Analisi fosfo-proteomica di campioni di tumore: il problema della stratificazione dei pazienti ed eterogeneità intra-tumorale nell'era della terapia personalizzata

Nella presente tesi indaghiamo la potenzialità di LCM e Reverse Phase Protein microarray negli studi clinici. Si analizza la possibilità di creare una bio banca con line cellular primarie, al fine di conseguire drug test di sensibilità prima di decidere il trattamento da somministrare ai singoli pazienti. Sono stati ottenuti profili proteomici da biopsie pre e post terapia. I risultati dimostrano che questa piattaforma mostra il meccanismo di resistenza acquisito durante la terapia biologica. Questo ci ha portato ad analizzare una possibile stratificazione per pazienti con mCRC . I dati hanno rivelato distinti pathway di attivazione tra metastasi resecabile e non resecabili. I risultati mostrano inoltre due potenziali bersagli farmacologici. Ma la valutazione dell'intero tumore tramite singole biopsie sembra essere un problema a causa dell’eterogeneità intratumorale a livello genomico. Abbiamo indagato questo problema a livello dell'architettura del segnale in campioni di mCRC e ccRCC . I risultati indicano una somiglianza complessiva nei profili proteomici all'interno dello stesso tumore. Considerando che una singola biopsia è rappresentativa di un intera lesione , abbiamo studiato la possibilità di creare linee di cellule primarie, per valutare il profilo molecolare di ogni paziente. Fino ad oggi non c'era un protocollo per creare linee cellulari immortalizzate senza alcuna variazione genetica . abbiamo cosiderato, però, l'approccio innovativo delle CRCs. Ad oggi , non è ancora chiaro se tali cellule mimino il profilo dei tessuti oppure I passaggi in vitro modifichino i loro pathways . Sulla base di un modello di topo , i nostri dati mostrano un profilo di proteomica simile tra le linee di cellule e tessuti di topo LCM. In conclusione, i nostri dati dimostrano l'utilità della piattaforma LCM / RPPA nella sperimentazione clinica e la possibilità di creare una bio - banca di linee cellulari primarie, per migliorare la decisione del trattamento.

Machine learning tools for protein annotation: the cases of transmembrane β-barrel and myristoylated proteins

Biology is now a “Big Data Science” thanks to technological advancements allowing the characterization of the whole macromolecular content of a cell or a collection of cells. This opens interesting perspectives, but only a small portion of this data may be experimentally characterized. From this derives the demand of accurate and efficient computational tools for automatic annotation of biological molecules. This is even more true when dealing with membrane proteins, on which my research project is focused leading to the development of two machine learning-based methods: BetAware-Deep and SVMyr. BetAware-Deep is a tool for the detection and topology prediction of transmembrane beta-barrel proteins found in Gram-negative bacteria. These proteins are involved in many biological processes and primary candidates as drug targets. BetAware-Deep exploits the combination of a deep learning framework (bidirectional long short-term memory) and a probabilistic graphical model (grammatical-restrained hidden conditional random field). Moreover, it introduced a modified formulation of the hydrophobic moment, designed to include the evolutionary information. BetAware-Deep outperformed all the available methods in topology prediction and reported high scores in the detection task. Glycine myristoylation in Eukaryotes is the binding of a myristic acid on an N-terminal glycine. SVMyr is a fast method based on support vector machines designed to predict this modification in dataset of proteomic scale. It uses as input octapeptides and exploits computational scores derived from experimental examples and mean physicochemical features. SVMyr outperformed all the available methods for co-translational myristoylation prediction. In addition, it allows (as a unique feature) the prediction of post-translational myristoylation. Both the tools here described are designed having in mind best practices for the development of machine learning-based tools outlined by the bioinformatics community. Moreover, they are made available via user-friendly web servers. All this make them valuable tools for filling the gap between sequential and annotated data.

