Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Service Provider Executive Series #17: Omar Serang of DNAnexus
32 Plays3 days ago
In this episode, host Steve Gens is joined by Omar Serang (EVP and GM of DNAnexus) to explore the evolution of trusted regulatory spaces. The conversation covers PrecisionFDA, Project PRISM, protocol digitalization, and the emerging shift toward faster, more collaborative global regulatory processes.
Learn more about DNAnexus through their website: https://www.dnanexus.com/
Recommended
Transcript
Introduction of Podcast and Guest
00:00:06
Speaker
Welcome to the Gens and Associates Regulatory Executive Podcast Series, where we explore innovation in the regulatory space. I'm your host, Steve Gens, and today I have Omar Sarang of DNA Nexus. So welcome, Omar. I think when we were prepping for this, the last time we were on stage out in Denver for the RENs Conference and had the benefit of each of us presenting. So I Been looking forward to this conversation because what you and your firm are doing are are very innovative. So before we get started, why don't you introduce yourself and the the firm to our listeners and then we'll get going.
00:00:44
Speaker
Certainly. Thanks, Steve. It was great to see you in Denver. And a lot of things have happened, even in the short time frame since then. I'm very happy to catch up on it. I'm Omar Sarang. I'm the executive vice president and general manager for DNA Nexus Regulatory and Government Solutions.
00:01:01
Speaker
And my claim to fame really is I'm the architect of Precision FDA. And for the last decade, that has been a large part of my life is working with the FDA on Precision FDA. And that is a big part of what we do.
00:01:14
Speaker
Prior to coming to DNA Nexus, 13 years ago it's been, I ran EC2 operations for AWS. So I've had a very strong grounding in the cloud and very strong grounding in security and compliance, which as we'll discuss today is really foundational for the innovation that we're accomplishing.
00:01:35
Speaker
Great, and welcome again. So probably a good place to get started for our listeners.
DNA Nexus's Cloud-Based Innovation Journey
00:01:40
Speaker
DNA Nexus was the first of what we're starting to call, or calling this year, of the cloud-based regulatory spaces.
00:01:47
Speaker
So I think it'll be really, really beneficial about sharing the story of the inception and where you're at today. Absolutely. So we call them trusted regulatory spaces, but I'll use those two terms synonymously. We're very happy to see an actual term being coined for this because the use of the cloud in the regulatory use cases is really important and we think it's going to make a big difference.
00:02:13
Speaker
And for us, the journey really started in 2015. And oddly enough, it was all because we had ISO 27001 compliant certification. which made us very much of an outlier.
00:02:25
Speaker
And we're a cloud-based genomic platform, which also made us an outlier back in 2015 as well. But the FDA approached us of all things through LinkedIn, and they asked us to design and deliver Precision FDA.
Impact of FedRAMP on Opportunities
00:02:40
Speaker
It was originally for presentation of scientific challenges. And I think we've done over 40 different challenges and hackathons, ranging from next generation sequencing, truth challenge, to representation of ethnicity in the EHR using VA data.
00:02:59
Speaker
But there have been a really wide range of scientific challenges hosted on precision FDA. But in 2018, the FDA realized that they could actually do more with the precision FDA platform.
00:03:12
Speaker
And at their urging, DNA Nexus embarked upon the arduous journey of obtaining FedRAMP authorization. It took us probably two plus years.
00:03:24
Speaker
First audit, I didn't quite make it, but second audit we did. And in 2018, we received the authorization. from HHS. And in truth, the FedRAMP authorization has been the keys to the magic kingdom. Without the FedRAMP authorization, we really couldn't do the work we're doing at the FDA, the work we're doing for ARPA-H, the work we're doing for NCI.
00:03:47
Speaker
All of these things are enabled by the strong security, privacy, and quality protections that we have. I was going to ask you, Steve, in the in your encounters, have you seen how compliance and things like FedRAMP authorization make or break a innovation?
00:04:08
Speaker
And I'm kind of curious if you've seen this from an international perspective, because obviously FedRAMP is a
Compliance vs. Innovation: A Global Perspective
00:04:15
Speaker
U.S. government. But I'm curious, as you see innovation and regulatory, do you see compliance come along with it Because I don't see ah oftentimes when people talk about innovation, I don't see them talking about the the sticky parts, the hard parts. I'm kind of curious what you've seen out there.
