Search

BackgroundIn this phase II clinical trial, we evaluated the efficacy of the nonanthracycline combination of carboplatin and nab-paclitaxel in early stage triple-negative breast cancer (TNBC).Patients and MethodsPatients with newly diagnosed stage II–III TNBC (n = 69) were treated with neoadjuvant carboplatin (area under the curve 6) every 28 days for four cycles plus nab-paclitaxel (100 mg/m2) weekly for 16 weeks. Pathological complete response (pCR) and residual cancer burden (RCB) were analyzed with germline mutation status, tumor-infiltrating lymphocytes (TILs), TNBC molecular subtype, and GeparSixto immune signature (GSIS).ResultsSixty-seven patients were evaluable for safety and response. Fifty-three (79%) patients experienced grade 3/4 adverse events, including grade 3 anemia (43%), neutropenia (39%), leukopenia (15%), thrombocytopenia (12%), fatigue (7%), peripheral neuropathy (7%), neutropenia (16%), and leukopenia (1%). Twenty-four patients (35%) had at least one dose delay, and 50 patients (72%) required dose reduction. Sixty-three (94%) patients completed scheduled treatment. The responses were as follows: 32 of 67 patients (48%) had pCR (RCB 0), 10 of 67 (15%) had RCB I, 19 of 67 (28%) had RCB II, 5 of 67 (7%) had RCB III, and 1 of 67 (2%) progressed and had no surgery. Univariate analysis showed that immune-hot GSIS and DNA repair defect (DRD) were associated with higher pCR with odds ratios of 4.62 (p = .005) and 4.76 (p = .03), respectively, and with RCB 0/I versus RCB II/III with odds ratio 4.80 (p = .01). Immune-hot GSIS was highly correlated with DRD status (p = .03), TIL level (p < .001), and TNBC molecular subtype (p < .001). After adjusting for age, race, stage, and grade, GSIS remained associated with higher pCR and RCB class 0/I versus II/III with odds ratios 7.19 (95% confidence interval [CI], 2.01–25.68; p = .002) and 8.95 (95% CI, 2.09–38.23; p = .003), respectively.ConclusionThe combination of carboplatin and nab-paclitaxel for early stage high-risk TNBC showed manageable toxicity and encouraging antitumor activity. Immune-hot GSIS is associated with higher pCR rate and RCB class 0/1. This study provides an additional rationale for using nonanthracycline platinum-based therapy for future neoadjuvant trials in early stage TNBCs. Clinical trial identification number: NCT01525966Implications for PracticePlatinum is an important neoadjuvant chemotherapy agent for treatment of early stage triple-negative breast cancer (TNBC). In this study, carboplatin and nab-paclitaxel were well tolerated and highly effective in TNBC, resulting in pathological complete response of 48%. In univariate and multivariate analyses adjusting for age, race, tumor stage and grade, “immune-hot” GeparSixto immune signature (GSIS) and DNA repair defect (DRD) were associated with higher pathological complete response (pCR) and residual cancer burden class 0/1. The association of immune-hot GSIS with higher pCR holds promise for de-escalating neoadjuvant chemotherapy for patients with early stage TNBC. Although GSIS is not routinely used in clinic, further development of this immune signature into a clinically applicable assay is indicated.

Favorable outcomes have been associated with high densities of tumor infiltrating lymphocytes (TILs) such as cytotoxic ([Formula: see text]) T cells. However, the clinical significance of the spatial distribution of TILs is less well understood. We have developed novel statistical techniques to characterize the spatial distribution of TILs at various length scales. These include a box counting method that we call “occupancy” and novel applications of fractal dimensions. We apply these techniques to the spatial distribution of [Formula: see text] T cells in the tumor microenvironment of tissue resected from 35 triple negative breast cancer patients. We find that there is a distinct difference in the spatial distribution of [Formula: see text] T cells between good clinical outcome (no recurrence within at least 5 years of diagnosis) and poor clinical outcome (recurrence within 3 years of diagnosis). The statistical significance of the difference between good and poor outcome in the occupancy, fractal dimension (FD), and FD difference of [Formula: see text] T cells is comparable to that of the [Formula: see text] T cell density. Even when we randomly exclude some of the cells so that the images have the same cell density, we still find that the fractal dimension at short length scales is correlated with cancer recurrence, implying that the actual spatial distribution of [Formula: see text] cells, and not just the [Formula: see text] cell density, is associated with clinical outcome. The occupancy and FD difference indicate that the [Formula: see text] T cells are more spatially dispersed in good outcome and more aggregated in poor outcome. We discuss possible interpretations.

