Lantern Pharma Inc. (LTRN) Earnings Call Transcript & Summary

Welcome, everyone, to the Diamond Equity Research conference. With me, I have our next presenter, Panna Sharma, the Chief Executive Officer of Lantern Pharma -- Lantern Technologies (sic) [ Lantern Pharma ]. And Panna, I'll pass the mic over to you to present.

Hunter, thanks a lot. I appreciate it. And thanks everyone for taking time today to listen to our story. Lantern Pharma is a publicly traded company. We trade on the NASDAQ under the symbol LTRN, and we will be making forward-looking statements today that we may or may not update and are obviously governed by the safe harbor statements. So I'd like to go ahead and give everyone who's new to the story, a quick overview of what would we do, talk a little bit about our platform that's very unique, our drug portfolio that we're developing and some of the milestones that we will be hitting over the next this quarter and, of course, next year. I think there's a lot of upside to the story. And more importantly, today, more than ever before, there is environment in which artificial intelligence and the use of big data is fundamentally transforming or has the ability to transform drug development. And that's what we're doing. There -- I think almost everyone in the audience has probably heard of drugs that have failed or they get abandoned in late stage. And it's not necessarily because these are bad drugs. Oftentimes, it's because there's not enough known about the mechanism of action, the types of patients that will respond to the drug and oftentimes, not enough is known about what combinations can be used with that drug or compound to better impact cancer outcomes. And so our AI engine analyzes and develops correlations across large-scale data sets that we're constantly inputting and cleaning and inputting and reviewing. And then looking at, is there a population of cancer subtypes and population of potential patients that will respond to that compound. And we can do that in a matter of months and weeks, and therefore, really shorten the discovery, lead optimization, hit development and getting it back into a trial. And because of this reduced risk and shortened time line, we believe that we have a transformational process for drug development. So today, we believe that oncology drug development is not only costly but very inefficient. And there's tons of data. So really, it's a perfect problem area for AI and machine learning. The average success rate of oncology development is about 3.5% -- under 3.5%. There's thousands and thousands of trials that can be mined. And the costs today are pretty significant, approaching $3 billion to bring an oncology drug to market. And so AI has the ability to transform the time line, the cost and insight. In fact, every week in the press, whether it be Fortune or Newsweek or Scientists, you're reading about medicines being increasingly driven by AI and machine learning. And oncology is not the only area, COVID-19 vaccine development would not have happened in the year or so that it's happened, if it weren't for computational virology approaches. So we're seeing the ability for AI approach to totally change and tackle medicine and biology. This comes at a time where we feel that they -- it's really critical. If you look at ROI in the top pharma companies, the ROI is actually negative -- almost negative if you look at the cost of capital, 1.8% last year according to Deloitte. And the reason is that it's 2 primary things: that the patient subgroup is not properly stratified. So the strong therapeutic effect that would be seen in one patient group gets diluted out. And so you don't get the kind of overall response rates or survival that you really should in a properly organized trial. Secondly, the genetic changes that are involved in cancer development are just rapidly increasing. We're learning so much more every month, every quarter. And so when a trial gets launched or a compound gets developed, you have to really understand the genetic basis of that drug and the catalog of genetic changes that are involved in the mechanism. But -- and oftentimes, it's not done in oncology drug development today. Our AI platform solves both of these problems that we believe are fundamental in oncology drug development, which is understanding patient heterogeneity and going for the right stratification of patients. And secondly and very importantly, understanding the genetic or biomarker basis that can -- where you can leverage the weakness of the tumor. This allows us for greater success, improved trial design and reduced cost. And so that is the core of what our platform does. These types of approaches, industry analyst say, will make up almost 40% of the global AI health care market. So 40%, imagine, 80% of the market being spent in drug discovery and drug development in health care, which is an outsized number. And it's going to be driven largely by smaller emerging companies like ourselves and like BioXcel and Schrödinger and Recursion. And we were actually featured in Fortune Magazine earlier this year in February. And also in ZDNet, where they talked about how our AI engine is leveraging your own networks to actually develop signatures quite rapidly. So this innovation is happening today because of a lot of tremendous emerging technologies and industry trends. You have more data available today than ever before, validated genomic and biomarker data, you have the ability to share and collaborate globally in industry consortiums. You have decreasing cost, but also increasing quality of getting genomic and biomarker data and generating it on your own. A big percentage of the RADR data is actually our own proprietary generated data from our own studies. And at the same time, you've got AI and machine learning technologies that are becoming more powerful and much more implementable. And so all this drives to being able to then use this to rescue and reposition and develop drugs for very specific indications in oncology faster than ever before. And at the same time, in oncology, you've got the acceptance of precision medicine and genomics. So this is really why we develop RADR. We think RADR solves a significant problem and allows us to have a platform in -- and a platform that actually showcases how oncology drugs can be developed. So where are we today? Our focus on advance is really on taking the platform to advance our own portfolio, places drugs that we own the therapeutic rights to. So we're not doing this as a service or a tech provider. We're really doing this to develop potentially therapeutic assets that are worth hundreds of millions or billions of dollars. Our platform today is over 1 billion data points, 1.1 billion actually, that covers the -- over 140-plus drug-tumor interactions. It's been validated in published studies at ASCO and AACR. There are some links to them in the end of this presentation. We are one of the few companies that has an active collaboration with the NCI, where we're constantly looking at new therapeutic classes. And we're using it today to accomplish some very important things. Number one, our signature for LP-100 was driven -- that's our first drug asset, was driven in large part by the RADR engine. Our mechanisms of action for LP-300, which has shown notable and statistically significant results