Latest / Let's Talk Risk! Podcast / Deep Dive: When FDA Qualifies the Ruler

Transcript

Deep Dive: When FDA Qualifies the Ruler · Let's Talk Risk! Podcast

• 0:00
  SPEAKER_01: Imagine spending like $50 million in five years developing
• 0:04
  SPEAKER_01: a revolutionary cardiovascular implant, right?
• 0:06
  SPEAKER_01: And you finally submit thousands of pages of clinical trial data to the FDA
• 0:11
  SPEAKER_01: and everything just stalls.
• 0:13
  SPEAKER_00: Yeah.
• 0:13
  SPEAKER_00: And it's usually not because your device failed.
• 0:15
  SPEAKER_00: It's because the reviewers,
• 0:17
  SPEAKER_00: well, they don't trust the ruler you use to measure your own results.
• 0:20
  SPEAKER_01: Exactly.
• 0:20
  SPEAKER_01: So listen, the absolute bottom line for you today is this.
• 0:24
  SPEAKER_01: The FDA Medical Device Development Tools Program,
• 0:27
  SPEAKER_01: or MDTT, Is basically the ultimate shortcut to market.
• 0:32
  SPEAKER_01: It fundamentally accelerates R&D and turns what used
• 0:36
  SPEAKER_01: to be agonizing regulatory unpredictability into this massive strategic advantage.
• 0:41
  SPEAKER_00: It really does solve one of the most frustrating inefficiencies in regulatory science.
• 0:44
  SPEAKER_00: But, you know, we have to be clear right up front.
• 0:47
  SPEAKER_00: Speed in this industry,
• 0:48
  SPEAKER_00: it only actually happens when it's backed
• 0:49
  SPEAKER_00: by bulletproof pre-qualified scientific evidence.
• 0:52
  SPEAKER_01: Right.
• 0:53
  SPEAKER_00: And that is the entire engine driving this program.
• 0:57
  SPEAKER_01: Absolutely.
• 0:57
  SPEAKER_01: Welcome to this deep dive.
• 0:59
  SPEAKER_01: If you are developing a medical device or maybe prepping for an R&D strategy meeting
• 1:04
  SPEAKER_01: or just looking for ways to capture market share faster,
• 1:07
  SPEAKER_01: you are in the exact right place.
• 1:08
  SPEAKER_00: Yeah, we're really glad you're here.
• 1:10
  SPEAKER_01: Our mission today is to decode the official FDA documentation on the MDDT program.
• 1:16
  SPEAKER_01: And we're pulling directly from their latest qualification updates through early 2026.
• 1:21
  SPEAKER_01: We're going to break down exactly how you can leverage this framework
• 1:24
  SPEAKER_01: for unprecedented speed to market and honestly a massive return on investment.
• 1:29
  SPEAKER_00: Right.
• 1:29
  SPEAKER_00: And to truly appreciate that strategic value you just mentioned,
• 1:32
  SPEAKER_00: we need to look at the historical bottleneck the FDA is trying to fix.
• 1:36
  SPEAKER_00: I mean, we need to look at the era before the MDDP program even existed.
• 1:40
  SPEAKER_01: Set that scene for us.
• 1:41
  SPEAKER_01: Walk us through the mechanics of that bottleneck.
• 1:43
  SPEAKER_01: Like how bad was it?
• 1:44
  SPEAKER_00: Well, before this program,
• 1:46
  SPEAKER_00: the FDA evaluated testing tools on a strict case-by-case basis
• 1:51
  SPEAKER_00: for every single medical device submission.
• 1:52
  SPEAKER_01: Wow.
• 1:53
  SPEAKER_01: Every single one.
• 1:54
  SPEAKER_00: Every single one.
• 1:55
  SPEAKER_00: Let's go back to your cardiovascular implant example.
• 1:58
  SPEAKER_00: To prove that implant works,
• 2:01
  SPEAKER_00: your engineers might use a highly specific computational fluid dynamics model
• 2:07
  SPEAKER_00: to simulate blood flow.
• 2:08
  SPEAKER_01: Right.
• 2:08
  SPEAKER_01: Because you have to generate empirical data to prove safety and efficacy.
• 2:12
  SPEAKER_00: Exactly.
• 2:13
  SPEAKER_00: Or maybe you use a specific biomarker test to check for inflammation.
