The PI3K/AKT/mTOR signaling pathway is commonly dysregulated in cancer. The goal of this thesis project was to assess the hypothesis of a strong correlation between PI3K/AKT/mTOR pathway activity and the response to targeted therapies, by using a protein quantitation technique called immuno-matrix assisted laser desorption/ionization (iMALDI). The use of iMALDI as a clinical tool was demonstrated by automating an established iMALDI assay for quantifying plasma renin activity. The results from the automated method gave high correlation coefficients of ≥0.98 with a clinical LC-MS/MS method and could be performed significantly faster than with manual sample preparation. The 7.5-fold faster analysis compared to LC-MS/MS, reduction in human error, and higher throughput, demonstrated the suitability of this assay for clinical use. The automated iMALDI platform was then adapted for use with cancer cell lines and tissue analysis, targeting the kinases AKT1 and AKT2 as surrogate proteins for signaling pathway activity. Using minute amounts (10 µg/capture), AKT1 and AKT2 expression and phosphorylation stoichiometry (PS) were successfully quantified via their C-terminal tryptic peptides, which encompassed key phosphorylation sites. After assay optimization, the assays were analytically validated for linear range, accuracy, and interferences. In addition, PS cut-off values based on measurement errors were established for confident PS quantitation. The functionality of the assay was demonstrated with cell lines, and flash-frozen and FFPE tissue lysates, with, on average, lower AKT1/AKT2 measurements obtained from FFPE samples. The developed assays were sensitive and precise enough to detect differences between matched normal and adjacent tumor tissues. To answer the hypothesis, patient-derived xenograft (PDX) mouse-model tumors treated with Herceptin, Everolimus, a combination of both (E+H), or with no treatment, were assessed for molecular patterns linked to tumor response. One mouse from the E+H group showed a partial response, with elevated total and phosphorylated AKT1/AKT2. Unfortunately, overlapping values between treatment groups were obtained in this study, and the large within-group spread and the low number of biological replicates made it difficult to confirm a definite correlation between PI3K/AKT/mTOR pathway activity and response to treatment. A follow-up study with additional protein targets, a larger number of samples, and serial biopsies will be required to determine if there is, in fact, a correlation between PI3K/AKT/mTOR pathway activity and response to treatment. / Graduate / 2019-10-05

The PI3-kinase/AKT/mTOR pathway plays a central role in cancer signaling. While p110α is the catalytic α-subunit of PI3-kinase and a major drug target, PTEN is the main negative regulator of the PI3-kinase/AKT/mTOR pathway. PTEN and p110α protein expression in tumors is commonly analyzed by immunohistochemistry, which suffers from poor multiplexing capacity, poor standardization, and antibody cross-reactivity, and which provides only semi-quantitative data. Here, we present an automated, and standardized immuno-matrix-assisted laser desorption/ionization mass spectrometry (iMALDI) assay that allows precise and multiplexed quantitation of PTEN and p110α concentrations, without the limitations of immunohistochemistry. IMALDI, which combines immuno-enrichment with analysis using a benchtop MALDI-Time-of-Flight (TOF) mass spectrometer, is an especially well-suited method for translating mass-spectrometry based assays into the clinical lab. We systematically optimized the iMALDI workflow regarding sensitivity, robustness, and throughput while developing highly flexible automation protocols using a Bravo 96LT liquid handling robot. We further developed custom R scripts to improve data visualization and analysis. One hour digestion using a protein to trypsin ratio of 1:2, followed by direct immuno-enrichment for 1 h yielded high and consistent peptide recoveries. We demonstrated that the PTEN and p110α iMALDI assays can be multiplexed using both simultaneous and sequential enrichment, reducing the amount of required sample material as well as simplifying the workflow. The PTEN+p110α iMALDI assay was validated and demonstrated high accuracy for both target proteins (90-112% recovery of known spiked-in concentrations) as well as high precision and 5-day reproducibility (overall CVs of 9%) across the linear range of the assay (0.6 to 20 fmol). Lower limits of quantitation below 1 fmol were achieved. Endogenous PTEN and p110α were quantified in cell lines as well as fresh-frozen tumor tissue samples. A novel two-point internal calibration strategy (2-PIC) was developed, based on spiking two peptide isotopologues into the sample as internal standards, avoiding the need for an external calibration. We quantified endogenous PTEN in a Colo-205 cell line using the PTEN iMALDI assay, as well an orthogonal PTEN immuno-multiple reaction monitoring (immuno-MRM) method to demonstrate this technique. Excellent agreement was shown