Physician-scientist Matthew Porteus, MD, PhD, has been a mainstay in the genome editing field for more than two decades. He trained at Stanford University Medical School before completing his residency and hematology/oncology fellowship at Boston Children’s Hospital/Dana Farber Cancer Institute. During his postdoctoral research at Caltech with Nobel laureate David Baltimore, Porteus began his journey in gene targeting, in particular homologous recombination, as a means to repair disease genes. He has been on the Stanford faculty since 2010, treating patients with sickle cell disease and other hematological disorders at the Lucille Packard Children’s Hospital.
Porteus is a scientific co-founder of CRISPR Therapeutics, the company that launched the exa-cel CRISPR trial that culminated in the approval of Casgevy in December 2023. With the latest companies he has co-founded—first Graphite Bio, now Kamau Therapeutics— Porteus remains steadfast in translating the promise of homologous recombination to the benefit of patients with sickle cell disease.
In this wide-ranging interview with GEN editorial director Kevin Davies, PhD, Porteus candidly discusses some of the scientific milestones along his 25-year genome editing journey, including an update on the clinical translation of homologous recombination to treat patients with sickle cell disease.
(This interview has been edited for length and clarity. A longer version has been published in the June 2024 issue of The CRISPR Journal.)
Kevin Davies: Matt, what’s your impression of the current state of gene therapy and gene editing?
Matthew Porteus: The approvals in December 2023 of Casgevy [Vertex Pharmaceuticals/CRISPR Therapeutics] and Lyfgenia [Bluebird Bio] are super exciting to see. Within a decade, a discovery made in a test tube translated to an approved therapy for a disease that has a large unmet medical need. I’m proud to have made my contributions to that process.
That said, this is a first step but we’re nowhere close to where we need to go on multiple axes. From an efficacy standpoint, I’d argue that it’s not as good as a bone marrow transplant. If I could do an allogeneic bone marrow transplant as easily as doing an autologous transplant, the results are still better. Then there’s the issue of cost and accessibility. Certainly, an autologous transplant (using the patient’s own cells) is going to be more accessible because it’s a much less complicated procedure.
The manufacturing, of course, is more complicated. If you had a patient and they had a perfect allogeneic transplant match, that’s the direction you would probably go. If you have a matched sibling donor—a perfect HLA match donor without sickle cell disease or even with sickle cell trait—the trials exclude you from getting a gene therapy or gene-edited product because the results there are so spectacular. It’s when you don’t have that donor that you look for alternative sources. That is 85% or more of patients who don’t have that gold-standard matched sibling donor.
Davies: You have a long history in the genome editing space, going back to the early 2000s. Can you summarize some of those early landmark studies?
Porteus: I started thinking about this during my MD/PhD training. During my PhD work, I would argue the work that led to the first Nobel Prize in genome editing was performed. Mario Capecchi and Oliver Smithies showed that you could get targeted integration at measurable frequencies in mouse embryonic stem cells. Maybe we should start calling that the first Nobel in gene editing—they called it ‘gene targeting’ but the reality is that’s what a lot of us are doing.
The second thing was taking care of patients with sickle cell disease during medical school and recognizing the gulf between the contributions this disease had made to our understanding. of almost everything—blood development, genomic regulation, genomic structure, molecular pathophysiology—and that we did nothing that was driven by that understanding. We had this tremendous genetic and pathophysiological understanding— but it wasn’t being translated.
At the beginning of my career, I thought I could go into lentiviral gene therapy. But I wanted to develop homologous recombination—the Smithies-Capecchi approach to this disease—because the basis is a single nucleotide change in a single gene. We know the cell that you need to fix because allogeneic transplant shows us—as Guido Lucarelli called it ‘allogeneic gene therapy.’ John Tisdale (NIH) called it ‘whole genome therapy.’ We knew fundamentally what needed to be done.
Luckily, I found a postdoctoral mentor—David Baltimore—who was willing to say go at it. He admits he treats his trainees with ‘benign neglect’ in his words! But that was what I needed. Maria Jasin and others had shown that if you make a DNA break, the homologous recombination pathway is stimulated.1 I was part of a group showing how to take an engineered nuclease at the time—zinc-finger nuclease (ZFN)—and create a genome-specific break.2 Other than Sangamo, no one can really make good ZFNs. There was a real barrier because it was so difficult to engineer them. Many of us spent a decade doing the best we could with them.
