Addressing a Shortage of Genetic Counselors with AI

As genetic testing continues to expand, it is bringing into focus a shortage of genetic counselors who can work with patients to explain results and answer questions. Igentify is helping genetic counselors manage more patients by providing an AI-based platform that can take some of the load off of them by helping onboard patients, obtaining consents, and educating them. We spoke to Doron Behar, co-founder and CEO of Igentify, about the company’s Digital Genetic Assistant, how it works, and why it will allow genetic counselors to handle a much larger volume of patients.

Daniel Levine: Doron, thanks for joining us.

Doron Behar: It’s a pleasure to be here Daniel, thank you very much for having me.

Daniel Levine: We’re going to talk about Igentify, its digital genetic assistant, and how it works. Let’s start with genetic counselors. Though I imagine most listeners will have a vague notion of what they do, can you explain the role of a genetic counselor?

Doron Behar: Absolutely. So, I must say the genetic counselors are starting to be one of the most important roles—that they might affect the future generations and the way we treat medicine. So genetic counselors are this intermediate between the bench side and the bedside, which means they’re the one responsible for explaining to patients what their genetic results will mean for them throughout their lifetime. And genetic counselors really have two major roles. Role number one is that they will play a major part of the future of precision medicine, the future of genomic medicine, the future of prevention medicine. Because one of the most important things that our genome can do for us is to identify potential risk. And while doing so, we can warn the individuals whether it is around reproductive health, whether it is about risk for early onset malignancies, whether it is about using certain drugs that will do better for the patient. And genetic counselors have to be the one that are making this information, making the tests, making the explanation available for them.

Daniel Levine: When do genetic counselors typically interact with patients and is this generally a one-time interaction or is there ongoing contact they have with patients?

Doron Behar: So genetic counselors can interact with patients in two different scenarios. Scenario number one is when there’s a healthy individual that we want to prevent a certain form of disease that is common in the population. So for example, genetic counselors will be the one that will explain to reproductive couples planning a family before pregnancy about what are the modalities that are available today to reduce the risk for a child from different long lists of diseases. So that is completely healthy individual and that is the major role because that is really affecting the entire population. So basically, what I’m saying is that each individual needs to see a genetic counselor under given medical circumstances. The second time is when there is a sick state, if there is an unfortunate situation of a malignancy or a sick kid in the family and people want to understand whether there is a genetic risk, or a genetic cause for the disease in order to prevent it from other family members in future pregnancies and so forth. So genetic counselors really see eventually the entire population, whether in the healthy state or at the sick state.

Daniel Levine: We’ve seen a large upswing in genetic testing. One of the consequences of this is that there’s a shortage of genetic counselors. How big a problem is this today and how does the supply of genetic counselors today measure up to the demand?

Doron Behar: So genetic counselors are really in a very tough situation. Now, there are only 7,500 to 8,000 genetic counselors worldwide. Not all of them are working as genetic counselors currently. And this is a scant amount of people. If you compare to the 8 billion people worldwide growing steadily up, and since I just said that, we are expecting that each individual will have to go through genetic counseling several times throughout one’s lifetime. Again, reproductive health risk for cancers, pharmacogenomics, and so forth. Then we are speaking about, let’s say 8,000 genetic counselors serving 8 billion people. This is impossible. But if you further look into this, then you will learn that in North America we have roughly 4,500 genetic counselors. So as we look to different places, the bottleneck is becoming really severe, really impossible to meet by human face-to-face meeting. And this is why we think that the future of genetic counselors must be related to technology that will allow each genetic counselor to supervise a hundred times more patients. Because again, we said that the vast majority of the role of genetic counselors is to help healthy individuals not get a disease. So, this is where I think that huge demand for technology can appear because as you can see, the magnitude of the bottleneck is really unparalleled in medicine. We can take better numbers to make it even more vivid. If you have roughly 4.5 million live births in North America and I’m saying that there are 4,500 genetic counselors, then you see that only for the purpose of explanation before pregnancy, completely healthy situation, each genetic counselor must see 1000 people every year. That is only for this particular cause. As you can see, that is a huge task. And genetic counselors are currently under a lot of demand and their number is simply not enough.

Daniel Levine: Igentify has developed what it calls an end-to-end platform to streamline the genetic pipeline and efficiently facilitate any screening service. Where does Igentify sit within the world of genetic testing? Who are Igentify’s customers?

