventis – Vascular Biogenics Treatment of High -Grade Gliomas Milest ones in Neuro‐OlOncology

AlApprovals

Radiotherapy TMZ up front TMZ for for GBM Lomustine relapsed AA Avastin for accelerated recurrent Carmustine Gliadel wafer approval GBM

1970 1980 1990 2000 2010

First US First US Macdonald commercial CT RANO commercial criteria: Brain Tumor Clinical Trial Criteria Levin criteria: MRI MRI + steroids; WHO Pathology Endpoints CT scans ASCO Criteria Workshop Workshop Technology Advances

AA=anaplastic astrocytoma; CT=computed tomography; GBM=glioblastoma multiforme; MRI=magnetic resonance imaging; RANO=Response Assessment in Neuro‐Oncology.

VEGF: VEGFR: , Brivanib, , , , Foretinib, , , , Pegdinetanib, , , Ang: CVX 060, Trebananib Tie-2: Cabozantinib EGF: ABT-806, , FGFR: Brivanib, Lenvatinib, Nintedanib , CXCR4:Plerixafor: Plerixafor HGF: IL-2: Basiliximab, Daclizumab ECM BLOOD VESSEL PDGFα: IMC-3G3, MEDI-575 PGF: RO5323441 VEGF: ,

NN Cilengitide Smo Notch INTEGRI c-KIT: Axitinib, Vismodegib MK0752 Cabozantinib, , Enzastaurin PKC LDE225 RO4929097 PLX3391, Sunitinib, Tandutinib Src , CXCR4:Plerixafor Dasatinib, ICN HSP EGFR: , Nintedanib Gli1/2 , , Client , , Protein Rind opepi mut (EGFR vIII) Tipifarnib HGFR/c-Met: PTEN PI3K Ras Cabozantinib, PDGFR: Axitinib, Brivanib, Vitespen HSP Crenolinib, Dasatinib, Raf Sorafenib Imatinib, Lenvatinib, PPifierifosine, Akt Nintedanib, Pazopanib, PX-866 Sorafenib, Sunitinib, Client Tandutinib MEK Protein TGF: LY2157299 Everolimus, Proteosome Sirolimus, mTOR Temsirolimus XL-765, ERK GDC-0084 AZD8055, Bortezomib CC223

Veliparib PARP HDAC Panobinostat, Chk Valproic Acid, Alexander, Lee DNA Vorinostat NUCLEUS et al. 2013

Outline

• Issues

• New therapies and targets

• New approaches to trial design RfLkfPiTtdMllReasons for Lack of Progress in Targeted Molecular Therapies in Glioblastomas

• Poor models • Blood-brain barrier • Co-actiiivation of tyrosi ikine kinases • Redundant sigggpynaling pathways • Spatial and temporal heterogeneity • Failure to genetically enrich patient population • Stem cells Bypassing The Blood Brain Barrier BhBenarroch NlNeurology 2012;78 :1268 Connolly et al, SNO 2012

Slide courtesy of May Han, Aveo Current Design

Surgery

Recurrent PK PK Tumor PET PET MRI MRI Blood biomarkers Blood biomarkers

Therapy Therapy

Convection‐Enhanced Delivery Low density Lipoprotein Receptor Related Protein (LRP‐1)

TPA (tissue plasminogen activator) Angiopep

2-Macroglobulin (MW ~ 700 kDa) • Transports small and large RAP molecules (> 40 ligands)

 • One of the most expressed Thyroglobulin Lactoferrin receptor at the surface of the BBB • Expressed on cancer cells

Cell Membrane ß • Expressed also in liver, lung and ovarian tissues

LRP-1: ~600kDa (515,  85) ANG1005 ANG1005

Low density lipoprotein receptor related protein (LRP1) ANG1005 ANG1005 ANG1005 ANG1005 ANG1005

2013; 19(6):1567‐1576 Science 2011;344:1727 Pulse Dosing Vivanco et al. Cancer Discovery 2012;2:458‐71

