Commercialization

Good morning, everybody. And welcome to the BioLaunchPad Seminar Series. This fireside chat today is co-hosted by the Center for Integrative Genomics. Thank you, Greg Gibson.

And today, we're really honored to have Ridhi Tariyal join us. Reedy is a Georgia Tech alum. Yay. Go Yellow Jackets. And ISYE, and she also has an MBA from Harvard.

She has a Master's of Science from MIT, and a lot of other great things. You're a Blavatnik Fellow, correct? Is that with Harvard or Yale or both? That's Harvard. Harvard, yeah.

Farrolyn Fellow, can you tell us what that is? It is a program out of the Fulgurty Institute based in Palo Alto, so their focus is medical devices. Great.

and she is the CEO and co -founder of Next Gen Jane. At Next Gen Jane, Riti has driven the development of novel menstrual data platform characterizing uterine biology at a molecular level.

In this effort, she has raised capital, she has established an IP position, and she's developed a team to create novel hardware and software to change how women access care.

Before NextGen Jane, she worked at the Broad Institute and then at Bristol-Myers Squibb. So what we're going to do today is, again, a fireside chat. I'm going to ask some questions just to get the conversation going.

And then I'm going to open it up for questions from the audience. We also have people that are remote. And I would just ask that if all the remote people, if you have any questions, please put them in the chat.

Christina is going to be monitoring the chat and then she'll be she'll be asking the questions here. So let's get started.

So Reedy, what inspired you to start your company? I'm gonna answer that question but I want to thank you guys first for inviting me and sponsoring this event.

I was telling everybody that this is my first time back on campus since 2002 and it is completely transformed and it was already sort of a place where a lot of innovation and novel IP was being filed like 23 years ago and I now it must be like a hundred times even more intense so I'm really excited for you guys and I don't think that there was an innovation center when I was going through no I'm so that's amazing but those are the types of resources that if anybody here is actually interested in taking the research that they're doing and I'm trying to grow it in a commercial

capacity that are going to be critical to making that transition. I started NextGenJain for two reasons. One was I had a lot of questions about the state of the state of my own reproductive care that when I engaged with the typical medical system I was given a lot of hand-waving answers or like you know blank stares in terms of why XYZ might be happening and everybody has an experience where they can fill in what XYZ is but specifically at the time I had just finished my tenure at my two-year stint at Broad and in that time you know you when you're at the Broad like you see

how genomics is transforming everything and your perspective is is that you're living in a future state where molecular medicine is here not some something in the future but when you actually engage with your own healthcare providers specifically for OBGYN. It's a completely different experience. The tools are very analog, they're not very digital, and so my desire to start Next Gen Jane was in part this realization there was an opportunity to try to build up more molecular understanding of diseases that are female-dominant.

So with your company, what is the problem that you're trying to solve? and then also what is your core product for your company? The second question is tough. I will try to answer it.

The problem we're trying to solve is just that there is, and I think everyone probably hears this, right? It's almost become a trope where there's a lack of data in women's health.

There's just, we can't answer the question because there's not enough knowledge, not enough data, not enough clinical research in that specific space.

And so that is our, the core problem at its most foundational is generating the data set that's able to answer some of these questions probably the higher level more accessible problem is that there is a lack of good diagnostics and a lack of good therapeutics because of the lack of data and so endometriosis is just a good example that we use because it's very easy for people to understand it affects about 10% of the female born population. You know, a highly debilitating disease that's accompanied by pain progresses to infertility.

And 10% is like common, right? That's not a rare disease. That's like very common. And yet the only way, the gold standard way of diagnosing disease is laparoscopy, which is surgery. So you have to have a surgery to confirm that you even have the disease.

There's still sort of lots of question marks as to the etiology of the disease. So we don't know why it happens in certain people and not in others, which means that because there's no clear molecular understanding of what's happening in the disease, that the most common way that we have of treating it is by sending women into menopause. And so we have GnRH analogs that we prescribe to say because this is an estrogen dominant disease, maybe if you shut off endocrine axis you could control the disease. That's a very blunt instrument to target this disease and if you ever talk to

patients who have endometriosis, they hate this modality. They don't want to take these drugs. The side effects are exactly what you would imagine that they are during menopause, but happening in women of reproductive age.

Endometriosis is great because it's a great example of how the lack of data can lead to bad diagnostics, so 10-year diagnostic odyssey to even find out you have a disease, and when you do, lack of therapeutic options.

Sadly, it's emblematic of a lot of women's health diseases, so that's the problem we're trying to tackle. The product is twofold, so we have a data set and a data asset that we are putting together that is categorizing and taking menstrual samples, which are essentially giving you a natural biopsy of the endometrium, right? We are generating high-dimensional data from those biopsies, so we're doing RNA-seq and exo-seq. And then we are doing clinical grade annotations, so we are taking labels from your clinical EMR as well as a next-gen Jane survey that we administer that asks very

specific questions, and we're putting it all in a female ontological framework, which just means there are specific female, there are specific female-specific labels that are necessary to stratify the data, that if you don't label the data with these labels, so a great example is like what day of the month you collected the sample which turns out is very important in understanding what you're looking at we put all of that together into a data asset that allows us to interrogate and ask questions that we have higher confidence we're going to get real signal and answers to so one

product is just having access to this data asset and a second so you know diagnostic companies biopharma companies anybody who has questions that are relevant to female biology, whether they're doing target identification or trying to find molecular phenotypes to improve their clinical trials, can access this data asset. The second product is we ourselves can access the data asset and spin out products such as a non -invasive diagnostic for endometriosis. So that is a second product that we ourselves are spinning out to say, can a woman mail in her tampon and find out she has

endometriosis prior to going down the path of surgery. With pharmaceutical companies, for instance, do you have a model where perhaps they would partner with you and provide their own data that would go into your analytics platform? Is that a model that you're thinking of in the future? No, you know, there's two things we're thinking about. One is of licensing our data to pharma companies, our data as well as the tools that we have on top of it.

