Predicting the target landscape of kinase inhibitors using 3D convolutional neural networks.

Many therapies in clinical trials are based on single drug-single target relationships. To further extend this concept to multi-target approaches using multi-targeted drugs, we developed a machine learning pipeline to unravel the target landscape of kinase inhibitors. This pipeline, which we call 3D-KINEssence, uses a new type of protein fingerprints (3D FP) based on the structure of kinases generated through a 3D convolutional neural network (3D-CNN). These 3D-CNN kinase fingerprints were matched to molecular Morgan fingerprints to predict the targets of each respective kinase inhibitor based on available bioactivity data. The performance of the pipeline was evaluated on two test sets: a sparse drug-target set where each drug is matched in most cases to a single target and also on a densely-covered drug-target set where each drug is matched to most if not all targets. This latter set is more challenging to train, given its non-exclusive character. Our model's root-mean-square error (RMSE) based on the two datasets was 0.68 and 0.8, respectively. These results indicate that 3D FP can predict the target landscape of kinase inhibitors at around 0.8 log units of bioactivity. Our strategy can be utilized in proteochemometric or chemogenomic workflows by consolidating the target landscape of kinase inhibitors.

KLIFS: an overhaul after the first 5 years of supporting kinase research.

Kinases are a prime target of drug development efforts with >60 drug approvals in the past two decades. Due to the research into this protein family, a wealth of data has been accumulated that keeps on growing. KLIFS-Kinase-Ligand Interaction Fingerprints and Structures-is a structural database focusing on how kinase inhibitors interact with their targets. The aim of KLIFS is to support (structure-based) kinase research through the systematic collection, annotation, and processing of kinase structures. Now, 5 years after releasing the initial KLIFS website, the database has undergone a complete overhaul with a new website, new logo, and new functionalities. In this article, we start by looking back at how KLIFS has been used by the research community, followed by a description of the renewed KLIFS, and conclude with showcasing the functionalities of KLIFS. Major changes include the integration of approved drugs and inhibitors in clinical trials, extension of the coverage to atypical kinases, and a RESTful API for programmatic access. KLIFS is available at the new domain https://klifs.net.

CXCR4 expression in glioblastoma tissue and the potential for PET imaging and treatment with [68Ga]Ga-Pentixafor /[177Lu]Lu-Pentixather.

PURPOSE: CXCR4 (over)expression is found in multiple human cancer types, while expression is low or absent in healthy tissue. In glioblastoma it is associated with a poor prognosis and more extensive infiltrative phenotype. CXCR4 can be targeted by the diagnostic PET agent [68Ga]Ga-Pentixafor and its therapeutic counterpart [177Lu]Lu-Pentixather. We aimed to investigate the expression of CXCR4 in glioblastoma tissue to further examine the potential of these PET agents. METHODS: CXCR4 mRNA expression was examined using the R2 genomics platform. Glioblastoma tissue cores were stained for CXCR4. CXCR4 staining in tumor cells was scored. Stained tissue components (cytoplasm and/or nuclei of the tumor cells and blood vessels) were documented. Clinical characteristics and information on IDH and MGMT promoter methylation status were collected. Seven pilot patients with recurrent glioblastoma underwent [68Ga]Ga-Pentixafor PET; residual resected tissue was stained for CXCR4. RESULTS: Two large mRNA datasets (N = 284; N = 540) were assesed. Of the 191 glioblastomas, 426 cores were analyzed using immunohistochemistry. Seventy-eight cores (23 tumors) were CXCR4 negative, while 18 cores (5 tumors) had both strong and extensive staining. The remaining 330 cores (163 tumors) showed a large inter- and intra-tumor variation for CXCR4 expression; also seen in the resected tissue of the seven pilot patients-not directly translatable to [68Ga]Ga-Pentixafor PET results. Both mRNA and immunohistochemical analysis showed CXCR4 negative normal brain tissue and no significant correlation between CXCR4 expression and IDH or MGMT status or survival. CONCLUSION: Using immunohistochemistry, high CXCR4 expression was found in a subset of glioblastomas as well as a large inter- and intra-tumor variation. Caution should be exercised in directly translating ex vivo CXCR4 expression to PET agent uptake. However, when high CXCR4 expression can be identified with [68Ga]Ga-Pentixafor, these patients might be good candidates for targeted radionuclide therapy with [177Lu]Lu-Pentixather in the future.

The TICking clock of EGFR therapy resistance in glioblastoma: Target Independence or target Compensation.