• 2:18
  SPEAKER_00: But historically,
• 2:19
  SPEAKER_00: when you submitted that data, the FDA didn't just review your physical implant.
• 2:23
  SPEAKER_01: They reviewed the test itself.
• 2:25
  SPEAKER_00: Yeah, they had to exhaustively review and mathematically validate
• 2:28
  SPEAKER_00: the computational model you used to test the implant.
• 2:31
  SPEAKER_00: And they did this every single time for every single submission.
• 2:34
  SPEAKER_01: Wait, so if my biggest competitor submitted a device last week using
• 2:38
  SPEAKER_01: the exact same fluid dynamics model, they had to prove the model worked.
• 2:42
  SPEAKER_00: Yep.
• 2:43
  SPEAKER_00: They sure did.
• 2:43
  SPEAKER_01: And then when I submit this week,
• 2:44
  SPEAKER_01: I have to prove the exact same model works all over again from scratch.
• 2:48
  SPEAKER_00: You have to prove it all over again.
• 2:49
  SPEAKER_01: Yeah.
• 2:49
  SPEAKER_00: The industry was essentially paying to validate the exact same math over and
• 2:55
  SPEAKER_00: over again in these isolated silos.
• 2:57
  SPEAKER_01: That is wild.
• 2:58
  SPEAKER_01: So you aren't just defending your proprietary product.
• 3:01
  SPEAKER_01: You're spending millions defending the basic methodology used to measure the product.
• 3:05
  SPEAKER_00: Precisely.
• 3:06
  SPEAKER_00: And, you know,
• 3:07
  SPEAKER_00: it drained vast amounts of capital from device sponsors
• 3:10
  SPEAKER_00: and it completely overwhelmed FDA reviewers.
• 3:13
  SPEAKER_01: I can imagine.
• 3:14
  SPEAKER_00: So the MDDT program was explicitly designed to shatter that bottleneck.
• 3:19
  SPEAKER_00: It's a voluntary program allowing R&D teams to pre-qualify a testing tool for
• 3:24
  SPEAKER_00: a highly specific context of use.
• 3:27
  SPEAKER_01: Okay, let's unpack this because context of use,
• 3:29
  SPEAKER_01: that sounds like the linchpin of the whole strategy.
• 3:31
  SPEAKER_01: How does that functionally work?
• 3:33
  SPEAKER_00: So the context of use is essentially a statistical and clinical boundary.
• 3:37
  SPEAKER_00: It defines exactly what the tool measures,
• 3:40
  SPEAKER_00: the specific patient population it applies to,
• 3:42
  SPEAKER_00: and the specific environment where it's valid.
• 3:45
  SPEAKER_01: Ah, so it's a very defined box.
• 3:46
  SPEAKER_00: Very defined.
• 3:48
  SPEAKER_00: And once a tool is qualified within that strict boundary,
• 3:51
  SPEAKER_00: the FDA accepts the tool's results indefinitely.
• 3:54
  SPEAKER_00: They will not ask you to reconfirm its scientific suitability.
• 3:56
  SPEAKER_01: That's incredible.
• 3:57
  SPEAKER_01: The ruler is officially calibrated, locked in, and trusted.
• 4:00
  SPEAKER_00: Exactly.
• 4:01
  SPEAKER_00: You don't have to prove the ruler works anymore.
• 4:03
  SPEAKER_01: I love thinking about this mechanically.
• 4:04
  SPEAKER_01: It's like TSA pre-check for your regulatory data.
• 4:09
  SPEAKER_00: Oh, that's a great way to put it.
• 4:10
  SPEAKER_01: Right.
• 4:11
  SPEAKER_01: Like you voluntarily submit to the rigorous deep dive background check on your methodology once.
• 4:18
  SPEAKER_01: And because you did that heavy lifting up front,
• 4:20
  SPEAKER_01: you get to skip the longest line at the airport every single time you fly with
• 4:23
  SPEAKER_01: a new device.
• 4:24
  SPEAKER_00: Yeah, that's exactly what it is.
• 4:25
  SPEAKER_01: But let's look at the actual execution here.
• 4:27
  SPEAKER_01: Does this mean a device sponsor literally just copies and pastes
• 4:31
  SPEAKER_01: an MDDT submission number into their FDA paperwork and bypasses
• 4:36
  SPEAKER_01: that entire review phase?
• 4:37
  SPEAKER_00: If we connect this to the bigger picture, yes, the mechanics are truly that streamlined.
