-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

592 Targeting the Hsp90 Oncoproteome in Burkitt Lymphoma

Lymphoma: Pre-Clinical – Chemotherapy and Biologic Agents
Program: Oral and Poster Abstracts
Type: Oral
Session: 625. Lymphoma: Pre-Clinical – Chemotherapy and Biologic Agents: Novel Therapies and Targets in Lymphoma
Monday, December 7, 2015: 11:15 AM
Tangerine 1 (WF1), Level 2 (Orange County Convention Center)

Lisa Giulino Roth, MD1,2, Herman van Besien3*, Anna Rodina, PhD4*, Tony Taldone, PhD5*, Hediye Erdjument-Bromage, PhD6*, Matthew J. Barth, MD7, Gabriela Chiosis, PhD5* and Ethel Cesarman1

1Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, NY
2Pediatrics, Weill Cornell Medical College, New York, NY
3Weill Cornell Medical College, New York, NY
4Memorial Sloan Kettering Cancer Center, New York, NY
5Department of Molecular Pharmacology and Chemistry, Memorial Sloan-Kettering Cancer Center, New York, NY
6New York University, New York, NY
7Department of Pediatrics, Roswell Park Cancer Institute, Buffalo, NY

 

Introduction: Novel therapies are urgently needed in pediatric Burkitt lymphoma (pBL) where the survival for relapsed disease is less than 20%.  Heat shock protein 90 (Hsp90) is molecular chaperone that protects proteins from proteolytic degradation including oncogenic signaling complexes.  The clinical development of broad-spectrum Hsp90 inhibitors has previously been limited by suboptimal target inhibition and off-target toxicities.  PU-H71 is a next-generation Hsp90 inhibitor that preferentially targets tumor enriched (te-Hsp90), the functionally distinct pool of Hsp90 present in tumor cells.  PU-H71 is not toxic to normal B-cells and has demonstrated pre-clinical efficacy in diffuse large B-cell lymphoma, but has not been studied in Burkitt lymphoma.  In the current study, we evaluated te-Hsp90 as a potential therapeutic target in pediatric Burkitt lymphoma. 

Methods and Results:  To evaluate overall Hsp90 protein expression in primary pBL tumors we performed immunohistochemistry on a tissue microarray.  Fifty-three of 59 cases (90%) demonstrated high levels of Hsp90 expression defined as >90% tumor cell positivity (Fig. 1A).  To evaluate the sensitivity of pBL to te-Hsp90 inhibition we performed in-vitro viability assays with the ATP-based CellTiter-Glo¨ in a panel of pBL cell lines (Ramos, DG-75, Raji, Namalwa, Daudi, Jiyoye, CA-46, Raji 2R, Raji 4RH) and in-vivo studies with a Ramos xenograft model of pBLpBL cells were sensitive to inhibition by PU-H71 with IC50s in the low nanomolar range (151-337nM, Fig 1B).  The Raji 2R and Raji 4RH pBL cell lines with acquired resistance to chemotherapy (Czuczman et al, Clin Cancer Res 2008) were also sensitive to PU-H71 (IC50 175-181nM).  In contrast, normal peripheral blood lymphocytes were resistant (IC50 >37,000nM).  In pBL xenograft studies, PU-H71 decreased tumor volume (p<0.01, Fig. 1C) and prolonged survival (p<0.01, Fig. 1D). 

To evaluate the targets of PU-H71 in pBL we performed high affinity capture followed by proteomic analysis using mass spectrometry.  Cellular lysates from pBL cell lines were incubated with PU-H71 conjugated agarose beads and the cargo was subject to tandem mass spectrometry.  Data was analyzed using Ingenuity Pathway Analysis¨.  PI3K and mTOR signaling were among the top networks identified.  Given the data supporting a crucial role for PI3K in BL (Sanders et al, Cancer Cell 2012), we further investigated protein targets in this pathway.  We found a significant number of proteins in the mTOR (p<0.0001) and PI3K (p<0.0001) signaling pathways including ERK, mTOR, and multiple isoforms of PI3K (p110-β; p110-γ, p110-δ).  PI3K targets were validated by pull-downs with PU-H71 conjugated beads (Fig 2A).

To investigate the activity of PI3K inhibition in pBL we evaluated a panel of PI3K inhibitors in pBL.  Agents with a narrow spectrum of PI3K isoform inhibition such as idelalisib (p110δ inhibition) and IPI-145 (p110 δ and γ inhibition) did not have activity in pBL.  In contrast, agents that broadly target PI3K/mTOR such as BEZ235 (pan PI3K and mTOR inhibition) and BKM120 (pan PI3K inhibition) demonstrated activity (IC50 35nM-2602nM).   Based on this we concluded that PI3K inhibition is effective in pBL only when multiple components of the pathway are targeted.  We hypothesized that PU-H71 will therefore synergize with PI3K inhibitors through dual targeting of PI3K signaling.  To test this we evaluated the combination of PU-H71 + BKM-120 and PU-H71 + BEZ235 in four pBL cell lines (Ramos, Namalwa, Daudi, DG-75).  The combined response at 48-72 hours was evaluated using the Chou-Talalay method.  Both combinations were synergistic in 3 of 4 cell lines with a combinatorial index at the IC50 of <1 (CI 0.44-0.79, Fig. 2B). 

Conclusion: Our work demonstrates that Hsp90 is over-expressed in primary pBL tumors and that te-Hsp90 inhibition with PU-H71 is toxic to pBL in-vitro and in-vivo.  The oncoprotein targets of Hsp90 in pBL include multiple components of the PI3K/mTOR signaling pathway, highlighting the importance of this pathway in pBL.  The anti-lymphoma activity of PU-H71 is synergistic with pan-PI3K inhibition as well as dual PI3K/mTOR inhibition.  Overall this work provides support for te-Hsp90 as a therapeutic target in BL and suggests the potential for combination therapy with PU-H71 and inhibitors of PI3K/mTOR. 

<Description: Macintosh HD:Users:hjv2001:Desktop:Figure 1.jpg

<!Description: Macintosh HD:Users:hjv2001:Desktop:Figure 2.jpg

Disclosures: Cesarman: Weill Cornell Medical College: Patents & Royalties: applied for patent for 6-ETI .

*signifies non-member of ASH