Ofer Sharon Presents OncoHost at LSI Europe '23

Transcription

Ofer Sharon  0:05  
Good morning everyone and thanks for having me. I'm really appreciate the opportunity to speak in front of you. My name is Ofer Sharon and the CEO OncoHost. OncoHost is developing biomarkers for immunotherapy. The reason we focus on biomarkers for immunotherapy is very simple. There are no good biomarkers for immunotherapy. Today what we have is PDL one and to some extent TMB. A, I would say low accuracy for those not very predictive sample agnostic element to those. And if we look at the current market, we will see that as we speak the 6000 ongoing clinical trials, looking at immunotherapy or immunotherapy combined with something. And if you look at the right hand side of this very busy slide, we can see that basically every tumor type is treated, or will be treated in the near future with immunotherapy. And with pretty moderate outcomes, you're talking about 20 to 40% are benefitting for treatments and would say this is an overestimation. And we are looking at the very near future where clinicians are going to have a lot of choice, and very little guidance on how to treat their patients, on course develop the profit platform. The profit platform combines machine learning proteomic pattern recognition in the plasma, and system biology. It's basically one blood test taken before treatment, liquid biopsy that supports clinical decision making. The way we're dunkers develop product is by focusing first on the clinical question, what is the clinical dilemma that the oncologist is facing, and then we are trying to solve it for them. And I'll show you later how we do it. We launched our first product in the USA earlier this year in February, as you can see with pretty good traction with leading academic centers, high volume centers, all across the country, including academic collaboration with the NIH and the NHS. And those same centers are also collaborating with us on the clinical trial. Clinical Data is on biochars. This is an ongoing prospective clinical trial, open it over 41 sites are coated to date more than 2000 patients. So really high quality of a data supporting the algorithm development. This is not some of those old databases analysis. But really robust clinical information that we can trust plasma base that we do not require tumor tissue very important, especially in the metastatic setting and easily integrated into the clinical workflow. Is it a single blood blood roll taken before treatment, we have clearly based in Cary, North Carolina and permit to sell in all of the United States New York is still pending, it's not us it's them bureaucracy, strongly position. And since it's a platform technology, we'll show that we have a pretty extensive clinical development plan and pipeline for the products. The first product that we launched is profit for non small cell lung cancer. The reason is simple, huge indication 130,000 new patients here in the metastatic setting 80% of them will be treated with immunotherapy first line, and there is a very clean, a very a clear clinical question there that the oncologist is trying to answer. In terms of algorithm development, we have a very robust approach. First of all, I show the results I believe in the next slide. Everything we do is based on band validation, the algorithm is trained and validated are two completely separated independent codes of patients. So we have one shot at success or failure. But what you see is what you get in terms of the performance and overfitting for the product. Second, instead of looking for a signature, we developed a new concept called resistance associated proteins. Those are proteins that are differentiated expressed between the responders and non responders, we identify those, make sure that what we see is indeed signal not noise. And then basically, for each patient, we measure the number of overexpressed reps. Patients that have a lot of those reps have expressed are not likely to benefit for treatment, and they will get a profit negative report. And patient did a few if any, will get a profit positive report and are likely to benefit from treatment. And when I'm talking about benefit from treatment, it's important to note, we are training an algorithm on progression free survival of over 12 months and not on response rates again, because it's more important clinically. So let's have a look at the results. So what we see here is basically our prediction versus the actual outcome of the patient. This was our first validation blinded to 50 patients firstline naive patients treated with immunotherapy or immunotherapy combined with chemo, as you can see almost perfect correlation between the prediction and the actual outcome of the patient with very good P value. We repeated those in another blind validation with a coat of almost 50 patients. Again, very strong performance, very strong correlation of the algorithm in terms of the ability to predict clinical benefit. And the last one, go to follow 600 patient and again the same performance of the algorithm. Then those are three separate advantage validations. The algorithm is very, very stable and performs very, very good. And if we stratify the patients based on the profit result, we see the patient with profit positive. It median overall survival of 26 months patient with perfect negative median overall survival of about 11 months. We're talking about metastatic lung cancer, of course. So we see very clear differentiation. But this is not enough, unfortunately. I mean, many companies would stop there and go to the market, but clinicians have clinical utility. And when you come to a clinician and you give him only big news, it's not likely that they will use a product, especially if it's before treatment. So clinicians asked us okay, so how do I use this tool? So what we did next is we combined the profit result with the PDL level of the patient's, as you may recall, Pdl, one is the current biomarker. This is what is approved, this is what we use. And