Roche CEO Severin Schwan | Cancer – Market and Mission (NZZ Standpoints 2016)

The pharmaceutical industry and the healthcare system are somehow connected. Next year, we are again seeing rising health insurance premiums, on average 4.5 percent, and for children the increase is even 6.6 percent. What goes through your mind each year when you hear these numbers?

I think you are absolutely right that the high value of health should not be an excuse for inefficient healthcare. I am convinced, also from my own experience, that there are inefficiencies in our healthcare system, not only in Switzerland but worldwide. We could talk for an hour about inefficiencies. Give us two examples anyway. Number one: we invest far too little in prevention and focus far too much on therapy. I always say our health ministers are not health ministers but disease ministers. Everyone knows it would be much cheaper to detect diseases early or avoid them altogether rather than treat them very expensively. That is the biggest lever. A concrete example: colorectal cancer. Through regular screening, which is easily possible today, you could completely avoid colorectal cancer. In developed societies like Switzerland, there would be no more colorectal cancer. Yet society has not yet managed to mandate such screening. The patients are to blame, the system is to blame, others are to blame. I think it requires cooperation from all involved, of course also the consent of patients. You cannot force patients to undergo screening. It also requires a certain social consensus and political pressure to establish new incentive systems, and the health insurance companies must also participate. Personally, I could imagine, sticking with the example, that people who regularly undergo such screening pay lower health insurance premiums or receive a bonus. I don't think you can force anyone, but you can set other incentive systems.

I personally believe that the greater the pressure in the system, the more these efficiency potentials will be used. On the one hand, we are all unhappy when health insurance premiums rise, but on the other hand, everyone wants to have a hospital around the corner. I live in a small suburb of Basel. We had our own municipal hospital there, five minutes by tram to the next hospital, and five minutes further to the university hospital in Basel. Yet people in the municipality even took to the streets when politics began to consider closing this very small, completely subcritical municipal hospital. So the same people who are always annoyed about high premiums and increases take to the streets, even though they only have to go five minutes further to the next hospital. I think over time, when the pressure is great enough, such efficiency potentials will be used more. There have already been hospital closures, and cantonal governments have stumbled over them, I must say.

May I first make an interjection to put the importance of medications into perspective? We talked about costs. Healthcare costs total eleven percent of value creation, as mentioned. Medications account for about ten percent of total healthcare costs. That means we spend about one percent of total value creation on medications. To put that in perspective, we spend more on telecommunications, and we spend more on tobacco alone than on cancer medications. But that should not be an excuse that we also need to be efficient in treatment with medications. Of course, the question of prices arises. You already mentioned the keyword: in the end, it is a negotiation process. It seems fundamentally impossible to determine the right price for a medication because ultimately you are setting a price for life. What does it cost if you can live a year longer in good health? What is that worth? You can never objectively determine that definitively. So it requires political consensus. I must say, what works very well in Switzerland is that all parties sit at the table: patient representatives, doctors, health insurance companies, manufacturers. What is our position? We always say, first, that the benefit should not be viewed too narrowly. There is the benefit for the patient, that is clear. What is often forgotten in Switzerland, and I must say in almost all countries of the world, is that the healthcare sector thinks in silos. That means individual areas are looked at, not total healthcare costs. With innovative medications, it is often the case that patients leave the hospital faster due to the medication, but if the hospital budget is a different budget than the medication budget, then no one brings it together. That is a very important point. What we also argue, and this makes negotiations with health insurance companies very difficult, is that if someone can be cured with a medication in the best case, they can work productively again for society, pay taxes, contribute to social security. These are, from a macroeconomic perspective, good investments. So our standpoint is always to move away from silo thinking, from small boxes, toward a holistic view.

That is precisely because you cannot objectively determine it, so you have to agree on a negotiation process. In Switzerland specifically, we also look at what the price is abroad. There is a benchmark. For political reasons, and ultimately in our interest, we do not want to move too far from prices abroad, either below or above. If we have lower prices in our home country, that will be held against us abroad, creating economic problems. Conversely, if we have higher prices in Switzerland, that also becomes a problem because political pressure on those responsible increases. So that is often the starting point, and then the discussion goes around what it means concretely for patients, for the healthcare system, and for society as a whole when this medication is brought to patients.

Of course, and in the long term, the business model is only sustainable if you recoup your costs. That is correct. The problem is you cannot calculate that down to a single product because we have a portfolio of products, most of which never come to market. So the medications that actually come to market must also finance the losses of the medications that naturally fail over the years. That is certainly the central question you already touched on: what is it worth to us as a society for health or therapies?

