Current and future diagnostic and treatment strategies for patients with invasive lobular breast cancer

BACKGROUND

Invasive lobular breast cancer (ILC) is the second most common type of breast cancer after invasive breast cancer of no special type (NST), representing up to 15% of all breast cancers.

DESIGN

Latest data on ILC are presented, focusing on diagnosis, molecular make-up according to the European Society for Medical Oncology Scale for Clinical Actionability of molecular Targets (ESCAT) guidelines, treatment in the early and metastatic setting and ILC-focused clinical trials.

RESULTS

At the imaging level, magnetic resonance imaging-based and novel positron emission tomography/computed tomography-based techniques can overcome the limitations of currently used imaging techniques for diagnosing ILC. At the pathology level, E-cadherin immunohistochemistry could help improving inter-pathologist agreement. The majority of patients with ILC do not seem to benefit as much from (neo-)adjuvant chemotherapy as patients with NST, although chemotherapy might be required in a subset of high-risk patients. No differences in treatment efficacy are seen for anti-human epidermal growth factor receptor 2 (HER2) therapies in the adjuvant setting and cyclin-dependent kinases 4 and 6 inhibitors in the metastatic setting. The clinical utility of the commercially available prognostic gene expression-based tests is unclear for patients with ILC. Several ESCAT alterations differ in frequency between ILC and NST. Germline BRCA1 and PALB2 alterations are less frequent in patients with ILC, while germline CDH1 (gene coding for E-cadherin) alterations are more frequent in patients with ILC. Somatic HER2 mutations are more frequent in ILC, especially in metastases (15% ILC versus 5% NST). A high tumour mutational burden, relevant for immune checkpoint inhibition, is more frequent in ILC metastases (16%) than in NST metastases (5%). Tumours with somatic inactivating CDH1 mutations may be vulnerable for treatment with ROS1 inhibitors, a concept currently investigated in early and metastatic ILC.

CONCLUSION

ILC is a unique malignancy based on its pathological and biological features leading to differences in diagnosis as well as in treatment response, resistance and targets as compared to NST.

Features of durable response and treatment efficacy for capecitabine monotherapy in advanced breast cancer: real-world evidence from a large single-centre cohort

PURPOSE

In metastatic breast cancer (MBC) population treated with capecitabine monotherapy, we investigated clinical-pathological features as possible biomarkers for the oncological outcome.

METHODS

Retrospective study of consecutive MBC patients treated at University Hospitals Leuven starting capecitabine between 1999 and 2017. The primary endpoint was the durable response (DR), defined as non-progressive disease for > 52 weeks. Other main endpoints were objective response rate (ORR), time to progression (TTP) and overall survival (OS).

RESULTS

We included 506 patients; mean age at primary breast cancer diagnosis was 51.2 years; 18.2% had de novo MBC; 98.8% were pre-treated with taxanes and/or anthracycline. DR was reached in 11.6%. Patients with DR, as compared to those without DR, were more likely oestrogen receptor (ER) positive (91.5% vs. 76.8%, p = 0.010) at first diagnosis, had a lower incidence of lymph node (LN) involvement (35.6% vs. 49.9%, p = 0.039) before starting capecitabine, were more likely to present with metastases limited to ≤ 2 involved sites (54.2% vs. 38.5%, p = 0.020) and time from metastasis to start of capecitabine was longer (mean 3.5 vs. 2.7 years, p = 0.020). ORR was 22%. Median TTP and OS were 28 and 58 weeks, respectively. In multivariate analysis (only performed for TTP), ER positivity (hazard ratio (HR) = 0.529, p < 0.0001), HER2 negativity (HR = 0.582, p = 0.024), absence of LN (HR = 0.751, p = 0.008) and liver involvement (HR = 0.746, p = 0.013), older age at capecitabine start (HR = 0.925, p < 0.0001) and younger age at diagnosis of MBC (HR = 0.935, p = 0.001) were significant features of longer TTP.

CONCLUSION

Our data display relevant clinical-pathological features associated with DR and TTP in patients receiving capecitabine monotherapy for MBC.

Efficacy of anti-HER2 therapy in metastatic breast cancer by discordance of HER2 expression between primary and metastatic breast cancer

PURPOSE

In stage IV breast cancer, the efficacy of human epidermal growth factor receptor 2 (HER2) targeted therapies in cases with discordance in HER2 expression between primary and metastatic site is not well known. We studied progression free (PFS) and overall survival (OS) by HER2 concordance when treating women with taxane-trastuzumab (± pertuzumab) in first or second line and trastuzumab-emtansine (T-DM1) or capecitabine-lapatinib in later lines.

PATIENTS AND METHODS

Retrospective monocentric study including all breast cancer patients receiving trastuzumab between January 2002 and September 2017 at the University Hospital in Leuven; we selected metastatic patients with an available HER2 status in primary and metastatic site. The Kaplan-Meier method was used for estimating PFS/OS and log-rank test for analyzing between group differences. A Cox model is used for testing difference between groups while correcting for Pertuzumab. Multivariable Cox regression is used to model overall survival as a function group, correcting for possible confounders.

RESULTS

We included 74 patients; 46 had an unchanged HER2 status (positive/positive), 9 lost HER2 (positive/negative), while 19 acquired HER2 amplification (negative/positive). 25 out of 28 cases with a discordant HER2 status were positive for ER and/or PgR in the primary site. HER2 positive/negative cases had a significantly lower PFS for taxane-trastuzumab-(pertuzumab) (PFS = 5.5 months), compared to HER2 positive/positive (PFS 9 months, p = 0.01) and HER2 negative/positive (PFS 14 months, p = 0.01) patients. PFS for later line T-DM1 (n = 30) was significantly higher for the HER2 positive/positive group (PFS 6.0 months) than for the discordant groups HER2 negative/positive (PFS 1.0 month, p = 0.04) and HER2 positive/negative (PFS 1.5 month, p = 0.01). After correcting for possible confounders, the HER2 positive/negative group had a significantly worse OS compared to HER2 positive/positive (HR 0.19, 95% CI 0.08-0.44) and negative/positive (HR 0.15, 95% 0.06-0.38).

CONCLUSION

Conversion of HER2 status was seen in 28 out of 74 cases and was mostly observed in hormone receptor-positive tumors. In contrast to patients with HER2 loss, patients with a positive conversion of HER2 status derived substantial benefit from first line treatment with taxane-trastuzumab-(pertuzumab). This study highlights the importance of re-biopsying the metastatic lesion and changing treatment according to the last HER2 result.

Abstract P4-06-16: Analysis of tumor infiltrating lymphocytes in three age categories of luminal B-like breast cancer patients and its correlation with lymph node involvement and systemic immunosenescence/frailty markers

BACKGROUND Immunosenescence, the age-related decrease in immune competence, which is characterized by decreased adaptive immunity and increased low-grade inflammation. It may lead to altered tumor immunity. However, immunosenescence markers have not been correlated yet with tumor infiltrating leukocytes (TILs) nor with clinical frailty.

METHODS This is the first study that investigates how age affects the relation between systemic immunity and tumor microenvironment in early luminal B-like breast cancer (BC) in function of age. Luminal B-like BC was defined as grade II-III, ER+, HER2- on core biopsies at inclusion. Three distinct age categories of BC patients were chosen: 35-45 years (y) (N=15), 55-65y (N=19), ≥70y (N=31). Stromal TILs (sTILs) % were assessed according to published guidelines, on representative tumor sections. Further characterization of the TILs by immunohistochemistry, using the antibodies against CD3, CD4, CD8, CD5, CD20, CD68 and FOXP3, is currently ongoing. Immunosenescence was evaluated by looking at eleven inflammatory plasma cytokines and chemokines, circulating insulin-like growth factor-1 (IGF-1), T-cell p16INK4a expression, PBMC subset profiles and expression levels of twenty immune-related microRNAs. In the old age category, geriatric assessment was performed.

RESULTS With increasing age, sTILs % significantly decreased, concomitant with significantly increased plasma levels of several inflammatory cytokines (IL-1α, IL-6) and chemokines (IP-10, IL-8, MCP-1), intermediate monocytes, as well as T-cell p16INK4a expression. Significant age-related decrease was seen for plasma IGF-1, naïve CD8+ T-cells and CD8+ T-cells expressing CD27 and/or CD28. Four immune-related microRNAs showed significantly different expression levels between the age groups: miR-18a decreased with age, miR-155 increased with age, miR-19b and miR-20a peaked in the middle group. As expected, various correlations exist between the different blood immunosenescence markers. The % of sTILs showed weak negative correlations with IL-6, IL-8, IL-1α, MCP-1 and the % of regulatory T-cells. Additionally the % of sTILs and several makers of immunosenescence (MCP-1, miR-20a, miR-155, intermediate monocytes) correlated with components of the geriatric assessment (activities of daily living (ADL), mini nutritional assessment (MNA), mini–mental state examination (MMSE)) and with the Leuven oncogeriatric frailty score (LOFS) in the oldest group. Conversely, lymph node involvement was not associated with the % of sTILs nor with any blood aging biomarker.

CONCLUSION sTILs % and several blood immunosenescence markers significantly differ between young and older luminal B-like BC patients. Some of these markers correlated with markers of clinical frailty as well. These findings suggest that interactions between tumor cells and immune/inflammatory cells differ with age and therefore applicable immune biomarkers and approaches for immunotherapy may vary depending on patients' age.

Citation Format: Berben L, Hatse S, Kenis C, Dalmasso B, Smeets A, Vos H, Neven P, Floris G, Wildiers H. Analysis of tumor infiltrating lymphocytes in three age categories of luminal B-like breast cancer patients and its correlation with lymph node involvement and systemic immunosenescence/frailty markers [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 P4-06-16.

