RESUMO
BACKGROUND: Human epidermal growth factor receptor 2 (HER2)-low breast cancer has emerged as a new subtype of tumor, for which novel antibody-drug conjugates have shown beneficial effects. Assessment of HER2 requires several immunohistochemistry tests with an additional in situ hybridization test if a case is classified as HER2 2+. Therefore, novel cost-effective methods to speed up the HER2 assessment are highly desirable. METHODS: We used a self-supervised attention-based weakly supervised method to predict HER2-low directly from 1437 histopathological images from 1351 breast cancer patients. We built six distinct models to explore the ability of classifiers to distinguish between the HER2-negative, HER2-low, and HER2-high classes in different scenarios. The attention-based model was used to comprehend the decision-making process aimed at relevant tissue regions. RESULTS: Our results indicate that the effectiveness of classification models hinges on the consistency and dependability of assay-based tests for HER2, as the outcomes from these tests are utilized as the baseline truth for training our models. Through the use of explainable AI, we reveal histologic patterns associated with the HER2 subtypes. CONCLUSION: Our findings offer a demonstration of how deep learning technologies can be applied to identify HER2 subgroup statuses, potentially enriching the toolkit available for clinical decision-making in oncology.
Assuntos
Biomarcadores Tumorais , Neoplasias da Mama , Aprendizado Profundo , Imuno-Histoquímica , Receptor ErbB-2 , Humanos , Receptor ErbB-2/metabolismo , Receptor ErbB-2/genética , Neoplasias da Mama/patologia , Neoplasias da Mama/metabolismo , Neoplasias da Mama/genética , Feminino , Biomarcadores Tumorais/metabolismo , Imuno-Histoquímica/métodos , Aprendizado de Máquina SupervisionadoRESUMO
INTRODUCTION: Lymph node fine needle aspiration (LN-FNA) is a minimally invasive method of evaluating lymphadenopathy. Nonetheless, its use is not widely accepted due to the lack of guidelines and a cytopathological categorisation that directly relates to management. We report our experience with LN FNA at a large Cancer Center in Latin America. METHODS: We retrospectively collected cytological cases of lymph node FNA from the department of pathology at AC Camargo Cancer Center performed over a 2-year period. Data extracted included LN location, age, sex and final cytological diagnosis. Patients that had undergone neoadjuvant chemotherapy and/or cases for which the surgery specimen location was not clearly reported were excluded. For those cases with surgical reports, risk of malignancy was calculated for each diagnostic category, along with overall performance of cytology. False positive cases were reviewed to assess any possible misinterpretation or sampling errors. RESULTS: A total of 1730 LN-FNA were distributed as follows: 62 (3.5%) non-diagnostic (ND); 1123 (64.9%) negative (NEG), 19 (1.1%) atypical (ATY), 53 (3.1%) suspicious for malignancy (SUS), and 473 (27.3%) positive (POS). Surgical reports were available for 560 cases (32.4%). Risk of malignancy (ROM) for each category was 33.3% for ND, 29.9% for NEG, 25% for ATY, 74.2% for SUS and 99.6% for POS. Overall sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) were 78.5%, 99.4%, 70.2% and 99.6%, respectively. CONCLUSION: Lymph node FNA is a very specific and accurate exam, which is reliable in the detection of lymph node metastasis and other causes of lymphadenopathy.
Assuntos
Linfonodos , Linfadenopatia , Biópsia por Agulha Fina/métodos , Humanos , Linfonodos/patologia , Linfadenopatia/diagnóstico , Linfadenopatia/patologia , Metástase Linfática/diagnóstico , Metástase Linfática/patologia , Estudos Retrospectivos , Sensibilidade e EspecificidadeRESUMO
DNA repair deficiency (DRD) is an important driver of carcinogenesis and an efficient target for anti-tumor therapies to improve patient survival. Thus, detection of DRD in tumors is paramount. Currently, determination of DRD in tumors is dependent on wet-lab assays. Here we describe an efficient machine learning algorithm which can predict DRD from histopathological images. The utility of this algorithm is demonstrated with data obtained from 1445 cancer patients. Our method performs rather well when trained on breast cancer specimens with homologous recombination deficiency (HRD), AUC (area under curve) = 0.80. Results for an independent breast cancer cohort achieved an AUC = 0.70. The utility of our method was further shown by considering the detection of mismatch repair deficiency (MMRD) in gastric cancer, yielding an AUC = 0.81. Our results demonstrate the capacity of our learning-base system as a low-cost tool for DRD detection.