And so that is for them to integrate, you know, any insights they find from our data into whatever repository they have. And I only say that because I'm assuming it will be much harder to get data out of pharma companies for a co-development or something like that.

Another model that we're interested in, though, is actually doing something like a co-development where taking our data as well as drug development, drug discovery tools that they may have where they're the experts and finding novel targets and spinning them out for particular indications.

and how are you getting all this data now so now it's voluntary people are sending their tampons to you and then also you have clinical trials that you are also planning to do or in the process yes everything's through a clinical trial so okay we we go through an informed consent process so that anybody who gives us a tampon is actually part of an IRB approved protocol and that we do that very very deliberately so that we have we're making sure that people understand and are permissed into the research and they they know exactly what they're participating in and that gives us the most

ability to look at the data in the most ambitious ways but everything has been done in that R&D capacity of and it is voluntary like there's a nominal gift card that we give them for their participation but I really do think the majority of people participate not for the Amazon gift card but because they they see an opportunity to transform research I'm gonna stop for a moment and see if at this point we have any questions so with this amount of data do you think you could identify maybe predictors of some diseases like early stage in the Mitosis for example if you have enough samples of

people with endometriosis or do you think it's more as a diagnostic tool you know for endometriosis in particular when you say predictor right I don't know if you mean like genetic predisposition to endometriosis like if you for example see certain genes that are being expressed expressed with in people with early endometriosis and then someone that is ion diagnosed sends you a sample and you see some of these markers being expressed you're like oh maybe you could have endometriosis yeah that's that and that is the that second product that I talked about is we absolutely think that that

is a great utility of the data is being able to see a difference between people who have there's actually four categories one is people who have symptoms and have endometriosis there are people who do not have symptoms and have endometriosis there are healthy people who don't have endometriosis and there are actually people who have no symptoms and and have sorry who have symptoms but don't don't have endometriosis there's some other cause for symptoms that look exactly like endometriosis and the ideal goal is to be able to parse through and and identify exactly what bin you fall into i

think this is such a great idea and understudied topic and i wonder if you've thought about how it's also an area where there's more risk for data privacy which is an issue in bio in general but you know we're in a state where women's reproductive care is criminalized. I've had personal experience with that. And in other states in the nation, really more so. So it's honestly, I think it's a more important topic for that reason. But this is also stuff if you find out somebody is pregnant, they're now in legal jeopardy that they wouldn't have been in otherwise. Obviously, clinical trial framework

helps for that. But I wonder how you think about that in your company and whether companies like yours can then sort of work on the policy side as well or have anything to add and to make that side better as well. Totally. I think that if you work in women's health, you sort of can't get away from engaging with the policy side. And so we've been on the Hill.

Our primary focus to date has been trying to advocate for better reimbursement. Sort of an added layer of interest and complexity in women's health is that women's procedures and drugs don't get reimbursed as highly as procedures that are done on male bodies. And I have Eurogine friends, I'm sure you guys can talk to your friends who say that they their male procedures subsidize their female procedures and so there's an entire like battle to be fought to say innovation goes where there is ROI right and so if if there are male interventions that are reimbursed more highly that's

where capital is gonna go because they know there's a path to getting a return on their investment and so we think it's very important not only for accessibility of any sort of diagnostic that there be a path to reimbursement but also even broadly for women's health innovation you have to show examples of where you're you're able to like fight that battle and get as good reimbursement as some of these other interventions so that's where our policy focus has been to date though I don't disagree with you that that we have to work on these privacy issues so that people feel

comfortable sharing their data I will say one of the direct changes we made to our entire process as a result of some of the the changes in recent years has been that we removed certain questions that we used to ask both because we thought they may be a deterrent that asking that type of question would make people not want to participate and as you can imagine that affects the science right if you're not asking certain questions because you're just you don't want to you don't want people not to answer and you don't want to be liable for holding on to that data but it felt like the right

thing to do foster trust to say this there's no risk here of this going down that particular loophole or path and so you know we've had to respond to it in that way but we care a lot about data privacy it's why we have we don't have just a voluntary you can drop your tampon off you have to be in for you go through an informed consent process so you understand exactly what what we're asking for and how your data will be used and then we have all sorts of mechanism by which we de-identify the data we remove your your like PII from it and we are looking at it in that completely de-identified

manner to address some of those privacy challenges yeah yeah absolutely this is sort of like a through line of questions that I've been getting asked today in general which is how do you get something like this funded it's you know it is when you're when you do it through the form factor of a startup you know it's not for basic research it's not just to advance the science and it's not for reaching our sort of global goals of equity and healthcare right those are there are different organizations that will fund endeavors like that and so you have to sort of prove out so you know they're they're

the only reimbursement currently for an endometriosis diagnosis right now for example is through surgery right you get you get reimbursed for actually doing the laparoscopy and so that that reimbursement code is not likely to translate into a esoteric genomic test based on menstruation right as far as I know there's no reimbursement code for a genomic menstruation test and if that's about a fight we'll climb that hill but it'll take a while and so do you hold off on going to the market until you have reimbursement locked up or do you start somewhere where there is some flexibility to

get payment and so we decided on the latter to say you've got to start somewhere and so one of the areas where we were able to find a good product market fit where there was a willingness to pay was the infertility context and you know there are many reasons that infertility is really interesting for endometriosis diagnosis.