Casgevy is not based on gene correction or homologous recombination. And the recognition that DNA breaks can lead to site-specific indels (insertions/deletions), we need to really give Dana Carroll credit. He published a paper using ZFNs showing that you could make mutations in specific genes.3 And since that was a relatively simpler process, it really has exploded. I thought TALENs were going to be really transformative but just as they were getting off the ground, we have the 2012–2013 discoveries of CRISPR.
Davies: You were a co-author on the famous Nature paper in 2005 that introduced us to the term ‘genome editing.’ How did your collaboration with Sangamo come about?
Porteus: I was talking to Srinivasan Chandrasegaran, who biochemically had said ‘I can link a ZFN to the Fok I-chimeric nuclease domain.’ He had collaborated with Carroll. We needed to engineer these systems into mammalian cells and had shown how to use nucleases in a mammalian cell to get high frequencies of gene targeting. I could fix the GFP gene. And I was finishing up my postdoctoral work with David who said, ‘you need to talk to Sangamo…’ I remember visiting Casey Case, the CSO of Sangamo at the time…That prompted a collaboration as I was transitioning to my independent faculty position and their enterprise on engineering ZF proteins and then coupling them to the nuclease domain, which led to the Nature paper.4 It was the first demonstration that you could engineer a nuclease to target an endogenous sequence in a genome. In all the prior experiments, you inserted your target site first, and then retargeted it.
Davies: Not long after the CRISPR gene editing papers in 2012–13, you became a scientific co-founder of CRISPR Therapeutics. How did you get involved with them?
Porteus: I make the distinction that gene editing is a process and CRISPR is a tool to start the process. But any protein that makes a DNA break initiates the gene editing process. CRISPR just happened to be this amazingly powerful, easy-to-use specific tool. There was a clear excitement—it caught the attention of investors who wanted to start biotech companies. Investors were looking to people who knew about gene editing and some came knocking on my door. I started talking to some of the VCs and through those conversations became a scientific founder of CRISPR Therapeutics, got to know [co-founders] Sean Foy, Rodger Novak, and Emmanuelle Charpentier. They clearly had a sense that they had a hold of something. One of the things I think I can take credit for is pointing them in the direction that sickle cell disease and β-thalassemia were the perfect first indication.
Sickle cell disease needed and deserved this attention. It was the right [patient] size, it had a large unmet medical need, the biology was well understood, and there were a lot of approaches that one could take. My lab has continued to work on directly correcting the mutation that causes the disease by homologous recombination because that’s where I got started.
But the biology of understanding that fetal hemoglobin (HbF) could counteract the effects of sickle hemoglobin and that there were pathways that controlled the levels of HbF, and that there were people in the world who had genetic changes causing hereditary persistence of fetal hemoglobin, pointed people to the idea that perhaps a faster approach was homologous recombination—to harness the indel-based approach that Dana Carroll had [published]. I helped CRISPR Therapeutics work through potential targets with CRISPR-Cas9.
They ended up building on the biology that came from Daniel Bauer and Stuart Orkin on BCL11A.5 The first approaches in fact were by Sangamo to target BC11A itself. We quickly learned that if you inactivate the protein, you disrupt stem cell function. When Dan identified this key enhancer—we helped them a bit with some TALENs [to show] that you could inactivate BC11A being turned on in the red cell lineage without inactivating all of its key functions in hematopoietic stem cells and B cells.
