Soluble PD-L1 shows no association to relapse and overall survival in early stage non-small cell lung cancer (NSCLC).

BACKGROUND: Cancer immune evasion is critical in non-small cell lung cancer (NSCLC) and has been targeted by immunotherapy. High soluble (s)PD-L1 is associated with reduced survival and treatment failure in advanced stages. Here we evaluated the effects of sPD-L1 on T cells, relapse free survival, and overall survival in early stage NSCLC. METHODS: In vitro T cell stimulation was performed in the presence of sPD-L1 to evaluate its immunomodulatory activity. Data from The Cancer Genome Atlas (TCGA) were investigated for PD-L1 splice variants and enzymes involved in proteolytic cleavage (i.e. ADAM10). Plasma from 74 NSCLC (stage IA-IIIB), as well as an additional 73 (control cohort) patients was collected prior to curative surgery. Thereafter sPD-L1 levels from an immunosorbent assay were correlated with patient outcome. RESULTS: In vitro sPD-L1 inhibited IFN-γ production and proliferation of T cells and induced a terminal effector CD4 T cell subtype expressing CD27. Data from the TCGA demonstrated that elevated mRNA levels of ADAM10 is a negative predictor of outcome in NSCLC patients. To investigate the clinical relevance of these in vitro and TCGA findings, we quantified sPD-L1 in the plasma of early-stage NSCLC patients. In the first cohort we found significantly higher sPD-L1 levels in relapsing NSCLC patients, with a multivariate analysis revealing high sPD-L1 (>1000 pg/mL) as an independent predictor of survival. However, these findings could not be validated in two independent control cohorts. DISCUSSION: Although in vitro and TCGA data support the suppressive effect of sPD-L1 we were unable to translate this in our clinical setting. These results may be due to the small patient number and their heterogeneity as well as the lack of a standardized sPD-L1 ELISA. Our inconclusive results regarding the value of sPD-L1 in early stage NSCLC warrant assay validation and further investigation in larger (neo-)adjuvant trials.

High-throughput sequencing of archival cerebrospinal fluid specimens defines B-lymphoblastic leukemia clonal composition.

Detailed characterization of the B-lymphoblastic leukemia (B-ALL) cells which invade the central nervous system (CNS) has been limited by practical challenges. To test whether the clonal composition of the cerebrospinal fluid (CSF) reflects the primary B-ALL tissue, we applied immunoglobulin (Ig) high-throughput sequencing (HTS) of archival CSF cytospin preparations from six patients with morphologically defined CNS involvement. We discovered that most CSF clones are detectable at some timepoint in the primary tissue, but that shifting clonal abundance is prevalent across tissue sites between diagnosis and relapse. Ig HTS of CSF cytospins may improve understanding of sanctuary site dissemination in B-ALL.

Emotion-Induced Frontal Alpha Asymmetry as a Candidate Predictor of Relapse After Discontinuation of Antidepressant Medication.

BACKGROUND: One in 3 patients relapse after antidepressant discontinuation. Thus, the prevention of relapse after achieving remission is an important component in the long-term management of major depressive disorder. However, no clinical or other predictors are established. Frontal reactivity to sad mood as measured by functional magnetic resonance imaging has been reported to relate to relapse independently of antidepressant discontinuation and is an interesting candidate predictor. METHODS: Patients (n = 56) who had remitted from a depressive episode while taking antidepressants underwent electroencephalography (EEG) recording during a sad mood induction procedure prior to gradually discontinuing their medication. Relapse was assessed over a 6-month follow-up period. Thirty five healthy control participants were also tested. Current source density of the EEG power in the alpha band (8-13 Hz) was extracted and alpha asymmetry was computed by comparing the power across 2 hemispheres at frontal electrodes (F5 and F6). RESULTS: Sad mood induction was robust across all groups. Reactivity of alpha asymmetry to sad mood did not distinguish healthy control participants from patients with remitted major depressive disorder on medication. However, the 14 (25%) patients who relapsed during the follow-up period after discontinuing medication showed significantly reduced reactivity in alpha asymmetry compared with patients who remained well. This EEG signal provided predictive power (69% out-of-sample balanced accuracy and a positive predictive value of 0.75). CONCLUSIONS: A simple EEG-based measure of emotional reactivity may have potential to contribute to clinical prediction models of antidepressant discontinuation. Given the very small sample size, this finding must be interpreted with caution and requires replication in a larger study.

