October 10, 2012
CDD Spotlight Highlights
"So I actually started out life as a physicist. My undergraduate degree was physics and during that time my undergraduate tutor was Professor David Blow, who was at Imperial College where I did my degree. He was one of the pioneers of x-ray crystallography, sadly passed away a few years ago, but I would say that David was certainly instrumental in fostering my interest in biology and as I went through my undergraduate degree, I realized that this was a field that was fascinating to me, that I really wanted to get into. So I guess that would be the first turning point was in making the decision to mutate from a physicist into a biologist."
David Matthews, Ph.D. joined Pathway Therapeutics as Vice President of Drug Discovery and Exploratory Development in 2010. Previously, Dr. Matthews was Executive Director of oncology discovery at Exelixis Inc. In this role, his responsibilities included oversight of the preclinical drug discovery portfolio and facilitating the transition of projects from discovery into preclinical development and early clinical studies, including the PI3-kinase inhibitors XL147 and XL765. During his time at Exelixis, Dr. Matthews also managed the structural biology group and led several multidisciplinary drug discovery and development teams. Prior to joining Exelixis in 1999, Dr. Matthews led the molecular and structural biology group at MetaXen LLC and was previously a scientist at Arris Pharmaceutical Corporation, where he led protein expression, protein engineering and biochemical/cellular assay initiatives for multiple projects. Dr. Matthews obtained his Ph.D. at the University of London (Imperial College) in collaboration with Celltech Plc, and was a postdoctoral fellow at Genentech Inc. He is the co-author of Targeting Protein Kinases for Cancer Therapy (John Wiley & Sons, 2010).
Interviewed by Barry Bunin, CEO Collaborative Drug Discovery, Inc.
Barry Bunin
Hey, Dave Matthews, this is Barry Bunin calling from CDD and recording for the CDD Spotlight.
David Matthews
Hey, Barry, how are you?
Barry Bunin
Good. I thought I would start out with a general question just to introduce yourself and talk a little bit about the research that you've done.
David Matthews
Sure. So I've worked in the Bay Area biotech community for somewhat over 20 years now, started out as a postdoctoral researcher at Genentech and from there I moved to my first industry post-Postdoc position, which was at Arris Pharmaceutical, which you'll recall is where you and I overlapped for some period of time. After that, I moved to a company called MetaXen, which was focused on drug lead optimization, and worked there for 18 months to two years before that company was acquired by Exelixis. So I moved to Exelixis I think it was 1999 and I was there for about 11 years, initially working on getting the drug discovery infrastructure set up. At the time, Exelixis was a model organism genetics company, so I arrived at the onset of drug discovery. For the next two years I ran a structural biology group there and we built that up to an organization that supported many drug lead optimization projects across the company. And then latterly, I was involved in coordinating the oncology discovery efforts, so essentially managing the different project teams that worked on discovery of oncology drugs as well as playing a role in early development. So from there I moved to Pathway Therapeutics, which is my current position, about two and a half years ago now and it was kind of an interesting fit for me at the time.
Pathway is a small company focused on PI3-kinase inhibitors, and a very different environment from Exelixis in that we're a small, mostly virtual company, although we do have some level of in-house lab work, so it's been a fascinating experience for the last two years. We've moved one molecule into the clinic. Our lead program, is a PI3-kinase/mTOR inhibitor that started phase one last year, and more recently we've been working on a program focused on PI3-kinase delta, which actually is the reason that we contacted you guys and got the database in-house and that's also moved along very nicely such that we're at development candidate stage with that program now.
So I hope that gives you a little bit of background about what I do and what I've been doing particularly at Pathway in the last couple of years.
Barry Bunin
It's a very hot area. I know there are entire focused conferences dedicated at the PI3-kinase field. So in general, what's been some of the excitement or let downs as well as the emerging general observations about the field and can you also talk about how your own work fits within that or not? In general, why is PI3-kinase so interesting at this time in history, if you will?
David Matthews
Sure. So I think the PI3-kinase pathway has really come into prominence over the last ten years, particularly initially in the field of oncology; PI3-kinase is a lipid kinase and is a key signaling molecule in cells which coordinates cellular growth, proliferation, and survival. In cancer cells, this pathway frequently becomes dysregulated and is probably one of the more dysregulated pathways in cancer, so that prompted a lot of the interest in both the academic and the pharma community over the last ten years and has led to a lot of molecules moving into the clinic, particularly in oncology.
