Jim Allison Commencement Address at UC Berkeley

May 31st, 2017

James P. Allison,

University of Texas, MD Anderson Cancer Center.

Commencement Address.

University of California at Berkeley.

Department of Molecular and Cell Biology PhD Program.

May 19, 2017

[Prepared remarks, lightly edited]

MCB graduates, Class of 2017 – Congratulations!

I have the honor today to offer my congratulations on finishing, or should I say surviving, ~5 years in a rigorous training program and preparing for a career as a scientist. I know the program well, having spent nearly 20 years, from 1984 until 2004, as a faculty member here, helping to put that rigor into this program. These were among the best years of my life, because UC Berkeley is an amazing place, with a rich history of world class science. This is in large part due to the quality of its students, who are the best, in terms of preparation, commitment, curiosity and drive that I have encountered during my career.

In addition to congratulating you, I would like to take a few minutes to make some observations, offer some advice, and tell a few stories that might be useful, some of which might help the non-scientists that are among your friends or relatives understand why we do what we do.

I think that for many of you, grad school may have been the best period of your life so far. It was for me. Why do I say that? Because we don’t choose science so much as science chooses us. I think most start because we are curious, we want to know how things, in our case living things, work. Science offers a way to do that. You may have had a chemistry set in your garage, or a microscope in your bedroom as a kid. In high school you liked the science courses, maybe worked in a lab as an undergrad. But it’s in graduate school you really learn how science works. You generate a hypothesis, design an experiment to test it, and then do the experiment to get the answer. Easy, right? Unfortunately, it’s not so easy. Too often the experiment doesn’t give a clear answer, or, worse, it generates two new hypotheses. Science is hard, very hard, because you mostly learn what is not correct.

My dad was a physician and he wanted me to consider medicine. I entered college as Pre-Med. But as I began to work in a lab, I realized there was a fundamental difference between practicing medicine and being a scientist. A physician needs to have his head full of facts that he can draw upon to generate a treatment algorithm according to how a patient presents. And it had better be correct or the patient will not benefit, and may even be harmed. I certainly didn’t have enough discipline for this. A scientist, on the other hand, has a different job. A scientist should generate new ideas and test them. Many, if not most will be wrong, especially on the leading edge of a field. Thus we are supposed to be wrong a lot. It is sufficient to be correct only some of the time, hopefully about important things to keep the field moving.

Hopefully grad school was one to the best moments of your life because you learned how to do this. It sometimes may not seem so great, especially if you are frustrated by a period of failures. I heard one my students say during such a period: “Is this really as good as it gets? If so, I think I’ll shoot myself”. You need to learn to love your data, even if it is disappointing, because data are like children. Once you have them, you have them. Some may be disappointing, but once you’ve got it, it’s yours. You can’t just throw it away. You need to spend time with your data, love it, understand what it’s telling you, and move on.

I think that there is a dirty little secret about what drives scientists to keep working when the going is tough, and there is no instant gratification That is to get the ego boost of being the first, and for a while the only, person on earth who knows the answer to an important problem.

Most of you, I expect are going on to postdocs. This next phase in your development that should be even better than grad school. You learn to develop your own ideas, with progressively decreasing guidance from your mentor, dive deeply into issues that interest you, free from the distractions that will burden you later in your career, such as the necessity to get funding, administrative duties, and workplace politics. Those distractions will come soon enough, when you secure your first permanent job, whether in a academia or pharma, so make the most of your postdoc.

If I may, I’d like to reflect on some aspects my training over the years. I did my both undergraduate and graduate training at the University of Texas at Austin in biochemistry. I should probably also mention that really enjoyed the music scene that was developing in Austin in the 70s. Stevie Ray Vaughan, Jerry Jeff Walker, Willie Nelson, and others and beer gardens near campus where we could relax after a long day in the lab.

Anyway, as an undergraduate I took a course in immunology and learned that a new type of cell, called the T cell, had just been discovered. These mysterious cells wander throughout the body, looking for signs of threats from bacteria or virus infections, and as we know now, cancer, and then dealing with the problem by killing the offending cells. I was fascinated. After dabbling with T cells a little bit in a side project in graduate school, I was hooked and decided to do my postdoc in an immunology lab, so I want to Scripps Clinic in La Jolla. I was in an immunology lab, but unfortunately my project was to make use of my training in biochemistry by sequencing MHC proteins. I was successful, but what I really wanted to do was work on T cells. Unbeknownst to my mentor, a fellow postdoc and I carried out a series of experiments on recognition of tumors by

T cells that culminated in a Nature paper and an offer for a position as Assistant from MD Anderson Cancer Center in Texas. Not at the main campus in Houston, but in an outpost facility in a park near Austin.