Understanding and predicting future relapse in depression from resting state functional connectivity and self-referential processing.

BACKGROUND: The recurrent nature of Major Depressive Disorder (MDD) asks for a better understanding of mechanisms underlying relapse. Previously, self-referential processing abnormalities have been linked to vulnerability for relapse. We investigated whether abnormalities in self-referential cognitions and functioning of associated brain-networks persist upon remission and predict relapse. METHODS: Remitted recurrent MDD patients (n = 48) and never-depressed controls (n = 23) underwent resting-state fMRI scanning at baseline and were additionally assessed for their implicit depressed self-associations and ruminative behaviour. A template-based dual regression approach was used to investigate between-group differences in default mode, cingulo-opercular and frontoparietal network resting-state functional connectivity (RSFC). Additional prediction of relapse status at 18-month follow-up was investigated within patients using both regression analyses and machine learning classifiers. RESULTS: Remitted patients showed higher rumination, but no implicit depressed self-associations or RSFC abnormalities were observed between patients and controls. Nevertheless, relapse was related to i) baseline RSFC between the ventral default mode network and the precuneus, dorsomedial frontal gyrus, and inferior occipital lobe, ii) implicit self-associations, and iii) uncontrollability of ruminative thinking, when controlled for depressive symptomatology. Moreover, preliminary machine learning classifiers demonstrated that RSFC within the investigated networks predicted relapse on an individual basis. CONCLUSIONS: Remitted MDD patients seem to be commonly characterized by abnormal rumination, but not by implicit self-associations or abnormalities in relevant brain networks. Nevertheless, relapse was predicted by self-related cognitions and default mode RSFC during remission, suggesting that variations in self-relevant processing play a role in the complex dynamics associated with the vulnerability to developing recurrent depressive episodes. CLINICAL TRIAL REGISTRATION: Netherlands Trial Register, August 18, 2015, trial number NL53205.042.15.

Machine learning based approaches for clinical and non-clinical depression recognition and depression relapse prediction using audiovisual and EEG modalities: A comprehensive review.

Mental disorders are rapidly increasing each year and have become a major challenge affecting the social and financial well-being of individuals. There is a need for phenotypic characterization of psychiatric disorders with biomarkers to provide a rich signature for Major Depressive Disorder, improving the understanding of the pathophysiological mechanisms underlying these mental disorders. This comprehensive review focuses on depression and relapse detection modalities such as self-questionnaires, audiovisuals, and EEG, highlighting noteworthy publications in the last ten years. The article concentrates on the literature that adopts machine learning by audiovisual and EEG signals. It also outlines preprocessing, feature extraction, and public datasets for depression detection. The review concludes with recommendations that will help improve the reliability of developed models and the determinism of computational intelligence-based systems in psychiatry. To the best of our knowledge, this survey is the first comprehensive review on depression and relapse prediction by self-questionnaires, audiovisual, and EEG-based approaches. The findings of this review will serve as a useful and structured starting point for researchers studying clinical and non-clinical depression recognition and relapse through machine learning-based approaches.

The Association of Non-Drug-Related Pavlovian-to-Instrumental Transfer Effect in Nucleus Accumbens With Relapse in Alcohol Dependence: A Replication.

BACKGROUND: The Pavlovian-to-instrumental transfer (PIT) paradigm measures the effects of Pavlovian conditioned cues on instrumental behavior in the laboratory. A previous study conducted by our research group observed activity in the left nucleus accumbens (NAcc) elicited by a non-drug-related PIT task across patients with alcohol dependence (AD) and healthy control subjects, and the left NAcc PIT effect differentiated patients who subsequently relapsed from those who remained abstinent. In this study, we aimed to examine whether such effects were present in a larger sample collected at a later date. METHODS: A total of 129 recently detoxified patients with AD (21 females) and 74 healthy, age- and gender-matched control subjects (12 females) performing a PIT task during functional magnetic resonance imaging were examined. After task assessments, patients were followed for 6 months. Forty-seven patients relapsed and 37 remained abstinent. RESULTS: We found a significant behavioral non-drug-related PIT effect and PIT-related activity in the NAcc across all participants. Moreover, subsequent relapsers showed stronger behavioral and left NAcc PIT effects than abstainers. These findings are consistent with our previous findings. CONCLUSIONS: Behavioral non-drug-related PIT and neural PIT correlates are associated with prospective relapse risk in AD. This study replicated previous findings and provides evidence for the clinical relevance of PIT mechanisms to treatment outcome in AD. The observed difference between prospective relapsers and abstainers in the NAcc PIT effect in our study is small overall. Future studies are needed to further elucidate the mechanisms and the possible modulators of neural PIT in relapse in AD.