One is I think that the stats that I said it was like 10% of the entire female population has this disease so it's very common. It turns out in the infertility context so if you are actually struggling with infertility, it is upwards of 20%. So there are a lot more women who have undiagnosed endometriosis where it's progressed infertility and they just don't know that are being worked up in IVF clinics. Now, the protocol actually might shift if you knew your patient had endometriosis. There's a couple of things that could influence that. One is they actually

try now to put you on three months of GnRH analog therapy to send you into menopause for a short time. And the hope is let's calm down the inflammation, bring you out of menopause, and then try implantation with the hope that implantation rates are higher because your body's not in such an inflamed state, right? But if you're going through the IVF process and you don't know you have endometriosis, you're probably not going to opt in to three months of chemically induced menopause unless there was a reason, right?

Unless you knew you had a disease that was going to be impacted should you go down that path. The second is some people actually want to get treated before they get pregnant, and that's fine, but given that the chance that having surgical intervention to treat your disease could impact your ovarian reserve, reproductive endocrinologists still suggest that you go ahead and do cryopreservation, form embryos, do the entire process, go get your treatment, and then come back and do IVF, right? And again, that is change in clinical management, that they can

only affect that change if they knew a priori that you had the disease, right? The final thing is the first -line treatment for infertility is they're going to do ovulation induction with something like Clomid, and they might do intrauterine insemination, but it turns out women who have endometriosis don't respond well to first-line treatment, so they should move on to second-line treatment, which is in vitro fertilization, and it costs like $3,000 per cycle to do that first -line treatment, and they really recommend like three to four cycles. And so if you have endometriosis, you

are wasting $12,000 in six months doing a therapy that you're not going to respond to. And so it's one of those rare instances in healthcare where an intervention or a diagnosis would both help the patient, right?

They wouldn't waste their time and money. It helps the doctor because if they knew, they would actually change your protocol. And it helps whoever is paying for this. So in a lot of states now, employers cover fertility benefits, some fertility benefits, like usually it's something like $10,000 to $20,000 a year.

And so they would benefit because they're not wasting time and money and supporting you to do something that you're not going to respond to. And so because that happens to benefit everybody, it ends up being an interesting go-to-market strategy before you get reimbursement, because there are multiple people who would be interested in paying for that because it would bring down the overall cost, whether the patient's paying out of pocket or whether your insurance company's paying or whether your employer's paying, everybody would benefit from having that knowledge. So the thinking

there when we think through these things is find the opportunity where everybody's going to benefit from paying for something like this while you work on reimbursement. And I say that because ultimately market adoption only pops in healthcare once you get reimbursement. You can do well without it, but not as well as once the reimbursement mechanism has kicked in.

And so there's generally people are thinking about what's my path to reimbursement and then how do I actually get someone to pay for it until I get reimbursement.

So infertility and IVF is one of those areas where we think there is an opportunity to do that. So this is not a direct-to-customer product? Oh, it's not.

It's not. And there's two reasons for that. It's a nice sort of segue of like there is a belief amongst a lot of life science investors that direct-to-consumer does not work in healthcare.

That either it's going to be seen as a fluff product that doesn't have staying power, there's no stickiness, right? You get your results once of like predisposition and your, you know, some fun facts that you use maybe like infrequently in your life and you never engage with a product again and or there's a limited number of the population that's actually going to pay out of pocket for something fun that's like genomic, right?

And so fundamentally, we are not necessarily of that camp. We do believe that the consumer, especially post-COVID, and especially with the rise of concierge medicine, like I don't know if anybody in here pays for one medical, you know, I don't know if anyone here pays for some of these TIA clinic or like there's these these really bespoke there are actually some of these concierge clinics that focus just on autoimmune disease or PCOS and they are showing that there is some market appetite if you have a particular conditions to up pay for access to more you

know accessible services better services more testing more access to drugs even you guys might have noticed that pharma companies are now starting direct -to-consumer mechanisms to sell their drugs. And in part, it was motivated by the success of GLP-1s.

There was such a high demand that so many compound pharmacies moved in that the big pharma companies were like, we should just be selling this directly ourselves.

And so there is this, how long that takes, 10 years, 20 years, until we see the full fruition of what that's going to look like, it's not here now. that's that's still on a you know one-off basis we do believe that you're gonna get better adoption especially for like a genomic test and for something like endometriosis or other things that were interested in our inflammatory and autoimmune diseases what are you going to do with that result yourself right you're gonna have to engage with the medical system to figure out the next steps anyway and so it just makes sense to have a physician

intermediated product maybe you want to differentiate yourself from the wellness and we do so can you tell us a little bit about the regulatory considerations for something like this it's so interesting I don't know how many of you guys just as a by a show of hands are familiar with the CLIA pathway awesome that's great so diagnostic tests do not have to go through FDA approval necessarily and There are a lot of reasons for that. If you have a high -complexity genomic test, one, you're constantly learning. So, like, as your data set grows, you continue to reevaluate

and refine the test. As you guys might know, once you submit a filing to the FDA, you can't change your assay, you can't adapt it, right?

