Where New Medical Devices Come From
Mapping the US FDA 510(k) and Pre-Market Approval Landscape
With Bioeconomy.XYZ this year I’ve been building a story about the United State’s medical device innovation pipeline.
The conversation started with a look at where NIH SBIR grant funding — one of the earliest types of funding for a new company — flows geographically across the nation.
America’s Biggest Biotech Angel
The NIH’s Billion Dollar Annual Boost to US Biotech Ecosystems
The follow-on piece traced medical device start-ups from ‘first money in,’ and considered in detail the discrepancy between where Angel-funding was being deployed and where formal venture capital investments were being made.
Mapping the Medical Device Start-Up Ecosystem
And identifying the niches where these companies most often thrive
The transition from Angel to institutional funding is a key step (but not quite an absolute requirement) in the evolution of a start-up bringing a new device to market. Broadening investments accompany deepening organizational sophistication, as start-ups more and more take on the business and regulatory trappings of larger device companies. However, even the best medical devices proposed by the best founding teams must cross the most critical governing threshold before reaching the market.
Regulatory approval, and the last step to market
In this essay, the thread continues. An essential step each new technology must take before impacting patients’ lives is regulatory sign-off. For medical devices in the United States, that clearance comes from the Food and Drug Administration’s Center for Devices and Radiological Health. For all but the simplest medical devices — like electric toothbrushes (but by no means to diminish their elegance) — some form of FDA evaluation is required.
A note about data sourcing: In this report, I draw extensively from the data collections at OpenFDA. These files report all devices receiving some kind of affirmative pre-market evaluation. That is, a good deal of information is provided about devices and their associated company when approval is given. OpenFDA does not provide item-by-item details for those devices that do not receive the go-ahead for clinical deployment. By necessity, my thoughts here apply chiefly to devices which have successfully navigated the FDA. I’ve provided more detailed notes about data and analysis at the foot of this essay.
Go-to-market clearance in the US proceeds through three general pathways: 510(k) Premarket Notification, Pre-Market Approval, and De Novo approval. Table 1 summarizes some of the formal differences between these regulatory pathways.
A simple way to approach the nuances in these classifications is to consider how the FDA approaches a new device’s evaluation. The central questions the agency will ask are:
1. Is this device similar to a device the FDA has been presented before, and
2. Is the clinical application for this device one in which patients’ well-being or even life might be at serious risk?
‘Familiarity’ — that is, a device’s similarity to previously reviewed devices — conveys not only an understanding of a technology’s likely use, benefits, and hazards, but also which aspects of its manufacturing, quality system, and post-market experience will need to be monitored. In general, devices that are similar to previously cleared devices are evaluated through the 510(k) process. The FDA has a specific term for new items meeting this criterion: ‘substantial equivalence.’
‘Criticality’ — that is, the likelihood that a device will have a direct role in saving or sustaining a patient’s life — and ‘Novelty’ — that is, the extent to which a device is unfamiliar — when present may prompt the FDA to pursue the other main evaluation path: pre-market approval, or PMA. Some of the devices requiring PMA might surprise you. They’re not all tools like deep brain stimulators or artificial organs. Diagnostic laboratory instruments whose outputs will inform life-or-death decisions also receive PMA assessment.
Some devices may be deemed a combination of ‘substantially equivalent,’ or 510(k)-like, yet need the type of assessment given to devices seeking pre-market approval. The ‘De Novo’ classification is used for new devices that are both unfamiliar and are expected to behave in such a way that usual safeguards and monitoring methods will be appropriate.
Between Q1 2018 and Q3 2021, the CDRH awarded 5,601 ‘501(k) clearances.’ That’s almost 6 each business day. During the same three-and-three-quarter years, only 86 devices received Pre-Market Approval. 510(k) clearances outnumbered PMA’s 65-to-1.