Purpose: Emerging data suggest immune checkpoint inhibitors have reduced efficacy in heavily pretreated triple-negative breast cancers (TNBC), but underlying mechanisms are poorly understood. To better understand the phenotypic evolution of TNBCs, we studied the genomic and transcriptomic profiles of paired tumors from patients with TNBC. Experimental Design: We collected paired primary and metastatic TNBC specimens from 43 patients and performed targeted exome sequencing and whole-transcriptome sequencing. From these efforts, we ascertained somatic mutation profiles, tumor mutational burden (TMB), TNBC molecular subtypes, and immune-related gene expression patterns. Stromal tumor-infiltrating lymphocytes (stromal TIL), recurrence-free survival, and overall survival were also analyzed. Results: We observed a typical TNBC mutational landscape with minimal shifts in copy number or TMB over time. However, there were notable TNBC molecular subtype shifts, including increases in the Lehmann/Pietenpol-defined basal-like 1 (BL1, 11.4%–22.6%) and mesenchymal (M, 11.4%–22.6%) phenotypes, and a decrease in the immunomodulatory phenotype (IM, 31.4%–3.2%). The Burstein-defined basal-like immune-activated phenotype was also decreased (BLIA, 42.2%–17.2%). Among downregulated genes from metastases, we saw enrichment of immune-related Kyoto Encyclopedia of Genes and Genomes pathways and gene ontology (GO) terms, and decreased expression of immunomodulatory gene signatures (P < 0.03) and percent stromal TILs (P = 0.03). There was no clear association between stromal TILs and survival. Conclusions: We observed few mutational shifts, but largely consistent transcriptomic shifts in longitudinally paired TNBCs. Transcriptomic and IHC analyses revealed significantly reduced immune-activating gene expression signatures and TILs in recurrent TNBCs. These data may explain the observed lack of efficacy of immunotherapeutic agents in heavily pretreated TNBCs. Further studies are ongoing to better understand these initial observations. See related commentary by Savas and Loi, p. 526

Background Alteration of the PI3K/AKT/mTOR pathway is a common genomic abnormality detected in triple-negative breast cancer (TNBC). Everolimus acts synergistically with eribulin in TNBC cell lines and xenograft models. This phase I trial was designed to test the safety and tolerability of combining eribulin and everolimus in patients with metastatic TNBC. Methods The primary objective of this study was to evaluate the safety and toxicities of the combination. Patients with metastatic TNBC who had up to four lines of prior chemotherapies were enrolled. The combination of eribulin and everolimus was tested using three dosing levels: A1 (everolimus 5 mg daily; eribulin 1.4 mg/m2 days 1 and 8 every 3 weeks), A2 (everolimus 7.5 mg daily; eribulin 1.4 mg/m2, days 1 and 8 every 3 weeks), and B1 (everolimus 5 mg daily; eribulin 1.1 mg/m2 days 1 and 8 every 3 weeks). Results Twenty-seven patients with median age 55 years were enrolled. Among 8 evaluable patients who received dose level A1, 4 had dose-limiting toxicities (DLTs). Among 3 evaluable patients treated with dose level A2, 2 had DLTs. Among 12 evaluable patients who received dose level B1, 4 had DLTs. The DLTs were neutropenia, stomatitis, and hyperglycemia. Over the study period, 59% had a ≥ grade 3 toxicity, 44% had ≥ grade 3 hematologic toxicities, and 22% had grade 4 hematologic toxicities. The most common hematological toxicities were neutropenia, leukopenia, and lymphopenia. Thirty-three percent had grade 3 non-hematologic toxicities. The most common non-hematological toxicities were stomatitis, hyperglycemia, and fatigue. The median number of cycles completed was 4 (range 0–8). Among 25 eligible patients, 9 patients (36%) achieved the best response as partial response, 9 (36%) had stable disease, and 7 (28%) had progression. The median time to progression was 2.6 months (95% CI [2.1, 4.0]), and median overall survival (OS) was 8.3 months (95% CI [5.5, undefined]). Conclusion Eribulin 1.1 mg/m2 days 1 and 8 every 3 weeks with everolimus 5 mg daily was defined as the highest dose with acceptable toxicity (RP2D). The combination is safe, and efficacy is modest. A post hoc analysis showed that participants that used dexamethasone mouthwash stayed on treatment for one additional cycle. Trial registration ClinicalTrials.gov, NCT02120469. Registered 18 April 2014

Regulatory T (Treg) cells play a major role in the development of an immunosuppressive tumor microenvironment. The origin of intratumoral Treg cells and their relationship with peripheral blood Treg cells remain unclear. Treg cells consist of at least three functionally distinct subpopulations. Here we show that peripheral blood CD45RA-FOXP3hi Treg cells (Treg II cells) are phenotypically closest to intratumoral Treg cells, including in their expression of CCR8. Analyses of T cell antigen receptor repertoires further support the hypothesis that intratumoral Treg cells may originate primarily from peripheral blood Treg II cells. Moreover, the signaling responsiveness of peripheral blood Treg II cells to immunosuppressive, T helper type 1 (TH1) and T helper type 2 (TH2) cytokines reflects intratumoral immunosuppressive potential, and predicts future relapse in two independent cohorts of patients with breast cancer. Together, our findings give important insights into the relationship between peripheral blood Treg cells and intratumoral Treg cells, and highlight cytokine signaling responsiveness as a key determinant of intratumoral immunosuppressive potential and clinical outcome.

Background: Metastatic triple negative breast cancer (mTNBC) is a heterogeneous disease with poor prognosis. Molecular evolution of TNBC through chemotherapy selection pressure is well recognized but poorly understood. PI3K/AKT/mTOR is one of the most commonly identified oncogenic-driver pathways in breast cancer. The current study is designed to understand the genomic and transcriptomic changes, focusing on the PI3K/AKT/mTOR pathway alterations in paired primary and metastatic TNBCs. Results: Genomic analysis of 7 paired specimens identified 67 known mutations including those from the following signaling pathways: cell cycle, p53, PI3K/AKT/mTOR, RAS/MAPK, and RTK/GF. Principle coordinate analysis (PCoA) identified 4 distinctive molecular groups based on the gene expression patterns of PI3K/AKT/mTOR pathway. Key differentially-expressed genes included AKT3, GSK3B, GNA11, PI3KR1, and GNAQ. Importantly, AKT-targeted therapy showed efficacy in a patient-derived xenograft (PDX) model of TNBC in vivo. Conclusion: Genomic discordance of paired primary and metastatic TNBCs was identified, with significant increase in tumor proliferation pathways seen in metastases. Among the differentially expressed genes, AKT3 can potentially serve as a target for novel combination therapy for treatment of metastatic TNBC. Methods: Paired specimens from 10 patients with TNBCs were identified through an IRB-approved protocol (2002–2015). FoundationOneTM sequencing was performed for genomic profiling, and Affymetrix Human Genechip 2.0st was used for mRNA expression profiling. The similarity among samples was calculated based on Pearson correlation coefficients, which were used to construct hierarchical clustering and heat maps.