• 4:42
  SPEAKER_00: When you prepare your pre-market submission,
• 4:45
  SPEAKER_00: you don't include a 300-page appendix proving your computational model is valid.
• 4:50
  SPEAKER_01: You just use a form.
• 4:51
  SPEAKER_00: Yeah.
• 4:51
  SPEAKER_00: You use the standard FDA CDRH pre-market review submission cover sheet.
• 4:55
  SPEAKER_00: That's Form 3514.
• 4:58
  SPEAKER_00: And there's a specific section on there just to cite the MDDT submission number.
• 5:02
  SPEAKER_01: You literally just drop the code in.
• 5:04
  SPEAKER_00: You drop the code in.
• 5:05
  SPEAKER_00: That code links the reviewer directly to the SEBQ.
• 5:08
  SPEAKER_01: And SEBQ stands for...
• 5:10
  SPEAKER_00: The summary of evidence and basis of qualification.
• 5:13
  SPEAKER_00: The reviewer checks that you are using the tool within its approved context of use,
• 5:18
  SPEAKER_00: and if you are, they immediately move on to evaluating the actual device.
• 5:22
  SPEAKER_01: You just saved months of potential pushback.
• 5:25
  SPEAKER_00: Easily months, sometimes years.
• 5:28
  SPEAKER_01: That is a staggering reduction in friction.
• 5:31
  SPEAKER_01: So if we understand that pre-qualification is the strategic goal,
• 5:35
  SPEAKER_01: We really need to transition to what these tools actually look like.
• 5:38
  SPEAKER_00: The actual categories.
• 5:40
  SPEAKER_01: Because reading through the FDA guidelines, they don't just accept one type of evidence.
• 5:43
  SPEAKER_01: It looks like a progression,
• 5:45
  SPEAKER_01: like an evolution of how we measure human health and device performance.
• 5:48
  SPEAKER_00: It really is an evolution.
• 5:50
  SPEAKER_00: Right.
• 5:50
  SPEAKER_00: So measuring how a patient feels.
• 6:06
  SPEAKER_00: Exactly.
• 6:07
  SPEAKER_00: A COA describes or reflects how a person feels, functions, or survives.
• 6:12
  SPEAKER_00: It could be a survey filled out by a clinician,
• 6:15
  SPEAKER_00: an observational checklist used by a parent,
• 6:17
  SPEAKER_00: or, you know, a direct questionnaire filled out by the patient themselves.
• 6:21
  SPEAKER_01: Here's where it gets really interesting.
• 6:23
  SPEAKER_01: I look at a patient questionnaire and my executive brain immediately screams, subjective.
• 6:28
  SPEAKER_00: Of course.
• 6:29
  SPEAKER_00: That's a natural reaction.
• 6:30
  SPEAKER_01: How does the FDA actually qualify human feelings as hard, empirical, regulatory evidence?
• 6:37
  SPEAKER_01: Like how does a survey pass the rigor test of a federal health agency?
• 6:40
  SPEAKER_00: What's fascinating here is how the FDA relies on psychometrics
• 6:44
  SPEAKER_00: and massive statistical validation to build a mathematical container around
• 6:48
  SPEAKER_00: that subjectivity.
• 6:49
  SPEAKER_00: They aren't saying human feelings are perfectly objective.
• 6:52
  SPEAKER_01: Right.
• 6:52
  SPEAKER_01: They're just measuring them in a controlled way.
• 6:54
  SPEAKER_00: Let's look at the Kansas City Cardiomyopathy Questionnaire,
• 6:57
  SPEAKER_00: the KCCQ, which they qualified back in 2017.
• 7:00
  SPEAKER_01: Oh, yeah.
• 7:01
  SPEAKER_01: Massive win in the cardiology space.
• 7:03
  SPEAKER_01: I know Abbott used the KCCQ when they were working on the Triclip device.
• 7:06
  SPEAKER_00: Yes, exactly.
• 7:07
  SPEAKER_00: The KCCQ doesn't just ask, hey, does your heart feel better?
• 7:10
  SPEAKER_00: It uses an incredibly precise matrix of questions that are validated
• 7:14
  SPEAKER_00: against thousands of clinical outcomes.
• 7:16
  SPEAKER_01: Ah, so it's tied to real-world data.