in prior trials also was uncovered using RADR. Our pathways for LP-184, which is now being used in 2 different cancer indications, again, significantly delivered by RADR and saved a lot of time and money in getting to the additional indications. We're also now beginning to use RADR to identify new drug candidates, which we believe we can then deliver shareholder value and develop programs for. So what is the future of this engine? And what kinds of data are incorporated into it? Again, we're hoping to achieve 1 billion data points before the end of the year. We're at that number as of last month. We expect to get to 3 billion to 5 billion next year and over 10 billion the following year. And the types of data we have is transcriptome data, gene expression data, drug sensitivity data, copy number and mutation data and then also information about the patients, age and race and then also their prior treatment history and response. And we get this data not only from our own studies, but also from a clinical history of the drug from proprietary sequencing campaigns that we're doing, from our collaborations that we've announced, such as with Fox Chase and Georgetown or Sloan-Kettering and then, of course, from public sources, where data is available. And all this data then is tagged, curated, cleaned, scored. All these data sets are -- then we have algorithms that went on top of that these data sets to give us insights about how these drug -- our drugs will work in monotherapy, in combination therapy, and then more importantly, what kind of signatures can we take to market? So what is our portfolio? Where are we pointing this powerful platform today? We're pointing this platform in our portfolio. Our most advanced asset is in metastatic hormone-refractory prostate cancer, that's in a Phase II trial being managed by Allarity Therapeutics, our partner, and they're paying for the trial and obviously have a lot of the economic rights. That was a deal done prior to the current management being at the company, but it really put the company on the map in terms of being able to use genomic approaches to rescue a drug that had failed a prior Phase III trial. We also have LP-300 today that we're launching into a Phase II for non-small cell lung cancer, but it's a very specific population, a population of never smokers that has a biologically different disease and reacts differently to different compounds. And then 2 programs with LP-184 are early-stage asset. It's in certain solid tumors that are genomically defined to overexpressed PTGR1. And we have some great data on that coming out of our Fox Chase research that we're doing. And also in glioblastoma, which is a new indication that was predicted by our engine. And actually, this time last year was not even in our literature. So this is an indication that in late December, early January, we really reviewed. And then in Q1 and Q2, we developed and now it's actively in a program. And more importantly, we'll be announcing additional data, very specifically in the glioblastoma program and how we can potentially fast track that. So this entire portfolio and the platform is really being accelerated because of the RADR engine. It is 106 issued patents, 8 pending applications across about 14 patent families. So let's take a look quickly at the market for this. We believe that even though these are very targeted indications, there's nearly 1 million patients that fall into this. This is several billion dollars in therapy sales. So as we add new indications, one of the most important things to look at in a biotech company is how much time does it take them to get to a new indication with an existing asset? How much time and money does it take them to launch a new program? And we believe because of our RADR engine, we can get the time to indication and time to new program much more rapidly with less risk, more certainty and with greater speed. And so we think these are very important measures going forward. So today, let's look at our molecules that we're developing, 100, 300, and 184, all small molecules. 100 is a DNA damaging agent. Again, this was a failed Phase III trial in pancreatic cancer, but we think this drug really needs to be pointed at metastatic castration-resistant prostate cancer. And in fact, there are some new interesting literature that's come out on this drug also and DNA damage repair-deficient cancers as well. And so that's going to be an area that we're also exploring. LP-300, which is, again, in non-small cell lung cancer. It works in multiple mechanisms. We think by modifying sustains on proteins that are involved in the non-small cell lung cancer pathway, and it also reduces toxicities associated with platinum and taxane chemotherapy, so it allows those drugs to work more potently. And then LP-184, which is a novel DNA damaged agent, it has nanomolar potency. And where we're pointing, it has been predicted by our RADR AI engine, and we validated and published a gene signature. And there are some links in the appendix of this presentation that will give you some reading material and publications on the signature. So 100. How is it derisked? Number one, it's been tolerated in historical trials. It's currently in a Phase II trial. First patient was dosed nearly in December of 2018. That trial is being managed by Allarity Therapeutics. And we have economic rights to that, up to $14 million in the future or about 26% of any future value. And more importantly, there's patents directed to the use of that drug with the signature in metastatic prostate cancer that allows us to extend the patent life through 2036. So this is an important asset and more importantly, really showcases how a signature can be used to actually guide a trial design and also guide the enrollment of patients. Again, this is historical data. Very important because in this data set, you can see the median 1 year survival. This is without a signature, was 86% greater and with -- because of Irofulven. And with the signature, we think we can improve this. And this trial failed because this indication was not pursued in Phase III. The indication that was pursued was pancreatic for whatever reasons. And again, this goes to the earlier point that I made, which is many of these trials failed not because the drug is bad, but because the wrong indications are chosen or lack of knowledge about the patient population. And here is a very clear case, which you had compelling data, but pancreatic was selected because there wasn't sufficient knowledge about the biomarker basis of this drug. Similarly, in LP-300, which is our second asset that's getting ready for a Phase II. And again, this failed to Phase III, not because there were bad results, but because it didn't affect enough patients. And the reason, again, has to do with stratification. How do we stratify and pick those patients that are most likely to respond? What do we understand about the drug's mechanism? And how can we use data to really feel strongly enough to derisk the asset. So the asset is very safe. It's been over 10 trials historically in both lung and breast cancer. We believe a non-small cell is the best area to point this drug, but more importantly, targeting never smokers. And so now we've designed a Phase II trial. And we're leveraging RADR to develop a biomarker signature that we can