RESUMEN
Recent advances in the field of immuno-oncology have brought transformative changes in the management of cancer patients. The immune profile of tumours has been found to have key value in predicting disease prognosis and treatment response in various cancers. Multiplex immunohistochemistry and immunofluorescence have emerged as potent tools for the simultaneous detection of multiple protein biomarkers in a single tissue section, thereby expanding opportunities for molecular and immune profiling while preserving tissue samples. By establishing the phenotype of individual tumour cells when distributed within a mixed cell population, the identification of clinically relevant biomarkers with high-throughput multiplex immunophenotyping of tumour samples has great potential to guide appropriate treatment choices. Moreover, the emergence of novel multi-marker imaging approaches can now provide unprecedented insights into the tumour microenvironment, including the potential interplay between various cell types. However, there are significant challenges to widespread integration of these technologies in daily research and clinical practice. This review addresses the challenges and potential solutions within a structured framework of action from a regulatory and clinical trial perspective. New developments within the field of immunophenotyping using multiplexed tissue imaging platforms and associated digital pathology are also described, with a specific focus on translational implications across different subtypes of cancer. © 2024 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.
Asunto(s)
Neoplasias de la Mama , Humanos , Femenino , Biomarcadores de Tumor/genética , Pronóstico , Fenotipo , Reino Unido , Microambiente TumoralRESUMEN
BACKGROUND: This study aimed to investigate the distribution and changes of HER2 status in untreated tumours, in residual disease and in metastasis, and their long-term prognostic implications. METHODS: This is a population-based cohort study of patients treated with neoadjuvant chemotherapy for breast cancer during 2007-2020 in the Stockholm-Gotland region which comprises 25% of the entire Swedish population. Information was extracted from the National Breast Cancer Registry and electronic patient charts to minimize data missingness and misclassification. RESULTS: In total, 2494 patients received neoadjuvant chemotherapy, of which 2309 had available pretreatment HER2 status. Discordance rates were 29.9% between primary and residual disease (kappa = 0.534), 31.2% between primary tumour and metastasis (kappa = 0.512) and 33.3% between residual disease to metastasis (kappa = 0.483). Adjusted survival curves differed between primary HER2 0 and HER2-low disease (p < 0.001), with the former exhibiting an early peak in risk for death which eventually declined below the risk of HER2-low. Across all disease settings, increasing the number of biopsies increased the likelihood of detecting HER2-low status. CONCLUSION: HER2 status changes during neoadjuvant chemotherapy and metastatic progression, and the long-term behaviours of HER2 0 and HER2-low disease differ, underscoring the need for obtaining tissue biopsies and for extended follow-up in breast cancer studies.
Asunto(s)
Neoplasias de la Mama , Progresión de la Enfermedad , Terapia Neoadyuvante , Receptor ErbB-2 , Humanos , Neoplasias de la Mama/tratamiento farmacológico , Neoplasias de la Mama/patología , Neoplasias de la Mama/metabolismo , Femenino , Receptor ErbB-2/metabolismo , Persona de Mediana Edad , Suecia/epidemiología , Anciano , Adulto , Metástasis de la Neoplasia , Pronóstico , Estudios de Cohortes , Quimioterapia Adyuvante , Neoplasia ResidualRESUMEN
PURPOSE: To evaluate the Stratipath Breast tool for image-based risk profiling and compare it with an established prognostic multigene assay for risk profiling in a real-world case series of estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative early breast cancer patients categorized as intermediate risk based on classic clinicopathological variables and eligible for chemotherapy. METHODS: In a case series comprising 234 invasive ER-positive/HER2-negative tumors, clinicopathological data including Prosigna results and corresponding HE-stained tissue slides were retrieved. The digitized HE slides were analysed by Stratipath Breast. RESULTS: Our findings showed that the Stratipath Breast analysis identified 49.6% of the clinically intermediate tumors as low risk and 50.4% as high risk. The Prosigna assay classified 32.5%, 47.0% and 20.5% tumors as low, intermediate and high risk, respectively. Among Prosigna intermediate-risk tumors, 47.3% were stratified as Stratipath low risk and 52.7% as high risk. In addition, 89.7% of Stratipath low-risk cases were classified as Prosigna low/intermediate risk. The overall agreement between the two tests for low-risk and high-risk groups (N = 124) was 71.0%, with a Cohen's kappa of 0.42. For both risk profiling tests, grade and Ki67 differed significantly between risk groups. CONCLUSION: The results from this clinical evaluation of image-based risk stratification shows a considerable agreement to an established gene expression assay in routine breast pathology.
