Opt-in, Opt-out, & Patient Led Databases

1 Duane Schulthess Managing Director Copyright Vital Transformation - CONFIDENTIAL Opt-in, Opt-out, & Patient Led Databases Better Patient Outcomes, Faster and Cheaper Researchers are: Dr. Daniel Gassull Duane Schulthess 2 Copyright Vital Transformation - 2014 12/11/2014 3 Copyright Vital Transformation - 2014 Sponsors and Supporters of the project: Assumptions & Limitations: The theoretical time impact of Databases on identification and sign-in has been discussed and presented to numerous clinical trial managers and researchers; these times are theoretical as well as untested, and should be seen as goals or targets We assume the use of Patient Databases in Europe is unaffected by the data protection legislation passed in the European Parliament in March of 2014 The ROI & Cost assumptions are based on averages across therapeutic areas.

2 Our model is illustrative of relative, not specific, value We assume there are many searchable health Databases operational in numerous international locations, this may not be the case 4 Copyright Vital Transformation - 2014 The Problem 5 6 The Problem Rate of success from Phase I 7 Negotiation for Reimbursement 28 member States EMA Filing Pre-Clinical Research Closed & Open Innovation Clinical Trials EMA Approval for Sale HTA Approval Phase 1 Phase 2 Phase 3 5,000 10,000 Compounds 250 Compounds 3 6 Years 6 7 Years 5 Therapies 1 Therapy 2 5 Years Number of patients /Subjects 20-100 100-500 1000-5000 Regulatory Review Drug Discovery Pre Clinical Testing PhV Monitoring Total Cost: ~$2 Billion USD Sources: Drug Discovery and Development: Understanding the R&D Process, ; CBO, Research and Development in the Pharmaceutical Industry, 2006; Forbes, Matthew Herper, The Truly Staggering Cost Of Inventing New Drugs , February 10, 2012 Current EU trial pathways are expensive and slow New EU therapies don t reach patients until here Real World Evidence - Health Databases 8 Time of recruitment represents about 30% of total CT Time - roughly 30 months of development Months Start-up Recruitment Time Trial Time Data Clean-up & Report Phase 3 Phase 2 Phase 1 30% of Clinical trial 30 months in total 9 Copyright Vital Transformation - 2014 Source: CMR International.

3 Pharmaceutical R&D Factbook 2011, Thomsom Reuters, May 2011 Current development times by clinical area (1) Target indications of the project FDIH = First-dose in Humans, here considered before Phase 1 Source: CMR International. Pharmaceutical R&D Factbook 2011, Thomsom Reuters, May 2011 0 1 2 3 4 5 6 7 8 9 10 New Chemical entity New Biological Entity Toxicity to FDIH Phase 1 Phase 2 Phase 3 Submission Years 0 1 2 3 4 5 6 7 8 9 10 Cardiovascular Anti-Cancer Central Nervous System Toxicity to FDIH Phase 1 Phase 2 Phase 3 Submission Years 10 Copyright Vital Transformation - 2014 Three queries from were selected to test Databases Alzheimer s Disease Cardiovascular disease Inclusion criteria Male and Female > 18 years Diagnosis of NSCLC Stage IIIB not amenable to curative treatment or Stage IV.

4 No prior chemotherapy for lung cancer. patients must have at least one uni-dimensionally measurable lesion. Prior radiation therapy to less than 25% of bone marrow, whole pelvis not allowed. Radiation must be completed at least 4 weeks prior to study enrollment. Exclusion Criteria: Treatment with any drug within the last 30 days that has not received regulatory approval. Serious cardiac condition. Serious medical disorder in addition to NSCLC that would make it difficult for the Patient to complete the study. Inability or unwillingness to take folic acid or Vitamin B12 supplementation. Presence of fluid retention that cannot be controlled by drainage. Inclusion Criteria Male and Female >50 Probable or possible Alzheimer's disease Living in skilled nursing home Exclusion Criteria Other types of dementia, psychiatric or neurologic disorders Musculoskeletal disease Inclusion Criteria Male and Female >18 years Must be on background lipid lowering treatment.