when we do that, we found something very interesting. Patients at a profit positive and PDL, one of over 50% There is basically no difference in overall survival between immunotherapy combined with chemo versus immunotherapy alone. So for those patients, there's no need for chemotherapy. However, patient at a profit negative and PDL one of over 50% Look at the difference between the immuno combined with chemo and the immunotherapy alone. If treated with chemotherapy, this group of patients will basically live three times more than if they would be treated with immunotherapy alone. So it will will clear, easy to understand and quickly adaptable tool with clinical utility for the oncologist and this is a resolution currently doesn't exist in the clinic clinicians do not see that. We do the same for the other PDL one levels as well to the one to 49 and below 1%. So basically what we did with our first product, we took the existing guidelines and improve the resolution for the clinicians instead of two groups of patients. We have six with clear commendation for each group of patients. And the report is very, very simple, because market research told us that we have 10 seconds or less in the clinic, profit positive negative, how positive how negative even by the prophets call, and then what to do in all three situations. So in less than 10 seconds, you can basically recommend something for your patient. very extensive pipeline. As you can see, we are working on more indications, earlier stages of disease in some interesting new technologies is not in the scope of this step, but I'd be happy to discuss them after that. The test is stalled in the US as a laboratory developed test. The lease price is $5,000. We have a MultiPlan approval rate of $4,000. And we are aiming for CMS rate of 3800. This is accepted expected by the end of next year. Very good commercial traction. This is basically the work of one man and we have a team of six just started. But over 150 tests sold in a very high order rate by declination and we just got approval for our own exclusive and PLA code for the profit this is going to be effective of as of January 24. We executed our first deal with a pharma company funded as a clinical utility trial. Very cool 94% of oncologist change their management based on the profit and destroy which I think is pretty unique for new technology. As I mentioned earlier, we have no results in more indication and we are going to expand those and take them to the market soon. Strong Scientific Advisory Board supporting the company on the oncology side on the biology side and on the data science side, but also a commercial advisory board with a leading individuals from the industry to support the process. We are currently raising a SAFE round of up to $15 million in the to take us to the end of 24 beginning of 25 and then we are planning on a crossover round of a larger amount at the end. Happy to discuss the details of the round. Later on after the talk. Thank you I appreciate your time and your activities.

Transcribed by https://otter.ai

Ofer Sharon  0:05  
Good morning everyone and thanks for having me. I'm really appreciate the opportunity to speak in front of you. My name is Ofer Sharon and the CEO OncoHost. OncoHost is developing biomarkers for immunotherapy. The reason we focus on biomarkers for immunotherapy is very simple. There are no good biomarkers for immunotherapy. Today what we have is PDL one and to some extent TMB. A, I would say low accuracy for those not very predictive sample agnostic element to those. And if we look at the current market, we will see that as we speak the 6000 ongoing clinical trials, looking at immunotherapy or immunotherapy combined with something. And if you look at the right hand side of this very busy slide, we can see that basically every tumor type is treated, or will be treated in the near future with immunotherapy. And with pretty moderate outcomes, you're talking about 20 to 40% are benefitting for treatments and would say this is an overestimation. And we are looking at the very near future where clinicians are going to have a lot of choice, and very little guidance on how to treat their patients, on course develop the profit platform. The profit platform combines machine learning proteomic pattern recognition in the plasma, and system biology. It's basically one blood test taken before treatment, liquid biopsy that supports clinical decision making. The way we're dunkers develop product is by focusing first on the clinical question, what is the clinical dilemma that the oncologist is facing, and then we are trying to solve it for them. And I'll show you later how we do it. We launched our first product in the USA earlier this year in February, as you can see with pretty good traction with leading academic centers, high volume centers, all across the country, including academic collaboration with the NIH and the NHS. And those same centers are also collaborating with us on the clinical trial. Clinical Data is on biochars. This is an ongoing prospective clinical trial, open it over 41 sites are coated to date more than 2000 patients. So really high quality of a data supporting the algorithm development. This is not some of those old databases analysis. But really robust clinical information that we can trust plasma base that we do not require tumor tissue very important, especially in the metastatic setting and easily integrated into the clinical workflow. Is it a single blood blood roll taken before treatment, we have clearly based in Cary, North Carolina and permit to sell in all of the United States New York is still pending, it's not us it's them bureaucracy, strongly position. And since it's a platform technology, we'll show that we have a pretty extensive clinical development plan and pipeline for the products. The first product that we launched is profit for non small cell lung cancer. The reason is simple, huge indication 130,000 new patients here in the metastatic setting 80% of them will be treated with immunotherapy