In terms of trials that have used MRD guidance, the phase 2 CAPTIVATE trial is one of the first. In this trial, patients with previously untreated CLL received 3 months of ibrutinib treatment before adding venetoclax, and then they completed 1 year of combination therapy. After 15 months of therapy, patients with confirmed undetectable MRD (uMRD) were randomized to continue or discontinue ibrutinib, whereas those who were not confirmed to have uMRD were randomized to receive ibrutinib monotherapy or ibrutinib with venetoclax. 

Among patients with confirmed uMRD, there was no difference in the 3-year rate of disease-free survival (DFS) between those who continued ibrutinib and those who stopped it. The DFS end point required MRD to remain undetectable and no disease progression. So, the findings of CAPTIVATE established a set of data for patients stopping therapy after uMRD. One caveat is that the trial criteria that were used to define uMRD were rather stringent in terms of frequency of assessment, with multiple points of confirmation, which may not be as viable in clinical practice.

When considering those patients in CAPTIVATE who did not have confirmed uMRD (and who were randomized between continuing ibrutinib or continuing ibrutinib plus venetoclax), the optimal duration of therapy remains an open question. That is, the durations that would allow most people to achieve uMRD are very unclear. This is partly because we have not done frequent sampling. Typically, clinical trials have tested MRD at 1 or perhaps 2 time points. Now that more trials are being done to evaluate MRD status–guided treatment decisions, hopefully we will learn if additional MRD assessments can be used to optimize the duration of therapy. 

The phase 3 MAJIC trial is comparing the combination of acalabrutinib and venetoclax with venetoclax plus obinutuzumab as frontline treatment for CLL. After 12 cycles of therapy, patients will have their MRD tested, and this will be used to guide the duration of their therapy. Patients with uMRD will stop treatment, and the remaining patients will continue to complete 24 cycles of treatment. That will be an interesting study to evaluate.

I think that when we look at the achievement of uMRD, to some extent, we are looking at a patient whose CLL has more favorable biology. The interesting thing is that standard measures of prognosis do not predict who will achieve uMRD. And that is something that I think we need to understand better so that we can intensify therapy early for those patients who do not or will not achieve uMRD. There is some evidence that a very rapid drop in disease burden during the first few months of therapy is associated with a much greater likelihood of uMRD and a better outcome. Future trials could investigate adding a therapy for patients who do not achieve that rapid drop or plan consolidation therapy for patients with positive MRD. These ongoing clinical trials are just a couple of examples of how we are studying MRD-guided decision making.

We are still in the infancy of using MRD for clinical decision making, and we do not routinely use MRD to guide therapy, but we all think that we are headed in that direction.