Baseline ctDNA Is Associated With Larger Pathological Tumors in HR+ Early Breast Cancer

Patients with primary hormone receptor (HR)–positive early breast cancer who had tested positive for baseline circulating tumor DNA (ctDNA) using a ultra-sensitive, tissue-free test were associated with larger pathological tumor size, according to data from the investigational phase 2 PELOPS trial (NCT02764541) shared at the 2024 San Antonio Breast Cancer Symposium. Notably, the tissue-free ctDNA can also detect a substantial proportion of patients evaluated in this study.

The presence of ctDNA at baseline in the patient population was linked with a larger pathological tumor size and higher residual cancer burden (RCB) scores after neoadjuvant endocrine therapy. Persistent ctDNA after neoadjuvant endocrine therapy was associated with higher rates of recurrence, Albert Grinshpun, MD, MSc and colleagues noted in the poster presentation.

“This study is the first investigation into the dynamics of ctDNA during neoadjuvant endocrine therapy using the Guardant Infinity platform,” Albert Grinshpun and colleagues wrote in their poster. “Our findings suggest that ctDNA has the potential to provide valuable insights into tumor burden, sensitivity to endocrine therapy and the emergence of endocrine resistance mutations.”

Grinshpun is the Goldfarb Advanced Breast Cancer fellow at the Breast Oncology Center at Dana-Farber Cancer Institute in Boston.

PELOPS Trial Design and Baseline Patient Characteristics

The phase 2 PELOPS trial included 49 patients with stages I-III HR–positive, HER2–negative breast cancer and were randomized to receive either neoadjuvant endocrine therapy plus palbociclib (Ibrance) or neoadjuvant endocrine therapy alone. Of note, neoadjuvant endocrine therapy with or without palbociclib was administered for 6 months until surgery. Plasma samples were also collected at baseline and presurgery. A total of 98 plasma samples from 53 patients were analyzed with successful sequencing achieved in 94.9% (93 of 98 samples) of samples. At baseline and presurgery, ctDNA detection rates were 38.8% (19 of 49) and 13.6% (6 of 44), respectively.

Cell-free DNA (cfDNA) was analyzed via Guardant Reveal powered by Infinity, a tissue-free epigenomic assay to assess cfDNA for present tumor-derived methylation signatures for minimal residual disease detection and tumor fraction (TF). Exploratory objectives in the study included samples with detectable ctDNA by methylation that were genotyped using Guardant360 on the Infinity platform. A categorical group comparison was performed using Fisher’s exact test and the Mann-Whitney and Wilcoxon tests to evaluate TF and ctDNA dynamics.

Baseline ctDNA detection using binary detection methods was demonstrated to be higher in patients with RCB scores of 3, compared with patients with RCB scores of 1 or 2 (P = 0.005). At the time of presurgery, ctDNA was only detected in patients with stage III invasive lobular cancer (ILC). At baseline, higher TF was associated with pathological stage at surgery (median, 0.049% for stage ypIII disease vs 0% for stages ypI or II; P = 0.0006). Higher RCB scores after neoadjuvant endocrine therapy were also associated with pathological stage at surgery (median, 0.049% for RBC scores of 3 vs 0% for RCB scores of 1 or 2; P = 0.0014).

The study’s main objective was to evaluate whether baseline ctDNA and dynamic changes during neoadjuvant endocrine therapy as a biomarker could help guide systemic treatment decisions, Grinshpun and colleagues noted on the poster.

Regarding patient characteristics at baseline, 49 patients were tested in total. Invasive ductal carcinoma (IDC) histology was present in 28.6% (n = 14), ILC in 61.2% (n = 30), and both or other in 10.2% (n = 5). Patients with stage I cancer included 8.2% of patients (n = 4), stage II in 73.5% (n = 36), stage III in 14.3% (n = 7), and unknown stage in 4.1% (n = 2). Pathological stage of ypI was shown in 12.2% (n = 6), ypII in 40.8% (n = 20), ypIII in 36.7% (n = 18), and unknown in 10.2% (n = 5). RCB scores of 1 were observed in 6.1% (n = 3), scores of 2 in 49.0% (n = 24), scores of 3 in 36.7% (n = 18), and unknown scores in 8.2% (n = 4). In the two treatment arms, 71.4% of patients (n = 35) received endocrine therapy plus palbociclib, 26.5% (n = 13) received endocrine therapy alone, and 2.0% (n = 1) received unknown treatment.

ctDNA detection was shown in 19 patients at baseline. Invasive ductal carcinoma (IDC) histology was present in 21.4% (n = 3), ILC in 50.0% (n = 15), and both or other in 20.0% (n = 1). Patients with stage I cancer included 25.0% of patients (n = 1), stage II in 38.9% (n = 14), and stage III in 57.1% (n = 4). Pathological stage of ypI was shown in 16.7% (n = 1), ypII in 20.0% (n = 4), ypIII in 66.7% (n = 12), and unknown in 40.0% (n = 2). RCB scores of 1 were observed in 0% (n = 0), scores of 2 in 20.8% (n = 5), scores of 3 in 61.1% (n = 11), and unknown scores in 75.0% (n = 3). In the two treatment arms, 34.3% of patients (n = 12) received endocrine therapy plus palbociclib, 37.5% (n = 6) received endocrine therapy alone, and 100% (n = 1) received unknown treatment.

