Desmond Pink, Ph.D.Chief Science Officer

Nanostics Inc

Studies in metastases:Or how chickens help fight

prostate cancer.

Borrowed from Gary Larson

2

Disclosures

Dr. Desmond Pink Ph.D.

Disclosure:

Financial Co-founder and employee of Nanostics Inc.

Out of every 100 men...

16 will be diagnosedwith prostate cancer in

their lifetime

In reality, up to 80 will haveprostate cancer by age 70

And 3 will die from it.

But which 3 ?

36% of newly diagnosed cancers, and 10% of all cancer deaths in men

The deadliest aspect of prostate cancer is its spread, or metastasis

In North America, the average 5 year survival rate for localized prostate cancer is 100%For metastatic cancer, it is less than 30%

Current diagnosis tools do not predict whether metastasis will occur

Current treatments do not prevent or cure metastasis

Lymph nodes

Bones

Bones

Prostate

Palmer et al., Advanced Drug Delivery Reviews 2011

Animal models for metastasis

Snapshots provide limited information…

1. Good Samaritan helps fallen woman during riot

2. Opportunistic rioter steals a kiss from an injured woman

3. Riot police defend couple's right to Public Display of Affection (PDA)

4. “At least someone from Vancouver can score on the road”

1. Man consoles distraught girlfriend after she was violently knocked down by riot police

60-8

0µm Endoderm

StromaCapillary bed

EctodermEggshell/membrane

chorioallantoic membrane (CAM)

Modeling cancer dynamics in chicken embryos

Immediately after injection 24 hours later

Modeling cancer dynamics in ex ovo avian embryos

Intravital imaging of tumour growth and metastasis

4 mm tumour growing over 4 days

Small metastasis of 200-300 cells

Individual cancer cells

When studying metastasis context is crucial to data analysis

When studying metastasis context is crucial to data analysis

When studying metastasis context is crucial to data analysis

3D reconstruction of single cancer cells (green) and their residence near a blood vessel. This type of image allows us to visualize cells moving toward a blood vessel in great detail.

Microvasculature

Blood vesselTumor cells

Imaging the CAM in 3D60

-80µ

m Endoderm

Stroma

Capillary bed

Ectoderm

Eggshell/membrane

chorioallantoic membrane (CAM)

Leong et al., Cell Reports, 2014

TOP BOTTOM

Rhodamine-Lectin (Lens culinaris agglutinin) 2MDa FITC-Dextran

Advanced Drug Delivery Reviews 2011

M- cell line

M+++ cell line

Antibody 1A5 targets tetraspanin CD151

Cyclophosphamide – “tolerizes” mouse immune system

Immunize mouse with low metastatic cancer variantM-

Immunize mouse with high metastatic cancer variantM+++

Isolate antibodies against targets in M+++ but not M-

Anti-CD151 antibody blocks spontaneous metastasis

Zijlstra et al., Cancer Cell, 2008

Avian Embryo

Con

trol

Anti-

CD

151

Dramatic differences in cell motility phenotype in vivo

Merged Topical View

Merged Topical View

Z-projection

Z-projection

Cont

rol (

IgG)

Anti-

CD15

1 (1

A5)Anti-CD151 antibody inhibits cell migration in vivo

Zijlstra et al., Cancer Cell, 2008

control mAb 1A5 treated

Average:(velocity: 24.6 µm/hr)(total distance: 271 µm)(productive distance: 101µm)

Average:(velocity: 5.9 µm/hr)(total distance: 85 µm)(productive distance: 23µm)

100µm

100µm

Productive Migration6.6 fold inhibition

1A5 antibody binds CD151 that is “free” from integrins

1A5TS151r

8C314A2.H1LIA1/1VJ1/16

11G5a11B1TS151

If CD151free marks cancer cells that have undergone a cell motility switch, perhaps we can use it as a test to detect or predict metastasis…

Palmer et al., Cancer Research, 2014

mAb 1A5

Tum

our

Beni

gn

adja

cent

Beni

gn

away

mAb 11G5a

CD151free is distinct from CD151all in prostate cancer

Palmer et al., Cancer Research, 2014

138 prostate cancer surgery patients

Follow up: 12.1 years

Recurrence: 34 casesMetastasis: 38 cases

1. Does CD151free predict recurrence after surgery?

2. Does CD151free predict metastasis?

CD151free predicts prostate cancer recurrence and metastasis

Palmer et al., Cancer Research 2014

• Tetraspanin CD151 and α3 integrin interactions comprise a cell migration “switch” between maintenance of epithelial structure and invasive cell migration