Abstract P4-06-14: Single-cell RNA sequencing to delineate changes in tumor microenvironment induced by immunotherapy

Introduction: Improved understanding of the interplay between the immune system and cancer would lead to more adequate response prediction to immune-checkpoint blockade (ICB) treatment. Furthermore, insight into immune interactions would allow identifying biomarkers for response to ICB. Single-cell RNA sequencing has emerged as a powerful technology to characterize heterogeneity in a large population of cells and opens up opportunities to predict response to therapy.

Purpose: Tracking the effect of a single dose of Pembrolizumab on the tumor microenvironment through single-cell RNA sequencing.

Methods: Single-cell RNA sequencing was performed on tumor tissue of one patient before and 10 days after a 200 mg dose of Pembrolizumab (Keytruda®). Changes in cell (sub)populations were analyzed. Fresh tumor material was obtained from a core needle biopsy at diagnosis and from the resection specimen.Single cell suspensions were converted to barcoded scRNA-seq libraries with the Chromium Single Cell 3' kit with 10X Genomics platform, aiming for an estimated 5,000 cells per library. The libraries were sequenced using HiSeq400. Expression matrices were generated using CellRanger and analyzed by Seurat package. Dimensionality reduction using principle component analysis was applied to identify major cell types and their subtypes.

Results: The analyzed tumor was a grade 3 invasive ductal adenocarcinoma, hormone receptor negative and HER2+ positive, pT2N0M0. Tumor infiltrating lymphocyte (TIL) count was 30% on core biopsy and 50% on resection specimen. We sequenced a total of 9867 transcriptomes at single cell resolution before and after Pembrolizumab treatment, consisting of 5808 and 4049 cells, respectively. Major cell types of the tumor microenvironment were identified by leveraging single-cell transcriptomics analysis. We observed a sharp decrease in cancer cells after treatment (74% versus 25%), which was accompanied by an increase of tumor infiltrating T cells (18% versus 50%). The residual cancer cells after immunotherapy showed a higher expression of the major histocompatibility complex (MHC), MHC-II in particular. In addition, we found enrichment of B-cells and endothelial cells and a downregulation of fibroblasts and myeloid cells. Furthermore, we analyzed subtypes of each cell type. We found a marked increase in cytotoxic CD4 (6% versus 18%) and cytotoxic CD8 cells (4% versus 12%) in the T cell population. These cytotoxic CD8 cells clearly expressed higher PD-1 after immunotherapy. A decrease in CD4 Tregs, naïve CD4 and intermediate CD8 cells was observed and the B cell enrichment after treatment was mainly driven by the increase of follicular B-cells. The increase of endothelial cells was driven by capillary tumor endothelial but not tip cell population, suggesting vessel normalization rather than neo-vascularization.

Conclusion: Single-cell RNA sequencing provides a powerful tool in detecting changes in the tumor microenvironment induced by immunotherapy, and thus offers new opportunities to predict response to immunotherapy. Thirty-three additional patient samples will be analyzed in the near future, with a special focus on T- and B- cell receptor repertoire.

Citation Format: Vos H, Qian J, Floris G, Nevelsteen I, Lambein K, Berben L, Neven P, Wildiers H, Lambrechts D, Smeets A. Single-cell RNA sequencing to delineate changes in tumor microenvironment induced by immunotherapy [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 P4-06-14.

Abstract PD3-09: Efficacy of HER2 inhibitors in metastatic breast cancer by discordance in HER2 amplification status between primary and metastatic breast cancer

Purpose:

In stage IV breast cancer (BC), discordance in the human epidermal growth factor receptor 2 (HER2) amplification status between primary and metastatic BC might affect efficacy of HER2-targeted agents. We studied progression free (PFS) and overall survival (OS) dependent on HER2 concordance in patients treated with a first line taxane-trastuzumab combination and later line trastuzumab-emtansine (T-DM1).

Patients and Methods:

This retrospective monocentric study included 76 patients with metastatic BC under treatment with trastuzumab in which a biopsy from both the primary and metastatic site was available. HER2 amplification status, sex-steroid receptor status, Nottingham prognostic index, distant metastasis-free interval and consecutive lines of therapy were retrieved from patients' reports. The Kaplan-Meier method was used for estimating PFS/OS and log-rank test for analyzing between group differences. A Cox model is used for testing difference between groups while correcting for Pertuzumab. Multivariable Cox regression is used to model OS as a function group, correcting for possible confounders.

Results:

Discordance in HER2 amplification status was seen in 30 out of 76 patients (39%), 11 patients lost HER2 amplification in the metastatic lesion (HER2lost) while 19 acquired HER2 amplification (HER2acquired). The other 46 patients had a HER2 amplification on both primary and metastatic site (HER2stable). The HER2lost group had a significant lower median PFS (PFS= 5.5 months) for taxane-trastuzumab, after correcting for pertuzumab, compared to the HER2stable group (PFS= 9 months, corrected p= 0.0146) and HER2acquired group (PFS=14 months, corrected p=0.0121). For T-DM1 treatment, both discordant groups, HER2acquired (PFS=1.1 months, p=0.0373) and HER2lost (PFS=1.5 months, p=0.0116), had a significant lower PFS compared to the HER2stable group (PFS=6.0 months). After correcting for possible confounders, HER2lost had a significant worse OS compared to HER2stable (HR 0.187, 95% CI 0.079 – 0.439) and HER2acquired (HR 0.147, 95% 0.058-0.378).

Conclusion:

Loss of HER2 amplification in metastatic lesions seems to have a negative predictive value for PFS on HER2-targeted agents and negative prognostic impact on OS. Acquiring of HER2 amplification was predictive for lower PFS on T-DM1 but wasn't predictive for lower PFS on taxane-trastuzumab.

Citation Format: Van Raemdonck E, Berteloot P, Laenen A, Han S, Van Nieuwenhuysen E, Salihi R, Concin N, Vergote I, Floris G, Wildiers H, Punie K, Neven P. Efficacy of HER2 inhibitors in metastatic breast cancer by discordance in HER2 amplification status between primary and metastatic 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 PD3-09.

Abstract P3-03-32: Monocentric experience with the sentinel lymph node biopsy prior to neoadjuvant chemotherapy in clinically lymph node negative early breast cancer

Background

In patients with clinically lymph node negative (cN0) early breast cancer (EBC) treated with neoadjuvant chemotherapy (NACT), the sentinel lymph node biopsy (SLNB) can be performed before or after NACT. We report safety of axillary staging performing the SLNB prior to NACT in cN0 EBC and estimate NACT-induced downstaging to ypN0 in previously NACT-treated cN1 EBC, to make an assumption for avoiding axillary lymph node dissection (ALND) if SLNB was done after NACT.

Patients and Methods

Monocentric retrospective study of consecutive triple negative (TNBC) and HER-2 amplified BC patients treated with standard NACT. cN0 patients had SLNB before NACT followed by local therapy. Axillary lymph node dissection (ALND) post-NACT was performed in all cN1 and in cN0 cases with a positive or failed SLNB. Using descriptive statistics, we here report SLNB-detection and SLNB-positive rate, SLNB-operative complications, complete tumor regression in the breast (ypT0/is) and disease-free survival (DFS) for cN0 cases and NACT-induced downstaging to ypN0 in previously NACT-treated cN1 EBC.

Results

We included 245 NACT-treated patients; 119 cN0 and 126 cN1. SLNB-detection rate in cN0 cases was 99,2%; 25 or 21% had ≥ 1 involved SLN, 21.8% experienced SLNB related-complications (e.g. infection, seroma, hematoma) leading to NACT-delay in 3 and interruption in 1 patient. Median start of NACT after SLNB was 7 days (range 1-20 days). In patients with a positive SLNB, there were no additional involved nodes in the ALND. In 5 of these patients, therapy response in a lymph node was described. Complete tumor regression in the breast (ypT0/is) was 52% in SLNB-positive and 59,1% in SLNB-negative cN0 cases. NACT-induced ypN0 was 61% in cN1 cases. At 30 months of median follow-up (range 1-86 months), DFS was 93,2% (4.2% metastatic; no axillary relapse) in cN0 cases. Median DFS was better for patients with complete tumor regression in the breast as compared to those with partial response; 95.6% and 90% respectively.

Conclusion

In conclusion, performing SLNB before NACT in cN0 cases is a safe and accurate method. While some pN1(sn) could have avoid ALND by NACT-induced axillary down-staging, based on our assumption, long term follow-up is needed to conclude whether SLNB after NACT is safe.

Keywords: Breast cancer, neoadjuvant chemotherapy, timing sentinel lymph node biopsy

Citation Format: Delameilleure M, Smeets A, Nevelsteen I, Han S, Van Nieuwenhuysen E, Berteloot P, Hoste G, Salihi R, Van Ongeval C, Keupers M, Prevos R, Wildiers H, Punie K, Van Limbergen E, Weltens C, Janssen H, Floris G, Vergote I, Neven P. Monocentric experience with the sentinel lymph node biopsy prior to neoadjuvant chemotherapy in clinically lymph node negative early 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 P3-03-32.

Abstract P4-02-07: Comparison of breast cancer molecular subtyping by Immunohistochemistry and by BluePrint® next generation RNA sequencing-based test at University Hospitals Leuven and Curie Institute Paris

Background

MammaPrint® (MP) and BluePrint® (BP) are microarray-based tests with MP being prognostic for distant recurrence and BP enabling stratification into breast cancer molecular subtypes (Luminal, HER2, Basal-type). Recently, a CE marked MP and BP targeted RNA Next Generation Sequencing (NGS)-based kit was developed at Agendia and validated at University Hospitals Leuven and Curie Institute Paris. Here we compare breast cancer molecular subtype stratification defined by immunohistochemistry (IHC) and by MP and BP NGS- and microarray- based tests.

