HER2 Subtype Offers Clues for Individualizing Breast Cancer Care

One-third of patients previously identified as having HER2-positive breast cancer were found to also have a luminal subtype that was resistant to chemotherapy and trastuzumab but still sensitive to the triplet neoadjuvant regimen of pertuzumab, trastuzumab, and chemotherapy.

Peter Beitsch, MD

One-third of patients previously identified as having HER2-positive breast cancer were found to also have a luminal subtype that was resistant to chemotherapy and trastuzumab but still sensitive to the triplet neoadjuvant regimen of pertuzumab, trastuzumab, and chemotherapy, according to results of the Neoadjuvant Breast Registry Symphony Trial (NBRST) study.

The phase IV registry study, which was presented during the 2015 San Antonio Breast Cancer Symposium in December, showed that 32.5% of patients with HER2-positive breast cancer also had luminal-type disease. In this resistant group, chemotherapy and trastuzumab alone showed a pathological complete response (pCR) rate of just 6% compared with 39% with the addition of pertuzumab (P = .0002), representing a distinct subgroup for tailoring treatment.

OncLive: Can you discuss the NBRST study?

What were the findings?

In an interview with OncLive, lead study author Peter Beitsch, MD, chief physician, Dallas Surgical Group, discussed the phase IV study, the significance of these findings, and the potential they have on changing clinical practice.Beitsch: The NBRST study is a neoadjuvant registry looking at both Mammaprint and BluePrint analyses prior to neoadjuvant chemotherapy, and then seeing how patients respond to that chemotherapy. At the end, they undergo surgery. The primary goal was to relate the Mammaprint and the BluePrint assays to either a pCR or a non-pCR.There were several major findings. The most important one is that we reclassified about 25% of the patients in the study. We looked at their functional molecular subtype, which is the way it was initially described with messenger RNA, as compared with the surrogate subtyping, which we do nowadays with immunostaining for ER-positivity and PR-positivity, and FISH testing for HER2.

The second major finding was quite a surprise, actually. We had known for a long time that triple-positive patients, which are patients who are ER+, PR+, and HER2+, respond less to chemotherapy plus trastuzumab than if you are ER-negative and HER2-positive. That group of patients has always done better with pCR, survival, etc.

This has always been a conundrum. In the first part of the study, we were seeing a trend where the triple-positive patients were subtyping into two groups: one that was HER2-driven and one that was of a luminal subtype. When we look at the pCR rates between the two, it was apparent that there was almost no pCR in the luminal subtype, and almost all of the pCR in the triple-positive patients came from the HER2-driven subtype.

However, over the course of the study, the standard of care changed regarding chemotherapy. For HER2-positive patients, it went from chemotherapy plus trastuzumab to chemotherapy plus trastuzumab plus pertuzumab.

Therefore, about 6 months ago, when we were analyzing the data, we saw this trend of a 3% to 4% pCR in the luminal patients, and then it tripled up to an 18% pCR. We thought, “Maybe we’re not as smart as we thought we were.”

Then, we thought that, perhaps, there was indeed a difference between response with trastuzumab and chemotherapy only versus the dual blockade plus chemotherapy. Low and behold, it’s almost like a dividing line. Before receiving pertuzumab, in patients who received chemotherapy and trastuzumab, they demonstrated a 4% pCR. When you add pertuzumab, there was a 40% pCR.

How could this information impact clinical practice?

It’s apparent that in the triple-positive patients who are luminal subtype, which is about one-third of all HER2-positive patients, they need not just trastuzumab and chemotherapy; they need trastuzumab, pertuzumab, and chemotherapy to have an improvement in the survival—if you consider pCR to be a surrogate for survival, and I think most people consider that to be the case. We are pretty excited about that.Pertuzumab is already approved in the neoadjuvant setting for HER2-positive patients. However, the vast majority of patients don’t get treated in the neoadjuvant setting; they get treated in the adjuvant setting. The cancer has already been removed, the lymph nodes have already been sampled and biopsied, and most of the patients are going to get chemotherapy with trastuzumab.

The problem is, in the patients who are triple-positive and are of the luminal subtype, it is probably not going to help much if you don’t add in pertuzumab. Our study creates an excellent argument to insurance companies to allow for pertuzumab to be given in addition to trastuzumab in the adjuvant setting.

Are there additional steps planned in this research?

I think it is going to make a major impact in the majority of triple-positive women and, probably, the subtyping should probably be routinely done on those patients who are triple-positive. This is to see if they are that luminal subtype, because they really need the pertuzumab/trastuzumab combination, even if it is in the adjuvant setting. I think this will have a major impact on women with HER2 disease.We are planning on following these patients for 5 years and beyond. We are going to determine if these pCRs translate into survival; there is still some debate about that. We have some ongoing studies looking at the luminal subtypes. There are some that don’t really respond well with chemotherapy, but some do and so we want to tease out the luminal subtypes that are responding. Perhaps, there is a signature we can find to determine the luminal patients who need chemotherapy, because the vast majority does not.

What are the challenges associated with finding and examining all of these subtypes?

We are probably planning on doing an NBRST “2” study to look at the low-risk group and see if they don’t need therapy beyond surgery—not even endocrine therapy. They are at such a low risk for recurrence. We will tease out the higher-risk groups that are basal. It is probably a heterogeneous group of patients, so we will try to split those up and see exactly what therapy might be directed at one of those basal subtypes.We have always conducted clinical trials of huge groups of patients with thousands of people in each arm, but not really subtyping them down into individual groups. When you do those big studies, you can get statistical significance because your groups are so big.

Could molecular subtyping also help avoid overtreatment?

However, the reality is: breast cancer is not breast cancer. Breast cancer is tens or maybe hundreds of different kinds of breast cancer, and we need to push them out into groups with more targeted therapies. Some patients may not need very much therapy at all to be cured of their cancer. Some may need a lot or some may need a target that we can tease out, find, and deliver a therapy with lower toxicities and improve survival. That is the name of the game: individualizing therapy for the patient. There is no question. There are both ends of the spectrum where we definitely need to treat some patients with chemotherapy more aggressively, but we better nail down their subtype correctly.

Then, there is the other group that does not really need chemotherapy. Once we subtype them appropriately, then we can decrease the amount of toxicity from chemotherapy, and perhaps they end up just needing endocrine therapy. Perhaps, they just need surgery. I think it is going to help give targeted therapy—an appropriate therapy—to the people who need it. It is personalized medicine, and we are getting there.

Beitsch P, Whitworth P, Baron P, et al. One-third of HER2 positive patients have 80-gene luminal subtype that is resistant to chemo-trastuzumab but sensitive to chemo-trastuzumab-pertuzumab: Critical implications for the adjuvant setting from the NBRST phase 4 neoadjuvant study. Presented at the 2015 San Antonio Breast Cancer Symposium; San Antonio, TX; December 8-12, 2015. Abstract P4-14-29.