Dr. Weaver featured in Faculty Perspective

Dr. Weaver was interviewed in depth by the Bioengineering Department and featured in a Faculty Perspective in BioE News:

A conversation with Professor Valerie Weaver, UCSF Professor of Surgery, about work in her highly collaborative cancer research lab.

What is the main focus of your lab?

We do a lot of work on cancer, mostly breast cancer, pancreatic cancer, and glioblastoma. There is another part of my lab that does stem cell work, primarily HESC and early gastrulation in a chick model. We also have a small project collaborating with two different labs on the liver, studying liver fibrosis and hepatocyte progenitor expansion.

Our primary focus in the lab however, is on tumor evolution. We work across length scales to achieve our goals. For example we do supra resolution imaging that involves examining cellular phenomena intensely at the single molecule or subcellular level, and then we move our observations into tissue-like structures called organotypic models. Then finally we take our observations in vivo into mouse models, and finally into human tissue. Our objective is to understand how cells respond to changes in tissue mechanics, and in particular the biochemical and biophysical properties of the extracellular matrix. The types of questions that we ask are, if we compare a metastatic tumor cell to a tumor cell that is non-metastatic, can we detect subtle differences in their biophysical behavior? Do they assemble different types of integrin adhesions, do these adhesions exert higher forces, do they have different dynamics, or do they interact with growth factor receptors differently? If they are different how does that influence their metastatic dissemination?

What I believe makes my laboratory unique is our ability to cross length scales from basic science to the clinic, to employ an interdisciplinary approach to our work, and to apply many of the concepts and approaches used by physical scientists together with classic cancer cell biology. We also use as frequently as feasible primary cell models and clinical specimens we obtain from colleagues from across the country, as well as genetically engineered mouse models, and three-dimensional tissue-like model systems.

Using these various model systems gives us the opportunity to cross-length scales and also to compare how mechanics can influence the evolution of tumors, to study their dissemination and their treatment response. We look at molecular pathways and use our 3D bioreactor systems and our acrylamide gels, and also work with micro-patterned surfaces, to begin to systemically deconstruct molecular pathways. Then we use our battery of classic molecular biology and quantitative approaches to carefully analyze mechanisms and to gain new insights into the biology.

So basically we are moving back and forth between monolayer studies, organotypic culture work, and mouse models, as well as in silico analysis, and then we obtain human clinical tumor specimens that we either directly analyze, inject into mice, or manipulate in one of our 3D bioreactors. Essentially we apply every system on deck and collaborate with a diverse assortment of colleagues to conduct state-of-the-art screening and various manipulations to fully interrogate the system at all levels. We are constantly flipping back and forth across length scales and across models to explore what might be going on in the cells and tissues, and we are having lots of fun doing it.

What is the big picture problem you’re trying to solve?

Put simply, I want to understand how good tissues develop and what happens in a diseased state such as cancer.

I’m very convinced that while the genetic material in the cell can go awry, during normal development the genetic material is secondary. It is what genes are turned on and off, or how the cell behaves, that determines whether you make a breast or a pancreas or a liver, and whether that liver functions well.

Nevertheless, if I was diagnosed with early stage luminal breast cancer or DCIS, I would likely opt for intensive therapy, regardless of the risk. I would swallow my concerns and go in, get the surgery, have radiation treatment, if needed anti-hormonal treatment, and perhaps chemo—despite the fact that there is a reasonably high probability that the DCIS or low grade tumor never would have progressed anyway. However, if there was a good biomarker that could be used to determine whether my DCIS would progress or my low grade invasive cancer would metastasize, then I would be empowered to make a more considered decision. Clearly as a clinical community we are over-treating a huge number of individuals, but without good biomarkers and information regarding tumor behavior we are flying half blind. So how do we figure out which ones to treat really aggressively, and which ones not? Those are the kinds of questions we are trying to answer.

Is there an upcoming potential application of your research?

In brain tumors, for example, the standard of care is to do MRI imaging, to surgically remove the tumor, and then to do radiation therapy and in many cases, particularly for the high grade glioblastomas, chemotherapy. Those are very difficult and invasive treatments. What everyone wants to know is, how do we enhance therapeutic efficacy for the high grade gliomas? For lower-grade gliomas, how do we set tumor margins, what is the optimal therapy to reduce recurrence, and for all brain tumors, can we develop better biomarkers and prognostic indicators? And are there other kinds of therapies that doctors can use for some tumors?

We have some really beautiful data in breast, brain, and pancreatic cancer that suggest that the mechanical features of the tumor enhance their aggression. One of the first things we have begun to ask is, can you actually obtain additional information from current MRI imaging that would provide important clues about the mechanical state of that tissue? We have been working with imaging colleagues to achieve that goal, to be able to obtain MRI-guided biopsies, do the mechanical analysis, find out what the biophysical properties of the extracellular matrix are like around the tumors, determine the phenotype of the tumor, and then back that up to find the perfusion and diffusion behavior on the MRI. The goal would be to be able to use the MRI data to directly predict tumor phenotype. While we started this work in gliomas we are anxious to expand to pancreatic cancers and breast cancers.

