Search

Supplementary Figure 1. AXL, MER and TYRO-3 expression in Triple Negative Breast Cancer (TNBC) cell lines and PDXs by Western blot analysis. Supplementary Figure 2. AXL mRNA expression in a large dataset of breast cancer tumors. Supplementary Figure 3. The 100 genes most correlated with AXL expression in 254 basal-like breast cancers. Supplementary Figure 4. Effect of the anti-AXL monoclonal antibody 20G7-D9 on matrix degradation by breast cancer cell lines. Supplementary Figure 5. GAS6 regulates expression of EMT markers in TNBC - Western Blot quantifications. Supplementary Figure 6. E-cadherin expression in TNBC cell lines

Suppl.Table 1: Analysis of the cellular localization, pathways and gene expression modification of the 100 genes most correlated with AXL expression in 254 basal-like breast cancers.

Purpose: AXL receptor tyrosine kinase has been described as a relevant molecular marker and a key player in invasiveness, especially in triple-negative breast cancer (TNBC).Experimental Design: We evaluate the antitumor efficacy of the anti-AXL monoclonal antibody 20G7-D9 in several TNBC cell xenografts or patient-derived xenograft (PDX) models and decipher the underlying mechanisms. In a dataset of 254 basal-like breast cancer samples, genes correlated with AXL expression are enriched in EMT, migration, and invasion signaling pathways.Results: Treatment with 20G7-D9 inhibited tumor growth and bone metastasis formation in AXL-positive TNBC cell xenografts or PDX, but not in AXL-negative PDX, highlighting AXL role in cancer growth and invasion. In vitro stimulation of AXL-positive cancer cells by its ligand GAS6 induced the expression of several EMT-associated genes (SNAIL, SLUG, and VIM) through an intracellular signaling implicating the transcription factor FRA-1, important in cell invasion and plasticity, and increased their migration/invasion capacity. 20G7-D9 induced AXL degradation and inhibited all AXL/GAS6–dependent cell signaling implicated in EMT and in cell migration/invasion.Conclusions: The anti-AXL antibody 20G7-D9 represents a promising therapeutic strategy in TNBC with mesenchymal features by inhibiting AXL-dependent EMT, tumor growth, and metastasis formation. Clin Cancer Res; 23(11); 2806–16. ©2016 AACR.

Background: Radiation therapy (RT), a novel approach to boost the anticancer immune response, has been progressively evaluated in the neoadjuvant setting in breast cancer (BC). Purpose: We aimed to evaluate immunity-related indicators of response to neoadjuvant chemoradiation therapy (NACRT) in BC for better treatment personalization. Patients and Methods: We analyzed data of the first 42 patients included in the randomized phase 2 Neo-APBI-01 trial comparing standard neoadjuvant chemotherapy (NACT) and NACRT regimen in locally advanced triple-negative (TN) and luminal B (LB) subtype BC. Clinicopathological parameters, blood counts and the derived parameters, total tumor-infiltrating lymphocytes (TILs) and their subpopulation, as well as TP53 mutation status, were assessed as predictors of response. Results: Twenty-one patients were equally assigned to each group. The pathologic complete response (pCR) was 33% and 38% in the NACT and NACRT groups, respectively, with a dose-response effect. Only one LB tumor reached pCR after NACRT. Numerous parameters associated with response were identified, which differed according to the assigned treatment. In the NACRT group, baseline hemoglobin of ≥13 g/dL and body mass index of

Introduction: Triple negative breast cancer (TNBC) is highly heterogeneous, but still most of the patients are treated by the anthracycline/taxane-based neoadjuvant therapy (NACT). Tumor-infiltrating lymphocytes (TILs) are a strong predictive and prognostic biomarker in TNBC, however not always available due to organizational and analytical issues. Peripheral blood counts, which reflect the systemic inflammatory/immune status, are easier to obtain than TILs. We investigated whether baseline white cell or platelet counts, as well as Neutrophil-to-Lymphocyte Ratio (NLR) or Platelet-to-Lymphocyte Ratio (PLR) could replace baseline TILs as predictive or prognostic biomarkers in a series of TNBC treated by standard NACT. Patients and methods: One hundred twenty patients uniformly treated by FEC/taxane NACT in a tertiary cancer care center were retrospectively analyzed. The presence of pathological complete response (pCR: ypT0/Tis, ypN0) or the presence of pCR and small residual disease (ypT0/Tis/T1ab, ypN0) were considered as good responses in data analysis. Baseline/pre-NACT blood count, NLR, PLR and TILs were evaluated as predictors of response, distant recurrence rate and distant recurrence-free survival (DRFS). TILs were assessed on breast tumor biopsies according to the recommendations of the International Immuno-Oncology Biomarker Working Group (www.tilsinbreastcancer.org). Results: TILs ≥30% and ≥1.5% were best predictors of pCR and distant recurrence risk, respectively (p=0.007, p=0.012). However, in this cohort, pCR status was not significantly associated with recurrence. Only the ensemble of patients with pCR and small residual disease had lower recurrence risk and longer survival DRFS (p=0.042, p=0.024, respectively) than the rest of the cohort (larger residual disease). The only parameter which could predict the pCR/small residual disease status was PLR: patients with values lower than 133.25 had significantly higher chance of reaching that status after NACT (p=0.045). However, no direct correlation could be established between baseline PLR and metastatic recurrence. No correlation either was found between TIL and individual blood counts, or between TILs and NLR or PLR. Conclusion: In this cohort, TILs retained their pCR predictive value, however PLR was better predictor of the ensemble of responses which had good outcome in terms of less distant recurrences or longer DRFS (pCR or small residual disease). Thus, baseline PLR is worth further, prospective investigation together with baseline TILs, as it might indicate a good TNBC response to NACT when TILs are unavailable. Citation Format: Nina Radosevic-Robin, Sejdi Lusho, Xavier Durando, Marie-Ange Mouret-Reynier, Myriam Kossai, Nathalie Lacrampe, Ioana Molnar, Frederique Penault-Llorca, Catherine Abrial. Platelet-to-lymphocyte ratio is worth using with tumor-infiltrating lymphocytes to predict good response to neoadjuvant chemotherapy in triple negative breast cancer: A study on 120 patients [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-08-24.

International audience; Vasculogenic mimicry (VM) formed by aggressive tumor cells to create vascular networks connected with the endothelial cells, plays an important role in breast cancer progression. WISP2 has been considered as a tumor suppressor protein; however, the relationship between WISP2 and VM formation remains unclear. We used the in vitro tube formation assay and in vivo immunohistochemical analysis in a mouse model, and human breast tumors were used to evaluate the effect of WISP2 on VM formation. Here we report that WISP2 acts as a potent inhibitor of VM formation in breast cancer. Enforced expression of WISP2 decreased network formation while knockdown of WISP2 increased VM. Mechanistically, WISP2 increased retention of oncogenic activators YAP/TAZ in cytoplasm, leading to decreased expression of the angiogenic factor CYR61. Studies using an in vivo mouse model and human breast tumors confirmed the in vitro cell lines data. In conclusion, our results indicate that WISP2 may play a critical role in VM and highlight the critical role of WISP2 as a tumor suppressor.

