PET imaging of ATM using 18F-ATMi

The integrity of the genetic material, a cell's DNA, is extremely important for an organism's survival. It is therefore no wonder that all cells possess a whole range of enzymes that detect and repair DNA damage. This DNA damage can be the accidental result of UV radiation of e.g. the skin, toxins from tobacco smoke, or the process of tumour formation, but can also be the result of purposely inflicted damage, such as during radiotherapy or chemotherapy of tumours. Because DNA damage repair is such a significant factor in tumour treatment and tumour formation, it presents a promising target for developing novel drugs in the fight against cancer that would target these mechanisms, and increase tumour cell DNA damage, and as a result, tumour cell kill.

ATM is an important enzyme involved in DNA damage repair, and several drugs have been developed to prevent ATM from performing its function, activation of other DNA damage proteins. This causes unprepared DNA breaks and, eventually, cancer cell death. However, it is still difficult to predict which patients will respond to the drug, and no markers are know to help with this. It is not always known whether the drug is actively pumped out of the tumour, or will even reach its target.

To address these concerns, we will generate a radiolabelled ATM inhibitor, which we will call 18F-ATMi. Here, we replaced a fluorine atom in the chemical structure of a known ATM inhibitor with a radioactive version, or isotope, fluorine-18 (18F). The radioactive emission from 18F can be measured using a dedicated medical scanner called a PET scanner. It can visualise the three-dimensional distribution of 18F in the body, with only the need of a single intravenous injection of the radiolabelled olaparib. PET imaging using 18F-ATMi will show exactly where the drug travels, whether and how much it is delivered to a tumour, and how fast it is removed. We therefore aim to develop 18F-ATMi as a novel imaging agent for PET imaging.

In addition, the ability for inhibitors to bind the target, the ATM enzyme, may increase after radiotherapy. We therefore also aim to use 18F-ATMi for visualising the biological effects of radiotherapy, to determine if the dose of irradiation will be sufficient to kill the tumour, or if changes need to be made to the therapy plan.

Given the increasing number of cancer drugs exploiting DNA damage repair (DDR) defects in tumour tissue, such as ATM inhibitors, there is a clear need for patient stratification, therapy prediction and evaluation tools. Previously, our group has developed several methods for visualising DNA damage and genetic instability (a key hallmark of cancer), based on medical imaging techniques used in routine clinical practice, such as positron emission tomography (PET), targeting DDR proteins such as yH2AX and PARP.

ATM is a member of the PIKK family involved in cell cycle regulation and DNA double strand break (DSB) damage recognition and repair. Upon ATM activation it phosphorylates a number of proteins involved in cell cycle checkpoint control, apoptotic responses and DNA repair, including p53, Chk2, BRCA1, and H2AX. To date, no companion biomarkers exist that predict the pharmacokinetic/dynamic behaviour of ATM inhibitors in patients. Such a tool would nonetheless be exceedingly useful to select patients that may or may not respond to the treatment.

Therefore, to a known ATM inhibitor (ATMi), we have tagged a radioactive isotope, 18F, while retaining its original chemical structure, to allow visualisation of olaparib biodistribution and ATM expression using a PET scanner. We hypothesize that PET imaging with radiolabelled ATMi can be used to determine the response to chemo- and radiotherapy of cancers. Molecular information about the response to treatment will allow much faster decision making regarding treatment, compared to the anatomical changes such as tumour shrinkage that is used routinely. Also, 18F-ATMi may, in future, be used as an early diagnosis system, detecting the increased genomic instability that accompanies tumorigenesis. Although here we are using PDAC and glioma cells in vitro and in vivo as a model system, given applications of PARP inhibitors in many other tumour types, the approach may be expanded to other disease sites.

18F-ATMi (a) Demonstrates use of a novel target for PET imaging, and its first application in a biological system; (b) for the first time, allows to study the distribution of an ATM inhibitor in a non-invasive fashion, in vivo; (c) Provides a tool for rapid pharmacodynamic study of novel competitive ATM inhibitors; (d) Provides a tool to study the biological effects of radiation therapy and other genotoxic insults in vivo, in a non-invasive fashion.
For patients: Radiolabelled ATMi will allow direct measurement of ATM expression and possibly activation in patients, enabling improved patient stratification, quantification of target engagement of therapeutic inhibitors, and will allow non-invasive monitoring of treatments, including radiation therapy. Radiolabelled ATMi allows the study of pharmacokinetics, pharmacodynamics, prediction of ATM inhibitor therapy, and, most importantly, prediction of olaparib delivery, a non-trivial point in highly fibrotic PDAC tumours, or in the tumour margins of glioblastomas or in brain metastases, where blood-brain-barrier may remain intact.
For drug development (academic and large pharmaceutical companies): this compound would allow non-invasive assessment of target engagement of novel, improved ATM inhibitors.
For the healthcare system: 18F-ATMi will allow triage of patients that will not benefit from ATM inhibitor treatments, and avoid unnecessary, yet expensive, treatment. It will also allow rapid decision-making for non-effective therapies, avoiding unnecessary side-effects and treatment delays.
Although we are using in vitro and in vivo models of mostly PDAC and lung tumours in this project, ATM inhibition may be used in colorectal, brain, ovarian and breast cancer, and would be equally valuable in those disease settings.