Are we ready to design oral PROTACs®?

PROTACs (R) are expected to strongly impact the future of drug discovery. Therefore, in this work we firstly performed a statistical study to highlight the distribution of E3 ligases and POls collected in PROTAC-DB, the main online database focused on degraders. Moreover, since the emerging technology of protein degradation deals with large and complex chemical structures, the second part of the paper focuses on how to set up a property-based design strategy to obtain oral degraders. For this purpose, we calculated a pool of seven previously ad hoc selected 2D descriptors for the 2258 publicly available degraders in PROTAC-DB (average values: MW= 972.9 Da, nC= 49.5, NAR= 4.5, PHI= 17.3, nHDon= 4.5, nHAcc= 17.7 and TPSA= 240 angstrom(2)) and compared them to a dataset of 50 bRo5 orally approved drugs. Then, a chemical space based on nC, PHI and TPSA was built and subregions with optimal permeability and bioavailability were identified. Bioavailable degraders (ARV-110 and ARV-471) tend to be closer to the Ro5 region, using mainly semi-rigid linkers. Permeable degraders, on the other hand, are placed in an average central region of the chemical space but chameleonicity could allow them to be located closer to the two Arvinas compounds.

Automated summary

The paper conducted a statistical study on the distribution of E3 ligases and POIs in PROTAC-DB, and focused on establishing a property-based design strategy for obtaining oral degraders. By comparing descriptors of publicly available degraders and orally approved drugs, a chemical space based on certain parameters was built to identify optimal permeability and bioavailability regions. Bioavailable degraders tended to be closer to the Ro5 region, while permeable degraders showed chameleonicity in their location within the chemical space.