They built on that target, but if you look at some of the subsequent programs that are targeting HbF, they’re targeting other pathways in BCL11A, the binding sites. They’re making changes that match what has been found in people. We often go back to the Orah Platt [Boston Children’s Hospital] paper showing that 8–9 is the level of HbF that starts to differentiate between less severe disease.6 And of course, hydroxyurea is a small molecule showing that if you upregulated HbF, you decrease the severity of the disease…
The second part that I contributed to was that CRISPR-Cas, as originally discovered, simply didn’t work in hematopoietic stem cells (HSCs). Delivering the system as a plasmid, as DNA expression molecules, simply didn’t work. We at Stanford, in collaboration with Agilent, showed that you could deliver it as a Cas9 protein complexed to a synthetic guide molecule, with the guide molecule having modifications at the end. This probably serves a couple of purposes, one of which is to protect it from degradation, the other maybe is to shelter it from being recognized as a foreign molecule. I suspect that was a highly effective way of getting genome editing to occur in a range of primary cell types and keeping those cells healthy. You could deliver plasmids to CD34 cells and they would just die like a dog because cytoplasmic DNA is sensed as a viral infection and you get this robust type 1 interferon response. This was a way to deliver that nuclease in a highly efficient fashion to get that break where you wanted and maintain the quality of the cells. It’s that approach using the CRISPR-RNP complex that’s being used to manufacture Casgevy.
CRISPR Therapeutics-Vertex found a good guide RNA, you have your target, you treat a patient, you see Victoria Gray doing fantastically well.7 The pace of their trial enrollment was appropriately slow, then you see this huge acceleration and the number of patients they enrolled. It is super exciting to see the results in Victoria basically generalize [to others] and the approval in December 2023.
Davies: Let’s talk about your former company, Graphite Bio. What was the origin of the name?
Porteus: We went through several names as most companies do, but we picked the name because the very first structure that Rosalind Franklin published was of graphite! So it was a call out to her contributions to our understanding of the structure of DNA that leads to sickle cell disease as a genetic disease.
Davies: Do you still believe that direct correction of the pathogenic mutation will work better than upregulating HbF?
Porteus: I certainly think it will be, it’s an approach that still is going to be the gold standard. Casgevy is still not a treatment of the root cause. This still has not been done for any disease—to take the pathologic variant and change it to the non-pathologic variant. That has not been done yet. [The HbF approach] is great but it’s a workaround, making mutations to compensate for the pathologic variant.…
Even as we were supporting and advising CRISPR Therapeutics on their program, in my academic lab, we were building the technology to get high frequencies of gene correction in the endogenous beta-globin gene. We developed it with CIRM support from Stanford, generous philanthropic support from some Bay Area supporters, and got to the place where we could submit an IND and then raise capital to run the clinical trial. We made the decision to start Graphite Bio to run that clinical trial, and then maybe to more broadly develop the HDR [homology directed repair] platform. The company got started when times were good in the biotech investing world. Raising capital is always challenging but they were able to raise capital.
Davies: The story of Graphite didn’t end the way you had planned it. There was an adverse event after you dosed your first patient?
Porteus: The patient received her own cells in August 2022. When you get chemotherapy following a bone marrow transplant, there’s always a period where you wait for the cells to engraft and start making new red blood cells and platelets. Her period lasted longer than we expected but she was discharged from the hospital, she was doing fine, she felt great. But the treating team decided to stimulate her bone marrow by putting her on a growth factor for stem cells—a thrombopoietin agonist—which is being used in other bone marrow failure settings. But it wasn’t written into the protocol. And because we started an off-protocol medicine for a finding that was related to the drug product—slow engraftment— that triggered the SAE [serious adverse event] reporting. She was home! There were some [erroneous] reports, one even reported that she had died. She called her treating physician and said, ‘I’m not dead!’ Nonetheless, it was clear that the process that had come out of Stanford needed improvement….
Davies: So Graphite was at a crossroads. One could imagine a scenario where the company regrouped and decided to carry on—but you didn’t do that?
Porteus: The company had appropriately decided to pause it on its own. After that [the patient’s] bone marrow started to recover. It was reported at ASH 2023 [American Society of Hematology] that she’s now transfusion independent, she’s off all her growth factors and clinically she’s doing great. There are some laboratory findings that we’ll continue to study and report in the future.
It was clear it was going to take another 12–18 months to take the improvements had been identified both academically and within Graphite to treat the next patient. It was clear that the manufacturing process was too hard on the cells, but it was going to take another 12–18 months to get the manufacturing down and reboot the clinical trial. Graphite, the board and the investors thought that it was not in the best interest for that company to continue the program.
So it came back to me because that makes sense. I got a lot of advice, thought long and hard. They were willing to do whatever I wanted to do. Some people said, ‘You should take it back to your Stanford lab and reboot it there.’ But I don’t think that works because we were ready to go. So instead, I started another company called Kamau Therapeutics….
