Group 5 Biomarkers & Chemoprevention: Where are we? March 4 th - PowerPoint PPT Presentation

Group 5 Biomarkers & Chemoprevention: Where are we? March 4 th /5th 2016

Miriam Rosin PhD Director, BC Oral Cancer Prevention Program, BC Cancer Agency GOCF Organizing Committee and Co-Chair, Group 5 Contact: mrosin.1987@gmail.com Risk prediction for dysplasia progression in developed countries Sok Ching Cheong, PhD Group Leader, Oral Cancer Research Programme, Cancer Research Malaysia Adjunct Professor, University of Malaya, Malaysia. Contact: sokching.cheong@cancerresearch.my What can be done now in less developed countries Anil K Chaturvedi, PhD Investigator, Division of Cancer Epidemiology and Genetics National Cancer Institute Contact:chaturva@mail.nih.gov Natural history of OPMDs & biomarkers: The establishment of national and regional programs

J. Silvio Gutkind, PhD Associate Director of Basic Science, UC San Diego Moores Cancer Center GOCF Organizing Committee and Co-Chair, Group 5 Contact: sgutkind@ucsd.edu Oral Cancer Prevention in the Era of Precision Medicine Scott Lippman, MD Director, UC San Diego Moores Cancer Associate Vice Chancellor for Cancer Research and Care Contact: slippman@ucsd.edu Contributor to white paper Eva Szabo, MD Chief, Lung and Upper Aerodigestive Cancer Research Group Division of Cancer Prevention, National Cancer Institute Contact: szaboe@mail.nih.gov Discussant Paul Brennan Head of Genetic Section International Agency for Research on Cancer Contact: brennanp@iarc.fr Moderator

Risk prediction for dysplasia progression in developed countries Miriam P Rosin, PhD Director, BC Oral Cancer Prevention Program BC Cancer Agency

Evolution of Biological Framework Evolution of a comprehensive progression biomarker set • Increased capacity to share knowledge & to craft new technology • What information do we collect? • How do we collect this information? • How do we share in process of creating knowledge? • Focus of collection – depth & breadth

Pure Nature’s Photos. Facebook

The need to look at the broader picture Variation in natural history of disease in different settings Universal vs specific to setting If yes, then what? Pictures from a biomarker/chemoprevention trial in Usbekistan in 1984. Nass users given multi-vitamin treatment

Evolution of progression models “Battle of the Clones” OPMD Invasive carcinoma Superficial Intermediate •Self-sufficiency in growth signals •Insensitivity to antigrowth signals Parabasal •Evading apoptosis Basal •Limitless replicative potential Basement •Sustained angiogenesis membran •Tissue invasion and metastasis e Fibroblast Blood vessel Inactivation of 9p21 Apoptosis Progression (genetic / epigenetic Senescence mutations, viral oncogenes) Genomic instability Cell cycle deregulation Heterogeneity – Impact of technology / new knowledge how many ways to • Detailed understanding of genetic change in clones, single cells end-game • Processes outside the clones / lesion, changing over time • Developmental windows (sets of genes on / off, induced by exposure) • Stressors, especially systemic, physiologic or tissue level with age Protectors (micro-environment, normal cells), e..g immune system • • Cross over studies, different cancers, chronic diseases: markers, processes, drugs Rosin et al., 2000; Kensler 2016

BC model of integration of dental & medical systems Readiness of Dental Clinics Back to community •Fluorescence Visualization Dental •Demographics • Guidelines Health •Risk Factors • Protocols Professions •Clinical • Referral pathway • Capacity building • Switching norms • Biopsy g • Social marketing Specialists • Histology Surveillance System Oral Biopsy • Cytology Service • QP • Register patients • Monitor patient flow Next •Assessment • Outcome Generation Clinic Oral Cancer Prediction Longitudinal Study BC Cancer Agency (Primary dysplasia 10-15 yrs) Today focus on mild and moderate dysplasia • Treatment

How to help patients now? Concept of Triage INFA-Arf region on 9p 21 1 β 1 α 3 2 2 1 Human 9p21 D9S171 D9S1751 D9S1748 21,550K 21,950K 22,000K 24,500K 24,550K ARF p16 INK4a p15 INK4b p16/CDKN2A CDK4 p14/ARF p21 MDM2 p53 Cyclin D p15/CDKN2B P RB RB Cell Cycle E2F Apoptosis E2F Progression Senescence Test: 9p21 a “gatekeeper” – 1 st separation • • Combined with change to markers on 2 other critical sites on 4q and 17p => Separate out high-risk (HR) cases - Genome marker based technology (gMART) 10

