Introducing nQuery Predict

Accurately predict your key trial milestones

The Autumn 2021 release (v9.0) sees the introduction of a brand new module to the nQuery platform - nQuery Predict

What is nQuery Predict?

nQuery Predict is a new module that has just been added to the nQuery platform for clinical trial design.

nQuery Predict uses simulation to help trialists make better predictions about when key trial milestones will occur.

With nQuery Predict, you gain access to tools that accurately predict key enrollment and event milestones both before and while a trial is ongoing. This gives you the ability to identify roadblocks and take action to keep your trial on schedule.

What are the main features of nQuery Predict in this release?

• Enrollment prediction
• Event prediction

nQuery Predict provides tools that project the expected trajectory of enrollment or event milestones based on current enrollment trends as well as tools to assess the effect of reduced or increased recruitment from that point onwards.

Enrollment prediction

Enrollment prediction (i.e. how long will it take to achieve the desired sample size of patients) is a common issue across all trials and has a direct impact on the administration of a clinical trial.

Delays in patient enrollment or trial events create additional costs. With nQuery Predict, you can ensure the best performance of your trial by examining projections at both the planning stage and interim monitoring.

Event prediction

Event prediction is a complex problem in trials, such as those in oncology that plan to use survival analysis.

Key decisions such as when to conduct an interim analysis or when to conduct right-censoring to finish the study are most commonly based on a priori unknown time when a given number of subjects will have had an event rather than based on the fixed follow-up more common in other areas. This means that accurate projections of when event milestones will be achieved are vital to provide realistic estimates of expected timings and costs.

nQuery Predict provides tools to project events based on both blinded (where treatment group is unknown) and unblinded (where treatment group is known) using multiple event models (exponential, piecewise, Weibull) and design considerations (e.g. fixed follow-up per subject).