00:04:33
Speaker
Yeah, I mean, an interesting question, and it you know kind of gets into compliance of validation. And I think the short answer is, if we think about it as as a dilemma, right, it's a clear dilemma in all this. And and in previous terms of technologies, you know the bottom line is we always find a way.
00:04:50
Speaker
We always find a compromise. and And certainly this is the big topic with AI today. you know There's guidance from the FDA, there's guidance from EMA and other different health authorities. And it's just a matter of kind of getting experience and that compliance innovation line. Where is it? Like, what's a red line? And I think we're dealing with a lot of gray area, but like in anything else,
00:05:13
Speaker
It's just a matter of having the right smart people figuring out, having some clear standards. And I think in this case, the standards have to be global and not country by country. Just as reading just the other day about AI, you know, does each state in the United States, is there going to be 50 state level perspectives, which would really hamper innovation versus just having some global quote unquote, common sense standards.
00:05:38
Speaker
Yeah, that's already happening in the privacy arena. We've stayed on state individual privacy and for companies like ours, that just adds to the regulatory burden. Fortunately, a lot of them are similar and we do have the GDPR data process processor experience.
Data Privacy and Trusted Research Environments
00:05:54
Speaker
Well, since we're talking about Europe, that takes us back to sort of the second a leg of the genesis of trusted regulatory spaces, which oddly enough was the UK Biobank's research analysis platform.
00:06:06
Speaker
So UK Biobank is a really unusual project and it's got an unusual data set because it's 500,000 individuals with whole genome, whole exome, proteomic imaging, clinical results. It is really a treasure trove of clinical genomic data and lots of research is being done on it.
00:06:29
Speaker
But in order to do that, We had to protect the privacy and health information of the participants, and that was done through something called a Trusted Research Environment, or a TRE.
00:06:40
Speaker
And you'll see this term generally associated with biobanks, but the idea and the principle is really straightforward. Keep the data located in one place. put tools with the data, presumably in the cloud, and then bring the researchers to the tools and the data.
00:06:57
Speaker
And this is going contrary to a, I'm sure, decades old paradigm of downloading the data to one's local compute infrastructure and performing the analytics there.
00:07:08
Speaker
The problem with that is the data gets out and then you really don't have control over its propagation. And hence you end up with your challenges with protecting the privacy and the information of the participants.
00:07:21
Speaker
So the trusted research environment have this design pattern, as I said, keep the data at one place. And this is really persisted through our thinking. We really don't think the data should be moving if it doesn't have to.
Trusted Regulatory Spaces and FDA's PRISM
00:07:34
Speaker
So by taking those two pieces, the trusted regulatory trusted research environment and precision FDA, we realized that these things have a lot of similarities. And in 2023, the FDA launched the PRISM Research Collaboration Agreement.
00:07:53
Speaker
This was between Cedar and Ciber and seven different pharma companies. And the idea was to use the Precision FDA with its FedRAMP authorization to host regulatory use cases. The first one they did was pre-submission validation.
00:08:10
Speaker
This is where we ran into our colleagues at Lorenz and imported the Lorenz products. onto the trusted regulatory spaces and they've done things with the m11 protocol for structured clinical trial representation real world data there are just a number of very tangible and meaningful use cases that are coming out of the project prism and it really has demonstrated that commercial software open source software
00:08:42
Speaker
Regulators, sponsors, all of these things can come together on a cloud platform to really transform regulatory efficiencies. So combining this concept of the trusted research environment from the UK Biobank with all the regulatory tools and experience we have from Precision FDA and PRISM, the concept of trusted regulatory spaces was born.
00:09:05
Speaker
Wolfgang and I were not sure who came up with it first, but both Lorenz and DNA Nexus have been propagating this term until you came out with the cloud-based regulatory spaces, but they really are synonymous. um We have deployed an initial trusted regulatory spaces commons in production now in our DNA Nexus Frankfurt region on AWS. And if you're actually sitting in the offices of the headquarters of Lorenz, you can look right out the window and see the AWS data center where all this stuff is housed.
