Supplements and Featured Publications
Post-surgery ctDNA status was prognostic for distant recurrence in early TNBC, and the prognostic value of ctDNA status per WGS outperformed pCR status.
Circulating tumor DNA (ctDNA) status following surgery was highly prognostic for distant recurrence in patients with early triple-negative breast cancer (TNBC), and this status—measured using a whole-genome sequencing (WGS) assay—outperformed pathologic complete response (pCR) status for prognostic risk stratification, according to findings from a prospective sub-study of the phase 2/3 PARTNER trial (NCT03150576).1
Findings presented at the
At a median follow-up of 5.0 years (range, 0.3-7.0), the median lead time for distant recurrence by the WGS assay was 1.05 years (95% CI, 0.26-2.50). With the WGS assay, the 3-year distant recurrence–free interval (DRFI) rate was 98.0% (95% CI, 86.4%-99.7%) in patients with ctDNA-negative disease (n = 52) vs 39.7% (95% CI, 14.8%-64.0%) in patients with ctDNA-positive disease (n = 14). With the WES assay, the 3-year DRFI rates in these respective populations were 90.3% (95% CI, 78.2%-95.9%; n = 57) vs 30.0% (95% CI, 1.2%-71.9%; n = 5). Although both assays were significantly prognostic for DRFI, the WGS assay showed an approximately 3-fold higher hazard ratio for distant recurrence (HR, 52.1; 95% CI, 5.15-414.4; P = .0002) compared with the WES assay (HR, 15.2; 95% CI, 3.5-66.7; P .0003).
At the mid-neoadjuvant chemotherapy time point, among 67 WES-evaluable samples and 68 WGS-evaluable samples, WES detected 1 ctDNA-positive sample, whereas WGS detected 11. Patients with ctDNA-positive results at this time point per WGS had a significantly lower 3-year DRFI rate vs those with negative results, at 40.9% (95% CI, 11.7%-68.9%) vs 94.3% (95% CI, 83.4%-98.1%), respectively (HR, 11.1; 95% CI, 3.3-37.6; P = .0001).
Although pCR status was significantly prognostic for DRFI (HR, 8.5; 95% CI, 1.9-38.4; P = .0055), ctDNA status 2 to 4 weeks after surgery was stronger prognostically. A post-surgical multivariable analysis showed that only ctDNA status (HR, 35.4; 95% CI, 3.7-339.4; P = .0020) and not pCR (HR, 2.3; 95% CI, 0.2-21.6; P = .4797), remained significantly predictive for DRFI.
“By tracking up to 5000 variants, the WGS assay achieves higher ctDNA detection rates than the WES assay at all time points through ultrasensitive detection, despite cell-free DNA [cfDNA] levels as low as 10 ng,” lead study author Andrew Dooley, BMBCh, MRCP, of the Precision Breast Cancer Institute in Cambridge, United Kingdom, and colleagues wrote in a poster of the data.
“Detection of ctDNA after treatment for early-stage TNBC is an indicator of MRD and is associated with a high risk of recurrence,” the authors wrote. “ctDNA clearance during neoadjuvant chemotherapy is associated with pCR and can provide prognostic value beyond pCR status alone.”
The PARTNER trial is a randomized, prospective, controlled trial investigating neoadjuvant carboplatin plus paclitaxel with or without olaparib (Lynparza) followed by anthracycline-based therapy in patients with basal-like, BRCA1/2 wild-type early breast cancer. Among patients who received neoadjuvant carboplatin plus paclitaxel with or without olaparib followed by anthracycline-based therapy, the pCR rate was 51.1%; this rate was 52.4% among patients in the control arm, who received the regimen without olaparib (difference, –1.3; 95% CI, –9.7 to 7.0; P = .753).2
The present sub-study compared the performance and prognostic value for DRFI of 2 tumor-informed MRD assays: a WES-based assay that tracks a maximum of 200 single-nucleotide variants (SNVs), and a minor allele enriched sequencing through recognition oligonucleotides WGS-based assay that tracks a maximum of 5000 SNVs.1
In total, 90% of patients with early TNBC who were included in this sub-study were those enrolled in the PARTNER study. All patients in the sub-study were enrolled to the Personalized Breast Cancer Programme whole-genome, whole-transcriptome profiling study. Blood for ctDNA detection was collected at baseline, twice during neoadjuvant chemotherapy, following neoadjuvant chemotherapy but before surgery, 2 to 4 weeks after surgery, 3 months after surgery, and 12 months after surgery.
The WES and WGS assays were both developed for each patient. Investigators tested plasma samples using both assays to determine ctDNA status (positive vs negative) and ctNDA concentration at each time point.
Among patients in the post-surgery analysis population (n = 66), the mean age was 48.6 years (standard deviation, 10.8). Patients had received control chemotherapy on the PARTNER trial (36.4%), chemotherapy plus olaparib on the PARTNER trial (54.6%), or control therapy not on the PARTNER trial (9.1%). Most patients had grade 3 tumors (89.4%), had not received adjuvant chemotherapy (83.3%), and had negative nodal status (63.6%).
Among 97 tumor samples, 97.9% yielded a valid WGS assay panel design. Additionally, among 477 plasma samples, 96.4% met cfDNA input and quality criteria to yield a valid result. The median WGS assay panel size was 3429 variants (range, 532-5000), and the median WES assay panel size was 156 variants (range, 47-200). The WGS assay had a higher sensitivity and therefore consistently detected more ctDNA-positive samples than the WES assay. Most samples were ctDNA positive at baseline, and surgery generated a decrease in ctDNA positivity detected by both assays. The WGS assay also detected more samples with ctDNA levels below 10 ppm than the WES assay.
Overall, the WGS assay had higher cumulative sensitivity for post-surgery DRFI vs the WES assay. The respective cumulative sensitivity rates at 1, 2, and 3 years were 100.0%, 85.0%, and 88.8% with WGS vs 66.7%, 49.6%, and 35.3% with WES.
In total, 5 patients with WGS ctDNA-positive results 2 to 4 weeks after surgery did not experience distant recurrence during the observation period. Four of these patients had ctDNA levels below 2 ppm. One patient experienced local recurrence at 1.21 years and was censored; this patient and 1 other received adjuvant therapy.
“The WGS assay sensitivity enables detection of low-level ctDNA positivity during neoadjuvant chemotherapy, supporting on-treatment molecular monitoring for treatment optimization,” the authors concluded. “The study provides further evidence for ctDNA status as a potential biomarker to inform therapy decisions. Patients with a ctDNA-positive test could be considered for adjuvant treatment intensification, or more frequent surveillance. Patients with a ctDNA-negative test could be considered for de-escalation strategies.”