Doron Behar: So, Igentify’s customers are either laboratories that are serving patients, so they are both producing genetic data at the lab and then take care of the genetic counseling sessions, and of course any other healthcare provider that is directly involved with genetic explanations and the genetic counseling. And basically, if you can look at an example of reproductive health, then we are going, again to what we know today, which is spreading quite fast, preconception screening, which means the ability to offer completely healthy couples without any family history of any disease that is genetic the ability to find out whether there are carriers for a disease or not, that is usually referred to as an autosomal recessive disorder, like cystic fibrosis, like Tay Sachs, like thalassemia, and so forth. And Igentify is here to help them in the full process. And we have to remember eventually we are looking to help the young people that have to go through a medical interview, have to then get an explanation about what is preconception screening, what is the test that is suggested to them based on the profile that they have reported on. They have to consent. You can’t just give a genetic test to a patient without having a consent. So Igentify takes all of this part, makes it very easy to scale up dramatically on the number of patients that one genetic counselor can supervise because the genetic counselor will dictate the exact digital interview that will be taking place. And then the computer, the machine, will generate on behalf of the patients, dynamic and personalized multimedia that will refer to them specifically to their medical history, will even speak in different languages, English or Spanish for example, and then will allow the consent process to be done. But this is only the beginning, then the sample is going to the genomic kitchen and there is some tests that are being done in this kitchen. And then there are the results. The results can also be interpreted by Igentify and then the results, which are interpreted, are being combined with the clinical information that we have collected before and we will now sort the patients from those that are at low risk, high risk, or average risk. So that is something very important. There is never no risk at genomics. There is only low risk, but all the low risk patients can now be sorted out. And again, if certain criteria that were pre-decided upon by the supervising genetic counselor are being met, which means both individuals that are being screened for preconception screening are both completely negative, let’s say then maybe we can immediately now shoot off a digital genetic counseling using again, dynamic and personalized multimedia content that will speak to the patients about their own situation. We’ll explain exactly what is the situation, we’ll explain that the risk was reduced, not eliminated, that was reduced. What are the other modalities that are possible? And allow the genetic counselors to have their time used to look to the patients who are at a high risk or are at actual risk detected by the genetic test that is going to have a huge effect on the population. And the good thing Daniel, is that now we know both from scientific studies that we did and from our production data that this solution is very well accepted, appreciated, and actually preferred by patients and genetic counselors for this group of low risk patients. So that’s where we sit with respect to how to help genetic counselors beef up and on their ability to support the entire population.

Daniel Levine: My sense is there are many points of interaction with the Igentify system. Can you explain how it works and what it’s capable of doing?

Doron Behar: So, the interaction with the Igentify system starts with the design of the solution for the particular healthcare provider. So, this is a platform, Igentify is a platform, and the platform allows you to do the pretest counseling for example. And there are differences in the ways that different healthcare provider would like to ask questions, would like to present it to the patients. So, the first interaction with the system is when a new healthcare provider is being onboarded to the system, when that is happening, then the team has to customize the system to make it exactly the way they would like their patients to receive it, whether by the questions, whether by the look and feel of the multimedia, and all the way to what exactly they want to answer under different clinical outcomes. And once this is done, then we can actually allow patients to interact now with the system. So, patients will now interact with the system. So, let’s say that the patient is going to the OBGYN before pregnancy or during pregnancy and a preconception screening test is being suggested. The patients are referred to the digital process by different ways, by allowing a QR code to be read or by sending an SMS method that we started. That is where the patients are interacting with the system and this is where they’re getting all of the information from the machine. Then if we take it further down the road, then now when the patient has consented and a sample was given, either blood sample or buccal swab and so forth, then now the lab of the healthcare provider can interact with Igentify directly so that Igentify can read the raw results of the platform used by the particular healthcare provider and analyze it and interpret it. So that is where the lab is actually in touch with the Igentify solution. Then when the lab is done, the lab is creating a report using the solution, using the Igentify platform, and then the lab is returning back the results to the clinicians, to the genetic counselors that are in their healthcare system. So now the physicians and the genetic counselors are actually in touch with the system. They have a way to revise whatever medical information was supplied by the patient and the genetic data itself. Or they can set rules that are going to say if this and this and this has been met, move directly to assist the patient. Or if this and this and this were not met, call the patients for an interview. So that is when the clinician, the clinical side is meeting the system. And then once the clinical side is done, the system will send once again a message to the patient saying, your results are ready, let’s explain them to you. And then the patient is again in an interaction with the system, the patients will receive both a formal PDF but also a multimedia content that will explain in a very, very vivid and self-explanatory way the situation and then the patients will be able to go back to the results, look at it again and again, if they would like to share it with people, and so forth. So actually, the system is allowing a patient to healthcare provider interaction, and within the healthcare provider it allows both for the lab and the clinician to be involved with the system. So really an end-to-end solution that is trying to assist the medical team to educate and to make as many patients use genetic tests for their patient care as we can allow in our days.