00:09:38
Speaker
Yeah, it's a pretty incredible journey. And I think the thing that I'm learning on this, because there's other players that have started and emerged in this space, and it's still early days. So and instead of just coming up, just think about the regulatory submission and kind of a submission of one-to-many versus it's always been a kind of one-to-one, even though we've been using this term for decades, simultaneous submissions, it's still one-to-one, not one-to-many, but hearing the story of inception to today, that you have a lot of general collaboration experience, being on the research, the different aspects of it.
00:10:13
Speaker
So there's different dimensions of of collaboration. So as we're introducing the dossier, it's just another thread, you know, in many, many about the the greater scientific collaboration when one place and the data's in one place.
00:10:26
Speaker
So that's really interesting. So we'll shift because we're in the environment, unfortunately, today is what have you done for me recently, which is a business benefit conversation, right?
00:10:37
Speaker
Yep. So with some of the initial experience and some people that I've been in pilots, or actually pilots are active projects, as far as things that are important to some, like time to filing, others more on the clinical development, it's time to patient.
00:10:54
Speaker
Hey, less individual submissions, you're reducing the complexity of a very complex process, and probably the list goes on. it'd be good if you could share with our listeners, like, what are some of the benefits that you're seeing? And I think more importantly, what do you envision in the future, in the coming years, from a clinical development, regulatory operations perspective, Omar.
Leveraging Cloud Technology in Regulatory Networks
00:11:16
Speaker
Sure. um Let's just checkpoint a bit in terms of where we're at today based on the two and of twopl plus years of PRISM and the learnings from Precision FDA and engaging FDA and sponsors.
00:11:29
Speaker
One of the things that we've noticed really is that the use of the cloud is pervasive throughout all industries. with the exception of cloud-based regulatory stakeholder interaction.
00:11:41
Speaker
Engaging national regulatory agencies is really challenging. There's data sovereignty issues, there's reviewer burden, um and reviewer burden is really a big deal. It's difficult to change processes when everybody knows how their current process and their current tools work, no matter how inefficient they might be.
00:12:01
Speaker
So what it comes down to is the technology is proven. So to create a cloud-based global regulatory network, there's no cloud technology that we need new.
00:12:12
Speaker
It's really, we have to address the process and governance aspects of the problem using the proven tools that we already have. So if you're thinking about this, there's SWIFT and SABER for the financial and travel organizations. Now, these are that I've thought a lot about this because our world in regulatory is probably the most complex mesh of information, regulation, laws, protections, security, privacy, quality. Now, when I say quality, I mean change control, things like GXP, computer system validation.
00:12:52
Speaker
But all of those unique dynamics I don't think that anything like Sabre or Swift will ever be implemented from a top-down in our industry.
00:13:03
Speaker
I cannot see ICH or WHO or any organization going off and actually being able to impart a top-down view of regulatory standards for interaction.
00:13:15
Speaker
What I do think would happen, and this is where we're at, I think it's going to be a groundswell. It's going to be a bottoms up effort. And a lot of the work we're doing is open source based. And I think that will also help to propagate a methodology and mechanism where there's a harmonized user experience, no matter If you're dealing with FDA or MHRA or PMDA, there should be a way for people to have an interaction that's similar across the different parties.
00:13:45
Speaker
And the approach needs to be driven through necessity. So if you're really looking at this, It's not an incremental change that's happening right now. These are really pretty revolutionary changes that are going on.
00:14:03
Speaker
And the blockers that exist are real. As I said, data sovereignty, not in my backyard. i didn't invent it here. Those things go on and on, but the benefits are making these advancements inexorable. The benefits are absolutely there.
00:14:18
Speaker
And it's really all about saving submission and review time without sacrificing regulatory rigor. And that's sort of what the trade-off we were talking about, Steve, with a the compliance, ah the overlay of all these things that slow you down.
00:14:36
Speaker
with all of the ability to go faster using the cloud.
Digitalization of Protocols and Submissions
00:14:41
Speaker
So again, it's really about saving time without sacrificing the rigor. And this is why security, privacy and quality are absolutely the foundation of everything we're doing with the trust of regulatory spaces.
00:14:54
Speaker
So if we're to look at what PRISM is currently demonstrating, you've got connection of multiple sponsors to the FDA. You've got integration of commercial and open source software in the cloud.
00:15:06
Speaker
And this is another area where we think a lot of progress can be made. If you think about people's software toolkits, everybody's got their own on the sponsor side, on the agency side, but we think there's benefit for sponsors and agencies to be looking at the submission through the same lens.