Therapies that modulate immune response to cancer, such as immune checkpoint inhibitors, began an intense development a few years ago; however, in breast cancer (BC), the results have been relatively disappointing so far. Finding biomarkers for better selection of BC patients for various immunotherapies remains a significant unmet medical need. At present, only tumour tissue programmed death-ligand 1 (PD-L1) and mismatch repair deficiency status are approved as theranostic biomarkers for programmed cell death-1 (PD-1)/PD-L1 inhibitors in BC. However, due to the complexity of tumour microenvironment (TME) and cancer response to immunomodulators, none of them is a perfect selector. Therefore, an intense quest is ongoing for complementary tumour- or host-related predictive biomarkers in breast immuno-oncology. Among the upcoming biomarkers, quantity, immunophenotype and spatial distribution of tumour-infiltrating lymphocytes and other TME cells as well as immune gene signatures emerge as most promising and are being increasingly tested in clinical trials. Biomarkers or strategies allowing dynamic assessment of BC response to immunotherapy, such as circulating/exosomal PD-L1, quantity of white/immune blood cell subpopulations and molecular imaging are particularly suitable for immunotreatment monitoring. Finally, host-related factors, such as microbiome and lifestyle, should also be taken into account when planning integration of immunomodulating therapies into BC management. As none of the biomarkers taken separately is accurate enough, the solution could come from composite biomarkers, which would combine clinical, molecular and immunological features of the disease, possibly powered by artificial intelligence., Highlights • At present, immune checkpoint inhibitors (ICIs) are the only approved immunotherapy drugs in BC. • Tumour PD-L1 and microsatellite status are current companion biomarkers for ICIs in BC; however, these need improvement. • Evaluation of tumour immune contexture and the dynamics of circulating immune cell counts are promising novel approaches. • Development of noninvasive monitoring and composite biomarkers will facilitate cancer immunotherapy, including in BC.

To find metastatic recurrence biomarkers of triple-negative breast cancer (TNBC) treated by neoadjuvant chemotherapy and anti-EGFR antibodies (NAT), we evaluated tumor genomic, transcriptomic, and immune features, using MSK-IMPACT assay, gene arrays, Nanostring technology, and TIL assessment on H&E. Six patients experienced a rapid fatal recurrence (RR) and other 6 had later non-fatal recurrences (LR). Before NAT, RR had low expression of 6 MHC class I and 13 MHC class II genes but were enriched in upregulated genes involved in the cell cycle-related pathways. Their TIL number before NAT in RR was very low (SOX2 and CXCR4. Our results indicate that high expression of cell cycle genes, combined with cold immunological phenotype, may predict strong TNBC resistance to NAT and rapid progression after it. This biomarker combination is worth validation in larger studies.

Background: Durable responses of triple negative breast cancer (TNBC) to pembrolizumab (anti-PD-1) or atezolizumab (anti-PD-L1) have been reported in the metastatic setting. Moreover, it is currently being hypothesized that immune checkpoint inhibitors might be more effective in the neoadjuvant setting, due to better preserved anti-tumor immunity in early TN disease. However, biomarkers predictive of response to anti-PD-1 or anti-PD-L1 agents, as well as biomarker-based strategies for testing those drugs in the neoadjuvant setting are still lacking. We evaluated PD-L1 protein expression and the composition of tumor-infiltrating lymphocytes (TILs) in untreated TNBC, in order to get a better insight into the TNBC sub-population(s) which would be suitable for neoadjuvant anti-PD-1 or anti-PD-L1 therapy evaluation. Methods: TNBC patients consecutively treated at the Jean Perrin Comprehensive Cancer Centre (Clermont-Ferrand, France), from 01/01/2010 to 12/31/2014, were included. FFPE tumor tissues were assessed for PD-L1 expression by immunohistochemical (IHC) laboratory-developed test (clone 28-8, Abcam), in tumor cells (tPD-L1) and in TILs. Positivity cut-offs evaluated were ≥1%, ≥5% and ≥10%. The amount CD8+, CD4+, FoxP3+ or PD-1+ TILs was determined by counting those cells, detected by IHC methods, within 5 consecutive HPFs (x400), from tumor invasive front toward tumor center. Clinical disease stage was determined using the TNM system. Results: One hundred and two TNBCs were assessed. There were 28.4%, 23.5% or 16.7% tPD-L1-positive cases (cs), for cut-offs ≥1%, ≥5% or ≥10%, respectively. Similarly, 32.4%, 15.7% or 5.9% of cs were positive for PD-L1 in TILs, using the same cut-offs. With ≥5% as cut-off, positivity for tPD-L1 significantly correlated with the amount of CD8+ (p=0.023), FoxP3+ (p=0.0036) or PD-1+ TILs (p=0.043). The same cut-off, applied to TILs, revealed significant correlations between PD-L1 positivity and the amount of each CD8+, CD4+ or PD-1+ TILs (p=0.025, 0.039 and 0.0042, respectively). Interestingly, when the ≥5% cut-off was used, tumors of T2 size were most frequently tPD-L1+ (11/41 cs, 26.8%), compared with the T1 (3/35 cs, 8.6%) and T3+T4 (3/18 cases, 16.7%) (p=0.04). With regards to TILs, with the ≥5% cut-off, the PD-L1+ cases belonged exclusively to the T1+T2 group (15/76), whereas the T3+T4 group was PD-L1-negative (0/18 cs). Other two cut-offs revealed only occasional correlations. Conclusion: This single-center, real-world TNBC cohort contained a high number of smaller tumors (T1-T2). The IHC-based PD-L1 assessment, with ≥5% as the positivity cut-off, revealed that approximately 1/4 of TNBC could be candidates for neoadjuvant anti-PD-1/anti-PD-L1 approaches. Combined with the amount of CD8+ and PD-1+ TILs, tumor PD-L1 positivity might make an easy-to-use composite biomarker for the 1st-level patient selection for PD-1 or PD-L1 inhibitors in neoadjuvant TNBC treatment. The 2nd level could exploit, for example, the assessment of mutation burden in tumors with low tPD-L1 or amount of CD8+ or PD-1+ TILs. Such tumors might be more frequent among larger TNBC (T3-T4). Citation Format: Finck W, Passildas J, Poirier C, Kwiatkowski F, Abrial C, Durando X, Penault-Llorca F, Radosevic-Robin N. The ≥5% cut-off reveals tumor PD-L1 positivity as potential selection biomarker for patient enrollment into the trials evaluating anti-PD-1 or anti-PD-L1 agents in neoadjuvant treatment of triple negative 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 P5-12-09.