What is important about this type of work is that clinical translation is feasible in only 3-5 years. In many other instances it takes a frustratingly long time to translate basic research findings to the clinic. From the researcher formulating ideas and conducting basic experimental studies, the work goes through multiple rounds of publishing, sharing, and validation, moving up through preclinical models and then into clinical studies. That of course does not mean that we do not continue to work hard, rather it underscores how many hurdles it takes to move a basic sciences finding to change clinical practice and how very important it is to ensure that what we study at the basic level and preclinical level is carefully thought out, well controlled, and held up to high scrutiny.

While I am not a clinician and I do not run clinical trials or conduct translational research, I am fully cognizant of the distance between basic sciences and clinical translation and I therefore try to ensure that what my group does is well done and relevant. What we are doing is basic and preclinical research that is critical for feeding the clinical pipeline. We recognize that once we make a discovery we must disseminate the observations to the general scientific community to be validated and expanded upon, and to our clinical colleagues to determine if the findings have any translational potential; at least with respect to the cancer work.

Collaboration seems to be an important part of your work?

My attitude is that we need to bring in people from across every discipline to work together so that we will be able to make a sustained difference. My goal is to assemble the smartest team that I possibly can, and it is my belief that that engineers are really bright, and they bring an incredibly important and unique perspective to cancer research.

I love working at UCSF because it’s a fantastic place for collaboration; UCSF is an ideal environment that fosters constructive cross fertilization of disciplines. Nevertheless, I also work with people at UC Berkeley, Stanford, Duke, Harvard, MD Anderson, Cornell, Georgia Tech, Memorial Sloan Kettering, UVirginia, Nebraska, McGill in Canada, the Beatson in the UK, and INSERM centers in France. These collaborations are essential to our work; I’ve spent years setting up these interactions and continue to exert much effort towards maintaining my network of colleagues. Experience has taught me that while not everything works as well as one would like, and not all collaborations are necessarily as constructive as one might have hoped, overall this this is the best approach to be able to successfully conduct interdisciplinary science. I personally have been fortunate to have assembled a wonderful cohort of colleagues that I cherish, and that have propelled my science forward in unexpected and exciting directions.

Of course I want to help my students and my postdocs get jobs and graduate, but my motivation is instilling them with this feeling that you have to search for the truth and do it the best way that you can. We also need to work across fields and with people who are out there in the trenches treating people.

Let’s face it, science is cool. It’s beautiful. But at the same time we have an obligation to the community, to actually see that what we discover is true and, at least when it comes to cancer research, that what we find has an application. I am a big proponent of basic science, which I believe is incredibly important, and given the pressures on scientists to publish publish publish or perish it becomes even more critical that the findings we generate must stand up to rigorous review, must be reproducible in other labs, and should be translated across systems.

Likewise, particularly with respect to cancer research, we need to ask what is actually important and what is needed, and to ensure that whatever we report is well done, relevant, and reproducible. Members of my group could spend hours trying to understand and cure mouse cancer or determine what makes cancer cells in a dish migrate the fastest, but that does not mean this will help us to understand and treat human tumor metastasis. Instead we need to be humble and invest the time and effort to understand what the important questions are, design the optimal model systems, and then work with clinicians to validate our findings and work towards translating our ideas. Maybe if we are lucky we will be able to make novel insights and actually help patients.

Ultimately my primary motivation is to truly understand and to be able to disseminate and translate my group’s research findings so that we make a difference. If that means that I need to invest in developing new model systems then I’m willing to do so, if it means that my group members and I need to work with ten other people from other institutions, then that’s what I’ll do. Basically my working model is that nobody can do everything and that if we work together we can make interesting and important discoveries.

What do you see as the big missing links in your field in the next few years?

In the cancer area, we really need to go into in vivo models and move toward the clinic. Right now there are a lot of people mucking around with cell lines and culture, and that area is ripe for really moving forward. The bottleneck at this point is that mainstream scientists are just starting to buy the idea that the environment is important. They don’t really believe that a cell running around in culture is reflecting something that goes on in vivo, so they need to see more definitive evidence in the clinic that supports it. They also need a model system.

The engineers with all of their technical prowess can really come in here and clean up, but they need to link up with the cancer biologists and they need to go in vivo and to begin to translate their findings to the clinic. And to that you need to set up a system, and they will need good in vivo models to manipulate mechanics. We are working hard at this. We need the engineers and biophysicists to take the work and really push this forward, and they need us.

I’m very committed to this area of investigation. If you go to the American Society for Cell Biology meeting now it’s very full of engineering and quantitative biology. When I first started doing this in 2000, we were one of the only posters in that area, and my engineering students were really depressed because they felt lonely. Now it’s completely blown up, and this is great. That’s what we need to happen in the area of cancer research and at meetings such as the American Association for Cancer Research (AACR) annual conferences, we need to work all together across all scales and disciplines so that we can truly make a difference and reduce cancer incidence and deaths.

Has working with the joint graduate bioengineering program impacted your collaborations?

I love the engineering students! I tell them they can own the world if they want to, because they’re not afraid of going in there and learning. They know the quantitative part, they know how to think about these things in different ways. We give them the depth in biology and the toolbox so that they can go forward and do what they want to do. I’m very proud of them, I love working with them. I think this program is fantastic and it’s one of the reasons I came to UCSF.