Asunto(s)
Biomarcadores de Tumor , Neoplasias de la Mama , Aprendizaje Profundo , Receptor ErbB-2 , Receptores de Estrógenos , Humanos , Neoplasias de la Mama/genética , Neoplasias de la Mama/patología , Femenino , Persona de Mediana Edad , Biomarcadores de Tumor/genética , Adulto , Anciano , Receptor ErbB-2/genética , Receptor ErbB-2/metabolismo , Receptores de Estrógenos/metabolismo , Medición de Riesgo/métodos , Pronóstico , Perfilación de la Expresión Génica/métodosRESUMEN
BACKGROUND: Studies identifying risk factors for death from breast cancer after ductal carcinoma in situ (DCIS) are rare. In this retrospective nested case-control study, clinicopathological factors in women treated for DCIS and who died from breast cancer were compared with those of patients with DCIS who were free from metastatic disease. METHODS: The study included patients registered with DCIS without invasive carcinoma in Sweden between 1992 and 2012. This cohort was linked to the National Cause of Death Registry. Of 6964 women with DCIS, 96 were registered with breast cancer as cause of death (cases). For each case, up to four controls (318; women with DCIS, alive and without metastatic breast cancer at the time of death of the corresponding case) were selected randomly by incidence density sampling. Whole slides of tumour tissue were evaluated for DCIS grade, comedo necrosis, and intensity of periductal lymphocytic infiltrate. Composition of the immune cell infiltrate, expression of oestrogen receptor, progesterone receptor, human epidermal growth factor receptor 2, and proliferation marker Ki-67 were scored on tissue microarrays. Clinical information was obtained from medical records. Information on date, site, and histological characteristics of local and distant recurrences was obtained from medical records for both cases and controls. RESULTS: Tumour tissue was analysed from 65 cases and 195 controls. Intense periductal lymphocytic infiltrate around DCIS was associated with an increased risk of later dying from breast cancer (OR 2.21. 95% c.i. 1.01 to 4.84). Tumours with more intense lymphocytic infiltrate had a lower T cell/B cell ratio. None of the other biomarkers correlated with increased risk of breast cancer death. CONCLUSION: The immune response to DCIS may influence the risk of dying from breast cancer.
Asunto(s)
Neoplasias de la Mama , Carcinoma Ductal de Mama , Carcinoma Intraductal no Infiltrante , Femenino , Humanos , Neoplasias de la Mama/patología , Carcinoma Intraductal no Infiltrante/patología , Estudios de Casos y Controles , Estudios Retrospectivos , Factores de Riesgo , Inflamación , Carcinoma Ductal de Mama/patologíaRESUMEN
A growing body of research supports stromal tumour-infiltrating lymphocyte (TIL) density in breast cancer to be a robust prognostic and predicive biomarker. The gold standard for stromal TIL density quantitation in breast cancer is pathologist visual assessment using haematoxylin and eosin-stained slides. Artificial intelligence/machine-learning algorithms are in development to automate the stromal TIL scoring process, and must be validated against a reference standard such as pathologist visual assessment. Visual TIL assessment may suffer from significant interobserver variability. To improve interobserver agreement, regulatory science experts at the US Food and Drug Administration partnered with academic pathologists internationally to create a freely available online continuing medical education (CME) course to train pathologists in assessing breast cancer stromal TILs using an interactive format with expert commentary. Here we describe and provide a user guide to this CME course, whose content was designed to improve pathologist accuracy in scoring breast cancer TILs. We also suggest subsequent steps to translate knowledge into clinical practice with proficiency testing.