Daniel Levine: What problems does this solve? What benefits does it provide?

Doron Behar: So, everything starts with education. I would say that if we want to get to better patient care, then we are certainly in a place where we need to educate the patients. Our patients will then explain to their family and friends about the benefits of genetic counseling. We need to understand what exactly genetic testing can allow for us. What are the limitations, what are the benefits of it? So, I would say that the system gives a very strong opportunity to educate the patient on the way to better patient care by allowing patients to be well aware and to have consented decisions in critical crossroads of their life. So that’s one: better patient care. Naturally the solution is also allowing scale up for the facilities. The facilities can now supply to many more patients information because we are relying on digital solutions, which means that they will be able to expand their realm of reach to patients that were not assisted thus far. But even more important than this, or equally important, is that if we want to create such a change in the way we look to genetic tests and generally to medicine use of digital technology, then naturally we should make sure that it is affordable, accessible, and certainly the use of digital tools is allowing better economic outcomes and we are removing some constraints that might be blocking us from reaching to each individual worldwide. So, I would say that the system has to create beneficial economic results as a condition for mass adoption. And again, all of it eventually has to lead to better patient care on our way to precision medicine and the ability to tailor specific treatments, specific diagnostic processes, and even specific drug utilization to different patients.

Daniel Levine: This is an AI platform. What role does artificial intelligence play in the system? Is it providing interpretation of genetic results? Is it answering questions posed by patients or is it doing something else?

Doron Behar: First it’s a great question. It’s a great question because certainly there’s a lot of discussion now about AI and there are different types of AI solutions. I would say that our system is allowing us to be labeled as an AI system that is allowing for mass production of data, mass analytics of data, and mass generating of results to patients in a way that is simply impossible by humans. And this particular form of AI is usually referred to as reactive machines because the machine will not decide about anything that the supervisor did not authorize. Naturally in medicine we are trying to be very conservative and very, very careful with the way we do our testing and our explanation. So, we don’t want, for example, a machine to decide that this new treatment is now good for this patient without first being approved by the medical team. So, this is why the system certainly uses AI technology, but in a way that is primarily focusing on the ability to create information that is simply impossible for humans to gather. If we’re speaking now about analyzing risk across our across 800 genes screened for prescreening for example, for millions of people, naturally that cannot be done by a human. But it is very important to say that the system is not allowing for any recommendations which were not tailored by the healthcare provider to any clinical scenario to be explained to patients. So that is the fine line that we have to be careful when we are in medicine, certainly allow AI protocols to help us, to assist us, but I think in a careful way that is allowing a genetic counselor, so to say, to duplicate themselves and say and support many more patients, rather than when sometimes people are referring to AI, which is an independent way of a computer to make new decisions which have not been approved yet. So that’s where we are with respect to this. It is very important to note we are here to augment the genetic counselors, not to replace them, to augment them and allow them to supervise a hundred times more patient to simply answer the demand.

Daniel Levine: What types of genetic testing does the system support today and are there plans to expand those at all?

Doron Behar: Yeah, so the system is ever expanding. It is currently supporting within the field of reproductive health preconception screening NIPS, which is a blood test being done during early pregnancy to detect chromosomal abnormalities like 21, like D trisomy to 21 trisomy or 18 trisomy and so forth. It is also assisting now within reproductive health for what we refer to as chromosomal microanalysis, which is a test being done following amniocentesis under certain clinical interactions. All of these are already working with patients with great success. The system is also now supporting pharmacogenomics applications, which is the science of suggesting or tailoring a particular drug for a particular patient, which means that not all patients with hypertension, for example, will respond to the same drug group and genomics can now help us to choose ahead of time which drugs will be better at shortening the time for successful treatment, reducing hospitalization days, and as important, avoiding adverse reactions that are never wanted. So, macro genomics is a very, very important portion of what we are doing and the company is currently releasing its applications about oncogene genetics, which is our ability to detect patients that are born with an inherited risk for cancer, not with the cancer, with an inherited risk for cancer like the famous BRCA 1 and BRCA 2 genes, which are putting carriers at higher risk for breast and ovarian cancer throughout their lifetime. And that is yet another major application that we should use for better patient care because it is always better to prevent cancer than to treat it. So, if we can assist to identify the patients at risk, the most important thing, and the term being used by the medical providers, is that these are actionable genes, which means that we would be able to look to the gene, understand that there is risk, and then suggest modalities that can prevent [disease].

