Trends in breast cancer death by stage at diagnoses in 2000-2018.

person Adriana Matutino Kahn Yale Cancer Center, New Haven, CT info_outline Adriana Matutino Kahn, Michal Marczyk, Andrea Silber, Mariya Rozenblit, Tara B. Sanft, Michael DiGiovanna, Sarah Schellhorn, Ian E. Krop, Maryam B. Lustberg, Eric P. Winer, Lajos Pusztai

Authors person Adriana Matutino Kahn Yale Cancer Center, New Haven, CT info_outline Adriana Matutino Kahn, Michal Marczyk, Andrea Silber, Mariya Rozenblit, Tara B. Sanft, Michael DiGiovanna, Sarah Schellhorn, Ian E. Krop, Maryam B. Lustberg, Eric P. Winer, Lajos Pusztai Organizations Yale Cancer Center, New Haven, CT, Silesian University of Technology, Gliwice, Poland, Yale University, New Haven, CT, Yale School of Medicine, New Haven, CT, Yale Cancer Center, Yale School of Medicine, New Haven, CT Abstract Disclosures Research Funding No funding received None. Background: Approximately 40-45,000, women die from breast cancer each year, most of which are due to metastatic recurrence of stage I-III cancers. Stage I and II cancers are considered highly curable, but they also include most patients. The goal of this analysis is to examine the contribution of stage I, II, III, and IV disease to annual breast cancer-specific death and assess time trends. Methods: Data from patients diagnosed with stage I-IV breast cancer between 1975-2018 (N = 1,545,502) were obtained from The Surveillance, Epidemiology, and End Results (SEER) Program. After filtering for the female sex, one primary site only, surgery performed, known AJCC stage > 0, and survival data available we included N = 1,018,331 patients. We estimated the year of breast cancer-specific death (BCSD) by adding time to BCSD to a year of diagnosis for each patient. We summarized the proportion of BCSD by stage at diagnosis in each year between 2000 and 2017. All temporal trends were assessed by fitting the linear model with interactions between stage and year, then running the ANOVA to test model coefficient significance. For each year of diagnosis, Kaplan-Meier curves were fitted to patient data grouped by stage, and 5-year survival was estimated. Results: The study population included 24% ER-, 73% ER+ (the rest unknown) cancers; 35% HER2-, 6% HER2+ (the rest unknown); 24% were < 50 and 76% > 50 years of age; 60% received adjuvant chemotherapy. Over time, the frequency of stage I cancers increased from 49% in 2000 to 54% by 2017 (slope = 0.37 [95% CI: 0.32; 0.42]), and stage II and III decreased from 35.2% to 34.9% (slope = - 0.09 [-0.15; -0.04]) and from 14% to 10% (slope = -0.23 [-0.29; -0.18]), respectively. Five-year breast cancer-specific survivals remained stable for stage I cancers and increased significantly for stage II (slope = 0.002 [0.001;0.003]), stage III (slope = 0.006 [0.005;0.007]) and IV cancers (slope = 0.011 [0.010; 0.012]). The contribution of stage I cancers to annual breast cancer death increased and reached 23% in 2017, while death from stage III disease decreased to 30%. Since 2003, > 50% of breast cancer deaths are from stage I-II cancers. Conclusions: The prevalence of stage I breast cancers increased in the past 20 years, and despite excellent overall prognosis, stage I-II cancers account for over 60% of all BCSD. To substantially reduce breast cancer death, we need to better identify the minority of Stage I-II patients who remain at risk for recurrence and death. Stage Breast Cancer Specific Death by Stage at Diagnosis by Year [%] 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 I 16 17 18 18.42 19 18 20 19 20 21 20 20 22 21 22 23 23 23 II 31 32 31 34.18 35 35 36 36 38 36 38 38 38 38 37 38 39 39 III 36 35 38 36.26 36 36 35 34 33 33 33 32 32 31 31 29 29 30 IV 17 15 13 11.14 11 11 10 10 10 9 9 10 9 9 9 9 8 7