use as a basis for selecting those patients. Let's take a quick look at some of the reasons why. So this is a historical trial, Phase III trial, and this actually also showcased the same things that were found in the Phase II is that it only had a very modest effect in 2-year survival in lung cancer for the overall group. But if you look at never smokers, which are on the far right or female member smokers, there is a tremendous increase, 125% increase in 2-year survival compared to the standard of care, which was just cisplatin and paclitaxel in the gray bar that -- and had the trial been designed to go after that patient population, this drug would be in the market today. And so this was also seen in the Phase II data as well. If you look at the Kaplan Meier curves, very similar in the Kaplan Meier curves, you see a tremendous separation between the [cisplatin/ paci ] group versus LP-300, especially in female never smokers and in never smokers, tremendous p-value and the potential to get this survival So our goal is how do we really feel strongly about the biology of this disease. And so we are going to select never smokers, validate it through our signature and validate that they'll be sensitive to the combination of these drugs and do a Phase II where LP-300 plus the existing standard of care is used to potentially improve the overall survival. And the interesting thing is in this group of nonsmokers or never smokers, actually, the ironic thing is that the survival is not that great, and they don't respond to PDL1 therapy. So there's a real clinical need. And most of the clinicians that we've spoken to have suggested this is becoming something they're seeing more frequently. And so as we leverage these historical results, we're also looking at a time line where we're meeting clinical need. So adenocarcinoma in non-small cell lung cancer patients and never smokers is now the seventh leading cause of death. It occurs much more frequently in women than in men. And there's a very different biological and mutational difference. And so we're actually working now to see if we can also get an orphan status for this indication and this population because we feel that this is a very different population. In addition, we've also filed to extend the patent on this drug to use in never smokers in non-small cell lung cancer that would give us patent protection until 2039 or 2040. So that's a key part, again, of our strategy is to find insights, replicate them in a signature and use that signature and the drug to go after very specific populations. LP-184, this is a new drug. So this is not -- does not have the historical clinical experience that the other 2 drugs have. But we know that where PTGR1 is overexpressed, and this occurs in a number of solid tumors, including glioblastoma sometimes that this drug works really well. And so we'll have more data this quarter on the blood-brain barrier profile of this drug. We did some work immediately after the IPO to start validating our AI-driven hypothesis. We have a number of collaborations where we're also validating this drug in other solid tumors. We announced one in pancreatic cancer with Fox Chase and in prostate cancer with Georgetown. We've also published on the broad antitumor effects of this drug, especially in aggressive cancers that are multidrug resistant. And again, as I mentioned, RADR engine also showcased that this had a very promising nanomolar potency, which we found was true in a number of studies. If you look at the relative tumor weight, when we dosed this drug in both 10 and 20 milligrams per kilo with a mouse that was engrafted with a lung cancer, and this was a multi-drug-resistant tumor, you can see that it almost went to 0, the relative tumor weight. So it's a drug that we're very excited about. We think it will have significant potency. And it's because of the way the drug works specifically in terms of inducing DNA damage. So this drug works across a number of solid tumors, much better than the existing standard of care. And if you can see in the green, which is much, much higher potency, several thousand times more potent than the existing chemotherapeutic regimens that are used. And the same is true in glioblastoma, which was predicted again by our engine. And when we went to the lab and replicated it, which we did recently, we found that both aspects of the engine predicted, which was the potency and its ability to cross blood-brain barrier were, in fact, true. So this allows us to really go after a very important indication where median survival is less than 1 year, and this is in glioblastoma. Right now, the existing drug of choice, TMZ, temozolomide, really only works well in about 50% of patients. So this rapid development for a young company in terms of our portfolio is driven by our RADR engine. We think we can generate insights to developing new indications in a matter of months as opposed to years. And more importantly, launch programs with only a few million dollars instead of tens or hundreds of millions of dollars. This engine will continue to grow and expand. And as I mentioned, we're in about 1.1 billion data points now, over 50,000 patient records, covering over 144 drug tumor interactions, and we have a pretty aggressive plan in for next year as well. For those that are interested have included details on the actual workflow, but the core is to come up with a tuned, optimized, predictive response model. We have machine learning algorithms that compete across different data sets and different parameters for training to give us that. And then we use that model not only to guide preclinical activity and clinical development, but also then to guide the signature that's used to give commercial value to the drug asset. So all this is protected by a significant amount of IP. We have over 108 patents issued across 14 patent families that cover composition of matter, how we're using it and what patients and then response signatures and biomarkers. So we have collaborations in multiple institutions. It's not just enough for us to believe in it, but we believe that doing this collaborative work allows us to have an asset-light biotech model, where the majority of our lab work is done in partners labs that are experts in these cancers and also can serve as potential clinical sites in the future. So that, I'll talk a little bit about the milestones. There's a lot of value-building milestones that we have coming up next year, including some launches of our trials. And more importantly, also continuing to grow RADR and pursuing big pharma partnerships. And we'll also have data coming out for the launch of the Phase I and Phase I/II for LP-184 as well as the potential readout from our existing metastatic prostate cancer trial. So we believe that will be very milestone rich over the next several quarters, and we may also secure additional compounds for development. We have about 7.3 million shares on a fully diluted basis, 6.22 common shares that are outstanding. And more importantly, a team that has not only assets in terms of a portfolio that we think can be any one of these assets can be worth several times more than our market cap, but also a platform that can generate new insights and new programs quite rapidly. So with that, I'll leave sometime, hopefully, for some questions from the audience.