Patients and Methods

In this study, 124 primary operable invasive breast cancer patients were included at University Hospitals Leuven and at Curie Institute (n=80 Leuven; n=44 Curie) with the following histological subtypes: ductal-NOS (n=100), lobular (n=16), mucinous (n=3), tubular (n=2), others (n=3). Patients with bilateral breast cancer or with &gt;3 positive lymph nodes were excluded. Surrogate breast cancer subtypes based on IHC were defined as follows: luminal if ≥10% estrogen receptor (ER) expression; triple negative if &lt;10% ER and progesterone receptor (PR) expression and HER2 stained negative by IHC and/or FISH; HER2+ if HER2 receptor stained positive (2+ or 3+) by IHC and/or FISH. Luminal subtypes were further stratified into Luminal A-like (HER2 negative, Ki-67&lt;14%, PR≥20%) and Luminal B-like (HER2 negative or positive, Ki-67 ≥14%, PR&lt;20%). When Ki-67% was not available, tumors with grade 1 or 2 were classified as Luminal A-like and with grade 3 as Luminal B-like. IHC subtypes were compared to the BP NGS and microarray molecular subtypes (Luminal-, HER2- and Basal-type). To further stratify BP luminal type tumors, MP test was used as follows: Luminal A (BP Luminal and MP low risk) and Luminal B (BP Luminal and MP high risk).

Results

Concordance between IHC and MP/BP NGS subtyping was 75.0% (93/124), while concordance between MP/BP on NGS and microarray was 89.5% (111/124). MP/BP NGS subtyping identified more low risk Luminal A tumors compared to IHC (54.0%, (67/124) vs 44.3% (55/124)). Notably, concordance was excellent for triple-negative and, to less extent for HER2 driven tumors (Luminal B-like-HER2 positive and HER2+).

IHC vs. MP/BP NGS molecular subtyping (n=124)  MP/BP NGSIHCLuminal ALuminal BHER2-positiveBasalTotalLuminal A-like4690055Luminal B-like, HER2-negative16210037Luminal B-like, HER2-positive565016HER2-positive00303Triple negative0101213Total6737812124Microarray vs MP/BP NGS molecular subtyping (n=124)  MP/BP NGSMicroarrayLuminal ALuminal BHER2 positiveBasalTotalLuminal A6040064Luminal B7310038HER2-positive028010Basal0001212Total6737812124

Conclusion

This study shows a discordance of 25.0% between IHC and BP/MP NGS subtyping. This is in line with previous findings where IHC was compared to molecular subtyping based on microarray (Viale 2017, Whitworth 2014) underlining the complementarity of genomic testing in early stage breast cancer. Moreover, we observed a high concordance between NGS and microarray molecular subtyping, which suggests a successful translation of the MP/BP microarray test to a MP/BP NGS test.

Citation Format: Darrigues L, Slembrouck L, Mittempergher L, Delahaye LJ, Vanden Bempt I, Vander Borght S, Vliegen L, Sintubin P, Raynal V, Bohec M, Reyes C, Rapinat A, Helsmoortel C, Jongen L, Hoste G, Neven P, Wildiers H, Smeets A, Nevelsteen I, Punie K, Van Nieuwenhuysen E, Han S, Laurent C, Vincent-Salomon A, Laas-Faron E, Witteveen AT, Neijenhuis S, Glas AM, Floris G, Reyal F. Comparison of breast cancer molecular subtyping by Immunohistochemistry and by BluePrint® next generation RNA sequencing-based test at University Hospitals Leuven and Curie Institute Paris [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 P4-02-07.

Abstract P4-08-25: Decentralized beta testing of MammaPrint and BluePrint NGS kit at University Hospitals Leuven and Curie Institute Paris

Background

Many countries restrict patient material exchange to central diagnostic laboratories abroad, limiting access to assays like MammaPrint® (MP) and BluePrint® (BP). Both assays are microarray-based, with MP being prognostic for distant recurrence and BP for molecular subtyping of breast cancer (Luminal-, HER2-, and Basal-type). To increase accessibility, decentralization is required with Next Generation Sequencing (NGS) being the preferred testing platform given that most diagnostic laboratories have the technology in place. The aim of this beta testing study is to validate a previously developed and centrally validated MP and BP NGS kit for RNA samples in two large tertiary academic hospitals in Europe.

Patients and Methods

Patients with early breast cancer diagnosed at the Multidisciplinary Breast Center at University Hospitals Leuven and Curie Institute Paris were prospectively included between September 2017 and January 2018. Patients with bilateral breast cancer or presenting with more than 3 positive lymph nodes were excluded. Only patients with invasive ductal and invasive lobular carcinoma were included. Twenty tissue sections were cut from formalin-fixed, paraffin-embedded (FFPE) blocks; 10 tissue sections were analyzed at the local site using the MP and BP NGS kit, and 10 tissue sections were analyzed at Agendia using the same kit and procedure, as well as with the golden standard method (gene expression microarrays). Targeted RNA sequencing of the 70 MP and 80 BP signature genes was performed on Illumina MiSeq instruments. The raw NGS data generated at the local test sites was sent through a secure file transfer protocol server to Agendia for interpretation and comparison with microarray and NGS performed in the Agendia laboratories. We aimed for a minimum concordance rate between MP and BP outcome of 90% between each local site and Agendia's centralized site.

Results

In this study, 116 early breast cancer patients were included (73 from University Hospitals Leuven and 43 from Curie Institute). Out of these patients, 52% were MP Low Risk and 48% MP High Risk according to microarray. The patients had a BP luminal, HER2 or basal subtype in respectively 83%, 9% and 8%. Concordance between MP microarray obtained from Agendia and MP NGS obtained from the local sites was 91.4%. Concordance between MP High and Low Risk classification between NGS Leuven versus NGS Agendia was 92.1% and between NGS Curie versus NGS Agendia 95.3%. For BP subtype outcomes, the results from microarray versus NGS for all patients combined from both local sites gave a 98.3% concordance and NGS Agendia versus NGS from each local site gave a 100% concordance.

Conclusion

The MP and BP NGS kit was successfully validated in a decentralized setting, showing high concordance between results obtained at three different sites. There was a clear benefit of having well-trained NGS experienced diagnostic technical teams. The MP and BP NGS kit the first FFPE targeted RNA sequencing based multigene signature for breast cancer care, will provide a high and equal standard of MP and BP gene expression testing for breast cancer in a decentralized setting.

Citation Format: Slembrouck L, Laurent C, Delahaye LJ, Mittempergher L, Vanden Bempt I, Vander Borght S, Darrigues L, Vliegen L, Sintubin P, Raynal V, Bohec M, Reyes C, Rapinat A, Helsmoortel C, Jongen L, Hoste G, Neven P, Wildiers H, Smeets A, Nevelsteen I, Punie K, Van Nieuwenhuysen E, Han S, Salomon AV, Faron EL, Cynober T, Witteveen AT, Neijenhuis S, Glas AM, Reyal F, Floris G. Decentralized beta testing of MammaPrint and BluePrint NGS kit at University Hospitals Leuven and Curie Institute Paris [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 P4-08-25.

Abstract P1-15-04: Breast cancer recurrence and predictors for recurrence despite pathologic complete response following neoadjuvant chemotherapy

Introduction

Breast cancer patients with a high-risk tumor (for example Triple Negative Breast Cancer) who achieve a pathological complete response (pCR) following neoadjuvant chemotherapy (NACT) have a better outcome compared to patients with residual disease at surgery. This study investigated Breast Cancer Free Survival (BCFS) and predictors for distant relapse despite pCR.

Methods

Monocentric retrospective study of 243 consecutive breast cancer patients who achieved pCR (ypT0/is ypN0/N0(i+)) after treatment with NACT in UZ Leuven between 01/2000 and 08/2017. 58% had stage III breast cancer, 40% Triple Negative Breast Cancer (TNBC) and 47% HER2 pos breast cancer. BCFS was defined as any breast cancer related event (local, contra-lateral, regional, metastatic) that first appeared. Primary endpoints were frequency of BCFS and predictors for metastatic relapse: patient demographics (age, body mass index (BMI)) and tumor characteristics (TNM stage, histological type, hormonal receptor status). Secondary endpoints were breast cancer specific survival (BCSS) and overall survival (OS). Statistical analysis was performed using the Statistical Package for the Social Sciences software (SPSS, version 25). The Kaplan Meier method was used for survival analysis.

Results

Of 1167 breast cancer patients undergoing neoadjuvant treatment, 243 patients (20,8%) achieved pCR and were included. Median follow up was 57 months (range 9-252 months). 22 (9.1%) developed tumor progression; 20 (8.2%) metastatic and 2 (0.8%) contralateral. First metastatic site was the brain in 11/20 patients (55%) and 14/22 (64%) died of breast cancer. Higher clinical tumor stage at diagnosis predicted metastatic relapse (stage I-II 2.9%; stage III 12.1%). Patients with a BMI ≤25 kg/m2 had less metastatic relapse than patients with BMI &gt;25kg/m2 (3.8% versus 12.0%), better OS (94.6% vs 88.0%) and BCSS (97.7 vs 91.7%). Neither tumor type (TNBC 8.2%; HER2-pos 8.1%; HR-pos/HER2 neg 9.3%) nor younger age &lt; 36yrs (3.3% versus 8.9%) was prognostic for post-pCR relapse. There is a lower OS (mean 174m versus 231m, 95% CI 158-190m, median 208m) and BCSS (mean 191m versus 253m, 95% CI 182-200m) in cN1-3 versus cN0 disease at diagnosis.

Conclusion

Despite NACT-induced pCR, a small proportion (9.1%) will develop a metastatic relapse after a median follow-up of 57 months. We found that a higher stage at diagnosis and a higher BMI were prognostic for worse BCFS while age &lt;36 y and negative hormonal receptor status were not prognostic. cN+ at diagnosis and a BMI &gt;25 predict worse OS and BCSS.

