Skip to main content
Premium Trial:

Request an Annual Quote

Q&A: MD Anderson's Mendelsohn Discusses WIN Consortium's Effort to Advance Personalized Cancer Drugs


mendelsohn.jpgA global consortium of 22 cancer centers is working with drug and diagnostics developers to advance treatments that are effective against genomic abnormalities underlying various cancers and speed them to market.

Launched by the Institute Gustav Roussy in France and the University of Texas MD Anderson Cancer Center in July 2010, the Worldwide Innovative Networking Consortium aims to help drugmakers change the way they develop drugs by focusing on treating diseases with common molecular features, rather than treating diseases by organ site. With access to the expertise of 22 cancer centers, drug developers can test their genomically guided drug development strategies in patient samples from different countries and identify the most appropriate markets for their drugs.

Additionally, through the consortium's diagnostic partners, biopharmaceutical companies can come to WIN with an investigational drug for which they want to pursue a personalized medicine approach and have access to the latest molecular analysis technologies. Currently, the technology partners that have joined WIN are Life Technologies, GE Healthcare, and Agilent.

At the WIN annual symposium in Paris in July, consortium members will be presenting two projects for approval by the organization's advisory committee. John Mendelsohn, chairman of the WIN Consortium and president of MD Anderson Cancer Center, couldn't provide details on the studies ahead of the committee's approval, but said that both projects will focus on pairing drugs with specific genetic abnormalities in cancer.

Mendelsohn recently spoke to PGx Reporter about the ongoing work of the WIN Consortium and how drug development is changing as biopharmaceutical firms are embracing genomic strategies. Below is an edited transcript of the interview.

What's the aim of the WIN Consortium? How did it start?

This is an idea that was begun at the Institut Gustave Roussy [in Paris]. We joined in early in making this collaboration happen. The idea is to pull together cancer centers from around the world, and pharmaceutical companies, and work together to try to move more quickly in advancing clinical trials, to test drugs against patients who are pre-selected to be likely to respond to the drug.

We are now able to take a human tumor, sequence the genes in it, and find out which genes are abnormal in a patient's tumor, and it isn't the same for every breast cancer and for every lung cancer. There are different genes that are abnormal. We couldn't do that in real time, in a week or two. You can't ask a patient to wait many months to decide what drug to give them. Only recently has the technology made it possible to do this in a short period of time. Meanwhile, the drug companies have made drugs recently that attack the products of the genes that cause cancer. We're able to actually study an individual patient's tumor and we're able to find out which genes are abnormal, and then select a drug to try on that patient, an experimental drug, that was designed to attack the product of that gene. We're finding that means that some of the new drugs that are being tried for the first time in human beings are … working in a large percentage of patients. That wasn't the case with the old way of studying drugs, where they were just given to advanced patients. We didn't know at that time, of course, what was wrong with the patient in terms of the genes.

With that as background, the next issue is that part of the response to drugs isn't that the gene's wrong in the tumor, it's how the patient handles the drug and how the patient responds to the cancer. We don't have any way of testing for that very well right now. Our hypothesis is that, if you're testing a drug, and you're finding tumors that have abnormalities in which that drug might be likely to work, you'd be wise to test it in a very diverse population of patients. Some patients may respond better than others, not because of the nature of their tumor, but because of their own, normal genetics, or the way they handle the drug and in the way they handle the tumor.

We've made this possible now by gathering together cancer centers from China, India, the Middle East, Arabic countries, Israel, Europe, and North and South America, and this gives the people exploring targets for cancer therapy a chance to look at very diverse populations. This is catching on.

The other aspect of the WIN Consortium is that it isn't just a group of cancer centers with academic credentials. We're also bringing in the companies that produce the equipment, so that we are able to sequence these genes, study the genes, and we're bringing in the companies that develop the new drugs. We're all working together so we can take an idea, convert it into an experimental drug, and do the trials that would bring it toward approval by the [US Food and Drug Administration] and other [regulatory] organizations. The idea is to develop better drugs that are more likely to work in patients, that are more rapidly developed, that are put out for general use, and that can become [the] standard of care.

What are the advantages for drug companies and diagnostics companies so that they would want to join a collaboration like the WIN Consortium?

