FACTS 6.1.6 Release Notes

February 26, 2018

FACTS 6.1.6 is a maintenance release for FACTS 6.1.0. The first product release of FACTS 6.1.0 was FACTS 6.1.3. Subsequent releases have introduced the following changes:

• FACTS 6.1.4: This was FACTS 6.1.3 with some additional logging when using the grid interface.

• FACTS 6.1.5: This was FACTS 6.1.4 with the Dose Escalation N-CRM re-compiled to allow a higher number (40) of dose strengths to be defined when using “Explicit Doses” rather than finely spaced doses.

• FACTS 6.1.6: This was FACTS 6.1.5 with 2 problems fixed in FACTS Core with a TTE endpoint & FACTS Staged Design with a TTE endpoint. In either engine the calculation of a “Current Trial Predictive Probability of Success at Current Enrollment” had 2 problems:

  1. There was an error in the way timings of future events were simulated in the calculation of the predictive probability. The result was approximately correct, and erred on the conservative side, the error is more manifest if the trial has long follow-up times.
  2. There was an error if the design also includes a predictor endpoint. This effects a much smaller set of designs, but the effect was much more marked and its impact was difficult to characterize in general. Our current advice is to not use predictive probability QOIs in combination with a “Predictor” endpoint using FACTS prior to FACTS 6.1.6.

Upgrading FACTS 6.1.6 should introduce no changes to the simulation or analysis results relative to FACTS 6.1.3 except in designs using a time-to-event endpoint and a “Predictive Probability of Success in the Current Trial at Current Enrollment” Quantity of Interest.