Citation Format: Borremans K, Berteloot P, Van Nieuwenhuysen E, Han S, Hoste G, Wildiers H, Punie K, Smeets A, Nevelsteen I, Floris G, Van Ongeval C, Keupers M, Prevos R, Van Limbergen E, Menten J, Weltens C, Janssen H, Vergote I, Neven P. Breast cancer recurrence and predictors for recurrence despite pathologic complete response following neoadjuvant chemotherapy [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 P1-15-04.

Abstract P5-09-05: Hereditary breast cancer beyond BRCA: Clinical and histopathological characteristics in patients with germline CHEK2, ATM, PALB2 and TP53-mutations

Background

The introduction of multi-gene panel testing in the diagnosis of hereditary breast and ovarian cancer (HBOC) has led to an important increase in the detection of breast cancer predisposition genes other than BRCA1 and BRCA2.

Methods

All individuals who underwent HBOC-testing at our institution since the introduction of multi-gene panel testing were included (March 2016-August 2017). In this retrospective analysis, the BRCA Hereditary Cancer MASTR Plus® panel is used (Multiplicom, Belgium), with sequencing of BARD1, BRCA1, BRCA2, BRIP1, RAD51C, RAD51D, TP53, MRE11A, RAD50, NBN, FAM175A, ATM, PALB2, STK11, MEN1, PTEN, CDH1, MUTYH, CHEK2, BLM, XRCC2, EPCAM, MLH1, MSH6, PMS2, MSH2.

In breast cancer patients with a recurrent germline alteration, age and TNM stage at diagnosis, histological subtype, grade of differentiation and molecular surrogate subtype were recorded. Given the low numbers of TP53-carriers diagnosed by HBOC testing, also patients with a germline TP53-mutation diagnosed by targeted sequencing at our institution were included. Statistical analysis were performed with SPSS version 25.

Results

In 11.9 % of 2806 patients who underwent panel testing, a germline pathogenic alteration was detected. BRCA1 and BRCA2 were the most prevalent alterations, detected in respectively 3.35 and 2.92 % of patients. Germline alterations in CHEK2, ATM , PALB2 and TP53 were detected in respectively 2.5 %, 1.1 %, 0.5 % and 0.1 %. In 1 % of patients, germline alterations were retrieved that only contribute to ovarian cancer risk (BRIP, RAD51C, RAD51D). Germline DNA mismatch repair alterations were detected in 0.39 % of patients.

The median age at onset of breast cancer in patients with germline CHEK2-, ATM-, PALB2- and TP53-mutations was 47, 53, 39 and 33 years respectively. The age of breast cancer diagnosis in patients with germline TP53-alterations was significantly younger compared to patients with CHEK2-mutations (p = 0.01), ATM-mutations (p = 0.01) and PALB2-mutations (p = 0.04). In situ carcinomas were diagnosed in respectively 9 %, 11 % and 11 % of patients with CHEK2-, PALB2- and TP53-mutations. Patients with CHEK2, ATM, PALB2 and TP53-alterations were diagnosed with ≥T3-tumors in respectively 13 %, 12 %, 33 % and 22 %. Nodal status at diagnosis was negative in 40-60 % in these 4 subgroups. Upfront metastatic disease was diagnosed only in 2/43 CHEK2-carriers. More than half of the breast cancer diagnoses were luminal tumors in CHEK2-, ATM- and PALB2-carriers, while cases with germline TP53-alterations only presented with luminal cancers in 22 % in our series.

Conclusion

Almost half of the pathogenic mutations detected in HBOC-genes are alterations in genes other than BRCA1 and BRCA2. CHEK2-mutations are by far the most prevalent, followed by ATM, PALB2 and TP53.

The range of the CHEK2- and ATM-population was wider then expected at the lower-age boundary. The age of breast cancer diagnosis in patients with germline TP53-mutations was significantly younger compared to patients with CHEK2-, ATM- and PALB2-mutations. The distribution of the histological subtypes and grade of differentiation was not suggestive of a specific correlation with germline mutation status.

Citation Format: Hoste G, D'Hoore P, Legius E, Van Buggenhout G, Floris G, Wildiers H, Han SN, Van Nieuwenhuysen E, Berteloot P, Smeets A, Nevelsteen I, Weltens C, Janssen H, Van Limbergen E, Neven P, Punie K. Hereditary breast cancer beyond BRCA: Clinical and histopathological characteristics in patients with germline CHEK2, ATM, PALB2 and TP53-mutations [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 P5-09-05.

Abstract P3-05-15: PD-1 expression and its correlation with tumour infiltrating lymphocytes and Ki67 in patients with breast cancer

Introduction: Immunotherapy targeting the programmed death-1 (PD-1) checkpoint has resulted in good outcomes in patients with melanoma, non-small cell lung cancer and head and neck cancer. Recent data have showed that patients with ER/PR negative breast cancer might benefit from immunotherapy as well. We investigated PD-1 expression on tumour infiltrating lymphocytes (TILs), the presence of TILs and Ki67. Furthermore, the association of PD-1, TILs and Ki67 with time to metastasis was statistically determined.

Methods: Resection specimens of 284 patients with an ER/PR negative invasive ductal adenocarcinoma of at least 2 cm were analysed. Haematoxylin-eosin staining was performed to determine the percentage of TILs. Subsequently these percentages were divided into three groups: less than 10%, 10%-30% and more than 30% which matches with respectively 30%, 50% and 20% of the patients. PD-1 expression on the TILs of 116 patients and Ki67 expression of 185 patients were assessed using immunohistochemistry.

Results: An increase in the fraction of TILs was significantly associated with a lower risk of metastasis (hazard ratio [HR] = 0.7, 95% CI = 0.6-0.9). PD-1 expression on TILs was observed in 97% of the patients. Three patients had no expression of PD-1, this lack of PD-1 expression significantly increased the risk of metastasis (HR = 5.541, 95% CI = 1.2-25.0). No association could be noted between PD-1 expression and the amount of TILs or Ki67 (p-values of respectively 0.272 and 0.637). Similarly, no association could be noted between Ki67 and the amount of TILs (p-value = 0.1954)

Conclusion: PD-1 was expressed on TILs in patients with breast cancer. Surprisingly, no correlation was found between PD-1 expression and TILs or Ki67. However, an increased amount of TILs was significantly associated with a lower risk of metastasis.

Citation Format: Vos H, Lambein K, Floris G, Nevelsteen I, Smeets A. PD-1 expression and its correlation with tumour infiltrating lymphocytes and Ki67 in patients with breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P3-05-15.

Body mass index, age at breast cancer diagnosis, and breast cancer subtype: a cross-sectional study

PURPOSE

Evidence suggests that premenopausal obesity decreases and postmenopausal obesity increases breast cancer risk. Because it is not well known whether this is subtype dependent, we studied the association between body mass index (BMI) and age at breast cancer diagnosis, or the probability of being diagnosed with a specific breast cancer phenotype, by menopausal status.

METHODS

All patients with non-metastatic operable breast cancer from the University Hospital Leuven diagnosed between January 1, 2000 and December 31, 2013 were included (n = 7020) in this cross-sectional study. Linear models and logistic regression were used for statistical analysis. Allowing correction for age-related BMI-increase, we used the age-adjusted BMI score which equals the difference between a patient's BMI score and the population-average BMI score corresponding to the patient's age category.

RESULTS

The quadratic relationship between the age-adjusted BMI and age at breast cancer diagnosis (p = 0.0207) interacted with menopausal status (p < 0.0001); increased age at breast cancer diagnosis was observed with above-average BMI scores in postmenopausal women, and with below-average BMI scores in premenopausal women. BMI was linearly related to the probabilities of Luminal B and HER2-like breast cancer phenotypes, but only in postmenopausal women. The relative changes in probabilities between both these subtypes mirrored each other.

CONCLUSION

BMI associates differently before and after menopause with age at breast cancer diagnosis and with the probability that breast cancer belongs to a certain phenotype. The opposite effect of increasing BMI on relative frequencies of Luminal B and HER2-like breast cancers suggests a common origin.

Abstract P6-09-12: A consecutive series of early breast cancers with a low estrogen receptor expression

Background: Estrogen receptor (ER) positivity is not a dichotomous biological phenomenon and within the ER positive cohort, major differences in amount and percentage of ER expression are observed. Women with tumors expressing low ER levels have worse outcome as quantitative ER (qER) expression is predictive for benefit from hormone therapy. The aim of our study was to further characterize these patients, tumors and the distant relapse free survival (D-RFS) based upon qER expression.

Patients and Methods: In this retrospective study we included women with primary operable human epidermal growth factor receptor 2 (HER2) negative breast cancers, diagnosed between January 1st 2000 and September 30th 2015 in the Multidisciplinary Breast Center of University Hospital of Leuven. ER-low tumors ((1-33% immunohistochemical (IHC)staining, Allred proportion score (2-3/5), or H-score (&lt;100/300 + &lt;33% staining)) were compared with ER-high tumors((&gt; 33% IHC staining, Allred proportion score (4-5/5), or H-score ≥ 100/300)) and triple negative (TN) tumors (Allred proportion score (0-1/5) or H-score 0% staining). PR and HER2 were scored according to ASCO/CAP guidelines. Local and adjuvant therapies were given according to the in house guidelines. The Fisher exact text was used for comparing groups on categorical variables and the Kruskal Wallis test or Mann-Whitney test were used for continuous variables. The Kaplan-Meier method was used for estimating the interval to relapse. The log rank test was used for the comparison between groups (2-sided test with 5% significance, SAS software version 9.4).

