For science to succeed, the pharma players all need to be on the same information technology page.
At the end of the day, the life sciences research industries (pharmaceutical, biotechnology, vaccines, medical devices, contract research organizations, etc.) exist to bring new therapies to patients. This process starts in the laboratory, traverses multiple business units (and companies) and ultimately results in review by a national regulatory agency. At the core of this process is information, with scientists assessing the value of the information and adding to it, and information technologists determining how best to provide and manage it.
For many years, the life sciences research industries have struggled with a fundamental problem. Although the value their business units add is rooted in science, the means to deliver the value is rooted in technology. Based on the competitive nature of each organization, this has led organizations to deploy their own proprietary information technology. Over time, these approaches become so entrenched in the business process that maintaining and enhancing them often distract from the value-added science that is being performed.
In recent years, individual life sciences research companies have recognized that they will differentiate themselves from their competitors through the value-added science they bring to market, and not through their proprietary information technologies. Industry-wide standards such as those put forth by the Clinical Data Interchange Standards Consortium (CDISC), provide one means of enabling research companies to focus on science. The increasing adoption of the CDISC standards provides value across the clinical trials information process, not the least of which can be seen in standardized and efficient safety reporting.
Most mature life sciences research companies have adopted corporate data standards because of the operational efficiencies they enable. By having such standards, businesses can create libraries of reusable data capture forms, data cleansing algorithms, and data analysis utilities. Although many organizations often joke about standards being so successful that they have multiple standards in place, there is little doubt that implementing corporate standards has been effective for research companies.
In recent years, the life sciences research industries have rallied around the consortium-led CDISC data models. With more than 230 corporate members, the CDISC organization is developing operational and analytical standards in support of the clinical research process. Begun almost 10 years ago, the industry conversation regarding CDISC has largely moved from “why” and “what” perspectives to more tangible discussions of “how” and “when.” Active communities are working to use the CDISC standards to better unify how data is collected for clinical trials, transferred between research companies and submitted to the FDA, but little work has been focused on the need for standard analyses and reports.
As with other corporate standards, most research companies have adopted an internal standard with regard to clinical trial safety analyses. (Currently, efficacy analyses are readily accepted as more difficult to standardize). Across organizations, the content of these analyses is almost identical although the structure varies. A sample safety summary is shown in Figure 1.
|BODY SYSTEM/Adverse Event||Placebo
N = 270
N = 368
N = 214
N = 93
|BODY AS A WHOLE|
|Figure 1. Sample safety table. (Source: SAS)|
Although the content in Figure 1 is straightforward, there are many different ways to represent this same type of information. Because of these variances, research companies sacrifice operational efficiency in exchange for customized content that adds little value to the overall analyses. Additionally, because each organization’s standards are different, the content submitted to regulatory agencies for approval will vary. This prevents the agencies themselves from providing an efficient regulatory review, which ultimately delays decisions regarding the approval of new therapies.
A changing regulatory landscape
A number of critical regulatory changes have been initiated in recent years. These include the following FDA documents:
- Guidance for Clinical Investigators, Sponsors, and IRBs; Adverse Event Reporting — Improving Human Subject Protection (April 2007)
- Guidance Drug Safety Information – FDA’s Communication to the Public (March 2007)
- Reviewer Guidance – Conducting a Clinical Safety Review of a New Product Application and Preparing a Report on the Review (February 2005)
- Guidance for Review Staff and Industry – Good Review Management Principles and Practices for PDUFA Products (April 2005)
- Clinical Review Template (July 2004)
Each of these provides valuable information that cannot be adequately summarized in a single journal article. Suffice it to say that all of these documents collectively point to the need for consistent and comprehensive review of safety information as part of the approval process.
This key safety review applies not only to the research companies submitting their therapies for approval, but for the reviewers as well. It may not be that surprising to learn that different regulatory agency divisions review submissions differently. Even within the same division, the reviewing process is often personalized by each reviewer. This is simply not efficient, for either the regulatory agency or for industry. It is not unusual, for example, for companies to realign their internal analysis standards to match those of an individual reviewer, rather than building submissions using a standard, company- and reviewer-neutral strategy.
The recent FDA guidances and documents now provide for a standardized review practice at the FDA, which will ultimately translate into more standardized submissions as they relate to safety. FDA is not only providing these new guidances, but actively training its staff to provide consistent reviews. Like all business process changes, this will take time to fully implement, but will ultimately provide an important level of synchronization across the agency, and across industry.
By training the individual reviewers to perform their work consistently, the FDA will indirectly train industry. There will be little value for an individual research company to apply its own corporate safety analysis standards as it becomes more established that the agency reviewer will consistently review the safety materials. Instead, research companies will recognize the value in producing and “pre-reviewing” the safety submission content in the same manner as the agency reviewer. Company standards, where they exist, will yield to regulatory expectations.
With the life science research industries adopting industry-wide data standards, and the FDA influencing not only the use of those standards in data submissions, but also influencing the standardization of the analysis content, it becomes clear that an important, but missing, step is the standardization of the analysis process and its associated tools.
The transition to an industry set of analysis tools will not be without obstacles because established research companies have invested significant resources in building their own analysis and reporting libraries. Companies must see operational gains in choosing to relinquish these established processes. Less mature research organizations will more rapidly accept these standards because they often struggle to implement robust and repeatable analytic processes, and many struggle even further in hiring and retaining qualified staff to perform these tasks.
Contract research organizations may see surprising benefits in more swiftly adopting these standard reporting practices. Because, historically, individual research companies operate under their own organizational standards, contract research organizations must provide customized safety analyses to each company. By accelerating the adoption of these emerging standards, contract research organizations can streamline their business processes and create their own operational efficiencies. These efficiencies can be passed on to their customers. By charging their customers a premium for deviating from the accepted industry standard, quicker industry adoption can occur.
With standard tools available, and data standards in place that support them, it seems reasonable to look at a standard method of applying these standard tools to the standard data. The best approach may be one that utilizes “software as a service” (SaaS). With SaaS, a centrally-hosted safety reporting service would exist on the Web, and users could call that service as necessary to produce their standard safety reports. The service would provide the means to specify the relevant data paths and report destination paths, and would ensure the security and integrity of the data and reports being processed.
Biostatisticians and technical scientists would still be responsible for reviewing and interpreting these results (just as regulatory reviewers are). Additional expertise and business processing would still be required for analyses that cannot be standardized, such as those associated with efficacy. These types of analyses tend to be very indication-specific, and in many cases are very trial-specific.
Tip of the iceberg
The analyses associated with bringing a new therapy to market represent a small fraction of the overall work that is performed; these analyses are also the most visible aspect of that work. By standardizing the means to report safety across the industry, there is tremendous opportunity to minimize iterations between FDA and research companies throughout the submission process, all the while providing operational efficiencies for both industry and the reviewing agency. The ability to ultimately deploy standard safety analyses as a service will provide further efficiencies in terms of system implementation and maintenance.
The value of industry standards cannot be overstated. By implementing the CDISC standards as part of the capture, transfer and reporting of business processes, research companies can renew and maintain their focus on the science of bringing new therapies to market. The science of bringing new therapies to market will remain extraordinarily complex for the foreseeable future; the business of bringing new therapies to market must not.
About the Author
David Handelsman leads the strategic development and marketing of life sciences research software at SAS. Previously, he was a senior director with ClinTrials Research and worked with PRA in Germany and the US.
Filed Under: Drug Discovery