IT Clouds in Life Sciences

Key to long term sustainability

Alan S Louie,  Research Director IDC Health Insights USA.

Cloud solutions offer immediate returns in improving operational performance and reducing costs. Multiple approaches to cloud solutions will be needed to satisfy life science industry needs, complicating the development of a coherent cloud strategy.

Patent expirations and their associated loss of blockbuster revenues across the life science industry have forced companies to take a hard look at ways to improve operational efficiency. As part of a deep introspection that has never before been considered, companies are critically assessing which activities are truly core to the business and which activities can be performed more efficiently by a partner. IT infrastructure is clearly an area under review and it is becoming clear that some, if not all of this activity can be performed more effectively and efficiently outside of the organisation. With data and information as core assets to the industry, companies are finding that not all information is equal and that the willingness to relinquish control of information varies, depending on where in the organisation the information resides. Ongoing innovation in the commercial IT cloud solution space has recognised these concerns and is developing (and evolving) product offerings for each need.

Key factors surrounding it cloud adoption in the life science industry
Fundamentally, the opportunity for IT cloud solutions is driven by the need to reduce costs while maintaining the same, or better, organisation-wide operational performance. While near term ROI is driving investment today, the shift to IT cloud solutions is also supported by the significant potential for long term benefits for the organisation. While external, public IT cloud solutions (e.g. Amazon Web Services Elastic Cloud 2) have the potential to reduce IT infrastructure costs by as much as 90 per cent and enable access to increased computational resources on-demand, this approach requires that company data reside outside of the organisation, which potentially exposes the organisation to the risk of accidental, or malicious data access by external parties. For efforts critical to the organisation (e.g. clinical trials or M&A), the risk of uncontrolled access, no matter how small, is typically too great to accept, and in these situations, the status quo, or better controlled options (e.g. internal IT clouds) are likely to remain the norm for the foreseeable future.

It is also clear that the industry is undergoing transformational change in a number of areas that are relevant to the discussion. The life science industry is becoming increasingly global, in part due to outsourcing to lower cost regions, but also due to growing opportunities in emerging markets. At the same time, M&A is increasingly becoming the near term path to backfilling anticipated revenue losses from blockbuster patent expirations, requiring organisations to regularly integrate and manage new assets and capabilities as they seek to build for the long term. This constant change would challenge the IT strategy and infrastructure of any organisation, but comes at an especially difficult time as the industry seeks rapid change while needing to maintain tight regulatory control to avoid compliance issues and potential penalties.

Additional pressures on IT are arising from the geometric growth in the amounts of data that are being generated by life science research. While most of the data is being generated during discovery research, data in clinical development is also growing (particularly from imaging and genomics), as is the data supporting M&A activities. Effective management of these growing data and information requirements will be key to organisational success over the long term.

Who, When and Where
A critical mistake for cloud vendors focused on the life science industry would be to assume a (one size fits all) approach to the adoption of IT cloud solutions. Information within the industry can be heavily siloed, often by design, and management of different data often carries regulatory compliance requirements that also bear financial risks. In looking at the way that data and information are managed in the industry, one needs to focus on the key industry segments as key differentiators as to how much control organisations are willing to relinquish in pursuit of direct cost savings and improved operational effectiveness. Key areas and their segment-specific requirements include:

Discovery research: This early part of the life science value chain is largely unregulated, highly diverse, and multi-faceted in its generation of and use of data and information. New innovations and technology advances are increasingly automating the generation of large quantities of data and it is not uncommon to need to create and manage terabytes of data on a routine basis. The very large datasets are often unstructured and the underlying purpose and design of the research that generated the data is typically not associated with the data directly. As a result, the risk to the organisation form accidental access or release of discovery research data is relatively low. In discussions with major life science companies, it is clear that this area is a prime opportunity for external IT cloud services. Key to the value delivered to discovery research efforts includes paying for only what you need from both storage and computing perspectives, the ability to access increased capacity on-demand, not paying for excess capabilities and significant reduction in IT infrastructure capital and operating expenses over time.

Clinical Development: Data generated and used in the management of clinical trials is tightly regulated and core to the ability of an organisation to successfully advance a new drug to market. Clinical development efforts routinely involve human patients, introducing additional issues and concerns over drug safety in addition to the fundamental issues over whether a drug works or not. While the diversity of data types used in the course of a clinical trial is much more limited than data generated during discovery research, additional complexity is introduced by the global nature of the clinical development effort, often requiring hundreds of trial principal investigators located in as many locations around the world. Timely access to information is critical even as companies work to tightly control the collection, transmission, and analysis of the data. With the regulated nature of the data also comes a financial penalty for non-compliance as well as the potentially more devastating risk of damage to the industry's high standards for ethics and value to humanity. As a result, external IT cloud services are not perceived as appropriate for this segment and alternatives, both private IT clouds and vendor-based private cloud approaches (i.e. hosted applications and software-as-a-service [SaaS] are likely to remain the key approaches to data and information management in this space over the long term.

