to cancer research? Jason B. Fleming, MD Chair, GI Oncology H. Lee - - PowerPoint PPT Presentation

How can a surgeon contribute to cancer research?

Jason B. Fleming, MD Chair, GI Oncology

• H. Lee Moffitt Cancer Center

No Relevant Disclosures

• None

Projected Increase in Deaths from Pancreatic Cancer in the US

www.pancan.org 2015

Surgery offers Hope of Cure

• National Cancer Data Base (~76% of cases in US)
• 4739 patients received PD in 587 hospitals from 2010 and 2011 Institutions categorized into

quartiles based on PD case volume

• 30-day and 90-day mortality

Contemporary Surgical Outcomes

Kulu and Conrad, unpublished, 2016. 1-5 >25 14-25 6-13 Cases per Year: >6%

10,000 Hours and Pancreatic Surgery

The Surgical “Learning Curve”

Tseng, Surgery 141: 456 (2007).

MDACC surgeons (n=3) ~ 60 pancreas resections are necessary before a significant reduction in EBL, OR time and hospital stay is realized.

EBL OR Time Hospital Stay

Improvements with Resection of PDAC

• 1970-2006 Johns Hopkins University
• Postoperative complications: 38%
• Mortality rate: 1%
• Median OS: 19 mo; 5-yr OS: 20% in 2000’s

Winter, J Gastrointest Surg 2006.

Progressively Limited Return

• n

the “investment’ of Pancreas Surgery

Treatment Phase Treatment Break

Best-evidence Therapy- but based upon incomplete information for individual patient

Staging Repeat OR Staging Establish Dx Dropout “Could not make the leap across”

The Preoperative Therapy Approach

Varadhachary et al., Ann Surg Onc 2006 Katz et al., JACS 2008

Observation: Patients fail mainly when systemic therapy ineffective.

Dx Preop Rx Surgery Resistant 30-40% 60-70% 65% Recurrence 35% Survivors Follow Up Clinicopathologic Information

Drop out

Observation: Distant Failure Predominates

Ann Surg Oncol. 2009 April; 16(4): 836–847.

Observation: Ideal systemic therapy and local control can cure

Gem/Cis Followed by XRT (50.4Gy) with Bevacizumab Erlotinib Capecitabine. Path CR 0/18 Nodes Pos. NED for >5yrs

Questions on Whipple Number 50….

How can we get more research going in pancreatic cancer?

What is something cool that I can do with all these pancreatic cancer tumors I am taking

• ut every week?

Kim, et al. Nature Protocols, 2009.

Tumor Implantation Subsequent Xenograft Tumor Harvest Tumor Preparation Original Tumor Xenograft Tumor Original Tumor Xenograft Tumor

Dir irect Xenografts from Surgic ical Specimens

Patient-Based Resource Building

Banked TMA Tumorgraft Repository Dx Preop Rx Surgery Resistant 30-40% 60-70% 65% Recurrence 35% Survivors Follow Up Clinicopathologic Information Banked Banked

1o Reagents 2o Reagents

Post-treatment RNA/DNA RNA/DNA Immortalized PBMCs Organoids Cell Lines blood sample = tumor sample Ex vivo Testing Platform tumor sample

Learning Strategy for Improvement

New Patients

Preop Rx Dx Follow Up Sx

Novel Rx Novel Rx Biomarker

Common Platform

PDAC PDX Program at MDACC

PATC (Pancreatic tumor cell line) Xenografts Cell lines DNA

Tumor tissue

F1 F2 F3 DNA RNA Protein PATX (Pancreatic tumor xenografts) Storage EVOC/LTSA TMA

EVOC: Ex Vivo Organotypic Culture LTSA: Live Tissue Sensitivity Assay

Number of Patients consented/implanted 220/204 cases Number of Xenograft tumors 131 (64%) PDAC cell lines from xenograft tumors 23 Characterized cell lines* 14/23 Fingerprinted 14/23 Collaborating PIs 60 PIs or Labs Annual new xenografts 15 to 25 cases

