ctDNA Refines Risk in Neoadjuvant-Resistant Breast Cancer

Data from the I-SPY2 trial (NCT01042379), published in Nature Communications, demonstrate that liquid biopsy can refine risk stratification for patients who have significant residual cancer burden (RCB) following neoadjuvant therapy (NAT). The study suggests that patients with resistant tumors who remain ctDNA-negative achieve survival outcomes comparable to those with a pathologic complete response, potentially opening avenues for treatment de-escalation.¹

"Refining risk stratification for NAT-resistant tumors using ctDNA and understanding ctDNA shedding in these tumors could guide treatment decisions to prevent or delay metastatic recurrence," wrote lead author Mark Jesus M. Magbanua, PhD, of the University of California, San Francisco, and colleagues.

Refining Prognosis in Resistant Disease

Pathologic complete response (pCR), or RCB-0, is a well-established predictor of favorable long-term outcomes, particularly in HER2-positive and triple-negative breast cancer (TNBC). However, approximately 50% of patients exhibit resistance to NAT, defined as moderate (RCB-II) or extensive (RCB-III) residual disease.

Current clinical staging poorly predicts outcomes for this resistant group; only 15% to 30% of patients with NAT-resistant tumors experience metastatic recurrence within 3 years. To address this heterogeneity, researchers analyzed plasma samples from 723 patients with high-risk early-stage breast cancer enrolled in the multicenter I-SPY2 platform trial. They utilized the Signatera test, a tumor-informed, personalized ctDNA assay, to detect patient-specific variants at pretreatment (T0), 3 weeks into treatment (T1), 12 weeks into treatment (T2), and post-NAT prior to surgery (T3).

The analysis revealed that ctDNA status is a potent independent predictor of metastatic recurrence. Among patients with NAT-resistant tumors (RCB-II and RCB-III), those who were ctDNA-negative post-NAT had significantly better 3-year distant recurrence-free survival (DRFS) compared with those who were ctDNA-positive:

• RCB-II: 88% DRFS for ctDNA-negative vs 57% for ctDNA-positive (adjusted HR, 0.29; P =.001)
• RCB-III: 83% DRFS for ctDNA-negative vs 22% for ctDNA-positive (adjusted HR, 0.14; P <.001)

Furthermore, patients with resistant tumors who cleared ctDNA early (by week 3 or week 12) had significantly higher 3-year DRFS rates (91% and 92%, respectively) compared with those with no clearance (35%).

Early Prediction of Response

The study also evaluated the utility of ctDNA dynamics in predicting therapeutic response. Clearance of ctDNA as early as week 3 (T1) was significantly associated with achieving a favorable response (RCB-0/I) across all receptor subtypes.

This predictive capability held true for patients treated with targeted therapies and immunotherapies. In patients with HER2-positive disease treated with paclitaxel plus trastuzumab (Herceptin) and pertuzumab (Perjeta), early ctDNA clearance was associated with a lower median RCB score (P <.001). Similar results were observed in patients with HR-positive/HER2-negative and TNBC subtypes treated with immune checkpoint inhibitors, such as pembrolizumab (Keytruda) or durvalumab (Imfinzi).

Biological Implications

Notably, less than 20% of patients with NAT-resistant tumors were ctDNA-positive post-NAT, despite having a high tumor burden. To investigate whether this was due to assay limitations or changes in tumor mutational landscape, the investigators performed serial mutation profiling on matched tumor tissues.

They found that the patient-specific variants tracked by the ctDNA assay were conserved in the residual tumor tissue in 94% of paired samples¹. This suggests that the absence of detectable ctDNA in resistant tumors is likely due to low shedding, which may characterize a biological phenotype with lower metastatic potential.

Clinical Relevance

These findings support the integration of liquid biopsy into neoadjuvant treatment paradigms. For patients with RCB-II/III who are ctDNA-positive, the high risk of recurrence warrants consideration for treatment escalation. Conversely, the lack of ctDNA detection in resistant tumors appears linked to less aggressive biology, potentially allowing for de-escalation strategies to spare patients from unnecessary toxicity.

"The goal is to combine ctDNA with imaging and pathology assessment... to improve early response prediction to NAT," the authors concluded.

About the Study

The I-SPY2 trial is an interventional phase 2 study designed to advance personalized medicine by identifying early markers of response before surgery through MRI and tissue and blood samples. The study is assessing the efficacy of novel drugs in sequence with standard chemotherapy. Treatment strategies are dropped from the study if they show a low probability of improved efficacy in any biomarker signature, while new drugs will continue to enter the study.²

The study plans to enroll 5000 patients with stage II to III breast cancer and is currently recruiting at UCSF, the University of Southern California, and City of Hope.

REFERENCES

1. Magbanua MJM, Manon NA, Wolf DM, et al. Circulating tumor DNA refines risk stratification of neoadjuvant therapy-resistant breast tumors. Nat Commun. 2025;16:9945. doi:10.1038/s41467-025-65432-5

2. I-SPY TRIAL: Neoadjuvant and Personalized Adaptive Novel Agents to Treat Breast Cancer. USCF Clinical Trials. Accessed December 18, 2025. https://clinicaltrials.ucsf.edu/trial/NCT01042379