00:15:24
Speaker
And this was proven out in the Project PRISM pre-validation use case, where sponsors and agencies could look through the same structural validation of the ECTD, the same validation of the clinical and non-clinical data that's in the submission, but seeing it through the same lens so that there wasn't any surprises by the time the submission went through the gateway.
00:15:48
Speaker
So sharing of the data and the tools in advance before things go through the gateway definitely proved to be beneficial. And real-time interactions where you're not sitting and waiting for a week or even an email ping pong, but to be able to interact with the data and the tools right on the platform.
00:16:09
Speaker
And so on the TRS, we've actually baked in whole discussion thread mechanisms secure where these things can take place directly on the platform. And last but not least is digitalization of protocols.
00:16:23
Speaker
And we've seen some are pretty amazing examples of what you can do if you take something like a clinical trial protocol and put it into structured format like in JSON so you can render it, visualize it, transfer it. And we'll talk a little bit more about a use case that was not quite so intuitive, but it really showed the benefit of digitalization.
00:16:44
Speaker
And this is a little further up in the clinical work stream. And I should say that all of these benefits are not confined just to the submission part. But, you know, you were talking about time.
00:16:54
Speaker
So if you could shave off time from last pace and last visit to database lock, you could save a lot of energy and time and money because ultimately everything you're doing that delays entry into market is all money that's sitting on the table.
00:17:12
Speaker
So let's let's take a look at what CVRS is going to do over the next two or three years. We really envision a global network of sponsors and agencies and standards development and trade associations using harmonized data interchange and communications to support parallelization. And parallelization is really a lot of the holy grail here.
Vision for a Global Regulatory Network
00:17:35
Speaker
um You have wave zero, wave one, wave two, go to market activities. But in those three words I just said, that could be five years for a market change in manufacturing.
00:17:50
Speaker
um that's just unconscionable. I mean, that stifles innovation. And worse than anything, in these other markets, patients don't get access to treatments in a timely manner.
00:18:02
Speaker
They're waiting on the wave zero and wave ones. So we'd like to flatten that and have agency to agency decision sharing and knowledge transfer so that all agencies can have a equitable, democratized place at the regulatory table.
00:18:19
Speaker
and to really improve access to patients globally by, as I said before, trying to knock down this multi-wave approach and to really accelerate protocol digitalization so that we can have better interchange, better visualization, and innovation. And we'll talk a little bit about, like I said, some innovation that we've seen in the clinical capturing of the EHR data into case report.
00:18:48
Speaker
I want to talk about that use case just briefly because it really struck me.
Digitalizing Clinical Trial Protocols with M11
00:18:53
Speaker
it was non-intuitive to me how this would work. So I wanted to talk just briefly about the M11 protocol. This is for clinical trial definitions. Clinical trial definitions lend itself really nicely to digitalization because it's so prescriptive. I mean, you've got your schedule of activities, you have all the detail that goes into those, you have to know exactly how you're doing the blood pressure, all that stuff has to be very, very rigorously specified.
00:19:21
Speaker
So it lends itself nicely to um the M11 protocol, which as I think it's just gotten, it's a level four approval. it's It's one of these things that's actually becoming reality.
00:19:32
Speaker
So we're working with a technology partner that has very cleverly looked at the problem of transcribing information from the EHR into the case report forms.
00:19:45
Speaker
So this is where site monitoring comes in. The site monitor has to go out and inspect the transcriptions. And again, at last patient, last visit, all of the all of that has to go through source data verification.
00:20:01
Speaker
And that's another big delay before you can do database lock. So this technology partner, they've come up with a very clever way to use the M11 protocol, which essentially defines everything you're capturing out of the EHR.
00:20:16
Speaker
And they built a screen... of fields essentially define the fields from the EHR that need to be captured. And then their tool, you take the screenshot literally of the EHR. You sit with the EHR on one screen and the tool on the other screen and you capture the EHR.
00:20:37
Speaker
The fields are all defined by the M11 protocol. And it literally takes all of that information, redacts the things it needs to, de-identifies the things it needs to, but it essentially fills in the case report form without transcription.