So we are not just doing it to do nothing. We can, for example in the case of breast and ovarian cancer, recommend more frequent MRI testing just as an example. So that is a big portion. And then the company will continue to different fields of genomics like cardio genomics, like early onset neurological disorder and so forth. And even more than this, the company has plans to grow outside of genomics because the same principle used that we are creating in our technology can also help patients in non-genetics fields—better consenting for different procedures, better explanation when people are recruited, when patients are recruited to different studies. and so forth. So I would say that this year and probably next year we are still completely focusing on genomics, but then we might spread outside of this field.

Daniel Levine: How do patients respond to the platform and do either patients or providers express concerns about a lack of human interaction?

Doron Behar: We have been very careful on this one. We actually went into IRB studies, which means that under a supervised protocol, patients were using the system and it was compared of course to face-to-face interaction. And we received over 96 percent satisfaction rate from patients in two different IRB studies showing that the patients not only approved the solution, they actually many times preferred it over a face-to-face meeting because it answered many more needs that they had like accessibility of the information, shorter turnaround time, and better explanation. So that is actually something that was very important to us. We are hoping that this is going to be released in scientific papers shortly and then everybody can read the results. The level of the pure IRB studies have been approved as positive, but it’s much more important than this. We have now known that production data is showing that only one in 1.000 patients are seeking a secondary explanation. Of course, we always allow for a secondary explanation by face-to-face meeting or telemedicine, but only one in 1000 patients is asking for such an appointment to better understand the results. And I think that generally speaking, we are getting into a better and better understanding of how to use digital solutions to assist genetic counselors and the medical networks to supply information to their patients. And again, we have to make it very clear the hands of genetic counselors are full by only treating the patients that are at risk or at a high risk. So we are certainly trying to augment their abilities to support more and more patients, but in no way to replace the genetic counselors because they are needed to set the protocols and they’re needed to treat the patients who are actually at risk.

Daniel Levine: The company was founded in 2016. You and I spoke several years ago about what you were doing. I’m wondering, given the rapid pace of innovation with AI, whether this has had any impact in what you’re doing and whether you’re incorporating any new capabilities in applying AI to Igentify.

Doron Behar: So, the last two years or two and a half years since we spoke, I believe, have been amazing for us. First, everything that we spoke about just a couple of minutes ago was proven in this two and a half years. So now we have a proven experience, a growing list of customers. And yes to your question about AI. AI can be implemented in writing code faster, in helping behind the scenes data mining, and so forth. So absolutely the company is utilizing such procedures to make it faster, more accurate to more clinical indications within the field of genomics in order for us to be able to support more and more demand for more and more clinical applications. So AI is certainly being used for such purposes, but once again, for the actual recommendation of patients, the AI is reactive in its nature. It is following exactly what the genetic counselors want to say to the patient, but it allowing it to do it in masses that humans can simply not do.

Daniel Levine: And as you think about growing the service, what do you think it’ll take providers to make greater use of the system?

Doron Behar: So, I think that the experience and the experience, they’re always the first birds, right? They’re all the first ones worth trying the system. And it tends that once a critical mass is being achieved and once enough scientific and production data is being shown that the actual solution is working and the patients are satisfied, then the appetite grows. The appetite grows both in the number of providers that want to do it, but the appetite also grows in the clinical applications of providers already using the system, and how to further expand to other genomic and nongenomic applications. So, I think that that is key, both patient acceptance of the solution—the growing number of patients that are accepting the results that way and are happy with it and are coming back with positive reactions to their healthcare provider, and similarly, when the healthcare providers are happy with this solution and they see it’s accurate, it’s reliable, it’s affordable, it’s accessible. Then as time passes, we are getting more and more requests to do also this and also this. And things that initially we tried not to do are now being examined and checked because people understand the huge saving of time and money and in the same time giving better patient care. That is the perfect solution, right? To do good at a lower expense level. So as the data is accumulated, we are now getting more and more requests from new customers and more and more requests from existing customers for new clinical indications.

Daniel Levine: Doron Behar, founder and CEO of Igentify. Doron, thanks so much for your time today.

Doron Behar: Thank you very much, Daniel. It’s been a real pleasure.

This transcript has been edited for clarity and readability.

The RARECast podcast is made possible through support from the Global Genes’ Corporate Alliance. The members of the Corporate Alliance support Global Genes’ mission and programs, work to meet the vital needs of people with rare diseases, and address inequities they face. To learn more about the Corporate Alliance or how your organization can become a member, click here.