Early relapse prediction after allogeneic hematopoietic stem cell transplantation for acute lymphoblastic leukemia (ALL) using lineage-specific chimerism analysis.

Relapse is a major cause of treatment failure after hematopoietic stem cell transplantation (HSCT) for acute leukemia. Here, we report a monocentric retrospective study of all HSCTs for B cell acute lymphoblastic leukemia (ALL) performed during the years 2005-2021 (n = 138, including 51 children), aiming to identify the optimal use of lineage-specific recipient-donor chimerism analysis for prediction of relapse. In adults, relapse was associated with increased recipient chimerism in CD3+ bone marrow cells sampled at least 30 days before a relapse. Relapse could be predicted with a sensitivity of 73% and a specificity of 83%. Results were similar for children but with a higher recipient chimerism cutoff. Additionally, adults that had at least one chimerism value <0.12% in CD3+ peripheral blood cells within the first 60 days after HSCT had 89% probability of being relapse-free after 2-years compared to 64%. Results were similar for children but again necessitating a higher chimerism cutoff. These results suggest that high-sensitive lineage-specific chimerism analysis can be used for (1) early ALL relapse prediction by longitudinal chimerism monitoring in CD3+ bone marrow cells and (2) relapse risk stratification by analyzing CD3+ blood cells early post-HSCT.

A Model of Normality Inspired Deep Learning Framework for Depression Relapse Prediction Using Audiovisual Data.

BACKGROUND: Depression (Major Depressive Disorder) is one of the most common mental illnesses. According to the World Health Organization, more than 300 million people in the world are affected. A first depressive episode can be solved by a spontaneous remission within 6 to 12 months. It has been shown that depression affects speech production and facial expressions. Although numerous studies are proposed in the literature for depression recognition using audiovisual cues, depression relapse using audiovisual cues has not been studied in the literature. METHOD: In this paper, we propose a deep learning-based approach for depression recognition and depression relapse prediction using audiovisual data. For more versatility and reusability, the proposed approach is based on a Model of Normality inspired framework where we define depression relapse by the closeness of the audiovisual patterns of a subject after a symptom-free period to the audiovisual patterns of depressed subjects. A model of Normality is an anomaly detection distance-based approach that computes a distance of normality between the deep audiovisual encoding of a test sample and a learned representation from audiovisual encodings of anomaly-free data. RESULTS: The proposed approach shows a very promising results with an accuracy of 87.4% and a F1-score of 82.3% for relapse/depression prediction using a Leave-One-Subject-Out training strategy on the DAIC-Woz dataset. CONCLUSION: The proposed model of normality-based framework is accurate in detecting depression and in predicting depression relapse. A prospective monitoring system is proposed for assisting depressed patients. The proposed framework is easily extensible and others modalities will be integrated in future works.

Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody.

Objective: We previously identified the independent predictors of recurrent relapse in neuromyelitis optica spectrum disorder (NMOSD) with anti-aquaporin-4 antibody (AQP4-ab) and designed a nomogram to estimate the 1- and 2-year relapse-free probability, using the Cox proportional hazard (Cox-PH) model, assuming that the risk of relapse had a linear correlation with clinical variables. However, whether the linear assumption fits real disease tragedy is unknown. We aimed to employ deep learning and machine learning to develop a novel prediction model of relapse in patients with NMOSD and compare the performance with the conventional Cox-PH model. Methods: This retrospective cohort study included patients with NMOSD with AQP4-ab in 10 study centers. In this study, 1,135 treatment episodes from 358 patients in Huashan Hospital were employed as the training set while 213 treatment episodes from 92 patients in nine other research centers as the validation set. We compared five models with added variables of gender, AQP4-ab titer, previous attack under the same therapy, EDSS score at treatment initiation, maintenance therapy, age at treatment initiation, disease duration, the phenotype of the most recent attack, and annualized relapse rate (ARR) of the most recent year by concordance index (C-index): conventional Cox-PH, random survival forest (RSF), LogisticHazard, DeepHit, and DeepSurv. Results: When including all variables, RSF outperformed the C-index in the training set (0.739), followed by DeepHit (0.737), LogisticHazard (0.722), DeepSurv (0.698), and Cox-PH (0.679) models. As for the validation set, the C-index of LogisticHazard outperformed the other models (0.718), followed by DeepHit (0.704), DeepSurv (0.698), RSF (0.685), and Cox-PH (0.651) models. Maintenance therapy was calculated to be the most important variable for relapse prediction. Conclusion: This study confirmed the superiority of deep learning to design a prediction model of relapse in patients with AQP4-ab-positive NMOSD, with the LogisticHazard model showing the best predictive power in validation.