Results: A total of 5390 women were included, 115 (2 %) were ER-low, 4658 (86.5%) were ER-high and 617 (11.5%) were TN. Average follow-up was 6 years. Patients with ER-low and TN breast tumors were younger at diagnosis compared to patients with ER-high tumors (respectively 55.5y and 56.7y vs 59.3 y; p&lt;0.001). Tumor size and nodal status did not differ between the three subgroups. ER low and TN were more often grade 3 (61% and 89% vs 24%, p&lt;0.001) which led to a higher mean NPI. Adjuvant chemotherapy was more likely to be given in ER-low and TN cases (57% and 75% vs 26%, p&lt;0.001). Women with an ER-low and TN breast cancer were more often referred for BRCA1/2/CHEK-2 testing as compared to those with an ER-high tumor (respectively 21%, 22% vs 10% p&lt;0.001); a mutation was found 5 to 7 times more in the ER low and TN group (5%-7% vs 1% p&lt;0.001). Metastatic relapse was 18% in ER low (21/114 patients), 15% in TN (90/617 patients) and 6% in ER high (283/4658 patients). Using a Kaplan Meier curve, the 5 year D-RFS was 84%, 85% and 96% in ER-low, TN and ER-high cases. The 10 year D-RFS was respectively 76%, 84% and 91% (p&lt;0.001).

Conclusion: ER-low breast cancers are rare and correlate better with TN than ER-high tumors regarding demographics, tumor grade, BRCA1/2/CHEK-2 mutation risk and breast cancer outcome.Background: Estrogen receptor (ER) positivity is not a dichotomous biological phenomenon and within the ER positive cohort, major differences in amount and percentage of ER expression are observed. Women with tumors expressing low ER levels have worse outcome as quantitative ER (qER) expression is predictive for benefit from hormone therapy. The aim of our study was to further characterize these patients, tumors and the distant relapse free survival (D-RFS) based upon qER expression.

Patients and Methods: In this retrospective study we included women with primary operable human epidermal growth factor receptor 2 (HER2) negative breast cancers, diagnosed between January 1st 2000 and September 30th 2015 in the Multidisciplinary Breast Center of University Hospital of Leuven. ER-low tumors ((1-33% immunohistochemical (IHC)staining, Allred proportion score (2-3/5), or H-score (&lt;100/300 + &lt;33% staining)) were compared with ER-high tumors((&gt; 33% IHC staining, Allred proportion score (4-5/5), or H-score ≥ 100/300)) and triple negative (TN) tumors (Allred proportion score (0-1/5) or H-score 0% staining). PR and HER2 were scored according to ASCO/CAP guidelines. Local and adjuvant therapies were given according to the in house guidelines. The Fisher exact text was used for comparing groups on categorical variables and the Kruskal Wallis test or Mann-Whitney test were used for continuous variables. The Kaplan-Meier method was used for estimating the interval to relapse. The log rank test was used for the comparison between groups (2-sided test with 5% significance, SAS software version 9.4).

Results: A total of 5390 women were included, 115 (2 %) were ER-low, 4658 (86.5%) were ER-high and 617 (11.5%) were TN. Average follow-up was 6 years. Patients with ER-low and TN breast tumors were younger at diagnosis compared to patients with ER-high tumors (respectively 55.5y and 56.7y vs 59.3 y; p&lt;0.001). Tumor size and nodal status did not differ between the three subgroups. ER low and TN were more often grade 3 (61% and 89% vs 24%, p&lt;0.001) which led to a higher mean NPI. Adjuvant chemotherapy was more likely to be given in ER-low and TN cases (57% and 75% vs 26%, p&lt;0.001). Women with an ER-low and TN breast cancer were more often referred for BRCA1/2/CHEK-2 testing as compared to those with an ER-high tumor (respectively 21%, 22% vs 10% p&lt;0.001); a mutation was found 5 to 7 times more in the ER low and TN group (5%-7% vs 1% p&lt;0.001). Metastatic relapse was 18% in ER low (21/114 patients), 15% in TN (90/617 patients) and 6% in ER high (283/4658 patients). Using a Kaplan Meier curve, the 5 year D-RFS was 84%, 85% and 96% in ER-low, TN and ER-high cases. The 10 year D-RFS was respectively 76%, 84% and 91% (p&lt;0.001).

Conclusion: ER-low breast cancers are rare and correlate better with TN than ER-high tumors regarding demographics, tumor grade, BRCA1/2/CHEK-2 mutation risk and breast cancer outcome.

Citation Format: Paredis D, Wildiers H, Vergote I, Smeets A, Van Limbergen E, Remmerie C, Laenen A, Jongen L, Floris G, Neven P. A consecutive series of early breast cancers with a low estrogen receptor expression [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 P6-09-12.

Abstract P5-14-06: The effect of adjuvant chemotherapy in a large consecutive series of ER-pos HER-2 negative early breast cancers

Background: The last EBCTCG-overview reported, in general, a significant better breast cancer outcome if adjuvant chemotherapy was added to surgery and endocrine therapy in ER-pos breast cancers, but the precise indication for adjuvant chemotherapy in this population remains controversial. We study the effect of adjuvant chemotherapy on breast cancer outcome in consecutive patients with an ER-pos HER-2 negative breast cancer treated with adjuvant antihormonal therapy.

Methods: Data were collected prospectively from consecutive patients with non-metastatic breast cancer that were primary operated between 2000 and 2012 at the University Hospitals Leuven (Belgium). A Propensity Score (PS) weighted analysis was performed to estimate the average treatment effect (ATE). The primary endpoint was recurrence free interval (RFI). Secondary endpoints were distant recurrence free interval (D-RFI) and breast cancer specific survival (BCSS). Covariates used to generate the propensity score and to include in the PS weighted analysis were age at diagnosis, body mass index, tumor size, grade, pN, lymph vessel invasion, PR, and radiotherapy. Cause-specific hazard models were fitted, using death not from breast cancer as competing risk. Treatment heterogeneity was examined by evaluating interactions of each covariate with adjuvant chemotherapy, using the Bonferroni-Holm method to correct for multiple testing.

Results: In the total cohort of 5609 patients, 4282 had a hormone sensitive HER-2 neg breast cancer and 4121 (96.2%) of these received adjuvant antihormonal therapy. Adjuvant chemotherapy was given in 1179/4121 patients (29%). Median follow-up was 8.5 years. Due to very strong differences between patients with and without adjuvant chemotherapy, a restricted PS weighted analysis was used according to a recent recommendation in the statistical literature. This analysis is based on 1750 patients with a PS between 0.1 and 0.9. In this group, 807 (46%) received adjuvant chemotherapy, 211 (12%) observed an event for RFI, 167 (10%) for D-RFI, and 108 (6%) for BCSS. Adjuvant chemotherapy was associated with better prognosis: the adjusted cumulative incidence of recurrence within 5 years was 9.7% without and 5.3% with adjuvant chemotherapy. The adjusted hazard ratio for RFI was 0.50 (95% CI 0.33-0.74). There were no strong interactions with adjuvant chemotherapy. Results for D-RFI and BCSS were similar.

Conclusion: Based on PS analysis to reduce confounding and chemotherapy indication bias, we observed clear benefit from adjuvant chemotherapy in ER-pos HER-2 negative breast cancers.

Citation Format: Van Calster B, Neven P, Wildiers H, Punie K, Jongen L, Han S, Berteloot P, Van Nieuwenhuysen E, Nevelsteen I, Smeets A, Floris G. The effect of adjuvant chemotherapy in a large consecutive series of ER-pos HER-2 negative early breast cancers [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 P5-14-06.

Abstract P4-03-01: The footprint of the aging stroma in older breast cancer patients

Introduction: Tumoral masses are not only composed of malignant cells, but also enclose a more or less ample stromal micromilieu, which has been shown to influence the cancer cell behaviour. As aging induces accumulation of senescent cells in the body, this micromilieu is thought to be different in cancers occurring in old patients compared to the younger counterparts. More specifically, senescence-related fibroblastic features, such as the Senescence Associated Secretory Profile (SASP) and the induction of Autophagy, are suspected to stimulate tumor growth and progression.

Abstract P6-09-32: The association of breast cancer subtype and breast cancer survival with parity and time since last birth

Background: Pregnancy affects breast cancer risk but it's influence on breast cancer subtype and prognosis remains controversial. We studied the effect of parity and time since last birth on breast cancer subtype and outcome in women aged ≤50 years at diagnosis.

Patients and Methods: A retrospective multivariate cohort study including all premenopausal women aged ≤50 years (N=1306) at diagnosis and primarily treated with surgery (N=1176) or neo-adjuvant chemotherapy (N=130) at University Hospitals Leuven (Jan. 2000 – Dec. 2009); local and systemic therapies were consistent with guidelines when treated. Tumor subtypes were defined by tumor grade and receptor expression for estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER-2) amplification; ER+PR+/-HER-2- cases were Luminal A- like if grade 1-2 and Luminal B like if grade 3; HER-2+ cases were Luminal HER-2 if ER+ and HER-2 like if ER-; triple negative breast cancer (TNBC) were ER-PR-HER-2-. Outcome endpoints were breast cancer subtype, disease free (DFS) and distant disease free survival (DDFS) by parity and in parous women comparing short (&lt;5 years) versus long (≥5 years) time since last birth. Statistics used were Cox proportional hazard model. Results were corrected for age at diagnosis, tumor size, lymph node status and tumor subtype.

Results: Breast cancer subtypes didn't differ between nulliparous and parous women but subtypes differed significantly in parous women by time interval since last birth (p&lt;0.001). Breast cancers within 5 years of last birth were proportionally more likely TNBC and HER-2 like compared to Luminal A (p=0.026 and p=0.003 respectively) than breast cancers ≥5 years after last birth even when corrected for age at diagnosis. After a mean follow-up period of 10 years, parous women had a better DFS compared to nulliparous women (DFS: HR 0.754; CI 0.593-0.959; p=0.021) but after correction for known prognostic factors, only a trend remained (HR 0.783; CI 0.611-1.004; p=0.054). In parous women, those with a longer time interval since last birth had a better DFS than women with a recent pregnancy (HR 0.965; CI 0.948-0.982; p&lt;0.001). However, after correction for known prognostic factors, this association was completely attenuated (HR 0.997; CI 0.972-1.023; p=0.828). Comparable results were seen for DDFS.

