Monday, December 3, 2012

Examining the Slow Transition to Risk-based Monitoring

In August 2011, the FDA published its draft guidance to industry, Oversight of Clinical Investigations – A Risk-based Approach.  The EMA, MHRA, and European Commission have also published papers in support of risk-based or adaptive monitoring.  Judging by the sheer number of conference presentations, forum discussions, and webinars, not to mention all the new vendor offerings built to facilitate the process, risk-based monitoring might be one of the most frequently discussed topics in clinical research today.

The FDA guidance encourages organizations to forgo the former one-size-fits-all, gold standard of monitoring:  100% source data verification (SDV) and on-site visits every 4 to 6 weeks.  There are two distinct themes at play here.  First, information technology reduces, if not obviates, the need to review study data on-site.  That site visits might be less frequent as a result cannot be a wholly unexpected consequence.  A move towards remote data monitoring, and any associated decrease in on-site visits, does not necessarily represent a philosophical shift for FDA, so much as an acknowledgement of the inevitability of advancing technology.  Had the new guidance been titled Oversight of Clinical Investigations – A Remote Monitoring Approach, the industry might not have raised a collective eyebrow.

A far more philosophical departure is FDA’s position that traditional absolutes, such as100% SDV, are often overkill -- they don’t necessarily produce more reliable results or offer greater subject protection than less intensive approaches would.  Instead, FDA says that decisions about frequency, intensity, and type of monitoring should be made according to the risk that each study, and even each site, presents.  Variables to consider in these decisions are therapeutic area, protocol complexity, investigational product safety, population vulnerability, investigator experience, EDC capabilities, and study phase.

Verifying anything less than 100% of source data is easier, cheaper, and faster than verifying all of it.  Applying more resources to critical areas and fewer resources to less important, less risky areas sounds obvious.  So why haven’t sponsors and CROs scrambled to implement these process changes as fast as possible? 

Assessing risk is hard.  You can download all the risk assessment tools you want, but it doesn’t change the fact that striking a smart balance between risk and benefit is the result of careful analysis, long experience, and good judgment.  You’ll probably never know if you overestimate the risk you face.  You’ll just end up performing unnecessary monitoring procedures that offer little benefit, but only after you’ve spent a lot of time and money on a useless risk assessment effort.  If you underestimate the risk you face and fail to sufficiently monitor the study, you defeat the entire essential purpose of study oversight, jeopardizing the safety of your patients and the reliability of your results.

Another reason sponsors and CROs have moved slowly towards an adaptive monitoring approach rests with the study sites themselves.  Without 100% SDV, sites need to integrate quality management into their processes, rather than rely on monitors to identify questionable data.  Risk-based monitoring requires more vigilance from site staff than is necessary with the traditional monitoring approach, and may require additional site training. 

Despite these difficulties, some sources report that companies are nevertheless beginning to implement risk-based approaches to clinical oversight.  In January, the Applied Clinical Trials website published a trend analysis developed by Medidata Insights.  Among their client companies, Medidata reports that median SDV percentage rates have been dropping steadily since 2007. 

Still, the perception is that transition to risk-based monitoring is slow.  An Outsourcing-Pharma article recently quoted the CEO of Annex Clinical who says firms are still checking every piece of data associated with a clinical trial, including data not central to data integrity or subject protection.  That same publication summarized the opinions of many attendees and speakers at this year’s Partnerships for Clinical Trials conference.  They believe that the guidances on risk-based monitoring lack clarity, and are urging regulatory authorities to make more solid rules.  Even Big Pharma concedes it has to understand risk-based monitoring better.  In September, Abbott, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, Eli Lilly, GlaxoSmithKline, Johnson and Johnson, Pfizer, Roche’s Genentech division, and Sanofi formed the non-profit TransCelerate BioPharma, with the goal of improving the quality of clinical studies and accelerating the development of new medicines.  Among their stated objectives is the development of risk-based site monitoring approaches and standards.

If Big Pharma is still struggling to understand comprehensive risk-based monitoring, smaller biotech companies are even further from implementation.  Smaller companies are less likely to have the resources and experience necessary to conduct thorough risk assessments and then translate them into adapted monitoring plans.  Also, small biotechs are more likely to conduct significantly smaller studies, which may translate into smaller risk tolerance.

And then there’s this:  Big Pharma is notoriously risk averse when it comes to deciding what biotech firms to purchase or what new products to license.  A due diligence team needs to present an airtight case to its management, who, in turn, wants to make sure the company’s investment will be sound.  Smaller firms understand this.  They know that they may only have “one shot on goal” to either partner or sell, and are concerned that their Big Pharma suitors will be dissatisfied with anything less than 100% SDV and frequent on-site visits. 

So what happens from here?  Transition to a risk-based monitoring approach may continue to inch along at its current pace.  Organizations like TransCelerate may develop standards which set the stage for large-scale adoption.  Maybe a final version of the FDA guidance will offer clearer parameters for implementation.   Whatever happens in the industry at large, it’s likely that the need to reduce clinical development costs will motivate some companies to aggressively move toward risk-based approaches, with or without accepted industry standards or further regulatory guidance.  They’ll be the ones to watch.

By Laurie Meehan