Pioneering the Field of Stem Cell Biology With Irving Weissman, MD

Transcript edited for clarity. Any views expressed above are the speaker's own and do not necessarily reflect those of Oncology Data Advisor. 

Oncology Data Advisor: Welcome to Oncology Data Advisor, I'm Keira Smith. Today, I have the honor of being joined by Dr. Irving Weissman, a Professor of Pathology and Developmental Biology and the Director of the Institute of Stem Cell Biology and Regenerative Medicine at Stanford University. Recently, Dr. Weissman received the Wallace H. Coulter Award for Lifetime Achievement in Hematology at the ASH Annual Meeting in New Orleans. Dr. Weissman, thank you so much for joining me today.

Irving Weissman, MD: Thank you.

Oncology Data Advisor: To start off, would you like to tell me about your background and how you came into this field?

Dr. Weissman: I was 10 years old, and a teacher gave me a book to read called Microbe Hunters, about the lives and discoveries by the first microbe hunters, semi-fictionalized, just to get the idea of the discoveries and how the intensity of making and translating these discoveries affected their families and the social impacts of them. It was a spectacular book. It had people like Pasteur, Ehrlich, Koch, all of the people who really did establish how you could study something, how you could know that what you were studying was reproducible and true, and, of course, how you could isolate the agents that cause the disease.

When I was 16 in high school, I heard that there was a pathologist in town, who, although he was at a private hospital, was carrying out a research program. He was the editor of the journal Transplantation Bulletin [now Transplantation] and was doing lots of experiments on the genetics of tissue transplantation in mice. I convinced him that it would be good if I could work in the lab starting in the summer, and he quickly transitioned me, even though I was a high school student, to owning projects. He did it because he was teaching me Socratically. He wouldn't tell me the results, but when we would go through a paper together, such as the paper by George Snell to define the H2 transplant locus-MHC, he would explain the Latin and Greek terms, which was the hardest part. Later in the summer, he showed me a paper by Medawar's group on immunological tolerance. That intrigued me enough that I started my own experiments.

From age 16 or 17 on, I've done my own experiments. I've had mentors or patrons who were role models on how they approached experiments. I talked back and forth with them, but I really have done my own experiments. Now, that was important because I was never disciplined enough to get great grades. I got good grades, but not great. The only reason I got into the medical school that I applied to was my research. I only applied to Stanford Medical School because it was required five years instead of four in 1960. The basic sciences classes were scheduled to be done in three years instead of two. We had didactic classes just half a day starting first year, leaving a half day every day for research or other scholarship. I did research a half-day during the year for three years, and full-time in the 3 summers. Well, that's a lot of time for research, and I took advantage of it.

I did a lot of work on the thymus and immunological tolerance. I went to Oxford to work with Jim Gowans in 1964 between my clinical years, and there showed that the thymus mainly made cells that emigrated out to populate the immune system. Now, that was important only because it was definitive. I would put labels like tritiated adenosine or tritiated thymidine in the thymus of a living rat and infused the cold nucleoside to prevent uptake at the nucleoside level by cells outside of the thymus. I think that was amongst the first lineage tracing experiments done. I knew for sure what the cells were. I knew that they went to a particular area of the lymphoid organs and avoided the other—the lymphoid primary or secondary follicles. The other turned out to be B cells.

After medical school, I decided to stay fulltime in science, I worked on defining lymphocyte homing receptors for T- and B-cell homing to lymph nodes—we discoveredthe protein-CD62L via specialized high-walled venous endothelial cells first discovered by Gowans—and to Peyer's patches-integrin a4b7. These cells, upon activation by their antigen receptors, switched to shut these off and express a4b1 the homing receptors for blood vessel luminal VCAM-1. We also cloned the genes expressing them. These experiments turned out to be important science, but also very practical in clinical translation.

For example, about 10 years ago, a company made antibodies against the integrin a4b7 homing receptor, which T and B cells used to go to the gut, like the Peyer's patch, appendix, and tonsil. They got stimulated in those lymphoid organs, and then their effector cell went to those organs selectively. Blocking antibodies to a4b7 have clinical usefulness for inflammatory bowel diseases. If you think about COVID and the fact that we immunize in the arm to trigger CD62L-positive lymphocytes, that doesn't trigger a4b7-positive lymphocytes. These immune and immunized lymphocytes might be important for protection of the body, but not protection against infection in the nasopharyngeal cavity.

