CRSC Meeting: Glioblastoma New Asset Strategy

• Most members had limited direct Phase 0 GBM experience.
• General consensus: Phase 0 can provide useful answer to exploratory or non-therapeutic questions PK/PD and BBB penetration data but often overlaps with Phase I objectives.

• The group discussed the importance of obtaining robust pharmacokinetic (PK) and pharmacodynamic (PD) data, especially regarding blood-brain barrier (BBB) penetration and target engagement.
• There was a focus on whether a phase 0 (microdosing) study would provide sufficient PK/PD and BBB penetration data, or if these objectives could be better addressed within a phase 1 trial.
• Suggested focusing on defining the key scientific question first (e.g., proof of mechanism, PK/PD, biomarker validation).
• Several members felt a small Phase I with a run-in cohort may be more efficient than a standalone Phase 0.
• The team considered the challenges of measuring drug levels in tumor tissue versus plasma, and the technical and ethical complexities of obtaining tissue and fluid samples (e.g., CSF taps, biopsies).
• The need for biomarker validation and the use of advanced imaging techniques (such as PET and volumetric MRI) were discussed as part of the strategy to assess target engagement and drug distribution.
• The group agreed that the primary objectives should include:
  • Demonstrating BBB and blood-tumor barrier penetration of the new asset drug.
  • Collecting PK/PD data to inform dosing and safety.
  • Validating biomarkers and imaging methodologies to assess target engagement.
  • Ensuring that the study design balances scientific rigor with operational feasibility and patient safety.

The most significant regulatory hurdles for phase 0 trials discussed were:

• The need for a comprehensive and robust preclinical data package.
• Potential flexibility in animal study requirements due to disease severity but still needing clear justification.
• Addressing safety concerns related to the novel aspects of the drug formulation, MOA or delivery system.
• Ensuring that ethical and operational challenges are considered, as they may impact regulatory acceptance.

• Surgical Timing & Biopsy Approaches:
  • The group discussed the logistical and technical challenges of obtaining tumor tissue in glioblastoma (GBM) studies.
  • Timing is critical: Pre-treatment and post-treatment tissue access can be difficult to coordinate, especially since GBM is a fast-progressing disease.
  • There are challenges in aligning neurosurgical procedures with study timelines to ensure relevant and high-quality samples.
• CSF (Cerebrospinal Fluid) Sampling:
  • The team considered different strategies for CSF collection, such as performing taps versus leaving an intrathecal reservoir in place for repeated sampling. (see questions 6,7 and 8)
  • They discussed the optimal frequency and volume for CSF collection, aiming to balance the need for reliable pharmacokinetic data by minimizing patient discomfort and risk. (see questions 6,7 and 8)
• Ethical and Operational Considerations:
  • Ethical concerns were raised about the invasiveness of repeated tissue and CSF sampling, but it was noted that patients and IRB and IBC support more invasive procedures in severe diseases like GBM.
  • Operational challenges, such as scheduling and coordinating with neurosurgical teams, were highlighted as significant factors in trial design.
• Biomarker and Imaging:
  • The use of advanced imaging (e.g., PET, volumetric MRI) and robust pharmacodynamic assays were discussed as complementary or alternative methods to direct tissue sampling for assessing target engagement and drug distribution.

• Possible flexibility in GBM due to urgent need, but regulators will scrutinize safety, preclinical package completeness, and rationale for microdosing.
• Recommended early FDA interaction (e.g., pre-IND/INTERACT) to clarify requirements.
• When the Scientific Question Doesn’t Require Phase 0:

  If there isn’t a specific, unanswered scientific or mechanistic question (such as the need for detailed PK modeling or BBB penetration data) that only a phase 0 study could address, it may be more efficient to move directly to phase I.

• When Preclinical Data Is Robust:
  If the available preclinical (nonclinical) data package is strong and provides sufficient evidence of safety and rationale for dosing, a phase I trial can be justified without the need for an exploratory phase 0.

• To Accelerate Development Timelines:
  If time is a critical factor (e.g., limited funding runway, urgency to reach clinical endpoints), skipping phase 0 can expedite the clinical development process, especially if phase 0 would not provide significant additional value.

• When Operational Complexity Outweighs Benefits:
  If phase 0 adds complexity, cost, or patient recruitment challenges without clear benefit, it may be preferable to incorporate key objectives (like PK/PD or tissue sampling) into the phase I design.

• If the Intervention Is Not Novel Enough to Warrant Phase 0:
  For drugs or delivery systems that are not highly novel (e.g., known chemotherapy agents with established safety profiles), phase 0 may not be necessary unless there is a unique mechanistic question.

• When Regulatory Feedback Supports Phase I:
  If regulatory agencies (e.g., FDA) indicate that the preclinical package is sufficient for phase I, or if a pre-IND meeting clarifies that phase 0 is not required, sponsors can proceed directly.

• While the group discussed the pros and cons and practicalities of using brain portal systems, there was no explicit statement from any attendee detailing their personal hands-on experience with Ommaya reservoirs, EVDs, or newer portal technologies in GBM trials.
• The discussion was informed and practical, suggesting familiarity with the concepts and challenges, but did not include specific anecdotes or case studies from the participants.
• Mixed views: reservoirs add risk and may not be justified unless frequent serial sampling is required.
• Standard lumbar puncture is often sufficient.

• The team deliberated on the optimal frequency and volume for CSF collection, aiming to balance the need for reliable data by minimizing patient discomfort and risk.
• There was consensus that repeated CSF sampling can be invasive and burdensome, so protocols should be designed to minimize both the number of collections and the volume drawn at each time point.
• The group emphasized the importance of patient safety and comfort, and the need to justify the sampling schedule to both ethics committees and patients.
• Members agreed that between 8-15 mL of CSF every 8 weeks was tolerated with minimal side effects
• Biopsies are feasible but invasive; frequency must balance data needs and patient burden.
• CSF tapping every 8 weeks is generally acceptable; some studies with shorter duration did CSF taps every 4 weeks. Reservoirs/portals may not be necessary.

• The members with experience in GMB studies did not have experience with an intrathecal reservoir claiming in their clinical trial they were able to perform at least 4 CSF taps and 2 biopsy (try to reduce the number of biopsy by using existing biopsies for baseline – verify TMZ treatment)

• The most critical regulatory considerations when designing trials that incorporate brain portals for ctDNA collection are ensuring robust preclinical safety data, device approval, risk-benefit justification, comprehensive informed consent, ethical oversight, operational feasibility, and early regulatory engagement. These considerations are essential to protect patient safety and ensure regulatory compliance.