Advertisement
Articles
Advertisement

Managing Pharmaceutical Data

Tue, 06/26/2007 - 12:02pm
Developing information management strategies for faster, safer and more cost efficient drug development

Anjali R. Kataria
Co-Founder & Chief Marketing Officer
Conformia


The pharmaceutical industry today faces heavy FDA regulations, consumers who demand accountability and shareholders that demand innovation. But, the industry is mired in massively fragmented, slow, undependable and expensive processes - costing the industry millions of dollars, putting consumers at risk and creating difficultly in meeting shareholder goals.

One of the biggest challenges amongst this mire is accessing and leveraging prior knowledge from R&D -- i.e. data and conclusions resulting from both successful and failed experimental runs that are often buried in multiple silos across the product/process lifecycle. It's often a mystery of where is the data? Moreover, it's often even more difficult to recall critical information and, even if the data is found, it's sometimes not reusable.

Recent research on a Cooperative Research and Development Agreement (CRADA) with FDA, revealed that one of the biggest challenges and opportunities across development organizations in implementing approaches to support Quality by Design (QbD) (ICH Q8 and Q9) stems from the way in which critical product/process science knowledge is captured, stored and retrieved. Of the nine companies surveyed in a blinded study, plus prior research conducted in part with Purdue University, it is apparent that informatics, data management, and knowledge management is a major problem and one that has largely been overlooked in terms of its impact on providing a competitive advantage to companies seeking to bring new products to market safer, faster, and for less cost.
Data management across the product/process lifecycle
In several studies, which all point to inefficiencies in the pharmaceutical development operations environment, information management challenges appear as a common thread. 1 In most pharmaceutical and biotech companies, key product/ process science information is often buried in organizational and system silos across each division of the lifecycle, making it difficult to find and even more difficult to pinpoint important information about key process steps and outcomes.

The problem in a nutshell, manifests itself when the right people don't have access to the right information at the time they need it. In most companies, mission critical data is buried in paper notebooks and spreadsheets. Even worse, there are hundreds of mini custom databases that are a source of hidden information and costs. Scientists sometimes spend upwards of 40 percent of their time trying to locate critical information needed for decision meetings. Unfortunately, meetings often occur with incomplete information and limited visibility into the rich knowledge that exists in organizations regarding the product / process science.

Typically, as scientists learn about the product/process science, they are gaining tremendous knowledge, but unfortunately the information is not captured in a way that it can be preserved and passed back to the global institution. Critical product/process science information ends up residing in the minds of key scientists, in paper notebooks or individual computer desktops, as opposed to a central master repository. The information gets further diluted as these highly skilled scientists pass the information forward to the next group of scientists, (i.e. - API to Drug Product, or across the lifecycle from early development to late development), because the transfer is akin to "throwing the information over the wall." These internal knowledge transfers often end up leaving behind critical process/product science knowledge resulting from failed experiments.

Numerous findings also indicate that these data silos are often scattered across the globe and isolated by islands of expertise, culture, and function -- leaving companies with a chasm between R&D and commercial manufacturing. For example, in Tech Ops or Manufacturing Sciences and Technology (MST) organizations (i.e. the front line for commercial manufacturing), technology transfer from R&D into commercial manufacturing is often another major bottleneck resulting in high costs, lengthy times and overall inefficiency. What goes unnoticed in many organizations is how rooted the tech transfer challenges are in poor to non-existent enterprise information management sources across development.

In fact most commercial manufacturing organizations that need to review the development history don't have a central enterprise master repository of product/process development history. Instead they spend precious time and resources pouring through a mismatch of paper printouts, hunting down key scientists, and sifting through mounds of uncorrelated data and reports -- all for suboptimal results that consists of-uncoordinated data, without good context, rendering information that is difficult to use.

In no other process industry does technology transfer pose as severe a challenge. The pharmaceutical industry's approach to data is not merely time consuming or costly but its rate limiting. Today, companies fly their best scientists around the globe to physically sit at the commercial manufacturing sites, sometime for six months to a year, just so they can transfer a fraction of the vast knowledge they have accumulated.

This might seem surprising given the world class large commercial enterprise systems like ERP, MES and others that exist in pharma organizations. These systems typically serve organizations outside of development i.e. commercial manufacturing, finance, sales/marketing, etc. They are not an antidote to the pharmaceutical development information management challenges. When you take a closer look at why these systems haven't been able to support development environments it becomes even clearer how significant the challenges and opportunities are for strategic information management across the pharmaceutical development organization.