Anyway, that's the kind of background. As I was looking at T-cell development, my patron mentor at Stanford Medical School, Henry Kaplan, had worked on how radiation causes acute lymphocytic leukemia, the T-cell type. He had discovered that a retrovirus was activated by the radiation, the retrovirus targeted cells in the thymus, and caused leukemia. Others showed each leukemogenic virus was generated by recombining to generate novel viruses. In the 1970s to mid-1980s, we showed that each lymphomagenic retrovirus bound to the lymphoma surface by its T-cell antigen receptors and coreceptors, and the binding led to virus infection. The T cells responded byentering cell cycle, continually produced the "antigenic" retrovirus that infected and triggered them into cell division as part of the leukemogenic process. We also showed that a single B-cell lymphoma engaged its produced retrovirus by its antibody cell surface molecules. All of that research was finished in the mid-1980s, and the model, which we called receptor-mediated leukemogenesis, was amongst the most unpopular set of papers I've published.

Probably our biggest line of discovery that I started in the 1970s was asking questions about whether blood formation and blood-forming stem cells can be tracked all the way back to embryonic life. At Oxford, again in the Gowans lab on sabbatical, I showed with Richard Gardner and Ginny Papaioannou that the yolk sac blood island cells—before the anastomoses of blood circulation between the yolk sac and the embryo proper—could be isolated. I transferred these yolk sac cells to the yolk sac of an allogeneic mouse fetus of the same stage, and these recipients developed lifelong yolk sac donor blood formation of all blood cell types .

A lot of people think that's wrong, but I did the experiments with my own hands. We transplanted cells from one embryo pre-circulation to another. Whenever you do a transplant, of course, it could be an artifact, but I'm pretty sure it wasn't. That led me to think, "Well, what is the origin of the cells that make the immune system?" By 1986, we were almost there to isolate the mouse hematopoietic or blood-forming stem cell (HSC). In 1988, we isolated and transplanted mouse HSC, which self-renewed and differentiated to restore mouse hematopoiesis for life.

At SyStemix, a company formed by Mike McCune and me, we isolated human HSC within three years. At that time, between 1990 and 2000, patients with metastatic breast cancer were being treated with high-dose combination chemotherapy (HDCT), approaching doses that wiped out the blood-forming system. Many labs championed rescuing the blood-forming system with the patient's own cells that had been taken before the HDCT. They harvested patient bone marrow (BM) or mobilized blood (MB), froze the viable cells, treated the women with HDCT, and then infused the autologous BM or MB cells. We found most of the mobilized blood had cancer cells in it, so the women could be transplanted with blood forming cells and also cancer cells. We showed the purification eliminated the cancer cell contaminants in MB, so we rescued the women from HDCT with their own HSC, free of detectable breast cancer cells.

SyStemix was purchased by a large pharma to transplant normal or gene-modified self HSC, to transplant allogeneic healthy HSC to regenerate defective or autoimmune blood systems with healthy immune systems, or even transplant HSC to induce transplant tolerance of organs or tissues from the HSC donor. By the early 2000s, the large pharma abandoned HSC transplant therapies. We knew that that project was gone. At that time I was also head of a US National Academies Committee on stem cells and nuclear transfer from somatic cells to enucleated oocytes to clone embryonic stem cells [ESC]. When we made our report, the next day, the US President by executive order defunded production of new ESC and of all nuclear transfer experiments to generate pluripotent stem cell lines.

That led to a number of parents of diabetic patients and children of Alzheimer's patients contacting me and really pushing pretty hard to ask whether ending that research could affect progress to understand and to treat their family's disease. I reported that in my view it would affect both, but that it takes a long time to accomplish research and translate it clinically. I would tell people that it's at least 10 to 20 years from a discovery to an approved product by the FDA , and very few discoveries made it that far. Several of us helped Robert Klein write a California initiative called Proposition 71 in 2004 to fund this research.The proposition passed, creating an agency called the California Institute of Regenerative Medicine (CIRM). Eventually, that allowed us and others to fund research through the agency as if we were starting a biotech but doing it at our university. We didn't have to worry about profits, just whether it works and whether you can do a clean enough experiment to show it works.

Around 2010 to 2012, I had just stepped down as Head of our Cancer Center. I was Head of the Stem Cell Institute, but I had asked physicians in our bone marrow transplant group to follow up and to see what happened to all those patients in the SyStemix trial who'd gone through the autologous cancer-depleted HSC rescue from HDCT for their metastatic breast cancer. Seventy-four contemporaneous patients got the usual mobilized blood rescue, the standard of care at that time. They weren't prospectively randomized, but it was the same doctors, same drugs, same protocol, except they got mobilized blood to rescue. The median survival of those given unpurified autologous MB was two years, the same as palliative intent chemotherapy, reported in 2000, that ended that line of therapy.Fifteen of our patients were rescued with their own cancer-free HSC. For our patients tested in the trial with metastatic disease between 1996 to 1998, their median survival was 10 years. Their overall survival at 15 years and now 25 years is 33% without disease.