Asunto(s)
Neoplasias de la Mama , Humanos , Femenino , Patólogos , Linfocitos Infiltrantes de Tumor , Inteligencia Artificial , PronósticoRESUMEN
Modern histologic imaging platforms coupled with machine learning methods have provided new opportunities to map the spatial distribution of immune cells in the tumor microenvironment. However, there exists no standardized method for describing or analyzing spatial immune cell data, and most reported spatial analyses are rudimentary. In this review, we provide an overview of two approaches for reporting and analyzing spatial data (raster versus vector-based). We then provide a compendium of spatial immune cell metrics that have been reported in the literature, summarizing prognostic associations in the context of a variety of cancers. We conclude by discussing two well-described clinical biomarkers, the breast cancer stromal tumor infiltrating lymphocytes score and the colon cancer Immunoscore, and describe investigative opportunities to improve clinical utility of these spatial biomarkers. © 2023 The Pathological Society of Great Britain and Ireland.
Asunto(s)
Neoplasias del Colon , Humanos , Biomarcadores , Benchmarking , Linfocitos Infiltrantes de Tumor , Análisis Espacial , Microambiente TumoralRESUMEN
The clinical significance of the tumor-immune interaction in breast cancer is now established, and tumor-infiltrating lymphocytes (TILs) have emerged as predictive and prognostic biomarkers for patients with triple-negative (estrogen receptor, progesterone receptor, and HER2-negative) breast cancer and HER2-positive breast cancer. How computational assessments of TILs might complement manual TIL assessment in trial and daily practices is currently debated. Recent efforts to use machine learning (ML) to automatically evaluate TILs have shown promising results. We review state-of-the-art approaches and identify pitfalls and challenges of automated TIL evaluation by studying the root cause of ML discordances in comparison to manual TIL quantification. We categorize our findings into four main topics: (1) technical slide issues, (2) ML and image analysis aspects, (3) data challenges, and (4) validation issues. The main reason for discordant assessments is the inclusion of false-positive areas or cells identified by performance on certain tissue patterns or design choices in the computational implementation. To aid the adoption of ML for TIL assessment, we provide an in-depth discussion of ML and image analysis, including validation issues that need to be considered before reliable computational reporting of TILs can be incorporated into the trial and routine clinical management of patients with triple-negative breast cancer. © 2023 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.
Asunto(s)
Neoplasias Mamarias Animales , Neoplasias de la Mama Triple Negativas , Humanos , Animales , Linfocitos Infiltrantes de Tumor , Biomarcadores , Aprendizaje AutomáticoRESUMEN
Ki67 has potential clinical importance in breast cancer but has yet to see broad acceptance due to inter-laboratory variability. Here we tested an open source and calibrated automated digital image analysis (DIA) platform to: (i) investigate the comparability of Ki67 measurement across corresponding core biopsy and resection specimen cases, and (ii) assess section to section differences in Ki67 scoring. Two sets of 60 previously stained slides containing 30 core-cut biopsy and 30 corresponding resection specimens from 30 estrogen receptor-positive breast cancer patients were sent to 17 participating labs for automated assessment of average Ki67 expression. The blocks were centrally cut and immunohistochemically (IHC) stained for Ki67 (MIB-1 antibody). The QuPath platform was used to evaluate tumoral Ki67 expression. Calibration of the DIA method was performed as in published studies. A guideline for building an automated Ki67 scoring algorithm was sent to participating labs. Very high correlation and no systematic error (p = 0.08) was found between consecutive Ki67 IHC sections. Ki67 scores were higher for core biopsy slides compared to paired whole sections from resections (p ≤ 0.001; median difference: 5.31%). The systematic discrepancy between core biopsy and corresponding whole sections was likely due to pre-analytical factors (tissue handling, fixation). Therefore, Ki67 IHC should be tested on core biopsy samples to best reflect the biological status of the tumor.
