Pharmacology and drug development

The aims of the Pharmacology and Drug Development Group (PDDG) are to optimise the pre-clinical development and science-led clinical application of novel therapies, including ‘first into man’ (phase I) studies.

We use pre-clinical model systems to inform the early clinical development of novel agents and identify interesting drug combination strategies. The PDDG is closely linked with the Early Phase Clinical Trials Team (EPCTT), also led by Duncan Jodrell, and the hepato-pancreatic-biliary (HPB) trials team, based in the Cambridge Cancer Trials Centre managing clinical trials in patients with pancreatic cancer, allowing us to implement our laboratory findings in the clinic.

In the laboratory, we generally use model systems representing pancreatic cancer, which complement our clinical links: Duncan Jodrell and Bristi Basu are members of the medical team at Addenbrooke’s Hospital, which cares for patients with pancreatic cancer, and Duncan Jodrell is also the Director of the Cambridge Pancreatic Cancer Centre. Pancreatic cancer is a major unmet clinical need and a priority cancer for Cancer Research UK. We are developing 2D and 3D co-culture models of pancreatic cancer, as the initial testing platforms for novel therapeutics. We also have access to the KPC GEM (genetically engineered mouse) model of pancreatic cancer. Using a 2D co-culture assay system (KPC pancreatic cancer cells and mouse fibroblasts), we have identified that cancer-associated fibroblasts (CAFs) confer resistance to gemcitabine in the cancer cells in vitro, whereas fibroblast-like normal pancreatic stellate cells do not. We are now investigating the mechanism of the induction of resistance using candidate approaches and RNA sequencing. We are also investigating whether 3D co-culture creates even greater resistance to gemcitabine. We will use these assay systems to identify drugs that overcome the resistance to gemcitabine, and investigate their utility as combination therapies for the treatment of patients with pancreatic cancer. We are continuing to assess the modulation of gemcitabine delivery to tumour tissue in various combination treatment regimes through both local (Neesse et al., Gut 2013; Epub 25 Sept; Neesse et al., PNAS 2013; 110: 12325) and international collaborations. In addition, we are studying how newly-discovered gemcitabine metabolites might impact on its anti-tumour activity. We are also investigating the DNA damage and cell signalling responses to gemcitabine in vitro and in vivo, in the presence and absence of cell cycle inhibitors. These data will be used to model the relevant signalling pathways, to identify new targets to enhance the activity of gemcitabine in vivo (Figure 1).

Figure 1: Abnormalities in mitosis induced by incubation of pancreatic cancer cells (MIA PaCa-2) for 24 hours with the S-phase active drug gemcitabine, in combination with a cell cycle checkpoint inhibitor. The four examples show failure of all the chromosomes to align and segregate correctly at the metaphase plate. The mitotic spindle is shown in red (alpha-tubulin stain) and the chromosomes are shown in blue (DAPI). Yellow arrow heads indicate examples of chromosomes that are not correctly aligned.

Our pre-clinical work often involves the assessment of combination strategies. We are taking two new approaches to evaluating those pre-clinical data. We are using mathematical models of the cell cycle, receptor/ligand interactions and the spindle assembly checkpoint to guide the pre-clinical studies we perform. We also use model based approaches to evaluate pre-clinical growth inhibition data and identify potentially synergistic ‘dose’ ratios of compounds by generating surfaces of interaction. We believe that current clinical trial design for evaluating new drug combinations may lead to missed opportunities, unless these pre-clinical data are used to guide trial design. We think that we should be trying to identify synergistic interactions in the clinic, as opposed to simply trying to combine the maximum tolerable doses of both agents, when used as single agents. In collaboration with the MRC Trial Methodology Hub (Adrian Mander and colleagues), we have reviewed adaptive designs for dual agent dose escalation studies (Harrington et al., Nat Rev Clin Oncol 2013; 10: 277) in the clinic and are developing novel Phase I trial designs that will be informed by our pre-clinical studies.

An example of the application of these combination approaches is the evaluation of a novel inhibitor of the cell cycle regulator CHK1, in combination with gemcitabine (in collaboration with Sentinel Oncology). We have observed synergy in pancreatic cancer cell lines in vitro, and the effect is now being investigated in mouse models of pancreatic cancer. We have also used these models to investigate the activity of capecitabine alone and in combination with gemcitabine (Courtin et al., PLoS ONE 2013; 8: e67330).

In general, it is assumed that combinations of agents have similar effects on normal and tumour cells, but this is not always the case. An optimal combination would lead to synergy in cancer cells and antagonism in normal cells, reducing the toxic side effects that often limit dosing. In studies with normal myeloid precursors (CFU-GM) and other diploid cells (e.g. IMR90 fibroblasts), we have demonstrated previously that the synergistic effects of combining an AK-A inhibitor and paclitaxel are not seen in non-malignant cells. This project is also utilising a mathematical model of the spindle assembly checkpoint to predict drug effects, through collaboration with Bob Jackson (Pharmacometrics Ltd). We ultimately intend to extend our pre-clinical findings into clinical trials.

In a new collaboration with AstraZeneca and Simon Cook (Babraham Institute), we have initiated studies to identify potential combination strategies, including the MEK inhibitor, selumetinib, using pancreatic cancer models. This will involve both a targeted approach, expanding on the Cook lab’s published work (Sale and Cook, Biochem J 2013; 450: 285) using a combination of MEKi and the pro-apoptotic agent ABT263 and a broad, medium throughput screening approach.

As a result of our collaboration with Steve Ley and Rebecca Myers (Department of Chemistry) and Fanni Gergely (CRUK CI), we have synthesised and evaluated biologically the first selective  inhibitor of the kinesin motor protein HSET (CW069) (Watts et al., Chem Biol. 2013; 20: 1399), which induces  phenotypic changes demonstrated previously to be related to HSET knockdown. We hope that  further synthetic efforts may facilitate the identification of a candidate molecule for pre-clinical development.

Our collaboration with Doug Fearon (CRUK CI) investigates two diverse features of pancreatic adenocarcinoma (local immunosuppression in the tumour and cancer associated cachexia) and their link to a particular cell lineage found in tumour cells and skeletal muscle. As outlined in a recent publication (Feig et al., PNAS 2013; 110: 20212), we have shown that administration of an inhibitor of CXCR4 (AMD3100), may overcome local immunosuppression in pancreatic adenocarcinoma, opening up new opportunities for immunotherapy in this disease, where agents active in other cancers (e.g. anti-CTLA4 and anti-PD1 in melanoma) have shown no activity in patients with pancreatic cancer. We initiated a clinical study with AMD3100 in early 2015.

Currently, the EPCTT is supporting 12 experimental medicine studies that are actively recruiting patients; five combination phase I trials, a further six single agent phase I trials and a biomarker study. We are continuing to explore novel PET and MR approaches in our trials and look forward to initiating a series of trials supporting the clinical development of hyperpolarised 13C pyruvate based PD studies (in collaboration with Kevin Brindle and Ferdia Gallagher).