Paula Hammond: So that was an excellent session. And we're about to have an even more exciting session in the sense that we're going to hear from our young scientists. And I am very honored to introduce a group of incredible young colleagues. The first is Angela Koehler, an Associate Director of the Koch Institute, an Associate Professor of Biological Engineering at MIT. Professor Koehler earned her bachelor's degree in biochemistry and molecular biology from Reed College in 1997. In 2003, she finished her doctorate in chemistry from Harvard University, and became an Institute fellow at the Broad Institute, where she served as the Director of Transcriptional Chemical Biology in the Chemical Biology Program. She was also a project leader in the NCI Cancer Target Discovery and Development Center at the Broad Institute, aimed at targeting casual cancer genes with small molecules. Angela then joined the Koch Institute in 2014. We're very fortunate to have her. And here, she serves as faculty director of the Swanson Biotechnology Center High-Throughput Screening Facility. She's also the Associate Director of the Koch Institute. She is also a co-director of the MIT Biomedical Engineering Undergraduate Program, and a member of the pre-health advising at MIT. She has served on the Chemists and Cancer Research Executive Advisory Board for AACR. And her awards include being named a Genome Technology Young Investigator and a Broad Institute Merkin Fellow. She also received the Ono Pharma Foundation Breakthrough Science Award, the Novartis Lectureship in Chemistry, the AACR Bayer Innovation and Discovery Award, and NSF Career Award, and the Junior Bose Award for Excellence in Teaching. Angela has founded three biotechnology companies, Ligon Discovery, Kronos Bio, and 76Bio. She serves on the scientific advisory boards of Kronos, MS2 Array, Nested Therapeutics, and Photys Therapeutics. She also advises Enveda Biosciences and ORIC Pharmaceuticals. So now I'll turn things over to Angela. [APPLAUSE] Angela Koehler: So as Paula mentioned, I've been here for the better part of the decade that we are celebrating today. And so in some sense, I'm really honored to be in the future science part of the lineup, because I guess it's a vote of confidence that I can still do cool things now in the next decade. Let's see. Oops. So as the old person in this session, I'm actually going to talk about an old problem, and that is trying to come up with strategies to modulate the prolific oncoprotein MYC. So here are my disclosures. Paula mentioned some of them. I've highlighted in green some of the companies that are focused on, well, they've publicly declared MYC as an area of interest to them. Although I think I don't violate any non-disclosure agreements by saying that some of the companies in black are also interested in MYC. I'll talk a little bit about Kronos, but I'm going to address the elephant in the room. 76Bio is a company that I co-founded with our Koch Institute colleague Dane Wittrup. The name really connects to the amino acid length of ubiquitin. But I think in this room, we can all pretend that this company is really an homage to our wonderful building. So if you're not familiar with my lab, I'm your friendly chemical biologist in the building. We spend a lot of time trying to think about how to expand the repertoire of so-called un-druggable targets, these proteins that have features like low complexity domains that make them very challenging from a drug discovery perspective. We develop different technologies to provide inroads to those historically challenging targets. And our goal is to find pinches of magic dust that we can use to clarify the relevance of those targets, and to convince our colleagues in industry that they may want to go after them as well. More recently, a way that I've remodeled since being at the Koch, is that I've actually started to think a bit more about nanomedicine, and how to collaborate with our colleagues in the building who care about drug delivery. We've connected and tried to basically use some of the same chemical biology or systems biology skill sets that we think about for small molecules, and try to understand cool mechanisms of action for nano care. So one example here, connecting with Amanda Facklam, who's a graduate student who has worked with Dan Anderson to think about how special chemical modifications of alginates give rise to certain phenotypes. And then we'll hear later in this session from Joelle Straehla about a collaboration with the Hammond lab and the Broad Institute. So I'll talk quite a bit about this prolific oncoprotein Myc. Many people have been thinking about how to drug Myc for decades, folks from all sectors. In part, because Myc deregulation really impacts all of the hallmarks of cancer. Basically uncontrolled proliferation, tumor invasion. I won't go through everything on this slide. But given the profound phenotypic events or consequences of Myc deregulation, and the suspicion that 70% of human tumors have Myc deregulation, I think it's fair to say that this is a target of keen interest to many folks. If you're thinking about drugging Myc, it's sort of important to recognize that Myc itself is a master regulator that governs the transcription of a whole host of genes that are involved in housekeeping functions. Again, thinking about growth, metabolism, et cetera. And then, Myc can become upregulated in certain settings, and in response to injury, for