In development organizations, scientists are working with multiple routes, pathways, materials, etc. The drivers of development are a mix of innovation and high variability in the beginning, moving towards an environment of more certainty, lock down and mere execution against a known standard. Hence development requires enterprise systems that can accommodate changes on the fly, multiple iterations, and at the end of the day are flexible and user friendly. Thus, commercial manufacturing systems have just not been able to support development needs.
Feeling the impact
Beyond the immediate impact these information management challenges pose on a company's top line growth and their efforts to reduce bottom line costs comes a greatly diminished productivity. Without the right technology-based scientific and regulatory strategies, companies wind up repeating experiments, using excess materials, adding unnecessary time to market, and creating inefficiency. Aside from the top line/bottom line growth and productivity are other levels of impact that like interest, compound -- i.e. the ability to gain a competitive advantage via platform based technology development and to comply with regulatory filings.

First, it's apparent that companies that aim to develop a platform-based approach to technology development will benefit significantly from an information management strategy that spans the product/process lifecycle and supports a central master development history repository. These companies will be in a better position to exploit the knowledge and leverage the science base.

Second, the lack of a coordinated information management strategy to support the product/process lifecycle will have a significant effect on driving implementation of approaches to support QbD. Because the development of a new drug occurs over many years, the product/process data needed to support and sustain QbD approaches will be difficult for development, quality, and regulatory organizations that rely on this long life data for filings. An inability to capture product/process knowledge in a centralized manner across the lifecycle fundamentally opposes the support needed to justify the rationale of the product/process understanding -- a key component of QbD. Without a strategic information strategy to support the entire lifecycle, companies end up at risk.

Finally, with multiple products in the development pipeline, for a company that wants to standardize QbD and regulatory compliance approaches across their multiple product portfolio they'll need to leverage the science via a platform technology. It will become extraordinarily expensive and very lengthy to create point to point integrations of each custom database compounding the information management problem, impacting a company's future growth and weakening their ability to drive new products to market rapidly and safely. Fortunately most major pharmaceutical companies have custom database creation and are developing information management strategies and roadmaps that consider the product/process lifecycle and master development history management. Those that aren't will undoubtedly be left behind as their operations are likely to become much more expensive, less productive, and less innovative, inhibiting the ability to obtain the right information at the right time and continuing to make compliance and product development expensive, unreliable, time consuming and difficult to scale.
Conclusion: The Silver Lining
In conclusion, it's clear that pharmaceutical industry executives face numerous challenges in their current development information environment. However, those companies that can quickly develop an information management strategy to support development operations stand a chance of gaining a very significant competitive advantage. Any strategic information management strategy should provide an opportunity to:

A. Integrate the Development information strategy across the entire product/process lifecycle
B. Ensure that the information strategy is coordinated with, and helps drive results for the business and science strategies
C. Supports their filings approach to FDA's Quality by Design
D. Create a Master Development Repository to support information access, retrieval, tracking and exchange among the entire global organization

This begins with:
1. Creating a strategic information roadmap that supports the development process map and views information as a strategic asset
2. Stopping. the development of short term custom data bases that have no long-term benefits- and replacing that technology with one enterprise system spanning all of development
3. Getting a handle on present information systems like LIMS, MES, EDMS etc and developing an integration strategy to for business advantage
4. Making a commitment to the strategic vision of a central master development repository

There are a number of companies presently working together to address the challenges that span the product/process lifecycle and better position pharmaceutical companies to "bridge" the chasms within development through commercial manufacturing.. They include- major technology companies, leading pharmaceutical companies and key services providers. The proposed solution lies in enabling electronic data capture and correlation of integrated resource flows of people, materials, process, equipment, environment/ facilities, and standards in a single system across the entire drug product/ process lifecycle.

By adding a strategic information centric view, development organizations can move to a pro-active, near real-time, electronic, integrated information management environment that enables PAT / process understanding , design space and other approaches to support QbD resulting in better scientific decision making, greater operational efficiency, and improved regulatory success. This, in turn, will bring new products to market safer, faster, and for less cost.

1 FDA- Conformia CRADA Research Study Part 1 Preliminary Phase 2007
Advertisement

Share this Story

X
You may login with either your assigned username or your e-mail address.
The password field is case sensitive.
Loading