Use of MRI in assessing response to neoadjuvant therapy in patients with HER2-positive breast cancer: An interview with Dr. Gabe Sonke

The relationship between a radiologic complete response (rCR) and pathologic complete response (pCR) in patients with HER2-positive breast cancer undergoing neoadjuvant therapy, if any, has not been established. The pCR is a clinically meaningful outcome measure in HER2-positive breast cancer patients receiving neoadjuvant chemotherapy, in addition to the traditional measures (recurrence rate, event-free survival, and overall survival). Thus, the current study, which was published April 21 in the journal Breast Cancer Research and Treatment, compared the rCR and pCR in patients with stage II-III HER2+ breast cancer who were treated with neoadjuvant trastuzumab-based chemotherapy. The study showed that the rCR correlates well with pCR. Although periodic clinical assessments are performed on-treatment to assure response to treatment, no protocol for imaging or imaging modality has been universally accepted, and the correlation between clinical and pathologic responses is poor.

Although it is aggressive, HER2-positive breast cancer has increased sensitivity to chemotherapy, especially when the signaling pathways involved in HER2 amplification are blocked. Indeed, pCR is achieved in a higher percentage of HER2-positive breast cancer patients than patients with other subtypes of breast cancer. 

Two hundred ninety-seven HER2-positive tumors in 296 patients were studied; 154 and 143 tumors were hormone receptor (HR)-negative and –positive, respectively. Two hundred six of 297 tumors were shown to have rCR (69%) compared with pCR in 181 of 297 tumors (61%). Of 206 patients with rCR, 150 had pCR (negative predictive value [NPV] = 73%). Of 91 patients without rCR, 60 had residual tumor (positive predictive value [PPV] = 66%). The NPV was 88% in HR-negative tumors compared with 57% in HR-positive tumors, while the PPV was 50% in HR-negative tumors compared with 78% in HR-positive tumors.  The 5-year recurrence-free survival was 88% and 68% in patients with and without a rCR, respectively.

MDLinx editors spoke to study author Gabe S. Sonke, MD, PhD, about the study.

MDLinx: Do you feel the results would have been different if you had separated stage II and III breast cancers?

Dr. Sonke: Although the likelihood of achieving an rCR and a pCR is higher in stage II than stage III breast cancer, we have no evidence that the association between rCR and pCR differs between stage II and III. It is important to realize, however, that we focused on imaging of the breast and did not take response in the axilla into account, which of course is a major difference between stage II and III. We also have to consider the relatively small sample size for subgroup analyses by stage as a possible reason why we did not detect subtle differences.

MDLinx: How do you think rCR will compare with the clinical response?

Dr. Sonke: The clinical response based on physical examination may be used to determine if the tumor shrinks upon treatment exposure and to exclude overt progressive disease. It is inferior to MRI, however, in determining complete disappearance of the tumor as not all complete responses as determined by physical examination will correspond to a rCR on MRI and a pCR. On the other hand, remaining DCIS or treatment-induced fibrosis may be mistaken for residual invasive tumor by palpation.

MDLinx: What effect if any do you feel proliferative breast cancer subtypes would have on the rCR?

Dr. Sonke: It is known that more proliferative tumors, reflected by a higher tumor grade or aggressive breast cancer subtype, are more likely to achieve a pCR (von Minckwitz, et al. Breast Cancer Res Treat. 2011) and therefore a higher rCR rate is expected as well. Additionally, hormone-receptor-positive tumors, which are generally lower proliferative tumors compared with hormone-receptor-negative tumors, show a different enhancement pattern on MRI (diffuse or non-mass) making interpretation of the response more difficult. We observed a difference in rCR rate and its correspondence with pCR according to the hormone-receptor status of the tumor, but not according to tumor grade.

MDLinx: Based on your findings, do you feel a time will come when imaging will be performed to determine responsiveness prior to completion of 6 to 8 cycles of neoadjuvant chemotherapy?

Dr. Sonke: It has been shown that chemotherapy plus dual HER2-blockade can lead to pCR in a subset of patients after a short course of treatment, even without the use of chemotherapy (Bundred, et al. EBCC10. 2016, Gianni, et al. Lancet Oncol. 2012). Based on these findings and the results of our present study, we have set up a prospective clinical trial to test whether de-escalating the number of treatment cycles based on early MRI evaluation during neoadjuvant treatment is feasible.

MDLinx: Did you find anything unusual or unexpected in your study?

Dr. Sonke: The clear association between rCR and pCR was in line with our expectation. The strong prognostic value of rCR considering long-term outcome was more of a surprise. This finding needs to be confirmed in independent cohorts, with longer follow-up and larger sample size.

The current study has shown the potential role for MRI and the rCR in patients with HER2-positive breast cancer during neoadjuvant chemotherapy. It is hoped that the early identification of patients with a rCR could lead to a reliable means to limit the number of chemotherapy cycles administered based on individual response.

Gabe S. Sonke, MD, PhD, is a medical oncologist in the Department of Medical Oncology at the Netherlands Cancer Institute in Amsterdam, The Netherlands.