• Cross-linking CD151free with 1A5 antibody blocks metastasis by stabilizing cell-cell adhesion

• mAb 1A5 detects a pool of CD151 (CD151free) that is distinct from that detected by other antibodies

• CD151free is associated with earlier biochemical recurrence and earlier onset of metastasis, independent predictor of outcome

Summary: Tetraspanin CD151free

Identified 27 clones with significant reduction in in vivo motility

Identified 11 single shRNAs required for in vivo cell motility

Prioritized 50 isolated clones by “compactness index” or CI

Stoletov et al., Nature Communications, 2018

Displacement (µm-2/sec)In vivo cell migration assays (spontaneous and experimental metastasis)

Stoletov et al., Nature Communications, 2018

Stoletov et al., Nature Communications, 2018

Expression in matched primary tumour/metastasis Staining of 98 patient prostate cancer TMA with

SRPK1 and Kif3b

If “cell motility switch” genes arerequired for prostate cancer spread,perhaps we can incorporate them into atest to detect aggressive prostatecancer

Screening for prostate cancercauses unnecessary harm

Screening• Symptoms/risk factors

(Family doctor)• PSA Blood test (20M per year)• DRE

InvasiveUncomfortable

Only 15-25% specific for prostate cancer,resulting in many unnecessary biopsies

DiagnosisBiopsy (1.3M per year)12 needles = Pain, Discomfort, Infection1.5% chance of life-threatening sepsis

More than 3/4 of men diagnosedwith prostate cancer have indolent,non-aggressive disease

32

33

1. Sanders A1, Buchan N. ANZ J Surg. 2013, 2. Loeb S. J Urol. 2011, 3. Robert K.Nam. J Urol. 2013, 4. Mohammed Shahait. Int Braz J Urol. 2016

Serious Adverse Events (SAE) from biopsies

• >1M biopsies are done per year in the US1,2

• Incidence of sepsis following transrectal ultrasound guided prostate biopsy ranges from 2-4% in developed countries and can go a high as >9% in developing countries3,4

• Antibiotic resistance and sepsis are on the rise1

Careful patient selection for prostate biopsy is essential to minimize the potential harms

https://www.ncbi.nlm.nih.gov/pubmed/?term=Sanders%20A%5bAuthor%5d&cauthor=true&cauthor_uid=23346881https://www.ncbi.nlm.nih.gov/pubmed/?term=Buchan%20N%5bAuthor%5d&cauthor=true&cauthor_uid=23346881https://www.ncbi.nlm.nih.gov/pubmed/23346881https://www.ncbi.nlm.nih.gov/pubmed/?term=Loeb%20S%5bAuthor%5d&cauthor=true&cauthor_uid=21944136https://www.ncbi.nlm.nih.gov/pubmed/21944136?dopt=Abstract&holding=npghttps://www.sciencedirect.com/science/article/pii/S0022534712054808?via=ihub#!https://www.ncbi.nlm.nih.gov/pubmed/?term=Shahait%20M%5bAuthor%5d&cauthor=true&cauthor_uid=27136468https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4811227/

12

34

5

Epstein et al. 2015, European Urology, 69, 428-435

Men with Gleason Grade Group 3-5 prostate cancers have significantly worse outcomes

34Prostate cancer is a heterogeneous disease: some forms are lethal, others are not

• Living cancer cells detached from the primary tumor and circulating through bloodstream

• Very rare (1 in a billion!)

• CTC counting has prognostic value for OS (>4 CTCs/7.5mL)

Circulating tumour cells (CTC) as biomarkers 35

Extracellular vesicles (EVs) provide much greater dynamic range than CTCs43% have

• Exomeres (< 50nm)• Small Exosomes (60-80nm)• Large Exosomes (80-120nm)• Microvesicles (50nm-1µm)• Oncosomes (1-10µm)• Apoptotic bodies (800nm-10µm)• As yet undescribed?

Extracellular vesicles (EVs) are released by all cells in the body 37

38

10’s to 100’s of receptorsHow small can we go?