Background: The role of tumour infiltrating lymphocytes (TILs) as a biomarker in non-invasive breast cancer is unclear. This meta-analysis assessed the prognostic impact of TIL levels in patients with non-invasive breast cancer. Methods: Systematic literature search was performed to identify studies assessing local recurrence in patients with non-invasive breast cancer according to TIL levels (high vs. low). Subgroup analyses per local recurrence (invasive and non-invasive) were performed. Secondary objectives were the association between TIL levels and non-invasive breast cancer subtypes, age, grade and necrosis. Odds ratios (ORs) and 95% confidence intervals (CI) were extracted from each study and a pooled analysis was conducted with random-effect model. Results: Seven studies (N = 3437) were included in the present meta-analysis. High-TILs were associated with a higher likelihood of local recurrence (invasive or non-invasive, N = 2941; OR 2.05; 95%CI, 1.03–4.08; p = 0.042), although with a lower likelihood of invasive local recurrence (N = 1722; OR 0.69; 95%CI, 0.49–0.99; p = 0.042). High-TIL levels were associated with triple-negative (OR 3.84; 95%CI, 2.23–6.61; p < 0.001) and HER2-positive (OR 6.27; 95%CI, 4.93–7.97; p < 0.001) subtypes, high grade (OR 5.15; 95%CI, 3.69–7.19; p < 0.001) and necrosis (OR 3.09; 95%CI, 2.33–4.10; p < 0.001). Conclusions: High-TIL levels were associated with more aggressive tumours, a higher likelihood of local recurrence (invasive or non-invasive) but a lower likelihood of invasive local recurrence in patients with non-invasive breast cancer., SCOPUS: ar.j, info:eu-repo/semantics/published

STING (Stimulator of Interferon Genes) is a well-known endoplasmic reticulum-anchored adaptor of the innate immunity that triggers the expression of inflammatory cytokines in response to pathogen infection. In cancer cells, this pro-inflammatory pathway can be activated by genomic DNA damage potentiating antitumor immune responses. Here we report that STING promotes cancer cell survival and resistance to genotoxic treatment in a cell-autonomous manner. Mechanistically, we show that STING partly localizes at the inner nuclear membrane in various breast cancer cell lines and clinical tumor samples, and interacts with several proteins of the DNA damage response (DDR). STING overexpression enhances the amount of chromatin-bound DNA-dependent Protein Kinase (DNA-PK) complex, while STING silencing impairs DDR foci formation and DNA repair efficacy. Importantly, this function of STING is independent of its canonical pro-inflammatory pathway. This study highlights a previously unappreciated cell-autonomous tumor-promoting mechanism of STING that opposes its well-documented role in tumor immunosurveillance.Graphical abstract

Background Triple negative breast cancer affects 10% to 20% of all women diagnosed with breast cancer. Due to its characteristics, treatment strategies are limited and metastatic recurrences are common in the first 5 years after treatment. However, not all patients affected by this disease develop metastases. Tumor-infiltrating lymphocytes have shown to be reliable predictive biomarkers of treatment response and metastatic recurrences. However, we need to develop simpler and faster ways to predict response to cytotoxic treatment and the possibility of eventual cancer relapse by identifying new biomarkers. Recently, new studies are emerging, suggesting a predictive role of circulating blood cells in different types of cancer. In this study, we will assess the correlation between tumor-infiltrating lymphocytes and different elements of the blood count in patients diagnosed with triple negative breast cancer. Methods The main objective of this study is to evaluate the correlation between the peripheral neutrophil-to-lymphocyte ratio and the amount of tumor-infiltrating lymphocytes, assessed in triple negative breast cancer patients at diagnosis. Secondary objectives include evaluation of the correlation between tumor-infiltrating lymphocytes at diagnosis and the baseline absolute neutrophil, lymphocyte, and platelet counts, as well as the platelet-to-lymphocyte ratio. The triple negative breast cancer patients will be enrolled in the PERCEPTION trial during the first year after the treatment completion. Two supplementary blood tests, at 12 months after the end of treatment and at the time of the first metastatic recurrence, will be performed. Discussion The discovery of new prognostic and predictive biomarkers is crucial for triple negative breast cancer. We set up the PERCEPTION clinical trial in order to evaluate certain blood counts as early biomarkers and to assess their correlation with tumor-infiltrating lymphocytes. Demonstration of comparative predictive and/or prognostic capacities of peripheral blood counts and tumor-infiltrating lymphocytes would allow introduction of the former as simple and cheap biomarkers in triple negative breast cancer patient management. Trial registration The PERCEPTION study has been registered in the French National Agency of Medical Security registry on the 2nd of July 2019 under the number 2019-A01861-56 and in the ClinicalTrials.org registry under the number NCT04068623.

Purpose Pathological complete response to the neoadjuvant therapy (NAT) for triple negative breast cancer (TNBC) is predictive of prolonged patient survival. Methods for early evaluation of NAT efficiency are still needed, in order to rapidly adjust the therapeutic strategy in case of initial non-response. One option for this is molecular imaging of apoptosis induced by chemotherapy. Therefore, we investigated the capacity of [18F]ML-10 PET imaging, an apoptosis radiotracer, to detect tumor cell apoptosis and early predict the therapeutic response of human TNBC. Results Initially, the induction of apoptosis by different therapies was quantified. We confirmed, in vitro, that paclitaxel or epirubicin, the fundamental cytotoxic drugs for breast cancer, induce apoptosis in TNBC cell lines. Exposure of TNBC models MDA-MB-231 and MDA-MB-468 to these drugs induced a significant increase (p < 0.01) of the apoptotic hallmarks: DNA fragmentation, membrane phospholipid scrambling, and PARP activation. Secondarily, apoptotic fraction was compared to the intracellular accumulation of the radiotracer. [18F]ML-10 accumulated in the apoptotic cells after 72 h of treatment by paclitaxel in vitro; this accumulation positively correlated with the apoptotic fraction. In vivo, [18F]ML-10 was rapidly cleared from the nontarget organs and mainly eliminated by the kidneys. Comparison of the in vivo [18F]FDG, [18F]FMISO, and [18F]ML-10 uptakes revealed that the tumor accumulation of [18F]ML-10 was directly related to the tumor hypoxia level. Finally, after the in vivo treatment of TNBC murine xenografts by paclitaxel, apoptosis was well induced, as demonstrated by the cleaved caspase-3 levels; however, no significant increase of [18F]ML-10 accumulation in the tumors was observed, either on day 3 or day 6 after the end of the treatment. Conclusions These results highlighted that PET imaging using [18F]ML-10 allows the visualization of apoptotic cells in TNBC models. Nevertheless, the increase of the chemotherapy-induced apoptotic response when using paclitaxel could not be assessed using this radiotracer in our mouse model.