between both calibration approaches (residual standard deviation between 2-PIC and external calibration of 1.6-5.8%), as well as high correlation between PTEN iMALDI and PTEN immuno-MRM (R²= 0.9966) and good agreement between quantified amounts (0.48±0.01 and 0.29±0.02 fmol/µg of total protein). Finally, we analysed a set of patient samples from a AKT inhibitor AZD5363 drug trial using a multi-site workflow combining the developed PTEN+p110α assay with established AKT1+AKT2 iMALDI assays and untargeted proteomics. We demonstrated how the combination of targeted and untargeted proteomics approaches may be used to gain novel insights into the tumor biology of patient tissue samples. Further, we showed that the PTEN iMALDI assay has good correlation with a comparable immunohistochemistry method (R²=0.86), and that our assays can be further multiplexed, reducing the required amount sample material. Thus, we showed that iMALDI is promising tool for biomarker quantitation. / Graduate / 2022-08-12

This work aims at developing a novel machine learning method to investigate heterogeneity in neurodevelopmental disorders, with a focus on autism spectrum conditions (ASCs). In ASCs, heterogeneity is shown at several levels of analysis, e.g., genetic, behavioral, throughout developmental trajectories, which hinders the development of effective treatments and the identification of biological pathways involved in gene-cognition-behavior links. ASC diagnosis comes from behavioral observations, which determine the cohort composition of studies in every scientific field (e.g., psychology, neuroscience, genetics). Thus, uncovering behavioral subtypes can provide stratified ASC cohorts that are more representative of the true population. Ideally, behavioral stratification can (1) help to revise and shorten the diagnostic process highlighting the characteristics that best identify heterogeneity; (2) help to develop personalized treatments based on their effectiveness for subgroups of subjects; (3) investigate how the longitudinal course of the condition might differ (e.g., divergent/convergent developmental trajectories); (4) contribute to the identification of genetic variants that may be overlooked in case-control studies; and (5) identify possible disrupted neuronal activity in the brain (e.g., excitatory/inhibitory mechanisms). The characterization of the temporal aspects of heterogeneous manifestations based on their multi-dimensional features is thus the key to identify the etiology of such disorders and establish personalized treatments. Features include trajectories described by a multi-modal combination of electronic health records (EHRs), cognitive functioning and adaptive behavior indicators. This thesis contributes in particular to a data-driven discovery of clinical and behavioral trajectories of individuals with complex disorders and ASCs. Machine learning techniques, such as deep learning and word embedding, that proved successful for e.g., natural language processing and image classification, are gaining ground in healthcare research for precision medicine. Here, we leverage these methods to investigate the feasibility of learning data-driven pathways that have been difficult to identify in the clinical practice to help disentangle the complexity of conditions whose etiology is still unknown. In Chapter 1, we present a new computational method, based on deep learning, to stratify patients with complex disorders; we demonstrate the method on multiple myeloma, Alzheimer’s disease, and Parkinson’s disease, among others. We use clinical records from a heterogeneous patient cohort (i.e., multiple disease dataset) of 1.6M temporally-ordered EHR sequences from the Mount Sinai health system’s data warehouse to learn unsupervised patient representations. These representations are then leveraged to identify subgroups within complex condition cohorts via hierarchical clustering. We investigate the enrichment of terms that code for comorbidities, medications, laboratory tests and procedures, to clinically validate our results. A data analysis protocol is developed in Chapter 2 that produces behavioral embeddings from observational measurements to represent subjects with ASCs in a latent space able to capture multiple levels of assessment (i.e., multiple tests) and the temporal pattern of behavioral-cognitive profiles. The computational framework includes clustering algorithms and state-of-the-art word and text representation methods originally developed for natural language processing. The aim is to detect subgroups within ASC cohorts towards the identification of possible subtypes based on behavioral, cognitive, and functioning aspects. The protocol is applied to ASC behavioral data of 204 children and adolescents referred to the Laboratory of Observation Diagnosis and Education (ODFLab) at the University of Trento. In Chapter 3 we develop a case study for ASCs. From the learned representations of Chapter 1, we select 1,439 individuals with ASCs and investigate whether such representations generalize well to any disorder. Specifically, we identify three subgroups within individuals with ASCs that are further clinically validated to detect clinical profiles based on different term enrichment that can inform comorbidities, therapeutic treatments, medication side effects, and screening policies. This work has been developed in partnership with ODFLab (University of Trento) and the Predictive Models for Biomedicine and Environment unit at FBK. The study reported in Chapter 1 has been conducted at the Institute for Next Generation Healthcare, Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai (NY).