A big investment in time and money
Companies might spend years developing a new medical device and preparing the documents requested by the FDA for pre-market assessment. For start-ups, a single application for a single device may be a company’s raison d'être, sustained sometimes for years and through several rounds of financing. Large public companies may submit dozens of applications annually. Regardless of the device, all-in development costs can easily exceed $1m. For complex devices with life-sustaining indications, the sky may be the limit for development costs.
For completeness, the submission fees required by the FDA for device evaluation are included above in Figure 1. These can be quite substantial for smaller companies, but likely pale in comparison to total development costs.
Reviewing the OpenFDA experience over the past 3 3/4 years (in Figure 1) suggests that companies submitting 510(k) pre-market notifications can expect notice of clearance — assuming it’s granted — in less than 12 months. While many Pre-Market Approval applications will be granted within one year, there’s a fairly long tail in this distribution.
Two years in the life of a global medical device company may not seem like much. Eighteen to twenty-four months can be a lifetime, or even more than a lifetime, for a start-up. Delays in receiving FDA clearance may be enough to empty the coffers and trigger an unplanned round of investment, an event that has profound consequences on the value of the company and on the amount of equity that needs to be surrendered by the founding team.
‘Familiar’ is not the same as duplicative.
It’s important for me to point out that just because a new device receives 510(k) clearance, that doesn’t mean it’s a knock-off of another commercially available product. Far from it. Given the lower costs in time and money of the 510(k) pathway, companies smartly may look for ways of drawing regulatory similarities between their technologies and other post-market devices. Remember: This is a matter of regulatory approval and the establishment of standards and quality systems around a device. 510(k) clearance in no way suggests degree of market differentiation or a technology’s promise in addressing an unmet need.
The Types of Devices Being Commercialized
The CDRH has almost two dozen device specializations that fall largely along the lines of medical subspecialty (see Figure 2). By nature of the differences between Class II and Class III devices, there’s a strong difference in medical specialties represented in the 510(k) and PMA pathways. Among procedural specialties, orthopaedic surgery outnumbers all others in terms of 510(k) approvals annually. In contrast, the largest share of new PMA notifications come from the cardiovascular space — where inventions like new blood pumps, pacemakers, and artificial valves all have immediate life-sustaining roles — and among clinical laboratory diagnostic applications using new sequencing and other molecular methods.
The types of companies developing new devices
Any company registered with the Food and Drug Administration can apply for commercial go-ahead. But given the differences in complexity and cost between developing a device intended for a 510(k) clearance and a device destined for a full Pre-Market Approval, different types of companies unavoidably dominate the two spaces.
Figure 3 plots the distribution of companies receiving notices through the 510(k) and PMA pathways. Here I use the following inexact definitions of company type:
- privately-held companies without evidence of receiving prior venture investment (‘Private’),
- venture-backed privately-held companies (‘Venture’),
- private equity-backed or owned companies (“Private Equity”),
- public companies,
- their subsidiaries, and
- non-profit organizations.
Note: In the study time interval, 1,999 companies received 510(k) clearance of a medical device. Therefore, my analysis of companies receiving these notifications is based on a random sample of the 5,601 notifications. In contrast, the PMA analysis includes all devices receiving that approval.
Devices submitted from public companies and their subsidiaries were the most common to receive FDA clearance of any type. However, there was much greater diversity in business type among applicants receiving 510(k) notifications. One-half of all awards were to private or venture-backed companies.
This was not the case with approvals of the more complex and expensive devices in the PMA pathway. Public companies and their subsidiaries received 75% of these approvals. Only 5% of companies receiving a PMA between 2018 and 2021 were privately held companies not on record as receiving outside investment.
The Geographic Medical Device Economy and the Importance of Location
For those of you who’ve been reading my other pieces this year, we are probably in agreement that where devices come from is important. The efficient creation of new devices relies on readily available facilities, talent (both within a company and among its consultants), and capital. Thus for medical devices, the economy of agglomeration is a real thing.
