Biocartis is creating the Nespresso of personalised medicine
Catching Rudi Pauwels, Biocartis’s founder and CEO, for an interview is not an easy task. Between advanced discussions with big pharmaceutical companies, negotiations with biomarker partners, and the launch of a first production facility in his native Belgium, his diary is fully booked. Still, on a quiet summer’s day, he has received technology by Bilan in his headquarters at the science park of the Swiss Institute of Technology of Lausanne (EPFL). There, the co-winner of BioAlps 2010 award (alongside with EPFL’s professor Philippe Renaud) takes time to explain what one of the first fully fulfilled promises of personalised medicine looks like. How did you come to personalised medicine? It is a very personal experience. To explain it, I need to go back to the early 90s, even before the term existed. Then, I was approached by a friend who had another friend, a doctor, who was dying after being infected with HIV in Africa. He had no therapy option left and only a few months to live. At that time I was managing Tibotec, a company I founded to develop new anti-HIV drugs. We had also developed a technique to measure the resistance of the virus to existing drugs. My friend asked me to try it on this desperate patient. To make a long story short, we tested his blood and saw that his virus was completely resistant to the few existing drugs he was taking. Fortunately, we were testing new drugs in clinical trials at that moment. So we rushed to the clinical centres to test the resistance of his virus against those drugs. The results were amazing. His critical conditions became a manageable chronic disease. To me it was also a revelation.What do you mean? We are transitioning from a period where treatments were based on symptoms to a medicine based on molecular diagnostics. The technique we had developed to test HIV virus resistance led me to create a second company, Virco, in 1995. Before, patients were prescribed a drug and if it was not efficient, they were prescribed another one. With Virco, we exited this trial-and-error process. We took a blood sample and measured the quantity of virus as well as its resistance to specific cocktails of drugs. We had developed an understanding of how the virus evolves and why its concentration in the patient increased again. So we were able to adapt the treatment to each patient’s conditions, thanks to molecular diagnosis of the virus itself. That mix of molecular diagnostics and therapeutics adapted to each patient is at the centre of personalised medicine. How your revelation did translate into what Biocartis is doing now?I realised this was the future of medicine, not only for HIV but for other infectious diseases, for oncology, etc. At the time, basic science started to advance at an unprecedented pace. The molecular mechanisms of diseases started to be understood. The number of biomarkers available to diagnose metabolic phenomena started to explode. For the first time, we were able not only to diagnose but to predict diseases and therefore to treat them – or even to prevent them – with all the benefits that implies for society as well as for healthcare economics. But at the same time, I realised that outside a handful of advanced research laboratories the necessary technologies were still lacking.And you found those technologies in Lausanne?Yes. I sold my companies to Johnson & Johnson in 2002. Soon after, I was attending the Nanotech conference organised each year by Alain Donzel in Montreux, to which EPFL’s professor Philippe Renaud contributes as scientific advisor. Late that night at the hotel bar, we discussed how nanotechnologies might solve the difficulties I had identified in making molecular diagnostics ubiquitous. Renaud offered me a sabbatical year in his lab to start a dialogue with engineers and understand how to adapt their knowledge to life sciences. I stayed three years, realising that to build a solution I needed contributions from a lot of fields, from chemistry to micromechanics. Then I felt it was time to begin with a new start-up.
What did you have at the time? Basically, we started to develop what is now our second-generation product (see “Peeping inside the box”). Microelectronics and specifically the semiconductor industry had developed techniques that have pushed the miniaturisation frontier to the limits. That is what Biocartis is about. We want to commercialise an instrument that is easy to use, cost efficient and small enough to embed the same molecular diagnostics instruments that today only university hospitals or research centres can afford. This means not only miniaturisation but speed, reliability, low cost and multiplexing – in other words the possibility to conduct multiple experiments on the same sample.Those are not trivial challenges… No. Technologically speaking, multiplexing was the most difficult to tackle. We need to look at multiple biomarkers linked to viral or bacterial infections, and at human cells at the same time. In the end, we want to come up with a system that is comparable to a Nespresso coffee machine. You have a touch screen that communicates with the healthcare information system and with the instruments that perform the tests. Our instruments are embedded in a box the size of a desktop computer, with four drawers that take disposable cartridges that contain the samples as well as all the reagents needed to perform the tests. There are 27 patents on this cartridge alone. 25 tests can be conducted simultaneously on each cartridge, with results in minutes. Basically, we have miniaturised a large laboratory into something that looks like a desktop PC. And just like the PC in the computer industry, this will drive down the cost of molecular diagnostics dramatically. By what factor will costs reduce?Ten or more. The cost of our system has to be finalised until the launch in the second part of 2012. But it will be below 50 000 US dollars. Equivalent kit today costs around 400 000 to 500 000 dollars. But there is more. Some of the costs are driven by the level of qualification and training of the operators. With our “Nespresso” model, our system comes in handy. We’ve incorporated the principle of cost savings right from the beginning, in the design as well as in the industrialisation processes. Why are costs so important? At present 90% of hospital laboratories in the Western world are not doing molecular diagnosis. They need special infrastructures, they need to train people, and the machines are too expensive. So you need a disruptive technology to let those molecular diagnostics become a routine for glucose monitoring or pregnancy test, for example. A second push comes from the reimbursement system. The healthcare budget has become such a large proportion of a nation’s budget. With the ageing of the population this trend seems unsustainable. Already insurance companies and government are saying: “You have those new treatments for cancer but only 20% or 30% of patients are going to respond. Why would we reimburse an inefficient treatment?” In the UK, there are already examples where the insurer reimburses only if the treatment proves its efficiency. That is an entirely new paradigm.
