Chirag Shah, MD from the Cleveland Clinic discusses the SABCS 2020 abstract - DCIS biosignature reclassified patients who met RTOG 9804 or ECOG-ACRIN E5194 ʻlow-riskʼ clinicopathologic criteria into an elevated invasive risk group who benefited significantly from radiation therapy.
The aim of treatment for DCIS is to prevent recurrences of local invasive breast cancer.
Randomized DCIS clinical trials have shown that patients benefit from adjuvant radiation therapy (RT) following breast preservation surgery (BCS). However, the selection of treatment for DCIS patients remains a problem. Studies assessing successful risk and clinical-pathological characteristics have not found a
Group of patients with regard to local regulation, do not benefit substantially from RT after BCS.
DCISionRT (PreludeDx, Laguna Hills, CA), a biosignature, has recently been validated in several cohorts. For patients treated with and without RT after BCS, the test offers a continuous 10-year chance of breast events. We tested the usefulness of the biosignature in this study to classify patients who met good risk criteria for RTOG 9804 or ECOG 5194 but remained at high invasive risk after BCS and benefited from RT.
In a composite cohort made up of four studies, the study was carried out.
FFPE tissue samples and patient results were collected between 1986 and 2004 from Uppsala University Hospital and Västmanland County Hospital, Sweden (UUH), from 1999 to 2008 from the University of Massachusetts, Worcester (UMass), from 1990-2007 from Kaiser Permanente Northwest (KPNW) and from the SweDCIS cohort study (1987-2000).
• Patients were treated after BCS with or without RT. Except for the randomized SweDCIS trial for RT, treatment decisions were neither randomized nor strictly rules-based.
Individual patient results and biosignature effects were independently analyzed at the University of South Florida. Using Cox proportional hazard analysis, hazard ratios (HR) were calculated and 10-year risks were assessed using survival analysis.
For 535 women meeting good risk clinical-pathological RTOG 9804-like criteria (negative margins vs broad margins) and for 660 women meeting ECOG E5194 grade 1 or 2 criteria, complete biomarker and clinical data were accessible. Within 10 years of diagnosis for RTOG 9804-like patients, there were 38 invasive breast cancer events in this subset and 49 for patients meeting the ECOG E5194 criterion.
• There was no substantial reduction in RT in the Low-Risk biosignature community (p>0.15), where the 10-year absolute invasive RT gain ranged from 1% to 2% for patients meeting the RTOG 9804 or ECOG5194 criterion.
• However, RT substantially decreased invasive cancer risk (p<.002) for patients with RTOG9804 (HR=0.16, 95% CI[0.05,0.47]) or ECOG 5194 Grade 1 or 2 requirements (HR=0.27, 95% CI[0.12,0.64]) in the biosignature Elevated Risk category. This correlates to a 10-year invasive relative risk reduction for RTOG 9804 of 84 percent, for ECOG 5194 Grade 1 or 2 of 73 percent. The 10-year absolute invasive gain from RT was 15% for RTOG 9804, 8% for Grade 1 and 2 ECOG E5194.
In comparison, a low-risk group of patients with modest (1-2 percent) risk reduction from RT was also defined by the biosignature.
The DCISionRT score was significantly correlated with the benefit of RT therapy relative to conventional clinicopathologic characteristics used to render RT recommendations.