Types of Genomic Testing

Transcript:

Hope S. Rugo, MD: Genomic testing is our new area that’s helping us determine which patients need what therapy within the broad classification of chemotherapy and hormone therapy. To understand that benefit, you really need to understand predictive and prognostic factors. Prognostic factors are those factors that we think about that help us understand when you stand here today with a new diagnosis. What’s your risk of recurrence, what’s your risk of death? What we really want to know from those prognostic factors is, what is your risk of invasive recurrence and what is your risk of distant recurrence? Those are prognostic factors.

Predictive factors are those that predict the benefit from specific therapies. We’ve had a lot of trouble with predictive factors. Right now, we can predict, to some degree, the benefit or lack of benefit from chemotherapy added to hormone therapy for early-stage hormone receptor—positive breast cancer. We can predict a little bit about the benefits of hormone therapy, but not a lot. There are emerging data, and particularly there is some help in a little bit of data that we might be able to predict longer-term hormone therapy benefit, but not definitive.

There’s also a lot of overlap between these predictive and prognostic factors. So, for example, hormone receptors are prognostic. They also predict benefit from hormone therapy. Grade is prognostic and it also predicts benefit, potentially, from chemotherapy. HER2 predicts benefit from trastuzumab and, to some degree, prognosis. But, now, it’s actually less of a prognostic factor because if you use trastuzumab, patients do very well. I suppose that gives them a good prognosis. It’s actually interesting. When you looked at MA.17R, the patients did so well with a 91% or 95% disease-free survival so far out. You pick patients who haven’t recurred, but, clearly, you could look at that population of patients and look at those prognostic factors quite well.

We really want to understand who’s at risk for late recurrence. We talked a lot about the clinical pathologic factors that seem to play a big role in predicting, using predictive factor here, the prognosis of patients or predicting the risk of late recurrence. But, they aren’t good enough because some women will be cured, and then, we all see women who had stage I breast cancer who have a 10-, or 12-, or 14-year recurrence. So, it would be nice to have a genomic test that would help us predict risk for late recurrence. And, there have been some data that suggest that a couple of tests have that potential ability, but none of them are perfect yet.

One of the problems is that it’s always a subset of patients who have a tumor available for testing, and then you have to have long-term outcome information on all of those patients. Two tests have actually looked at this in a variety of settings. One risk of recurrence test, the Prosigna test, has looked at the data in patients who were on the Austrian breast cancer trials, and suggested that they may be able to tell differences in risk of recurrence.

The Breast Cancer Index, though, had the benefit of being able to look at a subset of samples in patients enrolled on MA.17. And, those actually are really fascinating data. To me, those are the best data that suggest that a genomic test will be able to potentially tell us who’s at risk for late recurrence, but, more importantly, who would benefit from hormone therapy. So, in this subset of patients, they showed that if you were high risk by the Breast Cancer Index, you had a higher risk of recurrence. And, if you looked at the patients treated with letrozole, the patients who had a higher risk seemed to have less recurrence if they received letrozole versus placebo. Now, again, this is a subset of patients. Data from ongoing and completed trials, particularly data from the aTTom trial that they’re hoping to have in the near future, I think will help us a lot in understanding the benefit of those genomic tests. But, there are some early data that we could use cautiously in patients for whom we’re on the fence.

Transcript Edited for Clarity