Lantern Pharma Inc. Common Stock The ThinkEquity Conference

Panna. Yeah. Hello, everybody. Good evening. I am going to be introducing Panna Sharma from Lantern Pharma here for this presentation. All right, thank you. Sorry to keep you guys waiting, but let's go ahead and get started. I am going to be talking today about a pretty unique company, Lantern Pharma. We are advancing cancer drugs using our AI platform, and it is a platform that we started building several years ago. We will also talk about what the future of that platform holds in terms of spinning it out into a new entity in the future. We trade on Nasdaq, ticker is LTRN, and today I will be making some forward-looking statements. Like with most forward-looking statements, the actual results could differ significantly. I urge you guys to take a look at our disclosure, both on our website and as filed in our 10-K and on the SEC website.

What has our company and our platform accomplished since we focused our efforts on creating drug candidates using our AI? We have three molecules today in clinical trials that are going on. We have dosed over 100 patients. We have gotten 12 FDA designations ranging from Fast Track and Orphan to multiple rare pediatric disease designations. In fact, I believe one of our molecules has more rare pediatric disease designations than any other molecule out there. It has four of them. We have three additional trials that we have cleared recently to start, one in TNBC, triple-negative breast cancer, which is a terrible disease, one in bladder cancer, which just started in Europe under the EMA, and in pediatric brain cancers, that was just cleared with the FDA earlier this year. We have about 100 patents, both issued and pending.

For us, the cost that we will learn today to develop a new program and bring it to a human trial is $2 million-$3 million, about almost 1/10 to 1/20 the cost of what it typically takes. The way we were able to achieve this was our AI platform. In fact, the platform was really central because it allowed us to ask questions about cancer and how to take that question or idea and transform it into a potential molecule, understand the mechanistic rationale behind that molecule, develop a biomarker signature, and then ultimately target a unique patient population. The one thing that is common across all of our programs is that they are going after a very precision group of patients, where there is a high need, a known rationale, and along with a biomarker. So every single step of our platform was developed to run and make decisions about real drugs.

It was not some theoretical AI data mumbo jumbo. First thing we wanted to do was gather all the potential data across cancer studies. We did that in our platform. We have well over 250 billion data points from cancer studies. We then decided we wanted to create algorithms, models to understand why certain things worked or did not work. Many of the top models out there in the industry are created by Lantern Pharma, including five of the top 12 on the Therapeutic Data Commons. Then we wanted to understand which drugs can be combined together or not combined. We want to understand which biomarkers would give us insight, and then ultimately, how does that translate into an actual patient need?

So again, we weren't just creating protein folding examples, and we weren't just doing a bunch of data arithmetic and math, like a lot of these new data discovery companies are doing. We were really focused on solving real fundamental, real-world drug problems. Our platform evolved. Initially, it was internal only. It was called RADR®. We still use that platform today as the core machine learning platform. Then we decided I was at a conference, and another pharma executive said, "Can't you allow others to do what you guys are doing? What if we wanted to put in a biomarker or put in our molecule?" I thought, you know what, that really could be done. We raced for about a year to create a platform that we launched for the cancer community called withZeta.ai.

You can go there today and subscribe to it, we have a lot of subscribers, to basically do the full arc of development. Once we saw that was successful and scalable and highly functional, we decided we can do this not just for cancer, we can do it for almost any disease category. We created a new company called Open Medicine AI that's 100% owned by Lantern. For those of you that want to participate in Open Medicine AI as we roll that out, the best way to do that is to buy Lantern. It operates a platform with the goal of expanding it beyond oncology and opening it up to the entire world. Let's talk about our cost, because that's really part of our core differentiator, and very importantly, it's how we build. It's not just a marketing claim.

Industry norm is to spend about $20 million-$100 million to develop a new program. By new program, I don't mean the cost of the trial. I mean to go from an idea about a drug or a molecule or a target to actually having something to put into a human. That cost takes very easily $20 million. For us, it's $3 million. The number of years required can be 5-10. For us, it's 2-3, and we've done this now repeatedly with our programs. Let's talk about what the data told us and then where that led to. In our program for LP-184, the data showed that SMARCB1 was really important. That was a data-driven insight that came from the AI. We use that to go into pediatric cancers that we never set out to do.