Deep Learning-Based Multi-Omics Integration Robustly Predicts Relapse in Prostate Cancer.

Objective: Post-operative biochemical relapse (BCR) continues to occur in a significant percentage of patients with localized prostate cancer (PCa). Current stratification methods are not adequate to identify high-risk patients. The present study exploits the ability of deep learning (DL) algorithms using the H2O package to combine multi-omics data to resolve this problem. Methods: Five-omics data from 417 PCa patients from The Cancer Genome Atlas (TCGA) were used to construct the DL-based, relapse-sensitive model. Among them, 265 (63.5%) individuals experienced BCR. Five additional independent validation sets were applied to assess its predictive robustness. Bioinformatics analyses of two relapse-associated subgroups were then performed for identification of differentially expressed genes (DEGs), enriched pathway analysis, copy number analysis and immune cell infiltration analysis. Results: The DL-based model, with a significant difference (P = 6e-9) between two subgroups and good concordance index (C-index = 0.767), were proven to be robust by external validation. 1530 DEGs including 678 up- and 852 down-regulated genes were identified in the high-risk subgroup S2 compared with the low-risk subgroup S1. Enrichment analyses found five hallmark gene sets were up-regulated while 13 were down-regulated. Then, we found that DNA damage repair pathways were significantly enriched in the S2 subgroup. CNV analysis showed that 30.18% of genes were significantly up-regulated and gene amplification on chromosomes 7 and 8 was significantly elevated in the S2 subgroup. Moreover, enrichment analysis revealed that some DEGs and pathways were associated with immunity. Three tumor-infiltrating immune cell (TIIC) groups with a higher proportion in the S2 subgroup (p = 1e-05, p = 8.7e-06, p = 0.00014) and one TIIC group with a higher proportion in the S1 subgroup (P = 1.3e-06) were identified. Conclusion: We developed a novel, robust classification for understanding PCa relapse. This study validated the effectiveness of deep learning technique in prognosis prediction, and the method may benefit patients and prevent relapse by improving early detection and advancing early intervention.

Circulating Tumor DNA in Pediatric Cancer.

The measurement of circulating tumor DNA (ctDNA) has gained increasing prominence as a minimally invasive tool for the detection of cancer-specific markers in plasma. In adult cancers, ctDNA detection has shown value for disease-monitoring applications including tumor mutation profiling, risk stratification, relapse prediction, and treatment response evaluation. To date, there are ctDNA tests used as companion diagnostics for adult cancers and it is not understood why the same cannot be said about childhood cancer, despite the marked differences between adult and pediatric oncology. In this review, we discuss the current understanding of ctDNA as a disease monitoring biomarker in the context of pediatric malignancies, including the challenges associated with ctDNA detection in liquid biopsies. The data and conclusions from pediatric cancer studies of ctDNA are summarized, highlighting treatment response, disease monitoring and the detection of subclonal disease as applications of ctDNA. While the data from retrospective studies highlight the potential of ctDNA, large clinical trials are required for ctDNA analysis for routine clinical use in pediatric cancers. We outline the requirements for the standardization of ctDNA detection in pediatric cancers, including sample handling and reproducibility of results. With better understanding of the advantages and limitations of ctDNA and improved detection methods, ctDNA analysis may become the standard of care for patient monitoring in childhood cancers.

Multiple Sclerosis Severity Score (MSSS) improves the accuracy of individualized prediction in MS.

