With newer systemic oncology medicines beginning to demonstrate more activity within the brain, clinical trial designs need to be optimized to appropriately include patients with brain metastases and gather more reliable efficacy data, according to Response Assessment in Neuro- Oncology Brain Metastases, or RANO-BM.
While patients with active central nervous system disease are often excluded entirely from clinical trials of solid tumors, the group’s new trial guidelines provide recommendations and scenarios for when they should be included early in the clinical development process.
Previously, denying patients with brain metastases from clinical trials would discount half to two-thirds of potential late-stage populations. The researchers also cited a survey of over 400 clinical trials that showed only 41 percent enrolled patients after they had received CNS-specific treatment.
The RANO-BM guidelines also explored the limitations of retrospective studies of CNS efficacy data.
“Although CNS radiological response in a small number of patients might be encouraging, the conclusion that a drug has so-called CNS activity should not be reached without additional data,” the group wrote. Specifically, retrospective studies can overestimate the impact of a therapy, after characterizing CNS disease as non-target lesions.
“With so many patients with active CNS disease frequently excluded from clinical trials, direct evidence of the CNS activity of a drug cannot reliably be obtained,” the authors wrote, adding that information about a drug’s potential activity must be considered at the forefront of trial design.
The guidelines, published in the January 2018 issue of The Lancet Oncology, describe three scenarios: when study agents are considered likely to demonstrate CNS antitumor activity; when that is unlikely; and when their potential is unknown.
If a drug is considered very likely, the study’s goal should be to generate a robust CNS efficacy signal as well as a systemic signal. This would include mandatory baseline CNS imaging in all patients, and conducted as frequently as systemic imaging, the guidelines said. Sponsors should also clearly specify CNS-related endpoints separately from overall or extra study endpoints.
Drugs that fall under this scenario should have prospective or retrospective data showing CNS responses with the same dose or schedule in the same subtype of cancer. In addition, data from trials demonstrating improvements in non-progression-related CNS endpoints would be helpful.
Depending on the strength and reliability of previous evidence, developers can consider a step-up inclusion design for their clinical trials, moving from including asymptomatic patients to symptomatic, after gathering preliminary efficacy data.
For drugs that are unlikely to be effective against brain metastases, the aim is to minimize risks to patients with active disease, while maintaining the largest possible population of patients eligible to participate in the trial.
Here, baseline CNS imaging could be reserved to primary cancers with a high risk of brain metastases. Re-imaging can be limited to clinically indicated patients, although the guideline recommends follow-up CNS imaging at a lower frequency than systemic imaging.
Patients with treated, stable or no CNS disease can be included, as long as their previous therapy was completed at least one month prior to enrollment, and they display no evidence of radiographic progression. In addition, any CNS disease should be asymptomatic for two weeks.
Sponsors should work to determine any unknown CNS potential as quickly as possible, the guideline said, and to leverage Phase I development, including substudies for patients with untreated metastases, treated at the recommended Phase II dose or schedule.
Later-phase studies could include an early CNS cohort with measurable, untreated, asymptomatic disease. Protocols could be modified as data on CNS potential is gathered to inform future development paths, they wrote.
Broadening clinical trials eligibility criteria has been a work in progress, with some describing it as a necessary culture change.
A recent initiative from the FDA, the American Society of Clinical Oncology and Friends of Cancer Research explored templates for clinical trial protocols that could help increase accrual without diluting efficacy.
In addition to brain metastases, the project tackled other traditional areas of exclusion criteria, including HIV and AIDS, organ comorbidities, age requirements and prior and concurrent malignancies. The templates were developed by a panel of agency representatives, government scientists, academic researchers and members of the pharmaceutical industry.
“We’ve all become comfortable and change has to start someplace — not white papers,” said Edward Kim, chair of solid tumor oncology and investigational therapeutics at the Levine Cancer Institute, during 2016’s launch of the project at the Friends of Cancer Research annual meeting.
Kim described how much of today’s exclusion criteria are seen as boilerplate — grandfathered in from previous trial protocols — and that a one-size-fits-all approach is not appropriate in clinical trial design.
That group suggested that patients with treated or stable brain metastases for four weeks should not be routinely denied by exclusion criteria, with clearly defined safety exceptions. Their full recommendations were published in the Journal of Clinical Oncology in October 2017.
Of the approximately 300 commercial IND protocols submitted to FDA in 2015, 77 percent excluded known, active or symptomatic brain metastases, while 47 percent allowed treated or stable metastases, they cited. In addition, only 1.7 percent of trials allowed patients with stable HIV.
The researchers found that expanding eligibility in brain metastases can be justified and may accelerate overall clinical development, and recommended that patients with brain metastases should be included in trials where metastases are common in the intended population.