And so one of the great benefits of the CLIA infrastructure, which is CMS approves your lab, and then you are able to run tests within that lab setting through a doctor to help a doctor make their own clinical management decisions.

And you have to have a publication you can refer to, right? So a doctor is not going to prescribe your test unless there is a peer -reviewed paper that says here's how well the test performs. So generally, it still has a level of rigor that allows you to present your data. it just doesn't have the same regulatory apparatus, so it is like less of a regulatory burden.

It turns out for diagnostics, even if you get FDA approval, it does not guarantee reimbursement, and so it really disincentivizes people from going down that path. In drugs, it's different. If you get a regulatory approval, FDA approval for a drug, if it's first in class, you're going to get reimbursed.

It's very clear, and so from an investor perspective, when they're investing in a drug that's going through FDA approval, they're not really taking on payment risk, right? They're taking on scientific risk and they're taking on regulatory risk, but they're assuming reimbursement, so there's no payment risk.

For diagnostics, it's really bizarre. They're taking on scientific risk, regulatory risk, and payment risk, right? And so it actually erodes the value proposition of going down the FDA approval path for diagnostics, and it's something to consider, but most diagnostic tests in the United States just use this LDT pathway.

It turns out in 2024, the FDA released a rule that suggested that all LDT tests would come under FDA rubric within the next four years.

And as you can imagine, all the lobbying groups for big diagnostic companies sued the government. And when there was an administration change, that rule went away. And so at least for the next four-plus years, it's not something that's going to be in the FDA's line of sight.

And so it does mean that most diagnostic companies continue to use the LDT-CLIA pathway to market their tests and not the FDA IVD pathway.

I'd like to switch gears a little bit and talk about how you've funded your company so far and what your experience was in pitching to investors and talk a little bit about that. Yeah, and it's changed over time. So, you know, we actually formed a while ago. We've been alive for 10 years, and that's both a good and bad thing. It's a good thing because, like, that's how long it takes to do novel hard science, right? There's a lot of talk of, like, there should be more research in these white spaces where we don't know anything, but that means you have to

build things from the foundation from the ground up. and partly we pursued both paths of doing NIH funding so we've won four million dollars in NIH grants and continue to apply for more and you know in the early days we did everything from trying to pitch to DARPA, IARPA, BARDA I mean In-Q-Tel like every public entity you can imagine that writes grants in addition to the NIH.

I wish I could show you some of our early white papers as to how we are gonna be useful for the DoD. Very creative strategies. I still think that they're extremely useful.

Didn't really get any bites back then. But in addition to the classic ways of funding novel research of using public mechanisms, we actually found an opportunity in the private sector. And this was generally by investors that are interested in deep tech, you know, which is they have the patience. So generally, private sector dollars can be impatient, right?

They want to see their return relatively quickly. Deep tech tends to be a little bit more patient. They know that they're investing in infrastructure and big swings. And so we deliberately went after deep tech funds who would understand that to say that this might take a while, but it's going to be a big home run if it lands.

And so of the $22 million that we've raised to date, $18 million has been through venture capital, and $4 million has been through public funding.

And what has your experience been pitching? Can you give us some advice, or what have you learned, and what were some mistakes that you made?

Pitching's always challenging, and this has nothing to do with being in the space that we're in. Most people are going to tell you no.

And so in terms of advice, I would always say you've got to go in with that attitude that you're probably going to hear a no, but the intent of the meeting is what you can learn. These are smart people across the table. And from every meeting, you will walk away with at least one question that you should know the answer to or you should think about even if you don't know the answer to. And so they're all learning opportunities and they're all moments to continue to refine your pitch. So one advice is just to get in front of as many people as possible because by the hundredth pitch,

you will it'll be better the pitch will inevitably be better and that's useful in and of itself um the other thing is just it is a numbers game um and you have to um you know if you're going to assume that one percent of the people are going to say yes you're just trying to cycle through people as quickly as possible a mistake that i made early on was getting attached to certain conversations and feeling like i could convert them into believers and usually you can't uh you Usually, if someone has not leaned in and said yes, they're not going to. And so you don't want to waste time.

Your own personal highest ROI is in getting through people so you can find the person who actually is interested in saying yes, and then convincing them.