Policitemia Vera: scoperta di saggi di biopsia liquida finalizzata a identificare i biomarcatori correlati al rischio trombotico

La Policitemia Vera (PV) è una neoplasia mieloproliferativa con un aumentato rischio di trombosi e di progressione verso la Mielofibrosi. L'infiammazione cronica è comunemente osservata nelle neoplasie mieloproliferative, compresa la PV. La rete infiammatoria, tra le varie componenti, comprende le vescicole extracellulari (EVs), che svolgono un ruolo nella comunicazione cellula-cellula. Inoltre, le componenti microbiche circolanti sono state recentemente indicate come potenziali modificatori dell'infiammazione, della coagulazione e dell’emopoiesi in generale. Qui abbiamo studiato il DNA microbico delle EVs circolanti attraverso. Sangue periferico e feci sono stati raccolti da pazienti con PV (n=38) e da donatori sani (n=30). Le EVs circolanti derivate da megacariociti (MK) e piastrine (PLT) sono state analizzate mediante citometria a flusso. Dopo l'estrazione del DNA microbico dalle feci e dalle EV isolate, è stata sequenziata la regione V3-V4 del 16S rDNA. La percentuale di EVs di MK era ridotta nei pazienti con PV rispetto ai donatori sani. Al contrario, la proporzione di EVs di PLT era aumentata. La PV è stata associata anche a una firma del DNA microbico delle EVs isolate con una maggiore diversità e una composizione microbica distinta rispetto alla controparte sana. Nei pazienti con PV c’è una maggiore proporzione di EVs associate al lipopolisaccaride. Il profilo del microbioma intestinale non differiva tra PV e doantori. Inoltre, l'aumento della proporzione di EVs di MK e la riduzione di EVs di piastre identificavano i pazienti con pregressa trombosi. Le EVs dei pazienti con trombosi erano impoverite di DNA di Staphylococcus ma arricchite di DNA di Actinobacteria e Anaerococcus. Inoltre, questi pazienti avevano livelli più bassi di EVs associate al lipopolisaccaride. I pazienti con fibrosi midollare avevano una maggiore proporzione di PE-EV ed erano arricchite in DNA di Collinsella e Flavobacterium. Questi dati possono contribuire a perfezionare la prognosi della PV e a identificare nuovi bersagli farmacologici.

SMO inhibitor specifically targets the Hedgehog Pathway and reverts the drug-resistance of Leukemic Stem Cells

Abnormal Hedgehog signaling is associated with human malignancies. Smo, a key player of that signaling, is the most suitable target to inhibit this pathway. To this aim several molecules, antagonists of Smo, have been synthesized, and some of them have started the phase I in clinical trials. Our hospital participated to one of these studies which investigated the oral administration of a new selective inhibitor of Smo (SMOi). To evaluate ex vivo SMOi efficacy and to identify new potential clinical biomarkers of responsiveness, we separated bone marrow CD34+ cells from 5 acute myeloid leukemia (AML), 1 myelofibrosis (MF), 2 blastic phases chronic myeloid leukemia (CML) patients treated with SMOi by immunomagnetic separation, and we analysed their gene expression profile using Affimetrix HG-U133 Plus 2.0 platform. This analysis, showed differential expression after 28 days start of therapy (p-value ≤ 0.05) of 1,197 genes in CML patients and 589 genes in AML patients. This differential expression is related to Hedgehog pathway with a p-value = 0.003 in CML patients and with a p-value = 0.0002 in AML patients, suggesting that SMOi targets specifically this pathway. Among the genes differentially expressed we observed strong up-regulation of Gas1 and Kif27 genes, which may work as biomarkers of responsiveness of SMOi treatment in CML CD34+ cells whereas Hedgehog target genes (such as Smo, Gli1, Gli2, Gli3), Bcl2 and Abca2 were down-regulated, in both AML and CML CD34+ cells. It has been reported that Bcl-2 expression could be correlated with cancer therapy resistance and that Hedgehog signaling modulate ATP-binding (ABC) cassette transporters, whose expression has been correlated with chemoresistance. Moreover we confirmed that in vitro SMOi treatment targets Hedgehog pathway, down-regulate ABC transporters, Abcg2 and Abcb1 genes, and in combination with tyrosine kinase inhibitors (TKIs) could revert the chemoresistance mechanism in K562 TKIs-resistant cell line.