While tumor infiltration by CD8+ T cells is now widely accepted to predict outcomes, the clinical significance of intratumoral B cells is less clear. We hypothesized that spatial distribution rather than density of B cells within tumors may provide prognostic significance. We developed statistical techniques (fractal dimension differences and a box-counting method ‘occupancy’) to analyze the spatial distribution of tumor-infiltrating lymphocytes (TILs) in human triple-negative breast cancer (TNBC). Our results indicate that B cells in good outcome tumors (no recurrence within 5 years) are spatially dispersed, while B cells in poor outcome tumors (recurrence within 3 years) are more confined. While most TILs are located within the stroma, increased numbers of spatially dispersed lymphocytes within cancer cell islands are associated with a good prognosis. B cells and T cells often form lymphocyte clusters (LCs) identified via density-based clustering. LCs consist either of T cells only or heterotypic mixtures of B and T cells. Pure B cell LCs were negligible in number. Compared to tertiary lymphoid structures (TLS), LCs have fewer lymphocytes at lower densities. Both types of LCs are more abundant and more spatially dispersed in good outcomes compared to poor outcome tumors. Heterotypic LCs in good outcome tumors are smaller and more numerous compared to poor outcome. Heterotypic LCs are also closer to cancer islands in a good outcome, with LC size decreasing as they get closer to cancer cell islands. These results illuminate the significance of the spatial distribution of B cells and LCs within tumors.

This study addresses the core issue facing a surgical team during breast cancer surgery: quantitative prediction of tumor likelihood including estimates of prediction error. We have previously reported that a molecular probe, Laser Raman spectroscopy (LRS), can distinguish healthy and tumor tissue. We now report that combining LRS with two machine learning algorithms, unsupervised k-means and stochastic nonlinear neural networks (NN), provides rapid, quantitative, probabilistic tumor assessment with real-time error analysis. NNs were first trained on Raman spectra using human expert histopathology diagnostics as gold standard (74 spectra, 5 patients). K-means predictions using spectral data when compared to histopathology produced clustering models with 93.2–94.6% accuracy, 89.8–91.8% sensitivity, and 100% specificity. NNs trained on k-means predictions generated probabilities of correctness for the autonomous classification. Finally, the autonomous system characterized an extended dataset (203 spectra, 8 patients). Our results show that an increase in DNA|RNA signal intensity in the fingerprint region (600–1800 cm−1) and global loss of high wavenumber signal (2800–3200 cm−1) are particularly sensitive LRS warning signs of tumor. The stochastic nature of NNs made it possible to rapidly generate multiple models of target tissue classification and calculate the inherent error in the probabilistic estimates for each target.

BackgroundSingle-agent pembrolizumab treatment of hormone receptor-positive metastatic breast cancer (MBC) has demonstrated modest clinical responses. Little is known about potential biomarkers or mechanisms of response to immune checkpoint inhibitors (ICIs) in patients with HR+ MBC. The present study presents novel immune correlates of clinical responses to combined treatment with CDK4/6i and ICI.MethodsA combined analysis of two independent phase I clinical trials treating patients with HR+ MBC was performed. Patients treated with the combination of the CDK4/6i palbociclib+the ICI pembrolizumab+the aromatase inhibitor (AI) letrozole (palbo+pembro+AI) were compared with patients treated with pembrolizumab+AI (pembro+AI). Peripheral blood mononuclear cells collected at pretreatment, 3 weeks (cycle 2 day 1) and 9 weeks (cycle 4 day 1) were characterized by high-parameter flow cytometry to assess baseline immune subset composition and longitudinal changes in response to therapy.ResultsIn the peripheral blood, higher pretreatment frequencies of effector memory CD45RA+ CD8+ T cells and effector memory CD4+ T cells were observed in responders to palbo+pembro+AI. In contrast, this was not observed in pembro+AI-treated patients. We further characterized T-cell subsets of effector-like killer cell lectin-like receptor subfamily G member 1 (KLRG1+) ICOS+ CD4+ T cells and KLRG1+ CD45RA+ CD8+ T cells as baseline biomarkers of response. In comparison, pretreatment levels of tumor-infiltrating lymphocyte, tumor mutation burden, tumor programmed death-ligand 1 expression, and overall immune composition did not associate with clinical responses. Over the course of treatment, significant shifts in myeloid cell composition and phenotype were observed in palbo+pembro+AI-treated patients, but not in those treated with pembro+AI. We identified increased fractions of type 1 conventional dendritic cells (cDC1s) within circulating dendritic cells and decreased classical monocytes (cMO) within circulating monocytes only in patients treated with palbociclib. We also demonstrated that in palbociclib-treated patients, cDC1 and cMO displayed increased CD83 and human leukocyte antigen-DR isotype (HLA-DR) expression, respectively, suggesting increased maturation and antigen presentation capacity.ConclusionsPre-existing circulating effector CD8+ and CD4+ T cells and dynamic modulation of circulating myeloid cell composition denote response to combined pembrolizumab and palbociclib therapy for patients with HR+ MBC.Trial registration numberNCT02778685 and NCI02648477.