Thanks, Panna. So yes, we do have a couple of questions. So one was, can you tell us roughly how big of the time and money savings the AI engine can save relative to traditional oncology, drug discovery and development, so sort of what the delta would be or how you view that?

Sure. Let me give you a couple of data points. And I'll talk generally and give some Lantern specifics. So one of the very first things that's very important in oncology drug development is the creation of the signature that correlates to response, a companion diagnostic or patient stratification tool for these trials. Nearly 3/4 of trials today are using this type of approach. Historically, this takes about 18 to 24 months and costs in the range of $10 million to $15 million. To do this using the Lantern approach, it would take 3 months and there would be a couple hundred thousand. So that gives you one data point. Second is the time to new indication to develop insights about a new indication, whether it's monotherapy or combination therapy for a compound, that process can take three years, and it can cost anywhere low end of $5 million, $6 million, high end of $50 million traditionally. But timed indication is a 2-, 3-year process. And with our approach, that timed indication process can be six months. And obviously, significantly lower cost because a lot of that is people in time. So these are 2 benchmarks. So if you think about a program, getting it through a Phase II and ready for a Phase III to partner out, that cost typically in the pharma world has been close to $200 million plus. In our approach, we believe we can do that for probably close to 1/8 of that or 1/10 of that. And in a time line, that's probably 1/2 of what's being done today. So very significant.

