Abstract P1-18-05: Management of bone marrow involvement in advanced breast cancer: Retrospective multicenter cohort study

Abstract

Abstract Background: Optimal management of bone marrow metastasis (BMM) in advanced breast cancer (ABC) remains unknown. Associated severe cytopenias often urge treatment, but they are also a challenging factor for delivering chemotherapy (CT). Since BMM in ABC is infrequent, available data are scarce. Aim: Clinical and prognostic characterization of ABC patients with BMM and its management; evaluation of the effectiveness of treating BMM with more myelosuppressive CT regimen (that we hypothesized that could be more active in the BMM) Vs. less myelosuppressive regimen. Methods: Retrospective cohort study of patients with pathological confirmation of BMM (positive myelogram or osteomedullary biopsy) between Jan'2010 and Dec'2016 in the two major Portuguese cancer centers. Patients with diagnosis of a second carcinoma or active hemato-oncological condition within 5 years before BMM were excluded. We considered the more myelosuppressive regimens those with > 5% risk of febrile neutropenia according to Truong et al 2015(1). Kaplan-Meier and log-rank methodology were used to estimate survival and Cox regression to identify the covariates with independent prognostic value. Statistical level of significance was 5%. Results: We included 74 patients: 74% with disease stage I-III at presentation, 74% ductal and 12% lobular invasive carcinomas, 58% grade 2 and 27% grade 3, 80% hormone receptor + / HER2-, 4% HER2+ and 12% triple negative (TN). Median time from ABC diagnosis to BMM was 10 months (IQR 2-36), synchronous in 34%. At diagnosis of BMM, median age was 57 years-old (IQR 47-65), 57% were post-menopausal, 97% had bone metastasis, 50% had visceral metastasis, 53% performed ≥ 2 previous palliative systemic treatments and 53% were exposed to bisphosphonates. The most frequent immunohistochemistry change in BMM biopsy was the loss of progesterone receptor expression (37%). Bicytopenia (anemia/thrombocytopenia) was the trigger for BMM investigation in 69% of cases, with grade 3-4 anemia in 16% and grade 3-4 thrombocytopenia in 26%. Median survival after BMM diagnosis was 5 months (95% CI, 3-11); overall survival at 12 and 24 months were 35% (CI 26-48%) and 24% (CI 15-36%), respectively. First treatment after BMM was CT in 58% (median survival, 11 months) and endocrine therapy in 14% (median survival, 3 months). An anti-HER2 regimen was used in 4% and 22% did not receive any treatment after BMM. On multivariate analysis, TN subtype (HR 4.02, CI 1.46-11.01), thrombocytopenia (G0 reference; G1-2: HR 2.47, CI 1.11-5.56; G3-4: HR 4.89, CI 1.85-12.91) and ≥ 2 palliative systemic treatments (HR 2.77, CI 1.46-5.27) were associated with worse prognosis. Within those treated with CT, there was a trend for a deleterious survival effect of more myelosuppressive regimens (HR 2.19, CI 0.94-5.09; 5 months Vs. 14 months, n=31), after controlling for subtype, number of previous regimens and thrombocytopenia. Conclusion: BMM was not a late event in ABC disease course and had worse prognosis in multi-treated patients, in TN subtype and in the presence of thrombocytopenia. No benefit was shown with the use of more myelosuppressive CT regimens. (1)Truong J et al, Ann Oncol. 2016 Apr;27(4):608-18. Citation Format: Martins-Branco D, Ferreira SC, Pereira I, André S, Varelas A, Leal C, Esteves S, Abreu MH, Sousa S, Moreira A, Brito M. Management of bone marrow involvement in advanced breast cancer: Retrospective multicenter cohort study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P1-18-05.