transmart community meeting 5-7 nov 13 - session 5: recent transmart lessons learned

A tranSMART journey back to the real world at Deloitte

November 2013

tranSMART

Copyright © 2012 Deloitte Consulting LLC. All rights reserved.2

Agenda Topics

• About Recombinant By Deloitte• Hot topics from Deloitte client community• Real World Evidence + In Memory Computing• I2b2/AMC back translation• ‘integrated tranSMART’ demo (1.2 components)

preview

AboutRecombinant By Deloitte

- 4 - PP

T T

empl

ate_

Rec

ombi

nant

by

Del

oitte

.pot

x

Audit & Enterprise

Risk Services

Financial Advisory Services

Services

Deloitte U.S. Firms

ConsultingTax

Technology

Service Area

Recombinant

By

Deloitte

Human Capital

Service Area

Strategy & Ops

Service AreaInnovation

Dedicated US-India (USI) Resources

Information Management & Life Sciences Health Care

Consulting

4

Recombinant + Deloitte - Organization Within Deloitte

- 5 - PP

T T

empl

ate_

Rec

ombi

nant

by

Del

oitte

.pot

x

Pharma Payors

Recombinant Vision For Capabilities for Translational Medicine

- 6 - PP

T T

empl

ate_

Rec

ombi

nant

by

Del

oitte

.pot

x

Life SciencesProvider / Research

ACO / Payer Target

Markets

Clinical Performance Improvement Clinical Quality Operational Excellence Accountable Care

Key Capabilities

Translational Research Cost Effectiveness Comparative Effectiveness Pharmacovigilance

Federal

Services

Data StrategyData Governance

Data Warehousing/Bioinformatics Implementations

Professional Open Source Support Contracts

Products / Tools

Data TrustSelectrus AnalyticsMiner Suite

Data Integration Hub Open source tools (I2B2,

SHRINE, tranSMART)

6

General Market Approach

- 7 - PP

T T

empl

ate_

Rec

ombi

nant

by

Del

oitte

.pot

x

Client and Partner Ecosystem

http://hss.edu/index.htm

http://www.transmartproject.org/index.html

http://www.roche.com/index.htm

http://www.sanofi.com/

Hot Topics


Real World Evidence Objectives


Convergence of translational informatics data mining approaches

RWE Precision Medicine

Target identification

Target validation

Pharmacogenomic markers

Indication expansion

Cross-study analysis

System biology models

External Innovation

Comparative effectiveness

Unmet health system needs

Health economics

Safety signal sensitivity

Value based medicine

Competitive intelligence

Patient stratification

High volume assay multimodal data Observational data mining

(Dan Housman’s)Translational Research Enterprise Informatics

Infrastructure Maturity Model

Level 7 Cognitive computingLevel 6 Real time decision supportLevel 5 External innovation and validation/optimizationLevel 4 Business focused solutionsLevel 3 Enterprise utilization and standardizationLevel 2 Data integration – data warehouseLevel 1 Fragmented and siloed analysesLevel 0 Reliance on external vendors

Translational Research Enterprise Informatics nfrastructure Maturity Model

Level 7 Cognitive computing: Advanced ‘many to many’ unsupervised discovery algorithms with sufficient supporting underlying semantic models . Use of very large compute to identify hard to find insights. Advanced imaging feature detection analysis. Significant use of NLP enrichment and ‘on demand’ access to external data on ad-hoc basis. Broad access and use of phenotype and genotype for large populations. Use of in silico models for systems biology translated from inputs and molecular innovation.

Level 6 Real time decision support: Use of predictive analytics to drive decision support at multiple levels. Patient level decision support with use of molecular markers such as trial recruitment at point of care leveraging informatics services. Real time access to data from active studies. Rapid incorporation of a broad array of data from data platforms such as microbiome, PRO, home health devices. CFR 11 validation of translational analysis tools for use in active studies. Broad establishment of enterprise data driven culture within organization. Advanced rapid access to data visualizations.