And so I wish I had done more of that, just being OK with a no and moving on faster. But the other thing is that I also wish I didn't attach, and this is going to sound slightly contradictory, but I promise it's not, is don't attach too much valence to the no meaning as you're around because begins to come together and people are actually a term sheet someone is interested right they someone's priced around go back to the nose and don't spend time there just say hey just so you know there's movement there's momentum and this is just it is just a fact about capital

raising is that momentum begets people suddenly see that there's interest and other people are signing on and whether they have expressed a keen interest a mild interest or no interest that can sometimes flip them to a yes and for the longest time you know if I had worked on somebody and I was like after three meetings I couldn't get them there when something started to come together I didn't go back to them and I should have and again not don't spend time don't perseverate but to say hey by the way this is happening if you want to come on board now would be the time and I think that would

have been a useful strategy earlier I guess we call that FOMO yes missing out yes it is very real I'm gonna stop here and see if there are any further questions hey I was I really want to talk more about like the balancing the nose and so if you hear a lot of nose how do you stay confident that like you're doing the right thing and then how do you balance like knowing that like eventually you might get a yes and so you power through the nose but then also hearing the constructive criticism taking it in and using that to revamp your offer yeah um the open to the revamp is a really

good one this is just a tough question by the way like it's so hard right how do you even know when to throw in the towel right that you're wasting your time and other people's money and you all the things.

The critical part of it which is that you should be constantly adapting to what you're hearing and if people are not responding to X but you have Y in there and you can see that they're sort of like tell us more about why you know flip this script and start you know talk more about why.

We always do things like keep extra slides in the background because you never know what people are going to be interested in and you don't want to have a prefab pitch that you can't adapt right someone says actually that one experiment and you only have one slide in the real presentation, but you have 50 somewhere, have everything ready so that you can respond to that interest is probably the easy feedback of like, of course, you want criticism, you don't have the hubris. You don't walk in with the hubris of saying, I've thought of every question. I know exactly how to answer everything.

I think the how many no's do you hear before you're like, maybe there's nothing here is hard. There is no easy answer for that. I would say that, you know, what helps is having different subject matter experts and advisors in your life that can they can hold up the mirror that they're not as mired in the muck and they're not so overly invested in the thesis as you are, which you're going to be, which is as how it should be. Right. You can't do this job unless like you're overly invested in the thesis and and seek their involvement to say, hey, here's where I'm

at. Can I pitch this to you and you tell me, given your background and understanding of whatever their expertise is, whether there's something compelling here?

And for us, that was very useful to tap into people we deeply respected who had a lot of credibility in their field, who said, you know, I may not understand the path to market, but there's some really, really high potential science here, right? Or talking to a business subject expert who said, you know what, there's probably a way, I don't understand your science.

I can't comment on that. I hope it gets you there. But there is an actual viable path to market if you're able to solve this pain point of critical need in this call point.

And so not everyone has to say that. But you pick your jury of people whose opinion you really trust, again, who have credibility in their field. And if they're able to give you that level of assurance, it was enough for us to say we should keep going.

Hi. Thank you. It's very interesting. I think the whole project's actually interesting. We might have met a year or two ago at Irva, I think it was.

but as someone who does this kind of research over like the last couple decades I agree there's a lot of lacking of data and obviously your focus now is on endometriosis trying to do diagnostics but when you're doing these pitches when you're kind of trying to build the business how do you pitch it in terms of okay well we're focusing on endometriosis now there's all these other data we're collecting we're going to go in these different avenues like what avenues are you looking at how do you pitch that when you're talking to investors and things like that for future growth?

Totally. One is that, back to the first product, is the data asset. For us, we're very clear that we are trying to build relationships now with biopharma to make sure that we're asking them ahead of time, what would you be interested in a data set like this?

So that we don't miss the opportunity to build this data product that has secondary revenue, right? And there are people who have like really successfully monetize this entire product line they will sell diagnostics and then they will generate way more LTV which is lifetime value from the data that they generate from the diagnostics than they ever got from selling the diagnostics and so one useful thing one is that the business model exists to do that two is the others have successfully shown that you can use it and so that helps right I'm not introducing both a new product a new science and

then a new business model like that's way too much risk for people to take on but if you can say hey look there are people who are doing this in cancer very successfully and there is biopharma interests explicitly seen in their willingness to pay that they value that data set the question that comes back to us is okay have you talked to biopharma and seen if there's that same interest in the data set you are building and so one of the things that we do is go out to biopharma and we meet with those people and we say if you had access to this one are you interested it? How would you use it?

And what do we need to do to build it up that's missing or lacking right now that would make you even more interested? And that that shores up confidence that it's not just a grab, like a desperate grab of business models to say we can also do that. It's we've actually talked to potential customers who have indicated that they could find utility and would pay for something like this. So that's first. Second is that I think this is probably the case for everybody, but women in particular have a lot of comorbidities.

And women who have endometriosis in particular particular have a lot of comorbidities. So if you look at our our phenotypes and our metadata right now, these women who have endometriosis also have adenomyosis. They also have fibroids.

They also have PCOS. They also have IBD. They also have lupus. They have RA, right? And so it is not a stretch to show them, to say, we're not collecting an endometriosis data set. We are collecting a data set that's very well clinically annotated for many different diseases, and that's your future pipeline.

And so that's how we try to build a story both on data and indications of how we're going to have access to more than just one disease.

question. We're seeing that data ownership is becoming an issue with some companies that are collecting DNA, RNA sequences, especially when they change administrative hands or shut down. Can you talk about some of the ways your company's thinking about these challenges, and are there plans to mitigate these challenges? Yeah, since this is an online question, I don't know if there's an opportunity to ask for a clarification, but is it specifically that what happens to data ownership when a company shuts down or is it more that what's the sharing policy of this

data I'm gonna I'm gonna pretend that I it's gonna it's gonna go a certain way I mean are you know that the oh that's true that is true that the tricky thing here is is that having a proprietary walled-off data set is an asset in the commercial setting right just is and so it is gonna continue there's just some natural tension there that you can't get rid of.