• 7:19
  SPEAKER_00: Through exhaustive statistical modeling,
• 7:21
  SPEAKER_00: the creators prove that a specific change in
• 7:23
  SPEAKER_00: the patient's score perfectly correlates with a specific measurable change
• 7:28
  SPEAKER_00: in their actual heart failure status.
• 7:30
  SPEAKER_01: So they've proven the statistical variance.
• 7:32
  SPEAKER_01: If the score moves three points,
• 7:35
  SPEAKER_01: we know mathematically that the patient's physical capability has improved
• 7:38
  SPEAKER_01: by X amount.
• 7:39
  SPEAKER_00: Precisely.
• 7:40
  SPEAKER_00: And this is really expanding into new commercial realities.
• 7:43
  SPEAKER_00: Like in November 2024, they qualified the WoundQ instrument.
• 7:48
  SPEAKER_01: WoundQ?
• 7:49
  SPEAKER_00: Yeah.
• 7:50
  SPEAKER_00: This is a QOA designed to measure outcomes for individuals with non-healing chronic wounds.
• 7:54
  SPEAKER_01: Wow.
• 7:55
  SPEAKER_01: That is huge for the commercial space.
• 7:57
  SPEAKER_01: If you are developing, say, an advanced hydrocolloid dressing for diabetic ulcers...
• 8:02
  SPEAKER_01: You don't have to hire a team of statisticians to invent a new way
• 8:05
  SPEAKER_01: to prove your dressing improves a patient's daily life.
• 8:08
  SPEAKER_00: You don't?
• 8:08
  SPEAKER_01: You don't have to argue with the FDA about how to measure chronic pain.
• 8:12
  SPEAKER_01: You just license RuneQ, use it in your trial, and the FDA accepted the data.
• 8:17
  SPEAKER_00: You're using their approved ruler.
• 8:19
  SPEAKER_00: Which brings us to the next evolution of evidence, right?
• 8:22
  SPEAKER_00: Moving from subjective reporting to objective biological capture.
• 8:27
  SPEAKER_01: Okay, moving on to the hard biology.
• 8:29
  SPEAKER_00: This is the biomarker tests category or BT.
• 8:32
  SPEAKER_00: These are instruments or methods used to detect and measure a biomarker
• 8:37
  SPEAKER_00: to assess risk or predict treatment outcomes.
• 8:39
  SPEAKER_01: And this is where consumer tech is violently colliding with traditional medtech.
• 8:44
  SPEAKER_01: Look at the Apple atrial fibrillation history feature.
• 8:46
  SPEAKER_00: Yes, a prime example.
• 8:48
  SPEAKER_01: The FDA qualified that as a biomarker test in May 2024.
• 8:51
  SPEAKER_01: We're talking about an off-the-shelf consumer wearable being utilized as
• 8:55
  SPEAKER_01: a qualified regulatory tool for cardiology trials.
• 8:58
  SPEAKER_00: It's amazing.
• 8:59
  SPEAKER_00: But again, we have to look at the boundaries of the validation.
• 9:02
  SPEAKER_00: The FDA did not qualify the entire Apple Watch as a universal medical diagnostic tool.
• 9:06
  SPEAKER_01: Right, that would be insane.
• 9:08
  SPEAKER_00: That would be far too broad.
• 9:10
  SPEAKER_00: They qualified a highly specific software algorithm utilizing
• 9:14
  SPEAKER_00: the optical heart sensor to retrospectively estimate the burden
• 9:18
  SPEAKER_00: of atrial fibrillation.
• 9:20
  SPEAKER_00: And this is in a very specific demographic of users who already have a clinical diagnosis.
• 9:25
  SPEAKER_01: That distinction is critical for anyone building a strategy around this.
• 9:29
  SPEAKER_01: The rigor isn't in proving the device is flawless in all scenarios.
• 9:32
  SPEAKER_00: No, not at all.
• 9:33
  SPEAKER_01: The rigor isn't defining the exact narrow corridor where
• 9:37
  SPEAKER_01: the data is statistically indistinguishable from a clinical ECG.
• 9:41
  SPEAKER_01: If you stay in that corridor, the data is golden.
• 9:44
  SPEAKER_00: Exactly.
• 9:44
  SPEAKER_00: You have to stay in the corridor.
• 9:46
  SPEAKER_01: I also saw this in January 2026 with Moleculite DX, right?
• 9:49
  SPEAKER_00: Yes, the Moleculite DX.