Conclusion: After correction for age at diagnosis, parity does not but recent birth does affect breast cancer subtype. Such tumors are proportionally more likely ER-negative namely TNBC and HER-2 like. We observed a trend for better DFS for parous women. The prognostic value of time since last birth is mostly due to tumor characteristics and age at time of diagnosis.

Citation Format: De Mulder H, Laenen A, Wildiers H, Punie K, Poppe A, Remmerie C, Nevelsteen I, Smeets A, Van Nieuwenhuysen E, Van Limbergen E, Floris G, Vergote I, Neven P. The association of breast cancer subtype and breast cancer survival with parity and time since last birth [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 P6-09-32.

Abstract P1-03-14: Assessment of stromal characteristics in ductal carcinoma in situ of the breast: An inter-observer variability study

Aim. Ductal carcinoma in situ (DCIS) is considered to be a non-obligate pre-invasive precursor of invasive ductal carcinoma. We previously showed that DCIS with a predominantly myxoid periductal stromal architecture is associated with an increased risk of both overall and invasive recurrence. The aim of this study is to determine a cut-off for the assessment of myxoid stroma and stromal inflammation in DCIS, based on inter-observer variability. Here, preliminary results of the histopathological analysis of 285 DCIS patients are presented, in which the consistency of assessment of stromal features is compared to the reproducibility of other morphological characteristics.

Methods. Hematoxylin/eosin stained tissue sections of 285 DCIS lesions were retrieved from the archives of the Department of Pathology of Leuven University Hospitals, Leuven, Belgium. The following characteristics were independently scored by two pathologists: nuclear grade, intraductal calcifications, extensive comedonecrosis, DCIS architecture, stromal architecture and stromal inflammation. Nuclear grade was scored as low, intermediate or high grade. Intraductal calcifications were scored as absent or present. Extensive comedonecrosis was defined as eosinophilic necrotic debris in &gt;50% of ductal lumina. DCIS architecture was categorized as non-solid or solid, with a cut-off at 50% of ducts presenting with solid growth. Myxoid stroma was defined as loosely arranged collagen fibers interspersed with an amorphous, slightly basophilic substance. Stromal architecture was classified into 4 categories (0%, 1-33%, 33-66% or &gt;66% myxoid stroma). By applying identical cut-offs, stromal inflammation was subdivided into absent, mild, moderate or extensive periductal inflammation. All features were dichotomized, using different cut-offs. Kappa values were determined to assess inter-observer variability.

Results. Nuclear grade was dichotomized as low grade versus intermediate/high grade (κ 0,500), and as grade low/intermediate versus high grade (κ 0,507). The kappa value for scoring myxoid stromal architecture was highest by dichotomization with a cut-off at 33% (κ 0,566), compared to κ 0,454 and κ 0,501 when using 1% and 66% as a cut-off, respectively. A similar analysis for stromal inflammation revealed that the highest kappa value was obtained by dichotomization as 'absent to mild' versus 'moderate to extensive' inflammation (κ 0,724). Dichotomization with cut-offs of 1% and 66% resulted in lower kappa values of κ 0,564 and κ 0,670, respectively. Scores for extensive comedonecrosis showed substantial agreement (κ 0,604). Scores for solid versus non-solid DCIS architecture (κ 0,507), and presence or absence of calcifications (κ 0,664) showed moderate and substantial agreement, respectively.

Conclusions. Adequate prognostic markers require robustness of assessment, i.e. low inter-observer variability and thus high reproducibility. The dichotomous assessment of stromal features in DCIS resulted in similar or even higher kappa values compared to the dichotomous scoring of other histopathological characteristics. This study demonstrates the robustness of dichotomous assessment of both stromal architecture and stromal inflammation.

Citation Format: Van Bockstal M, Lambein K, Smeets A, Nevelsteen I, Neven P, Christiaens M-R, Libbrecht L, Floris G. Assessment of stromal characteristics in ductal carcinoma in situ of the breast: An inter-observer variability study [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 P1-03-14.

Abstract P6-09-12: Breast cancer characteristics and the levonorgestrel intrauterine device. A monocentric retrospective study

OBJECTIVE: The levonorgestrel-intrauterine device (LNG-IUD) is a widely used contraceptive method. It is not clear if LNG-IUD users are more likely to develop breast cancer. Breast cancer growth through the estrogen and/or the human epidermal growth factor receptor 2 (HER2) pathway could be influenced by a continuous low systemic dose of levonorgestrel. In this study, we compare breast cancer characteristics and the receptor expression of estrogen (ER), progesterone (PR) and HER2 in women with and without a LNG- IUD at the time of diagnosis.

METHODS: In this retrospective, observational study, we included 2599 consecutive breast cancer patients who were younger than 55 years at diagnosis and treated between 2000 and 2014 in the University Hospitals Leuven for a primary invasive, non-metastatic tumor. The non LNG-IUD group was matched by age and parity at diagnosis. ER, PR and HER2 status were reported according to ASCO/CAP guidelines. The Chi-square test was used to compare receptor status between groups. All tests were two-sided, and a 5% significance level was assumed. An additional analysis was performed to detect the occurrence of HER 2 expression with or without intake of oral contraception by diagnosis in the control group.

RESULTS: 366 LNG-IUD users and 2233 women without a LNG-IUD were included. Compared to the control group, the LNG-IUD users had a lower Nottingham prognostic index (4.2 vs 4.4; p=0.048), more PR expression (79.2% vs 73.4%; p=0.021) but less HER2 expression (11.6% vs 17.2%; p=0.009). A significant higher rate of ER+PR+HER2- was observed in the LNG-IUD group (63.26 % vs 73.46%; p&lt;0.001). These differences in receptor expression were mainly observed in the age group 45-49 years at diagnosis. Additionally, a trend of more HER2 positivity associated with oral contraceptive use was noticed in the control group.

CONCLUSION: We found in a breast cancer population, matched for age and parity, significant differences in the PR and HER2 expression according to use of LNG-IUD at time of diagnosis. ER positive, PR positive and HER2 negative breast cancers are more frequently seen in LNG-IUD users. There is a trend of less HER 2 positivity in LNG-IUD users and it is more common seen in oral contraception users.

Citation Format: Borowski E, Poppe A, Laenen A, Remmerie C, Van Asten K, Nevelsteen I, Smeets A, Weltens C, Peeters S, Leunen K, Berteloot P, Amant F, Vergote I, Van Limbergen E, Christiaens M-R, Wildiers H, Floris G, Poppe W, Neven P. Breast cancer characteristics and the levonorgestrel intrauterine device. A monocentric retrospective study. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-09-12.

Abstract P6-09-11: Independent prognostic value of age depends on breast cancer subtype

Background: Young women present more often with aggressive breast cancer phenotypes and have worse prognosis. It remains controversial whether age is an independent prognostic factor in early stage breast cancer. Arbitrarily chosen age cut-off values have been proposed in different studies. Furthermore, few studies have examined the impact of breast cancer subtypes on the prognostic value of age. This abstract represents an update of a prior analysis (San Antonio Breast Cancer Symposium, December 4-8 2012, P06-07-29).

Methods: We included all primary operable female breast cancer patients from our prospectively managed database in UZ Leuven, Belgium. We assessed the effect of age on locoregional free interval (LRRFI), distant metastasis interval (DMFI) and breast cancer specific survival (BCSS). In univariate analysis, using Cox regression models, we determined the best categorization of age at diagnosis into two or three age groups by considering all possible combinations of cut-off values. Best categorization was obtained with three age groups. We further determined, using multivariate analysis (correcting for phenotype, tumor size, nodal status, adjuvant chemo -, hormone – and radiotherapy, type of surgery and procedure of axillary staging), whether age at diagnosis remains an independent predictor of outcome (LRRFI, DMFI and BCSS). We further explored whether age at diagnosis is an independent predictor of event risk (LRRFI, DMFI and BCSS) in different breast cancer subtypes. Luminal A-like (grade I or II, ER and/or PR positive, HER 2 positive), Luminal B-like (idem but grade III), Luminal HER 2 like (ER and/or PR positive, HER 2 positive), HER 2 like (ER/PR negative, HER 2 positive), triple negative (ER/PR negative, HER 2 negative).

Results: We included 4180 patients with a mean/median age of 58/57 year and with a median follow up of 8.9 year. Multivariate analysis confirmed age as an independent prognostic variable for LRRFI, DMFI and BCSS.

Results multivariable analysis with age in 3 groups (HR (95% CI) P-VALUE) LRRFIDMFIBCSSYoungest versus middle1.61 (1.18-2.18) 0.00251.54 (1.23-1.93) 0.00021.72 (1.26-2.36) 0.0007Youngest versus oldest3.45 (1.85-6.45) 0.00011.25 (0.89-1.77) 0.19821.31 (0.85-2.02) 0.2220Middle versus oldest2.15 (1.22-3.79) 0.00820.81 (0.61-1.09) 0,17060.76 (0.55-1.06) 0.1014

We found optimal cut-off values for LRRFI at 44y and 72y, for DMFI at 47y and 71y and for BCSS at 41y and 70y.

In an exploratory analysis, with age as continuous variable, by subtype we found a significant independent association between age and LRRFI (P=0.0169), DMFI (P=0.0344) in luminal A-like, LRRFI (P=0.0022) in luminal B-like and DMFI (P=0.0010) and BCSS (P=0.0053) in triple negative breast cancer. No significant associations were found in luminal HER2 and HER2 like breast cancers.

