Abandoning Old Paradigms: Embracing Molecular Targeting in Breast Cancer Treatments

A Conversation With Pat Whitworth, Jr, MD

Pat Whitworth, Jr, MD, is the director of the Nashville Breast Center and a clinical associate professor of surgery at Vanderbilt University in Tennessee. He is currently involved in clinical breast cancer trials that evaluate pretreatment and on-treatment molecular profiles to target neoadjuvant regimens for breast cancer.

Our editors met with Whitworth at the Miami Breast Cancer Conference to discuss how breast cancer treatment has evolved from a “kitchen sink” approach with chemotherapy to a new paradigm that targets chemosensitivity.

OBTN: Can you describe how molecular profiling has affected how you treat patients with breast cancer in your practice?

Dr Whitworth, Jr: A lot of things have happened over the course of the past 10 to 20 years. There was a period of time when we had to debate about whether patients who were node-negative and estrogen receptor (ER)-positive even needed tamoxifen or endocrine therapy. The NSABP B-14 study showed us that indeed there was a benefit for those patients. That same question came up later about chemotherapy: Would chemotherapy improve the outcomes for node-negative and ER-positive patients? Frankly, we were excited to learn that we could improve outcomes for those patients with systemic adjuvant chemotherapy. But there’s a curse with the blessing. At that time, we weren’t able to target that treatment. So, for many years, we grew up giving chemotherapy to all of those patients, primarily based on their level of risk for recurrence. We didn’t use that same criterion for treatment with tamoxifen or endocrine therapy. If a patient was ER-positive and had any risk, we treated them with endocrine therapy. If the patient was ER-negative and high-risk, we didn’t add endocrine therapy because we knew it was targeted to ER-positive patients.

The same thing was true with HER2. We learned in 2005 that Herceptin could really improve outcomes in patients who expressed HER2 and was beneficial even in nodenegative patients. We didn’t generalize that to everybody (ie, patients who didn’t have HER2 overexpression); we were targeting that treatment, but we weren’t able to target chemotherapy to those patients who had chemosensitive tumors. The emotional history that we had with treating these patients was really based on their risk for recurrence. If there was a high risk for recurrence, we gave them chemotherapy. It was just all about risk. That’s why node status was critically important in cases with an apparently low-risk primary tumor; positive lymph node status would be critical in helping us make treatment decisions.

So, endocrine treatment and anti-HER2 treatment are based on targeted biological sensitivity; but chemotherapy is just based on overall risk, not chemosensitivity?

In 2006, we learned that we could in fact target chemosensitivity. The analysis of the B20 study using the Oncotype DX 21-gene recurrence score showed that patients who had a low score had no benefit whatsoever, and patients who had a high score had a lot of benefit from chemotherapy. That was a huge change in our understanding. However, translating that change in understanding to a change in practice is a different matter because we grew up with a very strong emotional attachment to this notion of a “kitchen sink” approach for someone at high risk. If we saw a patient with a lot of positive nodes or a high-risk tumor, we were conditioned to think, “This patient needs chemotherapy. Her risk is high. We know chemotherapy helps with high risk.” And that’s what we did—treated the patient with chemotherapy. Even though we understood targeting with ER and HER2, it still was very difficult to let go of that notion and embrace individualized chemosensitivity targeting.

There is a difference between understanding the new information about targeting chemotherapy only to patients who have chemosensitivity versus applying that in practice. When we began to understand that chemosensitivity can be measured and can be identified, and absence of chemosensitivity can be identified with a low recurrence score, it was still difficult not to use the approach that we have always used to treat the patient with everything we’ve got. I think the fact that we could understand targeting based on ER status and based on HER2 status enabled us to abandon the old paradigm based on risk, and move to a new paradigm based on targeting real chemosensitivity.

Would you say that most oncologists have overcome this hurdle and have started offering treatments other than chemotherapy?

I would say most oncologists are over this hurdle—this intellectual and emotional barrier where we reflexively think of chemotherapy benefit as a proportional benefit for all patients. We’re now moving into an era where we can actually target the patients who benefit and avoid unnecessary treatment in the patients who don’t. A few years ago most medical oncologists weren’t ready to make that change, except perhaps in very low-risk patients. In those very low-risk patients they would use an Oncotype DX 21-gene recurrence score to perhaps find a potential chemotherapy benefit for a patient who wouldn’t get chemotherapy based on traditional parameters We’ve gone from that very limited application just two or three years ago to a situation now where the vast majority of medical oncologists avoid chemotherapy in nodenegative patients with a low 21-gene recurrence score, in spite of traditional factors like larger tumor size or premenopausal status.

