Webinar Effective Trial Design: Clinical Trials in Small Patient Populations

To mark the occasion of Rare Disease Day 2021, on the 4th of March we were repeating our complimentary webinar “Effective Trial Design: Clinical Trials in Small Patient Populations”. Together with the participants, we discussed the benefits and challenges of three types of the following designs:

  • Designs where each patient serves as his/her own control
  • Designs where the control group is based on historical data
  • Designs that allow changes in the protocol of an ongoing trial

With the speaker, Thomas Zwingers, Statistical Consultant of CROS NT, we talked about how these designs can solve the challenges inherent in trials in the small patient populations: determining a meaningful difference in clinical outcomes that can be detected between treatment groups and extended time to recruit the necessary number of patients.

Continue reading to discover the questions from the webinar’s Q&A session answered by the speaker.

Click here to watch the replay of the “Effective Trial Design: Clinical Trials in Small Patient Populations” webinar

How many resources can I save with these designs?

The possible saving in resources depends on the design. For example, in a cross-over study design, usually up to 50% of resources can be saved.

How many treatments can be applied to patients in the N-of-1 design?

In a N-of-1 design, the number of treatments applied to a patient is not limited. As N-of-1 trials are often used to find the best individual therapy for a patient, no further treatments are applied thereafter.  The time needed to evaluate the response is also a limiting factor for the number of treatments.

What are the dangers of the trials where patients are their own control?

The biggest danger in studies where patients are their own control is a “carry-over” effect. This means, that the conditions at the start of each treatment period are not comparable. If this is due to the underlying disease, e.g. the disease severity changed due to the treatment applied, then there no workaround. If it is due to drug concentrations, then a longer wash-out period has to be applied.

In the adaptive design, there should be a wash-out period after the dose level choice in order to prevent carry-over effect on another patient?

After a choice of a dose level in a seamless adaptive design, no wash-out period is necessary, because carry-over effects can only affect an individual patient.

Is the cross-over design applicable to placebo-controlled studies, or this would be considered unethical?

A cross-over design where one treatment sequence is a placebo is not unethical.

How much time can I save with a sequential design?

Sequential designs offer the possibility to prematurely stop a trial for efficacy or futility.  The time-point when a stop is feasible depends on the effect-size evaluated in the study. If your effect-size is larger than anticipated in the sample size calculation, early stopping is realistic. From our experience very early stopping is unlikely. Most of the trials stop after 75% of all information is available.

What is the usual duration of the trial for sequential trials (patient duration of the treatment)?

The observation time per patient should be considered with respect to the targets of the interim analyses. If you want to adjust the sample size or stop for futility, then the respective interim analysis has to be performed before the last patient has been included in the study. Therefore the individual observation time has to be shorter than the recruitment period. If the target of the interim analysis is stopping for efficacy, then the observation time per patient is not limited.

If it takes 6 months to get an outcome, does it makes sense to use a sequential design?

A 6-month interval to get the outcome might be too long for sample size adjustment if your recruitment rate is high. But it is absolutely feasible if you want to stop early for efficacy. In cancer trials, the observation period is often much longer.

What do you think about Factorial design?

A factorial design is very effective to analyse two different interventions, e.g. duration of treatment and drug intensity, in one study. The saving with such a design is mainly time.

Click here to watch the replay of the “Effective Trial Design: Clinical Trials in Small Patient Populations” webinar

Do you have any questions about designing clinical trials in the small patient population that you did not manage to ask during the webinar? Get in touch now and talk to our experts!