first line, and there is a very clean, a very a clear clinical question there that the oncologist is trying to answer. In terms of algorithm development, we have a very robust approach. First of all, I show the results I believe in the next slide. Everything we do is based on band validation, the algorithm is trained and validated are two completely separated independent codes of patients. So we have one shot at success or failure. But what you see is what you get in terms of the performance and overfitting for the product. Second, instead of looking for a signature, we developed a new concept called resistance associated proteins. Those are proteins that are differentiated expressed between the responders and non responders, we identify those, make sure that what we see is indeed signal not noise. And then basically, for each patient, we measure the number of overexpressed reps. Patients that have a lot of those reps have expressed are not likely to benefit for treatment, and they will get a profit negative report. And patient did a few if any, will get a profit positive report and are likely to benefit from treatment. And when I'm talking about benefit from treatment, it's important to note, we are training an algorithm on progression free survival of over 12 months and not on response rates again, because it's more important clinically. So let's have a look at the results. So what we see here is basically our prediction versus the actual outcome of the patient. This was our first validation blinded to 50 patients firstline naive patients treated with immunotherapy or immunotherapy combined with chemo, as you can see almost perfect correlation between the prediction and the actual outcome of the patient with very good P value. We repeated those in another blind validation with a coat of almost 50 patients. Again, very strong performance, very strong correlation of the algorithm in terms of the ability to predict clinical benefit. And the last one, go to follow 600 patient and again the same performance of the algorithm. Then those are three separate advantage validations. The algorithm is very, very stable and performs very, very good. And if we stratify the patients based on the profit result, we see the patient with profit positive. It median overall survival of 26 months patient with perfect negative median overall survival of about 11 months. We're talking about metastatic lung cancer, of course. So we see very clear differentiation. But this is not enough, unfortunately. I mean, many companies would stop there and go to the market, but clinicians have clinical utility. And when you come to a clinician and you give him only big news, it's not likely that they will use a product, especially if it's before treatment. So clinicians asked us okay, so how do I use this tool? So what we did next is we combined the profit result with the PDL level of the patient's, as you may recall, Pdl, one is the current biomarker. This is what is approved, this is what we use. And when we do that, we found something very interesting. Patients at a profit positive and PDL, one of over 50% There is basically no difference in overall survival between immunotherapy combined with chemo versus immunotherapy alone. So for those patients, there's no need for chemotherapy. However, patient at a profit negative and PDL one of over 50% Look at the difference between the immuno combined with chemo and the immunotherapy alone. If treated with chemotherapy, this group of patients will basically live three times more than if they would be treated with immunotherapy alone. So it will will clear, easy to understand and quickly adaptable tool with clinical utility for the oncologist and this is a resolution currently doesn't exist in the clinic clinicians do not see that. We do the same for the other PDL one levels as well to the one to 49 and below 1%. So basically what we did with our first product, we took the existing guidelines and improve the resolution for the clinicians instead of two groups of patients. We have six with clear commendation for each group of patients. And the report is very, very simple, because market research told us that we have 10 seconds or less in the clinic, profit positive negative, how positive how negative even by the prophets call, and then what to do in all three situations. So in less than 10 seconds, you can basically recommend something for your patient. very extensive pipeline. As you can see, we are working on more indications, earlier stages of disease in some interesting new technologies is not in the scope of this step, but I'd be happy to discuss them after that. The test is stalled in the US as a laboratory developed test. The lease price is $5,000. We have a MultiPlan approval rate of $4,000. And we are aiming for CMS rate of 3800. This is accepted expected by the end of next year. Very good commercial traction. This is basically the work of one man and we have a team of six just started. But over 150 tests sold in a very high order rate by declination and we just got approval for our own exclusive and PLA code for the profit this is going to be effective of as of January 24. We executed our first deal with a pharma company funded as a clinical utility trial. Very cool 94% of oncologist change their management based on the profit and destroy which I think is pretty unique for new technology. As I mentioned earlier, we have no results in more indication and we are going to expand those and take them to the market soon. Strong Scientific Advisory Board supporting the company on the oncology side on the biology side and on the data science side, but also a commercial advisory board with a leading individuals from the industry to support the process. We are currently raising a SAFE round of up to $15 million in the to take us to the end of 24 beginning of 25 and then we are planning on a crossover round of a larger amount at the end. Happy to discuss the details of the round. Later on after the talk. Thank you I appreciate your time and your activities.