• 9:51
  SPEAKER_00: It's a point-of-care imaging tool qualified as a biomarker test
• 9:54
  SPEAKER_00: for identifying bacterial loads in wound care.
• 9:57
  SPEAKER_01: Which is a game changer for real-time assessments.
• 9:59
  SPEAKER_00: Which perfectly sets up the third,
• 10:01
  SPEAKER_00: and arguably the most technically advanced category,
• 10:04
  SPEAKER_00: non-clinical assessment models, or NAMs.
• 10:06
  SPEAKER_01: NAMs.
• 10:07
  SPEAKER_01: Okay, tell us about these.
• 10:08
  SPEAKER_00: Well, if COAs are subjective and biomarker tests are biological,
• 10:13
  SPEAKER_00: NAMs are entirely structural and mathematical.
• 10:16
  SPEAKER_00: An NAM replaces in vivo testing, meaning it replaces testing on live humans or animals.
• 10:22
  SPEAKER_01: We are basically talking about virtual reality for regulatory science.
• 10:26
  SPEAKER_00: Essentially, yes.
• 10:27
  SPEAKER_00: These are in vitro models like highly complex tissue phantoms used
• 10:31
  SPEAKER_00: to evaluate imaging devices or sophisticated computational physics models.
• 10:36
  SPEAKER_01: Let's dive into the mechanics of this because the ROI
• 10:39
  SPEAKER_01: of replacing physical testing is staggering.
• 10:42
  SPEAKER_01: If you don't have to build a physical prototype,
• 10:44
  SPEAKER_01: ship it to a lab and test it in a live biological subject,
• 10:48
  SPEAKER_01: you are shaving years and millions of dollars off your burn rate.
• 10:51
  SPEAKER_00: Absolutely.
• 10:51
  SPEAKER_00: It's a massive shift.
• 10:53
  SPEAKER_01: Look at the ND Point Pneuma Screw Virtual Pullout Test, the one qualified in April 2025.
• 10:59
  SPEAKER_00: A perfect example.
• 11:00
  SPEAKER_00: Historically,
• 11:00
  SPEAKER_00: if you design a new orthopedic screw,
• 11:02
  SPEAKER_00: how did you prove to the FDA that it wouldn't rip out of the bone under stress?
• 11:06
  SPEAKER_01: You'd have to physically do it, right?
• 11:07
  SPEAKER_00: Yeah, you had to physically drill it into cadaver bone or animal bone
• 11:11
  SPEAKER_00: and physically rip it out with a testing machine measuring the force.
• 11:14
  SPEAKER_01: And the problem there is biological variance.
• 11:16
  SPEAKER_01: No two cadaver bones have the exact same density.
• 11:19
  SPEAKER_01: So your data is inherently messy from the start.
• 11:22
  SPEAKER_00: Exactly.
• 11:23
  SPEAKER_00: The Numa Screw tool is a finite element analysis model.
• 11:27
  SPEAKER_00: It is a computational simulation that mathematically maps the physics
• 11:31
  SPEAKER_00: of bone density, thread pitch, and torque.
• 11:34
  SPEAKER_01: So no physical bone required?
• 11:35
  SPEAKER_00: None.
• 11:36
  SPEAKER_00: You feed the CAD file of your new screw into the software and
• 11:40
  SPEAKER_00: the algorithm simulates the pullout force.
• 11:42
  SPEAKER_00: The FDA qualified this because the math is actually more consistent and reliable
• 11:46
  SPEAKER_00: for baseline safety testing than messy biological reality.
• 11:50
  SPEAKER_01: That is mind bending.
• 11:52
  SPEAKER_01: The digital simulation is trusted more than the physical test.
• 11:55
  SPEAKER_00: It really is.
• 11:56
  SPEAKER_01: We see the same thing with the virtual MRI safety evaluations qualified back in 2021.
• 12:01
  SPEAKER_01: If you design a new titanium hip implant,
• 12:03
  SPEAKER_01: you don't have to physically build it and stick it in a 1.5 Tesla MRI scanner
• 12:08
  SPEAKER_01: with thermal sensors to see if it heats up and burns the patient.
• 12:11
  SPEAKER_00: Nope.
• 12:11
  SPEAKER_00: You run it through a qualified computational electromagnetic model.
• 12:14
  SPEAKER_01: Incredible.
• 12:15
  SPEAKER_00: And just to round out the categories,
• 12:16
  SPEAKER_00: the FDA also maintains an other category for tools that don't fit those three boxes.