Asunto(s)
Neoplasias de la Mama , Biomarcadores de Tumor/análisis , Biopsia , Neoplasias de la Mama/patología , Femenino , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Inmunohistoquímica , Antígeno Ki-67/análisis , Receptores de EstrógenosRESUMEN
Ki67, a nuclear proliferation-related protein, is heavily used in anatomic pathology but has not become a companion diagnostic or a standard-of-care biomarker due to analytic variability in both assay protocols and interpretation. The International Ki67 Working Group in breast cancer has published and has ongoing efforts in the standardization of the interpretation of Ki67, but they have not yet assessed technical issues of assay production representing multiple sources of variation, including antibody clones, antibody formats, staining platforms, and operators. The goal of this work is to address these issues with a new standardization tool. We have developed a cell line microarray system in which mixes of human Karpas 299 or Jurkat cells (Ki67+) with Sf9 (Spodoptera frugiperda) (Ki67-) cells are present in incremental standardized ratios. To validate the tool, six different antibodies, including both ready-to-use and concentrate formats from six vendors, were used to measure Ki67 proliferation indices using IHC protocols for manual (bench-top) and automated platforms. The assays were performed by three different laboratories at Yale and analyzed using two image analysis software packages, including QuPath and Visiopharm. Results showed statistically significant differences in Ki67 reactivity between each antibody clone. However, subsets of Ki67 assays using three clones performed in three different labs show no significant differences. This work shows the need for analytic standardization of the Ki67 assay and provides a new tool to do so. We show here how a cell line standardization system can be used to normalize the staining variability in proliferation indices between different antibody clones in a triple negative breast cancer cohort. We believe that this cell line standardization array has the potential to improve reproducibility among Ki67 assays and laboratories, which is critical for establishing Ki67 as a standard-of-care assay.
Asunto(s)
Biomarcadores de Tumor/análisis , Inmunohistoquímica/normas , Antígeno Ki-67/análisis , Índice Mitótico/normas , Neoplasias de la Mama Triple Negativas/patología , Adulto , Anciano , Anciano de 80 o más Años , Línea Celular Tumoral , Femenino , Humanos , Persona de Mediana EdadRESUMEN
Deep learning algorithms have shown benefits for pathology in the context of risk stratification of tumors. Although the results are promising, several steps have to be made to confirm clinical utility. In a recent issue of The Journal of Pathology, Colling et al present a perspective manuscript providing a roadmap to routine use of artificial intelligence in histopathology. In this commentary, we aimed to put these key points in the context of recent findings of AI and digital image analysis studies. © 2019 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
Asunto(s)
Inteligencia Artificial , Neoplasias , Algoritmos , Humanos , Reino UnidoRESUMEN
Programmed death 1 ligand 1 (PD-L1) Immunohistochemistry (IHC) is the key FDA-approved predictive marker to identify responders to anti-PD1 axis drugs. Multiple PD-L1 IHC assays with various antibodies and cut points have been used in clinical trials across tumor types. Comparative performance characteristics of these assays have been extensively studied qualitatively but not quantitatively. Here we evaluate the use of a standardized PD-L1 Index tissue microarray (TMA) to objectively determine agreement between antibody assays for PD-L1 applying quantitative digital image analysis. Using a specially constructed Index TMA containing a panel of ten isogenic cell lines in triplicate, we tested identical but independently grown batches of isogenic cells to prove Index TMAs can be produced in large quantities and hence serve as a standardization tool. Then the Index TMAs were evaluated using quantitative immunofluorescence (QIF) to validate the TMA itself and also to compare antibodies including E1L3N, SP142 and SP263. Next, an inter-laboratory and inter-assay comparison of 5 PD-L1 chromogenic IHC assays (US Food and Drug Administration (FDA) approved and lab developed test (LDT)) were performed at 12 sites around the USA. As previously reported, the SP142 FDA assay failed to detect low levels of PD-L1 in cell lines distinguished by the other four assays. The assays for 22C3 FDA, 28-8-FDA, SP263 FDA, and E1L3N LDT were highly similar across sites and all laboratories showed a high consistency over time for all assays using this Index TMA. In conclusion, we were able to objectively quantify PD-L1 expression on a standardized Index TMA using digital image analysis and we confirmed previous subjective assessments of these assays, but now in a multi-institutional setting. We envision commercial use of this Index TMA or similar smaller version as a useful standardization mechanism to compare results between institutions and to identify abnormalities while running routine clinical samples.