Background: Response to neoadjuvant therapy (NT) predicts progression-free and overall survival in triple negative breast cancer (TNBC). Carboplatin has shown efficacy in patients with TNBC. The current phase II prospective neoadjuvant trial was designed to decrease toxicities and improve efficacy. Methods: Patients with TNBC received carboplatin (carb) and nab-paclitaxel (nab). Pre-NT biopsies were procured to evaluate for biological predictors of pathological complete response (pCR). Newly diagnosed stage II-III patients with TNBC were treated with 4 cycles of carb (AUC 6, day 1 of 28 day cycle) and weekly nab 100 mg/m2 x 16. Targeted accrual goal is 70. RNA extracted from formalin fixed paraffin embedded (FFPE) biopsies pre-NT was tested for MammaPrint/BluePrint and custom Agilent full genome microarrays for gene expression (GE, by Agendia Inc). The raw gMeanSignal was log2 transformed and normalized to the 75thpercentile for GE analysis. Association between MammaPrint/ BluePrint results and pCR was tested by Fisher exact test. The linear model from R limma package was applied. Ingenuity Pathway Analysis (IPA) was applied to assess functional pathways associated with pCR. Cellular distribution by CIBERSORT analysis was carried out to estimate the abundance of 22 different cell types in each patient sample, and test whether the distribution of cell types is different between pCR and non-responders. Results: A total of 64 patients were enrolled. Two patients were deemed ineligible (Her2+), and three were too early, resulting in 59 patients evaluable for pathological response. The pCR rate was 47% (RCB0, 28/59). Eight patients had RCB I. RCB0 plus RCBI reached 61%. Sufficient quality RNA and DNA were available from the first 43 of 55 pts with TNBC. 44/59 (75%) required dose modifications (mostly hematologic), 5 patients had grade 3 peripheral neuropathy (PN), 3 had grade 2 PN, and 3 patients had grade 2 LFTs. In the 53 pts with GE assessment, pCR was inversely associated with luminal BluePrint type (p=0.04). With fold change >1.5 and p-value < 0.05, 36 genes were differentially expressed (DE) in TNBC. CIBERSORT analysis suggested that T-cell regulatory cells (TREGS) were associated with pCR in TNBC, and 5 cell types (plasma cells, TREGS, macrophage, dendritic cells and neutrophils) presented differently between all pCR and non-pCRs with P-value Conclusions: The combination of carboplatin and nab-paclitaxel given in the neoadjuvant setting reached a promising pCR rate of 47%. The MammaPrint non-luminal BluePrint subtype was predictive of pCR in TNBC. Preliminary analysis suggested that a 36-gene signature for TNBC was associated with pCR. CIBERSORT analysis revealed 5 cell types with different abundance between the pCR and non-responders, suggesting the need to target the tumor microenvironment. Citation Format: Yuan Y, Frankel P, Li M, Kruper L, Jones V, Treece T, Waisman J, Yim J, Tumyan L, Schmolze D, Hurria A, Yeon C, Mortimer J, Somlo G. Phase II trial of neoadjuvant carboplatin and nab-paclitaxel in patients with locally advanced 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 P1-15-07.

Purpose: Thyroid cancer is frequently difficult to diagnose due to an overlap of cytologic features between malignant and benign nodules. This overlap leads to unnecessary removal of the thyroid in patients without cancer. While providing some improvement over cytopathologic diagnostics, molecular methods frequently fail to provide a correct diagnosis for thyroid nodules. These approaches are based on the difference between cancer and adjacent thyroid tissue and assume that adjacent tissues are the same as benign nodules. However, in contrast to adjacent tissues, benign thyroid nodules can contain genetic alterations that can be found in cancer. Experimental Design: For the development of a new molecular diagnostic test for thyroid cancer, we evaluated DNA methylation in 109 thyroid tissues by using genome-wide single-base resolution DNA methylation analysis. The test was validated in a retrospective cohort containing 65 thyroid nodules. Results: By conducting reduced representation bisulfite sequencing in 109 thyroid specimens, we found significant differences between adjacent tissue, benign nodules, and cancer. These tissue-specific signatures are strongly linked to active enhancers and cancer-associated genes. Based on these signatures, we developed a new epigenetic approach for thyroid diagnostics. According to the validation cohort, our test has an estimated specificity of 97% [95% confidence interval (CI), 81–100], sensitivity of 100% (95% CI, 87–100), positive predictive value of 97% (95% CI, 83–100), and negative predictive value of 100% (95% CI, 86–100). Conclusions: These data show that epigenetic testing can provide outstanding diagnostic accuracy for thyroid nodules. See related commentary by Mitmaker et al., p. 457

Context.— Thyroid nodules are a common clinical problem. Cytologic evaluation via fine-needle aspiration is often employed in the diagnostic workup, and rapid on-site assessment of adequacy can help ensure an adequate sample is obtained. However, rapid on-site assessment of adequacy only examines part of the sample, a part that may not then be available for ancillary testing. Moreover, the procedure is time-consuming and poorly reimbursed. Objective.— To develop an automatable fluorescence-based image analysis system for assessing the adequacy of thyroid fine-needle aspirations that uses the entire aspirated sample. Design.— There were 12 previously diagnosed cases that served as a training set, and 11 cases were used for validation of an image analysis algorithm. The samples were fluorescently stained and imaged using a fluorescent microscope. The images were assessed for adequacy by an image analysis algorithm. Following image analysis, a ThinPrep slide was prepared and blindly scored by a cytopathologist. The standard and computer-derived results were then compared. Results.— The algorithm was optimized using the 12 cases in the training set and then applied to the 11 test cases. A total of 8 of 8 adequate samples in the test group were correctly scored as adequate, and 2 of 3 cases that were inadequate were correctly scored as inadequate by the algorithm. One case was erroneously designated as not adequate by the algorithm. Conclusions.— Our results demonstrate the feasibility of automating thyroid adequacy assessment using a fluorescent labeling technique followed by computer image analysis.