• 12:21
  SPEAKER_01: Right.
• 12:21
  SPEAKER_01: The catch all.
• 12:22
  SPEAKER_00: A great example there is the UCSF Lethal Arrhythmia Database, which was qualified in March 2024.
• 12:29
  SPEAKER_01: Right.
• 12:29
  SPEAKER_01: So instead of a simulation, it's a massive pre-qualified data set.
• 12:33
  SPEAKER_00: Exactly.
• 12:34
  SPEAKER_01: So if you build a new cardiac monitor,
• 12:36
  SPEAKER_01: you can test your software against their database of lethal arrhythmias,
• 12:40
  SPEAKER_01: and the FDA accepts that validation.
• 12:42
  SPEAKER_00: Because the database itself has already been rigorously vetted.
• 12:45
  SPEAKER_01: Okay.
• 12:46
  SPEAKER_01: So we've established that these tools,
• 12:48
  SPEAKER_01: whether they are psychometric surveys,
• 12:50
  SPEAKER_01: wearable algorithms,
• 12:51
  SPEAKER_01: or virtual physics simulations, they are absolute rocket fuel for R&D.
• 12:56
  SPEAKER_00: Rocket fuel is the perfect term.
• 12:58
  SPEAKER_01: But here's the critical operational pivot.
• 13:00
  SPEAKER_01: If I'm an executive at a midsize medtech company and we have
• 13:04
  SPEAKER_01: an incredible internal computational model we use for R&D,
• 13:08
  SPEAKER_01: how do we actually get it on this list?
• 13:10
  SPEAKER_00: The qualification process is highly structured into two phases,
• 13:13
  SPEAKER_00: the proposal phase and the qualification phase.
• 13:15
  SPEAKER_01: Well, let me play the skeptic for a second.
• 13:17
  SPEAKER_01: Doing the level of statistical validation we just talked about,
• 13:20
  SPEAKER_01: like proving a digital screw reacts exactly like a physical screw,
• 13:24
  SPEAKER_01: that has to be incredibly resource intensive.
• 13:26
  SPEAKER_00: Oh, it is.
• 13:27
  SPEAKER_00: Very much so.
• 13:27
  SPEAKER_01: What is the catch here?
• 13:28
  SPEAKER_01: I'm assuming the FDA user fees just to review these massive data packages are astronomical.
• 13:35
  SPEAKER_00: That is actually the most shocking aspect of the MDDT program.
• 13:38
  SPEAKER_01: Yeah.
• 13:38
  SPEAKER_00: There are absolutely zero user fees.
• 13:41
  SPEAKER_01: Wait, really?
• 13:42
  SPEAKER_01: Zero.
• 13:43
  SPEAKER_00: Zero.
• 13:43
  SPEAKER_00: The FDA does not charge you a single dollar to review your proposal
• 13:48
  SPEAKER_00: or your final qualification package.
• 13:50
  SPEAKER_01: In regulatory affairs,
• 13:51
  SPEAKER_01: zero fees usually means a massive catch,
• 13:54
  SPEAKER_01: like your submission goes to the bottom of the three-year backlog.
• 13:58
  SPEAKER_01: What is the reality here?
• 13:59
  SPEAKER_01: Why is the FDA doing this for free?
• 14:02
  SPEAKER_00: Because they desperately need the industry's help.
• 14:04
  SPEAKER_01: Oh, I see.
• 14:05
  SPEAKER_00: The FDA recognizes that regulatory science is evolving faster
• 14:08
  SPEAKER_00: than their internal budget can handle.
• 14:10
  SPEAKER_00: Okay, phase one, the proposal phase, is all about risk mitigation.
• 14:16
  SPEAKER_00: You don't run the expensive data collection yet.
• 14:18
  SPEAKER_01: Smart.
• 14:19
  SPEAKER_00: You submit a package describing your tool,
• 14:21
  SPEAKER_00: proposing the exact context of use,
• 14:23
  SPEAKER_00: and laying out a detailed plan for how you intend to validate it.
• 14:26
  SPEAKER_01: You are agreeing on where the goal posts are before you start spending money on the trial.
• 14:32
  SPEAKER_00: Exactly.
• 14:32
  SPEAKER_00: The FDA reviews the methodology.
• 14:35
  SPEAKER_00: If they agree, they accept you into the program.
