Comparative Assessment of Protein Kinase Inhibitors in Public Databases and in PKIDB

Abstract

Since the first approval of a protein kinase inhibitor (PKI) by the Food and Drug Administration (FDA) in 2001, 55 new PKIs have reached the market, and many inhibitors are currently being evaluated in clinical trials. This is a clear indication that protein kinases still represent major drug targets for the pharmaceutical industry. In a previous work, we have introduced PKIDB, a publicly available database, gathering PKIs that have already been approved (Phase 4), as well as those currently in clinical trials (Phases 0 to 3). This database is updated frequently, and an analysis of the new data is presented here. In addition, we compared the set of PKIs present in PKIDB with the PKIs in early preclinical studies found in ChEMBL, the largest publicly available chemical database. For each dataset, the distribution of physicochemical descriptors related to drug-likeness is presented. From these results, updated guidelines to prioritize compounds for targeting protein kinases are proposed. The results of a principal component analysis (PCA) show that the PKIDB dataset is fully encompassed within all PKIs found in the public database. This observation is reinforced by a principal moments of inertia (PMI) analysis of all molecules. Interestingly, we notice that PKIs in clinical trials tend to explore new 3D chemical space. While a great majority of PKIs is located on the area of "flatland", we find few compounds exploring the 3D structural space. Finally, a scaffold diversity analysis of the two datasets, based on frequency counts was performed. The results give insight into the chemical space of PKIs, and can guide researchers to reach out new unexplored areas. PKIDB is freely accessible from the following website: http://www.icoa.fr/pkidb.

Keywords: approved drugs; chemometrics analysis; clinical trials; database; kinome; molecular scaffolds; protein kinase inhibitors; rings system.

Figure 1

Progression of Food and Drug Administration (FDA)-approved protein kinase inhibitors (2001–2019) and their type of inhibition [12]. As of 11th December 2019, 55 kinase inhibitors were approved by the FDA. Not shown here: tivozanib, approved by European Medicines Agency (EMA) in 2017; anlotinib, apatinib, and icotinib, approved by the China Food and Drug Administration (CFDA) in 2018, 2014, and 2011, respectively; and fasudil, approved in China and in Japan in 1995. ND: not defined.

Figure 2

Distribution of physicochemical properties of PKIs: (a) number of hydrogen bond acceptors (HBA); (b) number of hydrogen bond donors (HBD); (c) ClogP (Rational Discovery Kit (RDKit)); (d) molecular weight (MW); (e) number of heavy atoms (NHA); (f) number of rotatable bonds (NRB); (g) topological polar surface area (TPSA); (h) number of aromatic rings (NAR); (i) number of chiral atoms (NCA). Pink areas represent values outside two standard deviation from the mean (95.4% confidence interval).

Figure 3

(a) Principal component analysis (PCA) of PKIs from ChEMBL and PKIDB, containing 76,504 and 209 compounds, respectively. Black, yellow, and red ellipses encompass 95% of the individuals from class “PKI_ChEMBL”, “Clinical_PKI”, and “Approved_PKI”, respectively; (b) correlation circle.

Figure 4

Principal moments of inertia (PMI) plot of PKIs in clinical trials (yellow), approved (red) and from ChEMBL database (light blue).

Figure 5

Representation of a molecular decomposition into scaffolds according to Bemis and Murcko (BM) and in graph framework.

Figure 6

Most retrieved Bemis and Murcko scaffolds in PKIDB dataset (a): (3Z)-3-(1H-pyrrol-2-ylmethylene)indolin-2-one and in PKI_ChEMBL dataset (b): N-phenylquinazolin-4-amine.

Figure 7

Application of the ring-system ensemble classification. Ring-system ensembles are obtained by removing substituents on acyclic bonds and by keeping attachment point (R1). The ring system unpositioned ensembles do not keep information on the attachment point. Rings are shown in bold.

Figure 8

Top ten bicycles retrieved in the PKIDB dataset (a) and in PKI_ChEMBL (b) with their occurrence and their frequency in brackets. In PKIDB there are 172 bicycles (56 unique) and in PKI_ChEMBL, there are 57,439 bicycles (918 unique).

Figure 9

Top ten most retrieved bicycles with their substituents in the PKIDB dataset (a) and in PKI_ChEMBL (b) with their occurrence and their frequency in brackets. In PKIDB, there are 172 bicycles (129 unique) and in PKI_ChEMBL, there are 57,438 bicycles (4480 unique). 1_*—connected to an atom not double bonded, not aromatic, not in a cycle and not halogen; 2_*—connected to non aromatic ring; 3_*—connected to aromatic atom; 4_*—connected to an halogen.