Where there are opportunities to be rewarded in academia for sharing massive data sets, there are. And there are opportunities to be rewarded in the commercial sector for not sharing those data assets, right? That is like your advantage, that's your competitive advantage.

And so we are a commercial concern. And so right now, part of the competitive advantage and the data moat is that, and that's why the word exists, right? We have a data moat.

It wouldn't be much of a moat if it was readily accessible to everyone and so from I don't again I don't know if I'm exactly answering the question but like I'm just acknowledging that there's a tension that data ownership in a commercial setting means that there is some advantage to not sharing the data and so there's there's two strategies one for indications that we're deeply interested in and doing our own either diagnostic or drug development we would not share and two in areas that we are not that interested in so I was mentioning to Greg today I I think there are

over 500 drugs in development for rheumatoid arthritis right now, 500, right? It's a very vibrant space. Next Gen Jane will probably not be launching an effort to be 501.

And so in those areas, we are interested in licensing the data to the people who are developing these 500 drugs to say, you should be looking at this data to find out if there is a way you can differentiate your drug from the other 499.

And in those instances, we are interested in sharing, right under some sort of productized model it's gonna be a data product that they can have access to and that those are the two ways we think about data ownership and data sharing are there any studies that can be done for women postpartum or perimenopause with these data sets because I think there's a huge gap in knowledge of what's going on with women during these stages or can any information for menopause be gleaned from this data? And then as a side note, they said very interesting work with significant potential. Thank you.

I went through the postpartum first. We're super interested in postpartum. Our ability to try to tackle that is twofold. One, we think that there are signs of your pregnancy health and potential postpartum health in your pre-pregnancy uterine lining, right? That is what ends up becoming the placenta. And so one of the, we are actually a subcontract in a potential RO1 where our objective is to try to collect pre-pregnancy tampons, so menstrual blood prior to getting pregnant, and then follow people to see what their outcomes were, whether they developed

preeclampsia, whether they went on to preterm labor and even whether after childbirth they develop postpartum depression to see if there's anything that we can find that's predictive in this pre-pregnancy tampon.

The second way is we actually have done studies before where we have women who are pregnant wearing tampons for 15 minutes and the intent there is to try to capture both vaginal background and so you're able to get you know, cells that are shed naturally from the reproductive tract in that tampon, as well as obviously a very robust microbiome signal.

There is actually a non -trivial amount of research in vaginal microbiome, just not menstrual microbiome. And so there have been early indications that there are microbiome shifts when you go into preterm labor, et cetera. And so we were interested in that. And so there is an opportunity, not postpartum, but just to continue to understand pregnancy health by using a tampon just as again as a form factor to collect information and then finally you know women do obviously start menstruating again after breastfeeding and so depending on how long they breastfeed whether it's three months

six months twelve months at some point we will have access to that the post pregnancy menstrual cycle as well and so we think that that could be informative to postpartum signals too so that's the postpartum and pregnancy question we are interested in it it's not a priority. All these other things are a priority, but it's obviously something that it could be facilitated, understanding that could be facilitated with this data.

Perimenophosphor, absolutely sure. There is just so much change happening and you know the entire thesis here is just to be explicit is we are capturing the uterine response to the sort of total set of hormonal signaling happening in your body, right?

It's a 28-day in quotes or however long your cycle is, lagging indicator of total hormonal health. And so we do think it is of the right, appropriate access point and aperture to understand whether or not you truly have begun the perimenopause transition.

And so, you know, yes, we actually do follow women in their 40 to 50s. So most of our data is in women from age 30 to 40.

Some of it is in 20 to 30, and some of it is in 40 to 50. We have like two patients who are minors because we think that pediatric menstrual health is an extremely important area It's just really difficult to recruit in that age group, but we continue to try Because we want to build up that data set so it ranges like we are interested in all of these groups in the ways that we could be Helpful the rate limiting factor is always capital Thank you for coming In our audience here, probably a lot of PhD women that are scientists like you and you have chosen to decide

to go into the startup world as opposed to academics. Can you tell a little about your history and how you made those types of choices? Oh, yes, absolutely.

One is I don't have a PhD and I made a deliberate choice. When I worked at BMS, it was very clear that lack of higher graduate degrees would be an impediment to promotion, right? If you looked at all the C -suite, you saw MDs, PhDs, or what have you. And I knew I wanted to stay in this general space, but not necessarily get a PhD or MD. And so I chose what I was jokingly calling earlier the worst of both options, which I got an MBA and a master's, which ended up being a three-year endeavor anyway. If I just tacked on one more year, I could have just done it.

I really appreciated the program, this program that I went through, and MIT. If you could either be getting your MBA at Harvard or Sloan, and they put you through medical classes at HMS through the HSD curriculum, and you took engineering classes at MIT, and you rounded in the clinics, and the thinking was is if you want to be a medical entrepreneur. You don't just need an MBA. You need frontline experience. You need to know the medical jargon. You need to understand the engineering behind biology and human medical systems.

And I completely agree. Coming through that curriculum, it was incredibly helpful in preparing me for being a medical entrepreneur.