Conclusion:

This study has shown that young age is an independent prognostic factor for LRRFI, DMFI and BCSS after correction for the most important clinical prognostic factors. The prognostic effect is most important in luminal A and triple negative subtypes. Additional analyses for subtypes with age as a categorical variable will be performed and optimal cut off values will be defined.

Citation Format: Poppe A, Brouckaert O, Laenen A, Soubry A, Remmerie C, Floris G, Leunen K, Berteloot P, Amant F, Vergote I, Nevelsteen I, Smeets A, Christiaens M-R, Weltens C, Peeters S, Van Limbergen E, Wildiers H, Neven P. Independent prognostic value of age depends on breast cancer subtype. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-09-11.

Abstract P5-08-31: Withdrawal of exogenous hormones affects prognostic multigene signature results in early luminal breast cancer

Background

Early breast cancer (BC) outcomes are mainly estimated based on clinicopathological parameters and rarely include proliferation markers: (Ki-67), and multigene signatures (MGS), which are typically measured in tumors from resection specimens (RS). We believe that tumor proliferation can change within days of anti-estrogen use. However, little is known how proliferation changes after withdrawal of hormone replacement therapy (HRT) or oral anti-conception (OAC) after a core needle biopsy (CNB) shows BC. Hence, this study compares the Ki-67 labeling index and the MGS results in CNB and RS collected from a cohort of patients under OAC/HRT.

Patients and Methods

This retrospective study included consecutive women diagnosed with a grade 1-2, any pTN0-1, primary operable estrogen receptor-positive, human epidermal growth factor receptor 2-negative, and invasive ductal carcinoma between January 2013 and July 2014 at the Multidisciplinary Breast Center of University Hospitals Leuven were selected from a prospectively managed database. Ki-67 staining was performed on RS (Ki-67RS) to compare those who used HRT/OAC for at least 3 months at diagnosis and those not using OAC/HRT at diagnosis. OAC/HRT was always stopped between CNB and RS. Subsequently, we compared Ki-67 of the CNB (Ki-67CNB) with the matched RS in 15 patients with a low Ki-67RS (14%) stopping OAC/HRT after CNB; revised standard pathology confirmed absence of tumor heterogeneity in all samples. In addition, in a subset of patients (≥50 years, Ki-67RS ≤5%) we compared Ki-67 index and MGS results (MammaPrint (MP) and BluePrint, Agendia) from the CNB versus RS.

Results

193 patients with a known Ki-67RS were included; 38 patients (mean age of 55 years) were on OAC/HRT at CNB and 155 patients (mean age of 64 years) were not. The median time between stopping OAC/HRT and resection was 23 days (range 8-48 days). Ki-67RS was &lt;6%, 6-14% and ≥15% in 30.0%, 36.8% and 33.2% in the whole group of patients, respectively. These figures were 44.7%, 44.7% and 10.5% in patients on OAC/HRT at CNB diagnosis and 26.5%, 34.8% and 38.7% in patients not on OAC/HRT. This difference was significant (p&lt;0.05) for patients ≥50 years on HRT at CNB diagnosis. Four of 15 patients, which stopped OAC/HRT after the CNB (26.7%), showed a low Ki-67RS (≤14%) and had a high matched Ki-67CNB (≥15%). In another subset of four out of 15 patients we compared CNB and RS (table1) where we observed changes in both Ki-67 and MGS.

MP and Ki-67 results from matched CNB and RS. CNBRSPatientKi-67 (%)MP indexKi-67 (%)MP index150.371-20.31280.1150.043100.0650.03425-0.702-0.13Low risk is defined as MP index ≥0.0575 and/or Ki-67 ≤14.

Conclusion

Women on OAC/HRT, at diagnosis of an early luminal BC, are more likely to have a lower Ki-67RS as those not using OAC/HRT. Our findings are likely explained by a sudden decrease in sex steroids after BC diagnosis, resulting in lower proliferation markers. As such, a lower hormonal environment by withdrawing HRT/OAC at BC diagnosis might underestimate proliferation markers used for prognostic and predictive purposes. Therefore we are currently testing this hypothesis in larger cohorts.

Citation Format: Jongen L, De Vries P, Van Asten K, Lintermans A, Laenen A, Wildiers H, Floris G, Neven P. Withdrawal of exogenous hormones affects prognostic multigene signature results in early luminal breast cancer. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-08-31.

Abstract P6-10-06: The impact of body mass index on age at breast cancer diagnosis and breast cancer phenotype

Background

Evidence suggests that premenopausal obesity decreases and postmenopausal obesity increases breast cancer risk. While it has been hypothesized that carcinogenesis may be accelerated by a disrupted metabolic homeostasis in obese women, it is unclear why this dual relationship is observed. We here study whether body mass index (BMI) affects (a) age at breast cancer diagnosis and (b) the probability of being diagnosed with a specific breast cancer phenotype, taking menopausal status into account.

Patients and methods

All patients with non-metastatic operable breast cancer from UZ Leuven diagnosed between January 1, 2000 and December 31, 2013 were included (n=7020). Luminal A like (= grade 1 or 2, ER and/or PR positive, HER2 negative), Luminal B like (= grade 3 ER and/or PR positive, HER2 negative), Luminal HER2 like (ER and/or PR positive, HER2 positive), HER2 like (ER and PR negative, HER2 positive) and triple negative breast cancer (TNBC = ER and PR and HER2 negative). For statistical analysis, linear models and logistic regression were used to study respectively the association between BMI and age at diagnosis and BMI and breast cancer phenotype by menopausal status.

Results

There was a quadratic relationship between BMI and age at breast cancer diagnosis studying the overall population (p&lt;0.0001). A 5kg/m2 increase in BMI was associated with the following increases in age at diagnosis: +1.8y (95% CI 1.4-2.3y) at BMI=18, +1.2y (95% CI 0.95-1.5y) at BMI=23 and +0.6y (95% CI 0.4-0.9y) at BMI=28 (corrected for menopause). This relationship was independent of the menopausal status, ER or HER2 status, histology and breast cancer phenotype.

We observed a linear relationship between BMI and the probability of being diagnosed with Luminal B like, Luminal HER2 like and HER2 like breast cancer (table 1). This linear relationship interacts with menopausal status for Luminal B like and HER2 like breast cancers (table 1).

Table 1: Probability of being diagnosed with a certain breast cancer phenotype by BMI (linear model) and the impact of menopausal status.Effect of BMI on the probabilityLinear modelof being diagnosed withp-valueeffectLuminal A like0.4430n/aLuminal B like0.0276BMI +5kg/m2 OR 1.07 (95% CI 1.01-1.14)Luminal HER2 like0.0367BMI +5kg/m2 OR 0.91 (95% CI 0.83-1.00)HER2 like0.0219BMI +5kg/m2 OR 0.88 (95% CI 0.78-0.98)TNBC0.5454n/aInteraction with menopausal statusp-valueeffectLuminal A0,2204n/aLuminal B0,0487Premenopausal OR 0,996 (CI 0,974-1,019), p=0,7449; Postmenopausal OR 1,023 (CI 1,008-1,038), p=0.0023Luminal HER20,2571n/aHER2 like0,0031Premenopausal OR 1,020 (CI 0,983-1,059), p=0,2923; Postmenopausal OR 0,947 (CI 0,919-0,976), p=0.0004TNBC0,1638n/a

Conclusion

We could not confirm the hypothesis that increasing BMI decreases (increases) age at diagnosis in postmenopausal (premenopausal) women. Obesity does affect the probability of being diagnosed with certain breast cancer phenotypes, but for certain breast cancer phenotypes an interaction with menopause was observed. We presume a potential biological link through BMI between Luminal B and HER2 like breast cancer that needs further exploration.

Citation Format: Brouckaert O, Poppe A, Laenen A, Floris G, Leunen K, Berteloot P, Amant F, Vergote I, Smeets A, Weltens C, Peeters S, Van Limbergen E, Wildiers H, Christiaens M-R, Neven P. The impact of body mass index on age at breast cancer diagnosis and breast cancer phenotype. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-10-06.

The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014

BACKGROUND

The morphological evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer (BC) is gaining momentum as evidence strengthens for the clinical relevance of this immunological biomarker. Accumulating evidence suggests that the extent of lymphocytic infiltration in tumor tissue can be assessed as a major parameter by evaluation of hematoxylin and eosin (H&E)-stained tumor sections. TILs have been shown to provide prognostic and potentially predictive value, particularly in triple-negative and human epidermal growth factor receptor 2-overexpressing BC.

DESIGN

A standardized methodology for evaluating TILs is now needed as a prerequisite for integrating this parameter in standard histopathological practice, in a research setting as well as in clinical trials. This article reviews current data on the clinical validity and utility of TILs in BC in an effort to foster better knowledge and insight in this rapidly evolving field, and to develop a standardized methodology for visual assessment on H&E sections, acknowledging the future potential of molecular/multiplexed approaches.

CONCLUSIONS

The methodology provided is sufficiently detailed to offer a uniformly applied, pragmatic starting point and improve consistency and reproducibility in the measurement of TILs for future studies.

