De-escalation therapy has become an important trend in the management of HER2-positive early breast cancer, yet accurately identifying patients who can safely omit chemotherapy remains a major challenge. At the 2025 San Antonio Breast Cancer Symposium (SABCS), the PHERGuide substudy innovatively incorporated circulating tumor DNA (ctDNA) monitoring into the neoadjuvant treatment framework. The study demonstrated that ctDNA clearance is strongly associated with pathological complete response (pCR), while baseline ctDNA positivity predicts poorer outcomes.These findings provide critical molecular evidence supporting individualized de-escalation strategies in HER2-positive early breast cancer and promote a shift toward more precise and dynamic treatment decision-making. Oncology Frontier invited Professor Wang Xin from the National Cancer Center / Cancer Hospital, Chinese Academy of Medical Sciences to introduce and comment on this important study.
Study Overview
Study Title
Circulating Tumor DNA (ctDNA) in Human Epidermal Growth Factor Receptor 2–Positive (HER2+) Early Breast Cancer: A Translational Analysis of the PHERGain Neoadjuvant Individualized Treatment Study
Background
HER2-targeted therapies have dramatically improved outcomes in patients with HER2-positive early breast cancer (EBC), driving growing interest in treatment de-escalation strategies. ctDNA has emerged as a promising tool for risk stratification and real-time disease monitoring, offering new opportunities for individualized treatment decisions.
The PHERGain trial demonstrated that, among patients with stage I–IIIA HER2+ EBC treated with neoadjuvant trastuzumab plus pertuzumab (HP) dual HER2 blockade, a FDG-PET–guided, pCR-adapted strategy allowed chemotherapy (CT) to be safely omitted in selected patients. Using this adaptive approach, 37.9% of patients avoided chemotherapy, while achieving a 3-year invasive disease-free survival (iDFS) rate of 94.8% in the adaptive treatment group.
The PHERGuide substudy aimed to further refine prediction accuracy by evaluating a tumor-agnostic epigenomic ctDNA assay for minimal residual disease detection. The goal was to improve prediction of pCR and 3-year iDFS, thereby enabling more personalized (neo)adjuvant treatment strategies within the PHERGain framework.
Methods
Among 356 randomized patients, 63 patients in arm A (standard treatment) and 267 patients in arm B (adaptive treatment) underwent surgery.
The primary objective of the PHERGuide substudy was to evaluate the association between ctDNA clearance after two cycles of treatment and pCR (ypT0/is ypN0) in all enrolled patients. Secondary objectives included assessing the relationship between ctDNA levels and patient prognosis.
Blood samples were collected at three time points:
- Baseline
- After two cycles of neoadjuvant therapy
- Preoperatively
A total of 932 plasma samples (336 baseline, 311 after cycle 2, and 285 preoperative) from 356 patients were analyzed using Guardant Reveal™, a tumor-agnostic epigenomic ctDNA assay that provides both binary ctDNA detection and estimated tumor fraction.
ctDNA clearance was defined as detectable ctDNA at baseline followed by undetectable ctDNA at subsequent time points. Categorical variables were analyzed using logistic regression, and survival outcomes were evaluated using Cox proportional hazards models.
Results
Of the 932 samples analyzed, 801 met PHERGuide eligibility criteria (161 from arm A and 640 from arm B).
Among 288 evaluable baseline samples, ctDNA was detected in 204 patients (71%). Detection rates were strongly associated with disease stage:
- Stage I: 33.3% (9/27)
- Stage II: 71% (154/217)
- Stage III: 93% (41/44) (P < 0.001)
Among patients with detectable baseline ctDNA:
- 75.4% (126/167) achieved ctDNA clearance after two treatment cycles
- 83.2% (124/149) achieved ctDNA clearance prior to surgery
ctDNA clearance after two cycles (P = 0.003) and preoperative ctDNA clearance (P < 0.001) were both significantly associated with achieving pCR. No significant association was observed between baseline ctDNA status and pCR (P = 0.583).
Notably, none of the patients with detectable ctDNA preoperatively (n = 25) achieved pCR.
Baseline ctDNA positivity was associated with worse 3-year iDFS:
- ctDNA-positive: 92.5%
- ctDNA-negative: 100% (HR 0.20; 95% CI: 0.02–0.98; P = 0.046)
Conclusions
These findings demonstrate a strong association between ctDNA clearance and pCR in HER2+ EBC patients receiving neoadjuvant HER2-targeted therapy. In addition, detectable baseline ctDNA is associated with poorer 3-year outcomes. Further prospective studies are needed to confirm these results and to define ctDNA-guided treatment algorithms.
Expert Commentary
In recent years, HER2-positive early breast cancer has become a major focus of treatment de-escalation research. With the widespread adoption of dual HER2-targeted therapy, pCR rates and overall survival have improved substantially, creating opportunities to safely omit or reduce chemotherapy in selected patients.
The PHERGain trial demonstrated that a FDG-PET–guided individualized treatment strategy allowed approximately 37.9% of patients to forgo chemotherapy while maintaining excellent 3-year iDFS. Building on this foundation, the PHERGuide substudy incorporated ctDNA analysis, exploring its molecular value in predicting treatment response and risk stratification, and providing a new biological rationale for precision de-escalation strategies in HER2-positive early breast cancer.
The most important contribution of PHERGuide is that it clearly establishes ctDNA as more than an ancillary biomarker—it is emerging as a central molecular node in the precision treatment pathway of HER2+ EBC. In the context of highly effective HER2-targeted therapy, traditional response assessments based solely on imaging or pathology may no longer be sufficient for guiding chemotherapy omission. ctDNA offers a true systemic biological readout of treatment response.
First, this study reinforces a critical concept: dynamic molecular response is more informative than single time-point tumor burden assessment. In HER2-positive disease, ctDNA dynamics more accurately reflect the underlying biological response to therapy than changes in tumor size or metabolic activity alone. This positions the combination of PET response and ctDNA clearance as a potentially powerful predictive strategy.
Second, the use of a tumor-agnostic epigenomic ctDNA assay enhances feasibility and scalability. By avoiding reliance on tumor tissue and lengthy sequencing workflows, this approach facilitates integration of minimal residual disease monitoring into routine clinical decision-making.
Nevertheless, challenges remain. Thresholds for ctDNA-guided decision-making have not yet been clearly defined, and optimal treatment escalation strategies for ctDNA-positive patients lack prospective validation. Furthermore, how ctDNA, PET response, and pathological findings should be hierarchically integrated into a unified clinical algorithm requires confirmation in future large-scale trials.
Overall, the PHERGuide substudy represents more than a supplementary analysis—it provides a critical biological anchor in the evolving roadmap of precision de-escalation therapy for HER2-positive early breast cancer. It paves the way toward a multidimensional, integrated assessment framework that may more accurately identify patients suitable for de-escalation, ultimately advancing personalized treatment strategies and the practical implementation of precision medicine in this field.
Professor Wang Xin
