Consistent and effective method to define the mouse estrous cycle stage by a deep learning-based model

The mouse estrous cycle is divided into four stages: proestrus (P), estrus (E), metestrus (M), and diestrus (D). The estrous cycle affects reproductive hormone levels in a wide variety of tissues. Therefore, to obtain reliable results from female mice, it is important to know the estrous cycle stage during sampling. The stage can be analyzed from a vaginal smear under a microscope. However, it is time-consuming, and the results vary between evaluators. Here, we present an accurate and reproducible method for staging the mouse estrous cycle in digital whole-slide images (WSIs) of vaginal smears. We developed a model using a deep convolutional neural network (CNN) in a cloud-based platform, Aiforia Create. The CNN was trained by supervised pixel-level multiclass semantic segmentation of image features from 171 hematoxylin-stained samples. The model was validated by comparing the results obtained by CNN with those of four independent researchers. The validation data included three separate studies comprising altogether 148 slides. The total agreement attested by the Fleiss kappa value between the validators and the CNN was excellent (0.75), and when D, E, and P were analyzed separately, the kappa values were 0.89, 0.79, and 0.74, respectively. The M stage is short and not well defined by the researchers. Thus, identification of the M stage by the CNN was challenging due to the lack of proper ground truth, and the kappa value was 0.26. We conclude that our model is reliable and effective for classifying the estrous cycle stages in female mice.

Deep learning based image analysis of liver steatosis in mouse models

The incidence of non-alcoholic fatty liver disease (NAFLD) is a continuously growing health problem worldwide, along with obesity. Therefore, both novel methods to efficiently study the manifestation of NAFLD and to analyze drug efficacy in pre-clinical models are needed. In the present study, we developed a deep neural network -based model to quantify micro- and macrovesicular steatosis in the liver on hematoxylin-eosin stained whole slide images (WSIs), using the cloud-based platform, Aiforia Create (Aiforia Technologies, Helsinki, Finland). The training data included a total of 101 WSIs from dietary interventions of wild-type mice and from two genetically modified (GM) mouse models with steatosis. The algorithm was trained for the following: to detect liver parenchyma, to exclude the blood vessels and any artefacts generated during tissue processing and image acquisition, to recognize and differentiate the areas of micro- and macrovesicular steatosis, and to quantify the recognized tissue area. The results of the image analysis replicated well the evaluation by expert pathologists, and correlated well with the liver fat content measured by EcoMRI ex vivo, and the correlation with total liver triglycerides were notable. In conclusion, the developed deep learning-based model is a novel tool for studying liver steatosis in mouse models on paraffin sections, and thus, can facilitate reliable quantification of the amount of steatosis in large preclinical study cohorts.

Automated image analysis of keratin 7 staining can predict disease outcome in primary sclerosing cholangitis

BACKGROUND AND AIMS

Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease that obstructs the bile ducts and causes liver cirrhosis and cholangiocarcinoma. Efficient surrogate markers are required to measure disease progression. The cytokeratin 7 (K7) load in a liver specimen is an independent prognostic indicator that can be measured from digitalized slides using artificial intelligence (AI)-based models.

METHODS

A K7-AI model 2.0 was built to measure the hepatocellular K7 load area of the parenchyma, portal tracts, and biliary epithelium. K7-stained PSC liver biopsy specimens (n = 295) were analyzed. A compound endpoint (liver transplantation, liver-related death, and cholangiocarcinoma) was applied in Kaplan-Meier survival analysis to measure AUC values and positive likelihood ratios for each histological variable detected by the model.

RESULTS

The K7-AI model 2.0 was a better prognostic tool than plasma alkaline phosphatase, the fibrosis stage evaluated by Nakanuma classification, or K7 score evaluated by a pathologist based on the AUC values of measured variables. A combination of parameters, such as portal tract volume and area of K7-positive hepatocytes analyzed by the model, produced an AUC of 0.81 for predicting the compound endpoint. Portal tract volume measured by the model correlated with the histological fibrosis stage.

CONCLUSIONS

The K7 staining of histological liver specimens in PSC provides significant information on disease outcomes through objective and reproducible data, including variables that cannot be measured by a human pathologist. The K7-AI model 2.0 could serve as a prognostic tool for clinical endpoints and as a surrogate marker in drug trials.

Abstract 4134: Stromal FAP expression is associated with MRI visibility and patient survival in prostate cancer

Some clinically significant prostate cancers (PCa) are missed by magnetic resonance imaging (MRI). We asked if the tumor stroma in surgically treated primary PCa lesions positive or negative with MRI are different in their cellular and molecular properties, and whether the differences are reflected to the clinical course of the disease.

We profiled the stromal and immune cell composition of MRI-classified tumor lesions by applying multiplexed fluorescence immunohistochemistry (mfIHC) and automated image analysis in a clinical cohort of 343 patients (Cohort I). We compared stromal variables between MRI-visible, invisible lesions, and benign tissue and assessed the predictive significance for biochemical recurrence (BCR) and disease-specific survival (DSS) using Cox regression and Log rank analysis. Subsequently, we carried out a prognostic validation of the identified biomarkers in a population-based cohort of 319 patients (Cohort II).

MRI true-positive lesions are different from benign tissue and MRI false-negative lesions in their stromal composition. CD163-positive cells and fibroblast activation protein (FAP)-positive cells were more abundant in MRI true-positive than in MRI false-negative lesions or benign areas. In MRI true-visible lesions, a high proportion of stromal FAP-positive cells was associated with PTEN status and increased immune infiltration (CD8+, CD163+), and predicted elevated risk for BCR. High FAP phenotype was confirmed to be a strong indicator of poor prognosis in two independent patient cohorts using also conventional IHC.

The molecular composition of the tumor stroma may determine whether early prostate lesions are detectable by MRI and associates with survival after surgical treatment.

Citation Format: Teijo Pellinen, Kevin Smith, Sami Blom, Riku Turkki, Annabrita Hemmes, Katja Välimäki, Juho Eineluoto, Anu Kenttämies, Stig Nordling, Olli Kallioniemi, Antti Rannikko, Tuomas Mirtti. Stromal FAP expression is associated with MRI visibility and patient survival in prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4134.