(02/2008-05/2016)

Summary of PDAC PDX Reagents

P rim a ry s ite (8 7 .7 % ) L iv e r m e t (6 .8 % ) B o n e m e t (1 .5 % ) M a lig n a n t a s cite (0 .7 % ) P e rito n e a l m e t (1 .5 % ) L u n g m e t (0 .7 % )

T o ta l= 1 3 1 P D X s P rim a ry site (8 7 .7 % )

L ym p h n o d e m e t (0 .7 % ) S u c c e s s fu l 1 3 1 /2 0 4 U n s u c ce s s fu l 7 3 /2 0 4

T o ta l= 2 0 4

Observation: Time is your enemy

Time from Specimen Removal (minutes) % TUNEL pos Nuclei/HPF

30 60 120 180 20 40 60 80 100

Goal: engraftment time <30 minutes Surgical Advantage

What Happens in Operating Room/Pathology:

• 1. Blood supply ligated
• 2. Specimen removed
• 3. Sits on back table
• 4. Circulator nurse
• 5. Path Tech paged
• 6. Path Tech arrives
• 7. Delivered to Path
• 8. Sits on table
• 9. Specimen registered
• 10. Sits on Dissection table
• 11. Tumor dissected
• 12. Research samples obtained
• 13. Research Tech paged
• 14. Research Tech arrives
• 15. Research Tech takes specimen
• 16. Downstream experiment

Several Hours!

Halling, et al. 2003

Observation: Tumorgraft growth reflects tumor biology

• Disease-free survival
• Overall survival

(n=14) (n=56) (n=14) (n=56)

Ann Surg Oncol (2015) 22: 1884-1892

Tumorgraft growth Yes No Median survival (days) 613 2067

PATX43 PATX50 PATX53 PATX66 MDA-PATC43 MDA-PATC50 MDA-PATC53 MDA-PATC66 Patient43 Patient50 Patient53 Patient66

Observation: Plastic Changes Cancer Cell Populations

Tumorgraft Primary PC Cell Lines

0% 50% 100% PT43 PATX43 SUB-… PT50 PATX50 SUB-… PT53 PATX53 SUB-… PT66 PATX66 SUB-… N-Cadherin 0% 50% 100% PT43 PATX43 SUB-… PT50 PATX50 SUB-… PT53 PATX53 SUB-… PT66 PATX66 SUB-… E-Cadherin

Kang, et al. Lab Investigation. 2015.

Ivanics, et al. 2017.

Technical Advance: Cryostorage and Reanimation of PDX

Xenograft source Prior therapy Engraftment rate with traditional method Engraftment rate with biopsy method Primary Yes 3/5 0/5 Primary No 3/5 1/5 Met Yes 5/5 5/5 Primary No 1/5 0/5 Primary Yes 0/5 0/5 Primary Yes 0/5 0/5 Primary Yes 4/5 0/5 Primary No 4/5 3/5 Met Yes 3/3 N/A Primary Yes 0/5 0/5

• Roife.Surgery. 2017.

Technical Advance: Fine Needle Aspirates Generate PDX

Questions on Whipple Number 150….

How can we use the xenografts to learn more?

I need some people way smarter than me….

Nat Med. 2011 Apr;17(4):500-3. Material: RNA extracted from microdissected PDAC cells + cell lines Hepatogastroenterology 55, 2016–2027 (2008) Classical QM-PDA Exocrine

62 gene signature

Subtyping by Gene Transcription Signature

Survival of 27 Resected Cases (UCSF) Classical is K-ras-driven

PDX Transcription Profiles

Collisson Moffitt Classical QM-Basal PDX provide excellent quality reagents for reproducible WES

PDX Chris Bristow and Tim Heffernan

Inference of regulators of basal /classical tumors

• Leverage compendium of PDX model data to infer pathway activity
• virtual inference of protein activity by enriched regulon analysis (VIPER)

data

analysis

RNAseq profilesA TCGA PDAC-specific interactome Subtype classification

VIPER

Alvarez et al. Nat Gen, 2016

VIPER Classical regulators Basal regulators Subtype associated regulators

Examples of context-specific regulators (VIPER)

Basal Classical PDX inferred activity

Martinelli et al, Gut, 2015

GATA6 suppresses EMT KLF5 regulates epithelial genes

Diaferia et al, EMBO, 2016

Goal: Cross-cutting Searchable Data Platforms

Imaging/Biophysical

Molecular

Tumor Sensitivity Clinicopathologic

Questions on Whipple Number 250….