Background: Epidermal growth factor receptor (EGFR) is expressed in ˜50% of triple negative breast cancer (TNBC) and has been proposed as a therapeutic target in this disease. However, trials testing EGFR blockade in TNBC failed to show significant clinical benefit. Probable reasons for such results were patient selection based on EGFR expression and the enrollment of heavily pretreated metastatic patients. Our team has conducted two neoadjuvant trials testing the activity of the anti-EGFR antibodies panitumumab (PTB) and cetuximab (CTX) combined with chemotherapy in locally advanced TNBC. Biomarkers predictive of pathological complete response (pCR) were the level of tumor cell EGFR protein expression and tumor-infiltrating lymphocytes' (TILs) profile (PMIDs 24827135, 26649807). The PTB-treated pts had a higher pCR rate (47%) than the CTX-treated pts (24%), but also a twice higher relapse rate, after 5 years of follow-up. Here we report results of genomic and TILs studies, performed in order to reveal possible determinants of recurrences in those trials. Methods: Tumor tissues sampled before and after neoadjuvant therapy (NAT) have been analyzed by next generation sequencing (NGS) using a targeted exome panel (MSK-IMPACT) of 410 cancer-related genes. Gene expression was evaluated by Affymetrix arrays. TIL density was assessed on pre-NAT samples according to Salgado et al, 2014 (PMID 25214542). The correlation between response, recurrences, genomic and TIL findings was analyzed in a case-by-case fashion. Results: Sixteen tumors that achieved pCR (PTB: 11, CTX: 5) and 23 non-pCR tumors (PTB: 11, CTX: 12) have been analyzed. For 14 non-pCR tumors (PTB: 6, CTX: 8) data have been obtained both from the pre-NAT and the post-NAT sample. Among those tumors, 6 recurred within 2 years after surgery (PTMB: 3, CTX: 3) and assays are on-going on several others that relapsed. Several genomic aberrations that potentially played a causative role in opposing to therapy were identified. We observed multiple mutations in the PI3K pathway in several non-pCR or relapsing pts. Interestingly, in a residual tumor (RT) of a non-pCR patient we found 3 different activating mutations in PIK3CA and one in PTEN. Another example of genomic selection induced by pharmacological pressure is the emergence of a HRAS G12S mutation in a RT after CTX. Additional novel findings include in-frame mutations and deletions in ARID1B and PARP1 amplification in non-pCR pts. Most of the tumors which recurred had ≤10% TILs (9/13) and only 4/13 had ≥30%. Among the tumors with a post-NAT RT but without recurrence, 17/33 had ≤10% TILs and 16/33 ≥30%. No particular link between TIL density and mutation pattern was observed. Conclusions: This is an example of a case-by-case approach where we captured the intrinsic inter-tumor heterogeneity, which is likely responsible for the different responses to EGFR-targeting in TNBC. Genes/pathways candidate of resistance to therapy are currently being validated. Citation Format: Radosevic-Robin N, Cocco E, Privat M, Abrial C, Penault-Llorca F, Scaltriti M. Potential recurrence markers of locally advanced triple negative breast cancer treated by combined neoadjuvant EGFR targeting and chemotherapy, revealed by genomic analyses and assessment of tumor-infiltrating lymphocytes [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 P2-09-29.

Background: Numerous studies have shown important impact of tumor-infiltrating lymphocytes (TILs) on natural or therapeutically-modified evolution of invasive breast cancer (IBC), however knowledge about TIL role in breast ductal carcinoma in situ (DCIS) is still limited. Because of the lack of reliable prognostic parameters, DCIS treatment is much less personalized than IBC therapy. BONBIS is a phase 3 French multicenter randomized trial designed to compare 2 schemes of adjuvant radiotherapy (adjRT) for DCIS (Azria et al, ASCO meeting 2011, TPS 131). It is accompanied by a translational study of DCIS pathobiology, aimed to discover predictive or prognostic biomarkers. Here we present results of TIL density (TIL-d) assessment, its correlation with pathobiology of the lesions and preliminary clues for further biomarker search in this DCIS cohort. Methods: Formalin-fixed, paraffin-embedded DCIS surgical specimens, obtained before adjRT, were prospectively collected and centrally reviewed for histology (architectural pattern, nuclear grade, proliferation, presence of necrosis), receptor status (ER, PR, HER2) and TIL-d. TIL-d was assessed on H&E-stained DCIS sections and reported as percentage of the DCIS specialized stroma area occupied by lymphocytes, lympho-plasmocytes and plasmocytes. Tumors were classified using the St Gallen 2011 criteria for IBC (PMID 21709140). For purpose of this study, the HER2+ category included all cases with HER2 protein expression scored 2+ and 3+, irrespective of the ERBB2 amplification status. Results: TIL-d was assessed in 495 cases, with distribution as follows: 0-4% TILs (D1): 85.5% (n=423); 5-9% TILs (D2): 9.3% (n=46); ≥10% TILs (D3): 5.2% (n=26). Molecular subclasses of those cases were: luminal A (LumA): 39% (n=192); luminal B (LumB): 25.5% (n=126), HER2+: 28.5% (n=141) and triple-negative (TN): 7% (n=33). TIL-d significantly correlated with molecular subclass: ≥5% TILs (D2) were found in 39.4% (13/33) TN, 22.7% (32/141) HER2+, 18.2% (23/126) LumB and only in 1% (2/192) LumA cases (p Citation Format: Bayol B, Tixier-Deves L, Dauplat M, Kwiatkowski F, Cayre A, Abrial C, Azria D, Penault-Llorca F, Radosevic-Robin N. Tumor-infiltrating lymphocytes in breast ductal carcinoma in situ: Correlations with tumor pathobiology in a French cohort of 495 cases (BONBIS) [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 P4-15-03.

Several studies have recently indicated the activation of the immune system against tumor cells as well as the targeting of cancer cell DNA damage repair mechanisms as effective strategies to target tumor growth. STING is a well-known DNA sensor of innate immunity mostly characterized as a transmembrane protein of various cytoplasmic organelles that senses cytosolic DNA as a danger signal and triggers inflammatory responses. A current cancer immunotherapy strategy relies on the use of STING agonists to boost the patient's immune system through a cytokine-mediated recruitment of immune cells that infiltrate and kill tumor cells. However, the role of the STING pathway in cancer is far to be fully understood as there is otherwise accumulating evidence that activation of the cGAS-STING pathway can have a deleterious outcome. We recently showed that genotoxic treatment of breast cancer PDXs and cell lines triggered the STING pathway. Genetic inhibition of this pathway in MCF7 cells increased genotoxic treatment efficacy by promoting cell death and delaying cell colony regrowth, indicating that STING pathway intrinsically promotes cell resistance to treatment. In this study, we show that STING silencing decreased cell viability in a panel of classical or PDX-derived breast cancer cell lines irrespective of their ER status and of the genotoxic treatment received. Cell fractionation indicates that part of the STING pool intrinsically resides in the nucleus of various malignant and non-malignant cells. Fluorescence and electron microscopy show that STING partly resides at the inner membrane of the nucleus, and mass-spectrometry analysis revealed that STING interacts with core proteins of the non-homologous end joining (NHEJ) DNA damage repair (DDR) complex. STING promotes NEHJ-related protein assembly with chromatin, and its silencing decreases DDR and cell viability, while STING overexpression protects cancer cells from genotoxic treatment. STING involvement in DDR is independent of the classical STING-TBK1-IFN inflammatory response, thus identifying a new functional pathway for STING. STING nuclear localization was confirmed in a panel of breast cancer patient-derived xenografts and in surgical samples from breast cancer patients that received neoadjuvant chemotherapy. Evaluation of the impact of STING expression on patient outcome via the Kaplan Meier plotter show that overall STING expression level is positively correlated with favorable outcome in breast cancer patients, however high STING expression in breast and ovarian cancer patients treated with adjuvant chemotherapy is associated with poor prognosis. These findings place STING at the crossroad of DDR and immune surveillance, two major pathways for tumorigenesis and tumor survival. Citation Format: Laura Cheradame, Ida Chiara Guerrera, Julie Gaston, Alain Schmitt, Vincent Jung, Marion Pouillard, Nina Radosevic-Robin, Mauro Modesti, Jean-Gabriel Judde, Vincent Goffin, Stefano Cairo. A non-canonical, cell-autonomous STING function protects breast cancer cells from intrinsic and genotoxic-induced DNA instability [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2038.