AI Model for Prostate Biopsies Predicts Cancer Survival

An artificial intelligence (AI) algorithm for prostate cancer detection and grading was developed for clinical diagnostics on biopsies. The study cohort included 4221 scanned slides from 872 biopsy sessions at the HUS Helsinki University Hospital during 2016–2017 and a subcohort of 126 patients treated by robot-assisted radical prostatectomy (RALP) during 2016–2019. In the validation cohort (n = 391), the model detected cancer with a sensitivity of 98% and specificity of 98% (weighted kappa 0.96 compared with the pathologist’s diagnosis). Algorithm-based detection of the grade area recapitulated the pathologist’s grade group. The area of AI-detected cancer was associated with extra-prostatic extension (G5 OR: 48.52; 95% CI 1.11–8.33), seminal vesicle invasion (cribriform G4 OR: 2.46; 95% CI 0.15–1.7; G5 OR: 5.58; 95% CI 0.45–3.42), and lymph node involvement (cribriform G4 OR: 2.66; 95% CI 0.2–1.8; G5 OR: 4.09; 95% CI 0.22–3). Algorithm-detected grade group 3–5 prostate cancer depicted increased risk for biochemical recurrence compared with grade groups 1–2 (HR: 5.91; 95% CI 1.96–17.83). This study showed that a deep learning model not only can find and grade prostate cancer on biopsies comparably with pathologists but also can predict adverse staging and probability for recurrence after surgical treatment.

Stromal FAP Expression is Associated with MRI Visibility and Patient Survival in Prostate Cancer

Some clinically significant prostate cancers are missed by MRI. We asked whether the tumor stroma in surgically treated localized prostate cancer lesions positive or negative with MRI are different in their cellular and molecular properties, and whether the differences are reflected to the clinical course of the disease. We profiled the stromal and immune cell composition of MRI-classified tumor lesions by applying multiplexed fluorescence IHC (mfIHC) and automated image analysis in a clinical cohort of 343 patients (cohort I). We compared stromal variables between MRI-visible lesions, invisible lesions, and benign tissue and assessed the predictive significance for biochemical recurrence (BCR) and disease-specific survival (DSS) using Cox regression and log-rank analysis. Subsequently, we carried out a prognostic validation of the identified biomarkers in a population-based cohort of 319 patients (cohort II). MRI true-positive lesions are different from benign tissue and MRI false-negative lesions in their stromal composition. CD163⁺ cells (macrophages) and fibroblast activation protein (FAP)⁺ cells were more abundant in MRI true-positive than in MRI false-negative lesions or benign areas. In MRI true-visible lesions, a high proportion of stromal FAP⁺ cells was associated with PTEN status and increased immune infiltration (CD8⁺, CD163⁺), and predicted elevated risk for BCR. High FAP phenotype was confirmed to be a strong indicator of poor prognosis in two independent patient cohorts using also conventional IHC. The molecular composition of the tumor stroma may determine whether early prostate lesions are detectable by MRI and associates with survival after surgical treatment.

Significance

These findings may have a significant impact on clinical decision making as more radical treatments may be recommended for men with a combination of MRI-visible primary tumors and FAP⁺ tumor stroma.

Development and validation of a supervised deep learning algorithm for automated whole-slide programmed death-ligand 1 tumour proportion score assessment in non-small cell lung cancer

AIMS

Immunohistochemical programmed death-ligand 1 (PD-L1) staining to predict responsiveness to immunotherapy in patients with advanced non-small cell lung cancer (NSCLC) has several drawbacks: a robust gold standard is lacking, and there is substantial interobserver and intraobserver variance, with up to 20% discordance around cutoff points. The aim of this study was to develop a new deep learning-based PD-L1 tumour proportion score (TPS) algorithm, trained and validated on a routine diagnostic dataset of digitised PD-L1 (22C3, laboratory-developed test)-stained samples.

METHODS AND RESULTS

We designed a fully supervised deep learning algorithm for whole-slide PD-L1 assessment, consisting of four sequential convolutional neural networks (CNNs), using aiforia create software. We included 199 whole slide images (WSIs) of 'routine diagnostic' histology samples from stage IV NSCLC patients, and trained the algorithm by using a training set of 60 representative cases. We validated the algorithm by comparing the algorithm TPS with the reference score in a held-out validation set. The algorithm had similar concordance with the reference score (79%) as the pathologists had with one another (75%). The intraclass coefficient was 0.96 and Cohen's κ coefficient was 0.69 for the algorithm. Around the 1% and 50% cutoff points, concordance was also similar between pathologists and the algorithm.

CONCLUSIONS

We designed a new, deep learning-based PD-L1 TPS algorithm that is similarly able to assess PD-L1 expression in daily routine diagnostic cases as pathologists. Successful validation on routine diagnostic WSIs and detailed visual feedback show that this algorithm meets the requirements for functioning as a 'scoring assistant'.

Automated assessment of Ki-67 proliferation index in neuroendocrine tumors by deep learning

The Ki‐67 proliferation index (PI) is a prognostic factor in neuroendocrine tumors (NETs) and defines tumor grade. Analysis of Ki‐67 PI requires calculation of Ki‐67‐positive and Ki‐67‐negative tumor cells, which is highly subjective. To overcome this, we developed a deep learning‐based Ki‐67 PI algorithm (KAI) that objectively calculates Ki‐67 PI. Our study material consisted of NETs divided into training (n = 39), testing (n = 124), and validation (n = 60) series. All slides were digitized and processed in the Aiforia^® Create (Aiforia Technologies, Helsinki, Finland) platform. The ICC between the pathologists and the KAI was 0.89. In 46% of the tumors, the Ki‐67 PIs calculated by the pathologists and the KAI were the same. In 12% of the tumors, the Ki‐67 PI calculated by the KAI was 1% lower and in 42% of the tumors on average 3% higher. The DL‐based Ki‐67 PI algorithm yields results similar to human observers. While the algorithm cannot replace the pathologist, it can assist in the laborious Ki‐67 PI assessment of NETs. In the future, this approach could be useful in, for example, multi‐center clinical trials where objective estimation of Ki‐67 PI is crucial.

Chronic cholestasis detection by a novel tool: automated analysis of cytokeratin 7-stained liver specimens

Background

The objective was to build a novel method for automated image analysis to locate and quantify the number of cytokeratin 7 (K7)-positive hepatocytes reflecting cholestasis by applying deep learning neural networks (AI model) in a cohort of 210 liver specimens. We aimed to study the correlation between the AI model’s results and disease progression. The cohort of liver biopsies which served as a model of chronic cholestatic liver disease comprised of patients diagnosed with primary sclerosing cholangitis (PSC).