Part of the reason in terms of just staying in academia. I paused on it when I was considering doing a startup. One was I did have some hesitation about whether or not what would be your ability to raise non-dilutive funding if you weren't like a professor in an academic setting and maybe that made the decision easier. But I will say even Even when I was first exploring this in 2013, there weren't many academic labs working in this space. There are now, thank God.

You can name the labs both in UK, Australia, and the US that are working on menstruation, isolating organoids, doing single cell sequencing, really trying to interrogate whether or not it is a useful substrate for medical science. but in 2013 they just weren't and so it was this double you know no one's currently working on it and I'm not sure I'm the best person to actually seek public funding right now but I bet I could I could try to convince some of these deep deep tech investors to take a swing at it and that sort of influenced the path that I took the NIH

grant they got was that SBIR it was it was both of So I can't believe I'm the one asking this question, because it's the only question I get from students these days, but what do you see the role of AI in interacting with your data asset?

Yeah, you know, this goes back to this data ownership question, which is like, all of these models are really good. I don't know which ones you guys prefer, but if you do chat GPT or Anthropic or, you know, they're all really good, right? The differentiation in my mind is how good is your data set, right?

And so we are big believers in data quality and these proprietary data sets, which give you the advantage in thinking through how AI could be applicable.

And so we did two things. One is we built what we call an AI-ready infrastructure, which means your data has to be high dimensional. Like it's one of the reasons that we do next generation sequencing and not like targeted probes or PCR, right? We need to generate lots of data.

And so we have, like we've deliberately done that. look we get we get probe down that a lot right we're like why are you doing NGS it's so expensive like is that really the best path to like a marketable diagnostic and for us we're saying yes because the the data product is equally important to the diagnostic that that could initially bring in revenue so that's a choice right to say where we want to generate high dimensional data we spend a lot of time on clinical annotation and that's a choice because the more well annotated your data is the more relevant it is to AI right and and we

needn't right we collect maybe like 2,000 clinical phenotypes like that seems like overkill for understanding some of these things if you you could actually check out some of the annotations we do by going to survey .nextengine.com which all the women in this room should do anyway but anybody can check it out if you are not in a female-born body don't hit submit because they will mess up with our algorithm and our answers you're welcome to go in and just check out the questions that we ask. They range from like, how many miscarriages have you had? Have you ever had a negative side effect to a

particular birth control? And it goes down to the form factor. Like, did you have a worse effect with an IUD versus like a transdermal patch, right? Because that's the level of annotation we want. We really want to know specifics because these specifics actually matter.

We ask about your entire menstrual experience, how long it lasts, how heavy it is, like really detailed annotation. And again, you know, we've gotten some pushback to say the survey thing takes an hour, like why is it so long? And it's because we're making a deliberate choice to say the more well annotated your data, the more sense you can make of it, the more AI ready it is. And the final thing is that we, you know, we developed this female specific ontological framework and it was very deliberate and curated over years and we have found that if you don't include certain labels in your data

sets it can be very confounding and so that the obvious sort of example that I find it's easiest for people to understand is cycle day so when you collect a sample matters to your data analysis I feel like a lot of people focus on sexes of a covariate now which they should it is important, but there are actually many more female-specific covariates that need to be tracked.

There are data now that are showing that when you collect a sample in the monthly cycle is impacted by when you collect beyond fertility hormones. We've always sort of known that, like obviously for like estrogen and progesterone, you want to collect in a specific timeframe because that's where all the reference ranges are built.

But it turns out things like cholesterol, small RNA there's many other biomarkers that are actually impacted by where you are in your menstrual cycle and yet if you look at any clinical data set that's available for analysis right now I would challenge you to find one that provides you with cycle day right and so it doesn't exist and so we were very deliberate in saying find all of these variables that actually need to be labeled in your data set because again AI like any computer endeavor garbage in garbage out and if you do not have the appropriate annotations to make the stratifications

that are relevant you're not really gonna see signal at the end of that process and so in that way we don't have the scale for AI yet obviously we want to get to the billions and trillions of data points but we have the data infrastructure to say we took a long time to set up this infrastructure and part of what we're trying to raise for now is to gas and scale that infrastructure so that the data structure which is ready is scaled to the to the extent that where AI is relevant.

How important is continuous hormone monitoring? You know there's two ways to answer that like if you if you are doing hormone monitoring the only approach should be continuous hormone monitoring because you can't actually make sense of single time point hormone data but I would say that for us the sort of the bigger point from that just how important is timing and longitudinal sampling in any endeavor that involves female biology. I mean, I would actually advocate any endeavor period across men, women should involve longitudinal sampling. It is especially important in female biology

where the hormone milieu is so dynamic and plays such an important role in everything that you see. And so I think anything to do with female biology should be continuous and it's part of the reason that we do so much longitudinal sampling in our own data set and yeah I have one I'm just curious how did you go about building your team so you're like co-founders your board your advisors and all those yeah great questions on all accounts you guys likely know that like the co-founder breakup thing is real like you often that relationship doesn't last it's very fraught and tense and like

you have to be very careful as to like how you pick your co-founders for mine in particular I told this story earlier which is we both were at the Broad I was the operational and financial person on a large GWAS that we were running in West Africa and he was a scientist and so we worked closely together a lot but it was especially in this one trip where we were tracking down convalescent patients who had survived a viral hemorrhagic fever known as Lassa fever and so we went into country in Sierra Leone and we essentially went backpacking over the course of two months and we would go out

