The discovery of CCS1477, first of its kind, p300/CBP inhibitor

20 Feb 2020

CCS1477 is first of its kind, a small molecule inhibitor of the conserved bromodomain of twin proteins, p300 and CBP. Small molecule inhibition of twin histone acetyltransferases not only promotes the turnover of key cancer signalling proteins (e.g. androgen receptor and androgen receptor splice variants), it also prevents the transcription of key cancer-driving genes (e.g. c-Myc and IRF4).

CCS1477 is currently in Phase 1 clinical studies for the treatment of late stage, castration resistant prostate cancer, as well as more recently, in a parallel study evaluating the treatment of certain haematological malignancies (multiple myeloma, non-Hodgkin’s lymphoma and acute myeloid leukaemia). Longer term, it can be applied to cancers with certain specific genetic drivers.

CellCentric developed the compound all the way from initial small molecule hit through to initial safety and efficacy evaluation in patients. The company operates with a core team working with a network of expert contractors, leveraging the best expertise wherever it may be. This blog describes the evolution of discovery of CellCentric’s pioneering drug.

CellCentric started as a spin out of the University of Cambridge, investigating epigenetic processes and their potential application to modifying cell function and fate. The initial business model was based on aggregating know-how and then parcelling up opportunities to be taken on by pharmaceutical companies. As the field matured, this model no longer had a clear market, so CellCentric pivoted, initiating and driving forward its own drug discovery programmes. From over 50 targets examined, the company worked actively on 7 before focussing wholly on p300/CBP.

The link between epigenetic-related control mechanisms, androgen receptor signalling and prostate cancer emerged relatively early in the company’s research. Initially CellCentric concentrated on another potential drug target to impact this pathway, USP22. This proved difficult to fully drug properly. Related research did however, highlight that focusing on p300/CBP may offer an alternative route forward to the same goal.

There are different ways to inhibit p300/CBP, with at least three pockets on the twin proteins that can be targeted. The catalytic HAT domain has proven hard to inhibit with good selectivity, with a wide range of biological activities affected. Other groups have tried focusing on the CH1 domain, but so far have not been able to progress an inhibitor blocking this region to the clinic. CellCentric chose to target the bromodomain pocket, that seemed to have more subtle effects when inhibited, but can still drive strong anti-tumour responses (see previous blog).

Early p300 small molecule compound hits were moderately potent and selective for the p300/CBP bromodomain, but demonstrated somewhat weak (micromolar) activity in cells. These compounds were however, able to confirm anti-tumour activity in cellular models of prostate cancer, both in terms of impacting cell proliferation, as well as inhibition of the androgen receptor pathway. They provided early reassurance to the overall approach.

Building a computational homology model of the p300 protein was key for CellCentric’s early progress, enabling the team to identify regions of these initial small molecules docked into p300 that could be modified to improve bromodomain binding. In parallel, in vitro metabolic profiling was undertaken using liver microsomes to identify sites of metabolic instability of the initial small molecule inhibitors. Combining the two strands allowed the best hits to be progressed for further optimisation; improving binding potency whilst also building in good metabolic stability.

X-ray crystal structures were then obtained of lead molecules bound into p300 protein. These confirmed the binding hypothesis generated using the computational model, and also identified further opportunities for prototype drug improvement. This was augmented with additional x-ray crystal structures of lead compounds bound into BRD4, the most closely related bromodomain to p300/CBP. These guided further optimisation, with the team actively designing in elements that would impede binding into BRD4 (and the wider bromodomain family of proteins), to maximise selectivity to p300/CBP.

The emerging chemical series which ultimately contained CCS1477, contained a chiral centre. Separation of the respective enantiomers demonstrated preferential binding of the (S) enantiomer into the p300/CBP bromodomain pocket. This fitted with the knowledge gained from the protein x-ray structure work. The improved binding of the (S) form was also accompanied by a significant increase in the cellular activity.

Further permutations of optimised lead compounds provided a panel of candidates which were highly potent, selective and active in cells. Wider profiling also confirmed no significant off target activities against panels of other bromodomains, kinases, enzymes, receptors and ion channels. This provided increased confidence that any activity observed in disease models would be specifically related to the inhibition of p300 and CBP.

In vivo pharmacokinetic studies showed the chemical series to be orally bioavailable with low to moderate plasma clearance across several species. Profiling the best compounds in mouse xenograft studies of late stage prostate cancer (models in which current marketed agents fail to work) indicated excellent tumour growth inhibition with complete tumour stasis achievable at well tolerated doses.

Pooling and reviewing the data of all the final panel of compounds, CCS1477 was selected as the molecule with the best overall efficacy, safety and developability profile. It was the one then progressed into full pre-clinical qualification, drug manufacture scale up and on to clinical studies.

At this point breadth of efficacy studies were carried out, where CCS1477 was tested against panels of hundreds of cancer cell lines representing different tumour types. These showed that the compound was not generally toxic, but that other tumours additional to prostate cancer were highly sensitive to p300/CBP bromodomain inhibition. This included AR+ driven prostate cancers, as would be expected. But it also included a strong number of haem cancers. Subsequent research confirmed that as well as knocking down c-Myc, CCS1477 also impacts IRF4 expression, key in a number of haematological malignancies. Follow on evaluation in xenograft models of multiple myeloma and acute myeloid leukaemia confirmed the compound’s efficacy, including causing tumour regression (shrinkage).

Identification and development of CCS1477 was only possible because of the strong, effective partnerships CellCentric has had with multiple contract research organisations, notably Sygnature Discovery, Proteros, Xenogesis, Axis Bio Discovery Services, Crown Bioscience, HD Biosciences and Cyprotex, to name just some. The Discovery Team worked as one, with combined enthusiasm and commitment, working towards a common goal. Thank you!