Are there any clinically available data that could help?

I need even more people way smarter than me….

Patient 1 Patient 2

Imaging Subtypes Observed

Koay, Truty, Cristini, et al. J Clin Invest. 2014 Apr 1;124(4):1525-36.

A

HU High delta Low delta Segmentation for delta Example: Histogram

B

Koay, unpublished, 2017.

Low and High Delta Subtypes

Overall survival (OS) stratified by delta measurement for patients

Summary

Stromal content/tumor cell proliferation ≈ Stability Parameter (L)

Koay, unpublished, 2017.

Imaging Characteristics Are Reflected in the Growth Kinects of Tumorgrafts

r=0.4752 R2=0.2258 P=0.0080 r=0.1288 R2=0.01659 P=0.7832

High Delta Low Delta

PATX-50 PATX-69 PATX-66 PATX-102 PATX-118

Hi D Lo D

Texture analysis of MR images of PATX tumors

• Survival curve of 6 xenografts

IBEX (open infrastructure software platform, imaging biomarker explorer)

Low delta High delta

Mutation Classification and Imaging Subtype

( B )

Proportion with

Koay, unpublished, 2017.

Example: PATX-69

Tissue segmentation

• Red: tumor
• Green: stroma
• Blue: normal pancreas

Observations: Stability of stroma with PDX

• Comparison of stroma content in tumor from patients and

from patient-derived xenografts

• Representative images

F0 F1 F2 F3 F0 F1 F2 F3 F0 F1 F2 F3 F0 F1 F2 F3 20 40 60 80 100

Tumor Generation Collagen Area Fraction % PATX1 PATX4 PATX7 PATX11

INDIVIDUAL PATIENT TUMOR PHENOTYPE DETERMINES DEGREE OF FIBROSIS

Low delta High delta H&E CT scan Imaging Phenyotype Low delta High delta Stroma score (by pathologist) Koay, unpublished, 2017

Low and High Delta Subtypes and Stromal Collagen

PATIENT CASES

Low delta High delta

Tissue category

PDX

Nuclei of cancer cells from high delta tumors are more elongated suggesting aggressive biology

• Test set (12 cases)
• Validation set (17 cases)

1 2 3 4 1 2 3 1 2 3 4 5 2 3 4 5 1 2 4 5 1 3 5 1 2 3 1 2 3 4 5

Stability of nuclei morphology of cancer cells with passaging

• Representative H&E stained images (PATX 118)

F1 F2 F3 F4 F5

1.0

PDAC Cell Shape and Imaging Phenotype

Koay, unpublished, 2017.

Nucleus morphology of cancer cells in PDX

Red: cancer cells, Green: stroma cells, Blue: lymphocytes

Low delta High delta

1.0

Goal: Cross-cutting Searchable Data Platforms

Imaging/Biophysical

Molecular Tumor Sensitivity Clinicopathologic

Questions on Whipple Number 350….

How do the medical

• ncologist know

what kind of chemo to give?

Could we make a simple test to find

• ut?