It is ultimately a balance that must be found. On the one hand, it is the question of what health is worth to us, as you rightly say, and societal expectations may differ in certain countries. That is indeed the case. The second is how we distribute the costs within society. It is ultimately about how much the individual has to pay and how much society covers. If you look at Great Britain, what you describe is all correct, except for those who are privately insured. The result in Great Britain is two-class medicine. There is the large majority of the population who cannot afford certain medications or therapies, or face long waiting lists, and a smaller part of the population who pay for it themselves or through private insurance and can afford the best care in private hospitals. That is socially accepted. That is not my world. I personally much prefer the system in continental Europe. This solidarity-based system suits me much more. But it comes down to two questions: not only what health is worth to us, but also how we distribute the costs within society.

Both aspects are true. On the one hand, the pressure to suffer may still be too low to realize efficiency potentials. On the other hand, surveys in Switzerland repeatedly show that the Swiss are very satisfied with this extremely good healthcare system. Many citizens have had the experience, either themselves or with family members, of being very well cared for in the event of a serious illness in Switzerland. I think that ultimately contributes to the high acceptance of healthcare spending.

It is absolutely fantastic what is happening. If you had asked me five years ago, I could not have imagined it. For the very first time, we are succeeding in reactivating the body's own immune system. What happens is that the body's own immune system, the killer cells, suddenly fight the cancer. The fantastic thing is that we now see cancer patients where the cancer has already spread, not just the primary tumor but metastases. In these patients, we see, not in all patients and all cancer types, complete remission, a complete regression of cancer and metastases. This is because when the body's own immune system takes over again, you have a continuous effect 24 hours a day, seven days a week, and it reaches everywhere in the body. It is fantastic. New doors are opening, and these new medications will increasingly be combined with existing medications to further increase efficacy.

You are right. Scientists have always wondered why the immune system does not attack these foreign, mutated cells from the start. As healthy individuals, we constantly have mutations, and the immune system constantly cleans them up. The question was: how do cancer cells manage to escape the immune system? A really exciting discovery was made. Imagine these immune cells, the killer cells, in the body. They want to function, but the body does not want them to kill everything all the time. They must function specifically when a foreign body is present, like a virus, bacterium, or cancer cell. Otherwise, they should not function. What was discovered is a switch on these killer cells. There is an on-off switch, a complex system of how these switches are activated when they should be. What cancer cells have developed is an arm that flips the switch. The killer T cell comes to the cancer cell, and the cancer cell has a protein, think of it as a switch, and it flips the switch on the cancer cell, making it invisible to the immune system. Once that was recognized, the approach for therapy was found. You can block the switch on the cancer cell by docking another molecule exactly at that switch, preventing the switch from attaching to the immune cells. These therapies are now effectively on the market, and we see incredible, outstanding successes where the immune cells are reactivated and can work normally again.

Yes, it is indeed the case, especially with cancer, that different patient groups respond very differently to therapy. Some respond, others do not, and others may even have side effects, even though the medication is the same. So logically, the differences must be in the patient. Today, we have much better diagnostic tools and technologies to recognize these differences, for example in genetics. Tumors are constantly changing and have different mutations. Today, we can measure these mutations. We increasingly see, though we still have many unanswered questions, that certain patients with a certain genetic mutation respond very well to a medication, while others without that mutation do not respond. Or, in the previous example with the switch, if the cancer does not have that switch at all, it is unlikely to respond to the therapy. These things can be measured today. What is happening is that diagnosis is becoming more precise and differentiated, and medications are only used in patient groups where they actually work. This also means that the collaboration between pharma and diagnostics across the entire value chain, from early research to clinical practice, is becoming much closer. That is something we as a company strongly focus on, having both diagnostics and pharma under one roof.

Let's stay with the example of the switch. The first hypothesis was that if we look at all cancer cells that have this switch, the therapy should work. That was the theory. It was brought into clinical trials, and in the meantime, these medications are approved. What do we see? It works for some but not for many. Naturally, we ask why. Then the next discovery was made: in certain cancer types, the T cells, the killer cells, do not even reach the cancer cell. So there is not only the mechanism with the switch; the cancer has developed other defense mechanisms. For example, it forms a certain tissue structure around the cancer cells that the T cells cannot penetrate. When you look at the cancer tissue, the cancer cell is in the middle of this structure, and all the killer cells are trying to get in but cannot. The next step was to further differentiate the diagnosis. We look not only at whether this switch is present but also whether these T cells actually reach the cancer. This can be done through imaging diagnostics. So differentiation advances another level, and it will continue. When we meet next time, there will be three or four additional levels of differentiation. Eventually, we will be able to cover more and more patient groups with specific medications. The ideal goal is to help 100 percent of cancer patients with these medications.

