Sponsors Collect Too Much "Non-Essential" Data, Say Tufts And TransCelerate

A conversation with Laura Galuchie, senior director, global clinical development, TransCelerate Program lead, oversight committee, MSD

Over the last 10 years, both clinical trial procedures and data points collected have skyrocketed, and everyone’s noticing.

The last decade has brought a 40% increase in total procedures and a 283% increase in data points collected during Phase 3 pivotal trials, according to a paper presentation by Getz et al. On the heels of the paper, TransCelerate BioPharma debuted a new initiative, Optimizing Data Collection. Under its auspices, TransCelerate teamed with the Tufts Center for the Study of Drug Development (Tufts CSDD) and the Tufts University School of Medicine on a collaborative study that dug into protocol design and data collection practices among 14 biopharmaceutical companies.

It found that:

• up to 32.5% of Phase 3 patient data is “non-core” or “non-essential,”
• 25%–30% of total participant and site burden is tied to lower-value data collection, and
• in non-oncology trials, as much as 40% of patient-level data may not be necessary.

To put the findings into context, Clinical Leader spoke with Laura Galuchie, senior director, TransCelerate program lead, oversight committee.

Clinical Leader: Clinical trial protocols are getting more complex and collecting more data points. Now we have data to illustrate that. How have trials become complex, and why?

Laura Galuchie: We did not assess that question in our research, but we could speculate quite a bit on the rationale for it. As clinical trials target more specific patient populations, which is good for patients because that means they'll have more targeted therapies, you need more assessments to fine-tune and identify those populations. Combined with some of the advances in technology, it makes it easier to collect more data. Even if you don't know what you're going to do with it, it's just easier to collect.

What are some problems that arise when we collect so much data?

The burden varies depending on your role. If you are a patient or a site, additional procedures or more frequent procedures may add to more time burden. If you are a sponsor or CRO, more data could mean more queries you'd write and the site would have to answer. That relationship between volume of data and impact on timelines, in terms of that overall protocol time period, could delay a compound getting approved and being made available to patients.

Looking back to the planning phase of this project, what questions were you seeking to answer?

TransCelerate recently launched a regulatory landscape highlighting the six recent ICH guidances that focus on fit-for-purpose, risk-appropriate clinical trials. With those in mind, the Tufts organization and participating member companies from TransCelerate met to fine-tune what questions we seek to answer. And one of the new questions was around non-essential data. It really highlights that it's not just the procedure that could add undue burden, but also the frequency.

What were some of the main takeaways from this study? What's something that surprised you?

The terminology of core and non-core has been used in past Tufts research. Core procedures produce data to support the primary or key secondary endpoints. They are fundamentally associated with a hypothesis. Required and standard procedures include informed consent, drug dispensing, adverse events, and similar fundamental clinical research elements.  The non-core procedures would be everything else, such as procedures that support exploratory endpoints and safety or efficacy procedures not tied to a study endpoint or objective.

There are two separate categorization activities: relation to the study objective supported (core/standard/required/non-core) and minimal frequency needed to support the objective, independent of the objective category. For example, if your primary study objective is to assess change from baseline to week 12 in X, then the assessment conducted at baseline and week 12 would be essential.  If an additional assessment was conducted at week 6, that timepoint would be non-essential.  So, you can have both core and non-core procedures with non-essential timepoints. Any procedure conducted beyond the minimum frequency is non-essential.

Several ICH guidances talk about unnecessary burden and unnecessary data collection.  By evaluating the purpose of each procedure and its associated frequency, trial teams may be able to identify opportunities for adjustment as appropriate for the indication, maturity of the compound, and information known about the compound.   

Prior to this exercise, were sponsor companies doing this on their own?

I don't know whether the companies that are participating do that as a routine process. But I can share that it was a new analysis for Tufts, and it was a learning experience for everybody involved.

I imagine this might prompt them to evaluate data collection on an ongoing basis.

Any sponsor can start thinking through this methodology within their own protocols and applying some of those mindset shifts. TransCelerate plans to put together materials for different frameworks and approaches, but that is still underway.

What are some things sponsors can do now to begin reflecting on their data collection practices?

On the TransCelerate website, there are six different ICH guidances plus the Declaration of Helsinki. So, if you look through those six, each one may resonate with a different department within the sponsor or CRO.

As teams are developing their protocols, they should think about whether they might want to include a particular procedure again. As you take a protocol from Phase 1 to Phase 2, you learn something. Once you have that maturity, you may adjust your focus for Phase 3. Being mindful of those guidelines and the methodology that was shared in the manuscript, they can apply the critical thinking steps along the way.

As I mentioned before, the specifics will differ for each indication, each compound, and the maturity of both.  As an example, if there is a theoretical risk of X, this may be an area of focus in Phase 1 and maybe even Phase 2 studies. As data is accumulated, the theoretical risk may be reduced or eliminated.  The Phase 3 studies may reduce the frequency or stop conducting the specific procedure to look for X, but still watch for it through medical monitoring or adverse event collection. 

Do you think sponsors reading this data will be more inclined to move on them?

One of the elements that's at the forefront of many of these ICH guidances is to reduce the unnecessary burden on sites and patients. In addition, you also see many of the guidances talking about having that input from sites and patients up front. These things together are now ICH guidances. It is a shift in the research paradigm to be more focused on the impact on the patient, and having material examples in the paper will really be the catalyst for this change.

About The Expert:

Laura Galuchie is senior director, global clinical development, TransCelerate Program Lead, oversight committee, at Merck & Co., Inc. Laura has a unique role which blends internal and external components. Laura is Merck’s assigned leadership to the Oversight Committee of TransCelerate BioPharma, Inc. The oversight committee is the primary governing body for decisions related to the operational direction of TransCelerate, including project direction and expected outcomes. Laura draws on her experience as director of clinical operations for a busy therapeutic area and head of clinical performance, analytics, and innovation to provide this perspective. Within Merck, she has responsibility for driving internal engagement with key stakeholders. She facilitates adoption and internalization of TransCelerate solutions in alignment with Merck's objectives. Finally, she watches for opportunities to match innovative approaches to potential process improvement initiatives or other areas of focus.