0.5 1 1.5 Relative viability

MDA-PATX121

0.5 1 1.5 Relative Viability

MDA-PATX124 H&E αSMA CD34 KI-67 Day-0 Day-3 Day-5

Trichrome

0.2 0.4 0.6 0.8 1 1.2 1.4 Ctrl AUR 1µM AUR 3µM AUR 10 µM Relative Vaibility

PATX137-F2 Growing xenografts

Taking cores Cutting slices Drug testing Viability Assay Reading result

Resazurin Resorufin (579/584 nm) Linear range: 50-50,000 cells/well in 96-well plate Resazurin-based viability assay

EVOC: Ex Vivo Organotypic Culture LTSA: Live Tissue Sensitivity Assay LTSA/EVOC Preclinical Drug testing Developing Novel Therapy PDX/Patient Tumors Optimizing SOC Therapy

Treatment Strategy Equipoise (Ambiguity)

Gemcitabine (1997) FOLFIRINOX (2011)

• 1. Von Hoff DD, J Clin Oncol 2011; 29:4548-54. [PMID 21969517]
• 2. Von Hoff DD,. N Engl J Med 2013; Oct 16. [PMID 24131140]
• 3. Therasse P, J Natl Cancer Inst 2000; 92:205-16. [PMID 10655437]
• 4. Conroy T, N Engl J Med 2011; 364:1817-25. [PMID 21561347]

Gem/Abraxane (2013)

P=0.0185 P=0.0168 0.2 0.4 0.6 0.8 1 1.2

Relative Viability

0.2 0.4 0.6 0.8 1 1.2

Relative Viability

P=0.124 P=0.199

A MDA-PATX76-F4

P=0.561 P=0.764 0.2 0.4 0.6 0.8 1 1.2 1.4 Contorl Irino 10 µM Irino 30 µM

Relative Viability

P=0.0717 P=0.0156 0.2 0.4 0.6 0.8 1 1.2 Relative Viability

MDA-PATX106-F4

PARP C-PARP β-Actin Caspase3 C-Caspase3 PARP C-PARP β-Actin Caspase3 C-Caspase3

MDA-PATX106 MDA-PATX76

30 Gem (µM) Irino (µM) 100 10 30 30 Gem (µM) Irino (µM) 100 10 30 Control Control

LTSA LTSA Gem Sens/Irino Res Gem Res/Irino Sens

** 100 200 300 400 500 600 700 Tumor Volume (mm3) PBS Irinotecan

Irinotecan PBS

R=-0.676, P=0.002

Cut-off AUC 0.8 0.9 0.75 0.95 0.7 0.99 0.69 0.99 0.68 1 0.67 1 0.66 0.94 0.65 0.92 0.5 0.97

ROC analysis

R e s is ta n t S e n s itiv e 1 0 2 0 3 0

PFS (months) P=0.011

0 .5 1 .0 1 .5

• 1 0

1 0 2 0 3 0

L T S A V a lu e v s P F S

L T S A V a lu e P F S M o n th s

PATX LTSA Sensitivity LTSA Value PFS Average PFS

MDA-PATX76

S 0.63936529 9

MDA-PATX81

S 0.65977384 11

MDA-PATX106

S 0.43088763 27 16.3

MDA-PATX107

S 0.66584785 26

MDA-PATX141

S 0.61646373 11 MDA-PATX142 S 0.30694747 14 MDA-PATX161 R 0.68007825 8

MDA-PATX97

R 0.73359011 5

MDA-PATX100

R 0.84183624 3.8

MDA-PATX104

R 0.99103035

MDA-PATX118

R 0.79856677 6

MDA-PATX124

R 0.80463465 4

MDA-PATX137

R 0.75039174 4

MDA-PATX136

R 1.02023144 8

MDA-PATX144

R 1.06664134 6

MDA-PATX140

R 1.18407554

MDA-PATX148

R 0.94166936

MDA-PATX153

R 1.00245356 5

A B C D PATX76 FOLFIRINOX

LTSA and Clinical Response

Do we need a mouse at all?

E x -v iv o ch e m o se n sitiv ity a ssa y P D A C 2 0 4 -F 0

F lu o re sce n ce

U n tre a te d T re a te d ( M )

• FOLFIRINOX 30:12:6 effectively suppressed tissue viability

48hrs after treatment compared to control (p=0.009). Newly Diagnosed Met. PDAC Laparoscopy w Liver Met Bx

Results Available within 48 hours of biopsy and within 5 days from initial evaluation!

