Search

Background: The current standard method for locating nonpalpable breast lesions is wire guided localization (WGL) despite several methodological difficulties. Radioactive Seed Localization (RSL) has been developed to reduce these difficulties. The aim of this randomized trial was to compare the rate of positive resection margins between RSL and WGL in patients with nonpalpable invasive breast cancer (IBC) or ductal carcinoma in situ (DCIS). Material and Methods: Patients with nonpalpable IBC or DCIS visible on ultrasound were randomized to either of the two localization methods. Primary outcome was margin status at the final pathological evaluation. According to Danish standard in the study period margins were defined positive if cancer cells were found < 2mm from the inked margin. Secondary outcomes were duration of the surgical procedure, weight of the excised specimen and patient's pain perception. χ2-test, Fisher's exact test and Wilcoxon rank-sum test, respectively, were used to test differences between groups. Level of statistical significance was set to 5%. The average activity of seeds used in the trial was 1.70 MBq (range 0.7-3.27). Results: 413 cases representing 409 patients were randomized; 207 were allocated to RSL and 206 to WGL. 23 cases, who did not meet inclusion criteria, chose to withdraw, or had a change in surgical management, were excluded. The remaining 390 were included in the analysis. Patient, surgical and pathological characteristics between the two groups were alike, except for significantly more patients with DCIS in the WGL group (5.1% vs 0.5%). Significantly more cases in the WGL group (9.7%) needed additional localization compared to the RSL group (2.1%) (p=0.0014). In all cases but one in the RSL group, the index lesion was removed. Margins were positive in 23 cases (11.8%) in the RSL group compared to 26 cases (13.3%) in the WGL group. We were not able to detect a difference in margin status between the two groups (p=0.65). For IBC only, the number of positive margins was 22 (11.3%) in the RSL group and 21 (11.4%) in the WGL group (p=0.997). There was no difference between the two groups in the amount of tissue removed whether the analysis was done on the primary excision (p=0.18) or the total weight including intraoperative re-excisions (p=0.33). There was no difference in pain perception between the two groups whether patients who received local anesthesia were kept in the analysis (p=0.28) or excluded (p=0.91). Local anesthesia was used more frequently in the RSL group. Finally, there was no difference in the duration of the surgical procedure (p=0.12), the complication rate (p=0.89) or the identification rate for SN (p=1.0). Conclusions: We were not able to detect any differences considering positive margins, patient's pain perception or duration of the surgical procedure between the two localization methods. However, RSL offers a major logistic advantage, as the seed localization can be done several days before surgery without any risk or discomfort for the patient, with a low proportion of patients needing additional localization. So the RSL procedure has now been found preferable at our institutions. Citation Format: Langhans L, Tvedskov TF, Klausen TL, Jensen M-B, Talman M-L, Vejborg I, Benian C, Roslind A, Hermansen J, Oturai PS, Bentzon N, Kroman N. Radioactive seed localization versus wire guided localization of nonpalpable invasive and in situ breast cancer: A Danish multicenter randomized controlled trial [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr S3-08.

Introduction: There is extensive research into immune-oncology (IO) treatment targets for triple negative breast cancer (TNBC). However, little knowledge exits on variations in expression levels of IO targets between paired primary tumors (PT) and metastases (MT) in TNBC. Objectives: To compare expression levels of IO drug targets in paired PT and distant MT from TNBC patients. PD-L1 protein expression was examined immunohistochemically, as well as differences in levels of tumor infiltrating lymphocytes (TILs). Material and methods:552 patients were primarily operated for TNBC at 2 university hospitals in Denmark between 2004 and 2014. Of these, samples from 22 patients were available from both PT (22 samples) and distant MT (24 samples, with 2 chronologically separate metastases for two patients). Localizations of metastases included 9 brain, 1 skin and 1 lymph node (both not regional), 2 bone, 5 liver, 2 soft tissue, 1 lung, and 1 adrenal gland. Gene expression analyses were performed on the Ion Torrent S5XL system, using the Oncomine Immune Response Research Assay. PD-L1 expression was evaluated immunohistochemically using the 28-8 clone (pharmdX). Cut-off levels for PD-L1 expression were 1% positive tumor cells. TILs were evaluated in accordance with guidelines from the International Immuno-Oncology Biomarker Working Group. Results: Gene expression levels of 29 established IO drug targets were tested on 18 of the 22 TNBC patients with matched PTs and MTs (Table 1). 4 genes showed significant downregulation in MTs compared to PTs: IDO1 (p=0.01), CXCR4 (p=0.01), KIR2DL1 (p=0.05) and TNFRSF9 (p=0.03). However, other important IO targets such as CTLA4, STAT3, TIGIT, and TNSFRSF18 (GITR) were not significantly different between PT and MT. CD276 (B7-H3), PD-1, IL2 and LAG3 showed a tendency towards significance (p= 0.08, 0.07, 0.05, 0.08), with LAG3 being the only gene upregulated in MT compared to PT. TILs were evaluated for all PTs and MTs. TILs in PTs ranged from 1 to 40% (median=13%). TILs in MTs ranged from 1 to 60% (median=5%). There was no significant overall change in TILs from PT to MT (p=0.2). However, 9 patients had a difference in TILs between PT and MT of > ±10% (-20 - +30). PD-L1 immunohistochemical evaluation was possible for both PT and MT in 20 cases. There was no significant overall change in PD-L1 expression (p=0.39). However, 4 patients with positive PTs, had negative MTs, and 4 patients with PD-L1 negative PTs, had positive MTs. Discussion: This is, to our knowledge, one of the largest studies of gene expression of IO drug targets in TNBC matched PTs and MTs. We show significant differences in gene expression levels for IDO1, CXCR4, KIR2DL1 and TNFRSF9, pointing to a generally less immune-active environment in MTs. This was also corroborated by lower levels of TILs in MTs, although these findings were not significant. For PD-L1 protein expression, we found that PD-L1 expression can change between PT and MT. In the clinical setting, treatment decisions are often based upon expression levels in a single biopsy or surgical specimen. Intra-patient heterogeneity should be taken into account in the diagnostic setting, and when making treatment decisions. Table 1IO drug targetRatio (Primary/metastasis)P-valueIDO19.65 (1.68-54.95)0.01CXCR41.83 (1.17-2.87)0.01KIR2DL12.17 (1.01-4.69)0.05TNFRSF97.11 (1.21-41.93)0.03CD276 (B7H3)1.51 (0.94-2.39)0.08PDCD1 (PD1)3.32 (0.88-12.64)0.07IL22.87 (0.99-8.28)0.05LAG30.28 (0.07-1.16)0.08CTLA41.96 (0.45-8.51)0.35STAT31.21 (0.92-1.60)0.16TIGIT1.87 (0.25-14.03)0.53TNFRSF18 (GITR)0.38 (0.06-2.20)0.26 Citation Format: Elisabeth S Stovgaard, Tim S Poulsen, Eva Balslev, Dorte L Nielsen, Anne Roslind, Maj-Lis Talman, Iben Kümler, Ib J Christensen, Estrid Høgdall. Triple negative breast cancer: Expression of immuno-oncology treatment targets and prevalence of tumor infiltrating lymphocytes in paired primary tumors and metastases [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-04-13.