Utilizing Circulating Tumor DNA as a Predictor of Treatment Response in MCL

Measuring circulating tumor DNA in the blood of patients with advanced MCL may be a viable way to predict how well they will respond to specific therapies, according to study findings reported at the 2018 ASH Annual Meeting.

Rahul Lakhotia, MD

Measuring circulating tumor DNA (ctDNA) in patients with advanced mantle cell lymphoma (MCL) may be a viable way to predict how well they will respond to specific therapies, according to study findings reported at the 2018 ASH Annual Meeting.

Watching the rise or fall of ctDNA very early in treatment can be a good way to determine whether therapies are working or should be changed, said lead author, Rahul Lakhotia, MD, from the Center for Cancer Research at the National Cancer Institute in Bethesda, Maryland. This is useful because, while therapies are chosen based on physician expertise, results are uncertain as the disease exhibits “striking clinical variability” from patient to patient.

“Mantle cell lymphoma is an incurable disease that can be aggressive, requiring multiple treatments, or (slower-growing and) not needing immediate therapy,” he said, adding that “treatments are usually not adjusted based on depth of response.”

Another advantage to testing tumor DNA is that this strategy is able to detect relapse earlier than CT scans can, Lakhotia explained.

In particular, determining at certain points during treatment whether patients have minimal residual disease (MRD) is an important predictor of patient health outcomes, the study found.

Lakhotia and his colleagues used a genomic sequencing assay that measures ctDNA; the assay relies on tumor samples collected at baseline to recognize circulating tumor cells.

The researchers applied the tests during a clinical trial that tested whether adding bortezomib (Velcade) as a maintenance drug after chemotherapy improved health outcomes in patients with previously untreated MCL.

Patients were initially given one round of bortezomib, with their ctDNA tested before and after that treatment. Then, the patients were given induction therapy with 6 cycles of rituximab (Rituxan), EPOCH (etoposide, prednisone, vincristine, cyclophosphamide and doxorubicin) and bortezomib. Their ctDNA was checked after every cycle.

Finally, the patients were divided into 2 groups: one that took bortezomib alone and another that was observed. The patients’ ctDNA was checked every 3 to 6 months for 5 years, and annually after that.

Patients also underwent CT scans every 4 months for two years, every 6 months during years 2 through 4, and then annually.

The trial included 53 patients, the majority of them men with stage 4 disease whose median age was 59. Most patients did not have a high MCL International Prognostic Index (MIPI) score or blastoid disease type, both of which are associated with poorer health outcomes. The patients were treated between September 2005 and January 2016.

After a median follow-up of 9.9 years, the median progression-free survival (PFS) of patients across both study arms was 29.3 months and their 5-year overall survival (OS) was 78.2 percent. When the 2 arms were compared, it was found that taking bortezomib after induction therapy had not increased PFS: Those who took the drug had a median PFS of 27.3 months, compared with 33.5 months for those who did not take it. However, there was a small group of patients whose PFS lasted years past the median without the need for consolidation therapy or stem cell transplant, suggesting that they were particularly sensitive to this treatment, Lakhotia pointed out.

Pathologists were able to detect circulating tumor cells in 50 of 52 patients (96%) eligible for analysis, Lakhotia reported, and baseline levels of circulating DNA correlated with patients’ metabolic tumor volume.

The researchers observed a reduction in circulating tumor DNA as patients moved through treatment.

Before they started treatment, 94% of evaluable patients had detectable tumor DNA in their blood. After 2 cycles of bortezomib followed by chemotherapy, about half (54%) had detectable cancer cells. Following completion of the chemotherapy induction, about 20 percent had detectable tumor DNA in their blood.

Imaging tests identified 3 nonresponders, 3 partial responders and 44 complete responders. Yet, 5 of the complete responders, who by definition should have been free of disease, were found to have circulating tumor DNA evident in their blood. This means that “traditional response criteria may not be sufficient in this disease,” Lakhotia said

.

When measuring PFS from study enrollment until disease progression, the researchers found that patients who were free of any circulating tumor DNA after induction fared better. Those with no detectable circulating tumor cells at that juncture had a median PFS of 31.5 months, compared with 17.9 months for those who did have circulating tumor DNA. The same held true for OS: Those with detectable tumor DNA in their blood after induction had a median OS of 47.2 months, compared with a number not yet reached for those with no detectable tumor DNA at that time point.

In addition, there were earlier moments in the course of treatment when circulating tumor DNA measurements were found to predict health outcomes.

After the first 2 cycles of induction therapy, median PFS was 21.4 months in those with detectable tumor DNA compared with 32.4 months in those with none; this was associated with a trend toward superior median OS. After just the first cycle of induction therapy, the difference between the groups was even more clear: The median PFS in those with detectable tumor DNA was 20.7 months, versus 76.4 in patients with no measurable tumor DNA; this was also associated with a trend toward superior four-year OS.

The difference could even be seen after the initial round of bortezomib, the researchers found. Patients with tumor DNA in their blood exceeding 35 cell equivalents/ml had a median PFS of 22 months, versus 34 months for those with tumor DNA below that level. This was an early indication of sensitivity to therapy, Lakhotia said.

The results of the study confirmed data from previous research showing that a post-therapy status of no MRD predicts a better health outcome, and that sequential MRD assessments can be used to guide therapy, Lakhotia said.