Targeted therapy against driver mutations responsible for cancer progression has been shown to be effective in many tumor types. For glioblastoma (GBM), the epidermal growth factor receptor (EGFR) gene is the most frequently mutated oncogenic driver and has therefore been considered an attractive target for therapy. However, so far responses to EGFR-pathway inhibitors have been disappointing. We performed an exhaustive analysis of the mechanisms that might account for therapy resistance against EGFR inhibition. We define two major mechanisms of resistance and propose modalities to overcome them. The first resistance mechanism concerns target independence. In this case, cells have lost expression of the EGFR protein and experience no negative impact of EGFR targeting. Loss of extrachromosomally encoded EGFR as present in double minute DNA is a frequent mechanism for this type of drug resistance. The second mechanism concerns target compensation. In this case, cells will counteract EGFR inhibition by activation of compensatory pathways that render them independent of EGFR signaling. Compensatory pathway candidates are platelet-derived growth factor ß (PDGFß), Insulin-like growth factor 1 (IGFR1) and cMET and their downstream targets, all not commonly mutated at the time of diagnosis alongside EGFR mutation. Given that both mechanisms make cells independent of EGFR expression, other means have to be found to eradicate drug resistant cells. To this end we suggest rational strategies which include the use of multi-target therapies that hit truncation mutations (mechanism 1) or multi-target therapies to co-inhibit compensatory proteins (mechanism 2).

WINDOW consortium: A path towards increased therapy efficacy against glioblastoma.

Glioblastoma is the most common and malignant form of brain cancer, for which the standard treatment is maximal surgical resection, radiotherapy and chemotherapy. Despite these interventions, mean overall survival remains less than 15 months, during which extensive tumor infiltration throughout the brain occurs. The resulting metastasized cells in the brain are characterized by chemotherapy resistance and extensive intratumoral heterogeneity. An orthogonal approach attacking both intracellular resistance mechanisms as well as intercellular heterogeneity is necessary to halt tumor progression. For this reason, we established the WINDOW Consortium (Window for Improvement for Newly Diagnosed patients by Overcoming disease Worsening), in which we are establishing a strategy for rational selection and development of effective therapies against glioblastoma. Here, we overview the many challenges posed in treating glioblastoma, including selection of drug combinations that prevent therapy resistance, the need for drugs that have improved blood brain barrier penetration and strategies to counter heterogeneous cell populations within patients. Together, this forms the backbone of our strategy to attack glioblastoma.

Sequencing of neuroblastoma identifies chromothripsis and defects in neuritogenesis genes.

Neuroblastoma is a childhood tumour of the peripheral sympathetic nervous system. The pathogenesis has for a long time been quite enigmatic, as only very few gene defects were identified in this often lethal tumour. Frequently detected gene alterations are limited to MYCN amplification (20%) and ALK activations (7%). Here we present a whole-genome sequence analysis of 87 neuroblastoma of all stages. Few recurrent amino-acid-changing mutations were found. In contrast, analysis of structural defects identified a local shredding of chromosomes, known as chromothripsis, in 18% of high-stage neuroblastoma. These tumours are associated with a poor outcome. Structural alterations recurrently affected ODZ3, PTPRD and CSMD1, which are involved in neuronal growth cone stabilization. In addition, ATRX, TIAM1 and a series of regulators of the Rac/Rho pathway were mutated, further implicating defects in neuritogenesis in neuroblastoma. Most tumours with defects in these genes were aggressive high-stage neuroblastomas, but did not carry MYCN amplifications. The genomic landscape of neuroblastoma therefore reveals two novel molecular defects, chromothripsis and neuritogenesis gene alterations, which frequently occur in high-risk tumours.

The allosteric AKT inhibitor MK2206 shows a synergistic interaction with chemotherapy and radiotherapy in glioblastoma spheroid cultures.