One of the things that we've learned that has both plusses and minuses over the last five years is that the efficacy of many of these drugs was maybe a little disappointing, not the sort of magic bullet that a lot of people had hoped for, and I think there may be a number of reasons for that, perhaps just that we weren't hitting the right target hard enough and it's possible that some of the drugs that moved into the clinic were limited in the ability to knock down the pathway by either off target effects or other toxicities such that you just weren't able to have a sufficiently robust inhibition of PI3-kinase in tumors. So one of the trends over the last few years has been to more selectively target specific isoforms and in particular PI3-kinase alpha is the isoform that's most involved in solid tumor growth and proliferation. There's an interesting story emerging around the beta isoform recently as well, so I think that's one of the trends that we've seen.
I mentioned PI3-kinase delta. That's a little bit different in that it's clinically validated now as a target in hematologic oncology but also has tremendous potential, we believe, as a target in various inflammatory and respiratory diseases, so that's where a lot of our attention has focused over the last year is in really developing selective and potent PI3-kinase delta inhibitors and demonstrating their utility, particularly in inflammatory and respiratory disease.
Barry Bunin
So maybe you can talk a little bit about how you use CDD and why.
David Matthews
Sure. We started our program to discover PI3-kinase delta inhibitors and initially it was a small exploratory chemistry effort. We were able to get to by with essentially just recording things in a spreadsheet, but as you'll appreciate there comes a point where if you have more than a few dozen compounds and more than one or two dimensions of data it becomes extremely difficult to interpret data. So we looked around for a solution. We are a very small capital efficient company, so didn't have a huge amount of money to spend and CDD seemed like the right solution for us in terms of allowing us the flexibility to incorporate both structural information as well as a wide variety of multidimensional compound assay information and we've made a lot of use of the flexibility that the database offers in terms of being able to record something as simple as a single binding assay number and something as complicated as pharmacokinetic profile and we also use the database’s curve-fitting routines, so we run a number of assays in-house, archive the primary data in the database and that's really sped up the workflow tremendously. It saves a lot of time for the guys working in the lab who then don't have to use some sort of ancillary programs to analyze the data and type it all into the database. They can just do it all on the fly, so it's worked out really well for us so far.
Barry Bunin
When giving individuals feedback, I like to give them positive feedback to feel good and critical feedback to get better. Maybe we could do the same on CDD, so what are the top two to three things you like about CDD, but also the top two or three things you wish for could be better in CDD, and it's okay do both the positive and the critical which is interesting for learning, even though we're recording.
David Matthews
Yeah, so… I think the most important thing for us is the flexibility. I think that the ability to be able to essentially make up our own assay protocols to record as many different readouts per assay as possible, that's been very impactful. And I think that the querying system, the ability to generate reports has been one of the features that we've really used a lot of. I know that's something that's evolved even over the last year. Certainly there have been some improvements even during the time that we've used the database, but that's been extremely useful for us in terms of generating reports both internally and for sharing with third parties outside the company. It's just been a real time saver for us.
As far as areas for improvement, I think there are always a few little quirks which if I were designing the database then maybe I'd design it a bit differently, but I think that has to be set against the fact that you're not just designing this database for us but it's being used by a wide variety of different people. I'm trying to think of a specific example where… I think it'll be more sort of small little technical issues like generating data in a format that we can very easily import into another spreadsheet or another format. Sometimes it requires a couple of extra steps over what I’d ideally like, but in general I think things worked out really well.
Barry Bunin
Before you started in the industry when you got your training, what was sort of the formative role model or person or persons who sort of got you set in deciding to become a biologist and how did that help you get to obviously where you got a post doc at Genentech and worked (in roles entrusted with responsibility) at all these biotech…
David Matthews
So I actually started out life as a physicist. My undergraduate degree was physics and during that time my undergraduate tutor was Professor David Blow, who was at Imperial College where I did my degree. He was one of the pioneers of x-ray crystallography, sadly passed away a few years ago, but I would say that David was certainly instrumental in fostering my interest in biology and as I went through my undergraduate degree, I realized that this was a field that was fascinating to me, that I really wanted to get into. So I guess that would be the first turning point was in making the decision to mutate from a physicist into a biologist.