Background: Androgen receptor (AR) targeted therapy and immunotherapy represent one of the most promising strategies for metastatic triple negative breast cancer (mTNBC), which accounts for 15-20% of all breast cancers. As a nonsteroidal selective androgen receptor modulator (SARM), GTx-024 demonstrated preclinical activity in AR+ TNBC PDX model. Pembrolizumab is a highly selective humanized monoclonal antibody of the programmed cell death 1 receptor (PD-1). The complementary modes of action and low potential for overlapping toxicity make the combination promising in patients with AR+ mTNBC. Trial Design: This is an open-label Phase 2 study for AR+ mTNBC. Eligible participants receive pembrolizumab 200mg IV every 3 weeks in combination with GTx-024 18mg po daily. Eligibility Criteria: Eligible patients must have AR+ (>10%, 1+ by IHC) TNBC; failed up to 2 lines of therapy in metastatic setting; and have measurable disease per RECIST1.1. Patients are excluded if they have had prior checkpoint inhibitors or AR targeted agents. Patients with current or prior use of testosterone, testosterone-like agents, androgenic compounds, or anti-androgens (including systemic steroids and immunosuppressive medications)are excluded, as well as current or prior history of noninfectious pneumonitis requiring systemic steroid therapy. Specific Aims: The primary objective is to evaluate the safety/tolerability of GTx-024 and pembrolizumab and determine the response rate (CR or PR via RECIST 1.1) in patients with advanced AR+ TNBC. We will use clinical benefit rate (CBR), duration of response (DOR), PFS, and OS to test the efficacy of this novel drug combination. Statistical Design: A Simon's MiniMax two-stage Phase 2 design will be utilized. Based on the previously reported response rate associated with single agent pembrolizumab (19%), we consider a response rate of 19% for the combination as discouraging, and a 39% response rate as encouraging. As a result, we will initially accrue 15 patients (including 6 patients from safety lead-in treated at the tolerable dose). If 2 or fewer patients respond, we will stop accrual for futility. Otherwise, the study will accrue an additional 14 patients for a total of 29 patients. With 29 patients, if only 8 or fewer respond (≤27.6%), the study will be considered discouraging unless secondary evidence of clinical benefit is substantial. With more than 8 patients responding out of the 29 patients, the combination would be considered promising. This design has 85% power to declare a true response rate of 39% as promising (power), and a 10% probability of declaring a true 19% response rate as encouraging (type I error). The probability of early termination if the true response rate is 19% is 44%. Target Accrual: 29 Study Contact: Yuan Yuan MD PhD, City of Hope Comprehensive Cancer Center; Duarte, CA 91030; Email: yuyuan@coh.org Citation Format: Yuan Y, Frankel P, Synold T, Lee P, Yost S, Martinez N, Tang A, Mendez B, Schmolze D, Apple S, Hurria A, Waisman J, Somlo G, Tank N, Sedrak M, Mortimer J. A phase II clinical trial of the combination of pembrolizumab and selective androgen receptor modulator GTx-024 in patients with advanced androgen receptor positive triple negative 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 OT1-05-02.

Purpose: Therapeutic strategies against hormonal receptor–positive (HR+)/HER2+ breast cancers with poor response to trastuzumab need to be optimized. Experimental Design: Two HR+/HER2+ patient-derived xenograft (PDX) models named as COH-SC1 and COH-SC31 were established to explore targeted therapies for HER2+ breast cancers. RNA sequencing and RPPA (reverse phase protein array) analyses were conducted to decipher molecular features of the two PDXs and define the therapeutic strategy of interest, validated by in vivo drug efficacy examination and in vitro cell proliferation analysis. Results: Estrogen acted as a growth driver of trastuzumab-resistant COH-SC31 tumors but an accelerator in the trastuzumab-sensitive COH-SC1 model. In vivo trastuzumab efficacy examination further confirmed the consistent responses between PDXs and the corresponding tumors. Integrative omics analysis revealed that mammalian target of rapamycin (mTOR) and ERα signaling predominantly regulate tumor growth of the two HR+/HER2+ PDXs. Combination of the dual mTOR complex inhibitor MLN0128 and anti-HER2 trastuzumab strongly suppressed tumor growth of COH-SC1 PDX accompanied by increasing ER-positive cell population in vivo. Instead, MLN0128 in combination with antiestrogen fulvestrant significantly halted the growth of HR+/HER2+ cancer cells in vitro and trastuzumab-resistant COH-SC31 as well as trastuzumab-sensitive COH-SC1 tumors in vivo. Conclusions: Compared with the standard trastuzumab treatment, this study demonstrates alternative therapeutic strategies against HR+/HER2+ tumors through establishment of two PDXs coupled with integrative omics analyses and in vivo drug efficacy examination. This work presents a prototype of future “co-clinical” trials to tailor personalized medicine in clinical practice. Clin Cancer Res; 24(2); 395–406. ©2017 AACR.

