A New Zealand Teen Could Help Rewrite Peanut Allergy Treatment

A teenage boy from New Zealand could make history, and he probably does not even know how much hope is riding on him. "He was the first to receive our injection, about three weeks ago," explains Prof. Yanai Ofran. "More people are gradually joining in Australia and New Zealand."

The injection is intended to eliminate, or at least weaken, peanut allergy. It is a treatment given two or three times over several weeks, and it introduces a molecule into the body that is meant to teach it not to recognize the proteins in peanuts as a threat. In practice, it is not yet really a treatment, but an experiment in treatment, after eight years of development in the laboratories of Ukko, one of Ofran's three companies. "We worked on this for eight years, tested our molecule in computer systems that simulate human cells, in test tubes, on human tissues, tested it on animals to make sure there was no damage or allergic reaction, and only then did we get approval for human trials. This is the real time, the money time."

The trial is being conducted in Australia and New Zealand. "Yes, because regulation there is relatively lenient. On both sides of the ocean, they want you to prove that the chance of harm is small and the chance of benefit is large, but each side differs on the question of how to balance them. The Australians place more emphasis on benefit, the Americans on harm. So the process in the East is more decentralized, faster and cheaper. These are countries that want to bring innovative treatments to their citizens. It is also their business strategy, a way to attract companies from around the world to do their trials there. We, for example, spend $100 million there and get $40 million back. So they have an interest, and we have an incentive. It is better than the United States, and certainly better than Israel, which does not have such a regulatory and economic system. No wonder Australia and China are the leaders today in clinical trials."

You are still only in a phase 1 clinical trial. When will you know whether the drug works? "Right now the participants, who came through hospitals, are receiving the first doses. Soon they will begin, gradually, to eat peanuts, under medical supervision of course. We will collect data, and I believe that after a year we will have some statistics on effectiveness and side effects."

Are you stressed? Not sleeping at night? "It is a long journey. Whoever cannot enjoy it should not come. Whoever has sleep disorders should move to another field."

Are you already thinking about what happens if it succeeds? "If it succeeds, and we have a drug that can enter the market, it could bring in many billions of dollars every year, because 1% to 3% of the population suffer from peanut allergy. When I wrote on LinkedIn that the trial had started, I got dozens of inquiries, only from Israel, from people asking to participate. Unfortunately, there is no trial in Israel right now, but people want a solution."

Will this be a revolution? "If we can prevent a life-threatening condition with two or three injections, that will be a revolution. Today, existing treatments for allergies are given in daily doses, on the assumption that the only way to remove the danger is years of daily treatment, perhaps for life, and most people cannot manage that. If our molecule works, even if it does not eliminate the allergy completely but does remove the fear of death, the entire concept of allergy treatment will change."

"In any case, there are already things here that have never been done before. First, this drug, Ukko18, is the world's first RNA vaccine in the field of food allergies to be in human trials. It is intended to reprogram the immune system so that it does not respond to peanuts with an allergic reaction, but not through daily exposure to peanuts, as mentioned, but through just two or three treatments. Second, the drug was designed with AI, using our special model. So every step of progress along the way is itself a revolution. In a few more months we will know whether the drug works."

And if it does not succeed? "Then we have nothing. When investing in AI or cyber, some companies do very well, some do a little well, some do not return the investment. In our world it is binary, billions, or zero. But in the two molecules already in trials, we have not failed yet. When you fail, you know quickly, and here we have already reached the stage that only one in 1,000 can reach. Now we are waiting for the results."

"If we can prevent a life-threatening condition with two or three injections, that will be a revolution. Today the assumption is that the way to remove the danger is years of daily treatment. Our treatment can change the entire concept of allergy treatment."

The problems are redefined, the solutions are redefined

Ofran's bottom line may be binary, but his thinking is the exact opposite. It grew in the home where he was raised, and later developed through an unusual career path. Ofran, 53, is the son of Avraham, a bank manager, and Mira, a PhD in physics. On his father's side, he is the grandson of Moroccan-born Rabbi Mordechai Zafrani, and on his mother's side, the grandson of Dr. Greta Leibowitz, a mathematician, and Prof. Yeshayahu Leibowitz, a scientist and one of the sharpest and most influential thinkers to emerge in Hebrew culture. His uncles worked in medicine, science and law. His siblings also went into medicine, psychology, religion, education and activism. In short, Ofran grew up in a home that combined a variety of fields of knowledge, where ideas were as accessible as breakfast.