CA 19-9

EVOC: Ex Vivo Organotypic Culture LTSA: Live Tissue Sensitivity Assay LTSA/EVOC Preclinical Drug testing Developing Novel Therapy PDX/Patient Tumors Optimizing SOC Therapy

Control MK2206 P-AKT P-AKT MDA-PATX135 MDA-PATX140 P-AKT P-AKT

D

Control MK2206 Control AZD6244 P-ERK 42/44 P-ERK 42/44 P-ERK 42/44 P-ERK 42/44 Control AZD6244

0.2 0.4 0.6 0.8 1 1.2 Relative Vibility

MDA-PATX135

P-ERK(42/44) P-AKT(S473) Pan-AKT Beta- actin

MDA-PATX135

A B

0.2 0.4 0.6 0.8 1 1.2 Relative Viability

MDA-PATX140

ERK(42/44)

MDA-PATX140

Agents/combinations being tested Oxphos Inhibitor, IACS:10759: single agents and combination Paclitaxel Gemcitabine/Digoxin combination Trametinib /2DG combination IL-1b inhibitor single agent β-Lapachone/PARP inhibitor combination DDR1 inhibitor MEK/CDK, MEK/HDAC combinations

In vivo validation EVOC/LTSA to Identify the responders and non responders TMA Staining, RNASeq, WES data analysis to identify the PDXs with/without targets presence Re-grow PDXs panel in mice Correlation analysis to identify and validate predictive biomarkers

EVOC Platform for drug activity testing

50 100 150 Suvival % IACS 10759

PATX60

50 100 150 Suvival % ICAS10759

PATX102

A B C

EVOC , Ex Vivo Organotypic Culture, also termed LTSA, Live Tissue Sensitivity Assay (Rofie et al., Clinical Cancer Research, 2016) ** ** **

Testing Pipeline: PDX Program at MDACC

“Evergreening” PDAC PDX Program at MDACC

PATC (Pancreatic tumor cell line) 14 primary cell lines with characterization Xenografts Cell lines DNA

Tumor tissue

F1 F2 F3 DNA RNA Protein PATX (Pancreatic tumor xenografts) Storage EVOC/LTSA TMA 48 PDXs with RNAseq, WES 80 PDXs in TMA

EVOC: Ex Vivo Organotypic Culture LTSA: Live Tissue Sensitivity Assay

150 PDAC PDXs GRANTS CCCT IACS SUSTAIN: Sponsored Research Agreements (Industry Testing Using PDX) Funding $ START: Patient Based Philanthropy (from direct patient care experiences)

Skip Viragh Family Foundation

EVOC: Ex Vivo Organotypic Culture LTSA: Live Tissue Sensitivity Assay LTSA/EVOC Preclinical Drug testing Developing Novel Therapy PDX/Patient Tumors Optimizing SOC Therapy

AZD2281 BEZ235 Everolimu s Ganetespib Palbocicli b Panobinos tat Sorafenib Sunitinib Trametini b PARP AZD2281 1 2 3 4 5 6 7 8 9 PI3K BEZ235 10 11 12 13 14 15 16 17 18 MTOR Everolimus 19 20 21 22 23 24 25 26 27 HSP90 Ganetespib 28 29 30 31 32 33 34 35 36 CDKS Palbociclib 37 38 39 40 41 42 43 44 45 HDAC Panobinostat 46 47 48 49 50 51 52 53 54 RAF/PDGF Sorafenib 55 56 57 58 59 60 61 62 63 TKI Sunitinib 64 65 66 67 68 69 70 71 72 MEK Trametinib 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 36 combinations in total arrayed in 96 well plate 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96

Application-3: Developing novel drug combinations

Highlighted numbers are single agent, red numbers are DMSO controls, green numbers are positive controls, 96 is blank.