Background: Prognosis for triple-negative breast cancer (TNBC) patients remains poor, due in part to the lack of effective targeted therapies in the advanced setting. Emerging clinical data indicates reduced efficacy of immune checkpoint inhibitors in heavily pre-treated TNBC, but the underlying mechanisms are poorly understood. To better understand the immune phenotypic evolution of paired TNBCs, we studied the genomic and transcriptomic profiles of tumors from patients undergoing treatment for TNBC. Methods: We analyzed primary and recurrent TNBCs from 55 poor-risk patients, including 44 paired primary-metastatic samples and 11 paired metastatic tumors. FoundationOne® and RNAseq was successful on 89 specimens and 97 specimens, respectively. In addition to somatic alterations, FoundationOne® provided tumor mutational burden (TMB). From RNAseq, we ascertained the TNBC molecular subtypes, and the mRNA expression of immune-related genes. Stromal tumor-infiltrating lymphocytes (stromal TILs), recurrence-free survival, and overall survival were also studied. Results: From FoundationOne® sequencing, a mutational landscape typical of TNBCs was observed across both primary and recurrent disease specimens, with TP53 mutated in 82.0% of specimens, and BRCA1 and BRCA2 mutated in 4.5% and 16.9% of specimens, respectively. Sample profiles revealed minimal shifts in copy number alterations and TMB over time, however, notable TNBC subtype shifts were observed between primary and recurrent tumors. These included an increase in the Lehmann/Pietenpol-defined basal-like 1 phenotype (BL1, 12.8% to 20.9%), an increase in the mesenchymal phenotype (M, 12.8% to 20.9%), and a significant decrease in the immunomodulatory phenotype (IM, 27.1% to 2.3%). Similarly, tumors exhibited a downward shift in gene expression delineating the Burstein-defined basal-like immune-activated phenotype (BLIA, 37.0% to 14.3%). Composite expression of immunomodulatory gene signatures representative of Th1/Th2 responses, IFNg-related inflammation, M1/M2 macrophage activation and suppression, etc., was decreased in the recurrent tumors compared to the primaries (p = 0.01), and histopathology-derived percent stromal TILs were significantly decreased in the recurrent TNBCs (p = 0.02). However, higher stromal TILs (≥30%) were not associated with improved overall survival when measured in primary specimens (p = 0.15), or with the time from relapse to death when measured in recurrent specimens (p = 0.65) in this cohort of immunotherapy-naïve patients. Conclusion: In this retrospective study of paired TNBCs, significant transcriptomic phenotype shifts were observed as patients progressed, while only minor genomic shifts were seen. Selective immune profiling showed significantly reduced TILs and immune-activating gene expression signatures in recurrent TNBCs, which may explain the lack of efficacy of immunotherapeutic agents in heavily pretreated TNBCs. Further studies are ongoing to understand the proteomic landscape shifts in TNBCs over time and to identify novel targeted agents appropriate for recurrent disease. Citation Format: Yuan Y, Yost SE, Chang C-W, Yoh KE, Johnson RM, Schmolze D, Liang J, Hutchinson KE. Comprehensive profiling of poor-risk paired primary and recurrent triple-negative breast cancers reveals immune phenotype shifts [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 PD5-07.

Gross intraoperative assessment can be used to ensure negative margins at the time of surgery. Previous studies of this technique were conducted before the introduction of consensus guidelines defining a “positive” margin. We performed a retrospective study examining the accuracy of this technique since these guidelines were published. We identified all specimens that were grossly examined at the time of breast conserving surgery from January 2014 to July 2020. Gross and final microscopic diagnoses were compared and the performance of intraoperative examination was assessed in terms of false positive and false negative rates. Logistic regression models were used to examine the effect of clinicopathologic covariates on discordance. 327 cases were reviewed. Gross exam prompted re-excision in 166 cases (61%). The rate of false negative discordance was 8.6%. In multivariate analysis, multifocality on final pathology was associated with discordance. We consider the false negative rate acceptable for routine clinical use; however, there is an ongoing need for more accurate methods for the intraoperative assessment of margins.

The evolutionary changes in immune profiles of triple negative breast cancer (TNBC) are not well understood, although it is known that immune checkpoint inhibitors have diminished activity in heavily pre-treated TNBC patients. This study was designed to characterize immune profile changes of longitudinal tumor specimens by studying immune subsets of tumor infiltrating lymphocytes (TILs) in paired primary and metastatic TNBC in a cohort of “poor outcome” (relapsed within 5 years) patients. Immune profiles of TNBCs in a cohort of “good outcome” (no relapse within 5 years) patients were also analyzed. Immune subsets were characterized for CD4, CD8, FOXP3, CD20, CD33, and PD1 using immuno-fluorescence staining in stroma, tumor, and combined stroma and tumor tissue. TIL subsets in “good outcome” versus “poor outcome” patients were also analyzed. Compared with primary, metastatic TNBCs had significantly lower TILs by hematoxylin and eosin (H&E) staining. Stromal TILs (sTILs), but not tumoral TILs (tTILs) had significantly reduced cytotoxic CD8+ T cells (CTLs), PD1+ CTLs, and total PD1+ TILs in metastatic compared with matched primary TNBCs. Higher PD1+ CTLs, PD1+CD4+ helper T cells (PD1+TCONV) and all PD1+ T cells in sTILs, tTILs and total stromal and tumor TILS (s+tTIL) were all associated with better prognosis. In summary, TIL subsets decrease significantly in metastatic TNBCs compared with matched primary. Higher PD1+ TILs are associated with better prognosis in early stage TNBCs. This finding supports the application of immune checkpoint inhibitors early in the treatment of TNBCs.

