"Dream Team" Leader Looks First to Existing Drugs in Exploring Epigenetic Therapies

Stephen B. Baylin, MD

Deputy Director

Sidney Kimmel Comprehensive Cancer Center

Johns Hopkins Medicine

Baltimore, MD

Stephen B. Baylin, MD, is conducting research in his laboratory at Johns Hopkins Medicine that aims to bring epigenetic therapy to the forefront of cancer management, in particular in gaining a better understanding of the abnormalities of chromatin and DNA methylation that may account for the development of epigenetic abnormalities during tumor development.

Baylin, who is the Virginia and D.K. Ludwig Professor of Oncology and Medicine, chief of the Cancer Biology Division, and associate director for Research at Hopkins, has received numerous awards and coauthored more than 350 peer-reviewed publications. Along with Peter A. Jones, PhD, DSc, at the University of Southern California, he leads the Stand Up To Cancer Epigenetics Dream Team.

1. Please briefly describe the research in your laboratory as it relates to epigenetic changes in cancer.

For several decades, we have been trying to determine how an epigenetic alteration can play a principal role in the initiation or progression of cancer. Essentially, we are attempting to understand the molecular processes that drive it, how it evolves over the course of a cancer, and, if we reverse the process, how it would give us new and better ways to manage cancer.

2. How do epigenetic modifications contribute to cancer development?

I like to make the analogy that your DNA is the hard drive, upon which all the information needed to instruct every single cell in your body to do what it needs to do is written. It can direct normal, healthy processes and, if it is damaged, it can govern abnormal processes that contribute to the development of disease. But a hard drive without software does not know how to play out its potential. The functions of the DNA are determined by which genes are active or inactive at any one time and by certain structural aspects of the DNA, both of which require the software. One of the principal software packages is epigenetics. It’s an oversimplification, obviously, but I think it is an effective analogy.

The way that this software package works is that we wrap our DNA up very tightly around protein structures (laid out flat, the DNA in any given cell would spread out a long way). How tightly or loosely you wrap the DNA determines how it functions. This is a complex process that involves how you move the proteins around and how you modify them. It also involves the biochemical modification of the DNA itself—a process called DNA methylation. Essentially, anything that goes wrong with the hard drive can also go wrong with the software in cancer and there is a delicate balance to these modifications. A building enterprise has developed to understand all the epigenetic abnormalities that we now know are present in the cell and their role in cancer.

DNA methylation was the earliest identified epigenetic change. It occurs all over the DNA, but one of the regions we focus on is where gene regions start and sites within these which are one key determinant whether the gene is switched on or off. About half of the genes in our body don’t have DNA methylation in and around these sites, as they need to be active or be ready to be activated. In tumors, even single ones in an individual patient, anywhere from 50 to several hundred genes have methylation at these sites when they shouldn’t. This can shut down genes inappropriately and, if this happens for a gene that is preventing the development of cancer (a tumor suppressor gene), it can help drive cancer formation. It has the same outcome as damage to the hard drive, but in this case the hard drive is still functioning underneath.

We are trying to figure out which genes are involved in these epigenetic processes and, if you could put the software package back in an appropriately functioning state, whether this could offer a therapeutic strategy for cancer. There are old drugs, which used to be toxic, but at lower doses will reverse the methylation of DNA. My colleague, Peter Jones, with whom I work closely, discovered the action of these drugs. They make it a possibility that we could reverse inappropriate methylation and, at low enough doses, they could be specific for these processes while not generally toxic to cells, and could be added to the management of common tumors. Ultimately that’s where we would like the field to go.

3. Which are the most promising epigenetic-targeting drugs currently under development?

I personally think the old drugs are the most promising at present, because of the way in which we use them, at a low dose, which doesn’t actually kill the cells straight away but gives them time to readjust the software package and reprogram it. These are drugs like azacitidine (Vidaza), which reverses DNA methylation, but there are also drugs that inhibit the proteins that work alongside DNA methylation. These drugs are very much still under development as we work out how to use them effectively, and they will remain an important part of the cancer armamentarium. I lead one of the Stand Up To Cancer projects in the United States, along with Peter Jones. Our charge is to take these older drugs but develop them in newer ways in both the laboratory and the clinic, and we are getting some very important trends to efficacy, particularly in lung cancer.

There are also newer drugs that are targeting other aspects of the epigenetic software, which may work alongside DNA methylation or may be less directly associated with it, processes that we didn’t understand five to 10 years ago. Pharmaceutically targeting these may make the older drugs work even better, because they help push the DNA methylation away by different processes. There are a number of these drugs that are being developed that may be useful alone or in combination with older drugs. That is going to be a significant area of development.

are being discovered and pharmaceutically targeted. We currently know very little about what cancer cells will do with these drugs and how they will be best used. A growing number of clinical trials are under way to develop these concepts and translate them into advances for patients. It is a great new enterprise, with tremendous promise, but it is really just starting out.

4. How do you think that epigenetic-targeted therapy will be best incorporated into existing cancer treatment regimens?

Most drugs today, although there are very exciting exceptions, are very difficult to employ as single agents to do the best therapeutic job. Usually in the exceptions, drugs generate big responses in just one or two different types of cancer and, while there may be dramatic early responses to these drugs, all too soon patients can develop resistance. So I think most drugs beg combinations, and the same will be true of epigenetics drugs.

In using the older drugs, our biggest challenge has been that they only work alone in solid cancers (the biggest killers, like lung cancer) in a small percentage of patients. What has become apparent, however, is that these drugs will sensitize patients to subsequent therapy, such as chemotherapy. We are formally testing this in lung cancer in the clinic as we speak, but there is a suggestion that combining epigenetic drugs with chemotherapy could improve responses or allow lower doses to be used so patients will do better.

We are also very excited about a low number of patients that formally suggest to us that epigenetic therapy could also sensitize them to immunotherapy. We are about to start a trial to formally test that also. So I’m a great believer that manipulating the software package can set up cancer cells to be more receptive to subsequent therapies, which include immunotherapy and chemotherapy, but maybe even include some of the newer targeted drugs (although this is an area that we as a community have yet to carefully address). I think that these approaches of using epigenetic therapy to sensitize to other therapies will be one of the best uses of this therapy approach.

5. What are the most significant challenges that need to be overcome in the development of epigenetic-targeted therapy for cancer?

I think that, although the promise is real and it’s a very exciting field, the challenge is huge, even for the older drugs. Although they have been around for a long time, we haven’t always known the proper dosage and sequencing and how we might be able to use them with other drugs, and these are questions we are still working on. We are trying to demonstrate these things in a laboratory—which is not always easy—to come up with the best prediction as to what will happen in the patient. We haven’t yet completely figured this out for these older drugs, so imagine the challenge for the newer drugs.

Pharmaceutical companies and academia are working to develop drugs that will target a given step in epigenetics—this is something that is being done in a big, big way. In this case, we then have the challenge once that is done to figure out what a cancer cell or an animal is going to do with the drug. The challenges are how do you investigate that, how do you come up with the best predictions of response, how do you determine whether this will be (either alone or in combination with another therapy) an effective drug against disease. Also, how do we go from developing these drugs in a laboratory to deciding that they should be moved forward into clinical trials? Then, of course, you have all the challenges associated with carrying out clinical trials. So there are years and years of work ahead. The biggest challenges will be not only the development of the drug itself, but also the prediction of how to best use it.