Asunto(s)
Antígeno B7-H1/análisis , Técnica del Anticuerpo Fluorescente , Línea Celular , Análisis de Matrices TisularesRESUMEN
PURPOSE: The proliferation-associated biomarker Ki67 has potential utility in breast cancer, including aiding decisions based on prognosis, but has unacceptable inter- and intralaboratory variability. The aim of this study was to compare the prognostic potential for Ki67 hot spot scoring and global scoring using different digital image analysis (DIA) platforms. METHODS: An ER+/HER2- breast cancer cohort (n = 139) with whole slide images of sequential sections stained for hematoxylin-eosin, pancytokeratin and Ki67, was analyzed using two DIA platforms. For hot spot analysis virtual dual staining was applied, aligning pancytokeratin and Ki67 images and 22 hot spot algorithms with different features were designed. For global Ki67 scoring an automated QuPath algorithm was applied on Ki67-stained whole slide images. Clinicopathological data included overall survival (OS) and recurrence-free survival (RFS) along with PAM50 molecular subtypes. RESULTS: We show significant variations in Ki67 hot spot scoring depending on number of included tumor cells, hot spot size, shape and location. The higher the number of scored tumor cells, the higher the reproducibility of Ki67 proliferation values. Hot spot scoring had greater prognostic potential for RFS in high versus low Ki67 subgroups (hazard ratio (HR) 6.88, CI 2.07-22.87, p = 0.002), compared to global scoring (HR 3.13, CI 1.41-6.96, p = 0.005). Regarding OS, global scoring (HR 7.46, CI 2.46-22.58, p < 0.001) was slightly better than hot spot scoring (HR 6.93, CI 1.61-29.91, p = 0.009). In adjusted multivariate analysis, only global scoring was an independent prognostic marker for both RFS and OS. In addition, global Ki67-based surrogate subtypes reached higher concordance with PAM50 molecular subtype for luminal A and B tumors (66.3% concordance rate, κ = 0.345), than using hot spot scoring (55.8% concordance rate, κ = 0.250). CONCLUSIONS: We conclude that the automated global Ki67 scoring is feasible and shows clinical validity, which, however, needs to be confirmed in a larger cohort before clinical implementation.
Asunto(s)
Antígenos de Neoplasias/análisis , Neoplasias de la Mama/química , Carcinoma/química , Estrógenos , Procesamiento de Imagen Asistido por Computador/métodos , Antígeno Ki-67/análisis , Neoplasias Hormono-Dependientes/química , Automatización , Neoplasias de la Mama/mortalidad , Carcinoma/mortalidad , Supervivencia sin Enfermedad , Femenino , Estudios de Seguimiento , Humanos , Estimación de Kaplan-Meier , Queratinas/análisis , Persona de Mediana Edad , Proteínas de Neoplasias/análisis , Neoplasias Hormono-Dependientes/mortalidad , Pronóstico , Modelos de Riesgos Proporcionales , Receptor ErbB-2/análisis , Receptores de Estrógenos/análisis , Reproducibilidad de los Resultados , Estudios RetrospectivosRESUMEN
Ki67 expression has been a valuable prognostic variable in breast cancer, but has not seen broad adoption due to lack of standardization between institutions. Automation could represent a solution. Here we investigate the reproducibility of Ki67 measurement between three image analysis platforms with supervised classifiers performed by the same operator, by multiple operators, and finally we compare their accuracy in prognostic potential. Two breast cancer patient cohorts were used for this study. The standardization was done with the 30 cases of ER+ breast cancer that were used in phase 3 of International Ki67 in Breast Cancer Working Group initiatives where blocks were centrally cut and stained for Ki67. The outcome cohort was from 149 breast cancer cases from the Yale Pathology archives. A tissue microarray was built from representative tissue blocks with median follow-up of 120 months. The Mib-1 antibody (Dako) was used to detect Ki67 (dilution 1:100). HALO (IndicaLab), QuantCenter (3DHistech), and QuPath (open source software) digital image analysis (DIA) platforms were used to evaluate Ki67 expression. Intraclass correlation coefficient (ICC) was used to measure reproducibility. Between-DIA platform reproducibility was excellent (ICC: 0.933, CI: 0.879-0.966). Excellent reproducibility was found between all DIA platforms and the reference standard Ki67 values of Spectrum Webscope (QuPath-Spectrum Webscope ICC: 0.970, CI: 0.936-0.986; HALO-Spectrum Webscope ICC: 0.968, CI: 0.933-0.985; QuantCenter-Spectrum Webscope ICC: 0.964, CI: 0.919-0.983). All platforms showed excellent intra-DIA reproducibility (QuPath ICC: 0.992, CI: 0.986-0.996; HALO ICC: 0.972, CI: 0.924-0.988; QuantCenter ICC: 0.978, CI: 0.932-0.991). Comparing each DIA against outcome, the hazard ratios were similar. The inter-operator reproducibility was particularly high (ICC: 0.962-0.995). Our results showed outstanding reproducibility both within and between-DIA platforms, including one freely available DIA platform (QuPath). We also found the platforms essentially indistinguishable with respect to prediction of breast cancer patient outcome. Results justify multi-institutional DIA studies to assess clinical utility.
Asunto(s)
Neoplasias de la Mama/metabolismo , Procesamiento de Imagen Asistido por Computador , Antígeno Ki-67/análisis , Adulto , Anciano , Anciano de 80 o más Años , Estudios de Cohortes , Humanos , Antígeno Ki-67/metabolismo , Persona de Mediana Edad , Variaciones Dependientes del Observador , Reproducibilidad de los ResultadosRESUMEN
In the second half of the 20th century research focusing to breast carcinomas at the Semmelweis University had been mostly linked to the 2nd Department of Pathology. Nowadays, following the rapidly improving treatment modalities in breast cancer there is an increasing need for defining new predictive and prognostic markers. The modern molecular pathological approach helps tremendously in mapping the biological behavior of individual cases of breast cancers and meanwhile, it is one of the prerequisites of a more efficient treatment both in neoadjuvant and adjuvant settings, as well as in metastatic disease. We provide a brief review of the relevant results we have obtained in breast cancer research between 2000 and 2015.
RESUMEN
The transanal endoscopic microsurgery (TEM) provides lower relapse and complication rate for the the surgical treatment of the neoplasms of the middle and lower third of the rectum in selected cases. Hence, it can be an alternative method of the conventional approaches, if it does not compromise oncological radicality. The TEM procedure has been started at the 1st Department of Surgery, Semmelweis University in the fall of 2013. In this short study we have evaluated the clinicopathological characteristics of patients undergoing TEM between September 2013 and September 2014. Fourty-four patients were included in our retrospective analysis. 12 patients had low grade adenoma, 14 patients had high grade adenoma, 17 patients had invasive adenocarcinoma, while one was operated for a neuroendocrine tumor. There was no difference in the size of neoplasms between the low and high grade adenomas or adenocarcinomas (p = 0.210), tumors below the size of 30 mm or over 30 mm displayed no significant difference either (p = 0.424). The surgical margins were free of tumor in 41 cases (95.3%). In 13 out of 44 cases the preoperative histology proposed a lower grade neoplasm than the final report (p < 0.001). These results demonstrate that the surgical treatment of large adenomas with TEM technique, which involves excision of the whole bowel wall, is more appropriate than the fractionated removal or polypectomy supplemented by mucosectomy. The pT2 stage tumours might be subjected to the TEM method in selected cases (e.g. following neoadjuvant treatment or palliative care), but this has to be confirmed with prospecively evaluated large series clinical studies which are currently ongoing.