BACKGROUND: Glioblastoma multiforme (GBM) is the most common, invasive and deadly primary type of malignant brain tumor. The Phosphatidylinositol-3-Kinase/AKT (PI3K/AKT) pathway is highly active in GBM and has been associated with increased survival and resistance to therapy. The aim of this study is to investigate the effects of AKT inhibition in combination with the current standard of care which consists of irradiation and temozolomide (TMZ) on human malignant glioma cells growing adherent and as multicellular spheroids in vitro. METHODS: The effects of the allosteric inhibitor MK2206 combined with irradiation and TMZ were assessed on glioma cells growing adherent and as multicellular 3D spheroids. The interaction was studied on proliferation, clonogenic cell survival, cell invasion, -migration and on expression of key proteins in the PI3K-AKT pathway by western blot. RESULTS: A differential effect was found at low- (1 µM) and high dose (10 µM) MK2206. At 1 µM, the inhibitor reduced phosphorylation of Thr308 and Ser473 residues of AKT in both adherent cells and spheroids. Low dose MK2206 delayed spheroid growth and sensitized spheroids to both irradiation and TMZ in a synergistic way (Combination index <0.35). In contrast, neither low nor high dose MK2206 did enhance therapy sensitivity in adherent growing cells. Effective inhibition of invasion and migration was observed only at higher doses of MK2206 (>5 µM). CONCLUSIONS: The data show that a 3D spheroid model show different sensitivity to irradiation when combined with AKT inhibition. Thereby we show that MK2206 has potential synergistic efficacy to the current standard of care for glioma patients.

Oncogenic composite mutations can be predicted by co-mutations and their chromosomal location.

Genetic heterogeneity in tumors can show a remarkable selectivity when two or more independent genetic events occur in the same gene. This phenomenon, called composite mutation, points toward a selective pressure, which frequently causes therapy resistance to mutation-specific drugs. Since composite mutations have been described to occur in sub-clonal populations, they are not always captured through biopsy sampling. Here, we provide a proof of concept to predict composite mutations to anticipate which patients might be at risk for sub-clonally driven therapy resistance. We found that composite mutations occur in 5% of cancer patients, mostly affecting the PIK3CA, EGFR, BRAF, and KRAS genes, which are common precision medicine targets. Furthermore, we found a strong and significant relationship between the frequencies of composite mutations with commonly co-occurring mutations in a non-composite context. We also found that co-mutations are significantly enriched on the same chromosome. These observations were independently confirmed using cell line data. Finally, we show the feasibility of predicting compositive mutations based on their co-mutations (AUC 0.62, 0.81, 0.82, and 0.91 for EGFR, PIK3CA, KRAS, and BRAF, respectively). This prediction model could help to stratify patients who are at risk of developing therapy resistance-causing mutations.

A Real-world Toxicity Atlas Shows that Adverse Events of Combination Therapies Commonly Result in Additive Interactions.

PURPOSE: Combination therapies are a promising approach for improving cancer treatment, but it is challenging to predict their resulting adverse events in a real-world setting. EXPERIMENTAL DESIGN: We provide here a proof-of-concept study using 15 million patient records from the FDA Adverse Event Reporting System (FAERS). Complex adverse event frequencies of drugs or their combinations were visualized as heat maps onto a two-dimensional grid. Adverse event frequencies were shown as colors to assess the ratio between individual and combined drug effects. To capture these patterns, we trained a convolutional neural network (CNN) autoencoder using 7,300 single-drug heat maps. In addition, statistical synergy analyses were performed on the basis of BLISS independence or χ2 testing. RESULTS: The trained CNN model was able to decode patterns, showing that adverse events occur in global rather than isolated and unique patterns. Patterns were not likely to be attributed to disease symptoms given their relatively limited contribution to drug-associated adverse events. Pattern recognition was validated using trial data from ClinicalTrials.gov and drug combination data. We examined the adverse event interactions of 140 drug combinations known to be avoided in the clinic and found that near all of them showed additive rather than synergistic interactions, also when assessed statistically. CONCLUSIONS: Our study provides a framework for analyzing adverse events and suggests that adverse drug interactions commonly result in additive effects with a high level of overlap of adverse event patterns. These real-world insights may advance the implementation of new combination therapies in clinical practice.

The Epigenetic Evolution of Glioma Is Determined by the IDH1 Mutation Status and Treatment Regimen.

Tumor adaptation or selection is thought to underlie therapy resistance in glioma. To investigate longitudinal epigenetic evolution of gliomas in response to therapeutic pressure, we performed an epigenomic analysis of 132 matched initial and recurrent tumors from patients with IDH-wildtype (IDHwt) and IDH-mutant (IDHmut) glioma. IDHwt gliomas showed a stable epigenome over time with relatively low levels of global methylation. The epigenome of IDHmut gliomas showed initial high levels of genome-wide DNA methylation that was progressively reduced to levels similar to those of IDHwt tumors. Integration of epigenomics, gene expression, and functional genomics identified HOXD13 as a master regulator of IDHmut astrocytoma evolution. Furthermore, relapse of IDHmut tumors was accompanied by histologic progression that was associated with survival, as validated in an independent cohort. Finally, the initial cell composition of the tumor microenvironment varied between IDHwt and IDHmut tumors and changed differentially following treatment, suggesting increased neoangiogenesis and T-cell infiltration upon treatment of IDHmut gliomas. This study provides one of the largest cohorts of paired longitudinal glioma samples with epigenomic, transcriptomic, and genomic profiling and suggests that treatment of IDHmut glioma is associated with epigenomic evolution toward an IDHwt-like phenotype. SIGNIFICANCE: Standard treatments are related to loss of DNA methylation in IDHmut glioma, resulting in epigenetic activation of genes associated with tumor progression and alterations in the microenvironment that resemble treatment-naïve IDHwt glioma.

