Evolving Concepts in Treatment for Advanced Breast Cancer - Episode 8
Joyce A. O’Shaughnessy, MD: I want to talk a little bit about heterogeneity. Sara, you can tell us first about the update you provided very nicely with the KRISTINE data, and using the neoadjuvant as a way to come up with even more effective and less toxic approaches.
Sara A. Hurvitz, MD: The KRISTINE trial was a phase III trial that enrolled 444 women with HER2-positive breast cancer that was greater than 2 cm or node-positive. They were randomized to receive either 6 cycles of T-DM1 [trastuzumab emtansine] plus pertuzumab or 6 cycles of standard TCHP [trastuzumab, pertuzumab, carboplatin, docetaxel]. We were hoping and praying the T-DM1 plus pertuzumab would be as good a path CR [pathologic complete response] with an equal or better toxicity profile. The path CR rate that we published last year was 44% with T-DM1, and it was 56% with TCHP [trastuzumab, pertuzumab, carboplatin, docetaxel]. It was statistically significantly inferior. We updated these data with the event-free survival at 3 years, and 3-year invasive disease-free survival [IDFS]. This is the final analysis. It was preplanned, descriptive, and not powered to show differences.
Interestingly, the event-free survival, which takes into account anything that happens from the time you go on study during that neoadjuvant phase, demonstrated that a lot more patients in the T-DM1 arm have locoregional progressions. There are 15 patients in the T-DM1/pertuzumab arm that progressed during that period, compared to 0 in the TCHP [trastuzumab, pertuzumab, carboplatin, docetaxel] arm. When you look at invasive disease-free survival starting at the time of surgery and going on 3 years, the IDFS was exactly the same for the 2 arms. It’s really those early progressors who are accounting for this difference in outcome.
We looked at them a little more closely from a biomarker perspective, and it was fun, because we got to present our results and then Otto Metzger Filho, MD presented his results relating to heterogeneity. Indeed, those patients with HER2 lower expression, either by mRNA [messenger RNA], IHC [immunohistochemistry], or increased HER2 heterogeneity—this idea where the HER2-positive cells are separated by HER2 normal-looking cells—had a higher association of having a locoregional progression than those patients who were in the T-DM1 arm with no progression. It’s a recurring theme. We saw this in EMILIA and TDM4450g. T-DM1 requires expression of the target to get the chemotherapy payload in, and if you have heterogeneity or low expression, you might need cytotoxic chemotherapy to get that response.
Joyce A. O’Shaughnessy, MD: There’s nothing right now that we need to look for specifically in practice, because the standard of care still remains chemotherapy—trastuzumab and pertuzumab—for the higher-risk preoperative patients, right? What do you think? Are there any clinical implications of the heterogeneity story, Aditya?
Aditya Bardia, MD, MPH: The heterogeneity would be an important piece as we think of antibody drug conjugates that targeted just HER2 or if we develop regimens that are just targeted to HER2, be it doublet therapies, or even triplet therapies, with anti-HER2 agents. If there’s no chemotherapy backbone, and everything is directed toward HER2, we need to ensure that the tumor is not HER2 heterogeneous because you’re not doing anything to target the HER2-negative cells. Maybe some degree of heterogeneity would be OK. If we think of an antibody drug conjugate, there might be some bystander effect that can take care of the cells that are negative, or with antibodies, some ADCC [antibody-dependent cellular cytotoxicity] The immune system might take care of some of the HER2-negative cells, but that should be minimal. For a homogeneous HER2 tumor, I think antibody drug conjugates or predominantly anti-HER2-directed therapy would be very effective.
We saw interesting data from Otto and the group at Dana-Farber Cancer Institute that looked at neoadjuvant T-DM1, and we participated in that trial. It was patients with stage II and III HER2-positive breast cancer. All of them received neoadjuvant T-DM1 with pertuzumab, and they looked at the PCR [pathologic complete response] rate based on the heterogeneity of HER2. Six biopsies were conducted and 6 regions of the tumor were looked at. If there was intratumor heterogeneity, that essentially would indicate that some cells were HER2-positive and others were not, and intratumor heterogeneity predicted for lack of PCR. In tumors that had intratumor heterogeneity, there was no PCR. If you take that group away, then the degree of PCR in the more homogeneous tumors increased. Moving forward in clinical trials, that might be something that we would consider. Obviously, it’s not ready for clinical practice yet.
Joyce A. O’Shaughnessy, MD: Am I right? About 10% were heterogeneous in the Otto Metzger Filho data. The reason it’s interesting is, it raised the question—I have a patient right now in whom I’m dealing with this. When you have one of these heterogeneous tumors when they are 2 distinct populations and with the multiple biopsies that were done, sometimes people have multifocal disease. Sometimes one will be HER2-positive and one will be HER2-negative; consider repeating the HER2 status on the residual disease. If you have eradicated the HER2 but they have residual disease, perhaps use AC [doxorubicin/cyclophosphamide] chemotherapy instead of T-DM1. It’s interesting. I know this is a relatively rare phenomenon. I don’t see it very often, but it’s a possible implication of the fact. It’s mostly the ER-positives who have lesser HER2 amplification levels and lesser expression.
Debu Tripathy, MD: That’s right.
Sara A. Hurvitz, MD: It does take up a whole lot of time and tumor board discussion, so it feels like it’s at least 2 or 3 cases a month in which we’re talking about this.
Joyce A. O’Shaughnessy, MD: Really.
Sara A. Hurvitz, MD: We check all patients’ residual disease for biomarkers again, and it seems invariable that ER goes up and HER2 goes down. In some of these cases, we’ll see that interplay there. Do we then intensely treat the ER-positivity and really go full throttle with the endocrine therapy, if they’re young, with a GnRH [gonadotropin-releasing hormone] analog, but make sure that we’re suppressing both pathways? I think this is an area that will be very rich for research looking at these residual cases, like the KATHERINE study design was, because this is an area where we can interrogate a lot of these very important questions.
Debu Tripathy, MD: In the studies that have looked at conversion to negative, 2 of the 3 of them have shown worse outcome.
Sara A. Hurvitz, MD: Absolutely.
Debu Tripathy, MD: It is something that needs to be studied more? We’re always stuck with what to do when you’re trying to practice ahead of the data. I don’t think there’s a right answer for that, Joyce. Like tumor board goes, sometimes you have to play out of the boundaries.
Transcript Edited for Clarity