Methods

In a cohort of patients with PSC identified from the PSC registry of the University Hospital of Helsinki, their K7-stained liver biopsy specimens were scored by a pathologist (human K7 score) and then digitally analyzed for K7-positive hepatocytes (K7%area). The digital analysis was by a K7-AI model created in an Aiforia Technologies cloud platform. For validation, values were human K7 score, stage of disease (Metavir and Nakunuma fibrosis score), and plasma liver enzymes indicating clinical cholestasis, all subjected to correlation analysis.

Results

The K7-AI model results (K7%area) correlated with the human K7 score (0.896; p < 2.2e^− 16). In addition, K7%area correlated with stage of PSC (Metavir 0.446; p < 1.849e^− 10 and Nakanuma 0.424; p < 4.23e^− 10) and with plasma alkaline phosphatase (P-ALP) levels (0.369, p < 5.749e^− 5).

Conclusions

The accuracy of the AI-based analysis was comparable to that of the human K7 score. Automated quantitative image analysis correlated with stage of PSC and with P-ALP. Based on the results of the K7-AI model, we recommend K7 staining in the assessment of cholestasis by means of automated methods that provide fast (9.75 s/specimen) quantitative analysis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13000-021-01102-6.

Fibroblast as a critical stromal cell type determining prognosis in prostate cancer

BACKGROUND

Tumor stroma associates with prostate cancer (PCa) progression, but its specific cellular composition and association to patient survival outcome have not been characterized.

METHODS

We analyzed stromal composition in human PCa using multiplex immunohistochemistry and quantitative, high-resolution image analysis in two retrospective, formalin-fixed paraffin embedded observational clinical cohorts (Cohort I, n = 117; Cohort II, n = 340) using PCa-specific mortality as outcome measurement.

RESULTS

A high proportion of fibroblasts associated with aggressive disease and castration-resistant prostate cancer (CRPC). In a multivariate analysis, increase in fibroblast proportion predicted poor cancer-specific outcome independently in the two clinical cohorts studied.

CONCLUSIONS

Fibroblasts were the most important cell type in determining prognosis in PCa and associated with CRPC. Thus, the stromal composition could be critically important in developing diagnostic and therapeutic approaches to aggressive prostate cancer.

Abstract 4447: Deep neural network for automatic histopathologic analysis of murine lung tumors

Lung cancer is the leading cause of cancer-related deaths worldwide, and genetically-engineered mouse models (GEMMs) of cancer provide important mechanistic and preclinical insights into this deadly disease. In particular, the “KP” model enables lung-specific inducible activation of oncogenic Kras G12D, and loss of Trp53, the two most common driver events of human non-small cell lung cancer (NSCLC). Importantly, the KP model is widely used and faithfully recapitulates molecular and histopathological features of the human disease, including progression from early hyperplasia and adenoma to invasive adenocarcinoma. However, the KP model results in multi-focal and heterogeneous tumor burden, and there is a need for improved tools to increase throughput and decrease subjectivity of tumor burden quantification and histopathological analyses.

To this end, we trained a convolutional neural network (CNN) for semantic multi-class segmentation using the Aiforia(R) platform. The CNN was trained to classify and detect lung parenchyma, NSCLC tumors, and NSCLC tumor grades (grade 1-4). For supervised training, we used selected areas from 93 hematoxylin and eosin stained slides. For validation, we analyzed 34 slides completely independent of the CNN training. Tumor grades were manually annotated on the validation slides blinded to the CNN results. The overall F1 score of the CNN in grade classification was 98% and total area error 0.3%. The grade-specific F1-scores were 89%, 97%, 99%, and 98% for grades 1, 2, 3, and 4, respectively. Corresponding grade-specific total area errors were 0.4%, 0.2%, 0.4%, and 0.1%. Manual scoring independent of the training and CNN yielded similar tumor burden and grading results. In addition, the algorithm accurately recapitulates the increased tumor burden and grade seen in KP tumors harboring additional mutation of the tumor suppressor Keap1, and the delayed kinetics of KP tumors harboring a strong T cell antigen, in independent datasets. We have also extended this methodology to identification of tumors in a GEMM of small cell lung cancer, a distinct class of lung cancer with poor prognosis.

In conclusion, we demonstrate that deep neural networks can be used for automated analysis and grading of preclinical models of lung cancer. We anticipate that this powerful technology will increase the throughput, sensitivity and reproducibility of hypothesis-driven studies of factors influencing tumor progression and immune response in mouse models of lung cancer.

Citation Format: Peter Maxwell Kienitz Westcott, Tuomas Pitkänen, Sami Blom, Thomas Westerling, Tuomas Ropponen, Nathan Sacks, Katherine Wu, Roderick Bronson, Tuomas Tammela, Tyler Jacks. Deep neural network for automatic histopathologic analysis of murine lung tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4447.

Combined epithelial marker analysis of tumour budding in stage II colorectal cancer

Tumour budding predicts survival of stage II colorectal cancer (CRC) and has been suggested to be associated with epithelial‐to‐mesenchymal transition (EMT). However, the underlying molecular changes of tumour budding remain poorly understood. Here, we performed multiplex immunohistochemistry (mIHC) to phenotypically profile tumours using known EMT‐associated markers: E‐cadherin (adherence junctions), integrin β4 (ITGB4; basement membrane), ZO‐1 (tight junctions), and pan‐cytokeratin. A subpopulation of patients showed high ITGB4 expression in tumour buds, and this coincided with a switch of ITGB4 localisation from the basal membrane of intact epithelium to the cytoplasm of budding cells. Digital image analysis demonstrated that tumour budding with high ITGB4 expression in tissue microarray (TMA) cores correlated with tumour budding assessed from haematoxylin and eosin (H&E) whole sections and independently predicted poor disease‐specific survival in two independent stage II CRC cohorts (hazard ratio [HR] = 4.50 (95% confidence interval [CI] = 1.50–13.5), n = 232; HR = 3.52 (95% CI = 1.30–9.53), n = 72). Furthermore, digitally obtained ITGB4‐high bud count in random TMA cores was better associated with survival outcome than visual tumour bud count in corresponding H&E‐stained samples. In summary, the mIHC‐based phenotypic profiling of human tumour tissue shows strong potential for the molecular characterisation of tumour biology and for the discovery of novel prognostic biomarkers.

