Editor's Note: Treatment for early HER2-positive breast cancer is moving away from a “one-size-fits-all” approach toward dynamic, individualized decision-making guided by treatment response and multidimensional biomarkers. Genomic risk stratification, pathological complete response (pCR), and emerging biomarkers are reshaping strategies for treatment de-escalation and escalation in the neoadjuvant setting.Against this background, at the 2026 Annual Meeting of the Chinese Society of Clinical Oncology (CSCO), Oncology Frontier invited Professor Kun Wang of Guangdong Provincial People’s Hospital and Professor Pier Paolo Berton Giachetti of the European Institute of Oncology to discuss precision risk stratification and dynamic treatment decisions in early HER2-positive breast cancer.
Professor Giachetti explained the value of HER2DX in guiding treatment for HER2-positive breast cancer. Drawing on the neoCARHP study, Professor Wang discussed findings showing that a carboplatin-free THP regimen was noninferior to TCbHP in patients with stage II–III early HER2-positive breast cancer. They also explored the potential for selecting de-escalated treatment according to clinical stage and genomic risk scores. Finally, the two professors discussed emerging technologies and biomarkers beyond pCR and genomic risk scores that may be most likely to change clinical decision-making, offering new perspectives on precision treatment.
Oncology Frontier: Professor Giachetti, you have long been involved in research on genomic risk-stratification tools for early breast cancer. What genomic assays are currently available to predict treatment response and assess prognosis in early HER2-positive breast cancer? How do you view their practical value in clinical decision-making?
Professor Pier Paolo Berton Giachetti: First, I would like to clarify a common misconception. When we talk about genomic assays, some people may think of Oncotype DX, the 21-gene assay, or MammaPrint, the 70-gene assay. However, these tests were specifically developed for patients with HR-positive/HER2-negative breast cancer.
For HER2-positive breast cancer, we have a completely different tool called HER2DX. It is a multiomics-based prognostic model that integrates four biological modules and two clinical variables. The four biological modules are the luminal differentiation score, immune infiltration score, HER2 amplicon expression, and tumor proliferation score. The two clinical variables are tumor size and nodal stage. I believe incorporating these clinical variables into the assay is very important.
HER2DX provides three main outputs. The first is a pCR prediction score, which estimates the probability that a patient will achieve a pathological complete response at surgery. The second is a recurrence risk score, allowing patients to be stratified according to their risk of recurrence. The third is ERBB2 messenger RNA expression, which reflects the degree to which a tumor depends on the HER2 signaling pathway.
HER2DX has undergone extensive evaluation. The DAPHNe study suggested a significant association between the HER2DX pCR score and pCR. The assay has also been validated in approximately 2,000 patients across 11 clinical trials, including the APT and ATEMPT studies.
I therefore consider this tool highly important. For example, when predicting treatment outcomes in patients with small tumors, findings from the APT study led by Professor Sara Tolaney suggest that not all patients with small tumors necessarily require the same course of adjuvant treatment with paclitaxel plus trastuzumab.
In my view, HER2DX may help us identify which tumors warrant different treatment strategies, such as neoadjuvant chemotherapy or treatment escalation in the postoperative setting. The assay clearly has prognostic value, but its predictive value has yet to be established.
An international, multicenter, open-label, randomized controlled phase III trial, Definitive, is currently evaluating the assay. Patients will be randomly assigned to receive either standard treatment or treatment guided by the assay. I hope the results will further establish the predictive value of HER2DX and support its incorporation into clinical guidelines.
Oncology Frontier: Professor Wang, the neoCARHP study you led demonstrated that de-escalating neoadjuvant treatment by omitting carboplatin can significantly reduce toxicity without compromising efficacy in early HER2-positive breast cancer. In light of Professor Giachetti’s discussion of genomic risk stratification, do you think it will be possible to use pretreatment genomic risk scores to predict which patients are suitable for dual-targeted THP therapy and which still require platinum-containing chemotherapy? How feasible is it to translate these scoring tools from research into clinical practice?
Professor Kun Wang: At the 2025 ASCO Annual Meeting, neoCARHP was selected for a late-breaking oral presentation, which I presented in person. The full results were also published in the Journal of Clinical Oncology (JCO) earlier this year.
neoCARHP was a multicenter, open-label, randomized, noninferiority phase III trial conducted at 15 hospitals. It enrolled previously untreated patients aged 18 years or older with stage II–IIIA invasive HER2-positive breast cancer. Patients were randomly assigned to receive treatment with carboplatin (the TCbHP group) or without carboplatin (the THP group).
The results showed that THP was noninferior to TCbHP. The pCR rate was 64.1% with THP and 65.9% with TCbHP.
These findings suggest that neoadjuvant treatment with the de-escalated THP regimen may be sufficient for some patients with early HER2-positive breast cancer. We can therefore use clinical stage to help guide treatment selection. For most patients with stage II disease, de-escalation to THP may be an appropriate option.
Professor Pier Paolo Berton Giachetti also discussed the HER2DX concept in his presentation. Based on their HER2DX pCR scores, patients can be categorized into high-, intermediate-, and low-score groups. Patients in the high- and intermediate-score groups have a high probability of achieving pCR following anti-HER2 therapy, whereas those in the low-score group have a lower probability of achieving pCR.
This suggests that we may be able to use both clinical stage and genomic test results to guide the selection of neoadjuvant treatment strategies.
Oncology Frontier: Treatment for early HER2-positive breast cancer is moving away from a “one-size-fits-all” approach toward dynamic, individualized decisions based on treatment response and multidimensional biomarkers. Beyond pCR and genomic risk scores, which emerging technologies or biomarkers do you believe are most likely to change clinical decision-making and help achieve precision treatment?
Professor Pier Paolo Berton Giachetti: This is an excellent question. So far, the tools we have discussed, including genomic assays and pCR, provide something like a “snapshot” of the tumor. In clinical practice, however, we see tumors changing dynamically throughout the natural course of the disease.
In this context, the biomarker that comes to mind is circulating tumor DNA (ctDNA). It allows us to characterize the tumor as it changes over time. For example, the I-SPY 2 study established an association between ctDNA clearance and better prognosis. I therefore believe ctDNA could be a valuable tool to incorporate into our treatment algorithms.
Professor Kun Wang: We want to monitor changes in tumors dynamically throughout treatment. These changes include both radiological responses and alterations in peripheral blood, such as changes in ctDNA.
In the future, we hope to use these dynamic assessments to make the best possible treatment decisions for patients receiving neoadjuvant therapy for HER2-positive breast cancer. In other words, six treatment cycles do not necessarily have to be the standard for every patient, and THP does not necessarily have to be the standard regimen for everyone.
By assessing radiological changes after two and four cycles, alongside ctDNA changes at the same time points, we may be able to determine whether a patient is suitable for treatment de-escalation or escalation. This represents a promising direction for future development.
Professor Kun Wang
Professor Pier Paolo Berton Giachetti