Then in my postgraduate work, my PhD program, I actually had an early exposure to industry, so I worked in conjunction with a company in Britain called Celltech, which was one of the early biotech companies in Britain. I worked for a guy there called Spencer Emtage, who again was a very accomplished molecular biology and it's really through him that I got my initial exposure to molecular biology and, well, really biology in general.
And then as far as my postdoctoral training, I moved to Genentech, as I said. I worked with Jim Wells in the protein engineering group there and that again was a fantastic experience in terms of working at a company but in what was at the time a very sort of academic environment. I got a chance to work on a number of fascinating projects there and not just the projects that I was working on but just the exposure to the broader environment there at Genentech in what I guess now are the relatively early years. That was a very formative experience for me as well so…
So there we go, there are three people that have definitely contributed to my career development who I think very highly of.
Barry Bunin
Great. That's three people and you'd mentioned Imperial College in Britain, so I'm going to ask a non-science fun question. I was recently watching the Olympics in London and from a creativity perspective noticing how it was done where the floor on the stage would have lines from Shakespeare, like: “To be or not to be” or the songs would be from The Beatles or the skits from Monty Python and so there's quite a bit of celebration of creativity and culture at the Olympics and I just was wondering for someone who grew up there what your perspective was on how London did the Olympics and just any thoughts with that you mentioned Imperial College.
David Matthews
Sure. So I spent six years in London doing my undergraduate and postgraduate work and the city I'm sure has changed a lot since I was there, but it's always fun to go back. In fact, I'll be back in Britain again towards the end of this year. I thought the Olympics was a lot of fun. It's always something I look forward to every four years and, as you can imagine, looked forward with particular attention since I grew up in Britain. So, yeah, that was a very interesting experience both to sort of look at it as a “used-to-be” insider as well as - I guess I'm now pretty much a full-fledged American. But, yeah, a fun experience and look forward to Rio in four years time.
Barry Bunin
Going back to the science, you mentioned being focused on the weeds on the PI3-kinase project, and that makes sense. That's sort of always the case in industry, you have to be focused, but what have been some of the technology or research advances? The field of biology in some ways is interesting in terms of how fast it evolves relative to chemistry in some ways, so I was just kind of curious as to what you felt were some of the maybe more surprising or more useful new trends that have come up since you've started in the biology field that have really mattered for either basic research or drug discovery?
David Matthews
One area that's really matured and continues to do so is what's broadly referred to as translational medicine. This covers a multitude of things depending on who you talk to, but I think in general the technologies available for monitoring and assessing biomarkers, whether they be biomarkers that predict the onset of disease, response of disease to therapeutic agents, I think that these technologies have really made leaps and bounds over the last five or ten years and continue to do so. I am particularly interested in the techniques that seem to be generating ever increasing sensitivity for looking at mutational status for example in tumor cells, in circulating tumor cells, but as well as just looking at serum biomarkers from both a proteomics as well as a genetic point of view and increasingly I think the metabolic markers, that's another area that's certainly up and coming seems to be receiving a lot of attention lately. So in general, I think that's a field that has already had an impact and will continue to do so and of course has particular relevance to the emerging field of personalized medicine where in the oncology space we’re starting to see some of the first truly targeted therapies where you're not treating for example the specific organ-based tumor type, you're treating a genetic lesion-based tumor type and so I think that that's something that will increase over the next few years.
Barry Bunin
That's excellent note to end it on. It's inspiring to hear your perspective having worked at everything from a large Genentech to a variety of other companies, including some of which I've worked at. So thank you for time. Unless there's any other questions, I think that's a good note to end it on.
David Matthews
Okay, thanks very much. Nice to talk to you.
Barry Bunin
All right, thanks again for your time. Bye.
David Matthews
Okay, all the best. Bye.
This blog is authored by members of the CDD Vault community. CDD Vault is a hosted drug discovery informatics platform that securely manages both private and external biological and chemical data. It provides core functionality including chemical registration, structure activity relationship, chemical inventory, and electronic lab notebook capabilities!
CDD Vault: Drug Discovery Informatics your whole project team will embrace!