Level 5 External innovation and validation/optimization: Extensive automated data exchange, broad data access contracting. Collaboration cloud with pre-competitive partners including AMCs, patient advocacy, peers, and commercial data providers. Execution of complex pipelines across multiple modes of data e.g. mRNA and NGS and literature. Federated queries across multiple institutions and modes to answer key questions. Collaborative environments with shared users and identity management and social networking. Automated tiered storage and compute to manage very large data sets and reanalysis pipelines. Use of semantic web tools to expose resources. Common internal and external tools and approaches such as OMOP.

Level 4 Business focused solutions: Differentiated solutions by business area such as health economics, safety, research, operations, marker discovery, lab/sample availability, competitive analysis, pre-clinical, etc. Demonstrated and published results driving key business decisions achieved from enterprise informatics frameworks. Integration of translational research informatics with multiple enterprise systems such as portfolio management. Significant curated library by use containing clinical studies and associated open/public data. Secure web service API access to data. Standard and shared algorithms and methods across disparate internal teams. Access to broad array of real world evidence sources e.g. Twitter, adverse events, surveillance partnerships.

Level 3 Enterprise utilization and standardization: Focus on use of data for decision making in major R&D cycle decisions. Documented governance of use of data, quality processes for data, and internal/external sharing. Semantic translation of studies into common formats. Cross study and multiple platform analysis enablement through integration of analytic pipelines and advanced standardization. Central informatics framework for interfacing to multiple commercial, open, and internally developed research platforms. Policy based self service access to data. Factory model and self-service curation. Acquired data sets from subscriptions converted into standard formats or repository system.

Level 2 Data integration – data warehouse: Centralization of translational research data sets in single DBMS repository. Data includes clinical studies, molecular assays, observational studies, 3rd party data. Linkage at patient level across data and between data and analyses. Access controlled by ad-hoc governance model with honest broker or service delivery focus on analyses on an as needed basis to share data across groups. Self service access via web for browsing and exploring data including basic analyses. Focused pilots engage early adopter users.

Level 1 Fragmented and siloed analyses: Silo approach to clinical data controlled through experts such as biostatistics groups. Data stored in primary forms such as SAS data sets and files in organized directories. Analyses produced are ad-hoc with specific tools. Internal development of systems to offer intranet or file server access to data files beyond. Recognition of governance needs. Subscription services manage reference data or to search external data. Basic catalog available through files or experts. Desktop analysis tools primary interface to data.

Level 0 Reliance on external vendors: Historical focus on clinical only data sets with no ‘Omics and data integration internally. External vendors exclusively generate analyses for combined clinical and molecular data. Infrastructure for storage is file servers with limited governance and generally report focus. Limited to no institutional knowledge of available data sets from historical work.


Deloitte Health Miner Capabilities

Precision Miner

PopulationMiner

Outcomes Miner

Recombinant Platform

• Omics analysis

• Transmart++

• Analysis archive

• Data delivery pipelines

• Research data warehouse

• Visually explore populations

• View temporal relationships

• Select cohorts to analyze

• Identify basic correlations

• Large population data sets

• Propensity matched subsets

• Identify advanced correlations

• Compare treatment effectiveness

• Access curated data sets

• Subscription access to reports

• Data consortia & licensed data sets

• Data integration, cleansing, enrichment tools

• Data models and analytics frameworks

• Cloud and on premise deployment tools

• Commercial open source support

• Informatics and statistical models

Translational Research Platform

Research Portal

Sec

uri

ty a

nd

Id

enti

ty M

anag

emen

t

Master PatientIndexing Metadata/

TerminologyServices

Data Trust (DT)Research Trust

Data Management, Storage and Processing Engine

‘Omic Data Management

Data Marts (ADM, Research Mart,

CFDM, OMOP, i2b2)

Data Processing Pipelines

Data Acquisition Custom ETLPackaged

Parsers/Adapters

Data IntegrationData Integration

Hub (DIH)Data De-ID/Re-ID

Services

OADM DT Extensions

Business and Analytical Services

Cohort Matching

Metrics Calculation

Statistical Model Execution

Knowledge Management

RIE Services

Application Layer

(Miner)

Precision Miner Cohort Identification

i2b2

‘Omic Explorer

tranSMART+

Study Design

Study Recruitment

Manager

Population Miner

In MemoryExploration

Outcomes Miner

Compare

Safety

Patient Journey

Primary Sources Research Datasets

EMR/Clinical Clinical Trials

Data/Messaging APIs

Clinical/Omics Terminology

Mapping

Real world evidence?