This thesis develops an Artificial Intelligence (AI) approach intended for accurate patient stratification and precise diagnostics/prognostics in clinical and preclinical applications. The rapid advance in high throughput technologies and bioinformatics tools is still far from linking precisely the genome-phenotype interactions with the biological mechanisms that underlie pathophysiological conditions. In practice, the incomplete knowledge on individual heterogeneity in complex diseases keeps forcing clinicians to settle for surrogate endpoints and therapies based on a generic one-size-fits-all approach. The working hypothesis is that AI can add new tools to elaborate and integrate together in new features or structures the rich information now available from high-throughput omics and bioimaging data, and that such re- structured information can be applied through predictive models for the precision medicine paradigm, thus favoring the creation of safer tailored treatments for specific patient subgroups. The computational techniques in this thesis are based on the combination of dimensionality reduction methods with Deep Learning (DL) architectures to learn meaningful transformations between the input and the predictive endpoint space. The rationale is that such transformations can introduce intermediate spaces offering more succinct representations, where data from different sources are summarized. The research goal was attacked at increasing levels of complexity, starting from single input modalities (omics and bioimaging of different types and scales), to their multimodal integration. The approach also deals with the key challenges for machine learning (ML) on biomedical data, i.e. reproducibility, stability, and interpretability of the models. Along this path, the thesis contribution is thus the development of a set of specialized AI models and a core framework of three tools of general applicability: i. A Data Analysis Plan (DAP) for model selection and evaluation of classifiers on omics and imaging data to avoid selection bias. ii. The histolab Python package that standardizes the reproducible pre-processing of Whole Slide Images (WSIs), supported by automated testing and easily integrable in DL pipelines for Digital Pathology. iii. Unsupervised and dimensionality reduction techniques based on the UMAP and TDA frameworks for patient subtyping. The framework has been successfully applied on public as well as original data in precision oncology and predictive toxicology. In the clinical setting, this thesis has developed1: 1. (DAPPER) A deep learning framework for evaluation of predictive models in Digital Pathology that controls for selection bias through properly designed data partitioning schemes. 2. (RADLER) A unified deep learning framework that combines radiomics fea- tures and imaging on PET-CT images for prognostic biomarker development in head and neck squamous cell carcinoma. The mixed deep learning/radiomics approach is more accurate than using only one feature type. 3. An ML framework for automated quantification tumor infiltrating lymphocytes (TILs) in onco-immunology, validated on original pathology Neuroblastoma data of the Bambino Gesu’ Children’s Hospital, with high agreement with trained pathologists. The network-based INF pipeline, which applies machine learning models over the combination of multiple omics layers, also providing compact biomarker signatures. INF was validated on three TCGA oncogenomic datasets. In the preclinical setting the framework has been applied for: 1. Deep and machine learning algorithms to predict DILI status from gene expression (GE) data derived from cancer cell lines on the CMap Drug Safety dataset. 2. (ML4TOX) Deep Learning and Support Vector Machine models to predict potential endocrine disruption of environmental chemicals on the CERAPP dataset. 3. (PathologAI) A deep learning pipeline combining generative and convolutional models for preclinical digital pathology. Developed as an internal project within the FDA/NCTR AIRForce initiative and applied to predict necrosis on images from the TG-GATEs project, PathologAI aims to improve accuracy and reduce labor in the identification of lesions in predictive toxicology. Furthermore, GE microarray data were integrated with histology features in a unified multi-modal scheme combining imaging and omics data. The solutions were developed in collaboration with domain experts and considered promising for application.