Implementation of deep neural networks to count dopamine neurons in substantia nigra

Unbiased estimates of neuron numbers within substantia nigra are crucial for experimental Parkinson's disease models and gene‐function studies. Unbiased stereological counting techniques with optical fractionation are successfully implemented, but are extremely laborious and time‐consuming. The development of neural networks and deep learning has opened a new way to teach computers to count neurons. Implementation of a programming paradigm enables a computer to learn from the data and development of an automated cell counting method. The advantages of computerized counting are reproducibility, elimination of human error and fast high‐capacity analysis. We implemented whole‐slide digital imaging and deep convolutional neural networks (CNN) to count substantia nigra dopamine neurons. We compared the results of the developed method against independent manual counting by human observers and validated the CNN algorithm against previously published data in rats and mice, where tyrosine hydroxylase (TH)‐immunoreactive neurons were counted using unbiased stereology. The developed CNN algorithm and fully cloud‐embedded Aiforia™ platform provide robust and fast analysis of dopamine neurons in rat and mouse substantia nigra.

Cell of Origin Links Histotype Spectrum to Immune Microenvironment Diversity in Non-small-Cell Lung Cancer Driven by Mutant Kras and Loss of Lkb1

Lung cancers exhibit pronounced functional heterogeneity, confounding precision medicine. We studied how the cell of origin contributes to phenotypic heterogeneity following conditional expression of Kras^G12D and loss of Lkb1 (Kras;Lkb1). Using progenitor cell-type-restricted adenoviral Cre to target cells expressing surfactant protein C (SPC) or club cell antigen 10 (CC10), we show that Ad5-CC10-Cre-infected mice exhibit a shorter latency compared with Ad5-SPC-Cre cohorts. We further demonstrate that CC10⁺ cells are the predominant progenitors of adenosquamous carcinoma (ASC) tumors and give rise to a wider spectrum of histotypes that includes mucinous and acinar adenocarcinomas. Transcriptome analysis shows ASC histotype-specific upregulation of pro-inflammatory and immunomodulatory genes. This is accompanied by an ASC-specific immunosuppressive environment, consisting of downregulated MHC genes, recruitment of CD11b⁺ Gr-1⁺ tumor-associated neutrophils (TANs), and decreased T cell numbers. We conclude that progenitor cell-specific etiology influences the Kras;Lkb1-driven tumor histopathology spectrum and histotype-specific immune microenvironment.

Abstract 5732: PI3K/Akt activity regulates androgen receptor expression and predicts poor clinical outcome in non-metastatic hormone-naïve prostate cancer

Activation of PI3K/Akt pathway is associated with adverse outcome and aggressive disease in many cancers. In prostate cancer (PCa), the activity of this pathway has been shown to promote disease progression and metastasis. However, it is still controversial how PI3K/Akt regulates androgen receptor (AR), a central signaling molecule in prostate pathophysiology, and whether it has an active role in hormone naïve non-metastatic PCa. Here, we show using immunohistochemistry (IHC) and advanced quantitative multiplexed IHC that the expression of phosphorylated-Akt(S473) and AR are highly correlated in clinical PCa, even at the cellular level. Furthermore, we found that high expression of p-Akt(S473) predicts poor clinical outcome in two independent hormone-naïve non-metastatic PCa cohorts. To study whether PI3K/Akt regulates AR expression, we performed an in vitro drug screen with 32 PI3K/Akt/mTOR inhibitors in PC346C, an AR expressing cell line derived from a hormone-naïve primary tumor of prostate. We observed a strong correlation between p-Akt(S473) and AR also in vitro in individual cells independent of the inhibitor used. Although both PI3K and Akt specific inhibition reduced cell viability, the response in nuclear expression of AR was highly dependent on the target of inhibition: Akt specific inhibition reduced AR nuclear expression and resulted in large, spindle-shaped cells, whereas PI3K specific inhibition increased AR nuclear expression and resulted in smaller, round-shaped cells. These data suggest that PI3K and Akt have different roles in sustaining AR activity in PCa as perturbations of the two components leads to differential responses in terms of AR nuclear expression and cell morphology. In conclusion, activated Akt associates with AR expression and predicts poor clinical outcome in hormone-naïve non-metastatic PCa. Furthermore, the differing roles of PI3K and Akt in AR regulation warrants for further studies as it may have implications in the design of PCa therapy targeting PI3K/Akt, especially when the inhibitors are administered in combination with anti-androgens.

Citation Format: Sami Blom, Petra Mäki-Teeri, Andrew Erickson, Lassi Paavolainen, Tuomas Mirtti, Antti Rannikko, Swapnil Potdar, Päivi Östling, Wytske van Weerden, Olli Kallioniemi, Teijo Pellinen. PI3K/Akt activity regulates androgen receptor expression and predicts poor clinical outcome in non-metastatic hormone-naïve prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5732. doi:10.1158/1538-7445.AM2017-5732

Abstract 1698: Systems pathology for characterization of cancer model systems in a multicenter IMI-PREDECT project

Despite of our increased understanding of cancer biology, the majority of anti-cancer therapies fail at late-stage clinical trials. Thus, there is an urgent need to develop and validate novel preclinical models that could predict drug efficacy in humans. For this purpose, as a part of IMI-PREDECT public-private research consortium, this study describes methodology and infrastructure for characterization and comparison of preclinical models for drug target validation applying a systems pathology approach. Formalin-fixed paraffin-embedded (FFPE) samples from 1050 different in vitro and in vivo models of breast, prostate and lung cancers, as well as 364 clinical tumors from the same origin, were collected from 26 PREDECT collaborators across the EU. We established standard operating procedures for central processing of FFPE samples, including tissue microarray (TMA) construction and immunohistochemistry (IHC) with 15 different antibodies (CK8, Ki67, p-histone H3, ER, AR, p-AKT, p-ERK, p-p38, γH2AX, cleaved caspase 3, p-MET, HIF1α, p63, vimentin, E-cadherin). We constructed 50 TMA blocks, from which sections were cut and stained as well as digitized using whole slide imaging. Images were hosted on a WebMicroscope digital pathology platform and sample metadata on a PREDECT Metadata database (MBase). We developed image analysis methods for the detection and quantification of IHC biomarkers in the 48,800 stained TMA spots. As a proof-of-concept, we compared MCF-7 on several preclinical platforms including cell cultures, xenografts and xenograft tissue slices. Our results of the integrated biomarker phenotype suggest that of the various MCF-7 in vivo and ex vivo complex cell culture models, the xenograft tissue slice model was the most similar model platform to human clinical samples. In summary, we established a systems pathology approach to analyse and compare novel preclinical cancer models with IHC and digital imaging. The intention is that this large database will be made publicly available on the web as images and summary data that could be broadly useful to the community of cancer researchers and drug developers in comparing cancer model systems. The established infrastructure and workflow integrating molecular and digital pathology in a large-scale consortium setting could be applied to quantitative characterisation of consortium data in collaborations similar to PREDECT.