While targeted therapies exist for human epidermal growth factor receptor 2 positive (HER2 +) breast cancer, HER2 + patients do not always respond to therapy. We present the results of utilizing a biophysical mathematical model to predict tumor response for two HER2 + breast cancer patients treated with the same therapeutic regimen but who achieved different treatment outcomes. Quantitative data from magnetic resonance imaging (MRI) and 64Cu-DOTA-trastuzumab positron emission tomography (PET) are used to estimate tumor density, perfusion, and distribution of HER2-targeted antibodies for each individual patient. MRI and PET data are collected prior to therapy, and follow-up MRI scans are acquired at a midpoint in therapy. Given these data types, we align the data sets to a common image space to enable model calibration. Once the model is parameterized with these data, we forecast treatment response with and without HER2-targeted therapy. By incorporating targeted therapy into the model, the resulting predictions are able to distinguish between the two different patient responses, increasing the difference in tumor volume change between the two patients by > 40%. This work provides a proof-of-concept strategy for processing and integrating PET and MRI modalities into a predictive, clinical-mathematical framework to provide patient-specific predictions of HER2 + treatment response.

BackgroundPoint of care (POC) glucose meters are widely used in hospitals to aid in monitoring blood glucose levels. We have evaluated the clinical performance of our meters in the critically ill population.MethodsTo evaluate the accuracy of glucose meters results, as obtained by our end users, we m

The goal of this study is to characterize the genomic and immune profiles of metaplastic breast cancer (MpBC) and identify the association with survival through an analysis of archived tumor tissue. A next-generation sequencing-based mutational assay (Onco-48) was performed for 21 MpBC patients. Clinicopathologic characteristics were captured, including relapse free survival (RFS) and overall survival (OS). Immunohistochemistry (IHC) for CD3, CD4, CD8, and programmed death-ligand 1 (PD-L1) was also performed. Recurrence free survival (RFS) at 5 years was 57% (95% CI 0.34-0.75) and overall survival (OS) at 5 years was 66% (95% CI 0.41-0.82). The most commonly altered genes were TP53 (68.4%, 13/19), PIK3CA (42.1%, 8/19), and PTEN (15.8%, 3/19. For patients with PIK3CA mutations, RFS and OS were significantly worse than for those without (HR 5.6, 95% CI 1.33-23.1 and HR 8.0, 95% CI 1.53-41.7, respectively). Cox regression estimated that PD-L1 expression was associated with worse RFS and OS (HR 1.08, 95% CI 1.01-1.16 and HR 1.05, 95% CI 1.00-1.11, respectively, for an absolute increase in PD-L1 expression of 1%). In conclusion, PIK3CA mutation and PD-L1 expression confer poor prognosis in this cohort of patients with MpBC.

Breast cancer diagnosis requires a pathologist to analyze the histology slides under various magnifications. An automated diagnosis method to aid pathologists that is magnification independent will significantly save time, reduce cost and mitigate subjectivity and errors in current histopathological diagnosis procedures. This paper presents a deep learning network, called MVPNet and a customized data augmentation technique, called NuView, for magnification independent diagnosis. MVPNet is tailored to tackle the most common issues (diversity, relatively small size of datasets and manifestation of diagnostic biomarkers at various magnification levels) with breast cancer histology data to perform the classification. The network simultaneously analyzes local and global features of a given tissue image. It does so by viewing the tissue at varying levels of relative nuclei sizes. MVPNet has significantly less parameters than standard transfer learning deep models with comparable performance and it combines and processes local and global features simulatenously for effective diagnosis. Additionally, NuView extracts tumor nuclei location and points the attention of MVPNet to the informative region specifically. The method gives an average magnification independent classification accuracy of 92.2% as compared to 83% reported in literature on the BreaKHis database.

Motivation: A major goal of biomedical research is to identify molecular features associated with a biological or clinical class of interest. Differential expression analysis has long been used for this purpose; however, conventional methods perform poorly when applied to data with high within class heterogeneity. Results: To address this challenge, we developed EMDomics, a new method that uses the Earth mover’s distance to measure the overall difference between the distributions of a gene’s expression in two classes of samples and uses permutations to obtain q-values for each gene. We applied EMDomics to the challenging problem of identifying genes associated with drug resistance in ovarian cancer. We also used simulated data to evaluate the performance of EMDomics, in terms of sensitivity and specificity for identifying differentially expressed gene in classes with high within class heterogeneity. In both the simulated and real biological data, EMDomics outperformed competing approaches for the identification of differentially expressed genes, and EMDomics was significantly more powerful than conventional methods for the identification of drug resistance-associated gene sets. EMDomics represents a new approach for the identification of genes differentially expressed between heterogeneous classes and has utility in a wide range of complex biomedical conditions in which sample classes show within class heterogeneity. Availability and implementation: The R package is available at http://www.bioconductor.org/packages/release/bioc/html/EMDomics.html Contact: abeck2@bidmc.harvard.edu Supplementary information: supplementary data are available at Bioinformatics online.