Early detection reduces mortality in lung cancer, but it is also considered as a challenge for oncologists and for healthcare systems. In addition, screening modalities like CT-scans come with undesired effects, many suspected patients are wrongly diagnosed with lung cancer. This thesis contributes to solve the challenge of early lung cancer detection by utilizing unique data consisting of self-reported symptoms. The proposed method is a predictive machine learning algorithm based on natural language processing, which handles the data as an unstructured data set. A replication of a previous study where a prediction model based on a conventional multivariate machine learning using the same data is done and presented, for comparison. After evaluation, validation and interpretation, a set of variables were highlighted as early predictors of lung cancer. The performance of the proposed approach managed to match the performance of the conventional approach. This promising result opens for further development where such an approach can be used in clinical decision support systems. Future work could then involve other modalities, in a multimodal machine learning approach. / Tidig lungcancerdiagnostisering kan öka chanserna för överlevnad hos lungcancerpatienter, men att upptäcka lungcancer i ett tidigt stadie är en av de större utmaningarna för onkologer och sjukvården. Idag undersöks patienter med riskfaktorer baserat på rökning och ålder, dessa undersökningar sker med hjälp av bland annat medicinskt avbildningssystem, då oftast CT-bilder, vilket medför felaktiga och kostsamma diagnoser. Detta arbete föreslår en maskininlärninig algoritm baserad på Natural language processing, som genom analys och bearbetning av ostrukturerade data, av patienternas egna anamneser, kan prediktera lungcancer. Arbetet har genomfört en jämförelse med en konventionell maskininlärning algoritm baserat på en replikering av ett annat studie där samma data behandlades som strukturerad. Den föreslagna metoden har visat ett likartat resultat samt prestanda, och har identifierat riskfaktorer samt symptom för lungcancer. Detta arbete öppnar upp för en utveckling mot ett kliniskt användande i form av beslutsstödsystem, som även kan hantera elektriska hälsojournaler. Andra arbeten kan vidareutveckla metoden för att hantera andra varianter av data, så som medicinska bilder och biomarkörer, och genom det förbättra prestandan.

Bladder carcinoma (BCa) is among the most common carcinomas in the Western world. Despite the availability of effective therapies, there is currently an urgent need to develop a stratification method, which would enable the accurate identification of patients responsive to therapy. In the theoretical part of my diploma project I describe the heterogeneity of BCa and the currently applied immunotherapeutic approaches. I specifically focused on the Bacillus Calmette-Guérin (BCG) vaccine instillation. For decades another use of BCG has been a prophylactic vaccination against tuberculosis (TB) infection. BCG serves as a model treatment because it is highly efficient when prescribed to the responsive patient. However, an effective stratification is yet to be developed for BCa and latent tuberculosis infection (LTBI) diagnosis and/or monitoring. In the experimental part of my project, I developed and tested a 10-parameter panel for T cell- specific activation test (TAT) applicable for a stratification of BCa patients as well as for the detection of LTBI. I tested the panel on positive controls using flow cytometry (FCM) method because it allows for detection and measurement of dozens of markers at a single cell level. It is easily applicable to available urine and blood samples obtained from BCa...

No description available.