The FDA has finalized guidance on the use of metastasis-free survival as an end point in clinical trials for nonmetastatic castration-resistant prostate cancer.
The FDA has finalized guidance on the use of metastasis-free survival (MFS) as an end point in clinical trials for nonmetastatic castration-resistant prostate cancer (CRPC).1
Nonmetastatic CRPC is defined by rising prostate-specific antigen (PSA) despite levels of testosterone without radiographic evidence of metastatic disease. Patients continue to experience rising PSA following local therapy, subsequent salvage approaches, and subsequent androgen deprivation therapy. Following the detection of rising PSA levels, patients with nonmetastatic CRPC can have a prolonged disease course.
“Such a prolonged assessment period (in which patients may receive multiple therapies) with low death rates may make the use of overall survival [OS] impractical as a primary end point to support approval of products in this disease setting,” the FDA wrote in the final guidance.
In 2011, the regulatory agency’s Oncologic Drug Advisory Committee met to further discuss these challenges. During these conversations, the committee conceded that certain end points like MFS, which can be measured earlier in the disease course could help evaluate the impact of products in patients with nonmetastatic disease.
The committee noted “that the transition from nonmetastatic CRPC to radiographically detectable metastatic disease is a clinically relevant event that can be associated with morbidity and the need for additional medical interventions.”
They noted that conversely, local progression events could be treated with local approaches, may never progress to distant disease, and may not result in systemic morbidity. As such, “a large treatment effect on MFS with an acceptable safety profile could demonstrate clinical benefit and support product approval,” the committee wrote.
The committee stated that sponsors for clinical trials should consider several points when using MFS as an end point in trials evaluating options for nonmetastatic disease. Sponsors should establish the definition of MFS prior to study start and stratification of randomization by previous local definitive therapy and by PSA doubling time should be considered.
Moreover, protocol for the trial should prespecify strategies to reduce attrition in both arms caused by patients who withdraw from the research due to concern about consistently rising PSA values. The guidelines suggest that a superiority trial design be utilized, that trials exclude patients who could derive benefit from local therapy to the prostate or pelvis, and entry criteria should include and justify the definition of castration-resistant disease.
Sponsors should prespecify imaging modalities and assessment frequencies and the same approaches should be utilized in the same patients for the entire trial. Moreover, regarding trial entry criteria, sponsors should also prespecify and justify the radiographic definition of nonmetastatic disease, as well as the radiographic definition of local disease or local progression and metastatic disease.
“For MFS to be interpretable, the expected magnitude of improvement in MFS should be substantially greater than the imaging frequency,” according to the guidance.
To evaluate any possible assessment bias, the FDA recommends that a blinded independent central review of imaging studies be performed. An audit can also be considered by sponsors. For the audit, a random sample of scans are sent for independent review.
According to the guidance, interim efficacy analyses of MFS are discouraged, as they may result in either an overestimation or underestimation of the magnitude of MFS improvement.
“The acceptable magnitude of improvement in MFS required to support drug approval will depend primarily on the trial design, toxicity profile, enrolled population, and overall benefit-risk assessment,” according to the guidance.
Sponsors are recommended to conduct a formal interim analysis of OS at the time of the final MFS analysis. For a positive benefit/risk assessment, the analysis needs to showcase a favorable numeric trend and provide evidence that OS is not adversely impacted by treatment. Moreover, to further inform the benefit/risk assessment, sponsors are permitted to collect patient-reported outcome and other clinical outcome assessment findings.
The guidance recommends that sponsors should describe the methodology for evaluating, measuring, and analyzing MFS in the trial protocol. To minimize missing data, which can complicate evaluating MFS, sponsors should prespecify in the protocol and statistical analysis plan methodology for evaluating incomplete and/or missing follow-up assessments.
“The statistical analysis plan should specify the primary analysis and 1 or more sensitivity analyses to evaluate the effect of missing observations on the results,” the committee concluded in the guidance. “Sponsors can consider additional analyses of progression-free survival to support the primary MFS analysis.”