He himself earned a bachelor's degree in physics and biology with a minor in philosophy, then went straight on to a doctorate, moving from linguistics to neuroscience until he arrived at biophysics and a specialization in computational systems biology, and began combining academic work with business entrepreneurship. Above all, all the time, he thinks a little like a philosopher.

For example, when we talk about the other leading developments of his companies. Imneskibart is already in phase 2 clinical trials, and offers an immunotherapy treatment for cancer, also developed with the help of artificial intelligence. In fact, this is the world's first drug of its kind based on a molecule designed by AI and brought into human trials. The peanut allergy drug is the world's first AI-based molecule in the food allergy field. And there are three more drugs whose development is similar and are expected to begin human trials in the coming year, one that strengthens and refines chemotherapy treatments, one intended for asthma and atopic dermatitis, and one aimed at autoimmune diseases such as multiple sclerosis and colitis.

Between a drug that will prevent peanut allergy and an immunotherapy treatment for cancer, which is more promising? "Who do you love more, dad or mom? The older child or the younger one? They are two projects I love very much. We have already seen the cancer drug work on people who had no hope, but I always remind the team that the road is long and there are endless ways to fail."

And that too is an AI product. Theoretically, are we moving toward an era in which artificial intelligence will allow us to treat all diseases? "That is almost true. But there is an interesting process with diseases: 150 years ago you could fall onto a rose bush, get an infection and die, but then antibiotics came. People stopped dying from infections, but there was an increase in heart disease, and people died from it at a young age. When treatment for heart disease improved and awareness of diet and fitness rose, people reached older ages, and then more people started dying of cancer. Now cancer drugs are improving, and after them there will be even uglier diseases, the degenerative diseases of the nervous system and the brain. In 10 to 15 years there will be more drugs for those diseases, and then others will appear. The machine of the human body can break down in so many ways that it guarantees we will always have something to work on. Anyone who wants job security should develop drugs."

What does it take to develop drugs? "We are in a period where the problems are being redefined, and the solutions are being redefined. To develop a drug, you need three things, three fronts that must converge and integrate. The first is good intelligence, knowing which protein, cells or gene needs to be attacked. You need a high-quality target bank, because without it, it is like shelling Iran without a plan. That is the biological research. In the second stage, artificial intelligence creates precise and focused armament. In the third stage, you have to prove statistically that the drug works. For that, you need to conduct trials on animals and humans, and on enough humans to reach a significant result. Most drugs fail at this stage, and in any case the whole process usually takes more than a decade and costs billions of dollars."

Isn't artificial intelligence supposed to shorten processes? "In the end, it is a matter of statistics. AI says you can learn principles from the existing data that will apply going forward. But there are 8 billion people in the world, and that is your data, and it may not help you treat the 8,000,000,001st person. Even if we manage to reach principles that help most people, there will always be something we did not take into account."

But this is a technology that is racing ahead, surprising even its developers. Are we close to the moment when it will let us know the chemical formula of pain, or of love? "Maybe we will be able to describe love in chemical terms, but we will not understand the connection between that and the emotion we feel, the mental experience. My grandfather always illustrated this with the story that when he goes to the dentist and says his tooth hurts, the dentist can see an infection in the root but cannot understand how much it hurts. Physics cannot describe all subjective feelings, even with powerful artificial intelligence."

"When investing in AI or cyber, some companies do very well, some do a little well, some do not return the investment. In our world it is binary, billions, or zero. But so far we have succeeded."

How a home brings curiosity under your skin

You grew up in a home that was perhaps as far as possible from artificial intelligence, in a family that seemed to try to conquer every field of knowledge at once, with a high dose of science. "I would not say I grew up in a science-oriented home, but in a home that greatly valued curiosity. And that gets under your skin."

What does that mean? How do you educate for curiosity? "You emphasize questions more than answers. When you sit at a meal and someone asks a question, you do not brush it off. You open a book, check, argue, and then open a dictionary or an encyclopedia. That was our life, before Google. Everything was relaxed, there were jokes, but when there was a question we tried to answer it. Good questions excited everyone. And if there was no question, well, it was strange when there was no question."

