Breaking through the limits of cancer medicine

Gordon Mills, M.D., Ph.D., the new director of precision oncology at Oregon Health & Science University, outlines a vision for the field.

Photo: Gordon Mills, M.D., Ph.D. (OHSU/Kristyna Wentz-Graff)

The cancer diagnosis, a rare sarcoma, was devastating for an old friend of Gordon Mills. “We spent lots of time trying to understand his disease, trying to characterize it, and he went to my colleagues to ask about what was the next therapy for him,” Mills says. The proffered advice was a searing reminder of the limits of cancer medicine. Between them, the consulted experts recalled an approach that had benefited a “somewhat” comparable patient two years earlier.

“That’s not acceptable,” says Mills, who has just been hired as director of precision oncology at the OHSU Knight Cancer Institute. “We need to know enough about every single patient so that we can pick what’s right for them, rather than saying, gee, I once saw somebody like you,” he says.  “We need to do this better.”

Mills has covered a lot of ground since he landed at the MD Anderson Cancer Center in Houston in 1994. Among many leadership roles there, he founded the first cancer systems biology department in the U.S. He oversaw the institute’s breast cancer and ovarian cancer programs. He established a center for molecular markers that evolved into the Institute for Personalized Cancer Therapy, which he co-directed. He has authored or co-authored close to 900 scientific papers, becoming one of the most widely cited medical scientists in the world.

Cancer translated met with Mills to talk about his vision for precision cancer medicine at OHSU.

CT:  Precision oncology has become a popular buzzword, but definitions of the phrase are surprisingly variable. How do you define it?

Mills:  What we want is to understand each patient’s cancer in sufficient breadth and depth to select the therapy mostly likely to give that individual the best outcome. And we want to select that therapy from the start to achieve the outcome that we want: a durable response with decreased toxicity. By understanding that each cancer is different and each patient is different, we should be able to select that with speed and efficiency.

CT:  The drug imatinib (Gleevec), ushered from lab to clinical success by Brian Druker, M.D., here at OHSU, is an outstanding example of precision medicine for people with chronic myeloid leukemia. A disease with a three- to five-year life expectancy became, for most patients, a chronic, long-term condition managed with a daily pill. Why haven’t there been more Gleevecs?

Mills: Gleevec was an incredible discovery – and an unbelievable boon for that population of patients. But most of the cancers that we treat, particularly the solid tumors, are vastly more complex. Single agents are not going to be enough because these cancers have many aberrations driving their progression, unlike chronic myeloid leukemia, in which only one genetic aberration is the primary driver of the cancer. So we’re going to need combinations of therapies to manage the more complex and challenging tumors.

CT: There’s been a little bit of pushback against precision oncology. What comes to mind is a New England Journal of Medicine commentary published last year. I’m sure you read it.

Mills: Oh, yes.

CT: The critique was that the benefit as it is currently practiced would be limited and that its prominence may divert funding from other promising avenues of research. What do you make of that?

Mills: That critique was based, in part, on one of my papers. And what we showed is that when you restrict the concept of precision oncology to phase 1 clinical trials evaluating new drugs, the number of patients that benefit really is fairly limited. But what the authors of that commentary forgot is that that analysis excluded all of the successful therapies. So imatinib mesylate and all of the drugs shown to work in melanoma, in lung cancer – drugs that have changed the history of those diseases – were not counted.

Now, I think the other criticism is real. As we look at single-drug therapy, we’re not going to have the success that we had hoped for in precision oncology. But we’re beginning a new chapter where we’re looking at combinations of targeted therapy drugs and immunotherapy drugs and we are seeing truly remarkable changes in diseases for which we had nothing 10 years ago.

CT: What if we use the experience with ovarian cancer to look at how the field is progressing?

Mills: Ovarian cancer is an interesting challenge. It is not one of the easier diseases. We find it late. We find it frequently when patients have kilograms of tumor, so it’s an advanced disease that is going to be very difficult to cure. Now, we have greatly improved the five-year survival rate. That has almost doubled in the last 10 to 15 years. So we are making progress, though we haven’t yet improved the cure rate.

But a whole new area has come to the fore, the so-called PARP inhibitor drugs. They target the aberrations that drive ovarian cancer by undermining the ability to repair damaged DNA. Over half of patients with ovarian cancer have an abnormality in that DNA repair pathway, and these new drugs specifically work on that population of patients.

And we are seeing responses like I’ve never seen before. We’re talking about studies where we show a 10-fold increase in survival with one agent over a period of time. No cures yet, but increased survival. And at some point we want to convert this into durable responses that approximate cures.

CT: So, what at this point are the greatest challenges facing precision oncology?

Mills: We’re going to need to find ways to begin therapy at an earlier stage of disease. Through efforts such as those at CEDAR [the Cancer Early Detection Advanced Research Center at the Knight Cancer Institute] we are going to have a much greater opportunity to have an impact. We now use most of our targeted therapies and our precision oncology agents after we’ve treated patients with chemotherapy and radiation therapy, and their tumors have had a chance to evolve and become even more complex and hard to deal with. So by moving things upfront to the first time we treat patients, and by treating earlier and earlier disease, we’re going to have remarkable improvements in patient outcomes.

CT: How will you measure your success in your new job at OHSU?

Mills: I want OHSU and the Knight Cancer Institute to be the destination of choice for all people in the Pacific Northwest. We here at OHSU will work to build a consortium with cancer centers in Seattle and San Francisco, and really make the Pacific Northwest a leading center for patient care and for patient research. And we will rival, if not exceed what you see in Southern California and in the Northeast.

◊ ◊ ◊

Further reading:

Limits to Personalized Cancer Medicine by Ian F. Tannock and John A. Hickman. NEJM (September 29, 2016)

Feasibility of Large-Scale Genomic Testing to Facilitate Enrollment Onto Genomically Matched Clinical Trials by Funda Meric-Bernstam, Gordon B. Mills and others. Journal of Clinical  Oncology (September 1, 2015)