Panel of FDA Approved Agents/Combinations Tested Across PDX

AZD2281 AZD2281-BE AZD2281-Ev AZD2281-Ga AZD2281-Palboc AZD2281-Panob AZD2281-Sor AZD2281-Su AZD2281-Tr BEZ235 BEZ235-Eve BEZ235-Ganetesp BEZ235-Palb BEZ235-Panobinosta BEZ235-Sora BEZ235-Suni BEZ235-Trametinib Everolimus Everolimus-Ganete Everolimus-Palb Everolimus-Panob Everolimus- Everolimus- Everolimus-Tr Ganetespib Ganetespib Ganetespib Ganetespib-Sora Ganetespib Ganetespib-Trame Palbociclib Palbociclib-Panobinos Palbociclib-Sorafe Palbociclib- Palbociclib-Trame Panobinostat Panobinostat- Panobinostat-Sun Panobinostat-Tram Sorafenib Sorafenib-Sunitini Sorafenib-Tra Sunitinib Sunitinib-Trame Trametinib PATX179 1.277123831 1.097522 0.857651 1.076887 0.323698033 1.141358126 0.731599 0.99721 0.612399 0.84898703 1.217026 1.111045326 1.074103 0.597673928 1.560291 1.3126187 0.878490867 0.403375384 0.706468216 1.511302085 0.872364996 0.972278 1.348565 0.79395385 0.426964 1.066857 0.726698 1.399002274 0.786192 0.850347035 0.62054189 0.756917251 1.182298872 1.116861 0.594889441 1.216615471 1.17424438 0.469212371 0.218948578 1.238469294 1.435147842 1.63354398 0.947774756 0.630888456 0.24541 PATX79 1.340372735 0.880797 0.66911 0.531631 0.760816516 0.50377883 0.961707 0.850329 1.044404 0.74067648 1.092329 0.549546952 0.804989 0.278912685 0.775571 0.6330004 0.350875206 0.676307661 0.384997941 0.483973435 0.514162891 0.749598 0.785961 0.42316722 0.592411 0.895222 0.534869 0.297636944 0.747606 0.370073105 0.965939044 0.24101112 0.675911244 0.426426 0.28886944 0.311892504 0.27035626 0.22720346 0.162525741 1.291541392 1.163514209 0.3833299 0.827481466 0.379159802 0.465862 PATX155 0.816903609 0.735477 0.793127 0.632242 0.911074592 0.859811568 0.793485 0.692821 0.981307 0.87027868 0.810886 0.94533907 0.797069 0.343140495 0.75694 0.6278166 0.503761432 0.947517215 0.783092786 0.70500703 0.750754089 1.09144 0.644124 0.75407989 0.696228 0.659284 0.805271 0.973471693 0.430122 0.901917369 0.992927788 0.716552702 0.75358718 0.749237 0.824053934 0.784682081 0.87973634 0.906766371 0.873904919 1.074288272 0.875848726 0.94228967 0.81285977 0.580807686 0.923251 PATX66 1.162642848 0.802585 0.816206 0.767339 0.888805691 0.390729351 0.925168 0.857132 0.704669 0.94376995 0.986205 0.471187891 0.861293 0.396559163 0.887287 1.0373607 0.556356635 0.926822485 0.724451338 1.008903389 0.569000374 1.137505 0.999412 0.69114475 0.492525 0.829502 0.913583 0.908826728 0.701489 0.842577241 1.019914068 0.705603394 0.986611043 1.042513 0.869369418 0.619172416 0.64549036 0.599882334 0.599176338 1.118824767 0.735622471 0.60917081 0.902230304 0.541587777 0.818563 PATX147 0.995793392 0.944742 0.831357 0.96386 0.862992046 0.684293144 1.267569 1.047777 1.148382 0.9181842 1.077952 0.77262099 0.9166 0.406361994 1.06582 1.0014495 0.437571021 1.052052679 0.810830104 0.939996649 0.706804872 1.241674 1.073133 0.89187742 0.696367 0.431532 0.6517 1.059151103 0.82789 0.335315842 0.832041607 0.088473529 1.126638938 0.789895 0.746281431 0.368473384 0.46001355 0.594544885 0.519674543 0.9762791 1.011636462 0.60794047 0.760802424 0.485723027 1.066202 PATX148 0.953252301 0.898149 0.898706 1.09981 1.117178471 0.55505044 1.163989 1.047252 0.900033 0.64257353 0.960216 0.714655518 0.791965 0.683330121 0.927776 0.9840606 0.691025492 0.846739238 0.768073209 1.031226686 0.383113295 0.866804 0.808917 0.8435832 0.799617 0.987549 0.990638 0.810243304 0.772582 0.893723975 1.02384692 0.837858672 1.199057619 0.899714 0.779011917 0.669134647 0.81257676 0.711261094 0.524178304 1.243339576 1.146556523 0.84776999 1.023222426 0.935403876 0.938832 PATX102 0.668213851 0.778479 0.773408 0.525746 1.033372956 1.363253427 0.599699 1.078699 1.146341 0.7015302 0.642138 0.57388749 0.908546 0.717426579 1.142115 1.1299258 0.886107416 0.441428693 0.580019623 0.888885744 0.555505241 0.741837 0.564784 0.73381353 0.427202 0.650049 0.66428 0.93661261 0.698455 0.530297456 0.76736573 0.996202794 1.289087633 1.15245 1.000410381 0.513382202 1.02438702 1.008514231 1.104472684 0.468490269 0.967424221 0.74427589 0.548679705 0.815064293 0.720243 patx122 1.24425532 0.717826 1.209781 1.044925 1.034780929 0.804230876 0.988239 0.936808 0.893869 0.63244029 0.788686 0.608698722 0.834221 0.906590263 0.761718 0.668643 0.82120421 1.001548416 0.849681106 0.850914645 1.036302564 0.753976 1.019946 0.62799711 0.768199 0.64958 0.879639 0.99844428 1.135692 0.834696042 1.165105942 1.147617282 0.895488412 0.923052 1.005253091 1.104910053 0.87228652 1.172121695 0.699436873 1.153772803 0.998854107 0.78586575 0.785768369 0.696860238 1.166494