Vivanco et al. Cancer Discovery 2012;2: 458‐271

GBM Lung Cancer Vivanco et al. Cancer Discovery 2012 Vivanco et al. Cancer Discovery 2012 Are we delivering the drugs appropriately? • Continuous dosing versus pulse dosing • Adm in is ter t arget ed d rug b ef ore or aft er ch emoth erapy • NSCLC – Solit et al CCR 2005;11:1983 – Riely JCO 2009:27;264 Targeted Molecular Therapies

The Cancer Genome Atlas Research Network. Nature. 2008;455:1061-1068; Stum et al. Cancer Cell 2012;22:425‐437 Failure To Genotype Patients Sequencing

Epigenetic Analysis

Set of activated Combinations of kinases and appropriate drugs pathways

Ivy Foundation Early Phase Clinical Trials Consortium DF/HCC MSKCC UCLA UCSF MDACC U Utah New Targets

• PI3K • FGFR/TACC • BRAF • CDK4 • WeeI PI3 Kinase Inhibitors

Growth Factors, etc

Ras

Raf PI3K inhibitor XL765 Mek XL147 BKM120 PX866 Erk GDC0084

Proliferation BKM120

• Oral pan-class I phosphatidylinositol-3-kinase (PI3K) inhibitor belonggging to the 2 ,6-dimorpholino pyrimidine derivative family • Inhibits p110α,,p p110β,,p p110δ and p110γ • Cross the blood-brain barrier (brain/blood ratio 2) • TkTaken orall y once d dilaily • Inhibits the growth of U87MG and GBM explants in vivo

39 Ivy Foundation Early Phase Clinical Trials Consortium A Phase II study of BKM120 for patients with recurrent glioblastoma and activated PI3K pathway BKM 120 T ri al

Patient Eligibility • Activation of PI3K pathway: – PIK3CA/PIK3R1 mutation or – PTEN mutation, loss of PET by FISH, or PTEN IHC negative Goal • 30 PTEN loss • 20 PIK3CA/PIK3R1 mutants

44 Next Steps

• Isoform specific Inhibitors – ? Beta specific isoforms better for PTEN loss – ? Alpha specific isoforms for PI3Ca mutants • Combinations – BKM120 +RT+TMZ – BKM120 + LDE225 (SMO inhibitor) – BKM120 + INC 280 (MET inhibitor) Science 2012 Phase II Trial of BGJ398 (FGFR inhibitor) TCGA

The Cancer Genome Atlas Research Network. Nature. 2008;455:1061-1068; PD-0332991 (CDK 4/6 inhibitor) Michaud et al: Cancer Res 2010;70:3228 Michaud et al: Cancer Res 2010;70:3228 LY2835219 (CDK4/6 Inhibitor) Sanchez Martinez et al Sanchez‐Martinez et al BRAF and/or MEK inhibitors for BRAFV600E mutated gliomas? Flaherty et al. Combined BRAF and MEK Inhibition in Melanoma with BRAF V600E Mutations NEJM 2012; 2012 Nov;367(18):1694-703 A Phase II, Open-label Study in Patients with BRAF V600EMutated Rare Cancers with Several Histologies to Investigate the Clinical Efficacy and Safety of the Combination Therapy of and

Dabrafenib

Trametanib Histologies

• Anaplastic Thyroid Carcinoma • Biliary Tract Cancer • Diffuse Large B Cell Lymphoma • GIST • Hairy Cell Leukemia • Hig h Gra de Glioma (GBM, AliPXAAnaplastic PXA, AliAnaplastic ganglioglioma) • Low-Grade Gliomas (PXA, Ganglioglioma, Pilocytic Astrocytoma) • Multippyle Myeloma • NSGCT/NGGCT • Small Intestine Adenocarcinoma Plummer: Clin Cancer Res 2010;16(18); 4527–31

Double strand break Single strand break

Non- homologous Homologous end joining Recombinati (NHEJ) on Plummer: Clin Cancer Res 2010;16(18); 4527–31 IiCttiitfTMZdRTIncreasing Cytotoxicity of TMZ and RT

• PARP Inhibitors – ABT 888 (ABTC; RTOG) • Wee1 Inhibitor – MK1775 W1Wee1

MK1775

G2 Arrest

G2 progression and mitotic catastrophe Glioma Initiating Cells

Wen PY, Kesari S. N Engl J Med 2008;359:492-507 Glioma Initiating Cells

Wen PY, Kesari S. N Engl J Med 2008;359:492-507 BKM120+LDE225 In vivo data: orthotopic xenograft of PTEN‐dfiideficient tumor