00:20:53
Speaker
So now all of a sudden you have an error-prone process. human that has to be checked leading to delays in database lock that now is validated by the fact you've photographed these screens off the EHR as defined by the M11 protocol and captured this data directly into the case report form.
00:21:18
Speaker
And, you know, this whole process of site monitoring and stuff 10% of a clinical trial budget can be absorbed in all of this activity.
00:21:28
Speaker
So this is where we get excited because if you think about just this one part being affected by digitalization and the ability to use the cloud to propagate all these parts, to connect all these parts.
00:21:42
Speaker
Just amazing potential for savings, both in time and in cost. So I really wanted to put that out there because, again, it's a non-intuitive use of a digitalization of protocols. I'm not just transferring it.
00:21:54
Speaker
But the digitalization enabled me to do things way more intelligently. So I really appreciated seeing how this digitalization had knocked- at knockoff benefits that weren't obviously, weren't intuitive.
Cost Implications and Efficiency in Clinical Trials
00:22:08
Speaker
Yes, that's a great use case. And, you know, I was making some notes as you were going through that because back when I was in the industry, I actually did did some work in clean ops with Janssen or, you know, Janssen, depending upon what side of the pond you're on. And, you know, certainly Johnson & Johnson Innovation now, I think they've renamed it. Yes, that's a topic for another podcast, you know that name change. so But anyway, you know especially here, where you and I are both in the US and just the the cost of prescription drugs.
00:22:39
Speaker
and And people that are not in the business, they don't understand the cost just to conduct and the cost to conduct a phase one, two or three clinical trial. So if you're shaving, 10 or 15 percent of that cost that is substantial and that gets built into every pill injectable inhaler whatever that is and then the other thing that last patient to database lock to the filing you know there's a big cost to that process and we compare some notes with our friends at mckinsey with their data and our data because they they track that and and they have best in class and then we look at it from an operational standpoint in our research
00:23:15
Speaker
On that, so like the median of that that thing was like almost like 30 weeks, like six years ago. And I think best in class as of last year was like 10.4. And people are talking about, which would have been unheard of five to six years ago, or even the inception of the work that you've been doing to say from last patient to database lock, the filing, the average medium time is going to be six to eight weeks.
00:23:39
Speaker
We would have said, what are you smoking? Right. You know, that's unrealistic, but here we are. We're getting very close to that. So your guys' contribution certainly can drive costs down. But the other two things, as I made it my business to really understand these cloud-based regulatory spaces, you know, earlier in this year, the first thing, and you've already mentioned it, is like, okay, if I'm a regulatory affairs person and I'm more in international markets, so say that wave two, wave three,
00:24:07
Speaker
Regulatory strategy as this matures fundamentally changes because I'm not thinking market to market reference, you know, that it just fundamentally because I think the average from like first filing to last filing might be like 36 months, even for an extension. That's kind of crazy. and Like, why couldn't that be cut in half?
00:24:25
Speaker
Substantial cost reduction there because everybody's struggling with the cost of medicines. But the other thing that became very clear to me is you think about all the health authorities.
00:24:37
Speaker
and you know them, I don't, they all have their different experts, maybe an expert for oncology, an expert for dermatology or whatever the therapeutic yeah area is.
00:24:48
Speaker
And they have their identical, say, twin and another health authority. We can start better leveraging the scientific expertise globally, which is really exciting. So So with that said, ah just one last question for our listeners.
Future Plans for DNA Nexus
00:25:02
Speaker
Like I said, the time flies and all that. So what should we expect from DNA Nexus kind of in the near term in the next one to two years? So what say ye, Omar?
00:25:12
Speaker
Well, as I had mentioned before, it's really not about technology. So the next steps are really going to be about process and adoption. um To build this trusted regulatory spaces network, we're working with sponsors, agencies, standards associations, excuse me, standards development organizations, and trade associations. We really think those are all the players.
00:25:32
Speaker
And our objective is to harmonize the data interchange, the protocols, the encryption standards, basically to create a secure global standard for interaction to develop omnibus and multilateral interparty agreements.
00:25:47
Speaker
Our our thesis is that all of this is going to be driven through data interchange agreements, whether it's a data use agreement, a trust agreement, a technology transfer agreement.
00:25:59
Speaker
These agreements are going to be the glue that holds the network together. We want to templatize artifacts for IT systems. So the first thing we do with any pharma company that we're bringing TRS in is we have to go through their security audit.