Give an example. "At age 13 or 14, for example, I noticed words referring to time through front and back, meaning before and after, and I asked about this connection between front and back and before and after. We began to check it, and discovered that in ancient Hebrew what is associated with the front is the past and with the back is the future, while in modern Hebrew it is the opposite. After all, ancient times are in the past, but to advance is to go forward, to face the future. It is as if in the modern world people see themselves as having the past behind them and their face toward the future, while in the ancient world people saw themselves as looking toward the past."

Were questions like these what led you to linguistics, as the first track for your doctorate? "Yes. I started research in pragmatics, a branch that studies the understanding of words within context. I examined the meaning of words not only in terms of syntax and semantics, but also from the pragmatic reality. A sentence like 'Yanai, your tie!' takes on different meanings in different contexts, if I am leaving for a meeting, or if I am sitting in front of a bowl of soup."

Why did you leave it? "Because I felt the questions in linguistics were amazing, but the answers were disappointing. I wanted research that advances answers."

But you did not move too far from language. "No, I moved to a doctorate in neuroscience that focused on handwriting. The interesting question in the brain field is where consciousness is, and I thought handwriting research was an interesting way to deal with the topic."

And you left that too. "Yes. I met Prof. Claude Ghez, one of the great researchers in neuroscience, a smart and funny man. He told me that in trying to understand how the nervous system learns, plans and executes complex movements, he studies a certain cat movement, and admitted, 'I have been studying this for 40 years, I have had 70 students, and I still do not know how cats do it.' I realized I needed to look for research in another field, one that would give me more concrete answers, and I arrived at biophysics."

Why biophysics, after all that? "Because there you can understand the connection between molecules and the phenomenon called life. How they create disease, how the body works as a distributed and sophisticated computing system in which every molecule knows what to do at the right time, in order to understand how it works and why it breaks down. Those questions led me to try to understand how to design a molecule that would dissolve cancer tumors, for example. And those are questions that even if you cannot find the answer, the research itself is a significant step along the way."

AI reduces the haystack and increases the number of needles

After a doctorate at Columbia University and teaching there and at Tel Aviv University, Ofran arrived at the laboratory at Bar-Ilan University, which he heads to this day. At the same time, he published articles in various forums and was also one of the creators and hosts of the documentary series "So Said Herzl." And for the past 20 years, he has also been founding companies. Before Ukko and its fight against peanut allergy, Biolojic Design and its subsidiary Aulos Bioscience were established, focusing on cancer treatments and responsible for a drug already in human trials. The two most recent companies are working with cooperation agreements with the pharmaceutical giants Merck, Teva and Nektar, on cancer, asthma and multiple sclerosis drugs, respectively.

Was academia not enough for you? "No. I could have made a career just publishing papers on antibody design, and on antibody design with artificial intelligence, but I wanted the antibodies to also work. And they do work, they are already dissolving cancers in humans. In academia I received government grants, but when I asked for budgets to design molecules using AI, the answer was usually, 'Listen, it's a cool idea but expensive. You will not reach the amount you need in government grants.' I understood that I needed to raise money from private investors, and to this day in all my companies I have raised more than $200 million for research. You could not raise such an amount in academia. You cannot do something at a large scale from academia, something that reaches humans, because when you have an idea you need money to test it. As a professor I feel like a peddler of ideas: 'Here is a nice idea, here is another idea.' It is a very fun job. But in the end you need to see whether the ideas work out there, in the world, and now I am still a peddler of ideas, but a little more advanced and proven."

When did you start thinking about using AI? "15 years ago, even more. That is what interested me, and there was also luck involved. When you examine new technologies, ideas and approaches, if you have a good eye it helps you be in the right place at the right time. It is like betting that Paris Saint-Germain will win the Champions League. There are hundreds of teams, but if you look at the players, the management and the fans, you already understand at the start of the season that PSG is not a bad bet. In science too, when you see innovative approaches and their quality, you understand that this is not going to be decided forever on penalties."

What is the rationale for AI-based drugs? "Breakthrough drugs are not discovered, they have to be designed and engineered. It is like the move from Google to ChatGPT, instead of taking a bucket and throwing a billion molecules into it, then hoping to hit on something that works, let's design the bucket better, so the chance of finding a drug is much greater. In short, reduce the haystack and increase the number of needles so we can find drugs faster."