Mariella Gruber-Olipitz et al. Nature Med. In Press LDE225+BKM120

Combo A MM an NU mm Hu MhiMany choices; limited resources Challenges

• Phase II trial s oft en n ot pr edi cti ve o f pos itive ou tco me in phase III studies – Imatinib mesylate and hyyydroxyurea – Enzastaurin – Cediranib – Cintredekin Besudotox (IL13-PE38QQR) – Cilengitide • More drugs and more combinations • Liidimited resources • Need more efficient trial designs • Need better response criteria, endpoints and more efficient trials and design

New Clinical Trial Designs

• Improve efficiency • Rapid elimination of ineffective regimens • Test multiple combinations simultaneously • Shorter path to definitive testing • DiDesigns • “Pick the Winner” • SliSeamless integrati ifh/on of phase II/III tri ilals • Sequential Accrual Design for Phase I/II studies • Factorial Design • Adaptive Randomization Adaptive Randomization Strategies

• Multiple arm study • Allocation of patients based on Bayesian probability of treatment efficacy • Treatment arms with success are more likely to accrue patients • Treatment arms with poor results are dropped , alternative arms added, and accrual continues until clear evidence of superior treatment(s) emerge

A + B

ACA + C Winner A + B + C

Adapti ve R and omi zed “B ayes ian ” Des ign

Courtesy of Mark Gilbert, MD. Trippa et al (JCO 2012; 30:3258-3263) Trippa et al (JCO 2012;30:3258 -3263) Trippa et al (JCO 2012 ;30:3258-3263) Ideally we can have trials that evaluate multiple drugs efficiently and identify predictive biomarkers

Wha t Mo lecul ar S ub groups ?

• EGFR amplification (45%) or mutation (20%) • PTEN loss (40%) / PIK3Ca or PIK3R1 mutations (15%) • CDK4/6 amp (18%); P16/15 loss (50%) • MDM2 amplification (15%) (with intact p53 • PDGFR amplification (13%) • Other less common subgroups Predictive or Prognostic Biomarkers

Predictive

Prognostic EGFR PI3K MDM2 CDK4/6 n ++++1 (1%) +++‐ 3 (3%) ++‐ +3 (3%) ++‐‐11 (12%) + ‐ ++4 (4%) + ‐ + ‐ 0 (0%) + ‐‐+1 (1%) TCGA (all) + ‐‐‐18 (20%) Courtesy of ‐ +++2 (2%) Brian Alexander ‐ ++‐ 1 (1%) ‐ + ‐ + 0 (0%) ‐ + ‐‐17 (19%) ‐‐++1 ((%)1%) ‐‐+ ‐ 1 (1%) ‐‐‐+4 (4%) ‐‐‐‐24 (26%) 41 (18) 38 (17) 13 (1) 16 (4) 91 How to prioritize overlapping classification

• Adaptive design with equal or weighted randomization will allow each subgroup to be evaluated Biomarkers

EGFR Pik3Ca/PIK3R1 CDK4/6 amplification/ mutations/ PTEN amplification/P16 mutation loss loss

+++ ++0 +0+ +00 00+ 0 + 0 0++ 000 Recurrent GBM Red: EGFR Amplification Green: PDGFRA Amplification PNAS 2013;110:4013 Recurrent GBM

Reoperation and Genotype

3‐5 weeks RANDOMIZE

EGFR MDM2 PIK3Ca/R1 CDK4 Amplified Amplified mutations amplification 40%; 14% 15% 16%; mutations or loss of P16 and 15 20% PTEN deletion 40% 50%

Drug A or Drug C Drug E or Drug G or A+B or C+D E+F G+H

Survival SilStdSurgical Study Possible Studies

• PI3K + MEK

• EGFR/mTOR inhibitor + Arsenic Trioxide

• etc PNAS 2013; 110:4339‐4344 Iwanami et al: Cell Cycle 2013:12:10, 1473–1474 U87 Human GBM We are slowly making progress! Thank You!