Abstract P6-09-10: Association of dermal lymphatic involvement and survival in inflammatory breast cancer

Abstract

Abstract Purpose: The pathologic hallmark of inflammatory breast cancer (IBC) is the presence of tumor emboli within the papillary and reticular dermis of the skin, termed dermal lymphatic invasion (DLI). DLI can be confirmed with a skin-punch biopsy in approximately 75% of cases, but its presence is not required for the diagnosis of IBC because of variability of tumor emboli in affected areas. The impact of confirming DLI by skin biopsy on the clinical outcome of IBC is unknown. We hypothesize that the ability to confirm DLI by biopsy is associated with a higher tumor emboli load and consequently a poorer disease prognosis. Therefore, we examined whether documented DLI was associated with poorer overall survival (OS) in IBC. Methods: Clinical characteristics were evaluated among 286 women presenting with IBC between 1999 and 2016 and enrolled in the IRB-approved IBC registry at Dana-Farber Cancer Institute. Kaplan-Meier curves were used to estimate overall survival and the log-rank test to examine survival differences based upon the presence of DLI. Multivariable Cox proportional hazards models were used to estimate hazard ratios (HR) and 95% confidence intervals for associations of DLI and risk of death among women with IBC. Results: A total of 102 deaths occurred in our study population over a median follow-up of 2.5 years (yr) (range 10 days to 16 yr). Compared to IBC without DLI, IBC with confirmed DLI was more likely to be associated with the presence of lymphovascular invasion in breast biopsies (73.1% vs. 37.7%; p-value <0.01) but less likely to be seen with edema of the skin (59.1% vs. 81.1%; p-value=0.02). Documented DLI was not related to the presence of de novo metastasis at presentation (11.5% vs. 28.7%; p-value=0.07). OS did not significantly differ among women with IBC based upon the presence of DLI (log-rank test p-value=0.68). In multivariable models, DLI was not independently associated with OS in inflammatory breast cancer, HR(95%CI) = 0.92(0.48-1.78). Conclusions: Our findings suggest that IBC presenting with documented DLI on skin biopsy may vary with regard to clinical characteristics at diagnosis, including lymphovascular invasion within the breast, edema of the skin of the breast, and the presence of de novo metastases. These clinical distinctions suggest potential differences in biology of IBC according to the presence or absence of DLI, and the extent of tumor emboli. However, in this study, DLI was not found to be an independent prognostic factor in IBC with respect to OS. Due to the variability in the clinical features of IBC at presentation and inherent complexities in selecting skin biopsy sites, studies to investigate the accuracy of determining DLI based on punch biopsy are necessary to more comprehensively assess the impact of DLI on clinical outcomes in IBC. Citation Format: Hirko KA, Schlossman J, Harrison BT, Yeh ED, Jacene HA, Nakhlis F, Schlosnagle E, Overmoyer BA. Association of dermal lymphatic involvement and survival in inflammatory breast cancer [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 P6-09-10.