E-cadherin and Choline Kinase: two challenging drug discovery targets

The data presented in this thesis was generated using molecular biology, protein chemistry and X-ray crystallography techniques. However, while the methodologies employed are essentially the same, the research work presented here refers to two different proteins, which are part of different research projects in the laboratory. For this reason, the content of this thesis is divided in two independent parts, each provided with an introduction and a general overview of the research topic and state-ofthe- art, a materials and methods section discussing the techniques used and the protocols followed, and a section where the results are presented and discussed in detail. The first half of the thesis deals with the structural characterization of the complex between human E-cadherin and three different small molecule potential inhibitors identified via a fragment-based drug discovery (FBDD) screening campaign that was conducted using a library of commercially available small fluorinated chemical fragments. For this screening phase, we used 19F-NMR as readout. The NMR experiments were done by our collaborator Dr. Marina Veronesi at the D3 PharmaChemistry division of the Italian Institute of Technology (IIT) in Genova (Italy). Functional cell adhesion assays to validate the inhibitory effects of the fragments thus identified were carried out in collaboration with Prof. Frédéric André at the University of Marseille (France). The second half of the thesis describes the structural characterization of Plasmodium falciparum Choline Kinase (PfChoK), an important pharmaceutical target in the fight against malaria, as well as the biochemical characterization of a library of potential inhibitors of PfChoK. These inhibitors were synthetized in the group of Prof. Luisa Carlota López-Cara at the Department of Pharmaceutical and Organic Chemistry of the University of Granada (Spain) in the framework of an ongoing collaboration between the two groups.

Anticancer drug screening from images of zebrafish embryogenesis

During the previous 10 years, global R&D expenditure in the pharmaceuticals and biotechnology sector has steadily increased, without a corresponding increase in output of new medicines. To address this situation, the biopharmaceutical industry's greatest need is to predict the failures at the earliest possible stage of the drug development process. A major key to reducing failures in drug screenings is the development and use of preclinical models that are more predictive of efficacy and safety in clinical trials. Further, relevant animal models are needed to allow a wider testing of novel hypotheses. Key to this is the developing, refining, and validating of complex animal models that directly link therapeutic targets to the phenotype of disease, allowing earlier prediction of human response to medicines and identification of safety biomarkers. Morehover, well-designed animal studies are essential to bridge the gap between test in cell cultures and people. Zebrafish is emerging, complementary to other models, as a powerful system for cancer studies and drugs discovery. We aim to investigate this research area designing a new preclinical cancer model based on the in vivo imaging of zebrafish embryogenesis. Technological advances in imaging have made it feasible to acquire nondestructive in vivo images of fluorescently labeled structures, such as cell nuclei and membranes, throughout early Zebrafishsh embryogenesis. This In vivo image-based investigation provides measurements for a large number of features at cellular level and events including nuclei movements, cells counting, and mitosis detection, thereby enabling the estimation of more significant parameters such as proliferation rate, highly relevant for investigating anticancer drug effects. In this work, we designed a standardized procedure for accessing drug activity at the cellular level in live zebrafish embryos. The procedure includes methodologies and tools that combine imaging and fully automated measurements of embryonic cell proliferation rate. We achieved proliferation rate estimation through the automatic classification and density measurement of epithelial enveloping layer and deep layer cells. Automatic embryonic cells classification provides the bases to measure the variability of relevant parameters, such as cell density, in different classes of cells and is finalized to the estimation of efficacy and selectivity of anticancer drugs. Through these methodologies we were able to evaluate and to measure in vivo the therapeutic potential and overall toxicity of Dbait and Irinotecan anticancer molecules. Results achieved on these anticancer molecules are presented and discussed; furthermore, extensive accuracy measurements are provided to investigate the robustness of the proposed procedure. Altogether, these observations indicate that zebrafish embryo can be a useful and cost-effective alternative to some mammalian models for the preclinical test of anticancer drugs and it might also provides, in the near future, opportunities to accelerate the process of drug discovery.

Different addictive drug exposure induces selective alterations of the Endogenous Opioid System : modulation of transcription and epigenetic mechanisms

Drug addiction manifests clinically as compulsive drug seeking, and cravings that can persist and recur even after extended periods of abstinence. The fundamental principle that unites addictive drugs is that each one enhances synaptic DA by means that dissociate it from normal behavioral control, so that they act to reinforce their own acquisition. Our attention has focused on the study of phenomena associated with the consumption of alcohol and heroin. Alcohol has long been considered an unspecific pharmacological agent, recent molecular pharmacology studies have shown that acts on different primary targets. Through gene expression studies conducted recently it has been shown that the classical opioid receptors are differently involved in the consumption of ethanol and, furthermore, the system nociceptin / NOP, included in the family of endogenous opioid system, and both appear able to play a key role in the initiation of alcohol use in rodents. What emerges is that manipulation of the opioid system, nociceptin, may be useful in the treatment of addictions and there are several evidences that support the use of this strategy. The linkage between gene expression alterations and epigenetic modulation in PDYN and PNOC promoters following alcohol treatment confirm the possible chromatin remodeling mechanism already proposed for alcoholism. In the second part of present study, we also investigated alterations in signaling molecules directly associated with MAPK pathway in a unique collection of postmortem brains from heroin abusers. The interest was focused on understanding the effects that prolonged exposure of heroin can cause in an individual, over the entire MAPK cascade and consequently on the transcription factor ELK1, which is regulated by this pathway. We have shown that the activation of ERK1/2 resulting in Elk-1 phosphorylation in striatal neurons supporting the hypothesis that prolonged exposure to substance abuse causes a dysregulation of MAPK pathway.