Genetic counselling in ALS: facts, uncertainties and clinical suggestions

The clinical approach to patients with amyotrophic lateral sclerosis (ALS) has been largely modified by the identification of novel genes, the detection of gene mutations in apparently sporadic patients, and the discovery of the strict genetic and clinical relation between ALS and frontotemporal dementia (FTD). As a consequence, clinicians are increasingly facing the dilemma on how to handle genetic counselling and testing both for ALS patients and their relatives. On the basis of existing literature on genetics of ALS and of other late-onset life-threatening disorders, we propose clinical suggestions to enable neurologists to provide optimal clinical and genetic counselling to patients and families. Genetic testing should be offered to ALS patients who have a first-degree or second-degree relative with ALS, FTD or both, and should be discussed with, but not offered to, all other ALS patients, with special emphasis on its major uncertainties. Presently, genetic testing should not be proposed to asymptomatic at-risk subjects, unless they request it or are enrolled in research programmes. Genetic counselling in ALS should take into account the uncertainties about the pathogenicity and penetrance of some genetic mutations; the possible presence of mutations of different genes in the same individual; the poor genotypic/phenotypic correlation in most ALS genes; and the phenotypic pleiotropy of some genes. Though psychological, social and ethical implications of genetic testing are still relatively unexplored in ALS, we recommend multidisciplinary counselling that addresses all relevant issues, including disclosure of tests results to family members and the risk for genetic discrimination.

FGF receptor genes and breast cancer susceptibility: results from the Breast Cancer Association Consortium

BACKGROUND

Breast cancer is one of the most common malignancies in women. Genome-wide association studies have identified FGFR2 as a breast cancer susceptibility gene. Common variation in other fibroblast growth factor (FGF) receptors might also modify risk. We tested this hypothesis by studying genotyped single-nucleotide polymorphisms (SNPs) and imputed SNPs in FGFR1, FGFR3, FGFR4 and FGFRL1 in the Breast Cancer Association Consortium.

METHODS

Data were combined from 49 studies, including 53 835 cases and 50 156 controls, of which 89 050 (46 450 cases and 42 600 controls) were of European ancestry, 12 893 (6269 cases and 6624 controls) of Asian and 2048 (1116 cases and 932 controls) of African ancestry. Associations with risk of breast cancer, overall and by disease sub-type, were assessed using unconditional logistic regression.

RESULTS

Little evidence of association with breast cancer risk was observed for SNPs in the FGF receptor genes. The strongest evidence in European women was for rs743682 in FGFR3; the estimated per-allele odds ratio was 1.05 (95% confidence interval=1.02-1.09, P=0.0020), which is substantially lower than that observed for SNPs in FGFR2.

CONCLUSION

Our results suggest that common variants in the other FGF receptors are not associated with risk of breast cancer to the degree observed for FGFR2.

Prognostic implications of lobular breast cancer histology: new insights from a single hospital cross-sectional study and SEER data

BACKGROUND

Invasive lobular breast cancer (ILC) is generally believed to have an increased risk for late relapse compared to invasive ductal breast cancer (IDC). However, the study most often referred to is a chemotherapy trial that mainly included node positive patients. We hypothesize that nodal status may influence the hazard of relapse since time of diagnosis differently in invasive ductal carcinoma (IDC) and ILC.

METHODS

Primary operable breast cancer patients from our institution diagnosed between 2000 and 2009 were studied. Multivariable analysis and subgroup analyses were performed to assess whether ILC carries a different prognosis compared to IDC. SEER data were used for external validation.

RESULTS

In lymph node negative patients, ILC carries a better prognosis regarding distant metastasis free interval (DMFI) (HR 3.242 (1.380-7.614), p = 0.0069) with a trend towards improved breast cancer specific survival (BCSS), over the entire study frame (UZ Leuven data). In lymph node positive patients, both DMFI (HR 0.466 (0.309-0.703), p = 0.0003) and BCSS (HR 0.441 (0.247-0.788), p = 0.0057) are significantly worse for ILC, especially after longer follow-up (>4-5 years) (UZ Leuven data). Similar results were found in the SEER cohort. Results remained identical when excluding screen detected cases (data not shown).

CONCLUSION

The prognostic impact of lobular histology not only depends on time since diagnosis but also on nodal status. The general believe that ILC have compromised late-term outcome compared to IDC seems untrue for the majority ( = node negative) of ILCs.

Abstract PD2-3: High body mass index (BMI) and worse response to neoadjuvant chemotherapy (NACT) by breast cancer phenotype: Own data and external validation on German breast group (GBG) patients

Introduction

It remains controversial to what extent BMI predicts pathologic complete response (pCR) following NACT. We evaluated pCR by ‘BMI category’ in different breast cancer subtypes and validated our findings in a larger dataset from the GBG.

Patients and methods

A retrospective study from UZL with consecutive breast cancer patients treated between 01-01-2000 and 31-12-2011 with different types of NACT +/- trastuzumab (dose capping BSA≥2m2) followed by surgery. pCR1 (ypT0N0) and pCR2 (ypT0/isN0) were reported in each of the intrinsic breast cancer subtypes based on ER and HER-2 expression. The effect of BMI [BMI 1(≤25kg/m2), BMI 2 (&gt;25-29,9 kg/m2) and BMI 3 (≥ 30 kg/m2)] on both pCR definitions was analyzed in univariate (Chi square test) and multivariate (logistic regression) model. Negative ER and HER-2 status were defined according to ASCO/CAP guidelines (e.g. &lt;1% ER positive tumor cells). Data were validated by GBG using different NACT +/- trastuzumab regimen (no dose capping BSA≥2m2) and similar biomarkers for defining breast cancer subtypes although ER-negative status was defined as &lt;10% ER positive tumor cells.

Results

We included 267 breast cancer cases while the GBG validation was done in 8874 cases. Multivariate model for UZL patients didn't retain ‘BMI category’ as a predictor for any of the pCR definitions. However, univariate analysis found a significant association between BMI and pCR2 (p = 0,048). The tables show the numerical differences for pCR2 in each BMI category by breast cancer subtype for UZL and GBG patients. UZL data suggest a numerical trend with lower pCR2 rate in obese women, which seems more pronounced in ER-negative breast cancer. Results from the much larger GBG dataset with more power to assess ‘BMI category’ for pCR2 in different breast cancer subtypes using a multivariate model are presented in another abstract.

Conclusion

Despite the small cohort of patients, our data suggest a lower pCR rate in obese women. External validation by GBG confirms a significant association between BMI and pCR2 in uni- and multivariate model, which is also significant for luminal A/B breast cancer. The most obvious reason for our observation was dose capping but based on the GBG dataset where dose capping was avoided where possible, we speculate that immunologic factors, the microenvironment or alternative signaling pathways may affect sensitivity to NACT +/- trastuzumab.

Numerical pCR2 differences by BMI category and breast cancer subtype for UZL patients Luminal A/BLuminal HER2HER 2 likeTNBCTotalUZLn = 111n = 40n = 39n = 75n = 267BMI 16.6% (4/61)35.0% (7/20)63.2% (12/19)31.4% (11/35)25.2% (34/135)BMI 23.1% (1/32)50.0% (7/14)56.3% (9/16)27.3% (6/22)27.3% (23/84)BMI 30% (0/18)33.3% (2/6)50.0% (2/4)15.8% (3/19)16.6% (8/48)

Numerical pCR2 differences by BMI category and breast cancer subtype for GBG patients Luminal A/BLuminal HER2HER 2 likeTNBCTotalGBGn = 3250n = 1077n = 806n = 1570n = 8847BMI 111.6%(187/1618)26.7%(147/550)47.3%(191/404)38.9%(300/772)22.5%(982/4358)BMI 210.5%(128/1026)24.9% (81/325)44.2%(111/251)37.4% (187/500)21.2%(596/2813)BMI 37.9% (48/606)22.8%(46/202)41.1%(62/151)31.9%(95/298)18.3%(312/1703)

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr PD2-3.

Breast cancer phenotype, nodal status and palpability may be useful in the detection of overdiagnosed screening-detected breast cancers

BACKGROUND

Breast cancer remains the leading cause of female cancer death despite improvements in treatment and screening. Screening is often criticized for leading to overdiagnosis and overtreatment. However, few have attempted to identify overdiagnosed cases.

PATIENTS AND METHODS

A large, consecutive series of patients treated for primary operable, screening-detected, breast cancer (n = 1610). Details from pathology and clinical reports, treatment and follow-up were available from our prospectively managed database. Univariate and multivariate Cox proportional models were used to study the prognostic variables in screening-detected breast cancers for distant metastatic and breast cancer-specific survival.

RESULTS

We included 1610 patients. The mean/median follow-up was 6.0/6.0 years. Univariate analysis: tumor size, palpability, breast cancer phenotype and nodal status were predictors of distant metastasis and breast cancer-specific death. Multivariate analysis: palpability, breast cancer phenotype and nodal status remained independent prognostic variables. Palpability differed by breast cancer phenotype.

CONCLUSION

Screening-detected breast cancer is associated with excellent outcome. Palpability, nodal status and breast cancer phenotype are independent prognostic variables that may select patients at increased risk for distant metastatic relapse and breast cancer-specific death. Overdiagnosed cases reside most likely in the nonpalpable node negative subgroup with a Luminal A phenotype.

The prognostic role of preoperative and (early) postoperatively change in CA15.3 serum levels in a single hospital cohort of primary operable breast cancers

Measuring CA15.3 serum levels in the early breast cancer setting is not recommended by current ASCO guidelines. In this large single center study, we assess the prognostic value of preoperative (n = 3746), postoperative (n = 4049) and change in (n = 3252) CA15.3, also across different breast cancer phenotypes. Preoperative, postoperative and change in CA15.3 were all significant (p = 0.0348, p < 0.0001, p < 0.0001 respectively in multivariate analysis) predictors of distant metastasis free survival. For breast cancer specific survival, only postoperative and change in CA15.3 were significant predictors (p < 0.0001 both). Multivariate prognostic models did not improve by incorporating information on preoperative CA15.3, but did improve when introducing information on postoperative CA15.3 for distant metastasis (p = 0.0365) and on change in CA15.3 for breast cancer specific survival (p = 0.0291). Change in CA15.3 impacts on prognosis (distant metastasis) differently in different breast cancer phenotypes. A decrease in CA15.3 may be informative of improved prognosis in basal like and HER2 like breast cancer.