We didn't set out to be a pediatric cancer company, but we have four pediatric disease designations because that's what the data aimed at. Today, we have four designations, and we have an IND in pediatric cancer that was cleared. It told us that PTGR1 was the prime driver of this molecule, LP-184. What we did is we decided to select patients by that biomarker. Today we have a patent, but we also have a launch in a biomarker, the only dual biomarker in bladder cancer that just launched in Denmark using PTGR1 and DNA damage repair deficiency. Every one of these decisions that the AI and data powered each led to successful clinical milestones for us. In drug LP-184, LP-284, and also in LP-300, which is our drug for never smokers, it is really focused on a group of mutations called L858R.

We narrowed our trial to specifically focus on L858R mutations. Let's dig into our portfolio. We have six clinical paths. Each one of these can be worth hundreds of millions of USD, and our goal is to sell these, of course, to larger biotech and pharma. LP-300 is in phase II. It's in the middle of a phase II program. It's enrolling in Taiwan and U.S. Again, we're focused on patients that have a very specific mutation called L858R, and that mutation shows some great promise. Those patients, unfortunately, don't do well with existing therapies, but with our drug, they extend life. LP-184, that's now launching in triple negative breast cancer. We hope to have some patients this year, or early next year, but probably this year, and we're starting up two sites.

The bladder cancer trial, which is fully paid for, that's in Denmark, that's an investigator-initiated trial, and the only one that's using two biomarkers to select patients. We also have an ongoing trial in LP-284 in soft tissue sarcoma. We have a number of shots on goal. I think this is very important because, again, any of these programs can be licensed or sold or partnered off to pharma companies. In terms of going back to our signal, RADR gave us a very, very clear signal and we followed it, and then we kept following it because it was right. It told us SMARCB1 loss was important. It told us that this drug would have nanomolar potency. We then went to xenografts, and then we moved to actual mice, and these mice lived a lot longer.

We multiplied their, again, it's a mouse, but we were able to extend their life 4 or 5X. These were in very, very aggressive cancers. We said, "This actually has real validity." We put it into patient trials. We had about a 50% disease control rate in that trial in the adult phase I-A, and now we're cleared to also go into pediatric. We also found that our drug, LP-184, now called zirdafulven, had a very unique mechanism, and once that mechanism works, which is basically to switch on the drug, tumors can't repair the DNA. What we found is that the next trial will select based on that biomarker. TNBC, bladder, GBM, pediatric CNS, all these trials, we know that the drug is safe at that dose level, which we're starting at, it was 0.39 mgs per kg.

We also know that there are other ways to make the drug even more active. The indications, again, we have Fast Track indications in TNBC and GBM, and in both of those very aggressive cancers, about 60%-70% of the patients have high levels of PTGR1. This is important because it shows clearly that there is a high chance that the drug will work on those patients and that they meet the right threshold. In TNBC, if you'll take a look at the red lines, you can see that our drug was able to destroy virtually every one of the TNBC models, including those that were resistant in the dark gray to PARP inhibitors. This is not like any other data we've seen in triple negative breast cancer, that basically every one of the models was sensitive.

We also saw this drug worked really well in combination, as predicted by our AI. If you look at our drug, it is synergistic not only with PARP inhibitors, meaning that it works better together versus alone, but it also is very synergistic with IO checkpoint inhibitors. In the 300 trial, what we saw, and this is very important, this is the HARMONIC trial that is going on in both Asia and the U.S., we saw that patients so far have lived about 8.9 months longer, their progression-free survival. If you take a look at our tolerability profile versus what is used today, it is vastly superior. Our treatment related AEs are significantly lower. There is no infusion related reactions, rash, which is a major issue because that causes dermatologic stress, and a dermatologist is involved because of the use of the other option.

What we have also seen is that the longer someone gets this drug, the higher their PFS goes up. So right now, the number in red, 8.9, those people are given six cycles of our drug. We are moving that now to eight cycles, and so we think we will even move the PFS up even higher. Out of the 14 patients that were evaluated, you can see none of the patients the disease progressed, and 10 out of the 14, there is tumor shrink. So we had one complete response, many partial responses, and a lot of stable disease. No one progressed, and in some cases, these responses held for over two years. So we are enrolling the next set of 15 patients. We think the data will continue to be this good, perhaps even a little bit better, because remember, in the next group of patients, we are giving eight cycles.