Citation Format: Sami Blom, Yinhai Wang, Tauno Metsalu, Tiina Vesterinen, Teijo Pellinen, Anne Grote, Nina Linder, Jenni Säilä, Katja Välimäki, Ruusu-Maria Kovanen, Outi Monni, Panu Kovanen, Emma Davies, Kristin Stock, Marta Estrada, Georgios Sflomos, Sylvia Grünewald, Catarina Brito, Julia Schüler, Ronald de Hoogt, Cathrin Brisken, Heiko van der Kuip, Wytske van Weerden, Simon Barry, Wolgang Sommergruber, Elizabeth Anderson, Matthias Nees, Juha Klefström, Jaak Vilo, Emmy Verschuren, Ralph Graeser, John Hickman, Johan Lundin, Olli Kallioniemi. Systems pathology for characterization of cancer model systems in a multicenter IMI-PREDECT project. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1698. doi:10.1158/1538-7445.AM2015-1698

Abstract 207: Caveolin-1 drives oncogenic TGFβ effects in prostate cancer: in vitro mechanistic insights integrated with systems pathology visualization in primary tumor samples

Caveolin-1 (Cav1) can have both a tumor suppressive or promoting role depending on cancer tissue, and its potential active role in the progression of prostate cancer remains poorly understood. In this study, we found out that Cav1 represses the expression of E-cadherin in prostate cancer cell lines, and that the expressions of these two genes show an inverse correlation both at the mRNA and protein levels in PCa patient samples. We show that the combined loss of E-cadherin and gain of Cav1 expression is an early event in a subgroup of patients (&lt;10%) (n(patients) = 80, n(glands) = 263, r = -0.35, p&lt;0.001). Cav1 positively regulates TGFβ targets with known oncogenic and pro-EMT functions (e.g. Slug, Pai-1), and its absence switches the TGFβ to become inhibitory to PCa cells by inducing p21 (CDKN1A) and STAT1 signaling. These changes were strongly associated with changes in 3D-invasive growth. By using RNAi image-based screening, we found out that Cav1 is positively regulated by beta1-integrins and cell adhesion to ECM. This association together with different alpha-integrins was visualized and quantified in human prostate cancer patient tissues (n = 80) using novel multiplexed fluorescent imaging. Both the integrin beta1 and alpha2 showed high correlation with Cav1 expression (p-values&lt;0.001) in tumor epithelium as well as in stroma.

In conclusion, the results suggest that Cav1 could be an important promoter of invasive growth in a subgroup of PCa, where it is upregulated by beta1-integrins, and where it promotes oncogenic TGFβ signaling.

Citation Format: Teijo Pellinen, Sami Blom, Sara Sanches, Katja Välimäki, John-Patrick Mpindi, Antti Rannikko, Raffaele Strippoli, Maria Montoya, Hind Azegrouz, Tuomas Mirtti, Miguel del Pozo, Olli Kallioniemi. Caveolin-1 drives oncogenic TGFβ effects in prostate cancer: in vitro mechanistic insights integrated with systems pathology visualization in primary tumor samples. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 207. doi:10.1158/1538-7445.AM2015-207

Aptamer-directed lanthanide chelate self-assembly for rapid thrombin detection

We report a sensitive assay method for homogeneous thrombin detection. The method is based on lanthanide chelate complementation, where the luminescent complex is split into two separate label moieties, which are intrinsically non-luminescent. A luminescent mixed chelate complex is formed only when the label moieties are brought into close proximity directed by two separate binding events of aptamers to the analyte. This results in high specificity in signal generation while time-resolved fluorescence detection eliminates the short lifetime autofluorescence, which is inherent to many homogeneous assays and limits their applicability. The developed method is also very rapid as the maximum signal is obtained in just five minutes. Lanthanide chelate complementation can be applied for the detection of other proteins when two binders recognizing separate epitopes of the analyte are available.

Abstract 721: Multiplexed systems pathology for in-depth analysis of the tumor microenvironment: a strong correlation between pAkt and androgen receptor in the epithelial component of prostate cancer

It is important to understand the heterogeneity across the cancer tissue as well as the spatial interactions and signaling between the different components that constitute the tumor microenvironment; epithelium, stroma, inflammatory cells, etc. The state of the tumor microenvironment and its variability within and in between tumors may give important insights on cancer initiation and progression, diagnostics, prognostics and therapeutic assessment. Thus, there is a major need to assess multiple molecular targets simultaneously, quantitatively, and within the specific tumor sub-compartments. Here, we demonstrate a systems pathology approach for quantitative detection of multiple proteins in formalin-fixed paraffin-embedded patient tissue samples using a combination of immunohistochemistry and high-throughput fluorescence imaging.

The study cohort consisted of 50 prostate cancer tissue samples from 25 patients exhibiting different disease stages in a tissue microarray (TMA) format. We developed a multiplexed assay for detecting pAkt, AR, and cytokeratin 8 (CK8) simultaneously using fluorescent nanoparticle labeling and high-throughput quantitative imaging. Fluorescent signals of pAkt and AR were automatically quantified within the segmented CK8-positive prostate epithelium. The stromal compartment and the non-malignant prostate tissue were analysed separately.

