1 of 6

Harnessing the Power of AI to Generate Real-World Evidence: A Biopharma Perspective

Christopher Boone, PhD

Global Real-World Evidence Center of Excellence

April 2019

Pfizer Confidential

1

Pfizer Confidential

“How to establish an effective RWE strategy across your product lifecycle”

Involve R&D colleagues from Phase II onwards to incorporate endpoints that align with RWE goals
Choose the appropriate data sources and methodological approaches to structure and analyze data to produce viable action plans
Establish a robust partnership ecosystem that enables the effective deployment of the RWE strategy

Items

�Application of RWE in R&D (Designing CT using RWD)

Hybrid&Virtual Clinical Trials-Patient Recruitment, Engagement, Retention and Close-outs

Real World Analytics (Descriptive (Data Scientists&Statisticians)&Predictive/Prescriptive Analytics)Real World Data as a Synthetic arm/Placebo to reduce overall burden of Clinical TrialsOverall Intelligent content authoring from Preclinical stages to NDA filing (Phase-3)

2 of 6

Disclaimer

The speaker is an employee of Pfizer Inc. Views expressed are the speaker’s own and do not necessarily represent those of Pfizer.
These slides are not intended for wider distribution outside the intended purpose without speaker’s approval.

2

Pfizer Confidential

3 of 6

Artificial intelligence is recognized as a powerful tool for enhancing research and development of novel therapeutics

Drug Discovery

Finding biomarkers within patient populations for drug targeting

Drug Development

Improving the probability of success in clinical trials by identifying subpopulations of patients who are responsive to specific drugs

Comparative Effectiveness

Leveraging real-world data (RWD) to generate the evidence needed to prove the value of drugs and interventions

Personalized Medicine

Identifying and optimizing the most effective care pathways at the individual level

3

Pfizer Confidential

Artificial intelligence is recognized as a powerful tool for enhancing research and development of novel therapeutics.

Biopharma companies are using real world data combined with AI to to accelerate drug development and deliver more personalized and predictive medicine by determining genomic profiles, recognizing a need for immediate clinical intervention, and monitoring medication adherence.

A key challenge will be ensuring that AI is developed and used in a way that is transparent and compatible with the public interest, whilst stimulating and driving innovation in the sector.

Drug discovery: Finding biomarkers within patient populations for drug targeting
Drug development: Improving the probability of success in clinical trials by identifying subpopulations of patients who are responsive to specific drugs
Comparative effectiveness: Leveraging real-world data (RWD) to generate the evidence needed to prove the value of drugs and interventions
Personalized medicine: Identifying and optimizing the most effective care pathways at the individual level

4 of 6

Lesson 1: How far would you like to go in this step of the journey?

What questions do we want to answer?

How will it implement AI?

What data it needs to perform tasks?

How it will use the results from AI to drive quantitative decision-making?

4

Pfizer Confidential

5 of 6

Lesson 2: Big data strategy, management, and curation are the foundational elements of any effective AI strategy

Research

Development

Medical

HEOR / Access

Integrated RWD and analytics infrastructure
Partner with key stakeholders for one-to-one value exchange
Plug gaps with select RWD aggregators and analytics vendors
Maximize value of RWD assets
Knowledge management and enterprise portfolio view on all RWE-related activities

GRWE Platform

Data Partners

Hosted Environment

Analytics Environment

5

Pfizer Confidential

6 of 6

Lesson 3: Understand the key challenges that could impede progress

Data sharing matters / Advanced analytics requires access to varied data from numerous sources, both within and among organizations.
Data security matters / health care data has replaced financial data as most valuable.
Use case identification / Advice: Create 1-2 “anchor” customers to build momentum and address high value opportunities.

Overhyping / Overhyping the capabilities of A.I. only creates mass confusion
Talent / Be sure to hire more business integration roles – individuals that can bridge Analytics, IT, and business decision-makers.
Leadership / Sustained, engaged leadership is key to success.

6

Pfizer Confidential