Search

Background: About 10–30% of breast cancer patients will develop brain metastases. In untreated patients with brain metastases the median survival is 1–2 months, and in those undergoing palliative radiotherapy — 3–6 months. The mechanism of brain metastases remains largely unknown. The identification of molecular markers might help in selecting high risk patients, and enable active surveillance, prevention and early treatment. The aim of this study was to analyze predictive value of expression of selected tumor proteins for the risk of brain metastases in breast cancer patients. Material and methods: This study included 198 advanced breast cancer patients treated between 2001 and 2007 in 11 oncology centers in Poland, including 96 woman with and 102 without overt brain metastases, respectively. The median age at diagnosis in these two groups was 52 and 60 years, respectively, with 52% and 32% of patients being premenopausal. Stage at diagnosis was similar in both groups and ductal carcinoma was a dominant histological type (76% and 86% of cases, respectively). Immunohistochemistry was performed on formalin-fixed paraffin embedded microarray cores derived from the primary tumor. Expression analysis included ER, PR, HER2, Ki67, CK5/6, EGFR, HER3, CXCR4, RAD51, E-cadherin, and claudin 3 and 4. Cox regression model was used to estimate the relative risk of brain metastases. Results: Expression of HER2, CK5/6, EGFR, RAD51 (both cytoplasmatic and nuclear staining), CXCR4 (cytoplasmatic staining) and Ki67 ≥14%, as well as ER or PR negativity was associated with increased risk of brain metastases in the univariate analysis (Table 1). Of those, Ki67 ≥14% (HR 2.76 [95%CI 1.70–4.48]; p < 0.001), cytoplasmatic expression of double strand DNA repair gene RAD51 (HR 1.87 [95%CI 1.14–3.08]; p = 0.014) and ER negativity (HR 1.72 [95%CI 0.36–0.94]; p = 0.029) were found to be significantly related to the risk of brain relapse in the multivariate analysis. Four molecular profiles composed of the latter three markers were created, of which a profile including ER, Ki67 and RAD51 was associated with the highest risk of brain metastases (HR 4.43 [95%CI 2.69–7.27]; p < 0.001). Molecular subtype analysis showed the highest risk of BM in the ER/PR/ HER2-negative (triple negative) subset (HR 1.21 [95%CI 1.11–1.32]; p < 0.001). Conclusion: Expression of proteins related to high tumor proliferation, DNA repair and ER negativity is associated with increased risk of brain metastases in breast cancer patients. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-12-09.

Background. Bones constitute the most frequent localization of distant failure in breast cancer patients. Treatment of bone metastases is virtually palliative, but in a proportion of patients can effectively prolong survival and improve quality of life. Currently, there are no effective strategies to prevent bone dissemination with systemic adjuvant therapies. Thus, better molecular selection and identification of patients who have particularly high risk of skeletal metastases may facilitate designing future clinical trials. In this study we analyzed expression of selected tumor proteins potentially associated with skeletal metastases in breast cancer patients. Patients and methods. The study group included 184 metastatic breast cancer patients; 113 with clinically diagnosed bone metastases and 71 with exclusively other sites of metastases, respectively. In all patients, using tissue microarrays technology, IHC expression of the following proteins was evaluated in the primary tumor: estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), Ki67, cyclooxygenase 2 (COX2), chemokine receptor CXCR4, osteopontin (OPN), calcium sensing receptor (CaSR), cytokeratins 5/6 (CK5/6) and parathyroid hormone-related protein receptor 1 (PTHrPR1). Additionally based on ER/PR, HER2 and Ki67 expression, following molecular breast cancer subtypes were selected: luminal A, luminal B HER2-negative, luminal B HER2-positive, nonluminal HER2-positive and triple negative. Results. Median survival in patients with bone vs. other site of metastases was 56 vs. 37 months respectively (p = 0.0098). ER expression was more common in patients who did, compared with those who did not develop bone metastases (74.% vs. 45% respectively; p = 0.0001), whereas cytoplasmic overexpression of OPN (1.9% vs. 14% respectively; p = 0.002) and PTHrPR1 (16% vs. 34% respectively; p = 0.007) was more common in patients with other sites of metastases. The impact of ER and cytoplasmic OPN expression on the risk of bone dissemination was confirmed in the multivariate analysis (Table 1). Luminal A breast cancer subtype (43% vs. 23% respectively; p = 0.009) and luminal B HER2-positive (16% vs. 4.9%, respectively; p = 0.032) were more common among patients with bone dissemination, whereas triple negative tumors prevailed in patients with other sites of metastases (16% vs. 38%; p = 0.002). Median survival of patients with luminal A subtype was significantly longer than that with remaining subtypes (67 vs. 38 months respectively; p = 0.0004). Conclusions. These results suggest that ER expression and lack of OPN expression are independent factors predicting increased risk of bone dissemination in breast cancer patients. Bone metastases are specifically associated with luminal A and luminal B HER2-positive subtypes. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-13-03.

Background Salivary gland neoplasms represent a heterogeneous group of benign and malignant tumors that comprise 6% of all head and neck cancers. Due to their low incidence, these tumors are poorly understood and remain a diagnostic and therapeutic challenge. Methods Out of 182 analyzed salivary gland tumors with various histologies, eight were positive for ALK immunohistochemistry. The cut-off was 15% positive cells in either membranous, nuclear or cytoplasmic compartments. The ALK positive tumors were first subjected to FISH analysis. Subsequently, these tumors were analyzed using hybrid capture based next generation sequencing to confirm possible ALK gene fusions or copy number alterations, and to detect additional genomic alterations of clinical relevance. Results An in-depth genomic analysis of the samples resulted in the detection of inactivating mutations in BRAF (p.D594N) and TP53 (p.C238S), as well as amplifications of ERBB2 and ALK. Strikingly, a novel MYO18A (Exon1-40)-ALK (exon 20-29) gene fusion was detected in a patient with adenocarcinoma not otherwise specified. This tumor was FISH positive and 100% of cells showed strong membranous staining for ALK. MYO18A (Exon1-40)-ALK (exon 20-29) gene fusion resulted in the retention of the kinase domain of ALK and the coiled-coil domain of MYO18A. Similarly to other ALK fusions, we hypothesize that the coiled-coil domain of MYO18A mediates the dimerization and activation of MYO18A-ALK, thereby resulting in an overexpression of constantly activated ALK. Conclusion Using hybrid capture based next generation sequencing, we identified in salivary gland tumors numerous genomic alterations in therapeutically relevant genes and a novel gene fusion (MYO18A-ALK). Patients harboring this fusion may potentially benefit from treatment with ALK inhibitors. Citation Format: Hanna Majewska, Judith Müller, Sotirios Lakis, Piotr Czapiewski, Adam Gorczyński, Mariola Iliszko, Alena Skalova, Rafal Dziadziuszko, Jacek Jassem, Wojciech Biernat, Roopika Menon, Johannes M. Heuckmann, Lukas C. Heukamp. Genomic profiling of salivary gland tumors identifies novel and targetable alterations. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3190.