to remote villages and try to consent into a study people who had survived this condition and there were just a lot of complicating factors so it was rainy season so there was like actual logistical constraints in terms of getting where you needed to go and like high -stress environments where like you crossed over a bridge and it wouldn't be there on your way back right so you were sort of like in a bind there was cultural influences where blood is very precious and sacred and so going into communities where you don't have cultural competence necessarily and asking for a blood sample

can be not a great thing to do and so that can cause really tense environments with you know there's not like hotels in in the remote areas some of these remote places so we would take tents and we would sleep outside and then finally it was it was 2012 so it was like a election year here in the US and we were because we remembered we were trying to get back to vote and it was election year in Sierra Leone and we got caught in two political riots it's just everything you could imagine that would cause like stress fighting and like extremely high emotions happened during that trip and so

we were like if we could survive this without like killing each other we could probably do a startup together and we we did survive it and so we ended up being like you're I trust you at this point I trust you in my life I might as well trust you with the startup and so we are I think one of the exceptions we've been you know in this journey for over a decade now and still still like happily co-founded so that was how I picked my co-founder and I would just one quick comment on the board your board your board is generally made up of like people who invested in you and then maybe some

independent directors is you cannot overestimate how important it is to find the right partners there you cannot do it and you know like people give a lot of lip service to like make sure you choose your board carefully they're going to be important and you sort of like when you're early in your entrepreneurship journey like yeah yeah yeah check the box yeah I should make sure my board is like on board but really like when you get to moments where like there's a shut down opportunity and your board either believes in you or they don't, right? Like that is when the power of that relationship

and whether you've made sure that they are actually backers of the big vision and not some sort of short-term outcome comes through. And so we've had instances where both advisors and investors where in future I probably wouldn't make that decision again and but like we've also been lucky in that way is and the investors that have been on our board and stayed with us have been long -term partners and like you know been through the hard times understand the big vision and are still supporting us today we are coming up on the top of the hour so there's two last questions

yeah that actually reminds me of my entrepreneurship journey many years ago yeah I do have a question because we're trying to build a business here and I want to put things together put things together and you are saying like you are not doing a DTC so I assume that you are you need to convince physicians you know to actually adopt the product right and then you need to convince the insurance company to pay for it at the back so like if you assume like I'm a physician how are you gonna convince me to actually be confident about your product so I think you know one critical

thing that needs to be considered probably is like okay so you are looking to replace the current golden standard which is surgical diagnosis and so how confident I can be what is you know the sensitivity and the specificity of your product like how much false positive how much false negative I can expect to be confident that to adopt your product yes totally yeah you have to you have to do You have to do the studies that show that. You have to put them up for peer review.

And you have to make sure that it's better than what they're currently have available to them. And so part of even finding your first indication, that plays a role.

If you're going up against a easy diagnosis pathway that is 95% accurate and great in the population that you're targeting, you would have to be really compelling to get them to replace it right so part of you know the the interesting dynamics about endometriosis is is that like why is there a 10-year delay because nobody wants surgery to find out that they have the disease right and so what are people using in lieu of that like teenagers aren't having surgery they're being treated empirically right so what you're going up against an empirical like I think you

have endometriosis is let's try progestin therapy, right? So that's the alternative. And so one is even picking through that morass and saying, all right, endometriosis is broad.

You could go into pediatrics where people have nothing, or they have no way, easy way. And if you think adult women don't want surgery, think about parents and their teenage daughters. They definitely don't want surgery, right? That could have been an extremely valuable call point that we went through.

I mentioned why infertility is an extremely important call point. But REIs don't do laps. They would have to lose the patient to go get a guy in surgery to find out that they had endometriosis. So essentially, they're also treating empirically.

And so fundamentally, you could go to GPs who deal with women who are having chronic pelvic pain, which is a different indication that brings people in.

And so fundamentally, part of that is what do they currently have, and what would they want to see to make this easier for them? But in the end of the day, you're not going to sell a test that doesn't have sensitivity specificity information that's published and so that that's sort of like table stakes you're we're gonna pop we're gonna before we ever get to market there will be a publication that helps them guide them so we we've done we have numbers for our current data set and we're doing our final pre-commercial study which is where we'll establish NPB and PBV.

As your method of collecting menstrual blood because like tampons and other menstrual products often have a lot of chemicals on them how does this affect the biomarker detection process and does it like cause a significant like disruption in that? That's such a great question. There's so much time that we spent go if you saw our early lab days going through like 50 different tampons trying to find one that didn't interfere with downstream sequencing chemistry so it absolutely does tampons add all sorts of weird stuff into their materials and even some like you know

ostensibly cotton organic tampons that were supposed to have nothing else we would we would try them and be like this this is not this clearly has something right and so the you know that's one of the reasons we don't say use whatever tampon you have at home there's too much variability right you have to control for that variable we provide the tampon and the tampon we landed on went through like many rounds of testing to establish that it didn't actually interfere with the sequencing and as like a footnote not many were able to survive our process and so you know it is it

is sometimes a little scary to see like what's going on in these like weird synthetic polymers that that some people are using well I'd like to thank Riddhi thank you so much for coming to Georgia Tech my pleasure yeah thank you so much for what you do thank you