Review: A Roadmap to Use Nonstructured Data to Discover Multitarget Cancer Therapies.

Therapy resistance to single agents has led to the realization that combination therapies could become the cornerstone of cancer treatment. To operationalize the selection of effective and safe multitarget therapies, we propose to integrate chemical and preclinical therapeutic information with clinical efficacy and toxicity data, allowing a new perspective on the drug target landscape. To assess the feasibility of this approach, we evaluated the publicly available chemical, preclinical, and clinical therapeutic data, and we addressed some potential limitations while integrating the data. First, by mapping available structured data from the main biomedical resources, we noticed that there is only a 1.7% overlap between drugs in chemical, preclinical, or clinical databases. Especially, the limited amount of structured data in the clinical domain hinders linking drugs to clinical aspects such as efficacy and side effects. Second, to overcome the abovementioned knowledge gap between the chemical, preclinical, and clinical domain, we suggest information extraction from scientific literature and other unstructured resources through natural language processing models, where BioBERT and PubMedBERT are the current state-of-the-art approaches. Finally, we propose that knowledge graphs can be used to link structured data, scientific literature, and electronic health records, to come to meaningful interpretations. Together, we expect this richer knowledge will lower barriers toward clinical application of personalized combination therapies with high efficacy and limited adverse events.

Explaining Blood-Brain Barrier Permeability of Small Molecules by Integrated Analysis of Different Transport Mechanisms.

The blood-brain barrier (BBB) represents a major obstacle to delivering drugs to the central nervous system (CNS), resulting in the lack of effective treatment for many CNS diseases including brain cancer. To accelerate CNS drug development, computational prediction models could save the time and effort needed for experimental evaluation. Here, we studied BBB permeability focusing on active transport (influx and efflux) as well as passive diffusion using previously published and self-curated data sets. We created prediction models based on physicochemical properties, molecular substructures, or their combination to understand which mechanisms contribute to BBB permeability. Our results show that features that predicted passive diffusion over membranes overlap with features that explain endothelial permeation of approved CNS-active drugs. We also identified physical properties and molecular substructures that positively or negatively predicted BBB transport. These findings provide guidance toward identifying BBB-permeable compounds by optimally matching physicochemical and molecular properties to BBB transport mechanisms.

Longitudinal drug synergy assessment using convolutional neural network image-decoding of glioblastoma single-spheroid cultures.

Background: In recent years, drug combinations have become increasingly popular to improve therapeutic outcomes in various diseases, including difficult to cure cancers such as the brain cancer glioblastoma. Assessing the interaction between drugs over time is critical for predicting drug combination effectiveness and minimizing the risk of therapy resistance. However, as viability readouts of drug combination experiments are commonly performed as an endpoint where cells are lysed, longitudinal drug-interaction monitoring is currently only possible through combined endpoint assays. Methods: We provide a method for massive parallel monitoring of drug interactions for 16 drug combinations in 3 glioblastoma models over a time frame of 18 days. In our assay, viabilities of single neurospheres are to be estimated based on image information taken at different time points. Neurosphere images taken on the final day (day 18) were matched to the respective viability measured by CellTiter-Glo 3D on the same day. This allowed to use of machine learning to decode image information to viability values on day 18 as well as for the earlier time points (on days 8, 11, and 15). Results: Our study shows that neurosphere images allow us to predict cell viability from extrapolated viabilities. This enables to assess of the drug interactions in a time window of 18 days. Our results show a clear and persistent synergistic interaction for several drug combinations over time. Conclusions: Our method facilitates longitudinal drug-interaction assessment, providing new insights into the temporal-dynamic effects of drug combinations in 3D neurospheres which can help to identify more effective therapies against glioblastoma.

Screening of predicted synergistic multi-target therapies in glioblastoma identifies new treatment strategies.