Copyright © 2013 Deloitte Development LLC. All rights reserved.16

Safety Solution Vision (Example)

Safety Case

ReportsArgus DB

Safety DW (internal)

OMOP

i2b2

RWE

RWE Data Trust

Others

High Quality Real World Data/Analytics from Collaborators

Purchased Real World Data and Federation

Safety & RWE PlatformReporting & Analytics

Safety Reports

Query Interface

Analytics

Export to SAS, Excel

RW

E P

orta

l

Reports

Population Miner

Outcomes Miner

OMOP Analytics

Internal Analytics

Population Stratification

Inventory of Data Assets

Reports

Research Trust

Randomized Clinical Trial (RCT) and L4

Data

tranSMART

Precision Miner

Social Media DW

Complaints

Sentiment

Cross Study

‘Omics Analysis

Signal Detection


Teaming to Enable Data Driven Healthcare Improvement

• Decrease variation in clinical processes

• Measure the processes through analytics

• Measure, adjust, measure, adjust…

“We need to build on the examples of outstanding medicine at places like ... Intermountain Health in Salt Lake City, where high-quality care is being provided at a cost well below average. These are islands of excellence that we need to make the standard in our healthcare system.”

- President Obama to the AMAJune 15, 2009Chicago, Illinois

“We selected Deloitte as the best partner to translate Intermountain’s pioneering work to other systems. The use of our technologies will allow clinicians and researchers to more quickly discover practices to help usher in a new wave of innovation throughout the nation’s health systems.”

- Marc Probst, CIO Intermountain Healthcare

Qua

lity

Cost

Lowering cost through quality improvement

Intermountain is the initial member of what will be a Consortium of preeminent health systems across therapeutic areas and from around the world


• Approximately 2.1 million patients

• 137,881,670 diagnoses

• > 10 years longitudinal data set

• At least 2 years visibility for all patients

Diabetes Mellitus

• HbA1c• Blood glucose

levels Heart Failure

• Echo data• Staging of CHF• EKG• Stress test

Hypertension• Blood

pressure• EKG• Cardiac

status

Ischemic Heart Disease

• Pulse oximetry

• Cath lab data• Inpatient

activity

Osteoporosis• Bone Mineral

Density• Fracture Risk• Menstrual and

HRT statusRheumatoid

Arthritis• Bone Mineral

Density• Fracture Risk• Biologics use

Alzheimer’s Disease

• Cognitive scores

• EEG results• Genetic data

Infectious Diseases

• Lab results• Microbiology

results

Renal Diseases• Glomerular

Filtration Rate• Creatinine

levels

Leukemia• Cancer staging• Chemo /

radiation therapy

• Genetic biomarkers

Breast Cancer• Tumor data• Cancer staging• Chemo / radiation

therapy• Genetic

biomarkers

Chronic Respiratory Diseases

• Pulmonary Function Test

• Respiratory Rate

Medication (prescription

and adherence)

Patient Demographic

(e.g., Age, Gender,

Ethnicity)

Mortality data (with primary /

secondary causes)

Clinical Diagnosis

and Symptoms

Patient Encounters

Lab Results (numerical values and

text information)

Treatment procedures

(medical and

surgical) Lifestyle Parameters (e.g., Smoking, Body

Mass Index)

Vitals

Available

Intermountain Data


Smart Data Approach to the Deloitte-Intermountain consortium

Customer Visualizations

& Tools

DH

I Po

rtal

Deloitte as a Service Provider

Analytic Platform

Analytic Platform

Analytic Platform

Provider Partners

Analytics Platform

Deloitte has jointly developed an analytic platform with Intermountain Healthcare’s Homer Warner