They are going to be quicker to get to market, because they are going to be testing their drugs on patients that are hopefully going to be more likely to respond [to their drugs]. They might pick up a drug that works very well on a Chinese population, and not in a European population, or one that works in Brazil where we know studies will be done. So, they are going to find out which markets are relevant for their drugs and get the trials done with a higher likelihood that they'll be positive and move toward regulatory approval quicker.

In the consortium, are drug industry partners willing to share information on investigational drugs with other industry collaborators?

No, each trial is going to be floating on its own tub. If a drug company wants to study their drug, they'll have to fund the trial, and they will have the infrastructure and the participation of the people in the consortium. They will keep confidentiality in the same way they would if they just privately worked with one or two of the cancer centers. This way, they'll be working with 20, and we think that number will probably increase to 30 or 40 centers, soon.

At the American Society of Clinical Oncology's annual meeting this year several drug firms announced they would work together to study combination drugs directed at more complex molecular targets, trying to figure out mechanisms of metastases and recurrence. Would the WIN Consortium offer opportunities for drug developers who want to pair up and study combination drugs directed at a number of gene targets?

That pairing is only beginning to happen, and it's going to happen more and more, because, again, it gives the company a competitive advantage where their drug alone may have a very low response rate. But their drug paired with another drug that attacks another gene that's abnormal in that cancer, might prove to be very effective. And, we will make that [type of collaboration] possible, but they don't need to go through WIN to do this. I think the companies are coming around to the idea that two- or even three-drug combinations may move them faster toward regulatory approval. But we will be there to help them at WIN.

The conventional wisdom about pharmacogenomics-driven personalized medicine strategies has been that they shrink patient populations and the market size of drugs. Do you think as drug companies are employing new business strategies, discovering biomarkers important in global populations, and collaborating in large consortia to develop drugs, that PGx can help drug companies redefine what it means to develop a blockbuster treatment?

The old model was, 'Let's try to develop a blockbuster drug that will cure all lung cancers.' The new model is, 'Let's find out which cancers will respond to our drug, find it out quicker, and not have to do huge trials.' I'm hoping that [with the new way] a drug can be developed with $200 million instead of $1 billion because the trials that will be done will have more positive results more quickly and move toward regulatory approval.

But the word "blockbuster" is leaving the lexicon. We're not going to have one drug that works for all lung cancers or all leukemias or all cancers in general. There are 200 to 300 genes probably that are relevant to causing cancer, and for any one cancer there [are] probably a half dozen — there may be more or may be less — that are the culprits. And it's going to be unusual to have one drug that's curative. There are few types of cancer where it's one gene that causes the problem and in that situation a drug can produce a very long response, but usually those patients develop resistance, too.

What near-term projects do you have in the works?

We have two projects that we're going to present at the [WIN Consortium Symposium] in July in Paris for approval by the assembly. At that point, we hope to launch [these projects] in the fall. For a new organization to be able to launch a new trial in a little over a year would be very unusual. We're working against a very aggressive time line.

What oncology areas are you conducting trials in?

Both of these would be trials where we're trying to pair drugs with genetic abnormalities. The type of cancer wouldn't be as relevant as the genetic abnormality. That's a different way of looking at cancer. We're treating the abnormal gene rather than the type of cancer.

You've said several times that working with a group like the WIN Consortium would allow industry to take their drugs through FDA much faster. Do think the FDA can make the path a little easier and clearer for drug companies developing drugs targeted at genomic markers?

I think the FDA is trying to do that. I think the main area where the FDA can speed things up on is guidance for [Rx/Dx] combination products. But even there I think they are trying to be helpful. They have a very tricky [charge] trying to balance protecting the public against side effects [of healthcare products] versus trying to speed drugs to market through the approval process so more patients can have access. They … could do things more quickly and with more efficiency. But then we could, too. When we analyze why it takes so long to get a clinical trial done we have to point the fingers at ourselves as well as at the FDA. I think all of us need to commit to moving expeditiously and accept certain risks and not aim over and over again for perfection. Because there is nothing perfect.

You said that you are applying sequencing technologies to your studies. Are you concerned at all that this will slow down the approval process for drugs that use the data generated through this consortium?

You know, if you use sequencing data to pick your patients, and you get a 50 percent response rate that lasts more than a year, whatever that sequencing technique was, it worked. If there are errors in it, it still worked … The proof in the pudding will be that whatever the biomarker detection mechanism is, if it's effective, it would be reasonable to approve it. There are mistakes made when you sequence, so we have to resequence many times in order to cover for those mistakes. There's no perfect test.