You can imagine it that way, but as the current state of science stands, it is not about individual patients. Perhaps the term 'individualized' or 'personalized' is not quite accurate. It is essentially always about patient groups. It might be about ten percent of patients, then again five percent of patients. It is not about finding a special therapy for each individual patient. But one can imagine that in the ultimate consequence, though we are still far from that.

That is correct, and I think it is conceptually right. Everyone agrees on that. When we sit down with health insurance companies and say we should find pricing models where payment is only made if the medication works, everyone says wonderful, it is in our mutual interest. So why does it fail in practice? It fails in the data systems. What you need is the information to be able to bill that way. Today, systems record how many medications leave the warehouse, and then you can bill based on that. But if you think of such a model, you need to track exactly for which cancer this medication is used. There are medications used for different cancer types. Then you need to define what a good outcome is, which can be debated for a long time. Once you agree, the doctor, every time he sees the patient, must electronically record in some way how the individual patient responded to the medication. You can imagine that is a huge effort and not easy to implement. But the good news is we are making progress. For example, in northern Italy, we have introduced a system with health insurance companies and the relevant provinces where we at least have the information for which cancer type this medication is used. We have a cancer medication used in different cancer types, and it responds differently. We know it works very well for one cancer type and less well for others. Previously, there was one price. Now, at least it is recorded for which cancer type the medication is used. It is not yet at the patient level to know how the patient responded, but it is a step in the right direction. Now we have billing systems where we bill differently depending on which cancer type the medication is used for. These are steps in this direction, and it will continue.

I believe that the whole topic of data processing, big data in healthcare, will have incredible significance and open up incredible possibilities. Let me say again: we are at the very, very beginning, and as an industry, we are far behind other industries. When you go to the doctor today, you probably know the experience: he usually pulls out a patient file, a physical, analog thing with your name on it, full of papers that only he can read. He makes notes, takes diagnostic data, maybe the result of a blood test, then you come back, he notes how you responded to the medication, maybe changes therapy, and at the end, the patient file goes back into the archive and is never seen again. You can imagine if these data are digitized, and it will happen, it is only a matter of time, then you can structure them, aggregate them, and analyze them. There is no question that when we analyze these data, we will see new correlations. We will see that certain patient groups, for example with a certain genetic mutation, do not respond to a medication at all. When we see that, we have the basis for new hypotheses, we can investigate what is going on, and develop a new medication based on that. You can imagine the impact on clinical practice when these data are known and statistically relevant. Treatments can be much more targeted. This will be an incredible revolution for healthcare, not only in terms of efficiency but especially in terms of quality of healthcare.

That is a very important aspect. It is possible to anonymize data. For research and development, we work with various partners on how to analyze data because we are not interested in the name of the individual patient; we are only interested in the insight from the sum of the data. That is technically possible. It needs to be done, it requires appropriate regulations, and it will come. I am convinced of that.

I do not believe that digitalization will replace our current business. I believe both need to come together. Data alone are not enough; you need great scientific understanding, medical understanding, to be able to do anything with the data. And do not forget, especially with medications, they will never be a digital product. You have to influence biological signaling pathways in the body. You need a molecule, something physical. I do not believe that in medicine, the operation, ultimately medicine is something quite physical. Robots may operate, but even if a robot operates, it has to operate on a human being who is not in a digital space but in a physical framework. So medicine will ultimately remain physical. But whether you are competitive, from an economic perspective, depends on whether you succeed in using the possibilities of digitalization and combining them. If you do not, the positive possibilities are there, but the negative is that your own research productivity over time will no longer be competitive.

That developed over time. Originally, I started at Roche because the finance department had a good reputation, and I specialized in finance in my studies. So for me, it was a natural entry. I could have just as easily ended up in another industry. But what is true is that when you are in a company for so long, you begin to identify with it. Over time, you see more and more what it means for patients when we develop these diagnostic procedures and medications. At the end of the day, you start working because you are curious, have specific knowledge, and also to earn a living and support a family. But at some point, it becomes more important why you do it and what comes out at the end of the day. It is an incredible privilege to work in the healthcare industry and personally experience the contribution you can make to patients and society. That is an important part of this situation, not only for me but for the company as a whole. It is a great advantage to know the company and be with it for so long.

That is true. You have to constantly challenge yourself and be careful not to believe everything you have told yourself over the years. You have to stay very open and develop a certain sensorium. Personally, I maintain a lot of external contacts. I really enjoy meeting people in completely different industries, partly because it is fun and fascinates me to hear and learn about them, but I think I always take away inspiration that I can implement in my own company. So the danger of becoming blinkered and only seeing things from your own perspective exists. The first step is to be aware of it, then stay open, go outside, and check if you are really on target.