At presurgery, 44 patients were tested in total. Invasive ductal carcinoma (IDC) histology was present in 29.5% (n = 13), ILC in 59.1% (n = 26), and both or other in 11.4% (n = 5). Patients with stage I cancer included 9.1% of patients (n = 4), stage II in 75.0% (n = 33), stage III in 13.6% (n = 6), and unknown in 4.5% (n = 2). Pathological stage of ypI was shown in 13.6% (n = 6), ypII in 40.9% (n = 18), ypIII in 40.9% (n = 18), and unknown in 4.5% (n = 2). RCB scores of 1 were observed in 6.8% (n = 3), scores of 2 in 45.5% (n = 20), scores of 3 in 43.2% (n = 19), and unknown scores in 4.5% (n = 2). In the two treatment arms, 70.5% of patients (n = 31) received endocrine therapy plus palbociclib, and 29.5% (n = 13) received endocrine therapy alone.

At presurgery, 6 patients were demonstrated ctDNA detection. ILC histology was present in 23.1% (n = 6) of patients. No patients had stage I cancer, stage II was shown in 14.7% (n = 5), and stage III in 16.7% (n = 1). Pathological stage of ypIII was shown in 33.3% (n = 6). RCB scores of 3 were shown in 31.6% (n = 6). In the two treatment arms, 9.7% of patients (n = 3) received endocrine therapy plus palbociclib, and 21.4% (n = 3) received endocrine therapy alone.

Additional ctDNA Detection Data

At baseline, ctDNA positivity was shown in 19% of patients and ctDNA negativity was shown in 30% of patients. Before patients underwent surgery and after treatment, ctDNA positivity was shown in 6% of patients and ctDNA negativity was shown in 38% of patients.

Undetectable ctDNA at baseline remained in most patients (65%; n = 26/40) with undetectable ctDNA after receiving neoadjuvant endocrine therapy. Specifically, 14.6% (n = 6/41) with lobular breast cancer and an RCB score of 3 had present ctDNA at baseline and presurgery. Among these 6 patients, 67% (n = 4) developed metastatic breast cancer recurrence within 3 years of surgery. Tumor fraction levels were reported to be lower before surgery compared with baseline in these respective patients (P = 0.0002). No patients became ctDNA positive at presurgery and 20% (n = 8/40) patients experienced ctDNA clearances from baseline to presurgery.

Notably, somatic variants were identified in 26% (n = 5/19) of baseline and 33% (n = 2/6) of presurgery ctDNA detected samples, respectively, including a common ESR1 mutation at presurgery. At baseline, mutated genes that were present included AKT1 E17K, MAP3K1 fs, PIK3CA E545K, PIK3CA E542K, RHOA G17E, and TERT. Presurgery, detectable somatic alterations in ctDNA included ATM (possible CHIP) R337C, BRCA2 fs, and ESR1 D539G.

Among 40 patients, ctDNA dynamics were categorized by persistent or emergent ctDNA, cleared ctDNA, or no ctDNA. At baseline and presurgery, ctDNA was persistent or emergent in 15% (n = 6) and 66.7% (n = 4) at 3-year distant recurrence. Of note, no patients in the study became ctDNA positive before surgery and 20% of patients (n = 8) experienced ctDNA clearance from baseline to before surgery. At baseline and before surgery, no ctDNA was reported in 65.0% (n = 26) and in 3.8% (n = 1) at 3-year distant recurrence.

“These results are limited by the sample size but merit further investigation of the role of ctDNA as a tool to predict sensitivity to endocrine therapy and long-term outcomes,” Grinshpun and colleagues concluded. “Such a tool could be valuable in tailoring surgical and systemic treatment approaches, particularly given the ongoing development of novel endocrine therapies and the addition of CDK4/6 inhibitors in the adjuvant setting.”

Reference

Grinshpun A, Dustin D, Cai M, et al. PS9-08: Ultra-sensitive detection of circulating tumor DNA (ctDNA) in patients undergoing neoadjuvant endocrine therapy for hormone receptor-positive (HR+) early breast cancer. Presented at: 2024 San Antonio Breast Cancer Symposium; December 10-13, 2024; San Antonio, TX. Abstract PS9-08.