Background: IDH-wildtype glioblastoma (GBM) is a highly malignant primary brain tumor with a median survival of 15 months after standard of care, which highlights the need for improved therapy. Personalized combination therapy has shown to be successful in many other tumor types and could be beneficial for GBM patients. Methods: We performed the largest drug combination screen to date in GBM, using a high-throughput effort where we selected 90 drug combinations for their activity onto 25 patient-derived GBM cultures. 43 drug combinations were selected for interaction analysis based on their monotherapy efficacy and were tested in a short-term (3 days) as well as long-term (18 days) assay. Synergy was assessed using dose-equivalence and multiplicative survival metrics. Results: We observed a consistent synergistic interaction for 15 out of 43 drug combinations on patient-derived GBM cultures. From these combinations, 11 out of 15 drug combinations showed a longitudinal synergistic effect on GBM cultures. The highest synergies were observed in the drug combinations Lapatinib with Thapsigargin and Lapatinib with Obatoclax Mesylate, both targeting epidermal growth factor receptor and affecting the apoptosis pathway. To further elaborate on the apoptosis cascade, we investigated other, more clinically relevant, apoptosis inducers and observed a strong synergistic effect while combining Venetoclax (BCL targeting) and AZD5991 (MCL1 targeting). Conclusions: Overall, we have identified via a high-throughput drug screening several new treatment strategies for GBM. Moreover, an exceptionally strong synergistic interaction was discovered between kinase targeting and apoptosis induction which is suitable for further clinical evaluation as multi-targeted combination therapy.

Longitudinal characteristics of T2-FLAIR mismatch in IDH-mutant astrocytomas: Relation to grade, histopathology, and overall survival in the GLASS-NL cohort.

Background: The T2-FLAIR mismatch sign is defined by signal loss of the T2-weighted hyperintense area with Fluid-Attenuated Inversion Recovery (FLAIR) on magnetic resonance imaging, causing a hypointense region on FLAIR. It is a highly specific diagnostic marker for IDH-mutant astrocytoma and is postulated to be caused by intercellular microcystic change in the tumor tissue. However, not all IDH-mutant astrocytomas show this mismatch sign and some show the phenomenon in only part of the lesion. The aim of the study is to determine whether the T2-FLAIR mismatch phenomenon has any prognostic value beyond initial noninvasive molecular diagnosis. Methods: Patients initially diagnosed with histologically lower-grade (2 or 3) IDH-mutant astrocytoma and with at least 2 surgical resections were included in the GLASS-NL cohort. T2-FLAIR mismatch was determined, and the growth pattern of the recurrent tumor immediately before the second resection was annotated as invasive or expansive. The relation between the T2-FLAIR mismatch sign and tumor grade, microcystic change, overall survival (OS), and other clinical parameters was investigated both at first and second resection. Results: The T2-FLAIR mismatch sign was significantly related to Grade 2 (80% vs 51%), longer post-resection median OS (8.3 vs 5.2 years), expansive growth, and lower age at second resection. At first resection, no relation was found between the mismatch sign and OS. Microcystic change was associated with areas of T2-FLAIR mismatch. Conclusions: T2-FLAIR mismatch in IDH-mutant astrocytomas is correlated with microcystic change in the tumor tissue, favorable prognosis, and Grade 2 tumors at the time of second resection.

Data-driven prioritization and preclinical evaluation of therapeutic targets in glioblastoma.

BACKGROUND: Patients with glioblastoma (GBM) have a dismal prognosis, and there is an unmet need for new therapeutic options. This study aims to identify new therapeutic targets in GBM. METHODS: mRNA expression data of patient-derived GBM (n = 1279) and normal brain tissue (n = 46) samples were collected from Gene Expression Omnibus and The Cancer Genome Atlas. Functional genomic mRNA profiling was applied to capture the downstream effects of genomic alterations on gene expression levels. Next, a class comparison between GBM and normal brain tissue was performed. Significantly upregulated genes in GBM were further prioritized based on (1) known interactions with antineoplastic drugs, (2) current drug development status in humans, and (3) association with biologic pathways known to be involved in GBM. Antineoplastic agents against prioritized targets were validated in vitro and in vivo. RESULTS: We identified 712 significantly upregulated genes in GBM compared to normal brain tissue, of which 27 have a known interaction with antineoplastic agents. Seventeen of the 27 genes, including EGFR and VEGFA, have been clinically evaluated in GBM with limited efficacy. For the remaining 10 genes, RRM2, MAPK9 (JNK2, SAPK1a), and XIAP play a role in GBM development. We demonstrated for the MAPK9 inhibitor RGB-286638 a viability loss in multiple GBM cell culture models. Although no overall survival benefit was observed in vivo, there were indications that RGB-286638 may delay tumor growth. CONCLUSIONS: The MAPK9 inhibitor RGB-286638 showed promising in vitro results. Furthermore, in vivo target engagement studies and combination therapies with this compound warrant further exploration.

p120-catenin-dependent collective brain infiltration by glioma cell networks.