The correlation between immune cells and the Lauren classification subtypes and their prognostic impact in advanced gastric cancer (AGC) are unknown. Circulating natural killer (NK) cells, CD4+ and CD8+ T cells, regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs) were quantified in peripheral blood mononuclear cells (PBMCs) from 67 patients with untreated AGC enrolled in the PRODIGE 17-ACCORD 20 trial. CD56+ cells (NK), CD8+, and FoxP3+ (Treg) tumor-infiltrating lymphocytes (TILs) were assessed in tumor samples. Circulating NK and Treg proportions were significantly lower in patients with diffuse/mixed-type AGC (n = 27) than those with intestinal type (n = 40; median 6.3% vs 11.5%; p = 0.02 and median 3.3% vs 5.2%; p = 0.03, respectively). Proportions of circulating MDSC, CD4+ and CD8+ T cells were not associated with one pathological type. Among tumor-infiltrating cells, CD8+ T cells, but not NK or FoxP3+ cells, were significantly lower in diffuse/mixed-type AGC (median 21 vs 59 cells/field; p = 0.009). Patients with high circulating NK cell counts (> 17%) had a better overall survival than those with 31 cells/field) had significantly longer overall survival (HR 0.44; 95% CI [0.21–0.92]; p = 0.02). The prognostic value of CD8+ TILs was maintained after adjustment for confounding factors, including the Lauren classification (HR = 0.42; 95% CI [0.18–0.96]; p = 0.039). Diffuse/mixed-type AGC has lower rates of CD8+ TILs and circulating NK cells and Tregs than the intestinal type. This “cold tumor” phenotype may be associated with a worse outcome.

International audience; Pancreatic ductal adenocarcinoma (PDAC) is an aggressive cancer characterized by poor response to chemo- and radiotherapy due to the lack of efficient therapeutic tools and early diagnostic markers. We previously generated the non-ligand competing anti-HER3 antibody 9F7-F11 that binds to pancreatic tumor cells and induces tumor regression in vivo in experimental models. Here, we asked whether coupling 9F7-F11 with a radiosensitizer, such as monomethylauristatin E (MMAE), by using the antibody-drug conjugate (ADC) technology could improve radiation therapy efficacy in PDAC. We found that the MMAE-based HER3 antibody-drug conjugate (HER3-ADC) was efficiently internalized in tumor cells, increased the fraction of cells arrested in G2/M, which is the most radiosensitive phase of the cell cycle, and promoted programmed cell death of irradiated HER3-positive pancreatic cancer cells (BxPC3 and HPAC cell lines). HER3-ADC decreased the clonogenic survival of irradiated cells by increasing DNA double-strand break formation (based on γH2AX level), and by modulating DNA damage repair. Tumor radiosensitization with HER3-ADC favored the inhibition of the AKT-induced survival pathway, together with more efficient caspase 3/PARP-mediated apoptosis. Incubation with HER3-ADC before irradiation synergistically reduced the phosphorylation of STAT3, which is involved in chemoradiation resistance. In vivo, the combination of HER3-ADC with radiation therapy increased the overall survival of mice harboring BxPC3, HPAC cell xenografts or patient-derived xenografts, and reduced proliferation (KI67-positive cells). Combining auristatin radiosensitizer delivery via an HER3-ADC with radiotherapy is a new promising therapeutic strategy in PDAC. This article is protected by copyright. All rights reserved.

Background: Several studies have confirmed the important role of the gut microbiota in the regulation of immune functions and its correlation with different diseases, including cancer. While brain-gut and liver-gut axes have already been demonstrated, the existence of a lung-gut axis has been suggested more recently, with the idea that changes in the gut microbiota could affect the lung microbiota, and vice versa. Likewise, the close connection between gut microbiota and cancer of proximal sites (intestines, kidneys, liver, etc.) is already well established. However, little is known whether there is a similar relation when looking at world's number one cause of death from cancer—lung cancer. Objective: Firstly, this study aims to characterise the gut, lung, and upper airways (UAs) microbiota in patients with non-small cell lung cancer (NSCLC) treated with surgery or neoadjuvant chemotherapy plus surgery. Secondly, it aims to evaluate a chemotherapy effect on site-specific microbiota and its influence on immune profile. To our knowledge, this is the 1st study that will analyse multi-site microbiota in NSCLC patients along with site-specific immune response. Methods: The study is a case-controlled observational trial. Forty NSCLC patients will be divided into 2 groups depending on their anamnesis: Pchir, patients eligible for surgery, or Pct-chir, patients eligible for neoadjuvant chemotherapy plus surgery. Composition of the UAs (saliva), gut (faeces), and lung microbiota (from broncho-alveolar lavage fluid (BALF) and 3 lung pieces: “healthy” tissue distal to tumour, peritumoural tissue and tumour itself) will be analysed in both groups. Immune properties will be evaluated on the local (evaluation of the tumour immune cell infiltrate, tumour classification and properties, immune cell phenotyping in BALF; human neutrophil protein (HNP) 1–3, β-defensin 2, and calprotectin in faeces) and systemic level (blood cytokine and immune cell profile). Short-chain fatty acids (SCFAs) (major products of bacterial fermentation with an effect on immune system) will be dosed in faecal samples. Other factors such as nutrition and smoking status will be recorded for each patient. We hypothesise that smoking status and tumour type/grade will be major factors influencing both microbiota and immune/inflammatory profile of all sampling sites. Furthermore, due to non-selectivity, the same effect is expected from chemotherapy.

In non-small cell lung cancer (NSCLC), the pathologist has contributed to the development of personalized medicine from the determination of the right histological type to EGFR and ALK/ROS1 molecular screening for targeted therapies. With the development of immunotherapies, pathologists intervene forefront with programmed death-ligand 1 (PD-L1) immunohistochemical testing, companion test for pembrolizumab monotherapy, first line and complementary test to the other programmed cell death-1 (PD-1) PD-L1 inhibitors. Recently, tumor mutational burden has emerged as a promising tool to evaluate sensitivity to immunotherapy (IO). The pathologist has a crucial role in the setting of tumor mutational burden (TMB) testing for the selection and the preparation of the sample for high throughput molecular analysis, and in the first steps of the next-generation sequencing (NGS) workflow.

Multiple independent studies have shown that tumor-infiltrating lymphocytes (TIL) are prognostic in breast cancer with potential relevance for response to immune-checkpoint inhibitor therapy. Although many groups are currently evaluating TIL, there is no standardized system for diagnostic applications. This study reports the results of two ring studies investigating TIL conducted by the International Working Group on Immuno-oncology Biomarkers. The study aim was to determine the intraclass correlation coefficient (ICC) for evaluation of TIL by different pathologists. A total of 120 slides were evaluated by a large group of pathologists with a web-based system in ring study 1 and a more advanced software-system in ring study 2 that included an integrated feedback with standardized reference images. The predefined aim for successful ring studies 1 and 2 was an ICC above 0.7 (lower limit of 95% confidence interval (CI)). In ring study 1 the prespecified endpoint was not reached (ICC: 0.70; 95% CI: 0.62-0.78). On the basis of an analysis of sources of variation, we developed a more advanced digital image evaluation system for ring study 2, which improved the ICC to 0.89 (95% CI: 0.85-0.92). The Fleiss' kappa value for