Basically, the entire genome editing program has been transferred from Graphite into Kamau. We’re in the process of closing our seed funding. The trial was never closed, so we’re just reopening it under different names, the same general sites, with this markedly improved manufacturing process, some tweaks to the clinical protocol around cell dose and utilization of growth factors. We hope to treat the next patient with Nula-cel in 2025.
Davies: We now have two approved therapies, priced at $2.2 and $3.1 million. Do you think those are fair prices?
Porteus: One part of me agrees with what Julia Kanter (University of Alabama) has publicly said, which is that these patients and these therapies deserve that price: the benefit to patients absolutely justifies the lifelong savings and the indirect costs. [These prices] are not out of line for the prices for similar therapies. Lenmeldy, which just got approved for metachromatic leukodystrophy, got listed at $4.25 million! Why should sickle cell drugs be priced any less than any of the other peer drugs?
That said, the price is clearly going to be a barrier to patients getting access. Where is the price coming from? Well, it’s coming from companies that are driven by shareholder value that need to make a return, and they develop the drugs and that’s the system we work in. It’s also driven by the fact that these autologous ex vivo manufactured therapies are expensive to make. Once you have a small molecule drug, you can make kilograms. If it’s an antibody and they do a 25,000-liter bioreactor run, one lot gives you doses for thousands, if not 10,000 patients. Every [gene editing] dose has to be manufactured specifically. So right now, there’s just a cost of goods that’s high.
Some people said that’s why we need to figure out how to give an in vivo gene editing drug, but I’d argue right now the in vivo gene therapies are just as expensive and just as inaccessible as ex vivo therapies. So yes, it could get you there, but AAV gene therapies are also $3 million. There’s not a big difference. In terms of treating patients in Africa and India where most patients are, ex vivo therapy is pretty much [considered] a non-starter. But I don’t believe that.
I think the other challenge with in vivo is the HSC is a pretty special cell. So in vivo editing and the Intellia programs are super exciting and the Verve Therapeutics programs are exciting in some senses. But delivery is challenging, even in the liver. The liver is designed to take up things. Macrophages are designed to take up things. HSCs are not designed to take up anything! They’re designed to be sheltered and protected. I think we’ve got a lot of biologic work to be done to find the right way to deliver an editor to that cell.
In contrast, I think ex vivo autologous engineering is now, to me, an engineering problem. How do you stick process engineers on this and make the cost of goods cheaper, make the process cheaper and scalable? Instead of having five people make one product, can you have one person make ten products? I think engineering is something we are really good at and there are solutions out there and solutions being developed. The story of monoclonal antibodies is the analogy I turn to: once considered impossible to be a scalable drug and now a platform that is considered standard. I hope to be part of the story where autologous genetically engineered cell therapies go through a similar developmental process.
I think we need competition. I’m going to put my effort in making ex vivo therapy as cheap as possible. Let’s set a bar that in vivo has to get to.
Matthew Porteus, MD, PhD, is a physician-scientist at Stanford University School of Medicine, Palo Alto, CA. Kevin Davies, PhD, is Editorial Director of GEN.
A longer version of this interview has been published in the June 2024 issue of The CRISPR Journal.
References
- Urnov F. Genome Editing B.C. (Before CRISPR): Lasting lessons from the “Old Testament. Crispr J 2018;1(1):34–46.
- Porteus MH, Baltimore D. Chimeric nucleases stimulate gene targeting in human cells. Science 2003;300(5620):763.
- Carroll D. Genome engineering with zinc-finger nucleases. Genetics 2011;188(4):773–782.
- Urnov FD, Miller JC, Lee Y-L, et al. Highly efficient endogenous human gene correction using designed zinc-finger nucleases. Nature 2005;435(7042):646–651.
- Orkin SH, Bauer DE. Emerging genetic therapy for sickle cell disease. Annu Rev Med 2019;70:257–271.
- Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease—Life expectancy and risk factors for early death. New Engl J Med 1994;330:1639–1644.
- Gray V, Thomas U, Davies K. Warrior spirit: An interview with Victoria Gray, sickle cell pioneer. Crispr J 2024;7(1):5–11.
