Comparison of conventional clinical models and a multigene assay to assess DCIS risk in a prospective single center study.

• DOI: 10.1200/jco.2023.41.16_suppl.e12578
• Publication Date: 2023-06-04T14:09:31Z
• Authors: Loise Wairiri; Janice N. Kim; Savannah C. Partr...; Sara H. Javid; Mark R Kilgore; Debosmita Biswas; Mary L. Bryant; Anum S Kazerouni; Daniel S Hippe; Habib Rahbar
• Keywords: Nomogram

e12578 Background: Ductal carcinoma in situ (DCIS) is a preinvasive breast cancer typically excised and treated with adjuvant therapy. While there consensus that this results overtreatment, little agreement on who may avoid radiation or endocrine Two freely available prediction models, Van Nuys Prognostic Index (VNPI) Memorial Sloan Kettering Nomogram (MSK-N), are commonly used to identify low-risk DCIS based standard clinicopathological features. Oncotype newer commercially tissue-based multigene assay also assess risk, but its use limited due cost unclear value over VNPI MSK-N. We sought compare these tests’ determining ipsilateral recurrence (IBR) risk potential de-escalate Methods: In subanalysis from prospective single center clinical trial, we analyzed 38 patients newly diagnosed pure confirmed at excision. Each assessment tool presents different manner. provides multiple assessments: raw score (0-100), Score Category (low, intermediate, high), Refined incorporating features (10-year IBR risk). The calculates (4-12) assigns category high). MSK-N 10-year risk. tests were dichotomized as low vs. not-low categories using thresholds: = low, ≤ 10% Agreement of the 4 models (DCIS Category, Score, VNPI, MSK-N) compared Spearman’s rank correlation for continuous data; percent Cohen’s kappa (k) categorized data. Results: There was poor assessments across only showing significant moderate (r 0.53, p 0.001; others ranged r 0-0.27, > 0.1). number cases identified each 14 (37%), 3 (8%), 1 (3%), 0. Percent categorizations between 92-97%, classified fewer not (k 0-0.48, .1). demonstrated 63-71% other assays 0-0.26). Only case tests. Conclusions: have initially many lesions; however, inclusion create decreased substantially, providing very few additional de-escalation Additional studies needed determine precise rates when therapy decision-making. Clinical trial information: NCT03495011 .