Platform trials for rare cancers—A complex innovation to accelerate knowledge

Rare tumor entities constitute a complex and heterogeneous group of tumors that differ considerably in terms of histological morphology, molecular genetic features, age predilection, response to therapy, and subsequent prognosis.¹ Conventional nonselected and personalized therapies that are adapted during the course of disease may increase treatment success and reduce side effects. However, especially in the field of rare cancers, there is little evidence-based clinical data for many entities justifying specific treatment decisions.

In addition to the infrequency of these diseases, the typical problems of conventional clinical trials lie in extended recruitment times and an inconsistency of the results, depending on the trial design. Making clinical trials more efficient and patient friendly is, therefore, a goal pursued by many institutions, including the European Patient-Centric Clinical Trial Platforms (EU-PEARL; https://eu-pearl.eu), but also individual trial platforms such as the NCT Master (https://www.nct-heidelberg.de/forschung/molecular-stratification/master.html) and GBM-AGILE,^2,3 to name but a few. This is usually achieved by setting up adaptive clinical trial platforms allowing for multiple interventions to be tested simultaneously, preferably against placebo or standard therapy comparison group. In an ideal world, these adaptive designs allow for easier recruitment of a sufficient number of patients. This also increases the likelihood that patients will be allocated to a promising treatment rather than a placebo, as placebo or standard therapies will only include the minority of patients and ineffective trial arms will be terminated early. An optimized approach would seamlessly enroll patients in such trials in a prospective, sequential, systematic, and unbiased manner to allow for rapid adaptation to changing biological and study evidence as well as treatment scenarios.

In their article,⁴ Dhaenens et al. present 2 proof-of-principle platform studies for patients with neurofibromatosis that were designed as part of EU-PEARL. EU-PEARL operates under the umbrella of the European Innovative Medicines Initiative (imi; http://www.imi.europa.eu) and is a strategic partnership between public and private sectors. It has developed a general framework for conducting patient-centered collaborative platform trials. The overall aim is to make clinical trials more efficient and to reduce the time it takes to develop new therapies. The concept of EU-PEARL is based on a clinical network of hospitals that form the environment for the clinical trials and a platform where different stakeholders, including patients, can come together and interact. EU-PEARL also aims to develop a comprehensive set of tools and methods for planning, conducting, and analyzing clinical trials.

The trials discussed in this issue of Neuro-Oncology Practice build upon this platform and were designed by a team of multidisciplinary neurofibromatosis and trial methodology experts and comprise an observation and a treatment period.⁴ Patients are randomly assigned to a range of available interventions that allow for additional interventions to be added during the study. Patients can switch from the observation arm to the intervention arm and back, making the design very flexible. The eligibility criteria and endpoints are adapted to the respective intervention. The platform thus executes a flexible basket design on a platform that enables maximum output through optimal knowledge acquisition from a relatively small number of patients.

We share the authors’ conclusion that platform studies such as the EU-PEARL-based neurofibromatosis proof-of-concept trials serve as a model for other rare diseases, as they increase the chance of identifying beneficial treatments through optimal learning strategies. Beneficial treatments can also be elucidated at a much faster rate and a lower cost than in traditional single-agent clinical trials. Another important point is that strongly committed alliances between the scientific community, third-party funders, industry, regulators, and patient representatives are the most efficient way to proceed. However, we caution that the structures are remarkably complex due to the multilevel organization of such studies and require clear governance structures, highly professional project management, and sufficient funding of the structures. In addition, regulatory authorities, which are not used to the concepts of basket or umbrella designs or platform studies, need to be convinced to support modern study designs. Furthermore, the regulatory landscape has been notoriously complex, at least in Europe until recently, and it remains to be seen whether the new Clinical Trials Regulation (https://www.ema.europa.eu/en/human-regulatory-overview/research-development/clinical-trials-human-medicines/clinical-trials-regulation), which aims to harmonize regulatory procedures in Europe, will accelerate clinical research in terms of complexity and speed.⁵ Moreover, funding structures need to be adapted for such a complex endeavor and additional funding should be made available beyond the “high-hanging fruit” of, for example, EU-based funding to enable transatlantic/transpacific and transnational funding especially in rare diseases. Multiple industry players will typically be involved in platform trials involving the use of pharmaceuticals. In addition to the governance, organization, regulatory, cooperation, and funding aspects, platform trials are complex from a logistical perspective and are likely to require a great deal of time and considerable resources to reach expert consensus, identify appropriate statistical designs, compose the trial, and allocate appropriate participating sites.

In the proof-of-concept studies discussed here, the authors use a sequential single-arm design without a randomized control or placebo group. As the EU PEARL proof-of-concept trials in neurofibromatosis are primarily aimed at testing agents that have the potential to induce objective tumor responses, this is likely to be less of an issue. However, this issue will become increasingly important once other types of agents or disease manifestations other than malignancy are included in the platform. Therefore, after proof of concept in the EU PEARL neurofibromatosis trials, it is of great importance to subsequently conduct randomized trials with robust study endpoints that consider both conventional endpoints such as overall survival and more modern patient-related endpoints, for example, health-related quality of life. This approach will ultimately combine the benefits of modern trial platforms and robust endpoints and thus accelerate the development of clinically meaningful treatments for patients with rare tumor entities and beyond.

Conflict of interest statement

None declared.