The challenge: single particle detection of prostate EVs

Leong et al., J of Thrombosis and Haemostasis, 2011

39MicroFlow Cytometry can detect and characterize a wide range of EVs

MicroFlow Cytometry resolves small biological particles

MLV Probiotics

Human Plasma

FASTED

40

Detection of prostate-derived EVs in complex biofluidsLocalized PCa

Metastatic PCa

PSMA-405

PSMA-405

CD15

1+CD

151+

PSM

A-40

5PS

MA-

405

66 patient cohort University Health Network - Toronto

Age, PSA matched

Biofluid: plasma

CD151+ PSMA+

CD151+ PSMA+

CD15

1+ P

SMA+

41

Microflow cytometry of plasma EVsis highly sensitive and reproducible

Reliable detection of 6 positive EVs(0.0003%) against a highly enriched blood EVbackground (2.5M)

Excellent %CV for clinical testing at a widerange of biomarker concentrations

0 1 2 2 4 3 6 4 8 6 0 7 2 8 4 9 60

1 0 0

2 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0

3 0 0 0

R e p lic a te W e ll

Bio

ma

rke

r C

on

ce

ntr

ati

on

% C V = 4 .2 1 % P a lm -G F P c o n d it io n e d m e d ia

% C V = 8 .7 1 % P S M A -A F 6 4 7 in p la s m a

% C V = 2 .2 1 % P a lm -G F P c o n d it io n e d m e d ia

% C V = 3 .5 6 % C D 6 3 F IT C in p la s m a

% C V = 4 .5 1 % C D 9 -P E in p la s m a

% C V = 2 4 .9 8 % P A L M -G F P c o n d it io n e d m e d ia

% C V = 4 .0 0 % P S M A -A F 6 4 7 in p la s m a

% C V = 1 9 .7 8 % P A L M -G F P s p ik e d in to p la s m a

42

Sample stability of EVs in human plasma and serumSample stability matrix

43

Sample stability quantification

Biomarker 3

No PCa PCa

0

100

200

300

400

500

6003000

3500

4000

Pos

itive

Eve

nts/

uL x

102

PSMA

N o P C P C A

0

5000

10000

15000

2000020000

30000

40000

PSM

A P

ositi

ve x

10

2 E

vent

s/uL

Ghrelin

No PCa PCa

0

2000

4000

6000

8000

Ghr

elin

Pos

itive

x 1

02

Even

ts/u

L

P=0.5337 P=0.0054 P

3D plot of ROC area under the curve

45Machine learning approach to generate classifiers from multi-dimensional microflow cytometry data

XGBoost provides highest AUCs for predicting clinically significant prostate cancer

46

Ensembled: Results of multiple decisions trees averaged into 1 result

Boosted: Each additional decision tree is design to correct misclassified observations

XGBoost is an ensembled, boosted, decision tree-based model.

Nanostics’ platform technology generatesEV fingerprints to predict disease with a liquid biopsy

47

48Generating the ClarityDX Prostate Risk Score

Prospective pre-diagnosis cohortin Alberta, Canada

49

Validation of ClarityDX Prostate in a 377 patient prospective cohort

Average AUC = 0.83

39% higher specificity for clinically significant prostate

cancer than PSA alone

50

Patient at risk 40-75 yrs PSA

PSA3 & ≥10 ng/ml

Repeat PSA at 1-4y intervals

Proposed prostate cancer diagnosis model

TRUSGuided Biopsy

Low Risk

Continue Monitoring

High Risk

51

Pivotal Clinical Validation Study

Participants enrolled

SST SerumAmbient

Annual ~ #s

NAUC: 500PCC: 500YUK: 100US: 250

Interim analysis in 6m ~700 pts

ShippingNAUC: DynaLIFE

PCC: Overnight Express door to door service

YUK & US: TBD

Processing

Within 48h

Flow Score

Quality & Regulatory Framework

FDA CFR 820.21 HC SOR/98-282 CE Mark 98/79/EC Regulations

ISO 13485:2016 Standard / framework to address regulatory requirements

52

AndriesZijlstra

KonstantinStoletov

Acknowledgements

Slide Number 1Slide Number 2Slide Number 3Slide Number 4Slide Number 5Slide Number 6Slide Number 7Slide Number 8Slide Number 9Slide Number 10Slide Number 11Slide Number 12Slide Number 13Slide Number 14Slide Number 15Slide Number 16Slide Number 17Slide Number 18Slide Number 19Slide Number 20Slide Number 21Slide Number 22Slide Number 23Slide Number 24Slide Number 25Slide Number 26Slide Number 27Slide Number 28Slide Number 29Slide Number 30Slide Number 31Slide Number 32Slide Number 33Slide Number 34Slide Number 35Slide Number 36Slide Number 37Slide Number 38Slide Number 39MicroFlow Cytometry resolves small biological particlesSlide Number 41Slide Number 42Slide Number 43Slide Number 44Slide Number 45Slide Number 46Slide Number 47Generating the ClarityDX Prostate Risk ScoreSlide Number 49Slide Number 50Slide Number 51Slide Number 52Slide Number 53