• 14:37
  SPEAKER_01: And that triggers phase two.
• 14:38
  SPEAKER_00: Yes, that triggers phase two, the qualification phase.
• 14:40
  SPEAKER_00: This is where you actually execute the agreed upon plan,
• 14:44
  SPEAKER_00: run the simulations or clinical surveys, and submit the hard data.
• 14:48
  SPEAKER_00: If the data holds up,
• 14:51
  SPEAKER_00: Okay, so
• 14:54
  SPEAKER_01: the FDA review is free, but the R&D to generate that hard data still costs money.
• 14:59
  SPEAKER_00: It definitely does.
• 15:00
  SPEAKER_01: But reading through the documentation,
• 15:02
  SPEAKER_01: there is a massive strategic advantage here regarding funding.
• 15:05
  SPEAKER_00: Yes, there is.
• 15:06
  SPEAKER_01: The FDA's Center for Devices and Radiological Health,
• 15:09
  SPEAKER_01: the CDRH,
• 15:10
  SPEAKER_01: actively promotes their Small Business Innovation Research,
• 15:14
  SPEAKER_01: or SBIR, grant program, specifically to fund the creation of these tools.
• 15:19
  SPEAKER_00: Yes.
• 15:20
  SPEAKER_00: And their definition of a small business is quite broad.
• 15:23
  SPEAKER_00: It covers domestic businesses with gross receipts under $100 million.
• 15:27
  SPEAKER_01: One hundred million.
• 15:29
  SPEAKER_01: That captures a massive portion of the MedTech startup ecosystem.
• 15:34
  SPEAKER_00: It really does.
• 15:35
  SPEAKER_01: So if you are under $100 million in revenue,
• 15:38
  SPEAKER_01: the government will literally grant you the capital to build a tool that
• 15:41
  SPEAKER_01: the FDA will then review for free.
• 15:43
  SPEAKER_00: And historically,
• 15:44
  SPEAKER_00: the FDA partners with major branches of the National Institutes of Health
• 15:48
  SPEAKER_00: to pool this grant money.
• 15:49
  SPEAKER_01: Oh, wow.
• 15:50
  SPEAKER_01: Like who?
• 15:51
  SPEAKER_00: Well, they've partnered with the National Cancer Institute to fund AI
• 15:54
  SPEAKER_00: and machine learning tools in oncology.
• 15:56
  SPEAKER_00: They've worked with the National Institute of Dental and Craniofacial Research
• 16:00
  SPEAKER_00: for novel biomaterial tools,
• 16:02
  SPEAKER_00: and even the National Institute on Drug Abuse for Digital Therapeutics.
• 16:06
  SPEAKER_01: This raises a massive strategic point.
• 16:07
  SPEAKER_01: Let's look at the competitive chessboard for a minute.
• 16:10
  SPEAKER_01: If I am a smart startup,
• 16:11
  SPEAKER_01: I secure an SBIR grant to build a proprietary non-clinical assessment model.
• 16:18
  SPEAKER_01: The government pays for the R&D.
• 16:19
  SPEAKER_00: Right.
• 16:20
  SPEAKER_01: The FDA reviews it for free and qualifies it.
• 16:23
  SPEAKER_01: Now, I own a calibrated ruler.
• 16:25
  SPEAKER_01: I can use it to push my own pipeline of devices through the FDA at warp speed.
• 16:30
  SPEAKER_01: But more importantly, well, I own a tool that my competitors desperately want to use.
• 16:36
  SPEAKER_00: This raises an important question about industry leverage.
• 16:39
  SPEAKER_00: Because you own that qualified tool and your SEQ is publicly listed on
• 16:44
  SPEAKER_00: the FDA website with your contact information, You can commercialize it.
• 16:48
  SPEAKER_01: You can license it out.
• 16:49
  SPEAKER_00: You can license the use of your computational model to your direct competitors.
• 16:52
  SPEAKER_01: You essentially transition from being a device manufacturer struggling
• 16:56
  SPEAKER_01: against regulatory friction to becoming a toll booth on
• 16:59
  SPEAKER_01: the regulatory highway itself.
• 17:00
  SPEAKER_00: The toll booth, yes.
• 17:01
  SPEAKER_01: That is a phenomenal competitive moat.
• 17:03
  SPEAKER_01: But we have to be intellectually honest about the risks here, too.