Asunto(s)
Microcirugia , Proctoscopía/métodos , Neoplasias del Recto/patología , Neoplasias del Recto/cirugía , Recto/patología , Recto/cirugía , Adenocarcinoma/patología , Adenocarcinoma/cirugía , Adenoma/patología , Adenoma/cirugía , Anciano , Canal Anal , Femenino , Humanos , Masculino , Microcirugia/métodos , Persona de Mediana Edad , Clasificación del Tumor , Tumores Neuroendocrinos/patología , Tumores Neuroendocrinos/cirugía , Estudios Retrospectivos , Resultado del TratamientoRESUMEN
Accurate detection of invasive breast cancer (IC) can provide decision support to pathologists as well as improve downstream computational analyses, where detection of IC is a first step. Tissue containing IC is characterized by the presence of specific morphological features, which can be learned by convolutional neural networks (CNN). Here, we compare the use of a single CNN model versus an ensemble of several base models with the same CNN architecture, and we evaluate prediction performance as well as variability across ensemble based model predictions. Two in-house datasets comprising 587 whole slide images (WSI) are used to train an ensemble of ten InceptionV3 models whose consensus is used to determine the presence of IC. A novel visualisation strategy was developed to communicate ensemble agreement spatially. Performance was evaluated in an internal test set with 118 WSIs, and in an additional external dataset (TCGA breast cancer) with 157 WSI. We observed that the ensemble-based strategy outperformed the single CNN-model alternative with respect to accuracy on tile level in 89 % of all WSIs in the test set. The overall accuracy was 0.92 (DICE coefficient, 0.90) for the ensemble model, and 0.85 (DICE coefficient, 0.83) for the single CNN alternative in the internal test set. For TCGA the ensemble outperformed the single CNN in 96.8 % of the WSI, with an accuracy of 0.87 (DICE coefficient 0.89), the single model provides an accuracy of 0.75 (DICE coefficient 0.78). The results suggest that an ensemble-based modeling strategy for breast cancer invasive cancer detection consistently outperforms the conventional single model alternative. Furthermore, visualisation of the ensemble agreement and confusion areas provide direct visual interpretation of the results. High performing cancer detection can provide decision support in the routine pathology setting as well as facilitate downstream computational analyses.
RESUMEN
Background: Estrogen receptor-low (ER-low) HER2-negative breast cancer has similar pathological and molecular characteristics as triple-negative breast cancer (TNBC), and it is questionable whether it should be considered a separate entity. When the international guidelines lowered the cutoff for ER positivity to ≥1% in 2010, the ≥10% threshold was kept in Sweden. ER-low breast cancer (ER 1-9%) is thus in Sweden treated as TNBC. We aimed to describe patient and tumor characteristics, treatment patterns and overall survival in a Swedish population-based cohort of patients with ER-zero and ER-low HER2-negative breast cancer treated as TNBC. Methods: All TNBC cases diagnosed in Sweden 2008-2020 were included in a population-based cohort study. Patient, tumor and treatment characteristics were analyzed by ER-status (ER 0% vs 1-9%), and associations between subgroups compared using χ2 test. Survival endpoint was overall survival (OS), and Kaplan-Meier curves were estimated. Cox proportional hazards models were used to estimate adjusted hazard ratios comparing ER-low to ER-zero. Findings: Of the 5655 tumors, 90.1% had an ER expression of 0%, while 9.9% were ER-low. ER-low tumors were grade III in 69.4% (80.8% in ER-zero tumors, p-value = 0.001), with a median Ki67 of 60% (63% in ER-zero tumors, p-value = 0.005). There were no significant differences in given chemotherapy (p = 0.546). A pathological complete response (pCR) was achieved in 28.1% of ER-low tumors (25.1% in ER-zero tumors). In the unadjusted analysis of OS, women with ER-low disease had a borderline but not significantly better OS than those with ER-zero disease (HR 0.84 (95% CI 0.71-1.00), p = 0.052). ER-status 1-9% vs 0% was not associated with OS in the multivariable analysis (HR 1.11 (0.90-1.36)). Distant disease-free survival did not differ by ER-status 0% vs 1-9% (HR 0.97 for ER-zero vs ER-low (0.62-1.53), p = 0.905). After preoperative treatment, the impact of pCR for OS did not significantly differ between ER-zero or ER-low disease. Interpretation: ER-low HER2-negative breast cancer has characteristics and prognosis similar to TNBC, when treated in the same way. Therefore, it seems reasonable to use a ≥10% threshold for ER positivity. This would provide patients with ER-low tumors the same treatment opportunities as patients with TNBC, within studies and within clinical routine. Funding: This work was financially supported by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, in accordance with terms and conditions of a Master Collaboration Agreement between the company and Karolinska Institutet.