Systemic therapy for triple negative breast cancer (TNBC) is mostly based upon chemotherapy. Epithelial Growth Factor Receptor (EGFR) is overexpressed in around 50% of TNBC and may play a role in its pathogenesis. Consequently, we performed a multicentric pilot Phase II neoadjuvant trial of cetuximab (anti-EGFR antibody) combined with docetaxel for patients with operable, Stage II-III TNBC. Therapy consisted of weekly cetuximab (first infusion: 400 mg/m(2), then 250 mg/m(2)) combined with six cycles of docetaxel (T: 100 mg/m(2)) q.3 weeks. Subsequently, all patients underwent surgery. The primary endpoint was pathological complete response (pCR) while clinical response, toxicity and ancillary studies were secondary endpoints. Paraffin-embedded and frozen tumor samples were systematically collected in order to identify predictive biomarkers of efficacy and resistance. From a total of 35 accrued patients, 25 were assessable for pathologic response. The pCR rate was 24% [95% CI: 7.3-40.7]. Complete clinical response rate (cCR) was observed in 22% of cases. Conservative surgery was performed in 75% of patients. Toxicity, mostly cutaneous and hematologic, was manageable. The pre-therapy ratio between CD8+ and FOXP3+ tumor-infiltrating lymphocytes equal or higher than 2.75 was predictive of pCR: 43% versus 0%, p = 0.047. Cetuximab in combination with docetaxel displays a modest activity, but acceptable toxicity as neoadjuvant therapy of operable TNBC. Similarly to previous observations using panitumumab, another anti-EGFR antibody, the immune component of the tumor microenvironment may play an important role in predicting TNBC response to the neoadjuvant therapy.

Background With the approval of anti-PD1 blocking antibodies in a growing number of indications, understanding the mechanisms responsible for potentiating response to these agents has been a critical avenue of research. However, most studies rely on the use of single cohorts and are relatively limited in their scope, thus limited their utility. Methods We created a consortium of investigators who share clinical and accompanying transcriptional data collected on the NanoString® nCounter® platform from patients treated with single agent immunotherapy to facilitate biomarker research. In this study, we apply a meta-analysis to a combined cohort of 4 lung cancer studies (n = 150) to identify transcriptional correlates of response to identify patients who are likely to experience clinical benefit. Results RNA was profiled with the NanoString IO360 and IO360 beta gene expression panels, and 43 signatures that describe facets of the immune response, tumor biology, and the tumor microenvironment were calculated. Included in these signatures is the Tumor Inflammation Signature (TIS), an investigational 18 gene signature of a suppressed adaptive immune response which enriches for clinical response to pembrolizumab. These signatures, as well as genes of interest identified in a given cohort, were compared to objective response, overall survival, or progression free survival across the cohorts. We confirm previously reported association of TIS with patient outcome and identify immune cell subsets that are also associated with response. In addition, we examine differences between patients with distinct mutational backgrounds, smoking histories, or histological classifications, as well as differences attributable to biopsy location. Conclusion This multi-cohort study allows for the development of multi-variate predictors of response to anti-PD1 monotherapy using a single research assay to transcriptionally profile the tumors. We can generate robust predictions from real-world cohorts, which may lead to the development of improved diagnostic assays to guide treatment decisions. Legal entity responsible for the study NanoString Technologies. Funding NanoString Technologies. Disclosure N. Radosevic-Robin: Research grant / Funding (institution): NanoString Technologies. J. Reeves: Full / Part-time employment: NanoString Technologies. K. Leroy: Research grant / Funding (self): NanoString Technologies. M. Duruisseaux: Research grant / Funding (self): NanoString Technologies. P. Morel: Full / Part-time employment: NanoString Technologies. M. Bhagat: Research grant / Funding (institution): NanoString Technologies. F. Penault-Llorca: Advisory / Consultancy, Research grant / Funding (institution): NanoString Technologies. D. Damotte: Research grant / Funding (institution): NanoString Technologies. F. Goldwasser: Research grant / Funding (institution): NanoString Technologies. A. Brindel: Research grant / Funding (institution): NanoString Technologies. M. Cumberbatch: Research grant / Funding (institution): NanoString Technologies. S. Ong: Full / Part-time employment: NanoString Technologies. J. Lopez: Research grant / Funding (institution): NanoString Technologies. S. Warren: Full / Part-time employment: NanoString Technologies.

Breast cancer is heterogeneous in clinical, morphological, immunohistochemical and biological features, as reflected by several different prognostic subgroups. Neoadjuvant approaches are currently used for the "in vivo" efficacy assessment of treatments. Pathological complete response (pCR) has been reported as a reliable predictive factor of survival in that setting. However, pCR remains a subject of controversy in terms of definition and its evaluation methods. In addition, its predictive value for patient outcome in various breast cancer biological subtypes has been under debate. In this review, we will present the existing definitions of pCR, the impact of its evaluation methods on its rate and the assessment of its predictive value for patient outcome in the molecular subtypes of breast cancer (luminal A and B, Triple Negative and HER2-positive).

Purpose: AXL receptor tyrosine kinase has been described as a relevant molecular marker and a key player in invasiveness, especially in triple-negative breast cancer (TNBC). Experimental Design: We evaluate the antitumor efficacy of the anti-AXL monoclonal antibody 20G7-D9 in several TNBC cell xenografts or patient-derived xenograft (PDX) models and decipher the underlying mechanisms. In a dataset of 254 basal-like breast cancer samples, genes correlated with AXL expression are enriched in EMT, migration, and invasion signaling pathways. Results: Treatment with 20G7-D9 inhibited tumor growth and bone metastasis formation in AXL-positive TNBC cell xenografts or PDX, but not in AXL-negative PDX, highlighting AXL role in cancer growth and invasion. In vitro stimulation of AXL-positive cancer cells by its ligand GAS6 induced the expression of several EMT-associated genes (SNAIL, SLUG, and VIM) through an intracellular signaling implicating the transcription factor FRA-1, important in cell invasion and plasticity, and increased their migration/invasion capacity. 20G7-D9 induced AXL degradation and inhibited all AXL/GAS6–dependent cell signaling implicated in EMT and in cell migration/invasion. Conclusions: The anti-AXL antibody 20G7-D9 represents a promising therapeutic strategy in TNBC with mesenchymal features by inhibiting AXL-dependent EMT, tumor growth, and metastasis formation. Clin Cancer Res; 23(11); 2806–16. ©2016 AACR.

International audience; Chromosomal instability (CIN), a feature of most adult neoplasms from their early stages onward, is a driver of tumorigenesis. However, several malignancy subtypes, including some triple-negative breast cancers, display a paucity of genomic aberrations, thus suggesting that tumor development may occur in the absence of CIN. Here we show that the differentiation status of normal human mammary epithelial cells dictates cell behavior after an oncogenic event and predetermines the genetic routes toward malignancy. Whereas oncogene induction in differentiated cells induces massive DNA damage, mammary stem cells are resistant, owing to a preemptive program driven by the transcription factor ZEB1 and the methionine sulfoxide reductase MSRB3. The prevention of oncogene-induced DNA damage precludes induction of the oncosuppressive p53-dependent DNA-damage response, thereby increasing stem cells' intrinsic susceptibility to malignant transformation. In accord with this model, a subclass of breast neoplasms exhibit unique pathological features, including high ZEB1 expression, a low frequency of TP53 mutations and low CIN.

