Minimal Residual Disease in CLL

Alan P. Skarbnik, MD: Minimal residual disease [MRD] in CLL [chronic lymphocytic leukemia] is a complicated and controversial question. CLL is not, in general terms, a curable disease. The holy grail is to allow patients to enjoy the longest possible time without the disease being active and causing symptoms, without the need for any treatment for the longest period of time. Since chemoimmunotherapy has been used, particularly FCR [fludarabine, cyclophosphamide, rituximab], the question of anywhere near minimal residual disease undetectability is never negative because there’s a threshold of detection of the tests.

MRD undetectability is a predictor of PFS [progression-free survival]; patients who have a deeper response seemed to enjoy a longer progression-free survival. This was shown initially in the CLL-8 trial of FC [fludarabine, cyclophosphamide]vs FCR [fludarabine, cyclophosphamide, rituximab]. In the University of Texas MD Anderson Cancer Center data set, patients who had MRD undetectability and were IGHV mutated, over that long PFS there were some differences compared with patients who had detectable minimal residual disease.

There are some caveats to this because the way to test minimal residual disease is in the blood or in the bone marrow. You can eliminate the component of CLL in the bone marrow but not necessarily eliminate it in the molecular compartment. You can have a partial remission with an MRD undetectability. Based on some data from a couple of different trials, we know that there is a difference in patients who show FCR [fludarabine, cyclophosphamide, rituximab] with undetectable MRD. It is a strong predictor, but we have put everything into perspective and put all the data together to understand what’s going on.

In the MURANO data, which was venetoclax-rituximab vs bendamustine-rituximab, undetectable MRD was the main predictor of recurrence of disease, after discontinuation of therapy. Venetoclax in that trial was given for 2 years. For patients who did achieve undetectable MRD, it was done by real-time qPCR [quantitative polymerase chain reaction]. The lowest threshold of detectability was 10-4, 1 in 10,000 cells. For patients who had undetectable MRD at the median follow-up of about 10 months, 97.6% of them were progression-free. Patients who had what they call low-MRD positive, between 10-4 and 10-2, 87% were progression-free at 10 months postdiscontinuation of venetoclax. Patients who had high MRD up to true positivity of MRD, which was more than 10-2. Only 21% of patients were progression-free after 10 months postdiscontinuation; 78.6% of the patients did progress after discontinuation of treatment.

This showed that achieving an undetectable MRD status with that particular agent was important because the goal was to discontinue the therapy. Should the patients who have positive MRD at the end of the treatment continue on therapy or not? We don’t know the answer. They relapse a little quicker than the ones who are MRD undetectable. It does not seem that increasing length of treatment, particularly with a venetoclax-based regimen, increases the depth of MRD for these patients. If patients don’t achieve an MRD-undetectable state during the first 10 to 12 months, that’s not going to convert afterward. Transitioning to the frontline setting, looking at the patients who received venetoclax plus obinutuzumab, we know that obinutuzumab is a decent driver of MRD undetectability based on the results of the CLL11 trial. The levels of undetectable MRD were much higher in the obinutuzumab-chlorambucil arm than in the rituximab-chlorambucil arm.

In the CLL14 trial of venetoclax plus obinutuzumab vs obinutuzumab and chlorambucil, after 18 months they address the MRD state. Close to half the patients in the VenG arm [venetoclax, obinutuzumab] had undetectable MRD; it was 47.2%, and the level of detectability was 10-4; 13% had low MRD, between 10-4 and 10-2; 71.9% had high MRD, above 10-2. The real number we’re looking for is undetectable, less than 10-4. That’s 47.2% in the VenG [venetoclax, obinutuzumab] arm, whereas in the chlorambucil-obinutuzumab arm it was 7.4% undetectable MRD. In this trial after discontinuation of therapy, it seemed that patients who had undetectable MRD at the time of fixed duration enjoyed a longer progression-free survival.

That question comes into play with newer therapeutic approaches. We know that the use of BTK [Bruton tyrosine kinase] inhibitors as monotherapy doesn’t drive MRD undetectability as much. The levels of undetectable MRD, particularly with the monotherapy with BTK inhibitors, are lower. These patients are on therapy continuously, so is it important to achieve that for those patients? We don’t know for sure. But by combining novel agents—BCL3 inhibitors with the BTK inhibitor and anti-CD20 monoclonal antibody—with the intent of stopping therapy after a fixed duration of time, achieving undetectable MRD is of high importance. It may be the guide of therapy for those patients.

There are more data coming. Things are going to be updated at ASH [American Society of Hematology Annual Meeting], particularly in MRD undetectability for CLL. We understand the importance of checking for MRD, particularly for predicting the patients who may progress after discontinuation of therapy. Taking into consideration other disease states with the hematology malignancies—CML [chronic myelogenous leukemia] and ALL [acute lymphoblastic leukemia]—we’re looking for an MRD-undetectable state, and that does make the difference in survival. We’re certain that this will translate and speak to a significance in CLL as well.

Transcript Edited for Clarity