N = 296 Genome marker based technology (gMART) By Risk Pattern (gMART) By Histology Overall Progression Hyperplasia 9p Retention Mild dysplasia Mild Dysplasia RR 1 RR 0.86 (0.33- 2.22)* Moderate dysplasia Moderate dysplasia 9p & either 4q/17p LOH RR 0.96 (0.39-2.46)** RR 11.6 (2.7-49.9)* 14% progressed Histology did not predict (Median = 44 months) 9p, 4q & 17p LOH RR 52.1 (11.8-230.6)** outcome *p=0.75 *p=0.001 **p=0.98 **p<0.0001 5-year lesion progression risks for gMART categories • Low risk (LR) = 3% (47% of cases) TRIAGE • Medium risk (MR) = 16% (44% of cases) • High risk (HR) = 63% (10% of cases)

Development-Validation Process Prospective Cohort Retrospective Cohort Progression 100 100 HR 2-year HR 3% Progression Rate (%) 10% 80 80 LR IR 3-year LR 47% 38% IR 5-year 59% 43% 60 60 40 40 20 20 0 0 LR MR HR LR MR HR Similar results for independent retrospective cohort • • gMART alone – identifies 54% of cases as low or high risk • What about medium risk (MR) group? Further sorting

Integration of technologies: QP - LOH • High throughput quantifiable indication of risk of progression • Automated -objectively measures subtle changes to tissue phenotype (amount and distribution of DNA in nuclei) Likely to be globally relevant marker • Quantitative pathology alone 1.0 0.9 0.8 Low Risk, RR = 1 0.7 0.6 0.5 High Risk, RR = 10 0.4 Cumulative Proportion Sur 0.3 0.2 0.1 0.0 0 20 40 60 80 100 120 140 160 180 200 220 What happens when done together with gMART?? => Nuclear Phenotype Score (NPS)

gMART alone gMART+ QP Performance of gMART+ • Addition of QP places up to 80% of patients with oral dysplasia into the HR and LR categories • Independent validation study – 12/15 progressors detected as high- risk (80% sensitivity) • Other fusions: QP-LOH-TB, LOH-FISH (EPOC study) etc

Clinical application of tool: “Next Gen” Clinic • Goal: Move to community settings to triage OPLs • Treatment of HR cases (metformin) • Watchful waiting LR • Further genomic profiling for MR Look at specific settings • – South Asian community • High-risk communities

Importance of depth & breadth in focus – Data collection for future research ??? What is stopping clones with MR or HR genetic patterns from progressing ???

Group 5 What can be done now in less developed countries ? March 4 th /5th 2016 Sok Ching Cheong Cancer Research Malaysia University of Malaya

Oral cancer A significant burden in less developed countries and Asia ( all regions of Africa, Asia (excluding Japan), Latin America and the Caribbean, Melanesia, Micronesia and Polynesia) DISTRIBUTION OF LIP/ORAL CANCERS IN DISTRIBUTION OF LIP/ORAL CANCERS IN DIFFERENT GEOGRAPHICAL REGIONS MORE DEVELOPED AND LESS DEVELOPED Northern GLOBAL REGIONS Africa America Latin 6% 10% More America & develope Caribbean Oceania d 7% 1% 34% Europe 20% Less develope d Asia 66% 56% Source: Globocan 2012 (Total cases: 300,373)

Chemoprevention and biomarker development Opportunities in the less developed world Biomarker Randomized validation clinical trials Large number of High statistical power specimens Large number of patients Understanding Validation of genetic genetic progression drivers Genetic heterogeneity Robust identification of represented in models genetic drivers

Genetic alterations in the different stages of oral cancer development Dionne et al. IJC 2014

Weaknesses and potential limitations • Endpoint of dysplasia weak: oral submucous fibrosis (OSF) • Endpoint: status of malignant transformation • Large number of association studies – causative role of biomarkers unclear • The tumor as an organ – understanding progression includes an understanding of the tumor microenvironment and systemic changes (important for OSF in particular) • Wide implementation of biomarkers in low resource setting challenging – expertise, infrastructure, who pays?