1069 Background: Androgen receptor (AR) targeting therapy has shown single agent activity in triple negative breast cancer (TNBC). GTx-024, a nonsteroidal selective androgen receptor modulator (SARM), demonstrated preclinical and clinical activity in AR+ breast cancer. The current study is designed to test the safety and efficacy of GTx-024 and pembrolizumab in patients with AR+ metastatic TNBC (mTNBC). Methods: This is an open-label phase 2 study for AR+ mTNBC. Eligible participants receive pembrolizumab 200mg IV every 3 weeks in combination with GTx-024 18mg po daily. Key eligibility criteria include patients with AR+ ( > 10%, 1+ by IHC); mTNBC; ECOG 0-1; measurable disease per RECIST 1.1. Patients are excluded if they had prior checkpoint inhibitors or AR targeted agents. The primary objective is to evaluate the tolerability of GTx-024 and pembrolizumab, and determine the response rate. Results: Seventeen patients were enrolled in the study. One patient was ineligible due to previously undiagnosed brain metastases. Ten of 16 patients had visceral metastasis (lung or liver), and 15% of patients had received ≥ 3 previous lines of therapy for mTNBC. Among 16 patients evaluable for response, 2 patients achieved a best response of partial response (PR), 2 patient had stable disease (SD, 18 and 19 weeks ), 11 patients had progressive disease (PD), and 1 patient is too early for restaging imaging. Durable response was found in 1 patient. Grade 3 toxicities include 1 diarrhea and 1 dry skin. Grade 2 adverse events include 3 elevated liver function, 1 adrenal insufficiency, 1 hyperthyroidism, 1 palpitation, 1 diarrhea, 1 hyperhydrosis, 1 hot flashes and 1 headache. Three patients had dose delay and two patients had dose reduction. Conclusions: AR targeted therapy GTx-024 combined with pembrolizumab is well tolerated with clinical activity. Clinical trial information: NCT02971761.

Rapid histopathological examination of surgical specimen margins using fluorescence microscopy during breast conservation therapy has the potential to reduce the rate of positive margins on postoperative histopathology and the need for repeat surgeries. To assess the suitability of imaging modalities, we perform a direct comparison between confocal fluorescence microscopy and multiphoton microscopy for imaging unfixed tissue and compare to paraffin-embedded histology. An imaging protocol including dual channel detection of two contrast agents to implement virtual hematoxylin and eosin images is introduced that provides high quality imaging under both one and two photon excitation. Corresponding images of unfixed human breast tissue show that both confocal and multiphoton microscopy can reproduce the appearance of conventional histology without the need for physical sectioning. We further compare normal breast tissue and invasive cancer specimens imaged at multiple magnifications, and assess the effects of photobleaching for both modalities using the staining protocol. The results demonstrate that confocal fluorescence microscopy is a promising and cost-effective alternative to multiphoton microscopy for rapid histopathological evaluation of ex vivo breast tissue.

The assessment of protein expression in immunohistochemistry (IHC) images provides important diagnostic, prognostic and predictive information for guiding cancer diagnosis and therapy. Manual scoring of IHC images represents a logistical challenge, as the process is labor intensive and time consuming. Since the last decade, computational methods have been developed to enable the application of quantitative methods for the analysis and interpretation of protein expression in IHC images. These methods have not yet replaced manual scoring for the assessment of IHC in the majority of diagnostic laboratories and in many large-scale research studies. An alternative approach is crowdsourcing the quantification of IHC images to an undefined crowd. The aim of this study is to quantify IHC images for labeling of ER status with two different crowdsourcing approaches, image-labeling and nuclei-labeling, and compare their performance with automated methods. Crowdsourcing- derived scores obtained greater concordance with the pathologist interpretations for both image-labeling and nuclei-labeling tasks (83% and 87%), as compared to the pathologist concordance achieved by the automated method (81%) on 5,338 TMA images from 1,853 breast cancer patients. This analysis shows that crowdsourcing the scoring of protein expression in IHC images is a promising new approach for large scale cancer molecular pathology studies.

A heavily pretreated patient with triple negative breast cancer distinguished by cutaneous metastases received p53MVA vaccine in combination with pembrolizumab. Her cutaneous metastases regressed and after 2 cycles of therapy, a skin biopsy showed a complete pathological response. Systemic response was confirmed with restaging CT and bone scans. Activation of p53-specific T cell responses and elevation of multiple immune response genes in peripheral blood correlated with the rapid clinical response which lasted for 6 months after the initiation of combined therapy.

e12041 Background: Metaplastic breast cancer (MpBC) is a rare malignancy with aggressive clinical presentation and poor prognosis. The goal of this study is to characterize MpBC at the genomic level with targeted exome sequencing and correlate with clinical-pathological characteristics. Methods: We have Identified 23 patients (pts) with a diagnosis of MpBC through the COH Cancer Registry from 1995 to 2013. A next generation sequencing based panel (Onco48) was performed on 21 pts on using the Ion AmpliSeq™ Kit. Molecular testing for two pts was performed at FoundationOne®. Patient’s clinical-pathological characteristics were also captured (age, race, stage, chemotherapy, radiation therapy history, and 5-year relapse free survival (RFS)). Results: Of the 21 specimens tested with COH Onco48, 3 failed to yield results. Of the 20 pts included in the analysis, 1 pt did not have survival information available. The median age at time of diagnosis was 68 years (range 35-93). All of the tumors were TNBC. The subtypes of MpBC were: squamous (35%), spindle (30%), mixed squamous and spindle (15%) and chondroid (20%). A total of 70% of pts received (neo)adjuvant chemotherapy, and 65% received radiation therapy. Median follow-up of alive pts is 63 months. Median RFS and OS have not been reached for entire cohort with a lower 95% of PFS of 22.2 months. The most commonly mutated genes included TP53 (65%), PIK3CA (45%), PTEN (15%).In a univariate analysis, the association of p53, PTEN, PIK3CA mutations and breast cancer specific survival. RFS at 2 years was 33% (0.03-0.64) for those with PIK3CA mutations, and 100% for those without (p