Britney Spears' Take On Building Platform Analytics for mRNA Therapeutics

By Anna Rose Welch, Editorial & Community Director, Advancing RNA

Seeing as this article is being published right around April Fool’s Day, it would make sense you may have clicked on this article expecting to be met with an (albeit slightly belated) April Fool’s Joke.

As it turns out, you’re (kind of, sort of) in luck.

To be clear, this article itself is not a joke; it’s based off of very real takeaways from an incredibly important (and serious) virtual summit US Pharmacopeia hosted a few weeks ago: “Joining Forces To Advance the Quality of mRNA Therapeutics.” However, I daresay even something as serious as the analytical development of mRNA-LNPs deserves some levity. And as it turns out, the Internet had just the knock-knock joke for the occasion.

Knock, knock.
Who’s there?
Britney.
Britney who?
Knock, knock.
Who’s there?
"Oops!...I Did It Again."

If it’s been a while since you last heard Britney Spears’ classic “anthem” “Oops… I did it again,” perhaps treat yourself to another listen.

Regardless of whether you’re a Britney Spears fan or not, however, the punchline of this joke/the title of the song felt strangely appropriate for an article on RNA analytical development. After all, our biggest long-term goal is to approach the FDA and confidently demonstrate both the reproducibility of our analytical procedures and our manufacturing processes. (By the way, don’t let Spears’ “Oops” fool you; she knew just what she was doing.)

To be clear, we still have a long way to go until we have the confidence in our analytical methods to achieve such reproducibility and a true analytical platform (i.e., one comprising a suite of what ICH Q(2) (R2) defines as platform analytical procedures). But as many of the presentations throughout the two-day USP quality forum revealed, we’re doing what we must to understand the benefits and the limitations of our methods. Yes, this is painstaking and expensive work. But understanding our methods is also one of the critical steps toward a future in which we can rely upon platform analytical procedures and — channeling our inner Britney Spears — say “[Ooo], I did it again!” to the regulatory agencies for each of the subsequent molecules in our pipelines.  

From “Snapshot” To “Story”: The Path To Platform Analytics

For those of you interested in unpacking some of the more granular learnings shared about individual methods, I’d highly encourage you to check out the USP event, which will be available to stream here (eventually). However, as we continue to turn our attention toward achieving the efficiency that platform processes promise us, there was one presentation by an expert at BioNTech that I felt was exemplary of the work we must (continue) to do to achieve the necessary level of confidence in our analytical conclusions.

I once heard an FDA official make the lovely comparison that our iterative CMC efforts are akin to the work we do as humans to consistently improve ourselves. I was reminded of this comparison again while listening to the BioNTech expert’s high-level analysis of ICH Q(14) and ICH(Q2). As the expert explained, validation of an analytical method, for example, has evolved from simply following a “check-box approach” providing a “snapshot of a singular event in time” to a more holistic, “knowledge-driven validation strategy” that evolves as we iteratively garner more data throughout a product’s development.  

“We are always talking about data-driven decisions,” he explained. “How do we identify risks? It comes back to knowledge. Without knowledge, we don’t know what the risk is, and we cannot identify it” nor can we adequately rank/prioritize the greatest risks to our development program. In turn, we find ourselves unable to properly manage and mitigate those risks.  

This foundational knowledge takes many different forms. In the earliest stages of development, we often talk about the importance of “unbridled” (ish…) characterization to help us understand all the facets of our molecule’s structure. But as we were also reminded by the USP event itself: The more information we can gather about an analytical method and its performance early, the more confidently we can accept the information that method provides as well as pinpoint the source of any deviations in the future.

Practice makes perfect, and in many ways, this cliched phrase applies just as much to the work we do in analytical development as it does to stepping up to the microphone for Thursday night karaoke. Not only does performing the same analytical procedures repetitively promote constant data generation and continuous improvement of methods, but it also makes the work we do with each method more akin to muscle memory.

“Doing the same procedure over and over again ultimately improves the team’s understanding of the method, and, in turn, establishes a foundation for more efficient development and QC of subsequent products,” the expert added.

Likewise, much of the painstaking analytical work we’re doing today — and will continue to do throughout the lifecycle of our product — is essential for building up our prior knowledge. As the BioNTech expert continued, to garner this essential prior knowledge, there must be a symbiotic relationship between development and quality control teams.

“Development work needs to be much more closely connected to QC” to create the holistic, end-to-end approach ICH Q(2) and ICH Q(14) emphasize, he added. “Routine testing and the increasing knowledge we garner over time from these tests serves as feedback for development, adding to our overall prior knowledge. This feedback loop between development and QC teams ensures constant knowledge generation and improvement of our understanding of each product.” Not to mention, such close collaboration promises accelerated development of subsequent products using the hoped-for platform approach.

Though the platform approach is our desired future, we can’t and shouldn’t underestimate the work we still need to do to arrive at this future. It goes without saying, the last thing we ever want to say is “oops!” when it comes to our drug development — especially once we arrive at what we thought would be a platform. After all, as the expert went on to explain, a true platform approach will affect multiple products — not just one.

“Establishing platform procedures require Herculean effort, and huge data sets must be generated upfront to reliably put a platform in place,” he concluded. “It’s important we understand that the platform approach is an investment in the future, but to reach that future, we need to put a lot of effort into generating this knowledge and managing this data set. Doing so will allow us to speed up subsequent product development.”