That may simplify the process, but drug development is still a difficult and complicated field. "Yes. In high-tech, two kids who were discharged from army computer unit 8200 yesterday and have an idea for protecting data with cyber can try to sell it to Google and Microsoft, and if they do not want it, the developers can put the idea in the app store. If the public likes it, it can become Wiz and Google can buy it for $32 billion. In biology it is not like that, where someone says, 'I have an idea for curing cancer,' Pfizer and Merck tell him no, and he goes and puts it in a pharmacy."

What does that mean? "That a lot of excellent drugs have been thrown away. Because the investment was too expensive and stopped, because they could not expand the trials to enough patients, because the drugs were given to patients in poor condition after all existing lines of treatment. This is a central paradox in the drug world, that sometimes approval for a trial is given only to patients for whom all hope is almost gone, and then it is not clear how much the treatment can help, it may also be dangerous to give the experimental drug, and the trial cannot always be completed. If the patient dies, of course the trial stops, and you cannot know whether it was because of the disease or because of the drug."

What do you do about that? "You design a trial that gives the drug the best chance to prove itself, so it will be worth continuing to invest. You need the right design, you need to identify the patients for whom hope has not yet been exhausted, and you need luck."

You have a drug in hand that you see works, and you cannot give it to everyone who needs it. "In the drug world you cannot tell companies, 'Sell it, let the patient decide whether to buy and take the risk.' Drugs have to be approved, and there has to be great expertise to approve them, and regulators have to decide about the risks. There are situations where they make a decision that is not necessarily the right one, and drugs are lost. There are companies that only after several failures managed to prove that the drug does work. So now we have a drug that shows efficacy in disappearing tumors in advanced melanoma, kidney cancer, lung cancer, every cancer we tested showed good responses, but there is not yet enough information to approve it and I have no ability to give this drug outside the trial, only through the hospitals working with us."

That must not be easy. "True, but it is also inevitable. If we do not insist on the process, the regulation, the long-term monitoring of efficacy and toxicity, the statistics, we will not have drugs at all. So yes, it is very frustrating, but no one has invented another way yet."

It is not a story about a scientist whose wife died of cancer

The dilemmas and pain involved in the disease are not foreign to Ofran. His wife, Dr. Ranna Samuels, was a climate researcher, and died in 2014. "It looks like the story of a scientist whose wife died of cancer and who is trying to cure the disease, but it is not," he rejects any attempt to connect the personal biography to the professional one. "I founded the company before my wife died. But yes, I may have some idea of what families in this situation go through."

He lives in Tel Aviv with their four children, ages 22, 21, 18 and 15. "These children have a mother, she is just not alive. I always tell the HR teams in my companies not to use the phrase 'the Biolojic Design family' and 'the Ukko family.' Family is not work, and work is not family. It is not even in the same league. The only place where a person is irreplaceable is in their family. Only for my children am I irreplaceable, in a company you are never irreplaceable."

From the family he comes from, when asked about the legacy of his grandparents Greta and Yeshayahu Leibowitz, he says that "they had a sharp view of reality, a view that was not captive to anything. That becomes sharper now, when all of our view is completely dependent on which team we are on. They had an affinity for the tribe, but the truth is the truth regardless of that affinity, and the power of that is enormous. When you ask a person a question and know that the answer will not be biased and will not depend on the tribe's blessing, that is amazing. That is the essence of the difference between someone who seeks truth and someone who seeks to defend a camp, position and ideology. If what is right in your eyes is a matter of position, then you are not committed to the truth. And you cannot do good science, research or even philosophy if you step onto the field wearing the jersey of a certain team, because in that case the truth no longer interests you. If tomorrow your interest is reversed, then truth becomes something else?"

Can one understand the power of belonging? "Yes, it is hard to overcome our biases, with the tendency to identify with one side and not see its mistakes, or to oppose another side and not see the places where it is right, or to see what you want, imagine, hope for, and not the reality hiding behind it. But my grandfather could say the most unexpected thing without thinking whether it was an own goal for his team."

In this context, one cannot help thinking of a controversial expression associated with Leibowitz, which he used when he wanted to warn of the dangers the occupation of the territories posed to Israel as a society. Ofran is presumably used to hearing the term "Judeo-Nazis" brought up in conversations about his grandfather. "It was a rhetorical debate," he says. "Everyone understood what my grandfather actually meant. Whether it was nice or not nice to speak that way was not part of the truth debate for him. You can say the truth in a gentle way or in a blunt way, and you can also tell lies bluntly. In the end it is a debate about style, and my grandfather had a style... thank God."