Detection of circulating tumor DNA following neoadjuvant chemotherapy and surgery to anticipate early relapse in ER positive and HER2 negative breast cancer: Analysis from the PENELOPE-B trial.

502 Background: The PENELOPE-B phase III trial investigated the addition of one year of palbociclib to endocrine therapy (ET), in patients with hormone receptor positive HER2 negative breast cancer with residual invasive disease after neoadjuvant chemotherapy. Prior research has demonstrated that detection of circulating tumor DNA (ctDNA) in the adjuvant setting is associated with a high risk of disease relapse. We assessed the potential of ctDNA analysis to predict future clinical relapse for patients enrolled in the PENELOPE-B trial. Methods: Patients who were endocrine naïve at the time of study entry were selected for ctDNA analysis. Plasma samples were collected at baseline (after completion of neoadjuvant chemotherapy and surgery), prior to cycle 7 (approximately 6 months into ET +- palbociclib), end of treatment (EOT), and progressive disease. A tumor sample was subjected to exome sequencing, and up to 50 tumor somatic mutations were tracked in plasma using error-corrected sequencing combined with a proprietary algorithm for ctDNA detection (RaDaR assay). Detection of ctDNA was associated with invasive disease-free survival (iDFS) and distant metastasis-free survival using Cox proportional hazard models. Results: Of 1250 patients enrolled in PENELOPE-B, 129 were endocrine naïve at trial entry, and 78 had a baseline ctDNA sample analyzed. The ctDNA analysis group was representative of the overall endocrine naïve group, with median follow-up of 42.9 months. Seven patients had baseline ctDNA detected, with detection strongly associated with iDFS (HR 8.8, 95% CI 3.3-23.4, p < 0.0001). Detection of ctDNA at cycle 7 (4 patients) was also strongly associated with iDFS (HR 25.5, 95% CI 6.5-99.6, p < 0.0001). Of the 7 patients with baseline ctDNA detection, 2 had undetectable ctDNA at cycle 7 and remained progression free at 30 months, although one later relapsed; the 3 patients with detectable ctDNA at cycle 7 all relapsed within 25 months. Of the 12 patients with a distant relapse within 24 months, only 4 had ctDNA detected at baseline and 3 first at cycle 7/EOT. Of the 8 patients with distant relapse after 24 months, 2 had ctDNA detected at baseline and none first at cycle 7/EOT. Conclusions: Detection of ctDNA following neoadjuvant chemotherapy, and surgery, is associated with a very high risk of early relapse suggesting limited efficacy of adjuvant ET. Clinical imaging and studies of experimental therapy are warranted in this patient population. Testing ctDNA after recent neoadjuvant chemotherapy in luminal-A like breast cancer has relatively low ‘sensitivity’ for predicting future relapse, in particular for later relapses, in part suggesting that response to neoadjuvant chemotherapy may reduce ctDNA detection. Clinical trial information: NCT01864746 .