The third General Session at the 2024 San Antonio Breast Cancer Symposium® included presentations on a risk-adapted treatment approach based on circulating tumor DNA (ctDNA)-based detection of minimal residual disease (MRD), assessment of anthracycline benefit in genomic risk-stratified groups, and a novel integrated clinicopathologic and genomic feature-based machine learning model (MLM) for improving CDK4/6 inhibitor treatment response prediction.

A recording of the session, held Friday, December 13, is available on demand for registered 2024 SABCS® participants through March 31, 2025, on the virtual meeting platform.

GS3-01: Circulating tumor DNA surveillance in ZEST, a randomized, phase III, double-blind study of niraparib or placebo in patients w/ triple-negative breast cancer or HER2+ BRCA-mutated breast cancer with molecular residual disease after definitive therapy.

The ZEST phase III clinical trial, designed to evaluate the efficacy of the poly (ADP-ribose) polymerase (PARP) inhibitor niraparib in preventing breast cancer recurrence in patients with ctDNA positivity, failed to accrue enough ctDNA-positive patients, according to results presented by Nicholas Turner, MD, PhD, Director of Clinical Research and Development at The Royal Marsden Hospital and Institute of Cancer Research in London.

“There is no standard of care available to detect MRD and direct future therapy to prevent or defer relapse,” Dr. Turner said.

ctDNA in blood, associated with high relapse risk, may help identify MRD-positive patients who may benefit from additional treatment.

Patients with stage I-III triple-negative or BRCA-mutated, hormone receptor (HR)-positive human epidermal growth factor 2 (HER2)-negative breast cancer who had completed their recommended treatment were eligible for ZEST. The patients also had to have detectable ctDNA without evidence of radiographic recurrence.

Overall, the proportion of ctDNA-positive patients was low (7.7% of the 1,901 patients tested). Patients with detectable ctDNA were more likely to have positive lymph nodes, larger tumors, stage III disease, and residual disease after neoadjuvant therapy, compared to ctDNA-negative patients.

The study was terminated after random assignment of only 40 patients, 18 to niraparib and 22 to placebo. Although the low enrollment and early termination precludes definitive conclusions, recurrence-free survival was numerically longer with niraparib.

Dr. Turned noted that the low randomization rate likely reflects the broad entry criteria, which included low-risk patients who are more likely to have low rates of ctDNA detection in the cohort.

ZEST provides insights on improving the design of future MRD-directed studies, including: selecting relatively high-risk patients and early ctDNA testing respective to treatment completion.

GS3-03: Impact of Anthracyclines in High Genomic Risk Node-Negative HR+/HER2- Breast Cancer

Addition of anthracyclines to taxane-based chemotherapy yielded better outcomes, compared with adjuvant taxane-based regimens alone, for patients with early-stage, node-negative, HR-positive, HER2-negative breast cancer who have a high risk of recurrence based on a recurrence score (RS) ≥31 with the OncotypeDX test, according to post hoc analysis of data from the TAILORx trial shared at the 2024 SABCS^®.

Nan Chen, MD, Assistant Professor of Internal Medicine at the University of Chicago Medicine, presented the study.

“We noted that this benefit is primarily limited to tumors greater than 2 cm in size,” she said, “and this benefit increases with increasing RS above 31,” while no trend toward benefit was seen in patients with a score between 26 and 30.

“In previous studies, RS [cutoffs] of 26 and 31 have been used to evaluate the benefit of chemotherapy. In this analysis, we chose the RS cutoff of 31 to best isolate the patients at the highest genomic risk,” Dr. Chen continued.

After controlling for age, grade, tumor size, and HR status, the use of taxane with anthracycline/cyclophosphamide and similar regimens (T-AC) in patients with an RS ≥31 and tumors ≥2 cm was associated with improved survival outcomes after five years. Additional benefits with T-AC, compared to taxane with cyclophosphamide, included higher distant recurrence-free interval, distant recurrence-free survival, and recurrence-free interval, along with a trend towards improved overall survival at five years.

Based on these data, “anthracycline should be considered in patients with high genomic-risk HR-positive, HER2-negative lymph node-negative disease,” Dr. Chen said.

GS3-09: Multimodal integration of real world clinical and genomic data for the prediction of CDK4/6 inhibitors outcomes in patients with HR+/HER2- metastatic breast cancer

A machine learning model that integrated both clinical and genomic factors outperformed models based solely on clinical features or genomic data in predicting improved outcomes with addition of CDK4/6 inhibitors to endocrine therapy as first-line treatment for patients with HR-positive, HER2-negative metastatic breast cancer, according to results shared by Pedram Razavi, MD, PhD, Scientific Director of the Global Research Program at Memorial Sloan Kettering Cancer Center.

Combination therapy with CDK4/6 inhibitors and endocrine therapy has significantly improved outcomes in patients with HR-positive, HER2-negative metastatic breast cancer.

“But there is extensive heterogeneity in the outcomes among patients,” ranging from a robust response to a complete lack of response to development of resistance over time, Dr. Razavi noted.

Given the wealth of genomic information regarding the mechanisms of resistance to CDK4/6 inhibitors, Dr. Razavi and colleagues sought to develop strategies for improving prediction of response by integrating both traditional clinicopathologic features and genomic profiles. They generated three MLMs — based on clinicopathological features (CF), genomic features (GF), or both (CGF) — to predict progression-free survival (PFS) with CDK4/6 inhibitors, using a training cohort of 761 patients. The performance of the MLMs were also evaluated in a test cohort of 326 patients.

The CGF model performed better, identifying four risk groups characterized by significant separation of PFS outcomes, compared to the GF and CF models, which identified three risk groups based on PFS outcomes. The integrated model proved robust, performing with comparable precision in the training and test data sets.

Dr. Razavi said that the next steps would be to validate the MLMs in external cohorts to confirm their generalizability and robustness.

“We are committed to developing a practical single-patient outcome predictor tool based on maximum parsimony, so that we have the minimum [required] amount of clinical and genomic data included, and we hope that we can validate that as well,” he concluded.

---

Following are additional abstracts presented during Friday’s General Session 3:

GS3-04: (Neo)adjuvant nab-PAC weekly vs. sb-PAC q2w, followed by EC q2w, in genomically or clinically high-risk HR+/HER- early breast cancer according to ET-response: final survival results from the WSG ADAPT-HR+/HER2- chemotherapy-trial.

GS3-05: NSABP B-59/GBG-96-GeparDouze: A randomized double-blind phase III clinical trial of neoadjuvant chemotherapy with atezolizumab or placebo followed by adjuvant atezolizumab or placebo in patients with Stage II and III triple-negative breast cancer.

GS3-06: Neoadjuvant camrelizumab plus chemotherapy (chemo) for early or locally advanced triple-negative breast cancer (TNBC): a randomized, double-blind, phase III trial.

GS3-08: In situ detection of individual classical MHC-I gene products in breast cancer identifies gene- and subtype-specific biased antigen presentation loss.

GS3-10: Paired DNA and RNA analysis of CALGB 40603 (Alliance) reveals insights into the molecular and prognostic landscape of stage II-III triple-negative breast cancer.