Drug safety: Although primarily an effort under the umbrella of clinical development, drug safety is a key area that merits separate consideration. Individual drug safety issues are defined as ‘adverse events’ in the industry and severe adverse events (SAEs, including death or other major life threatening impacts) have specific regulatory requirements that companies must meet to maintain regulatory compliance. In particular, SAEs must typically be reported to regulatory agencies within 10 days or severe penalties may result. This issue becomes more complicated as clinical trials are conducted globally, since multiple jurisdictions become involved. The critical nature of safety data, combined with the regulated nature of the clinical trials suggest that cost savings and general operational efficiency improvements are likely to be secondary to meeting regulatory requirements. That said, drug safety is increasingly being looked at as a non-core competency within life science companies and is increasingly being outsourced to qualified application vendors (i.e. general contract research organisations [CROs] or specialty drug safety service providers) along with their private IT infrastructures [i.e. private IT clouds].

M&A: Mergers and acquisitions, along with other collaborative efforts, is a growing effort within the life science industry. Partnerships of all kinds are changing the way that organisations manage and share data, requiring new thinking regarding how data is held and used. For traditional M&A, data security and control are paramount, since any lapses can result in lost opportunity. In these cases, the high risk necessitates the tightest control over data and information and, in some cases; companies are only now shifting away from paper, human couriers, and face-to-face signoffs. Access across corporate firewalls is tightly managed and IT clouds are unlikely to deliver sufficient value to drive change in this space over the foreseeable future. Broader partnership and collaborative efforts, on the other hand, are also growing (e.g. academic collaborations, pre-competitive consortia, and translational research efforts) and to a large extent are being driven by data and information sharing. In these cases, it becomes possible to exploit both external and internal IT cloud solutions as separate, isolated IT ecosystems where key information is available and accessible only to collaborators, separate from other corporate data.

Companies today are actively seeking to consolidate data on an organisation-wide basis as part of ongoing efforts to apply business intelligence to both scientific and business management. Almost by default, these efforts are running into concurrent efforts to apply IT cloud solutions to improve operational efficiency at the operational level. The diversity and scope of implemented IT cloud solutions is likely to complicate these data consolidation efforts. Regardless of the specific approaches selected, it will be critical for organisations to incorporate IT clouds into an overall, organisation-wide data management strategy, based on the additional complexity that is likely to be introduced from increased dependence on external partners for IT infrastructure.

Where we are going
The life science industry is clearly in the midst of transformational change and it is reasonable to expect that IT will be a core element in the change process. The industry overall is finally reali sing that it is not significantly different from other major industries in many ways and, as a result, realising that it could benefit from best practices proven in other industries. As an example, the financial industry regularly addresses data security and privacy concerns and M&A issues are regularly addressed by industry in general. It can be expected that IT infrastructure in the life science industry is likely to become increasingly commoditised over the near term, with particular industry segments seeing early adoption as companies look to gain near term cost savings and improved operational effectiveness. Risk and perceived risk can be expected to slow the adoption of IT cloud innovations in other industry segments, with limited or no progress in specific areas where cost concerns are secondary.

Globalisation, geometric data growth, and increasingly complex partnering relationships are major challenges that will require changes in the life science industry IT landscape. As companies expand to emerging regions, they are increasingly finding that existing IT infrastructure is inadequate to do science and conduct business effectively. This shortcoming, while transient, can slow progress at a time when the industry can least afford it. Continuing rapid growth in the generation of data of all kinds is increasingly limited by our ability to digest and extract meaningful insights and is continually complicated by islands of diverse data silos. And finally, the promises of partnering to more efficiently advance pre-competitive research, reduce the overall risk in advancing new drugs to market, and improve patient outcomes over the near term will all require new ways of sharing data and insights that were historically discouraged, requiring new mindsets and social change as organisations seek to work smarter, faster and more efficiently.

While only a single component in broader efforts to advance information management, IT cloud solutions have arrived at a time when they can be effectively leveraged into both short term solutions and long term strategic planning. While taking the jump into IT clouds can be easy, fast, and cheap, it will be important for both senior management and IT administrators to take a step back and reflect on the larger picture to ensure that their efforts are consistent with long term goals and expectations. The life science industry is large, complex, and ripe for opportunity for those companies willing to properly think through their near and long term goals for information management. It is clear that information is, and always has been, key to success and a broader information management strategy that includes IT cloud solutions is likely to be important as both an enabler in the near term and a requirement over the long term.

-- Issue 20 --

Author Bio

Alan S Louie
TOP