Background: Neoadjuvant chemotherapy (NAC) is the standard treatment for early-stage triple negative breast cancer (TNBC). The primary endpoint of NAC is a pathological complete response (pCR). NAC results in pCR in only 30–40% of TNBC patients. Tumor-infiltrating lymphocytes (TILs), Ki67 and phosphohistone H3 (pH3) are a few known biomarkers to predict NAC response. Currently, systematic evaluation of the combined value of these biomarkers in predicting NAC response is lacking. In this study, the predictive value of markers derived from H&E and IHC stained biopsy tissue was comprehensively evaluated using a supervised machine learning (ML)-based approach. Identifying predictive biomarkers could help guide therapeutic decisions by enabling precise stratification of TNBC patients into responders and partial or non-responders. Methods: Serial sections from core needle biopsies (n = 76) were stained with H&E and immunohistochemically for the Ki67 and pH3 markers, followed by whole-slide image (WSI) generation. The serial section stains in H&E stain, Ki67 and pH3 markers formed WSI triplets for each patient. The resulting WSI triplets were co-registered with H&E WSIs serving as the reference. Separate mask region-based CNN (MRCNN) models were trained with annotated H&E, Ki67 and pH3 images for detecting tumor cells, stromal and intratumoral TILs (sTILs and tTILs), Ki67+, and pH3+ cells. Top image patches with a high density of cells of interest were identified as hotspots. Best classifiers for NAC response prediction were identified by training multiple ML models and evaluating their performance by accuracy, area under curve, and confusion matrix analyses. Results: Highest prediction accuracy was achieved when hotspot regions were identified by tTIL counts and each hotspot was represented by measures of tTILs, sTILs, tumor cells, Ki67+, and pH3+ features. Regardless of the hotspot selection metric, a complementary use of multiple histological features (tTILs, sTILs) and molecular biomarkers (Ki67 and pH3) resulted in top ranked performance at the patient level. Conclusions: Overall, our results emphasize that prediction models for NAC response should be based on biomarkers in combination rather than in isolation. Our study provides compelling evidence to support the use of ML-based models to predict NAC response in patients with TNBC.
Background: Although triple-negative breast cancer (TNBC) is associated with an increased risk of recurrence and metastasis, the molecular mechanisms underlying metastasis in TNBC remain unknown. To identify transcriptional changes and genes regulating metastatic progression in TNBC, we compared the transcriptomic profiles of primary and matched metastatic tumors using massively parallel RNA sequencing. Methods: We performed gene expression profiling using formalin-fixed paraffin-embedded (FFPE) TNBC tissues of patients from two cohorts: the Zurich cohort (n = 31) and the Stavanger cohort (n = 5). Among the 31 patients in the Zurich cohort, 18 had primary TNBC tumors that did not metastasize, and 13 had primary tumors that metastasized (11 paired primary and locoregional recurrences). The Stavanger cohort included five matched primary and metastatic TNBC tumors. Significantly differentially expressed genes (DEGs; absolute fold change ≥2, p < 0.05) were identified and subjected to functional analyses. We investigated if there was any overlap between DEGs from both the cohorts with epithelial-to-mesenchymal-to-amoeboid transition (EMAT) gene signature. xCell was used to estimate relative fractions of 64 immune and stromal cell types in each RNA-seq sample. Results: In the Zurich cohort, we identified 1624 DEGs between primary TNBC tumors and matched metastatic lesions. xCell analysis revealed a significantly higher immune scores for metastatic lesions compared to paired primary tumors in the Zurich cohort. We also found significant upregulation of three MammaPrint signature genes (HRASLS, TGFB3 and RASSF7) in primary tumors that metastasized compared to primary tumors that remained metastasis-free. In the Stavanger cohort, we identified 818 DEGs between primary tumors and matched metastatic lesions. No significant differences in xCell immune scores were observed. We found that 21 and 14 DEGs from Zurich and Stavanger cohort, respectively, overlapped with the EMAT gene signature. In both cohorts, genes belonging to the MMP, FGF, and PDGFR families were upregulated in primary tumors compared to matched metastatic lesions. Conclusions: Our results suggest that distinct gene expression patterns exist between primary TNBCs and matched metastatic tumors. Further studies are warranted to explore whether these discrete expression profiles underlie or result from disease status.
The most prevalent form of bladder cancer is urothelial carcinoma, characterized by a high recurrence rate and substantial lifetime treatment costs for patients. Grading is a prime factor for patient risk stratification, although it suffers from inconsistencies and variations among pathologists. Moreover, absence of annotations in medical imaging renders it difficult to train deep learning models. To address these challenges, we introduce a pipeline designed for bladder cancer grading using histological slides. First, it extracts urothelium tissue tiles at different magnification levels, employing a convolutional neural network for processing for feature extraction. Then, it engages in the slide-level prediction process. It employs a nested multiple-instance learning approach with attention to predict the grade. To distinguish different levels of malignancy within specific regions of the slide, we include the origins of the tiles in our analysis. The attention scores at region level are shown to correlate with verified high-grade regions, giving some explainability to the model. Clinical evaluations demonstrate that our model consistently outperforms previous state-of-the-art methods, achieving an F1 score of 0.85.