example. And then in cancer, you really have this sort of super physiologic level of Myc, often because of amplification, that really drives the upregulation of tumor specific oncoproteins. And so what we really want to do is think about attenuating this amplified Myc-driven transcription to these somewhat so-called normal activities. Unfortunately, this has been hard. Myc is considered a graveyard from a drug discovery perspective. It's challenging, in part, because Myc functions primarily through biomolecular interactions, protein-protein interactions, protein-DNA interactions. And this suite of interactions can remodel in different cells states and different cell types. For the last couple of decades, most folks have thought about how you could disrupt, typically using small molecules and peptides, this really important interaction of Myc with its obligate heterodimer partner Max. They come together and coordinate transcription. But unfortunately, Myc is intrinsically disordered. It's essentially a piece of spaghetti that will roll up into an ordered structure, like a piece of rotini, when it finds its heterodimer partner Max. And most of the small molecules that have been found to date lack specificity. They don't-- you find something in a test tube that doesn't really translate in terms of potency into a cellular model. Before coming to the Koch Institute, I spent a lot of time trying to find small molecules that would disrupt this interaction, and we found plenty of them. I won't bore you with all the details, but none of them were progressable. So we had to kind of think about another strategy. And I will tell you two vignettes today about how our lab has thought about Myc at the Koch Institute. And so this is the first vignette about a chemical probe that's directed to the obligate heterodimer partner Max. So our rationale here was to use high throughput screening to go after the Max protein. Max is a central node in the Myc Max Mxd family. It turns out that as a smaller protein, this is a little bit easier to work with in a variety of biophysical assays. And it's just really a much easier protein to handle. But all of you should sort of see Max as sort of a central node in this network. We just wanted to find chemical tools that would enable us to understand the consequences of perturbing Max. So what we did was we applied purified Max protein to our small molecule microarrays, these little chips with tens of thousands of drug-like small molecules. We found a bunch of putative binders. We promoted those to a Myc-driven reporter gene assay that was developed in our colleague Ben Ebert's lab across the river. We found about six molecules that modulate Myc-driven transcription at reasonable potencies, and then assessed whether some of those compounds disrupted the Myc-Max interaction. We really focused on a molecule that appeared to be stabilizing Max-Max homodimers in a dose-dependent manner. And we evaluated that compound and the other compounds in a broad spectrum of different viability assays, including a special panel of cell lines that you'll hear more about at the Broad Institute. And we partnered with colleagues at the Stanford School of Medicine to evaluate these compounds in cellular models with Myc dependency. We evaluated this compound from a global transcriptional perspective using RNA seq, and noticed that the chemical perturbation matched very well with removing, basically removing Myc from a system. And then we used a whole host of techniques that we routinely use in the lab to basically confirm that the molecule could confirm, or could engage Max in a cellular setting. So our favorite molecule in the end is this one shown here, KI-MS2-008, which had a potency of around 1 micromolar. So the special thing about KI-MS2-008 is that while it binds to Max, we noticed that at different doses and at different times, we observed a decrease in Myc protein levels, and in mg 132 rescue experiments that suggested that this was a proteasome dependent process. And in fact, global proteome-wide measurements showed that this was a somewhat selective process. Myc is certainly degraded or perturbed greater than any of the other genes, which was interesting. We thought this molecule originally might impact Myc- driven functions by shifting the dynamic equilibrium of heterodimer and Max-Max homodimers, which really bind to the same e-box sites and repress transcription. So a competitive mechanism. But now, we think that by shifting this dynamic equilibrium, you do something to Myc that renders it very unstable, giving rise to a degradation phenotype. So we've spent time collaborating with, again, colleagues like Ben Ebert, colleagues at the Broad Institute, to try to understand how this is happening. So I don't have time to say much about this molecule, other than we've spent a lot of time trying to map its structural binding to Max homodimers. We've continued to work with our colleagues at Stanford School of Medicine to evaluate this compound in vivo. We took a step back and made a bunch of analogs. We've scaffold hopped. We've tried to make better molecules. And we've also collaborated with the Hammond Lab to coformulate this compound with other compounds that we think could be of interest for modulating Myc-driven transcription. And then where I really get excited, again, is trying to understand the systems biology associated with the Myc degradation phenotype. And in fact, we've nominated a whole host