It seems as though there’s been an explosion of information and commercial options for targeting breast cancer treatment. What’s the next hurdle / challenge?

Well, there’s a tremendous amount of excitement about next-generation sequencing. But at this point we haven’t fully incorporated more established genomic profiles in practice in a lot of cases. We’re just learning to use those to help us make decisions about patients who might benefit from neoadjuvant treatment. Some very innovative thinkers and clinical leaders have found ways to treat patients with a short course of either endocrine therapy or chemotherapy, assess the response in a tumor, and make a decision—at least in the clinical trial setting— about the best course of treatment for that patient based on tumor parameters, such as proliferation and other parameters of response. I think that’s the next step now that we’re beginning to understand these molecular profiles and molecular profiling. We’re not going to just jump from here to next-generation sequencing to full genome sequencing and variant analysis for patients. The goal is to individualize treatment for these patients. And short-course treatment based on initial assessment of the tumor biology, with rebiopsy to assess biologic response, will be the next step in individualized treatment.

Are there specific new assays that are available now that are the next step from the recurrence score?

Well, I don’t think we have Recurrence Score 2.0 yet. The MammaPrint 70-gene assay is in clinical trials, mostly in Europe right now in the MINDACT trial. Again, the goal is to begin to target treatment to those patients who benefit. The MINDACT trial has an elegant design, and it will be interesting to see what we learn from that because it is really comparing clinical judgment with the molecular assay and whether putting those together or leaving them apart is appropriate. Right now, that trial is still accruing patients, but it’s accruing both node-positive and node-negative patients. The 21-gene score is now being tested in node positive patients in the SWOG 1007 trial in the United States. So the next step is to extend these assays and newer profiles like PAM50 and others to higher risk patients.

Are there any specific things that you see on the horizon that you find very exciting?

Well, one of the areas I’m really excited about is the neoadjuvant setting, both in terms of treatment and in terms of research. The ACOSOG Z1031B study showed we can target endocrine sensitivity and resistance using a short course of neoadjuvant endocrine treatment. The followon trial comparing different anti-endocrine strategies should open soon.

For other tumor types, we are presently conducting a neoadjuvant registry, the NBRST registry, looking at the 70-gene assay in patients who are getting neoadjuvant treatment, chemotherapy, or endocrine therapy. Our goal is to determine whether the assay has an impact on the neoadjuvant treatment decision making and to determine how well the genomic information we get from that assay and from the underlying 4500 gene chip platform correlates with the patient’s actual response to the treatment. There are some other neoadjuvant studies using molecular profiling to try to design better treatment for these patients individually, too.

Have community oncologists been conditioned to think in these terms now, or are there specific recommendations or advice that you would offer them?

I think doctors are adopting the newer biology and chemosensitivity-based approach, and certainly they’ve changed their practice for the earlier stage tumors. In the node-negative tumors we’re seeing much more reliance on molecular profiling, specifically the use of the 21-gene score in the United States and the use of the 70-gene assay in Europe. I think we are seeing some excitement and enthusiasm about biological targeting in the community because the results have been so good for neoadjuvant treatment in patients who have a molecular profile that demonstrates HER2 overexpression. In an MD Anderson study, two-thirds of patients had complete pathologic disappearance of the tumor if they received Herceptin as part of the neoadjuvant treatment and were HER2 over expressers. Any doctor that has seen that, and most medical oncologists are well aware of that, has started saying, “Gosh, if she’s HER2-positive, my threshold for thinking about neoadjuvant treatment just went down.” They’re much more inclined to treat her in advance because of the information they get from her response and the fact that she can have her chemotherapy before surgery. We know she’s going to need it. So, she can have chemotherapy done before her surgery, and if she’s having mastectomy and reconstruction, that’s a lot less complicated because that’s out of the way.

We’ve known for a long time that ER-positive tumors, in absolute terms, have a much bigger benefit from endocrine therapy than from chemotherapy. Even though we’ve known this a long time, there’s been a bit of a disconnect in that our sense is that chemotherapy is strong and if the risk is high, we need to give something strong. But if you look at the numbers, patients who are ER-positive have about twice the benefit from endocrine therapy that they get from chemotherapy. So, it makes sense now to look at patients who have that particular molecular profile (ie, ER-positive patients, and in particular, patients who have a low recurrence score) and think about endocrine therapy as their neoadjuvant strategy instead of chemotherapy.

The goal there is not to see a complete pathologic response because we rarely see that with endocrine therapy, but to shrink that tumor and make that patient a better candidate for surgery, whether that’s breast conservation or mastectomy.