Background/Objectives: Breast cancer (BC) is the most commonly diagnosed cancer in women and the leading cause of cancer-related deaths globally. Molecular subtyping is crucial for prognosis and treatment planning, with immunohistochemistry (IHC) being the most commonly used method. However, IHC has limitations, including observer variability, a lack of standardization, and a lack of reproducibility. Gene expression profiling is considered the ground truth for molecular subtyping; unfortunately, this is expensive and inaccessible to many institutions. This study investigates the potential of an artificial intelligence (AI) model to predict BC molecular subtypes directly from hematoxylin and eosin (H&E)-stained whole-slide images (WSIs). Methods: A pretrained deep learning framework based on multiple-instance learning (MIL) was validated on the Stavanger Breast Cancer (SBC) dataset, consisting of 538 BC cases. Three classification tasks were assessed, including two-class [triple negative BC (TNBC) vs. non-TNBC], three-class (luminal vs. HER2-positive vs. TNBC), and four-class (luminal A vs. luminal B vs. HER2-positive vs. TNBC) groups. Performance metrics were used for the evaluation of the AI model. Results: The AI model demonstrated strong performance in distinguishing TNBC from non-TNBC (AUC = 0.823, accuracy = 0.833, F1-score = 0.824). However, performance declined with an increasing number of classes. Conclusions: The study highlights the potential of AI in BC molecular subtyping from H&E WSIs, offering an easily applicable and standardized method to IHC. Future improvements should focus on optimizing multi-class classification and validating AI-based methods against gene expression analyses for enhanced clinical applicability.
Purpose: Triple-negative breast cancer (TNBC) is a molecularly complex and heterogeneous breast cancer subtype with distinct biological features and clinical behavior. Although TNBC is associated with an increased risk of metastasis and recurrence, the molecular mechanisms underlying TNBC metastasis remain unclear. We performed whole-exome sequencing (WES) analysis of primary TNBC and paired recurrent tumors to investigate the genetic profile of TNBC. Methods: Genomic DNA extracted from 35 formalin-fixed paraffin-embedded tissue samples from 26 TNBC patients was subjected to WES. Of these, 15 were primary tumors that did not have recurrence, and 11 were primary tumors that had recurrence (nine paired primary and recurrent tumors). Tumors were analyzed for single-nucleotide variants and insertions/deletions. Results: The tumor mutational burden (TMB) was 7.6 variants/megabase in primary tumors that recurred (n = 9); 8.2 variants/megabase in corresponding recurrent tumors (n = 9); and 7.3 variants/megabase in primary tumors that did not recur (n = 15). MUC3A was the most frequently mutated gene in all groups. Mutations in MAP3K1 and MUC16 were more common in our dataset. No alterations in PI3KCA were detected in our dataset. Conclusions: We found similar mutational profiles between primary and paired recurrent tumors, suggesting that genomic features may be retained during local recurrence.
29 July 2021 in World Journal of Surgical Oncology
Background: Extramedullary leukemia, also known as myeloid sarcoma, is a rare form of acute myeloid leukemia and often accompanies bone marrow involvement. Myeloid infiltration of the thyroid gland is extremely rare. Here, a unique case in which thyroid cancer tissue was infiltrated with myeloid cells is presented. Case presentation: We present a case of thyroid papillary cancer infiltrated by blastic cells and bilateral breast and axillary myeloid sarcoma in a 30-year-old Caucasian female patient with a history of osteosarcoma and MDS-RAEB2. The patient firstly received 6 cycles of chemotherapy for osteosarcoma, and allogeneic hematopoietic stem cell transplantation was performed after anthracycline-based chemotherapy due to MDS-RAEB2. The patient remained in remission on follow-up in terms of both osteosarcoma and MDS-RAEB2. Malignant features (Bethesda VI) were observed in the fine needle aspiration biopsy performed from a newly developed firm, fixed thyroid nodule approximately 4–5 cm in length in the left thyroid lobe. Because of the Bethesda VI thyroid nodule, the patient underwent total thyroidectomy. In the pathological evaluation, CD34-, CD117-, MPO-, and HLA-DR-positive blastic cells which infiltrated into follicular variant papillary thyroid carcinoma were detected. In the evaluation performed due to blastic cell infiltration, multiple lesions showing increased 18-fluorodeoxyglucose activity in bilateral breast and axillae were detected. Myeloid sarcoma was found as a result of tru-cut biopsy from these lesions. A fungal cystic lesion was detected in the frontal region of the patient who developed altered consciousness after the second cycle of treatment of myeloid sarcoma. During her follow-up in the intensive care unit, she died of cranial septic embolism and acute infarction. Conclusions: Here, we present a very interesting case that is the first. A staged approach to diagnosis with methods including immunohistochemical staining, radiological imaging methods, and cytogenetic and molecular analyses can help make the definitive diagnosis.