Abstract C74: Screening outcomes of digital Breast tomosynthesis versus digital mammography in an underserved population with multiple disparities

Abstract

Abstract Purpose: The purpose of this report is to compare screening outcomes in an underserved population in North Louisiana with two different imaging techniques: digital breast tomosynthesis and standard digital mammography. Digital breast tomosynthesis (DBT) is a relatively new technology that images the breast in 3-D which is thought to overcome limitations of standard (2-D) digital mammography (DM). The outcome measures for this report are: 1) recalls for further mammography and evaluation and 2) breast cancers confirmed by biopsy. Background: The LSU Feist-Weiller Cancer Center has operated a free cancer screening clinic for uninsured and under-insured women since 1999. We started with one clinic in a fixed location in the LSU hospital in Shreveport, Louisiana. Since 2009 cancer screens are also routinely performed by 2 mobile units in 28 rural, underserved locations in 24 parishes in North Louisiana. Permanent mammography facilities are not available in 40% of the parishes we serve. For this population, in the remote rural areas, or the working poor in the non-rural areas, recalls are a significant extra burden on the patients. Methods: In 2009 we started mobile mammography screening with a Hologic Selenia Digital mammogram machine onboard. In 2014 we obtained another vehicle which was especially designed for the Hologic Dimension Tomographic mammography machine. DBT in a mobile setting is rare, with very few implementations in the U.S. In Louisiana, we have the only mobile DBT unit, statewide. This presents an unique opportunity to evaluate this new technology, in a mobile screening program in a high-risk population with multiple disparities. The subjects used in this presentation are those seen on one of the two mobile screening vehicles between April 1, 2009 and December 31, 2015. All patients screened qualified as either uninsured or under-insured and their mammograms were paid for by one of a multiple number of grants. All cancer screens were performed according to American Cancer Society guidelines by technologists certified in mammography. All mammographic images were digital. Images were read by one of two board certified radiologists, specialists in mammography and both trained to read DBT images. Of special interest to this presentation is the number of patients who are called recalled for additional views or other follow-up and the results of those recalls visits. Results: Results of screening for only the mobile vehicles are presented below, with emphasis on comparing results from tomosynthesis mammography (DBT) to those from standard digital mammography (DM). Chi-Squared tests were used to calculate p values. For the meeting presentation the numbers will be updated to include more patients screened and demographics and area health measures will be included. 4,619 patients were given mammograms with the DM machine, and (since it is new and has not long been in service) 2,731 with DBT. The number of recalls for additional study was 713 (15.4%) with DM and 131 (4.8%), with DBT. DBT resulted in significantly fewer recalls, p<0.0001. The number of cancers detected by DM was 20 (2.8% of recalls) and by DBT was 15 (11.5% of recalls). Again this is highly significant, p<0.0002. Conclusion: tomosynthesis mammograpy is significantly better at reducing unnecessary recalls, and at dectecting breast cancer than usual digital mammography. Recall visits are an especially serious burden on underserved populations with many access to care issues. Tomosynthesis mammography requires a more expensive machine, but its value, especially in minority underserved populations which are hard to reach and have major access to cancer care issues is undeniable. Citation Format: Jerry W. McLarty, Jennifer Lance, Towanno Collins, Sharon Grubbs, Stacey Massey, Nelson Luraguiz. Screening outcomes of digital Breast tomosynthesis versus digital mammography in an underserved population with multiple disparities. [abstract]. In: Proceedings of the Ninth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2016 Sep 25-28; Fort Lauderdale, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(2 Suppl):Abstract nr C74.