Our results showed that the expression of pAkt and AR correlated strongly in the prostate cancer epithelium (Pearson correlation coefficient 0.89, p &lt; 0.001). This suggests that the androgen-dependent signaling is strongly associated with the pAkt signaling in cancer, with potential ramifications for administrating and monitoring anti-androgens and/or Akt-inhibitor therapy. The systems pathology approach enables quantitative in situ multianalyte detection for any key cancer signaling proteins. The analyses can be readily expanded to TMAs of large prospective or clinical trial cohorts with automated imaging. The ability to quantitatively study the spatial arrangement of the tumor microenvironment could have decisive advantages and improve the diagnostic and therapeutic predictions.

Citation Format: Sami Blom, Petra Mäki-Teeri, Tuomas Mirtti, Antti Rannikko, Jenni Säilä, Teijo Pellinen, Olli Kallioniemi. Multiplexed systems pathology for in-depth analysis of the tumor microenvironment: a strong correlation between pAkt and androgen receptor in the epithelial component of prostate cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 721. doi:10.1158/1538-7445.AM2013-721

Effects of food deprivation and handling stress on head kidney 17alpha-hydroxyprogesterone 21-hydroxylase activity, plasma cortisol and the activities of liver detoxification enzymes in rainbow trout

The 21-hydroxylation of 17alpha-hydroxyprogesterone is one step in the biosynthesis of corticosteroids. Both 7 days of handling-induced stress and 7 weeks of food deprivation significantly elevated head kidney microsomal 17alpha-hydroxyprogesterone 21-hydroxylase activity in juvenile rainbow trout. The increased 21-hydroxylase activity was not paralleled by changes in plasma cortisol levels induced by handling stress whereas food deprivation for 3 and 7 weeks increased both 21-hydroxylase activity and plasma cortisol levels significantly. Food deprivation in rainbow trout affected detoxification enzyme activities, namely glutathione-S-transferase (GST), uridine-di-phosphate glucuronosyltransferase (UGT) and glutathione reductase (GR) activities in the liver. Together our observations suggest that experimental conditions can affect experimental results, especially the values of parameters like GST, UGT and GR. Furthermore, alterations in the metabolic state of the liver caused by stress or food deprivation can alter the balance between detoxification enzymes in rainbow trout liver.

Surgical treatment for epilepsy: a retrospective Swedish multicenter study

The characteristics of patients suffering from drug resistant epilepsy, including the results of the preoperative evaluation and epilepsy surgery were retrospectively analyzed in a Swedish multicenter 10-year cohort of children and adults. Altogether 152 patients (65 children and 87 adults) treated during the period 1980-1990 in three epilepsy centers were included and followed-up 2 years after surgery. Median age at onset of seizures was 4 years for the children and 12 years for the adults. A localization related epilepsy was present in 85% of the children and in 95% of the adults. The mean number of seizure types in the children was 1.7 (range 1-4) and in the adults 1.8 (range 1-4). The median monthly seizure frequency was 52 and 15 for children and adults respectively. Resective surgery was performed in 143 cases (94 temporal, 31 extratemporal, 9 multilobar and 9 major resection procedures) and palliative procedures in 16 cases (13 callosotomies and 3 stereotactic amygdalotomies). Postoperative neurological deficits were detected in 9% of the patients after temporal lobe resections and in 15% of the patients after extratemporal and multilobar resection procedures. Two years after resective surgery 53% of the children and 49% of the adults were seizure free. Another 25% of the patients had a more than 50% reduction of seizure frequency. In the postoperative non seizure free group of patients there was a negative correlation between decrease in weighted seizure severity and decrease in seizure frequency. This finding stresses the need for including other parameters than seizure frequency when evaluating the outcome of epilepsy surgery.

Effect of platelet-derived growth factor-BB on bone formation in calvarial defects: an experimental study in rabbits

The effect of recombinant human platelet-derived growth factor-BB (rhPDGF-BB) on bone healing was examined in calvarial defects in rabbits. Bicortical circular (critical size) defects were prepared in the calvarial bone of 16 rabbits. The defects were closed on the dural side and covered externally with expanded polytetrafluoroethylene membranes to prevent interference with osteogenesis within the defect by the surrounding tissue and to keep the growth factor in place. A single dose of methylcellulose gel (4.4%) with (n = 8) or without rhPDGF-BB (50 micrograms/ml) (n = 8) was applied to the defects, and the bone formation was evaluated after 8 weeks. Healing of defects in both groups was characterized by the presence of newly formed bone along the edges of the original defect and by a central area of fibrous connective tissue. The newly formed bone in the rhPDGF-BB treated defects had a trabecular structure; in contrast, a more compact structure was found in the control defects. In the rhPDGF-BB-treated defects, the bone ingrowth was 51.8 +/- 7.1% compared to 30.5 +/- 3.3% in the control defects. Furthermore, the amount of mineralized tissue was increased 112% in the rhPDGF-BB group. The amount of bone marrow was increased 75% in the rhPDGF-BB-treated defect. The porosity of cortical lamella in the newly formed bone was 84% higher in the rhPDGF-BB-treated defects compared to the control. These results show that administration of a single dose of rhPDGF-BB stimulates bone formation in critical size calvarial defects.

Conductivity ratios of the scalp-skull-brain head model in estimating equivalent dipole sources in human brain

The dipole tracing (DT) method estimates the position and vector dipole moment of an equivalent current dipole by minimizing the mean squared error of the dipole potentials at the surface electrode positions. In the scalp-skull-brain/DT (SSB/DT) method, which we have developed, the head model consists of three compartments of uniform conductors corresponding to the scalp, skull and brain. The accuracy of the calculations are mainly dependent on the ratios of the conductivities of the three compartments. The best result was obtained with the conductivity ratios of 1:1/80:1 for the scalp, skull and brain compartments, respectively.

