The global pharmaceutical industry struggles with integrating the multiple clinical data streams necessary to operate a global clinical operation into one seamless accessible database. There have been many attempts. Different vendors such as Oracle and Accenture have made noble attempts to support the industry. At the time of this writing, we cannot be certain how well these “life science data hub” integrations fared. We have anecdotal evidence at least some did not turn out well.
It is a topic near and dear to us. Hence, why we were intrigued with an obscure white paper published April 2019 from a team of LabCorp (Covance) authors. Titled “A Dimensional warehouse for integrating operational data from clinical trials”, it appears these authors have designed a truly exceptional system. It is not clear if this exact system is in production and there can be a big difference between a proof of concept and a system fully validated and in production.
The authors introduce to the reader the universal goal in clinical trials— industry sponsors need timely, consistent and integrated access to clinical trial data. They continue they have developed a “data warehouse that can integrate operational data from any source, conform it to a canonical data model and make it accessible to study teams in a timely, secure and contextualized manner to support operational oversight, proactive risk management and other analytic and reporting needs.
The LabCorp (Covance) built solution consists of a dimensional relational data warehouse, a set of extraction, transformation and loading processes to coordinate data ingestion and mapping, a generalizable metrics engine to enable the combination of operational metrics and key performance, quality and risk indicators and a set of graphical user interfaces to facilitate configuration, management and administration. When combined with the appropriate data visualization tools, the warehouse enables convenient access to raw operational data and derived metrics to help track study conduct and performance, identify and mitigate risks, monitor and improve operational processes, manage resource allocation, strengthen investigator and sponsor relationships and other purposes.”
A truly exceptional system if it can accommodate all of this, in production, at scale. The value would be tremendous. Big pharma has spent decades attempting to realize just this application!
We did some digging and found commercial artifacts pointing toward some of this underlying technology. Enter the Xcellerate® platform. Written up in marketing material under the Amazon S3 News (meaning that LabCorp’s Covance Xcellerate sits in the AWS S3 cloud) we found Covance (LabCorp) has incorporated many of the concepts above in its Xcellerate Informatics Suite designed with the operational data model described in the academic paper.
Xcellerate Data Warehouse Architecture
Underlying the core principles are two distinct data repositories as part of the Xcellerate Clinical Trial Optimization Suite: an operational data warehouse and a clinical data warehouse. The operational data warehouse is designed to support operational objectives and is based on a universal data model. Operational data remains consistent across clinical studies and can be represented by a general data model that tracks site performance, subject visits, data quality, etc. On the other hand, clinical objectives cannot be generalized and require an unconstrained data model. Often data is not harmonized; it may be in unstructured manner, etc. Hence Covance has embraced as well a non-relational (NoSQL) approach.
The Covance marketing collateral thereafter includes a section for “Creating an Operating Data Model” where they sought to “decouple” operational and clinical objectives.
They note “Our approach at Covance to clinical development analytics is based on the principle of separation of concerns, that is, the uncoupling of operational and clinical objectives and hence can segment, capture and manage the different data streams.” They move on to Implementing a Data Integration Layer where they describe their dependence on the TIBCO Enterprise Service Bus and of course the pervasive ODM format. TIBCO ruled the enterprise integration roost a decade ago but with the advent of cloud and cloud scale a whole new batch of vendors have positioned higher than TIBCO on the Gartner Magic Quadrant. The piece then moves to Organizing the Operational Data Warehouse which delves into heart of the original topic of this snapshot. Based on a set of relational schemas in Microsoft SQL Server, they include an A) landing area for the data B) staging area where data reconciled, harmonized C) aggregated and of course the Operational Data Warehouse (ODW) itself where the data is organized for consumption. We believe Covance is utilizing the very popular vendor product Snowflake. Covance includes other sections focusing on the gathering of data, capturing of metrics & KPIs, security model, etc. For those interested in this topic, we recommend at least a bookmarking for future reference.
We sought to understand how well Xcellerate was doing in the market. Was this product being distributed and consumed as a distinct and separate part of Covance offerings or is it purely an integrated part of their contract research services? Was it a primary reason why industry sponsors were selecting LabCorp/Covance over other tier 1 CROs? Unless one sits in the inside this is hard to know. We must note that even for those that are (and we know plenty of them), they cannot talk about any of this due to non-disclosure agreements. So we conclude a brief and cursory review of recent publicly available evidence of Xcellerate market acceptance and consumption. We undertake this by 1) reviewing Covance press releases and news article 2) publicly available presentations and other artifacts and 3) via self-generated publicly available content such as LinkedIn. If demand is skyrocket for the comprehensive data warehouse we will find evidence.
In December 2018 Xcellerate Clinical Research Associate Dashboard was awarded “Best Technological Development in Clinical Trials at the 14th Annual Scrip Awards. In October 2018 LabCorp reported 7% organic CRO growth and noted “Technology is a critical component of streamlining trials and we continue to see a broad interest in our powerful investigator performance data, real-world evidence insights and Xcellerate platform.” Melissa Fassbender, an excellent writer for Outsourcing Pharma reported GSK selected Covance and noted “as per agreement GSK will use Covance’s Xcellerate Monitoring, Xcellerate Insights, and Xcellerate Clinical Data Hub solutions. “As of 2018, GSK was the sixth largest pharmaceutical company in the world. Clearly a serious indication of the Xcellerate data warehouse viability. A number of generic case studies are published—mentioning a sizeable industry sponsor was utilizing the system but apparently Covance couldn’t secure permission to mention the actual name. A search in LinkedIn turned up over 250 names associated with Xcellerate. This was more than we expected. We will not summarize personal information here. We conclude Xcellerate has been accepted, at least in part, by the market. We cannot be sure as to the magnitude of its success in the market. There is surely evidence that global industry sponsors express interest and may even include the system access as part of a major contract research outsourcing deal.
There is a school of thought out there that market forces, competition and specialization will yet again transform drug development over time. From this theory, CROs will start divesting from heavy technology stacks and will move toward an ecosystem of cloud-based software-as-a-Service vendors and utilize cloud-scale integration services such as Mulesoft so the CRO in effect acts as the orchestrator to tie together the cloud services for a unified trial or program “experience.” The various services are modular and hence they can tailor the exact offering to the industry sponsors’ needs. To some extent they do this today but with what is in fact heavy stacks requiring lots of labor. On the other hand, CROs have built deep and extensive strategic partnerships where they even co-develop drugs with industry sponsor partners. In these scenarios the CRO must invest and manage an comprehensive technology portfolio to manage data and documents for the long haul. The economic returns factor in the necessary investments ongoing. Perhaps some CROs that excel at technology will become continue to invest heavily while those that are not will focus on the services and activities that indeed differentiate their CRO services. It would appear Xcellerate represents an impressive suite of robust, complex and comprehensive technologies. As long as it drives more business the executives at LabCorp will continue to invest.