of targets that are involved in this process that our lab has already identified new molecules for. So in the last couple of minutes, I'll pivot to our other strategy to think about Myc. And this is actually an important story, because it involves an important relationship that the Koch Institute had with our colleagues at Janssen Pharmaceuticals as part of the Transcend Project, which I had the good fortune to enter into starting out here at the Koch. And I'll also make reference to a really cool Bridge project involving Steve Balk. So the Bridge-- or sorry, the Transcend project wasn't inspired by Myc. It was inspired by a splice variants of the androgen receptor in prostate cancer. And our colleagues at Janssen were very much interested in trying to identify small molecules that would bind to this region of the androgen receptor. And so what we did was run our small molecule microarray screen again. And this time, we screened in cell lysates, not with purified androgen receptor, but cell lysates that expressed an engineered version of splice variants. We worked with our colleagues at Janssen to promote hits from that assay into downstream PSA expression assays and reporter assays that they had developed internally. And this was about the time that the Transcend project ended. But we found three really cool molecules that we wanted to study further. And so we applied for a Bridge project with our colleague across the river, Steve Balk, and continue to advance the compound. One notable piece of data was that they use this VCaP 16 enzalutamide resistant cellular model, and found that this compound shown here-- oops-- KI-ARv-03 really led to rapid and robust degradation of ARV7 in ARVs. But by my eye, this was not an interesting molecule, because it looked like a kinase inhibitor. And my guess was it was actually a non-selective kinase inhibitor. But the post-docs in our lab were really interested in continuing to push this compound forward. They did the right thing, and that was a kind of profiling assay. And lo and behold, very much to our surprise, this was actually a selective CDK9 inhibitor. And moreover, we could actually see that this CDK9 inhibitor impacted phosphorylation status of relevant targets, including Pol 2. So how the heck did we find a CDK9 inhibitor in an assay focused on androgen receptor splice variants? It turns out that there's an established binding relationship between CDK9 and androgen receptor, as well as the splice variants. And so it's tempting for us to think that in our original assay, these proteins had come together. But we think that the dominant mechanism of action of the molecule is actually transcriptional, because Myc regulates the expression of androgen receptor, and CDK9 comes together with the P-TEFb complex and Myc to regulate that expression. And we have a whole host of data to demonstrate that this is the case. So at that point, we did a little bit of chemistry in our lab. But we punched KI-ARv-03 into a company called Kronos Bio. Within a couple of months, they developed a better candidate called KB-0742 that retains selectivity with enhanced potency. And continuing on initially with the theme of prostate cancer, they evaluated this compound in a xenograft model, and saw reasonable potency. But we went back to Myc. As a team, we started to think again about this opportunity. CDK9 regulates the expression of Myc itself, and coordinates with Myc downstream to regulate a whole host of other transcriptional programs. And there's a hypothesis that intermittent or partial modulation of Myc may be sufficient to disrupt the whole network. So I don't have time to go through a lot of the preclinical data. But one key partnership, again, was with the Broad Institute and the Prism Platform, this sort of panel of barcoded cell lines, hundreds of barcoded cell lines. They were able to notice that this compound-- that the cancer cell lines that had Myc amplification, genomic copy number amplification were more sensitive to this molecule relative to other cell lines. And that gave rise to a lot of additional preclinical data. This molecule, however, is now in a phase 1, 2 clinical trial. And the phase 1, it's really just sort of standard safety experiments. But when you move to stage 2, the focus is actually on Myc amplified tumors, based on some of that original data. And mostly, I should say, focusing on solid tumors. I'm running out of time. But one thing I wanted to say here was that in November, Kronos announced their interim phase 1 report, and so far so good. No serious adverse events. They've demonstrated target engagement, excellent PK. What I wasn't prepared for was what would come next for me as a non-physician sitting here in my office, was a whole host of calls from parents who just saw in a press release Myc amplification. And so over the last six months, I've actually spent a lot of time talking to those parents, thinking about my own children, quite frankly. And I think it's really changed how I think about my role here at the Koch Institute over the next 10 years. While I really geek out about molecular-level details, I think now I would like to be more translational, and specifically try to think about pediatric tumors moving forward. So I will end there, and thank the amazing trainees in our lab who've contributed to this project, and to thank our collaborators on both projects. And then, obviously, to thank the funding sources that we've had. [APPLAUSE]