5 Must be at high risk of a CV event. Must have an LDL C >/=70 mg/dL ( mmol/L) and < 100 mg/dL ( mmol/L) or OR, Must have non-HDL-C >/= 100 mg/dL ( mmol/L) and < 130 mg/dL ( mmol/L) Exclusion Criteria An LDL C < 70 mg/dL ( mmol/L) or >/= 100 mg/dL ( mmol/L) or non HDL-C < 100 mg/dL ( mmol/L) or >/=130 mg/dL ( mmol/L) Planned coronary (PCI or CABG) or other arterial revascularization. New York Heart Association Class IV congestive heart failure or left ventricular ejection fraction < 25% by cardiac imaging. Chronic renal insufficiency with creatinine clearance of <30 ml/ ^2 by MDRD formula or with end state renal disease on dialysis. History of hemorrhagic stroke. Non-Small Cell Lung Cancer We studied these queries in three data systems that can be classified into two groups Region # patients in database Healthcare setting Year created Databases with electronic Patient records* CPRD TrialViz UK M GPs 2012 Penn Medicine Oracle Health Science Network US M Hospitals 2013 Databases led by patients (*) patients -Know-Best UK 150k Hospitals 2012 * Data is Anonymised/De-identified Results CPRD Trial Viz Alzheimer disease.

6 Result of Query with CPRD Geographical Distribution of Hits Distribution of # of patients per practice # of patients per practice Inclusion Criteria Male and Female >50 Probable or possible Alzheimer's disease Living in skilled nursing home Exclusion Criteria Other types of dementia, psychiatric or neurologic disorders Musculoskeletal disease Number of patients identified: 1,353 Query results Anonymised Data Alzheimer s CPRD Penn Medicine patients identified in query 1,343 890 Acceptance Rate Based on Historical Averages 60% (1,2) 60% (1,2) Total patients Needed in in Clinical Total Potential Patient Trial Pool 670 445 249 (3) (1)Avalon L. Am J Geriatr Psychiatry. 2009 Jan;17(1):65-74 (2)Fouad MN.

7 Et al. J Oncol Pract. 2013 Mar;9(2):e40-7 (3)Number of patients needed in entire trial (world wide) (4)Number of trial sites used in the actual trial (*) Data shown in previous slide 11 (*) 1 53 (4) Total Trial Locations Registered in Clinical Total Implied Locations Required using revised Data Base Search Result of Query with CPRD: Cardiovascular Disease Geographical Distribution of Hits Distribution of # of patients per practice Inclusion Criteria Male and Female >18 years Must be on background lipid lowering treatment. Must be at high risk of a CV event. Must have an LDL C >/=70 mg/dL ( mmol/L) and < 100 mg/dL ( mmol/L) or OR, Must have non-HDL-C >/= 100 mg/dL ( mmol/L) and < 130 mg/dL ( mmol/L).

8 Exclusion Criteria An LDL C < 70 mg/dL ( mmol/L) or >/= 100 mg/dL ( mmol/L) or non HDL-C < 100 mg/dL ( mmol/L) or >/=130 mg/dL ( mmol/L) Planned coronary (PCI or CABG) or other arterial revascularization. New York Heart Association Class IV congestive heart failure or left ventricular ejection fraction < 25% by cardiac imaging. Chronic renal insufficiency with creatinine clearance of <30 ml/ ^2 by MDRD formula or with end state renal disease on dialysis. History of hemorrhagic stroke. Number of patients identified: 136,053 Copyright Vital Transformation - 2014 Query Results Anonymised Data Cardiovascular CPRD Penn Medicine patients identified in query 136,065 982 Total patients Needed in Clinical Acceptance Rate 10% (1) 10% (1) Total Patient Trial Pool 13,065 98 650 (2) (1)Harper D.

9 B. Projecting realistic enrolment rates . Monitor, Winter 2004, pages 15-18 (2)The clinical trial included 12,000 patients world-wide. Based on the number of hospitals in the UK and the rest of the world, 650 patients were allocated in the UK (3)The clinical trial required and registered 461 trial locations. Assuming the acceptance rate to be the same world-wide, 24 locations were allocated in UK Copyright Vital Transformation - 2014 7(*) 7 (**) 24 (3) Total Implied Locations Required using revised Data Base Search Total Trial Locations Registered in Clinical (*) Data shown in previous slide, (**) Number extrapolated, assuming all locations have ths same number of patients ( 98) Query Results Anonymised Data (1)Harper D.

10 B. Projecting realistic enrolment rates . Monitor, Winter 2004, pages 15-18 (2)The clinical trial included 12,000 patients world-wide. Based on the number of hospitals in the UK and the rest of the world, 650 patients were allocated in the UK (3)The clinical trial required and registered 461 trial locations. Assuming the acceptance rate to be the same world-wide, 24 locations were allocated in UK Copyright Vital Transformation - 2014 Why is there a difference of two orders of magnitude between CPRD and Penn Medicine? CPRD is GP data, Penn Medicine is hospital/outpatient/pharmacy data Very few CVD patients go to the hospital every two years (*) Data shown in previous slide CPRD and Penn Medicine differ in population base, offering a different number of patients and depth of information patients with Lipid Lowering Treatment patients with LDL > & < patients without Renal Failure OR stroke patients without CABG or PCI CPRD (5M patients ) Penn Medicine ( patients )