Center

Data Providers

Deloitte is implementing the analytics platform at provider organizations participating in a consortium and providing their EHR / EDW data

Produce analytic resultsHelp generate analytics algorithms for new products

Conduct follow-on studies

Aggregate analytic resultsDeliver content to customersBroker follow-on studies

Subscription & Custom Study Fees

Data/Technology

Analytic Platform

Analytics Provisioning

DHI Provides Analytic Results to Customers

EHR / EDW

EHR / EDW

EHR / EDW

Access the results through subscription portal

Acc

ess

Via

Su

bsc

rip

tio

n

Customers

1

2

Business Model:

Custom StudiesCustom

Informatics & Insights On-Demand

3

Deloitte is focused on providing life sciences companies with the deepest insights from “near real time” medical record data in the world through: Analytics based upon Intermountain (in the future other consortium health data) systems; ~200 systems covering the

entire patient clinical experience Insights from leading health systems who are mastering the care processes Innovative business model to facilitate rapid learning with collaborative research with systems including Intermountain

health care

Illustrative

Study of patient outcomes for selected therapeutic area (TA)


Subscription PortalRWE Reports from RWE Data Sets

Safety DW Report Report based on purchased claims

Report summarizing social media analytics


Demos

Population Miner

Outcomes Miner

22 Copyright © 2013 Deloitte Development LLC. All rights reserved.

Development of a learning loop that leverages RWE and the experience of healthcare providers

Evaluating Evidence

from Studies

Validating Evidence in Real World

Collaborate to Develop

New Insights

Focused Studies to Generate

New Evidence

Implement

Learning


Vision: Leveraging tranSMART + workbench to identify insights from existing study

Evaluating Evidence from

Studies


Collaborate to Develop New

Insights

Focused Studies to

Generate New Evidence

Implementation of Learnings

Using Deloitte’s translational research tools suite of tools for evaluating current studies

Studying phenotypic and genotypic profile of patients participating in a recent Asthma study

Variants of PDE4 gene and CYP 450 gene indicate variation in outcomes (however not statistically significant)

Step 1: Viewing at insights into a single research study, specifically, a box plot of a gene signature list against all participants in an asthma study who have genomic data loaded. This shows us large variants in two distinct subgroups (Type I and Type IV)

Step 2: Heatmap view limiting our selections to those subgroups showing the variance in genetic markers. It shows variations, but they are not as significant to generalize insights


Vision: Pooled analysis of asthma studies to identify impact of genomics on treatment outcomes


Studies



Insights

Focused Studies to



Performing a pooled analysis of “multiple studies” across various asthma studies

Larger sample size enables studying phenotypic and genotypic profile of patients with greater confidence

Analysis indicates variants of PDE4 gene and CYP 450 gene showing significant variation in outcomes for certain treatments

Step 3: Now we perform a comparison of multiple different study groups to observe first the phenotypic differences (Age, Sex, etc.) and then compare the specific variances of two gene variants between the study groups

Step 4: Heatmap view now indicates significant difference in terms of how the genetic variations are impacting the outcomes of treatments


Vision: Overview of Asthma patients in real-world to enable better characterization of disease


Studies



Insights

Focused Studies to



Overview of all the asthma patients treated in the real-world setting in the past decade

Evaluation of current treatment paradigms in the real-world and correlation with outcomes

Identification of two key treatments medications that are the cornerstone of treatment for further evaluation

Step 5: Evaluating all the patients having ‘Asthma’ at Intermountain Healthcare to identify age, gender, disease frequency distribution. Identification of treatment, lifestyle, ethnicity and comorbidity patterns for the patients

Step 6: Ability to identify two most common medications used in patients with severe asthma condition for further evaluation using Outcomes Miner


Vision: Evaluation of outcomes for two asthma medications in real-world setting


Studies



Insights

Focused Studies to



Comparison of patients on ‘Drug A’ versus ‘Drug B’ to identify difference in outcomes

Evaluating ‘Emergency Visits’ as an outcomes and then filtering them by ‘Emergency Visits specific to Asthma’

Patients with CHF as comorbidity and treatment with Beta-blocker treatment indicate higher Emergency Visits