One approach that has worked well at Roche is to start small pilots. You don't know yourself if the new path will work, if it is the right product or approach. What works well in our culture is that a department or function tries something new. If it works, word spreads quickly, and others do it too. Initially, many may be skeptical, but as soon as success is achieved on a small scale, everyone wants that success. That concept has worked quite well for us. Instead of prescribing from the top down, saying this is the new line, this is the future, and now we go this way, we do it the other way around. We say, what could the future look like? There are three or four scenarios. Let's start pilots and see what comes out. If it proves itself, then people's willingness to go in that direction is much greater.

It is true that when you are in a business like ours, which is about innovation and new things, you have to open up a bit. You have to have the courage to fail. You have to take the risk that things might go wrong and not work out. You cannot focus on innovation if it has to work every time.

When you talk to researchers, in my experience, when I talk to scientists in my company, to my research directors, to project leaders, eventually we get to budget time. My experience is that each of them could and wants to spend at least three times as much as we can afford, precisely for the reasons you described. At some point, the whole thing has to be sustainable: only spend as much as you earn. But I guarantee you, the scientist will demand about three times as much. That initially reassures me enormously because the worst thing would be if your researchers had no more ideas for spending money. How do we handle it? We set globally how much we want to spend on research, how much we can afford. We spend quite a lot, by the way, more than most companies in the world. We are the healthcare company that spends the most on research and development in absolute terms, namely 10 billion Swiss francs. That is about 20 percent of our sales, so relatively we are also at the upper end of our industry. That has a lot to do with our strong focus on innovation. So we spend a lot, but within a limit. We could spend 30 billion immediately, but then it would not be sustainable. Once this limit is set, we can give the individual research departments a lot of freedom. Then we can leave it to the researchers to make trade-offs: which things do we invest in, where do we believe the chances are better, where is the medical breakthrough potentially greater, where are the possible big successes? That is the balance. You have to set the framework so that everything does not explode, but in detail, when it comes to weighing how to achieve the goals, you have to give maximum freedom.

Of course, we watch each other. But there is no big difference between the two companies in Basel and other companies we compete with. It is always this balance: on the one hand, we have common interests, industrial policy interests. When it comes to Switzerland, the overlaps with the second company in Basel and other companies in Switzerland are naturally greater. Then there are areas where we are in direct competition. Accordingly, we talk less or more with each other depending on the interest situation. Overall, it has worked well.

On a scale of 1 to 10, I would say 9, simply because I never believe that anything can be perfect. But I believe that Switzerland, especially northwestern Switzerland, really plays in the top league worldwide. I am convinced of that. That is why we invest so much in Switzerland. Otherwise, we would not invest so much, especially in research and development. The decisive point is clear: when it comes to innovation, you need the most creative, the best people. We find them in Switzerland. We also get them from abroad; people like to come to Switzerland. We have an environment where these people can work well. It has a lot to do with the good education in Switzerland. The dual system is fantastic because you need not only Nobel Prize winners but also laboratory technicians for the whole system to function. And we have an excellent university network. That is the decisive reason why we invest so much in Switzerland: we get good people there.

We have 90,000 employees worldwide, over 13,000 in Switzerland. Of those, 60 percent are foreigners, if I remember correctly. Half of those are foreigners who settle here, and in research and development, half are cross-border commuters. We are really dependent on that; it is existential. As the head of the company, I would have wished for a different outcome with the mass immigration initiative, of course. But on the other hand, I said from the beginning, as an Austrian living in Switzerland, my experience is that in all important matters, the Swiss people and Switzerland have always found an incredibly pragmatic solution. We will see how the decision-making process develops and what comes out in the end. But my fundamental trust in Switzerland and pragmatic solutions is unbroken. I am convinced that a solution will be found.

Well, trust does not mean not working on it and contributing. For example, in background discussions, we recently had the topic of third-country quotas. It was debated whether these quotas should remain as low as they were reduced two or three years ago or be increased again. We also advocated for these third-country quotas to go up. In Basel, the quotas were exhausted in February, which creates uncertainty, and that is poison for the research location. We have direct discussions with the canton of Basel-Stadt, who are close to it and know immediately, and then the discussions are escalated to the federal level. The canton advocated for it, and we advocated for it. I am very glad that the quotas have been increased, though not quite to the old level. We as an industry feel we were able to contribute. So we have to work on it and take care. But you see again, when it really comes down to it, a solution is found.

I am optimistic about that.