I was really lucky and had a fantastic time. No administrative or teaching duties, and the freedom to work on whatever I wanted. By then it was known that we have 10 of millions of T cell clones, each of which can recognize something different structures called antigens, but nobody knew what the T cell receptor for those antigens was. I decided to identify the T cell antigen receptor (TCR). After about 2 years work, I identified a protein composed of 2 chains that was expressed by all T cells, but that the fine structure of both chains differed from clone to clone, and proposed that this was the TCR. This was quickly confirmed by other laboratories.

As a result, in 1984 I was offered and accepted a position as Professor here at UC Berkeley. I couldn’t turn down the vast opportunities this offered, and accepted. I never regretted it. The intellectual atmosphere was inspiring, the collegiality of the faculty and students amazing. All made for an atmosphere that gave the feeling that one could accomplish anything.

By the late 80s it had become clear that althouth the T cell antigen receptor was required for T-cell activation, it was not sufficient, and another signal, called a costimulatory signal was required. We next sought to identify the molecule that was responsible for providing that second signal. In collaboration with David Raulet, we showed a molecule called CD28 was responsible for this second signal. We then began a search for genes that were similar to CD28 that might encode molecules with similar function. It turned out that a molecule called CTLA-4 was structurally very similar to CD28. As we began studies of the function of CTLA-4, another lab reported that it was another costimulatory molecule. However, Max Krummel, a graduate student in my lab, conducted a series of experiments showing that it actually did the opposite – it was inhibitory and its job is to terminate T cell responses.


Basic vs Translational

In thinking about possible applications of this knowledge, I thought that if we could block this inhibitory pathway, we might be able to enhance the responses against tumors. We embarked on a series of experiments that tested whether this was indeed true. Using mouse models, we showed that if we treated mice carrying experimental tumors with anti-CTLA4 antibodies, the tumors were rejected, and the mice were permanently immune to rechallenge with the same tumors. We found this work for many different tumors in mice, although with some tumors, we required an additional treatment such as radiation chemotherapy or a vaccine.

This finding changed my life again and raised the possibility of an entirely new approach to treating cancer by focusing on immune system rather than targeting the tumor itself. Since we were basically ignoring the tumor cell, I realized that with this mode of therapy, we could possibly treat virtually any kind of cancer by blocking this checkpoint. I began to search for a company that would help us develop a drug targeting human CTLA-4 to move our work to the clinic; however this idea was so radical that it took almost 3 years to do so, despite the overwhelming amount of data that we had.

Finally, a company named Medarex decided to help us. They made an antibody to human CTLA-4 named Ipilimumab, and took it into phase 1 trials. The results were very encouraging. In the first trial, one patient with late stage metastatic melanoma had complete disappearance of her tumors after about six months, and she is still tumor free 16 years later with no additional treatment. There were additional small trials showing responses in other cancers, including lung, prostate, kidney.

BMS, a large pharma company, teamed up with and then bought Medarex and extended the trials. Despite the success of the early trials, several large trials designed to obtain FDA approval failed. I felt this was at least in part because the trial was designed as if ipiliimumab was a cancer drug, that it would rapidly kill cancer cells. However, Ipi targets T cells, not cancer cells, and it takes time to accumulate enough T cells to eliminate the tumor. I felt that it was necessary to get more involved in the trials to make sure that they followed the biology accurately. So, in 2004 I decided to move to MSKCC in NYC, where many of the trials were underway.

After a series of trials, Ipilimumab was approved by the FDA in 2011 for the treatment of metastatic melanoma, meaning that any doctor can prescribe it, and insurance companies will reimburse. When we started this work, the median survival of patients with metastatic melanoma was 11 months. In 2014, a retrospective study of about 5,000 patients treated with ipi showed that 22% were still alive after 10 years.

[The prepared remarks end here; during the address, Dr. Allison went on to discuss additional checkpoint immunotherapies that have been discovered and that have also shown remarkable clinical successes. In particular, he highlighted the work on anti-PD1 antibodies by several pharmaceutical companies. He noted that there are currently more than 1000 cancer immunotherapy trials underway. He highlighted his current work at MD Anderson. He ended his address by discussing the challenges faced by newly minted PhD students. He lamented the diminished funding for basic science and the challenges and importance of doing science in what is increasingly a ‘fact-free’ world. He again congratulated the graduates and wished them the best in their future endeavors.]

Back to Top