Oncology Data Advisor: That's amazing.

Irving Weissman: It is amazing. So, I negotiated back from the pharma that purchased SyStemix the hybridomas that make the antibodies to isolate stem cells pure, to put them at Stanford in a not-for-profit setting. I didn't want us to end up in this problem where money was going to be the issue, rather than, "Does this work?" Now, that took a long time, but I got them there. I've shown they work in testing.

But when I talked to many breast cancer oncologists, they countered by saying that stem cells don't work. I'd say, "How do you know that?" They said, "Well, here are the papers." Every one of the papers were on mobilized blood contaminated with cancer. I'd show them the data, and they said, "Yeah, but we know that stem cells don't work. Sadly, the titles of papers called mobilized blood "stem cell transplants." I couldn't convince either the journals or the bone marrow transplant community to name the transplants for what they were—MB or BM cells—rather than what seemed most attractive to them, HSC. The single most important thing I could get across in this interview is that the main barrier to taking these translations to people isn't an ignorant public. It's the clinical transplanters and breast cancer oncologists who have been taught that MB transplants are HSC transplants, and that such transplants don't change the fate of metastatic breast cancer patients

The last thing that led to the award, I think, was because we had blood-forming stem cells in mice and in humans, we had shown that the very first step of differentiation from HSC was to a cell. At the single-cell level, this could make all blood cells but didn't self-renew like the stem cell. It was multipotent but not self-renewing. Over the next decade, we worked out nearly all the steps of differentiation from HSC to blood in mice and humans. Because we could purify every cell in the hierarchy, that would allow us to compare human adult onset myeloproliferative diseases such as human chronic myelogenous leukemia myeloid blast crisis, acute myelogenous leukemia (AML), MPN and myelodysplastic syndrome. We could ask the question, "Do the mutations that lead to human acute myelogenous leukemia accumulate in HSC, or do they accumulate in the cell that normally doesn't self-renew, but these mutations might give self-renewal to it?"

We showed even back in 2000 that the Hiroshima Hospital samples where people got acute myelogenous leukemia—many with translocated AML1 to ETO AML, that the multipotent progenitors (MPP) that normally don't self-renew were the leukemia stem cells (LSC]) that self-renew.and transfer the leukemia to immune-deficient mice. Most striking was that same exact translocation was in a fraction of HSC, the precursors of the LSC MPP; this aml1/ETO-translocated HSC, didn't transfer leukemia. Then we did a study of taking fresh Stanford patient AML samples and sequenced the exons in their DNA to find the mutations that the patient had. We made DNA primers so we could look at the mutant allele and the normal allele, and then we looked in their hematopoietic stem cell one HSC at a time. In all patient samples that Ravi Majeti and I and our labs analyzed, all but the last mutation were in hematopoietic stem cells, and the driver early mutations expanded the clone at the expense of normal HSC. Then we realized those genes didn't have to turn a non–self-renewing cell into a self-renewing cell, because they were hitchhiking in self-renewing HSC.

The biggest surprise was that the last mutation was almost always the oncogenes that people had previously discovered: FLT3, internal tandem duplication (ITD), NRAS, KRAS, activation of beta-catenin, etc. The first mutations in almost all of them altered the hematopoietic stem cell's ability to give rise to daughter cells that shut off some genes and turned on others—epigenetic alterations in gene expression. Often the first mutations were loss of function changes in TET2, which is required to demethylate cytosines in DNA, likely to open the chromatin for transcription. Another initiating loss of function gene was DNMT3A. It can methylate cytosines in the DNA to close down and therefore shut off gene expression. Each of those were in most of the leukemias, so we learned that there was a progression to it—the early mutations likely altered opening or closing suites of genes usually necessary for self-renewing HSC to differentiate to non–self renewing-progenitors.

When we compared the gene expression profile of the LSC and normal HSC , we found that the cell surface CD47 protein—a "don't eat me" signal for macrophage—was overexpressed on all LSCs. We made blocking antibodies to CD47 and found that it would enable macrophages to eat it, because CD47 is a "don't eat me" signal working through SIRPα. Patient-derived AML LSC xenografts in NSG immune-deficient mice could be "cured" if the leukemia was treated when the leukemic burden wasn't too large.

