Why did we change the design to and calculate our sample size based on a non-inferiority trial with local recurrence rate at 3-year follow-up as primary endpoint?


The COLOR III collaboration started in the end of 2014 and was based upon facts that were available from published cohorts. The circumferential resection margin (CRM) rate was chosen as primary endpoint within a superiority design; calculated sample size of 1098 patients based on decrease of involved CRM from 7 to 3 percent. We have waited with the real start of the trial because the majority of centers were still in their learning curve, which would negatively influence the new technique. Currently, the experience of TaTME surgeons is expanding and now is the time to start the trial. However, since 2014 more data is published demonstrating that the CRM rate after TaTME is about 4.7% and on beforehand the primary endpoint can no longer be justified.

Therefore we have changed the design to a non-inferiority design with a patient relevant endpoint, namely local recurrence rate. A superiority design based upon CRM has a risk of failing on the primary endpoint, subsequently downgrading the possible benefit of TaTME in the secondary endpoints. The non-inferiority design was used in the previous COLOR trial and COLOR II trial, and has shown to be an effective design in evaluating a possible difference between open and laparoscopic surgery.

The TaTME technique has the potential to improve clinical outcome of patients with mid and low rectal cancer in terms of specimen quality, conversion rate, end colostomy rate, morbidity and efficacy. These will be all investigated as endpoints within the safety of a non-inferior design. Due the imaging of the pelvis after three years and the possibility of salvage surgery patients can still be offered therapy translating the potential 4% difference to a lower local therapeutic failure.

The COLOR III group will publish their results in stepwise fashion;

  • CRM rate and short term outcomes will be available after finishing the inclusion
  • Quality of Life and functional outcome after one and two years
  • Long-term oncological outcome after 3-year follow up and after 5-year follow-up

The change of design and primary endpoint has already been approved by the ethical committee of the VU University Medical Center in Amsterdam, The Netherlands.