Abstract P4-03-02: Up-classification of suspicious breast masses using opto-acoustic imaging

Abstract

Abstract Purpose: Breast cancer diagnostic methodologies have been optimized to achieve increased sensitivity at the expense of relatively low specificity. Seno Medical's opto-acoustic (OA) imaging fuses real time co-registered, temporally interleaved laser optic and ultrasound imaging showing dual functional (hemoglobin de-oxygenation) and morphology findings for breast masses using a hand-held probe. We present data from the PIONEER Pilot study (n=100). We have shown improved specificity for OA relative to the ultrasound component (IUS) and the site determination by conventional diagnostic ultrasound (CDU). We now examine the BI-RADS upgrades for 36 malignant masses achieved by OA versus IUS and the site determinations using CDU. Materials and Methods: A total of 7 independent registration readers (IRRs) blindly assessed all 36 malignant masses using IUS first and OA second without any knowledge of clinical data or outcome. Among the cancers, there were 2 BI-RADS 4b, 12 BI-RADS 4c, and 22 BI-RADS 5 according to participating site radiologists' CDU evaluations. IRRs trained to identify and score three OA internal features and two OA external features for all masses were immediately offered the results of two nomograms based on their OA feature scores to predict the Probability of Malignancy (POM), which was then used to assign a BI-RADS category. Results: Combining data from all 7 readers, OA findings enabled upgrades of site CDU-determined BI-RADS categories 43% of the time for BI-RADS 4b and 29% for BI-RADS 4c; in contrast, the overall percentages of IUS upgrades versus site CDU were 21% for BI-RADS 4b and 10% for BI-RADS 4c. Overall, 12% of all OA reads resulted in upgrades in contrast to 4.4% for IUS compared to site CDU BI-RADS classifications. Relative to IUS, the overall percentages of OA upgrades were 58% for BI-RADS 4b and 34% for BI-RADS 4c. OA has comparable sensitivity to IUS. Conclusions: OA was more likely than IUS to result in a BI-RADS upgrade of a malignant mass. If subsequently confirmed, OA findings may help identify more cancers prior to biopsy. The 2,095 PIONEER Pivotal Study will allow for confirmation. Clinical Relevance: The ability to upgrade BI-RADS 4b and 4c for cancer masses is an unmet need. If verified, these findings could provide additional evidence to confirm a malignant mass earlier and spare subsequent diagnostic evaluations. This may help plan the efficient identification and excision of malignant masses. Citation Format: Lavin PT, Stavros AT, Ulissey MJ. Up-classification of suspicious breast masses using opto-acoustic imaging. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-03-02.