PDX PATC

Ide Identification of

• f Activ

ctive Lea Lead Com Combin inations

Agents/combo Targets BEZ235-Panobinostat PI3K/HDAC Panobinostat-Trametinib HDAC/MEK Sunitinib-Trametinib TKI/MEK Palbociclib-Trametinib CDK4/MEK Panobinostat HDAC Genetispib HSP90

1 2 3 4 5 6 7 8 9 10 11 12 A DMSO A1 A3 A10 B1 B3 B10 AB1 AB3 AB10 AD1 AD10 Drug A: B DMSO A1 A3 A10 B1 B3 B10 AB1 AB3 AB10 AD1 AD10 Drug B C DMSO A1 A3 A10 B1 B3 B10 AB1 AB3 AB10 AD1 AD10 Drug C D DMSO A1 A3 A10 B1 B3 B10 AB1 AB3 AB10 AD1 AD10 Drug D E DMSO C1 C3 C10 AC1 AC3 AC10 D1 D3 D10 DMSO AD3 F DMSO C1 C3 C10 AC1 AC3 AC10 D1 D3 D10 DMSO AD3 3 Combination G DMSO C1 C3 C10 AC1 AC3 AC10 D1 D3 D10 DMSO AD3 H DMSO C1 C3 C10 AC1 AC3 AC10 D1 D3 D10 DMSO BLANK

Dose finding studies against panel of PDX

0.2 0.4 0.6 0.8 1 1.2 0.3 1 3

Relative Viability (uM)

PATX102

G1T38 Trametinib G1T38/Trametinib 0.2 0.4 0.6 0.8 1 1.2 0.3 1 3

Relative Viability (uM)