Step 7: Evaluating ‘Emergency Room Visits’ as the outcome on the dashboard, overall more Emergency Room Visits for Asthma patients with CHF disease as comorbidity

Step 8: Evaluating ‘Emergency Room Visits for Asthma’ as the outcome, high Emergency Room Visits for patients with beta blocker treatment for CHF


Vision: Evaluation of outcomes for two asthma medications in real-world setting (contd …)


Studies



Insights

Focused Studies to



Comparison indicates patients on ‘Drug A’ depression, neurological conditions and psychosis have high degree of correlation with Emergency Visits

CHF and Beta Blockers believed to have an association with CYP 450 gene variants

Indicates the need to further study impact of CYP 450 genes in drug outcomes

Step 9: Evaluating ‘Emergency Room Visits for Asthma’ as the outcome for patients on Drug A shows same degree of correlation with depression, neurological conditions and psychosis

Step 10: Evaluating ‘Emergency Room Visits for Asthma’ as the outcome for patients on Drug B shows limited correlation with depression, neurological conditions and psychosis

Knowing more about i2b2 data?

- 29 -*Note: Representative diagram – not all integrations are shown

Big picture type solution for ‘AMC’ genomics initiatives

RI Analytics & Care DeliverySource Data

Clinical Trials,Registries,

Internal/ExternalResults

BiobanksLIMS

‘Omics Platforms(CLC Bio)

Clinical EMRs& Claims

Labs

Partner Clinical data

Master Data Management

MPI/ProviderScientific

ReferenceTerminologyReference

CommonServices MPI HPCRef Data Mgmt Hub Security Collaboration Portal Storage

Data Trust

Research Trust

Data Warehouse /Research Stores

Clinical

Research

Omics

ETL

DataCuration

Data De-Identification

Data Workflow/ Enhancement

Closed Loop

Translational Research Applications

Statistical Analysis

R SPSS SAS

Re

se

arc

h P

orta

l Research Open Source

i2b2tranSMART/

Sample Explorer

Extended SystemsStudy

Recruitment Manager

Omics/Cohort Explorer

Honest BrokerData Pipeline

Research Information Exchange

File Storee.g. genomics (BAM, VCF, CEL)

Publications, PDF, Pathology

Research Data Marts

- 30 -

Representative View: Select Cohorts via i2b2 Query & Analysis interface

- 31 -

Representative View: i2b2 passport profiles available data

- 32 -

Representative View: i2b2 Passport, cont.

- 33 -

Representative View: i2b2 Passport – summary of data over time

- 34 -

Automated research request data mart production system

Getting AMC registry data into i2b2(for tranSMART)

Harvard/CHiPJonathan Bickel M.D., M.S., FAAP

- 36 -

REDCap Study Representation

- 37 -

XML to i2b2

REDCapArchive

(ODM XML)i2b2

Stagingi2b2PRD

File system Oracle Schema• Ontology• CRC

EDC system

of choice

- 38 -

Choose your Stud(ies)

• Choose studies to be imported

• Supply token to be used for study

• Click to initiate export

- 39 -

Choose your Stud(ies)

• If a project that has been previously exported is selected, the export process begins by cleaning out all references to the project from the i2b2 staging database.

- 40 -

CDISC ODM XML

<?xml version="1.0" encoding="UTF-8" standalone="yes" ?> <ODM ODMVersion="1.3.1" CreationDateTime="2012-02-03T10:59:14.175-05:00" FileOID="000-000-000" FileType="Transactional" xmlns:ns2=http://www.w3.org/2000/09/xmldsig# xmlns="http://www.cdisc.org/ns/odm/v1.3"> <Study OID="10"> <GlobalVariables> </GlobalVariables> <BasicDefinitions /> <MetaDataVersion Name="Version 1.3.1" OID="v1.3.1"> {YOUR METADATA HERE} </MetaDataVersion> </Study> <ClinicalData MetaDataVersionOID="v1.3.1" StudyOID="10"> {YOUR STUDY DATA HERE} </ClinicalData></ODM>

http://www.w3.org/2000/09/xmldsig

- 41 -

REDCap Study Representation

Modifiers POC (Kimmel Cancer Center)