So, we got a CIRM grant. Both in the UK and in the US, we did all the early studies as if we were a biotech, but we did it at Stanford. Within four years, we filed an Investigational New Drug Application (IND), going from a mouse antibody, no pre-clinical, to an approved IND, and began phase I clinical trials all while at Stanford, supported by CIRM and the Ludwig Foundation. By that time, we had discovered all cancers have CD47. At that point, we formed a company, Forty Seven, Inc, and Stanford licensed us and another independent company.

At Forty Seven, the team continued development of the anti-CD47 antibody and looked for ways to increase the "eat me" signal—cell surface calreticulin was the dominant entity we had found at Stanford—by increasing calreticulin or adding other antibodies that were relatively tumor selective and were using the human immunoglobulin IgG1which opsonized the tumor for macrophage phagocytosis via its high affinity FcR. In older folks who get AML and myelodysplastic syndrome (MDS), a preleukemic syndrome, we found that if you add anti-CD47—which we'd shown was a safe, humanized particular immunoglobin isotype (IgG4)—to azacitidine, a non-curative drug, that helps increase the "eat me" signal on those cells. With the combination of blocking "don't eat me" and increasing "eat me" selectively on the pathological cells, all patients respond.

By the time my company got sold—I'm going to be serious about that in a second—we had at least two or three years where the disease didn't come back in roughly half of the patients with leukemias and MDS. We showed that by combining anti-CD47 with rituximab, the IgG1antibody that binds lymphoma CD20 and interacts with the high-affinity FC receptor on macrophages, this increases the "eat me" signal, even more than what azacitidine did to increase the "eat me" signal. If you combined rituximab and anti-CD47 in patients for whom nothing worked anymore—not rituximab alone, not chemo, not rituximab plus chemo—in over half of them, the tumors started shrinking. I think the last time I saw it, about half of those went into molecular remission.

Forty Seven was sold about 5 years after it was founded. By that time, I realized the function of a company is to make a profit, while the function of academic institutions is simply to advance medicine for the benefit of people. Back in 1994, I wrote an editorial as President of the American Association of Immunologists that opposed the Bayh-Dole act of gifting to academic institutions the rights to license intellectual property to commercial organizations without contributing to the federal agency that funded the discovery research. I thought that arms-length methods could have provided evergreen compensation to the government agency that funded the research, and so included that in the CIRM initiative. Thus, at Stanford with CIRM and Ludwig funding we developed anti-leukemia, anti-myelodysplastic syndrome, and anti-lymphoma treatments that by state law required the university—here, Stanford—to contribute part of its CD47 licensing funding to the state.

With that knowledge, we're bringing back HSC transplants for each of the diseases that we're looking at, including autologous cancer-free HSC rescue from HDCT treatment of patients with metastatic breast cancer. We just have to convince breast cancer physicians to read the data from the purified, cancer-free HSC rescue study published in 2012, and let their patients come into the trial. 

As I said, all of the experiments I got to do are because I learned in high school that I could think about a biological phenomenon and design an experiment, even though it may have been crude at the time, to approach the mechanisms involved.

Oncology Data Advisor: That's absolutely fascinating. Thank you so much for sharing all about your work. It was really interesting to hear about.

You mentioned the barriers in translating the research from the lab to the clinic. Would you have any advice either for the broader scientific community or even oncology clinicians about how to overcome these barriers?

Irving Weissman: Sure. First, I would recommend all oncology residencies to include basic science lab time to see how tough it is to come up with answers via experiments; and not being so glib that you think you understand how and which treatments have a chance, even though the paper and topic weren't considered by their mentors during their training. Second, the experiment in California at an academic center had way more value than I could even believe, because we had to get together the professional people to help us who had done the regulatory and other preclinical testing for presentation to the FDA.

For that, I thank the leadership and training during the run up to the anti-CD47 trials provided by Susan Jerian and David Essayen. They would meet with us every week. We didn't have a CEO. We didn't have a business development person. We were just meeting every week with the team to say, "What do we do this week? What happened? What went wrong? What could we change?" and so on. The MD students, the PhD students, and others on our project—especially us-- learned during this period how preclinical work is done that leads to filing INDs and starting phase I trials. None of us got that training during our doctoral degree educations.

I wanted to keep doing that until we finished the phase 1 trial, because we got enough money to do the lead-up to the trial, filing our IND in the UK and the US, and then fund at least phase 1. The university made the decision that there was too much financial desire from large and small pharma and venture capital companies, so we had to form a company. Now, that company, Forty Seven, was the leader in this kind of antibody research. All of the others that got licenses started without the ability to have both the discoverers of the field and their continuation during the nonprofit phases of development. I'm convinced that the people who discovered CD47 and how it works were seeing everything that was happening in the preclinical and clinical trials, and they had the position to opine on what it meant and where you should go.