Background Tumour-infiltrating lymphocytes (TILs) have been demonstrated to significantly influence prognosis and response to therapy of invasive breast cancer (IBC). Thus, it has been suggested that TIL density or/and immunophenotype could serve as biomarkers for selection of IBC patients for immunotherapy. However, much less is known about significance of TILs in breast ductal carcinoma in situ (DCIS). Methods We retrospectively investigated TIL density and immunophenotype in 96 pure DCIS and 35 microinvasive carcinomas (miCa). TIL density was assessed on H&E-stained breast biopsy sections as the percentage of tumour stromal area occupied by TILs, and classified into 4 grades: 0 (0%–9%), 1 (10–29%), 2 (30–49%) and 3 (50%–100%). TIL immunophenotype was assessed by immunohistochemistry for CD8, CD4, FoxP3, CD38 or CD20. Results Compared to pure DCIS, miCa contained significantly more cases with TIL density grade 3 (p = 0.028). Concordantly, CD8+, CD4+ and CD38+ cells were more numerous in miCa than in pure DCIS. In the pure DCIS subgroup with TIL density grades 2 and 3, all TIL subpopulations were more numerous than in the pure DCIS with TIL density grades 0 and 1, however the ratio between T-lymphocytes (CD8+ and CD4+) and B-lymphocytes (CD20+) was significantly lower (p = 0.029). On the other side, this ratio was significantly higher in miCa, in comparison with pure DCIS having TIL density grades 2 and 3 (p = 0.017). By cluster analysis of tumour cell pathobiological features we demonstrated similarity between miCa and the pure DCIS with TIL density grades 2 and 3. The only significant difference between those two categories was in the ratio of T- to B-TILs, higher in miCa. Conclusion Results indicate that TIL density level can distinguish 2 biologically different DCIS subgroups, one of which (DCIS with ≥30% TILs, the TIL-rich DCIS) is like miCa. Similarity of TIL-rich pure DCIS and miCa as well as the role of B-lymphocytes in DCIS invasiveness are worth further investigating with regards to the potential development of immunotherapy-based prevention of DCIS progression. Electronic supplementary material The online version of this article (10.1186/s12885-018-4013-6) contains supplementary material, which is available to authorized users.

Background: Our group has previously reported (SABCS 2012, abstract 1081) a 2-fold higher pathologic complete response (pCR) rate in TNBC treated with panitumumab+FEC100 (47%) than with the standard therapy alone. However, in a half of the patients (pts) a residual tumor (RT) was detected. Having in mind the detrimental impact of resistant TNBC on patient outcome, we have investigated the tumor tissue factors that might have been important in development of resistance to this treatment. Methods: 40 TNBC pts, treated with 8 cycles the anti-EGFR antibody panitumumab combined with FEC 100 for the first 4 and with docetaxel only for the last 4 cycles, were eligible. All pts underwent breast surgery after the neoadjuvant chemotherapy (NACT) completion. Formalin-fixed, paraffin-embedded tumor samples were collected before and after NACT for biomarker analysis. EGFR, HER3, IGF-1R, MET and the number of FOXP3+ or CD8+ tumor-infiltrating lymphocytes (TIL) were evaluated by immunohistochemistry (IHC). Expression of the kinases was graded semi-quantitatively (histoscores). Results: Out of 40 treated pts, 21 demonstrated a RT (53%, according to the Chevallier classification, PMID 8338056). The IHC data were available for 17 pts. Since high EGFR expression was important for pCR achievement in this study (SABC 2012), we first investigated the expressions of EGFR, HER3 and MET in the RT. The results are presented in Table 1. Table 1post-NACT RTn = 17EGFR down or flatEGFR up143HER3 upHER3 flatHER3 upHER3 flat8512MET upMET down or flatMET upMET down or flatMET upMET down or flatMET upMET down or flat17220120 Overall, among the 9 cases (cs) with increased post-NACT HER3 expression (HER3 up), only 1 had increased MET (MET up), while among the 7 cs with no change in HER3 post-CTNA (HER3 flat), 4 had MET expression increased. Interestingly, although the pre-NACT IGF-1R expression was a strong predictor of pCR (p = 0.028, SABCS 2012), no significant changes in this kinase expression were observed in the RT, compared to the pre-NACT levels. Most RT had low IGF-1R scores ( The density of TIL post-NACT correlated principally with the% of tumor mass (TM) reduction. Among the 7 pts with average reduction of 20%, there was only 1 cs with a good InSitu Immune Response (ISIR: dense TIL, with CD8+/FOXP3+ > 1.2). On the contrary, among the 9 pts with average TM reduction of 90%, 8 demonstrated a good ISIR. Conclusion: The results indicate that HER3 and MET play significant roles in development of TNBC resistance to the anti-EGFR agents. Those molecules seem to drive 2 different resistance pathways in TNBC, as it has been shown in other malignancies. As several anti-HER3 or anti-MET drugs are nowadays available, we find very important investigating the expression of HER3 and MET in each TNBC pt pre- and post-NACT, in order to rapidly detect and combat the resistance. The ISIR-stimulating agents may bring an additional therapeutic benefit to those patients. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-08-07.

Background: TNBC is a heterogenous group of tumors for some of which the Epithelial Growth Factor Receptor pathway (EGFR) may play an important role. We evaluated the efficacy and toxicity of an anti-EGFR antibody (cetuximab) combined with docetaxel, which were given to the TNBC patients (pts) in the neoadjuvant setting. A biomarker analysis accompanied this trial, aiming to identify predictors of response. Methods: 35 patients with stage II-IIIA TN breast disease were prospectively included in this multicentre pilot study. Systemic therapy (ST) consisted of 6 cycles of docetaxel (100 mg/m2) each 3 weeks, in combination with weekly cetuximab (first dose 400mg/m2 then 250 mg/m2/week) for 6 cycles. All patients underwent surgery at completion of ST. Patient characteristics : mean age 48 [28-67]; TNM: T1: 3%, T2: 73%, T3: 24%; N0: 61%, N1-N2: 39%; mean tumor size 40 mm [15-100]; SBR: grade III: 73%, grade II: 27%. The median number of cycles was for docetaxel 6 [1-6] and for cetuximab 15 [1-18]. Pathological complete response (pCR) rate was 24% according to the Chevallier and Sataloff classifications; 28% if we consider breast pCR only. Overall clinical response rate was 57% (22% CR). Conservative surgery was performed in 75% of cases. The main side effect was skin toxicity: grade II: 39%, grade III: 36%, grade IV: 3%. Other side effects were: neutropenia grade IV: 12.7%, febrile neutropenia: 1.3%, hand-foot syndrome grade III: 3%, grade II: 3%, ungueal toxicity grade III: 3%, grade II: 33%. Paraffin-embedded and frozen samples were systematically collected before and after ST for biomarker studies. Germinal BRCA1 mutations and EGFR, KRAS, BRAF and PI3KCA somatic mutations were analyzed by NGS. EGFR, MET, cytokeratins 5/6 and 8/18, PTEN, P-cadherin, ALDH1, Ki67, p53 and the number of FOXP3+ or CD8+ tumor-infiltrating lymphocytes (TIL) were evaluated by immunohistochemistry. Results: The biomarker analysis was interpretable on the samples from 21 pts (3 pCR and 18 non-pCR). We applied the ROC curve to identify the best cut-off value for Ki67, EGFR, MET, cytokeratin 5/6 and 8/18, p53, ALDH1, PTEN, P-cadherin and the FOXP3+ or CD8+ TIL counts. None of these biomarkers was predictive of pCR except for the CD8+/FOXP3+ TIL count ratio. pCR rate was higher in the pts with the ratio equal or higher than 2.75 than in the others (43% versus 0%, p = 0.047). BRCA1 mutations were detected in 16% of pts. PI3K and EGFR somatic mutations were observed in 1 and 3 patients, respectively. The presence of the mutations was not predictive of pCR. Conclusion: Similarly to the previously reported trial by our group (SABCS 2012, abstract 1081), the immune component of the tumor microenvironment plays a very important role in the TNBC response to cytotoxic therapy. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-34.