BACKGROUND: The MSBase prediction model of treatment response leverages multiple demographic and clinical characteristics to estimate hazards of relapses, confirmed disability accumulation (CDA), and confirmed disability improvement (CDI). The model did not include Multiple Sclerosis Severity Score (MSSS), a disease duration-adjusted ranked score of disability. OBJECTIVE: To incorporate MSSS into the MSBase prediction model and compare model accuracy with and without MSSS. METHODS: The associations between MSSS and relapse, CDA, and CDI were evaluated with marginal proportional hazards models adjusted for three principal components representative of patients' demographic and clinical characteristics. The model fit with and without MSSS was assessed with penalized r2 and Harrell C. RESULTS: A total of 5866 MS patients were started on disease-modifying therapy during prospective follow-up (age 38.4 ± 10.6 years; 72% female; disease duration 8.5 ± 7.7 years). Including MSSS into the model improved the accuracy of individual prediction of relapses by 31%, of CDA by 23%, and of CDI by 24% (Harrell C) and increased the amount of variance explained for relapses by 49%, for CDI by 11%, and for CDA by 10% as compared with the original model. CONCLUSION: Addition of a single, readily available metric, MSSS, to the comprehensive MSBase prediction model considerably improved the individual accuracy of prognostics in MS.

Risk Factors and Nomogram for Predicting Relapse Risk in Pediatric Neuromyelitis Optica Spectrum Disorders.

BACKGROUND: Neuromyelitis optica spectrum disorders (NMOSDs) are attack-relapsing autoimmune inflammatory diseases of the central nervous system, which are characterized by the presence of serological aquaporin-4 (AQP4) antibody. However, this disorder is uncommon in children, and AQP4 antibody was often found to be seronegative. However, some pediatric patients diagnosed with NMOSDs were tested to be positive for myelin oligodendrocyte glycoprotein (MOG) antibody. The previous investigations of pediatric NMOSDs were usually focused on the clinical presentation, treatment responses, and long-term prognoses, but little is known about the risk factors predicting NMOSD relapse attacks in a shorter time, especially, for Chinese children. METHODS: We retrospectively identified 64 Chinese pediatric patients, including 39 positive for AQP4 antibody, 12 positive for MOG antibody, and the rest negative for AQP4 and MOG antibodies. Independent risk factors predicting relapse in 1-year follow-up were extracted by multivariate regression analysis to establish a risk score model, its performance evaluation was analyzed using receiver operating characteristic (ROC) curve, and the independent risk factors related to relapse manifestation were also explored through multivariate logistic analysis. A nomogram was generated to assess relapse attacks in 1-year follow-up. Thirty-five patients from 3 other centers formed an external cohort to validate this nomogram. RESULTS: Four independent relapsed factors included discharge Expanded Disability Status Scale (EDSS) (p = 0.017), mixed-lesion onset (p = 0.010), counts (≧1) of concomitant autoantibodies (p = 0.015), and maintenance therapy (tapering steroid with mycophenolate mofetil (MMF), p = 0.009; tapering steroid with acetazolamide (AZA), p = 0.045; and tapering steroid only, p = 0.025). The risk score modeled with these four factors was correlated with the likelihood of relapse in the primary cohort (AUC of 0.912) and the validation cohort (AUC of 0.846). Also, our nomogram exhibited accurate relapse estimate in the primary cohort, the validation cohort, and the whole cohort, but also in the cohorts with positive/negative AQP4 antibody, and noticeably, it performed predictive risk improvement better than other factors in the concordance index (C-index), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). CONCLUSIONS: The risk score and nomogram could facilitate accurate prognosis of relapse risk in 1-year follow-up for pediatric NMOSDs and help clinicians provide personalized treatment to decrease the chance of relapse.

The temporal dynamics of sleep disturbance and psychopathology in psychosis: a digital sampling study.