• 17:06
  SPEAKER_01: What happens if a company licenses someone else's MDDT,
• 17:10
  SPEAKER_01: but their new device slightly pushes the boundary of that tool's approved context
• 17:15
  SPEAKER_01: of use?
• 17:16
  SPEAKER_00: Ah, that is the primary risk factor.
• 17:19
  SPEAKER_00: The FDA is completely rigid on the context of use.
• 17:23
  SPEAKER_00: If you license the Apple AFib feature,
• 17:26
  SPEAKER_00: but you try to use it to validate a device intended for pediatric patients,
• 17:30
  SPEAKER_00: the FDA will reject it immediately.
• 17:32
  SPEAKER_01: Because it wasn't approved for kids.
• 17:34
  SPEAKER_00: Exactly.
• 17:35
  SPEAKER_00: Because the tool was only qualified for adults with a prior diagnosis.
• 17:39
  SPEAKER_00: The moment you step outside the mathematical container, the tool loses all its regulatory power.
• 17:44
  SPEAKER_01: You are back to square one defending the ruler.
• 17:46
  SPEAKER_00: You're starting completely from scratch.
• 17:48
  SPEAKER_01: Which means your regulatory strategy team has to be ruthlessly precise
• 17:53
  SPEAKER_01: when matching your product's intended use with a tool's context of use.
• 17:56
  SPEAKER_00: Exactly.
• 17:57
  SPEAKER_00: It requires high-level strategic alignment between your engineering team
• 18:00
  SPEAKER_00: and your regulatory affairs team from day one of development.
• 18:03
  SPEAKER_01: Day one.
• 18:04
  SPEAKER_01: All right.
• 18:04
  SPEAKER_01: Let's bring this all together for you,
• 18:06
  SPEAKER_01: the listener, so you can take this directly into your next strategy meeting.
• 18:09
  SPEAKER_00: Let's do it.
• 18:10
  SPEAKER_01: The MDDT program is a fundamental shift in how you should approach R&D.
• 18:14
  SPEAKER_01: We are moving away from the custom-built,
• 18:16
  SPEAKER_01: agonizingly slow process where you have to prove your testing methods
• 18:20
  SPEAKER_01: from scratch every single time.
• 18:21
  SPEAKER_00: You are stepping onto a prefab high-speed highway where
• 18:24
  SPEAKER_00: the tools are already verified and the ruler is already calibrated.
• 18:28
  SPEAKER_01: Whether you are leveraging virtual physics models to entirely skip live testing,
• 18:33
  SPEAKER_01: utilizing consumer technology like the Apple Watch
• 18:35
  SPEAKER_01: to gather qualified biomarker data,
• 18:38
  SPEAKER_01: or securing SBIR grants to build your own proprietary tool and create
• 18:42
  SPEAKER_01: a competitive moat, Well, the speed to market is undeniable.
• 18:46
  SPEAKER_00: You are actively removing friction from the most expensive part of device development.
• 18:50
  SPEAKER_01: Exactly.
• 18:51
  SPEAKER_00: In a landscape where regulatory complexity is only increasing,
• 18:55
  SPEAKER_00: understanding these pre-qualified pathways is no longer just a nice to have.
• 18:59
  SPEAKER_00: It is a critical requirement.
• 19:00
  SPEAKER_00: You can walk into your next R&D meeting knowing exactly how to bypass
• 19:04
  SPEAKER_00: the validation bottlenecks that are currently trapping your competitors.
• 19:07
  SPEAKER_01: You walk in with the strategic cheat code.
• 19:10
  SPEAKER_01: But I want to leave you with one final thought to explore on your own.
• 19:13
  SPEAKER_01: We've traced the evolution of these tools today,
• 19:16
  SPEAKER_01: moving from subjective patient surveys to digital biomarkers,
• 19:20
  SPEAKER_01: and finally to advanced virtual computational models.
• 19:23
  SPEAKER_01: If the FDA is increasingly trusting finite element analysis to predict how
• 19:27
  SPEAKER_01: a screw interacts with bone and electromagnetic models
• 19:30
  SPEAKER_01: to predict how titanium reacts in an MRI,
• 19:33
  SPEAKER_01: how long will it be until a complex medical device is entirely tested,
• 19:37
  SPEAKER_01: mathematically validated,
• 19:38
  SPEAKER_01: and fully cleared for human use without ever having touched a physical prototype?
• 19:42
  SPEAKER_01: Think about that.