Diffuse brain infiltration by glioma cells causes detrimental disease progression, but its multicellular coordination is poorly understood. We show here that glioma cells infiltrate the brain collectively as multicellular networks. Contacts between moving glioma cells are adaptive epithelial-like or filamentous junctions stabilized by N-cadherin, ß-catenin and p120-catenin, which undergo kinetic turnover, transmit intercellular calcium transients and mediate directional persistence. Downregulation of p120-catenin compromises cell-cell interaction and communication, disrupts collective networks, and both the cadherin and RhoA binding domains of p120-catenin are required for network formation and migration. Deregulating p120-catenin further prevents diffuse glioma cell infiltration of the mouse brain with marginalized microlesions as the outcome. Transcriptomics analysis has identified p120-catenin as an upstream regulator of neurogenesis and cell cycle pathways and a predictor of poor clinical outcome in glioma patients. Collective glioma networks infiltrating the brain thus depend on adherens junctions dynamics, the targeting of which may offer an unanticipated strategy to halt glioma progression.

A cancer drug atlas enables synergistic targeting of independent drug vulnerabilities.

Personalized cancer treatments using combinations of drugs with a synergistic effect is attractive but proves to be highly challenging. Here we present an approach to uncover the efficacy of drug combinations based on the analysis of mono-drug effects. For this we used dose-response data from pharmacogenomic encyclopedias and represent these as a drug atlas. The drug atlas represents the relations between drug effects and allows to identify independent processes for which the tumor might be particularly vulnerable when attacked by two drugs. Our approach enables the prediction of combination-therapy which can be linked to tumor-driving mutations. By using this strategy, we can uncover potential effective drug combinations on a pan-cancer scale. Predicted synergies are provided and have been validated in glioblastoma, breast cancer, melanoma and leukemia mouse-models, resulting in therapeutic synergy in 75% of the tested models. This indicates that we can accurately predict effective drug combinations with translational value.

The Landscape of Atypical and Eukaryotic Protein Kinases.

Kinases are attractive anticancer targets due to their central role in the growth, survival, and therapy resistance of tumor cells. This review explores the two primary kinase classes, the eukaryotic protein kinases (ePKs) and the atypical protein kinases (aPKs), and provides a structure-centered comparison of their sequences, structures, hydrophobic spines, mutation and SNP hotspots, and inhibitor interaction patterns. Despite the limited sequence similarity between these two classes, atypical kinases commonly share the archetypical kinase fold but lack conserved eukaryotic kinase motifs and possess altered hydrophobic spines. Furthermore, atypical kinase inhibitors explore only a limited number of binding modes both inside and outside the orthosteric binding site. The distribution of genetic variations in both classes shows multiple ways they can interfere with kinase inhibitor binding. This multilayered review provides a research framework bridging the eukaryotic and atypical kinase classes.

To Combine or Not Combine: Drug Interactions and Tools for Their Analysis. Reflections from the EORTC-PAMM Course on Preclinical and Early-phase Clinical Pharmacology.

Combination therapies are used in the clinic to achieve cure, better efficacy and to circumvent resistant disease in patients. Initial assessment of the effect of such combinations, usually of two agents, is frequently performed using in vitro assays. In this review, we give a short summary of the types of analyses that were presented during the Preclinical and Early-phase Clinical Pharmacology Course of the Pharmacology and Molecular Mechanisms Group, European Organization for Research and Treatment on Cancer, that can be used to determine the efficacy of drug combinations. The effect of a combination treatment can be calculated using mathematical equations based on either the Loewe additivity or Bliss independence model, or a combination of both, such as Chou and Talalay's median-drug effect model. Interactions can be additive, synergistic (more than additive), or antagonistic (less than additive). Software packages CalcuSyn (also available as CompuSyn) and Combenefit are designed to calculate the extent of the combined effects. Interestingly, the application of machine-learning methods in the prediction of combination treatments, which can include pharmacogenomic, genetic, metabolomic and proteomic profiles, might contribute to further refinement of combination regimens. However, more research is needed to apply appropriate rules of machine learning methods to ensure correct predictive models.