Small Molecules as Modulators of Different Targets Involved in Tumor Progression

Tumor is a lesion that may be formed by an abnormal growth of neoplastic cells. Many factors increase the risk of cancer and different targets are involved in tumor progression. Within this thesis, we have addressed two different biological targets, independently connected with tumor formation, e.g. Hsp90 and androgen receptor. The ATP-dependent chaperone Hsp90 is responsible for the conformational maturation and the renaturation of proteins. “Client” proteins are associated with the cancer hallmarks, as cell proliferation and tumor progression. Consequently, Hsp90 has evolved into promising anticancer target. Over the past decade, radicicol has been identified as potential anticancer agent targeting Hsp90, but it is not active in vivo. With that aim of obtaining radicicol-related derivatives, we developed the design and synthesis of new chalcones analogs. Chalcones, which are abundant in edible plants, own a diverse array of pharmacological activities and are considered a versatile scaffold for drug design. Antiproliferative assays and western blot analysis on the new compounds showed that some of those display an interesting cytotoxic effect and the ability to modulate Hsp90 client proteins expression. Androgen Receptor (AR) hypersensitivity plays crucial role in prostate cancer, which progression is stimulated by androgens. The therapy consists in a combination of surgical or chemical castration, along with antiandrogens treatment. Casodex® (bicalutamide), is the most widespread antiandrogen used in clinic. However, hormonal therapy is time-limited since many patients develop resistance. Commercially available antiandrogens show a common scaffold, e.g. two substituted aromatic rings linked by a linear or a cyclic spacer. With the aim of obtaining novel pure AR antagonists, we developed a new synthetic methodology, which allowed us to introduce, as linker between two suitably chosen aromatic rings, a triazole moiety. Preliminary data suggest that the herein reported new molecules generally decrease PSA expression, thus confirming their potential AR antagonistic activity.

Synthesis of Biologically Active Small Molecules: Different Approaches to Drug Design

In the past years, genome biology had disclosed an ever-growing kind of biological targets that emerged as ideal points for therapeutic intervention. Nevertheless, the number of new chemical entities (NCEs) translated into effective therapies employed in the clinic, still not observed. Innovative strategies in drug discovery combined with different approaches to drug design should be searched for bridge this gap. In this context organic synthetic chemistry had to provide for effective strategies to achieve biologically active small molecules to consider not only as potentially drug candidates, but also as chemical tools to dissect biological systems. In this scenario, during my PhD, inspired by the Biology-oriented Synthesis approach, a small library of hybrid molecules endowed with privileged scaffolds, able to block cell cycle and to induce apoptosis and cell differentiation, merged with natural-like cores were synthesized. A synthetic platform which joined a Domino Knoevenagel-Diels Alder reaction with a Suzuki coupling was performed in order to reach the hybrid compounds. These molecules can represent either antitumor lead candidates, or valuable chemical tools to study molecular pathways in cancer cells. The biological profile expressed by some of these derivatives showed a well defined antiproliferative activity on leukemia Bcr-Abl expressing K562 cell lines. A parallel project regarded the rational design and synthesis of minimally structurally hERG blockers with the purpose of enhancing the SAR studies of a previously synthesized collection. A Target-Oriented Synthesis approach was applied. Combining conventional and microwave heating, the desired final compounds were achieved in good yields and reaction rates. The preliminary biological results of the compounds, showed a potent blocking activity. The obtained small set of hERG blockers, was able to gain more insight the minimal structural requirements for hERG liability, which is mandatory to investigate in order to reduce the risk of potential side effects of new drug candidates.