00:26:14
Speaker
So that stuff should be templatized. that There should be a very straightforward way to enable these IT teams and security teams to assess a platform, again, so they can adopt the cloud rather than running hiding from it.
00:26:28
Speaker
And we really want to accelerate this use of digitalization of protocols. I think the other place we're going to see this is in M4Q in CMC. ah That's another area that is really, really ripe for digitalization.
00:26:43
Speaker
And as we know, there's been a lot of efforts to do that with PQ, KMS and others of that um haven't really come to fruition. And I'll have to admit some frustration on my part, seeing how long it's taking to actually represent CMC data in a structured
Challenges in Digitalizing CMC Data
00:27:01
Speaker
manner.
00:27:01
Speaker
it It actually came to my attention just yesterday, and I should have known this, that module three is all PDFs currently. And I found that to be just like, what? there's no There's no SDTM standard. There's no standards for it.
00:27:16
Speaker
So that's another area that I think we really want to tackle through bringing people together on the TRS to start hammering out a better way to represent CMC. And I know that M4Q R2 separates out the PDF and the data, but there's no standard for the data yet. So how do we do that without waiting three more years?
00:27:37
Speaker
That's really yeah where we're coming from. Time. Time is what we want to save. Yeah, and that that PDF story with the SCMC, kind of flashes me back. So I basically had three jobs you know when I was at Janssen with J&J.
00:27:51
Speaker
And my last rotation was to the Janssen Research Foundation. So I was still in technology, but everything was so customized and you build it, right? So I had ah such a unique opportunity to run part of clinical ah operations along with being the steward and the driver for the tech that was highly customized.
00:28:11
Speaker
So I'm walking into the research building and I'm like, well, what's what are all those tractor trailers for? So it was the classic three tractor trailers. Submission was going out to the FDA. They had yeah had do three copies. They had different routes going down to Washington.
00:28:28
Speaker
And there is this guy they just hired. He was a PDF expert. It's like, oh, we're going to make all these PDFs. We're no longer going to make the trucks and or have the trucks transport it. But you know that was 30 years ago. And I don't think nobody could imagine really ah kind of what you're doing today, let alone having a cell phone with what cell phone is doing.
00:28:48
Speaker
So like you said, how you opened it up, it's really revolutionary kind of times. And it's like you said, it's not so much about the text and the enabler. It's about a different way for a worldwide collaboration to get things to patients quicker, ah hopefully to drive down the cost of doing that and just take out so much complexity in today's process.
Contact Information for Omar Sarang
00:29:12
Speaker
So Omar, thanks so much. I'm sure some of our listeners might want to get a hold of you. So what's the best way to get a hold? Is it websites, LinkedIn? What would you suggest? Well, LinkedIn is always a great way to just go out there. And there's very few Omar Sarang's out there, S-E-R-A-N-G at DNAnexus. So it's really easy to find me on LinkedIn, but you can also get a hold of me at osarang at dnanexus.com.
00:29:37
Speaker
And of course, there's the DNA Nexus website at dnanexus.com. And you'll find our trusted regulatory spaces in there alongside all of the other work we do for diagnostic companies, healthcare care companies, and global biobanks. so So just look for it and it'll be there. But again, you can reach me at osarang at dnanexus.com.
00:29:58
Speaker
And Steve, thank you so much for having me and for your patience in working through this. I really appreciate it. Yeah, now now my pleasure because there's there's always updates, something's always changing in the regulatory environment, but this is very significant.
00:30:12
Speaker
you know We're entering a new era, a new wave of collaboration enabled to know by by the tech, like you say. And for our listeners, we love LinkedIn too. I have an easy way to get a hold of me like you, Omar, Steve Gens. There's not too many Gens's, you know, kind of out there or certainly off our website.
00:30:29
Speaker
And Omar, I mean, things are changing so rapidly. So, you know, maybe next fall we'll have you back and you can give an update about what happened and more importantly, what's on deck. So it's it's definitely very exciting time. So thanks so much again.
00:30:44
Speaker
And again, thanks to Gens and Associates for really spearheading this pharma, biopharma viewpoint. It's great to have this insight and to participate in all of the work that you folks are doing. Really appreciate it.