Great. No, I appreciate the response. So the other question we had was on the RADR platform. So you mentioned the 1 billion data points, then 2 billion, 5 billion, 10 billion. Can you just explain how investors should interpret that retail institutional? What that sort of means and how they can gauge sort of the advancement of that.

Yes, great question. So one, the more data we have, the more robust our signatures will be when we go into using them to stratify for trials or ultimately in the commercial setting. So the more robust the signature you have, the more you derisk the asset and most likely to also ensure the highest response rates. So that's very important. Second, as we increase the number of data points, we'll get more knowledgeable about different cancers. So we'll be able to find new subtypes of cancer where our drugs can be used, either stand-alone or in combination therapy. So that's very important because those new programs can add hundreds of millions of dollars of value. And then third, we also, as we add additional data points, we find new compounds that may be repurposeable or new compounds that we should go after that have been abandoned. So those can be entirely new programs that we can develop and license out. So those are the reasons why we continue to expand more robust signatures, more subtypes of cancer that we can target and then most -- and then new programs entirely.

Great. No, I appreciate the clarification on that. So I think that's all we have in terms of the questions, and I think we're approaching the end of the presentation time frame. So if anyone has additional questions, they can reach out to the Diamond Equity Representative. We'd be happy to connect you to the CEO or management team for additional questions. And Panna, where can they find additional information on Lantern Pharma and about the company.

lanternpharma.com is probably the best place to go for additional information about us, or obviously, you can look up our ticker, LTRN and then use that to get to our website. But our website has a lot of information on our compounds, including also white paper on our technology.

Okay. Perfect. Thank you very much for the presentation. I appreciate you supporting Diamond Equity.

Thank you, Hunter.

Have a good day.