A preclinical study of spinal cord injury focused on cellular and molecular modifications as potential targets for innovative therapies

Spinal Cord Injury (SCI) is a devastating condition for human and animal health. In SCI particularly, neurons, oligodendrocytes precursor cells, and mature oligodendrocytes are highly vulnerable to the toxic microenvironment after the lesion and susceptible to the elevated levels of noxious stimuli. Thus the regenerative response of the organism in case of SCI is significantly reduced, and only little spontaneous amelioration is observed in lesioned patients during the early phases. This work mainly focuses on studying and characterizing the modification induced by the SCI in a preclinical animal model. We investigated the ECM composition in the spinal cord segments surrounding the primary lesion site at a gene expression level. We found Timp1 and CD44 as a crucial hub in the secondary cascade of SCI in both spinal cord segments surrounding the lesion site. Interestingly, a temporal and anatomical difference in gene expression, indicating a complex regulation of ECM genes after SCI that could be used as a tool for regenerative medicine. We also investigated the modification in synaptic plasticity-related gene expression in spinal and supraspinal areas involved in motor control. We confirmed the anatomical and temporal difference in gene expression in spinal cord tissue. This analysis suggests that a molecular mapping of the lesion-induced modification could be a useful tool for regenerative medicine. In the last part, we evaluated the efficacy of an implantable biopolymer loaded with an anti-inflammatory drug and a pro-myelinating agent on the acute phase of SCI in our preclinical model. We found a consistent reduction of the inflammatory state in the spinal lesion site and the cord's surrounding segments. Moreover, we found increased preservation of the spinal cord tissue with a related upregulation of neuronal and oligodendroglial markers after lesion. Our treatment showed effective ameliorating functional outcome and reducing the lesion extension in the chronic phase.

Modeling Alzheimer disease with iPSCs and mouse model for the identification of risk factors, new targets, and potential therapeutical strategies

Alzheimer's disease (AD) is the most common neurodegenerative disease in elderly. Donepezil is the first-line drug used for AD. In section one, the experimental activity was oriented to evaluate and characterize molecular and cellular mechanisms that contribute to neurodegeneration induced by the Aβ1-42 oligomers (Aβ1-42O) and potential neuroprotective effects of the hybrids feruloyl-donepezil compound called PQM130. The effects of PQM130 were compared to donepezil in a murine AD model, obtained by intracerebroventricular (i.c.v.) injection of Aβ1-42O. The intraperitoneal administration of PQM130 (0.5-1 mg/kg) after i.c.v. Aβ1-42O injection improved learning and memory, protecting mice against spatial cognition decline. Moreover, it reduced oxidative stress, neuroinflammation and neuronal apoptosis, induced cell survival and protein synthesis in mice hippocampus. PQM130 modulated different pathways than donepezil, and it is more effective in counteracting Aβ1-42O damage. The section two of the experimental activity was focused on studying a loss of function variants of ABCA7. GWA studies identified mutations in the ABCA7 gene as a risk factor for AD. The mechanism through which ABCA7 contributes to AD is not clear. ABCA7 regulates lipid metabolism and critically controls phagocytic function. To investigate ABCA7 functions, CRISPR/Cas9 technology was used to engineer human iPSCs and to carry the genetic variant Y622*, which results in a premature stop codon, causing ABCA7 loss-of-function. From iPSCs, astrocytes were generated. This study revealed the effects of ABCA7 loss in astrocytes. ABCA7 Y622* mutation induced dysfunctional endocytic trafficking, impairing Aβ clearance, lipid dysregulation and cell homeostasis disruption, alterations that could contribute to AD. Though further studies are needed to confirm the PQM130 neuroprotective role and ABCA7 function in AD, the provided results showed a better understanding of AD pathophysiology, a new therapeutic approach to treat AD, and illustrated an innovative methodology for studying the disease.

Development of advanced analytical methodologies for Alzheimer’s disease drug discovery