Every one of our discoveries that's made its way to big pharma goes to silos. These are tried and true ways to develop the kinds of projects that made big pharma in the past. It's hard for them to talk to each other, and usually, there's nobody there who made the original discovery. If you think about it, we have set up systems largely because all of us are under the same economic system. A biotech or pharma company is appropriately more risk-averse and is driven largely from the fact that the function of the company is to make a profit.

I think that if there's any lesson to come from this, the first lesson is that you need to find a way to fund things that are just post-discovery with the discoverers until at least an IND can be filed. Then they carry it forward, because only they will see when a mouse or a non-human primate has an aberrant reaction. They'll know enough to see how to pivot, and therefore what to do. They know the system so well. They know how many variables you can look at, and it would be way more efficient. Biotechs spend a lot of money for early phase to try to chase down these unexpected events, when they might have gotten the answer from the inventor of the field.

The final one is that many people who get into great colleges from high schools and get into medical schools and get into high-end training and hematology/oncology—the real high-level, high-income, but highly interesting subspecialties—they get there by memorizing and being disciplined to memorize all the way. They can become great physicians in medical practices, but they do so without becoming medical scientists. We should all be grateful to land in the care of these great physicians. But we shouldn't think they have any special insight into the biology and pathogenic mechanisms of the diseases their patients have. To advance a field to the next step requires scientists who study mechanisms. These could be molecular mechanisms, where the elements of molecular pathways become revealed, and errors in those pathways become revealed by studying diseased patient materials for pathway alterations. To people like me, the fundamental irreducible element in stem cell biology and regenerative medicine is the stem cells themselves. No single molecule controls stem cell features. I suspect that to understand how to lead these nascent fields, one must do a retrospective analysis of how the discoverer got to the stage of being a discoverer. I doubt it is grades.

I think that a study needs to be done to think about, "Can you find out outcome measures by measuring outcomes?" Look at what they did early on. Look at what the granting agencies did with their grants, and on and on. There are properties of mind and character are that allows somebody to go on and inquire and do experiments, versus those qualities which are critical in my field to become a doctor. And both then have to work together, openly admitting the advantages and disadvantages of their previous training in trying to make a discovery into a real therapeutic.

Every good doctor has terrific recall and well-trained methods to translate observations to actions. You see A, B, C, and D, and you do F for the patient. If you didn't have that smart memorization and lot of experience and reinforcement to make the transition between observations and actions, we'd have a lot of trouble. But don't ever ask those people what is going to work or what they would do for research to change an incurable disease to something that might be cured, because they have no background.

Oncology Data Advisor: One last question I have for you is, do you have any advice for medical students or residents who are interested in becoming involved in basic science research?

Dr. Weissman: Yes. Take the time in medical school to do research, to look for the field as early as your first or second year of medical school—or better yet, as a high school student or as an undergraduate. That's what gets you into graduate school; nobody comes into graduate school just with great grades. They must have done research, and we evaluate them by being able to talk to them about the research they've done. I would say get involved early, as early as you can, and then that will help you see the medical field you should be in. If you're doing research and you see it will apply to cardiology but not hematology, then you can start tailoring your career early, not late.

Oncology Data Advisor: That's really great advice. Thank you. Well, those are all the questions I had for you, so thank you again.

Dr. Weissman: Yes, nice talking to you.

Oncology Data Advisor: Nice talking to you too.

About Dr. Weissman

Irving Weissman, MD, is a Professor of Pathology and Developmental Biology and the Director of the Institute of Stem Cell Biology and Regenerative Medicine at Stanford University in California. Throughout his career, Dr. Weissman has made monumental contributions to the field of hematology, including his research on hematopoiesis and hematopoietic stem cells. Dr. Weissman is considered the father of hematopoiesis and a trailblazer in stem cell biology and drug development. His contributions to the field include being the first to discover and isolate blood-forming stem cells in mice, publishing the first isolation of human blood-forming stem cells, and discovering that CD47 can be targeted with blocking antibodies for the treatment of blood cancers and drug-resistant lymphomas. Dr. Weissman has been the recipient of numerous awards, including the 2022 Wallace H. Coulter Award for Lifetime Achievement in Hematology.

Transcript edited for clarity. Any views expressed above are the speaker's own and do not necessarily reflect those of Oncology Data Advisor.