RESUMEN
Developing AI models for digital pathology has traditionally relied on single-scale analysis of histopathology slides. However, a whole slide image is a rich digital representation of the tissue, captured at various magnification levels. Limiting our analysis to a single scale overlooks critical information, spanning from intricate high-resolution cellular details to broad low-resolution tissue structures. In this study, we propose a model-agnostic multiresolution feature aggregation framework tailored for the analysis of histopathology slides in the context of breast cancer, on a multicohort dataset of 2038 patient samples. We have adapted 9 state-of-the-art multiple instance learning models on our multi-scale methodology and evaluated their performance on grade prediction, TP53 mutation status prediction and survival prediction. The results prove the dominance of the multiresolution methodology, and specifically, concatenating or linearly transforming via a learnable layer the feature vectors of image patches from a high (20x) and low (10x) magnification factors achieve improved performance for all prediction tasks across domain-specific and imagenet-based features. On the contrary, the performance of uniresolution baseline models was not consistent across domain-specific and imagenet-based features. Moreover, we shed light on the inherent inconsistencies observed in models trained on whole-tissue-sections when validated against biopsy-based datasets. Despite these challenges, our findings underscore the superiority of multiresolution analysis over uniresolution methods. Finally, cross-scale analysis also benefits the explainability aspects of attention-based architectures, since one can extract attention maps at the tissue- and cell-levels, improving the interpretation of the model's decision. The code and results of this study can be found at github.com/tsikup/multiresolution_histopathology.
RESUMEN
Programmed death-ligand 1 (PD-L1) expression is currently used in the clinic to assess eligibility for immune-checkpoint inhibitors via the tumor proportion score (TPS), but its efficacy is limited by high interobserver variability. Multiple papers have presented systems for the automatic quantification of TPS, but none report on the task of determining cell-level PD-L1 expression and often reserve their evaluation to a single PD-L1 monoclonal antibody or clinical center. In this paper, we report on a deep learning algorithm for detecting PD-L1 negative and positive tumor cells at a cellular level and evaluate it on a cell-level reference standard established by six readers on a multi-centric, multi PD-L1 assay dataset. This reference standard also provides for the first time a benchmark for computer vision algorithms. In addition, in line with other papers, we also evaluate our algorithm at slide-level by measuring the agreement between the algorithm and six pathologists on TPS quantification. We find a moderately low interobserver agreement at cell-level level (mean reader-reader F1 score = 0.68) which our algorithm sits slightly under (mean reader-AI F1 score = 0.55), especially for cases from the clinical center not included in the training set. Despite this, we find good AI-pathologist agreement on quantifying TPS compared to the interobserver agreement (mean reader-reader Cohen's kappa = 0.54, 95% CI 0.26-0.81, mean reader-AI kappa = 0.49, 95% CI 0.27-0.72). In conclusion, our deep learning algorithm demonstrates promise in detecting PD-L1 expression at a cellular level and exhibits favorable agreement with pathologists in quantifying the tumor proportion score (TPS). We publicly release our models for use via the Grand-Challenge platform.