Marriage and cohabitation in a changing society: experience of Norwegian men and women born in 1945 and 1960

"Retrospective survey data are used to estimate the effect of various factors on the transitions to first marriage or first cohabitation among single Norwegian men and women born in 1945 and 1960. A high educational level is not found to reduce marriage intensities for women, although educational enrollment appears to be less compatible with marriage for women than men. The effect of employment varies according to prevailing sex-role expectations. The data support the assumption that modern cohabitation developed from two socially opposite origins, the educated elite and the working class. A social value dimension is assumed to have a major effect upon the present-day choice between marriage and cohabitation." (SUMMARY IN FRE)

Location of electric current sources in the human brain estimated by the dipole tracing method of the scalp-skull-brain (SSB) head model

Using a realistic, 3-shell head model including the scalp (S), skull (S) and brain (B) with conductivity ratios of 1:1/80:1, respectively, the electrical activity in the human brain recorded by conventional electroencephalography was approximated by 1 or 2 equivalent current dipoles. The dipole locations and vector moments were estimated by minimizing the squared difference between the potentials actually recorded from the scalp and those theoretically calculated from the equivalent dipoles. The validity of this dipole tracing method (the DT of the SSB head model) was tested in patients with focal epileptic seizures undergoing presurgical evaluation with intracranial subdural strip electrodes. Weak currents were passed through 1 or 2 pairs of subdural electrodes to create artificial dipoles. The dipole estimations correctly distinguished between single and double generator sources, but there were certain dislocations of the calculated dipoles. The average error of dislocation was found to be 8.5 mm for the 1-dipole model. That for the 2-dipole model was 6 mm for one of the components and 18 mm for the other. It was concluded that the DT method of the SSB head model can be a valuable clinical tool in 3-dimensional localization of focal epileptic discharges in the human brain.

A comparison of the effect of epidermal growth factor, platelet-derived growth factor, and fibroblast growth factor on rat periodontal ligament fibroblast-like cells' DNA synthesis and morphology

An enhanced formation of bone, dentin, and collagen fibers in periodontal wounds after application of polypeptide growth factors has recently been reported. However, the complex environment in vivo makes it impossible to determine the specific effects of growth factors on various cells involved in the wound-healing process. We have therefore investigated the mitogenic and morphogenic effects of recombinant epidermal growth factor (rEGF), natural platelet-derived growth factor (nPDGF), and natural fibroblast growth factor (nFGF) on periodontal ligament fibroblast-like cells. A cell line was established from rat PDL tissue. The cell line was characterized according to morphology, growth pattern, cytoskeletal proteins, and growth kinetics. The mitogenic effect of growth factors was assessed by incorporation of [3H]thymidine in the cellular DNA for 4 hours. Differences between groups of observations were assessed by the Student t-test. The morphogenic effects of growth factors were described with respect to growth pattern, cell orientation, and cell and nucleus form after a random photographic recording. The fibroblast-like cell type and the non-transformed phenotype of the cell line have been identified by the presence of parameters considered to be characteristic of a normal fibroblast-like cell line. The morphogenic analysis of both experimental and control cultures showed a monolayer of adherent cells with spindle or stellate morphology, a random alignment and round or elongated nuclei. Incorporation of [3H]-thymidine was increased in a dose-dependent manner by all growth factors. Maximal effect on the DNA synthesis was: rEGF, 131%; nPDGF, 274%; and nFGF, 182%.(ABSTRACT TRUNCATED AT 250 WORDS)

The effect of insulin-like growth factor-I and human growth hormone on periodontal ligament fibroblast morphology, growth pattern, DNA synthesis, and receptor binding

Repopulation of the detached root surface by cells from the periodontal ligament (PDL) is a prerequisite for new attachment formation. Stimulation of PDL-cell growth may therefore serve as an essential method to enhance formation of new attachment. Studies have demonstrated that insulin-like growth factor-I (IGF-I) has a mitogenic effect on fibroblasts originating from various connective tissues and cell-lines. Further, human growth hormone (hGH) is known to regulate the plasma concentration of IGF-I and to mediate cellular biological effects. In the present study we examined the effect of IGF-I and hGH on morphology, growth pattern, and DNA synthesis. The expression of IGF-I and hGH receptors on the surface of cultured PDL fibroblasts is also described. A primary fibroblastic cell line was established from rat PDL tissue, and blind, photographic recordings of morphology and growth pattern, as well as incorporation of [3H]thymidine in cellular DNA, was carried out in the presence and absence of IGF-I and hGH. The presence of specific membrane receptors was investigated by binding of [125I]IGF-I and [125I]hGH. The analysis of photographs showed that IGF-I and hGH had no effect on morphology and growth pattern. Incorporation of 3H-thymidine, however, was increased in a dose-dependent manner by IGF-I, whereas hGH alone or in combination with IGF-I produced no dose-dependent response. Maximum effect (% of control) on DNA synthesis was 176% for IGF-I and 91% for hGH.(ABSTRACT TRUNCATED AT 250 WORDS)

Generator mechanisms of epileptic potentials analyzed by dipole tracing method

A new dipole tracing method, based on a realistic head model, was used to determine dipole locations and vector moments of interictal convexity sharp waves recorded (with conventional EEG technique) from the right fronto-temporal region in a patient with partial complex seizures. When the dipole locations in the head model were compared to MRI scans, the majority of the sharp wave dipoles were found to be located in the right hippocampal area. For individual sharp waves, the hippocampal dipoles moved along tracks corresponding to the vector moment directions, suggesting that the electrical sources of the convexity sharp waves were somato-dendritic currents which spread rapidly from one neuron group to the next in the hippocampal area. Previous long-term subdural recording had shown seizure onset in this area. After right-sided anterior temporal lobectomy including the hippocampus the patient has been seizure-free for three months.

Arsenic exposure to smelter workers. Clinical and neurophysiological studies

Forty-seven copper smelter workers, exposed to airborne arsenic for 8-40 years, were examined clinically with electromyography, and the motor and sensory conduction velocities in their arms and legs were determined. Fifty age-matched industrial workers not exposed to arsenic formed a reference group. The level of arsenic in the air at the smeltery was estimated to be below 500 micrograms/m3 before 1975 and approximately 50 micrograms/m3 thereafter. Urine analyses of arsenic showed a mean value of 71 micrograms/l (1 mumol/l) in the exposed group; this value is lower than that found in earlier studies reporting clinically detectable neuropathy. Only minor neurological and electromyographic abnormalities were found. A slightly reduced nerve conduction velocity in two or more peripheral nerves was more common among the arsenic workers than the referents, and a statistically significant correlation between cumulative exposure to arsenic and reduced nerve conduction velocity in three peripheral motor nerves was found. This occurrence was interpreted as a sign of slight subclinical neuropathy. In conclusion the risk of clinically significant neuropathy is small when exposure is kept below 50 micrograms/m3 in workroom air. The subclinical findings may be of interest in relation to the prevention of early adverse health effects from arsenic exposure.

