The FDA has officially released its finalized recommendation for the use of metastases-free survival (MFS) as an end point in clinical trials evaluating treatments for patients with nonmetastatic castration-resistant prostate cancer (nmCRPC).1
In addition to including editorial changes for clarity, the guidance also has recommendations for sponsors on aspects such as general trial design, imaging, and considerations for interpreting data from the trial and MFS analyses.
A prolonged assessment period and low death rate means that the utility of overall survival (OS) can be impractical as a primary end point to support drug approvals for patients with prostate cancer. These issues were explored during the 2011 Oncologic Drugs Advisory Committee meeting wherein members recommended that other end points be explored that can be measured earlier in the disease course, such as MFS which could prove to be useful in evaluating the impact of treatment in this patient population.
The guidance highlighted several general considerations for clinical trial designs that utilize MFS as an end point. For example, the FDA recommends defining MFS prior to starting the trial—which should exclude local progression events such as progression in pelvic lymph nodes below the aortic bifurcation—as well as considering stratification of randomization by prior local definitive therapy or lack of prior definitive therapy. Prostate-specific antigen (PSA) doubling time should also be considered, according to the guidance.
Other recommendations include utilizing a superiority trial design and excluding patients who are likely to benefit from local therapy. This population of patients could be enrolled following local therapy provided that their PSA continues to rise and other enrollment criteria for CRPC and minimum PSA value are met.
In terms of imaging consideration, the guidance states that sponsors ought to prespecify acceptable imaging modalities and how often assessments should be completed. Additionally, it is important to prespecify and justify the radiographic definition of nonmetastatic disease. Radiographic definition of local disease and progression, as well as metastatic disease, should also be prespecified. In order to interpreted results with MFS, the significance of MFS improvement should be notably greater than frequency of imaging.
“FDA recommends a blinded independent central review of imaging studies to assess any potential assessment bias. Sponsors may also consider using an audit where a random sample of scans are sent for independent review. Sponsors should seek advice from the Agency about the design of an audit for MFS,” the guidance stated.
When it comes to interpreting the results of trials utilizing MFS as an end point, interim efficacy analyses of MFS are discouraged to avoid over overestimating and underestimating MFS benefit. Moreover, the magnitude of MFS improvement that is required to support drug approval will be based on trial design, safety, patient population, and overall risk/benefit assessment. Additionally, although demonstration of OS benefit is not required, a formal interim OS analysis should be conducted at the time of final MFS analysis.
Lastly, the guidance highlighted several final considerations for MFS analyses. As missing data may complicate MFS analyses, it is imperative to put procedures in place that minimize missing data.
“The statistical analysis plan should specify the primary analysis and one or more sensitivity analyses to evaluate the effect of missing observations on the results,” the guidance read. “Sponsors can consider additional analyses of progression-free survival (including both local and metastatic progression) to support the primary MFS analysis.”
Before the draft guidance was released, the FDA had approved the use of apalutamide (Erleada) in February 2018 for patients with nmCRPC, marking it as the first drug approved on the basis of a clinical trial with an MFS primary end point.2 The phase 3 SPARTAN trial (NCT01946204) enrolled 1207 patients with nmCRPC who were treated with either apalutamide or placebo. Results indicated that apalutamide yielded a median MFS of 40.5 months vs 16.2 months in those who received placebo (HR, 0.28; 95% CI, 0.23-0.35; P <.001).3
Similarly, enzalutamide (Xtandi) received approval in July 2018 for the treatment of nmCRPC with MFS, which was defined as the time from randomization to loco-regional and/or distant radiographic progression or death up to 112 days following treatment discontinuation with no radiographic progression, as the primary end point of the supporting trial.4 Findings from the supporting phase 3 PROSPER trial (NCT020032924) indicated that enzalutamide yielded a median MFS of 36.6 months vs 14.7 months in the control arm (HR, 0.29; 95% CI, 95% CI, 0.24-0.35; P <.001).5
Lastly, the FDA approval of darolutamide (Nubeqa) on July 30, 2019, was based on the results of the phase 3 ARAMIS (NCT02200614) trial, in which 1509 patients with nmCRPC achieved a significant improvement in MFS, the primary end point of the study.6 For this trial, MFS was defined as the time from randomization to first signs of distant metastases or death from any cause within 33 weeks of the patient’s last evaluable scan. Those who received the experimental agent experienced a median MFS of 40.4 months vs 18.4 months among those in the placebo arm (HR, 0.41; 95% CI, 0.34-.050; P <.0001). Notably, the OS data were not mature.7