Advanced analytical methodologies were developed to characterize new potential active MTDLs on isolated targets involved in the first stages of Alzheimer’s disease (AD). In addition, the methods investigated drug-protein bindings and evaluated protein-protein interactions involved in the neurodegeneration. A high-throughput luminescent assay allowed the study of the first in class GSK-3β/ HDAC dual inhibitors towards the enzyme GSK-3β. The method was able to identify an innovative disease-modifying agent with an activity in the micromolar range both on GSK-3β, HDAC1 and HDAC6. Then, the same assay reliably and quickly selected true positive hit compounds among natural Amaryllidaceae alkaloids tested against GSK-3β. Hence, given the central role of the amyloid pathway in the multifactorial nature of AD, a multi-methodological approach based on mass spectrometry (MS), circular dichroism spectroscopy (CD) and ThT assay was applied to characterize the potential interaction of CO releasing molecules (CORMs) with Aβ1-42 peptide. The comprehensive method provided reliable information on the different steps of the fibrillation process and regarding CORMs mechanism of action. Therefore, the optimal CORM-3/Aβ1−42 ratio in terms of inhibitory effect was identified by mass spectrometry. CD analysis confirmed the stabilizing effect of CORM-3 on the Aβ1−42 peptide soluble form and the ThT Fluorescent Analysis ensured that the entire fibrillation process was delayed. Then the amyloid aggregation process was studied in view of a possible correlation with AD lipid brain alterations. Therefore, SH-SY5Y cells were treated with increasing concentration of Aß1-42 at different times and the samples were analysed by a RP-UHPLC system coupled with a high-resolution quadrupole TOF mass spectrometer in comprehensive data-independent SWATH acquisition mode. Each lipid class profiling in SH-SY5Y cells treated with Aß1-42 was compared to the one obtained from the untreated. The approach underlined some peculiar lipid alterations, suitable as biomarkers, that might be correlated to Aß1-42 different aggregation species.

New chemical modalities for leishmaniasis drug discovery

Leishmaniasis is one of the major parasitic diseases among neglected tropical diseases with a high rate of morbidity and mortality. Human migration and climate change have spread the disease from limited endemic areas all over the world, also reaching regions in Southern Europe, and causing significant health and economic burden. The currently available treatments are far from ideal due to host toxicity, elevated cost, and increasing rates of drug resistance. Safer and more effective drugs are thus urgently required. Nevertheless, the identification of new chemical entities for leishmaniasis has proven to be incredibly hard and exacerbated by the scarcity of well-validated targets. Trypanothione reductase (TR) represents one robustly validated target in Leishmania that fulfils most of the requirements for a good drug target. However, due to the large and featureless active site, TR is considered extremely challenging and almost undruggable by small molecules. This scenario advocates the development of new chemical entities by unlocking new modalities for leishmaniasis drug discovery. The classical toolbox for drug discovery has enormously expanded in the last decade, and medicinal chemists can now strategize across a variety of new chemical modalities and a vast chemical space, to efficiently modulate challenging targets and provide effective treatments. Beyond others, Targeted p Protein Degradation (TPD) is an emerging strategy that uses small molecules to hijack endogenous proteolysis systems to degrade disease-relevant proteins and thus reduce their abundance in the cell. Based on these considerations, this thesis aimed to develop new strategies for leishmaniasis drug discovery while embracing novel chemical modalities and navigating the chemical space by chasing unprecedented chemotypes. This has been achieved by four complementary projects. We believe that these next-generation chemical modalities for leishmaniasis will play an important role in what was previously thought to be a drug discovery landscape dominated by small molecules.

Anticancer drug discovery using artificial intelligence: an application in pharmacological activity prediction

Hematological cancers are a heterogeneous family of diseases that can be divided into leukemias, lymphomas, and myelomas, often called “liquid tumors”. Since they cannot be surgically removable, chemotherapy represents the mainstay of their treatment. However, it still faces several challenges like drug resistance and low response rate, and the need for new anticancer agents is compelling. The drug discovery process is long-term, costly, and prone to high failure rates. With the rapid expansion of biological and chemical "big data", some computational techniques such as machine learning tools have been increasingly employed to speed up and economize the whole process. Machine learning algorithms can create complex models with the aim to determine the biological activity of compounds against several targets, based on their chemical properties. These models are defined as multi-target Quantitative Structure-Activity Relationship (mt-QSAR) and can be used to virtually screen small and large chemical libraries for the identification of new molecules with anticancer activity. The aim of my Ph.D. project was to employ machine learning techniques to build an mt-QSAR classification model for the prediction of cytotoxic drugs simultaneously active against 43 hematological cancer cell lines. For this purpose, first, I constructed a large and diversified dataset of molecules extracted from the ChEMBL database. Then, I compared the performance of different ML classification algorithms, until Random Forest was identified as the one returning the best predictions. Finally, I used different approaches to maximize the performance of the model, which achieved an accuracy of 88% by correctly classifying 93% of inactive molecules and 72% of active molecules in a validation set. This model was further applied to the virtual screening of a small dataset of molecules tested in our laboratory, where it showed 100% accuracy in correctly classifying all molecules. This result is confirmed by our previous in vitro experiments.