Drugging Hijacked Kinase Pathways in Pediatric Oncology: Opportunities and Current Scenario

Abstract

Childhood cancer is considered rare, corresponding to ~3% of all malignant neoplasms in the human population. The World Health Organization (WHO) reports a universal occurrence of more than 15 cases per 100,000 inhabitants around the globe, and despite improvements in diagnosis, treatment and supportive care, one child dies of cancer every 3 min. Consequently, more efficient, selective and affordable therapeutics are still needed in order to improve outcomes and avoid long-term sequelae. Alterations in kinases' functionality is a trademark of cancer and the concept of exploiting them as drug targets has burgeoned in academia and in the pharmaceutical industry of the 21st century. Consequently, an increasing plethora of inhibitors has emerged. In the present study, the expression patterns of a selected group of kinases (including tyrosine receptors, members of the PI3K/AKT/mTOR and MAPK pathways, coordinators of cell cycle progression, and chromosome segregation) and their correlation with clinical outcomes in pediatric solid tumors were accessed through the R2: Genomics Analysis and Visualization Platform and by a thorough search of published literature. To further illustrate the importance of kinase dysregulation in the pathophysiology of pediatric cancer, we analyzed the vulnerability of different cancer cell lines against their inhibition through the Cancer Dependency Map portal, and performed a search for kinase-targeted compounds with approval and clinical applicability through the CanSAR knowledgebase. Finally, we provide a detailed literature review of a considerable set of small molecules that mitigate kinase activity under experimental testing and clinical trials for the treatment of pediatric tumors, while discuss critical challenges that must be overcome before translation into clinical options, including the absence of compounds designed specifically for childhood tumors which often show differential mutational burdens, intrinsic and acquired resistance, lack of selectivity and adverse effects on a growing organism.

Keywords: chemical inhibitors; childhood cancer; clinical trials; kinases.

Figure 1

(A) Protein kinases may be triggered or deactivated in many ways, acting as key regulators of many features of cell behavior and specialized functions by coupling (1) reception of extracellular signals and (2) intracellular signaling transduction, leading to (3) direct cellular responses or (4) changes in gene expression. (B) In cancer, increased kinase activity may result from gene amplification (1,4), mutations that stabilize the kinase in an active conformation and destabilize cis-inhibitory interactions (2) and translocations that encode chimeric proteins with novel/increased activity (3). This figure was created using Servier Medical Art templates, which are licensed under a Creative Commons Attribution 3.0 Unported License; https://smart.servier.com (accessed on 14 December 2022).

Figure 2

(A) Protein–protein interactions accessed through the software STRING v11.5 (available at https://string-db.org/ (accessed on 2 November 2022)). The parameters evaluated were text mining, experiments and databases. Network edges denote confidence and the minimum required interaction score was 0.700, considered high. (A) K-means clustering; (B) enrichment analysis for biological processes.

Figure 3

(A) Polar plots of differentially expressed kinases in pediatric tumors obtained through the analysis of available data on the R2: Genomics Analysis and Visualization Platform (http://r2.amc.nl (accessed on 15 October 2022)). Tumor abbreviations: EPN—ependymoma; EWS—Ewing sarcoma; MB—medulloblastoma; NB—neuroblastoma; OS—osteosarcoma; PA—pilocytic astrocytoma; RB—retinoblastoma; RMS—rhabdomyosarcoma. p-values are represented by differential coloring gradients. The external inner circle corresponds with “tumor versus normal tissue” results. The other concentric layers represent data related to associations with clinical features: metastasis, death, molecular subgroup (MB), MYC status (NB) and Huvos grade (OS). For actual p-values, refer to Supplementary Table S2. (B) Percentage of tumors with altered expression of each kinase. Few commonalities were found.

Figure 4

Percentage of pediatric cell lines dependent on the selected group of kinases. Dependency data were imported from the DepMap consortium (CRISPR (DepMap 22Q2 Public + Score, Chronos; https://depmap.org/ portal/ (accessed on 30 October 2022)) and classified as highly dependent, dependent or not dependent. The data were plotted on a histogram where it is possible to see the vulnerability of pediatric cell lines mainly to cell cycle kinases and PI3K/AKT/mTOR families. Cell lines selected included Ewing sarcoma (EWS), osteosarcoma (OS), glioma (GB), medulloblastoma (MB), neuroblastoma (NB), neuroblastoma (NB), retinoblastoma (RB) and rhabdomyosarcoma (RMS). Dependency scores for each cell line are shown in Supplementary Table S3.

Figure 5

(A) Schematic illustrations of RTK druggability identified by the CanSAR database, including the total number of compounds with predicted interaction capacity with each kinase, as well as FDA-approved drugs and clinical candidates. (B) Interaction networks of RTK inhibitors and associated binding proteins according to STITCH (search tool for known and predicted interactions between chemicals and proteins available at http://stitch.embl.de (accessed on 1 November 2022)). Compounds are represented as pill-shaped nodes, while proteins are shown as spheres. Small nodes represent proteins of unknown 3D structures, while large nodes show proteins with known or predicted structures. Nodes that are associated with each other are linked by an edge: thicker lines represent stronger binding affinities. Networks were constructed considering a minimum required interaction score of 0.700, and based on associations reported in curated databases (gray lines), or on both databases and experimental/biochemical data (green lines). Purple lines represent functional links between proteins.

Figure 6

(A) Venn diagram showing the number of kinase inhibitors retrieved from the CanSAR platform versus those already tested in clinical trials. Around 30% of the compounds have been tested in patients. Of note, eight kinase inhibitors tested in clinical trials were not found in the CanSAR platform. (B) Venn diagram comparing kinase inhibitors found in the CanSAR platform without evidence of preclinical tests in pediatric cancer versus clinical trials. Three of the drugs are already tested in pediatric patients, without in vitro or in vivo evidence.