I started my first company in 1987 because I realized it was an effective path for transforming science into life-saving and life-improving inventions. Startup companies provide one means for accomplishing ends that interested me: creating products that have a positive effect on human health. I did this first with a colleague (Box 1), but through the years I have also started many companies with students and postdocs in my MIT lab.

Beginning the journey

I had been drawn to chemistry since I was a boy, and this led me to obtain BSc and ScD degrees in chemical engineering from Cornell University and MIT, respectively. While in graduate school, I hoped to use my background in chemistry and chemical engineering to improve people's lives, so after earning my doctorate, I went into health-related research. This was against the norm in the chemical engineering field at the time, and most of my colleagues followed more traditional career paths, such as petroleum engineering, which was popular among chemical engineers. I realized early in my career as an assistant professor at MIT that breakthrough science can help save lives. My goal was to discover or create such technologies and then publish the research, hoping that commercial ventures would recognize the value in these ideas and develop them. Though I had some success publishing in academic journals, at first there was no commercial interest.

Around this time I also realized that the path from invention and discovery to benefitting patients was both difficult and expensive. I decided that if my ideas were to be translated into products, they would need patents, which would help incentivize companies to invest. I was awarded my first patent for plastics that could slowly and continuously release large molecules—the patenting process took seven years and taught me a great deal. In some regards this plan worked: two large companies licensed the patent—one in animal health and one in human health—and the patent also helped bring in grant money. However, the companies quickly gave up on these projects and essentially shelved the patent (Box 1). This was my first lesson: if I wanted my inventions to have an impact on people's lives, then I would need to get involved in the commercialization process myself. So I did.

A formula for a startup

My criteria for starting a company involves three Ps: platform, paper and patent.

Platform. A platform technology should be able to be used over and over in different applications. Good examples of platform technologies are genetic engineering, which has been used to make various protein therapeutics, as San Francisco–based Genentech does, and polymer microspheres, which can deliver different drugs for long time periods, as Dublin-based Alkermes' products do. You know you have a platform if you can envision a number of distinct applications for the new idea, even when you are in the early stages of research.

A technology could certainly be interesting and make an intellectual contribution to science or engineering while still not being a platform—but for it to be ripe for a startup, ideally it should have product potential in several areas.

Can you be successful with a one-off idea? Absolutely. I think, however, that it will generally be easier to find support (funding, collaborators and more) and your company will have a greater chance to have an impact with a platform technology rather than a single-shot product.

Paper. I generally want our research to be published in high-profile journals. This was also desirable for starting a company in the 1980s, and it is still desirable today. It is not about being elitist; rather, getting through the peer review process of a highly selective and competitive journal validates that the idea may be a significant breakthrough. For example, a paper in Science1 led to an aerosol company we started called AIR, and a paper in Nature2 led to the company MicroChips, based in Waltham, Massachusetts. In my experience, an idea without a high-profile paper describing the science behind it is harder to turn into a company—and starting a company at all is already a high hurdle to leap. Without the published research, I have found it more difficult to acquire potential partners for both the science and business sides. One may also want additional papers to demonstrate the proof of concept, particularly if the idea is especially groundbreaking.

Patent. In my lab, patents flow directly from the publications. Ideally, these are blocking patents, written as broadly as possible, to protect the platform technology. A blocking patent has claims so broad it not only allows one to practice the invention but also will prevent others from practicing anything close to it. Such patents are important for getting investors and partner companies to put in the required funds for development; without a patent, investors will not feel their high-risk investment is safe from others simply copying the process or product once it is approved. There are people who say patents restrict the use of a discovery, but in my experience I have found that is rarely the case, particularly in early stage R&D, which is almost always the stage at which academic discoveries are made. Without substantial funding, those discoveries will never get to the point at which they can help people.

Other elements

Those three elements are important for launching a startup, but there are at least two others needed to solidify a bioentrepreneur's prospects. The first is convincing proof that the platform technology is viable. This is called proof of principle and almost always requires going beyond test tubes; it involves showing strong in vivo or animal data. Proof of principle, as well as everything else mentioned above—platform, paper and patent—generally falls in the category of doing good science. For example, when we developed a new nanotechnology approach for targeting drugs to tumors, we wanted to show that it worked in animals and shrinks tumors. This research was completed3,4 before Omid Farokhzad and I started BIND Biosciences, based in Cambridge, Massachusetts.

The second element gets into the human factor, and it is at least as important as the rest: behind every successful platform is a champion (or champions) who had great desire to get products to patients and were practically willing to walk through walls to do so. John Santini, who did his thesis on creating a drug delivery microchip2 and started MicroChips with Mike Cima and me, is a good example. He solved problem after problem, taking his PhD thesis to animals5 and then humans6. I have been lucky to find several people like this at MIT and in the greater Boston area, with its wealth of universities and biotech companies.

Final points

I established this approach to building companies early on, and it has not changed over the decades even as the appetites of investors have shifted (Box 2). The number of ideas looking for funding is probably larger today, which may make it harder for a good idea to stand out, but that is why I think having a solid foundation (like the three Ps) may be even more important now.

This is not to say that our way is the only way; it is just what has worked for us (see “The Langer lab's secret sauce” p. 490), and the above criteria are just the beginning—having the right business partners, a great CEO and supportive and wise investors all become essential. At startups, the CEO and investors need to work together, and a good venture capitalist can be very helpful in recruiting the right CEO and/or management team for a new company. Through the years I identified the venture capitalists I want to work with, and I have worked with them over and over.

Beyond the brass tacks of building companies, you have to believe in yourself (even when others do not) and never give up. I can say this from experience as my research on plastics for controlled-release macromolecule delivery was generally thought unachievable at first7, but became the science behind Cambridge, Massachusetts–based Enzytech (Box 1). I sometimes joke that the one thing I had going for me was that I had not read the literature saying my goal was impossible, and thus I was not as discouraged as I might have otherwise been. When I finally had the data I wanted and presented my results to a group of chemists and engineers, the audience members confronted me, saying they did not believe anything I had just said. It took a number of years before the validity of these findings became widely accepted. During those years, I had to persevere despite enormous difficulty in obtaining support for my research. Funding agencies such as the National Institutes of Health rejected nine consecutive grant proposals during that time, scornfully in some cases. Yet I never stopped believing in the work8.