Facilitating Access and Optimizing Patient Selection in Precision Oncology: Towards a Learning Healthcare System

Cancer treatment has undergone a major transformation over the past decades. Whereas therapy was traditionally based on tumor location and consisted largely of chemotherapy, surgery, and radiotherapy, modern oncology increasingly focuses on precision medicine. Precision oncology uses the genetic and molecular characteristics of a tumor to guide treatment decisions, enabling more personalized therapies that can improve outcomes and reduce unnecessary toxicity.

Two major breakthroughs have driven this shift: targeted therapies and immunotherapies. Targeted therapies inhibit specific oncogenic proteins essential for tumor growth, while immunotherapies stimulate the immune system to better recognize and attack cancer cells. These approaches are often most effective in patients whose tumors harbor specific biomarkers.

Despite its promise, precision oncology also introduces new challenges. Many new drugs are approved only for one cancer type, even though the same actionable biomarker may occur across tumor types. Patients may therefore benefit from therapies outside their registered indication, so-called off-label use, but gaining access to such treatments is often difficult. Moreover, off-label prescribing is frequently not accompanied by systematic data collection, limiting the ability of healthcare systems to learn which treatments are truly effective.

A second challenge is that as more biomarkers are identified, patient subgroups become smaller. This makes large randomized controlled trials increasingly impractical. New drugs are therefore often tested in single-arm studies without control groups, which can lead to uncertainty about real-world effectiveness, particularly because trial participants are often younger and fitter than the average patient. These limitations may delay reimbursement decisions and restrict patient access.

To address these barriers, two Dutch initiatives were established: the Drug Rediscovery Protocol (DRUP) and the Drug Access Protocol (DAP). DRUP evaluates whether patients with metastatic cancer, who have no remaining standard treatment options, can benefit from approved targeted or immunotherapies matched to their tumor’s molecular profile but used outside their original indication. Mandatory fresh biopsies and whole-genome sequencing allow identification of biomarkers associated with response or resistance, supporting improved patient selection.

This thesis highlights several key DRUP findings. In non-small cell lung cancer patients with MET mutations, crizotinib showed substantial activity, with tumor shrinkage in 62.5% and clinical benefit in 70.8%. Tumor-agnostic cohorts further demonstrated that cancers with high mutational burden may respond to immunotherapy combinations. Broader analyses revealed that tumors with multiple disrupted oncogenic pathways derive less benefit from single targeted agents, emphasizing the need for refined selection strategies and the potential role of combination approaches. Overall, DRUP has treated more than 1,600 patients, with about one-third experiencing clinical benefit, underscoring the value of structured frameworks for off-label precision treatment.

DAP complements DRUP by providing temporary access to newly approved drugs awaiting reimbursement while collecting real-world evidence in broader patient populations. In squamous cell skin cancer, DAP confirmed the effectiveness of cemiplimab in routine practice, supporting reimbursement in the Netherlands.

Together, DRUP and DAP demonstrate how systematic data collection can improve access to precision medicines, optimize patient selection, and support a learning healthcare system in oncology.