Background: Panitumumab is an antibody targeting the epidermal growth factor receptor (EGFR) for which an important role has been suggested in TNBC. Consequently, we evaluated a combination of the standard chemotherapy (FEC 100 followed by docetaxel) with panitumumab as neoadjuvant therapy of operable TNBC. Complete pathologic response (pCR) was the primary endpoint, with clinical response, toxicity, and outcome as secondary endpoints. An investigation of biomarkers possibly predictive of pCR accompanied this trial. Here we focus on tumor recurrence, after a median follow up of 33 months [25-40] as on April, 1, 2013. Methods: Sixty patients (pts) with stage II-IIIA TNBC were prospectively included. Systemic neoadjuvant treatment (ST) consisted of the anti-EGFR antibody panitumumab combined with FEC 100, followed by 4 cycles of docetaxel. All pts underwent surgery after ST completion. Patient characteristics: median tumor size: 40 mm [20-120]; SBR grade III: 71.7%; pCR rate: 55.3% and 46.8% (the Sataloff and the Chevallier classifications, respectively). Paraffin-embedded and frozen tumor samples were collected before and after ST for biomarker analysis. EGFR, IGF-1R, MET, cytokeratins 5/6 and 8/18, PTEN, P-cadherin, ALDH1, Ki-67, p53, tumoral FOXP3 expression and the number of FOXP3+ or CD8+ tumor-infiltrating lymphocytes (TIL) were evaluated by immunohistochemistry. Results :.We have observed 9 recurrences: 1 local and 8 distant recidives, including 1 both local and distant. The distant recidives (metastases) were as follows: brain (4 pts); brain and lungs (1 pt); lungs only (1 pt), pleura (1 pt); liver (1 pt). 6 out of the 8 metastatic pts died and all were non-pCR post-ST. The 2 alive pts had brain metastases, but reached a pCR after the ST. Among the 9 relapsed pts 6 were 55 years old or less at the diagnosis. Seven out of those 9 pts had tumors with the clinical size equal or higher than 4 cm. As previously reported (SABCS 2012, abstract 1081), the pCR-predictive biomarkers in this study were high CD8+ TIL count (p = 3.4.10−6) and high ratio between the CD8+ and FOXP3+ TIL counts (CD8+/FOXP3+ > 1.23, p = 8.5.10−5). With this in mind, we have evaluated whether those parameters, assessed before or after the ST, could predict the recurrences. No difference was found in the preoperative CD8+ and the FOXP3+ TIL counts, as well as in the CD8+/FOXP3+ ratio, between the patiens who have recurred and the others. Conclusion : As it has been reported in previous studies, in our cohort of TNBC pts, the relapses occurred early after the administration of the last systemic treatment. The patients who relapsed died rapidly and most of them have not reached pCR after the ST. In addition, half of the metastatic pts got brain deposits. This implies that research on the resistance factors in TNBC should focus on those important for seeding of the “sanctuaries”, like brain. This research is ongoing in our group. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-14-19.

Our understanding of the early natural history of epithelial ovarian carcinoma is limited by the access to early lesions as the disease is very often diagnosed at advanced stages. The incessant ovulation theory from the last century that indicated the ovary as the site for the initiation of high-grade serous cancers is contrary to the newly emerging idea that ovarian cancer could arise from the distal fallopian tube. In view of the recent pathological and molecular studies, we propose to discuss the genesis of high-grade serous ovarian cancer.

Background: The incidence of infra-centimetric breast cancer (BC) is increasing due to mass screening. The management remains controversial opposing locoregional treatment only based on low stage and potentially aggressive tumor biology (especially for triple negative and HER2 positive tumors). The objectives of the prospective multicenter ODISSEE study were i) to describe daily practice management of pT1ab BC, ii) to identify potential new biomarkers and iii) to describe long term outcome (10-year follow up). Methods: 616 patients with unifocal pT1a, b pN0M0 BC were included after surgery from May 2009 to March 2010 in 116 centers. Paraffin blocks were available for central pathology review in 350 patients. Due to the small tumor size, tissue microarray (TMA) construction was finally achieved for 287 cases. Collection of additional tumor blocks is ongoing. Review of the cases was performed independently by two pathologists and discordant cases were reviewed under a multihead microscope. Histological characteristics (histological type, tumor size and grade, presence of in situ component, presence of lymphovascular invasion) were assessed on whole tissue sections. Immunohistochemical detection of ER, PR, HER2, Ki67, EGFR, cytokeratin 5/6, Bcl-2 was performed on TMAs. Results: Clinicopathological characteristics of the 287 centrally reviewed cases were similar to those of the 616 patients included in the cohort. Median age: 60.1 years (range [31–89] years). Median tumor size: 7.8 mm (range [1–10] mm) with 11% pT1a and 89% pT1b. Histological types: 72% ductal, 11% lobular and 17% other types (among which 53% of tubular). Minor in situ component was found in 36% of the cases, and invasive carcinoma with an extensive in situ component in 1.9%. Most of tumors were histological grade I and II (52% and 40% respectively), 8% were grade III. Proliferation was low as assessed by mitotic count (low 82%, intermediate 13% and high 5%) or by Ki67 index ( Conclusions: The pT1a, bpN0M0 BC ODISSEE patients were mainly grade I or II, low proliferative HR+/HER2− tumors (i.e. the so-called luminal A subtype). Adjuvant treatment decision was mainly based on the presence of an aggressive phenotype such as HER2+ and triple negative tumors. A further exploration of ER/luminal pathway (FoxA1, GATA3, androgen receptor, Cox-2) and of PI3K/Akt/mTOR pathway (eIF4E, 4EBP, p70S6) is ongoing and will be presented. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-21-01.

Patient stratification has been instrumental for the success of targeted therapies in breast cancer. However, the molecular basis of metastatic breast cancer and its therapeutic vulnerabilities remain poorly understood. Here we show that PML is a novel target in aggressive breast cancer. The acquisition of aggressiveness and metastatic features in breast tumours is accompanied by the elevated PML expression and enhanced sensitivity to its inhibition. Interestingly, we find that STAT3 is responsible, at least in part, for the transcriptional upregulation of PML in breast cancer. Moreover, PML targeting hampers breast cancer initiation and metastatic seeding. Mechanistically, this biological activity relies on the regulation of the stem cell gene SOX9 through interaction of PML with its promoter region. Altogether, we identify a novel pathway sustaining breast cancer aggressiveness that can be therapeutically exploited in combination with PML-based stratification., Targeting PML in acute promyelocytic leukaemia has changed the outcome of patients with this disease. Here, the authors demonstrated that PML is also a potential therapeutic target in breast cancer where it specifically regulates cancer initiating cells and tumour progression through the transcriptional regulation of SOX9.