BACKGROUND: Sleep disruption is a common precursor to deterioration and relapse in people living with psychotic disorders. Understanding the temporal relationship between sleep and psychopathology is important for identifying and developing interventions which target key variables that contribute to relapse. METHODS: We used a purpose-built digital platform to sample self-reported sleep and psychopathology variables over 1 year, in 36 individuals with schizophrenia. Once-daily measures of sleep duration and sleep quality, and fluctuations in psychopathology (positive and negative affect, cognition and psychotic symptoms) were captured. We examined the temporal relationship between these variables using the Differential Time-Varying Effect (DTVEM) hybrid exploratory-confirmatory model. RESULTS: Poorer sleep quality and shorter sleep duration maximally predicted deterioration in psychosis symptoms over the subsequent 1-8 and 1-12 days, respectively. These relationships were also mediated by negative affect and cognitive symptoms. Psychopathology variables also predicted sleep quality, but not sleep duration, and the effect sizes were smaller and of shorter lag duration. CONCLUSIONS: Reduced sleep duration and poorer sleep quality anticipate the exacerbation of psychotic symptoms by approximately 1-2 weeks, and negative affect and cognitive symptoms mediate this relationship. We also observed a reciprocal relationship that was of shorter duration and smaller magnitude. Sleep disturbance may play a causal role in symptom exacerbation and relapse, and represents an important and tractable target for intervention. It warrants greater attention as an early warning sign of deterioration, and low-burden, user-friendly digital tools may play a role in its early detection.

Early assessment of circulating tumor DNA after curative-intent resection predicts tumor recurrence in early-stage and locally advanced non-small-cell lung cancer.

Circulating tumor DNA (ctDNA) has demonstrated great potential as a noninvasive biomarker to assess minimal residual disease (MRD) and profile tumor genotypes in patients with non-small-cell lung cancer (NSCLC). However, little is known about its dynamics during and after tumor resection, or its potential for predicting clinical outcomes. Here, we applied a targeted-capture high-throughput sequencing approach to profile ctDNA at various disease milestones and assessed its predictive value in patients with early-stage and locally advanced NSCLC. We prospectively enrolled 33 consecutive patients with stage IA to IIIB NSCLC undergoing curative-intent tumor resection (median follow-up: 26.2 months). From 21 patients, we serially collected 96 plasma samples before surgery, during surgery, 1-2 weeks postsurgery, and during follow-up. Deep next-generation sequencing using unique molecular identifiers was performed to identify and quantify tumor-specific mutations in ctDNA. Twelve patients (57%) had detectable mutations in ctDNA before tumor resection. Both ctDNA detection rates and ctDNA concentrations were significantly higher in plasma obtained during surgery compared with presurgical specimens (57% versus 19% ctDNA detection rate, and 12.47 versus 6.64 ng·mL-1 , respectively). Four patients (19%) remained ctDNA-positive at 1-2 weeks after surgery, with all of them (100%) experiencing disease progression at later time points. In contrast, only 4 out of 12 ctDNA-negative patients (33%) after surgery experienced relapse during follow-up. Positive ctDNA in early postoperative plasma samples was associated with shorter progression-free survival (P = 0.013) and overall survival (P = 0.004). Our findings suggest that, in early-stage and locally advanced NSCLC, intraoperative plasma sampling results in high ctDNA detection rates and that ctDNA positivity early after resection identifies patients at risk for relapse.

Resting Hypoconnectivity of Theoretically Defined Addiction Networks during Early Abstinence Predicts Subsequent Relapse in Alcohol Use Disorder.

Theoretical models of addiction suggest that alterations in addiction domains including incentive salience, negative emotionality, and executive control lead to relapse in alcohol use disorder (AUD). To determine whether the functional organization of neural networks underlying these domains predict subsequent relapse, we generated theoretically defined addiction networks. We collected resting functional magnetic resonance imaging data from 45 individuals with AUD during early abstinence (number of days abstinent M = 25.40, SD = 16.51) and calculated the degree of resting-state functional connectivity (RSFC) within these networks. Regression analyses determined whether the RSFC strength in domain-defined addiction networks measured during early abstinence predicted subsequent relapse (dichotomous or continuous relapse metrics). RSFC within each addiction network measured during early abstinence was significantly lower in those that relapsed (vs. abstained) and predicted subsequent time to relapse. Lower incentive salience RSFC during early abstinence increased the odds of relapsing. Neither RSFC in a control network nor clinical self-report measures predicted relapse. The association between low incentive salience RSFC and faster relapse highlights the need to design timely interventions that enhance RSFC in AUD individuals at risk of relapsing faster.

Predicting Relapse in Patients With Triple Negative Breast Cancer (TNBC) Using a Deep-Learning Approach.