In Figure 4, I’ve plotted the past several years’ 510(k) and PMA awards by city. Awards of 510(k) approvals strongly correspond geographically with previous maps I’ve shown of NIH funding and early-stage investment. That is, there is broad regional participation in new device creation, with less activity in the Midwest and sparsity of activity in the Mountain West.
Pre-Market Approval is a much rarer phenomenon, and there are only a few geographic centers in the United States participating (right panel, Figure 4).
I’ve previously shown data suggesting the importance of location to biotech activity by looking at the distance between companies receiving NIH start-up grants and the nearest university.
Research schools are strong regional anchors for new science and technology start-ups, and most NIH small business awards are within kilometers of such institutions. In Figure 5, I show a similar analysis of the correlation between where companies receiving 510(k) clearances are located and where Pre-Market Approvals are granted. The majority of 510(k) clearances are given to companies within 100 km of a company commercializing devices receiving Pre-Market Approvals. My interpretation is that the receipt of a PMA is a useful surrogate for a region’s abundance of talent, facilities, and capital needed to foster medical device companies. As with NIH funding and universities, the data suggest tight clustering of companies around localities where complex medical devices are being created.
Seeing these data through an early-stage investment lens
To size up an early-stage medical device investment is to grapple with the many risks that can upend or even bring down a start-up. Often the most glaring risks are technical and in the founders’ wheelhouse — will the device actually work, can it be made? Other risks are very difficult to do much about — is a superior technology in stealth mode, are there large market factors that will negate a new device’s value proposition?
But other risks are more manageable and often involve providing a start-up with essential ingredients for growth. These ingredients can be an important contribution by non-technical investors and include access to talent, facilities and shared equipment, and capital.
I don’t believe that where on the planet someone invents a new device is critically important.
However, where a founding team decides to build a company around an invention is a different matter. My three recent pieces have focused on this topic — where the NIH puts its earliest resources, where Angel and Venture investors place their bets, and the locations from where new FDA approvals arrive.
Although the data are imperfect, and causality not guaranteed, the evidence supports a theme that medical device companies sprout more often and grow more effectively in communities and regions where other device companies are doing likewise. I would argue that these regions become hyper-fertile in terms of talent, facilities, and cash and by extension become generators of early-stage deal flow. Experienced early-stage investors know in which streams to drop their hooks for the best fishing. That predilection for some cities is reflected in the variation in funding advancement from Angel to institution investment.
I don’t have an opinion whether regional concentration is ‘good’ or ‘bad,’ although, from a macroeconomic and national strategic perspective, diversity of all types leads to more robust systems. But from the perspective of attracting early investment, I believe founding teams positioned far from the usual locations will need to go the extra mile to show how their remoteness either is either a strategic advantage or, more likely, is a disadvantage for which they have a corrective plan in place.
ArgoPond is forever upping its game, and our team is interested in questions and comments that arise among readers and what types of analysis you’d like to see in future research notes. Get the conversation started and leave comments below. You can also reach out to us directly via firstname.lastname@example.org.
Notes on data sources and modeling
- The data stack in this note includes OpenFDA, the Google maps geocoding API, and geographic-specific information from the US Census and the Federal Communications Commission. Contact me for details regarding the query structure, working with the json data format at the FDA site, and the fuzzy and brute force methods used to identify and correct some of the anomalies within the FDA data.
- Here is the main FDA page linking to their data sets.
About the Author
A physician-scientist by training, John Younger is the Managing Director at ArgoPond, LLC, a life science advisory and investment company that provides analytics and diligence services to companies and funds engaging the life science space. John sits on the Board of Directors of New View Surgical, a Boston-based start-up creating new visualization tools for minimally invasive surgery. He has served for years as an advisor to the NIH and has provided congressional testimony on how best to support early-stage biotech companies. That testimony can be seen here. He represents ArgoPond as a member of the Life Science Committee of New York Angels, one of the most active angel investment groups in the world.
John’s LinkedIn profile and contact information are here.