Informatics Core Director, KCC, TJU Director of Research Informatics, and Research Professor of Cancer BiologyDr Jack London

Kimmel Cancer Center Deputy Director for Basic Science and Professor of Cancer BiologyKaren Knudsen, PhD

Professor, Cancer BiologyHallgeir Rui, MD, PhD

Informaticist, KCC Informatics Shared ResourceDevjani Chatterjee, PhD

Assistant Professor Medical OncologyHushan Yang, PhD

Vice President and CIOStephen Tranquillo


Jefferson Kimmel Cancer Centeri2b2 v1.6 Biospecimen Ontology

Specimen type (frozen or paraffin)

Pathologic status (normal or malignant)

Specimen class (solid tissue, fluid, serum, etc.)

de-identified case ID de-identified specimen

ID


Jefferson Kimmel Cancer Centeri2b2 v1.6 Tumor Registry Ontology

Tumor identifier modifier links different facts about the same tumor.

Changing these facts from concepts to modifiers allows multiple occurrences of the same fact for the same individual to be associated with the correct corresponding tumor.


Jefferson Kimmel Cancer Centeri2b2 v1.6 Genomic Profile Ontology

chromosome number

Results for a 58 gene assay panel.

mutation classification


Jefferson Kimmel Cancer Center tranSMART (prototype 1)

- 48 -

The FDA needed to explore new approaches to data management and analysis for effective evaluation of product safety and efficacy

Business Problem

• The FDA has committed to improving their overall submission review process

• Resources were spending too much time on basic tasks to aggregate data across clinical trials

• As a result, fewer resources were available for high-value data and regulatory analysis

Strategic Goals

• Implement improved data management systems across the following multiple FDA Centers• Enable the ability to:

• Automate the process of loading clinical trial data from multiple source formats• Correlate data across clinical trials through a simple and intuitive user interface• Conduct advanced analytics across multiple data sets to better inform regulatory decisions

• Shift the utilization of resources from basic data management to high-value regulatory science

Current Effort Ideal Effort

Regulatory ScienceData Management

Data Curation & Loading

DataSelection

DataAnalysis

InnovationLearningSharing

Eff

ort

Review Activities

tranSMART 1.2 prep


Demo tranSMART ModifiersIntegrated faceted GWAS results Cross trialsQuery by ‘sequence’Workspace ‘save’

- 51 -

Switch data center co-locates multiple hosting options for Life Sciences

Cloud and Hosting Eco-System

Deloitte – Internal UsersDeloitte – SaaS Solutions Deloitte – Client Hosting

Amazon

• Open burstable compute

• Client managed cloud purchases

• Connectivity• Low

cost/commodity on demand

Bluelock

• Approved for PII, PHA, HIPPA

• High Performance• Metered

applications• Subscription data• Cloud provisioning

of Deloitte managed burstable resources

Client Hosting

• Traditional Hosting• Local sensitive

data

Switch

Copyright © 2013 Deloitte Development LLC. All rights reserved.

Disclaimer: This publication contains general information only, and none of the member firms of Deloitte Touche Tohmatsu Limited, its member firms, or their related entities (collective, the “Deloitte Network”) is, by means of this publication, rendering professional advice or services. Before making any decision or taking any action that may affect your business, you should consult a qualified professional adviser. No entity in the Deloitte Network shall be responsible for any loss whatsoever sustained by any person who relies on this publication.

Definition footnote: As used in this document, "Deloitte" means Deloitte Consulting LLP, a subsidiary of Deloitte LLP. Please see www.deloitte.com/us/about for a detailed description of the legal structure of Deloitte LLP and its subsidiaries. Certain services may not be available to attest clients under the rules and regulations of public accounting.

Copyright notice: Copyright © 2013 Deloitte Development LLC. All rights reserved.

Member of Deloitte Touche Tohmatsu Limited

transmart community meeting 5-7 nov 13 - session 5: recent transmart lessons learned

Health & Medicine