Positron emission tomography imaging of drug concentrations in the brain

See the article by Jucaite et al in this issue pp. 687–696.

Many factors contribute to the lack of progress in developing more effective therapies for brain tumors. These include redundancy of signaling pathways, tumor heterogeneity, suppressed immune environment, and the blood-brain barrier (BBB), among other factors.¹ The BBB, in particular, has been a major impediment to progress by preventing the majority of anticancer therapies (>90%) from achieving therapeutic concentrations within brain tumors.^2,3 Frequently, pharmaceutical companies develop therapies that are designed not to penetrate the BBB to reduce unwanted neurotoxicity. In addition, there are often limited data regarding the ability of agents to pass through an intact BBB. While preclinical models have often been used as a surrogate to evaluate the ability of agents to cross the BBB, the reliability of these models has been questioned. To determine whether an agent adequately crosses the BBB surgical “window-of-opportunity” trials are frequently performed. In these studies, the study drug is given for a period of time to patients with recurrent tumors requiring reoperation, and the tumors are then resected and drug concentrations within the tumor analyzed in both enhancing and non-enhancing areas.^2,4 These studies can be difficult and time-consuming and often delay development by several years. In addition, only the total drug concentrations are usually measured. Approaches to determine the more relevant free drug concentrations using techniques such as microdialysis or modeling using techniques such as the physiologically based pharmacokinetic approach⁵ can be difficult to perform. There is a critical need to develop easier, reliable, and widely available approaches that accurately evaluate the ability of novel agents to cross the BBB, achieve adequate concentrations within brain tumors, and inhibit the targeted pathways.

One potential strategy to more efficiently evaluate the ability of novel agents to cross the BBB is to radiolabel drugs and evaluate their concentrations in the brain using high-resolution positron emission tomography (PET). Recently, there has been renewed interest in targeting the DNA damage response pathways to augment the antitumor activity of radiotherapy and chemotherapy. In this issue of Neuro-Oncology, Jucaite and colleagues evaluated the ability to cross the BBB of AZD1390, a small molecule inhibitor of ataxia-telangiectasia mutated (ATM) protein kinase, which mediates cellular response to DNA damage induced by radiation therapy.⁶ They radiolabeled AZD1390 with carbon-11 and used microdose PET to evaluate the exposure of this agent in the brains of eight normal healthy subjects with an intact BBB. The brain radioactivity concentration of [¹¹C] AZD1390 was 0.64 SUV (standard uptake value) and reached a maximum of 1.00% of the injected dose at T_max (brain) of 21 min (time of maximum brain radioactivity concentration) after intravenous injection. The whole-brain total distribution volume was 5.20 ml × cm⁻³. Using this approach, they were able to demonstrate that [¹¹C] AZD1390 crosses the intact BBB, supporting the development of the agent in combination with radiation therapy for patients with glioblastomas and brain metastases (NCT03423628). It also offers the potential of using PET microdosing to predict and guide dose range and schedule for subsequent clinical studies.

Despite interest in these types of studies, they have only very rarely been used to evaluate the ability of novel agents to cross the BBB in neuro-oncology. One major obstacle is the effort and cost required to radiolabel novel agents and the need to obtain an Investigational New Drug (IND) application from the Food and Drug Administration. To date, most pharmaceutical companies have had limited interest in developing agents for brain tumors, which have traditionally been considered to be a small and difficult market. However, the growing interest in developing therapies for brain metastases may help change this mindset, as would streamlining the regulatory process. There are also challenges in evaluating the results of these neuroimaging studies. The assessment of an agent’s ability to cross the BBB in normal brains used in the study by Jucaite and colleagues may be the cleanest approach. Evaluation of tumor concentrations of radiolabeled agents in patients with brain tumors can sometimes be difficult to interpret. For instance, a study evaluating the epidermal growth factor receptor targeting monoclonal antibody ch806 included single-photon emission computed tomography (SPECT) studies which showed increased uptake of radiolabeled drugs within brain tumors and were interpreted as indicating that the drug crossed the BBB.⁷ This provided some of the basis for subsequent phase II and III trials of the antibody-drug conjugate depatuxizumab mafodotin, derived from this antibody for the treatment of patients with glioblastomas. Unfortunately, these trials showed no therapeutic benefit and ongoing preclinical studies suggest that this agent does not readily cross an intact BBB. The SPECT studies showing uptake of the drug in tumors may have simply reflected the passage of the agent across disrupted BBB within the enhancing tumor mass. Potentially advanced PET imaging and detailed evaluation of drug uptake in enhancing and non-enhancing areas of tumor may provide more reliable assessment of an agent’s ability to cross the BBB.

It should be noted that even when therapeutic drug concentrations can be measured this may not necessarily translate into adequate inhibition of the targeted molecular pathways.⁸ Conversely, an agent with low plasma-to-brain drug concentration ratios may still achieve therapeutic tumor concentrations if it is potent and adequate doses can be administered.⁹ Detailed evaluations of the in vivo pharmacodynamic effects are also needed. Currently, these studies are conducted on tissue obtained by surgery in window-of-opportunity trials in patients with recurrent tumors.² Since a prior biopsy of untreated tumor for comparison cannot be easily obtained, tumor from the initial surgery or from untreated patients are used instead and increases the difficulty of evaluating the results of these pharmacodynamic studies. Fortunately, there is increasing interest in using PET imaging to evaluate inhibition of targeted signaling pathways, proliferation, and apoptosis, as well as evaluating the effects on the tumor immune environment.¹⁰ These approaches, perhaps complemented by imaging of drug concentrations by PET, will hopefully accelerate the development of therapies for brain tumors and obviate the need for the cumbersome and time-consuming “window-of-opportunity” surgical trials that are currently being performed.

Conflict of interest statement. Research was supported by Astra Zeneca.