PATX102

G1T38 Pictilisib G1T38/Pictilisib

Evaluate for Synergy

Efforts to Identify Novel Combinations

PDX Sunitinib- Trametinib Synergy PATX147 0.12872865 YES PATX155 0.18308739 YES PATX137 0.22761044 YES PATX106 0.35197237 YES PATX66 0.39579509 YES PATX122 0.54923073 YES PATX79 0.86034729 YES PATX148 0.94693182 YES PATX102 1.56068081 NO PATX179 2.04434139 NO Palbociclib- Trametinib Synergy PATX122 0.24747413 YES PATX148 0.378103527 YES PATX155 0.47331694 YES PATX106 0.48910406 YES PATX79 0.773424613 YES PATX137 0.82123102 YES PATX147 0.879089621 YES PATX66 1.401413016 NO PATX179 2.092546838 NO PATX102 10.33692038 NO PDX Panobinostat

• Trametinib Synergy

PATX137 0.723647 YES PATX148 0.781375 YES PATX122 0.84487 YES PATX179 0.898059 YES PATX79 1.02057 NO PATX106 1.043018 NO PATX66 1.230388 NO PATX147 1.264003 NO PATX155 2.185187 NO PATX102 7.995594 NO BEZ235- Panobinostat Synergy PATX106 0.378565066 YES PATX179 0.410411599 YES PATX155 0.482763681 YES PATX66 0.68878945 YES PATX137 0.989547347 YES PATX79 1.066429185 NO PATX147 1.114607814 NO PATX148 1.800082123 NO PATX102 2.113408852 NO PATX122 4.645222963 NO

Sensitive PDX to Lead Compounds by Combination Index

Ex-vivo Drug Optimization Workflow

Final optimized drug and dose tested in-vivo

Xenograft harvested from mouse and tested against 30-45 different drug combos

Algorithm suggests new drug combos Initial ex-vivo testing Drug

• ptimization

round 1 Drug

• ptimization

round 2 Drug

• ptimization

round 3 Algorithm suggests new drug combos Algorithm suggests new drug combos

Xenograft harvested from mouse and tested against 30-45 different drug combos Xenograft harvested from mouse and tested against 30-45 different drug combos Xenograft harvested from mouse and tested against 30-45 different drug combos 4 days later 4 days later 4 days later

Geoffrey, 2017

Patrycja Nowak-Sliwinska, et al, Nature protocol, 2016

Clinical Therapy

• Gem/Abraxane
• FOLFIRINOX

Tumorgraft (~60 days) Progression Identify Unique Combination

Co-Clinical Approach: Individual Patient

Testing (~15-30) days)

MDA-PATX121

Tissue Evaluation (3-5 days)

MEK + CDK4/6 inhibitors Respond (25%) Establish PDX Classify PDX By Response Resistant (75%) AIM 1 Genomic/RPPA Analysis Biostatistical Signal Pathway Evaluation By Response AIM 3 Rational Combinations + + + + + + + + + + + + + +

• Biomarker Discovery

Novel Therapeutics AIM 2 CRC (52% activating KRAS/NRAS mts) PDAC (90% activating KRAS mts) x20 x48

Genomic/RPPA Analysis Drug Sensitivity Testing

+

PRELIM DATA

Co-Clinical Approach: Discovery

Goal: Cross-cutting Searchable Data Platforms

Imaging/Biophysical Molecular

Tumor Sensitivity

Clinicopathologic

Questions on Whipple Number 450….

What’s next?

Gray hair underneath surgical cap.

PDAC specimens PDX

Organoids

Banking Sequencing Sequencing Drug testing Panel-20, panel-40 New therapies LTSA Biomarkers New therapies Biomarkers Drug testing Panel-20, panel-40

Living Organoid Biobank of PDAC

Cell 160, 324–338, January 15, 2015

PDAC PDX-Organoid Program at MDACC

PATX136-F3 PATX136- F3-Day 3 PATX136- F3-Day 6 PATX136- F3-Day 10

Conclusion: Aim high, Embrace the struggle, and Cherish the friends.

Acknowledgements

• Jason. Fleming@moffitt.org

713-855-8551