The abundance and/or location of tumor infiltrating lymphocytes (TILs), especially CD8+ T cells, in solid tumors can serve as a prognostic indicator in various types of cancer. However, it is often difficult to select an appropriate threshold value in order to stratify patients into well-defined risk groups. It is also important to select appropriate tumor regions to quantify the abundance of TILs. On the other hand, machine-learning approaches can stratify patients in an unbiased and automatic fashion. Based on immunofluorescence (IF) images of CD8+ T lymphocytes and cancer cells, we develop a machine-learning approach which can predict the risk of relapse for patients with Triple Negative Breast Cancer (TNBC). Tumor-section images from 9 patients with poor outcome and 15 patients with good outcome were used as a training set. Tumor-section images of 29 patients in an independent cohort were used to test the predictive power of our algorithm. In the test cohort, 6 (out of 29) patients who belong to the poor-outcome group were all correctly identified by our algorithm; for the 23 (out of 29) patients who belong to the good-outcome group, 17 were correctly predicted with some evidence that improvement is possible if other measures, such as the grade of tumors, are factored in. Our approach does not involve arbitrarily defined metrics and can be applied to other types of cancer in which the abundance/location of CD8+ T lymphocytes/other types of cells is an indicator of prognosis.

Differential response to cytotoxic therapy explains treatment dynamics of acute myeloid leukaemia patients: insights from a mathematical modelling approach.

Disease response and durability of remission are very heterogeneous in patients with acute myeloid leukaemia (AML). There is increasing evidence that the individual risk of early relapse can be predicted based on the initial treatment response. However, it is unclear how such a correlation is linked to functional aspects of AML progression and treatment. We suggest a mathematical model in which leukaemia-initiating cells and normal/healthy haematopoietic stem and progenitor cells reversibly change between an active state characterized by proliferation and chemosensitivity and a quiescent state, in which the cells do not divide, but are also insensitive to chemotherapy. Applying this model to 275 molecular time courses of nucleophosmin 1-mutated patients, we conclude that the differential chemosensitivity of the leukaemia-initiating cells together with the cells' intrinsic proliferative capacity is sufficient to reproduce both, early relapse as well as long-lasting remission. We can, furthermore, show that the model parameters associated with individual chemosensitivity and proliferative advantage of the leukaemic cells are closely linked to the patients' time to relapse, while a reliable prediction based on early response only is not possible based on the currently available data. Although we demonstrate with our approach, that the complete response data is sufficient to quantify the aggressiveness of the disease, further investigations are necessary to study how an intensive early sampling strategy may prospectively improve risk assessment and help to optimize individual treatments.

Utilizing Machine Learning on Internet Search Activity to Support the Diagnostic Process and Relapse Detection in Young Individuals With Early Psychosis: Feasibility Study.

BACKGROUND: Psychiatry is nearly entirely reliant on patient self-reporting, and there are few objective and reliable tests or sources of collateral information available to help diagnostic and assessment procedures. Technology offers opportunities to collect objective digital data to complement patient experience and facilitate more informed treatment decisions. OBJECTIVE: We aimed to develop computational algorithms based on internet search activity designed to support diagnostic procedures and relapse identification in individuals with schizophrenia spectrum disorders. METHODS: We extracted 32,733 time-stamped search queries across 42 participants with schizophrenia spectrum disorders and 74 healthy volunteers between the ages of 15 and 35 (mean 24.4 years, 44.0% male), and built machine-learning diagnostic and relapse classifiers utilizing the timing, frequency, and content of online search activity. RESULTS: Classifiers predicted a diagnosis of schizophrenia spectrum disorders with an area under the curve value of 0.74 and predicted a psychotic relapse in individuals with schizophrenia spectrum disorders with an area under the curve of 0.71. Compared with healthy participants, those with schizophrenia spectrum disorders made fewer searches and their searches consisted of fewer words. Prior to a relapse hospitalization, participants with schizophrenia spectrum disorders were more likely to use words related to hearing, perception, and anger, and were less likely to use words related to health. CONCLUSIONS: Online search activity holds promise for gathering objective and easily accessed indicators of psychiatric symptoms. Utilizing search activity as collateral behavioral health information would represent a major advancement in efforts to capitalize on objective digital data to improve mental health monitoring.