USER FRIENDLY The molecular diagnostic platform aims at simplicity.
And why would molecular diagnostics limit this cost inflation?We are in a phenomenal phase of medical advance. Thanks to the increasing number of biomarkers, we can diagnose a disease earlier and therefore prevent it. Prevention is very cost effective. And when the disease is there, we can better predict the response to a specific treatment. In other words, it is possible to use specific molecular diagnostics to sub-classify patients to give them only the drugs that we know are effective. But in order to fulfil this promise of personalised medicine, molecular diagnostics still have to reach the final points of care. How will you do this?The system we have created will be efficient for the 10% of hospitals that do molecular diagnosis today. It is cheaper, easier of use and it offers random access 24/7. Still, our philosophy was from the start to create a system that can be deployed in all kinds of settings. We are not going to reach the individual nurses or doctors. But the revolution will happen in places that have no test facilities today: retirement homes, pharmacies, primary care physicians, ambulances, airports, military installations, and so on.And what about the laboratories? Are they going to become irrelevant if diagnostic tests become ubiquitous?Absolutely not. Take the Nespresso comparison again. Coffee bars have not disappeared. It has just increased access to excellent coffee. Laboratories have contributed enormously to healthcare and will continue. They have qualified and experienced people that will remain crucial to diagnostics. What will happen is that with accessible molecular diagnostics, we will increase their repertoire of technologies and solutions. Still, your molecular diagnostics for the masses may be very disruptive for the pharma industry, whose model is built on one-size-fits-all drugs and blockbusters?We are entering a period where there is no other option. The status quo is not sustainable: the costs of drug development have spiralled while productivity has declined. The vast majority of pharma companies already see biomarkers and molecular diagnostics as a central piece of drug discovery and development. Actually, the fastest growth rate in the industry is now seen in the molecular diagnostics field, mostly DNA-based tests.Why?Pharma companies need more efficient processes to survive. Tens of thousands in revenues have disappeared because of non-reimbursement. Blockbusters won’t disappear but they are already becoming rare. Failures during clinical trials have increased. So researchers need to understand the deep mechanisms of diseases better to be able to design predictable models for a new drug. During clinical trials, biomarker tests could also play a critical role. They will allow patient populations to be stratified into categories for which a specific drug is efficient or not. So you can design better clinical trials and drastically reduce failures. It may be even that some promising drugs that failed in clinical trials in the past will get a second chance. Does this mean that in the future regulators will allow a drug to reach the market if is accompanied with a specific molecular diagnostic?Swiss pharma companies like Roche and Merck Serono, already have such combinations authorized by the FDA or EMA. In the US, the Path initiative is a move toward this direction. And in Europe the European Medicines Agency is looking at a system of authorising a drug with a diagnostic. For those agencies the priority is patients’ safety. That is why we could not compromise on the quality of information extracted from our system. The person who is operating this system will receive information from which to take their final decision.
TEAMWORK Nader Donzel (left) and Rudi Pauwels are both seasoned entrepreneurs.
PEEPING INSIDE THE BOX
Created in 2007 in Lausanne, Biocartis raised 10 million euros in October 2009 from investors such as Biovest and Aescap Venture and another 30 million euros, six months later, with Johnson & Johnson and Debiopharm joining as lead investors. That unusual speed between financing rounds is partly explained by the acquisition of a DNA/RNA automated molecular diagnostics testing technology developed by Netherlands-based Philips.
This acquisition complements the various technologies developed in collaboration with EPFL and will be at the centre of Biocartis’ first-generation product. In this system, named MDx, the instruments that detect and analyse biomarkers are mostly miniaturised and integrate technologies currently used by large molecular diagnostic laboratories. Biocartis chose to not introduce new tests yet, in order to reduce hurdles in market authorisation.
Still, the leakproof disposable cartridge that contains the reagents and the samples was extremely challenging to realise. That is why it is protected by 27 patents. For example, in order to be able to test not only liquid but also solid samples such as tumour cells, a sonic technology that liquefies tissues in 40 seconds has been embedded in the instruments. Users will also find all the technologies needed to extract biomarkers, purify and amplify them – in a box the size of a desktop...Silicon microcarriers
But using Swiss know-how in miniaturisation was not enough for Biocartis. With its second-generation system, operators will be able to conduct not only 25 tests on the same sample but hundreds, potentially thousands, and faster. That is because this technology is based on semiconductors and microfluidic techniques mastered by co-founder Philippe Renaud and his team at EPFL.
Biocartis has created extremely small microcarriers out of silicon wafers. Each of these nanoscopic pellets is dug with a series of little holes, creating a binary code of 1 (hole) and 0 (no hole). Each microcarrier is also chemically bonded with a type of molecule (antigen, DNA, protein, etc…) that will react when meeting specific biomarkers. On the cartridge, those reactions take place within seconds in a channel less than one millimetre deep where microcarriers are floating when the sample is introduced. Outside detectors read the reactions, identified thanks to the binary code imprinted on the microcarriers.