Intracranially recorded compound action potentials from the human trigeminal nerve

Potentials from the sensory root of the trigeminal nerve in the cerebello-pontine angle, evoked by electrical stimulation of the different trigeminal divisions, were recorded during neurosurgical operations. Two components of the compound action potential could be distinguished: one early with large amplitude and one late with small amplitude. It is concluded that the two components represent activity in A alpha and A delta fibers respectively.

Phrenic nerve stimulation (diaphragm pacing) in respiratory paralysis

Phrenic nerve stimulators (diaphragm pacers) were implanted in 16 patients with partial or total respiratory insufficiency due to high cervical medullary lesions and brain stem lesions (14 cases) or central hypoventilation syndrome (2 cases). At 5-72 months' follow-up (M = 35) 12 patients are entirely independent of conventional respirator, 2 of them after 8 years of total respirator dependency. Two patients are dead and the final 2 cases had limited help from diaphragm pacing.

Effects of amobarbital and methohexital on epileptic activity in mesial temporal structures in epileptic patients. An EEG study with depth electrodes

Bilateral intracarotidal Amytal (amobarbital) tests for evaluation of speech and memory function were performed during preoperative evaluation of 30 patients with drug-resistant epilepsy. In 8 of these patients (16 tests), having partial complex epilepsy, EEG was recorded with depth electrodes, implanted bilaterally in anterior mesial temporal structures. The EEGs during 13 tests could be quantified with regard to spike activity. A rapid increase in spike frequency was observed ipsilateral to the injection in all tests but one. No seizure activity or clinical seizures were provoked. This previously unnoticed effect of amobarbital could be due to a direct excitatory effect of the drug on epileptic temporal neurones or, alternatively, to a release of interictal inhibition, exerted upon these neurons by other structures. In 4 patients, the effect was compared with that of methohexital, another barbiturate known to have excitatory effects upon epileptic activity.

Preoperative and postoperative memory testing of epileptic patients

The present paper has four main objectives. First, it proposes a general theoretical orientation of memory functions as a basis for the research to be discussed. This framework is functionalistic/interactionistic in nature, in that memory is seen as an interaction between available cognitive capabilities of the individual and specific demands of the situation in which the individual is to remember a certain set of materials. Secondly, on the basis of this framework certain methodological requirements are discussed. These considerations take into account the need for a proper analysis of the to-be-remembered (TBR) information and those cognitive functions involved in remembering the TBR information presented. Thirdly, a preoperative and postoperative memory test is presented. This was designed on the basis of the theoretical and methodological considerations mentioned. Data from one such ongoing study is presented, taking into account various measures of memory functions; immediate free recall, final free recall, final cued recall, serial recall, final final free recall, short-term memory and long-term memory capacity as measured by the Tulving & Colotla [1] lag measure, primacy, asymptote and recency effects, and semantic memory. Finally, on the basis of the experience from this memory test a few aspects of memory testing in epileptic patients are suggested for future research.

Phrenic nerve stimulation (diaphragm pacing) in chronic singultus

The authors implanted a diaphragm pacer in a 71-year-old man who had suffered for two years from continuous hiccup of unknown cause with resulting insomnia. Preoperative transcutaneous stimulation of the phrenic nerves in the neck resulted in diaphragm contractions but did not affect his hiccup. Postoperatively, the patient was free of symptoms for two weeks. Electrophrenic stimulation was then initiated and after three days his hiccup disappeared again. After a ten day period without stimulation his hiccup recurred. For ten months the patient has suffered from hiccup only in the day time during which time he has also used the pacer. With a few exceptions he has been free of symptoms and without pacing during the night, which enables him to sleep normally. The effect of phrenic nerve stimulation on hiccup could be due to interference with the abnormal activation of the phrenic nerve, whatever its cause.

Effects of intravenous progesterone infusions on the epileptic discharge frequency in women with partial epilepsy

Earlier work showed that in a group of women suffering from partial epilepsy, there is a decrease in seizure frequency during the luteal phase of the menstrual cycle. Moreover, iv progesterone infusions decrease the discharge frequency from penicillin epileptic foci in cats, when given in doses that reach plasma concentrations as observed during pregnancy. In the present study, iv progesterone infusions, reaching plasma concentrations as during the luteal phase, were given to 7 women with partial epilepsy. The inclusion criterion was that they should have more than one epileptic discharge per 5 min on an ordinary EEG . A 6 h EEG-registration was made. Blank and progesterone solutions were infused for 2 h respectively and the progesterone infusion was followed by a 2 h EEG recording period. 4 of the 7 patients showed a significant decrease in spike frequency during the infusion. In the patients not showing this effect, the plasma progesterone binding capacity was high. The 2 patients with the most marked progesterone effect had low progesterone binding capacity and no antiepileptic treatment. Somnolence during the test might also have influenced the results in 2 of the patients.

Gaucher disease--Norrbottnian type. Neurodevelopmental, neurological, and neurophysiological aspects

Neurodevelopmental, neurological and neurophysiological aspects of the Norrbottnian type of Gaucher disease (type III) were studied in 22 patients. The findings were related to clinical stages and to whether or not the patients had been splenectomized. Neurological abnormalities were mainly found in splenectomized patients who had passed the first decade of life. The EEG findings were not diagnostic for the disease, but changes were more common in the advanced stages and among splenectomized patients. The patterns of development of neurological and neurophysiological features indicate that CNS deposition of glucosylceramide increases after splenectomy, and favor a restrictive attitude towards surgery.

Benign epilepsy of children with centrotemporal EEG foci: a follow-up study in adulthood of patients initially studied as children

A group of patients with benign epilepsy of children with centrotemporal EEG foci, consisting of 40 patients reported 10 years ago, has been reinvestigated. Thirty-seven of these patients were reinvestigated. They were 26-34 years old (mean, 29 years). The observation time from the first seizure was 13-27 years (mean, 21 years). Thirty-six patients had been seizure free for 14-23 years (mean, 18.5 years); 32 patients had stopped medication 5-22 years ago (mean, 18.5 years); and 5 patients had had problems in obtaining their drivers' licenses. No epileptic discharges were found in those 24 patients who volunteered for a new EEG study. The conclusion is drawn that this common seizure disorder in children has a good long-term prognosis.