The histone deacetylase inhibitor panobinostat can produce durable responses in certain patients with relapsed diffuse large B-cell lymphoma (DLBCL), according to a phase 2 trial.
Though less than 30% of the patients in this trial responded to treatment, the median duration of response was 14.5 months, and the longest ongoing response has lasted more than 3.5 years.
In addition, researchers found the presence of mutations in MEF2B and levels of circulating tumor DNA (ctDNA) could be used to predict response.
Sarit Assouline, MD, of Jewish General Hospital in Montreal, Quebec, Canada, and her colleagues reported these findings in Blood.
“DLBCL is one of the more common forms of lymphoma and is highly aggressive,” Dr Assouline noted. “Following relapse, there are no effective standards of treatment, and life expectancy averages 6 months.”
“Our challenge is to identify new biomarkers and target specific mutations in order to improve the prognosis. While many clinical trials simply report on how patients respond to a therapy, we devised this study to reveal the mechanisms on which the therapy works in order to understand which patients would benefit.”
The trial enrolled 40 patients with relapsed or refractory de novo (n=27) or transformed (n=13) DLBCL. The patients’ median age was 59.8 (range, 28.9-78.6). They received a median of 3 prior therapies (range, 1-9), and 75% of patients (n=30) were refractory to their most recent therapy.
The patients received panobinostat at 30 mg three times a week, with rituximab (n=19) or without (n=21).
Adverse events
The most common grade 3 or higher adverse events (AEs) that were thought to be related to panobinostat (with and without rituximab, respectively) were thrombocytopenia (68% and 71%) and neutropenia (11% and 24%).
Twenty-three patients (58%) required dose reductions, and 22 (55%) required dose interruptions. All of these were due to AEs, and there was no difference between patients who received rituximab and those who did not.
All 19 serious AEs were a result of disease progression.
Response
Twenty-eight percent of the patients (11/40) responded to treatment, and receiving rituximab did not increase the likelihood of response. Six patients (29%) responded to panobinostat alone, and 5 (26%) responded to panobinostat and rituximab.
There were 7 complete responses—5 of them among patients who received panobinostat alone. And there were 4 partial responses—3 among the patients who received panobinostat plus rituximab.
The median duration of response was 14.5 months overall, 9.4 months among patients who received panobinostat plus rituximab, and it was not reached among patients who received panobinostat alone.
At the data cutoff, 6 of the 11 responders had not progressed. The longest response had lasted more than 43 months and was ongoing at the cutoff point.
Predicting response
The researchers performed genome and exome sequencing in relapsed tumor biopsies and quantified ctDNA in serial plasma samples.
They found that mutations in MEF2B were significantly associated with response to panobinostat. The likelihood ratio for response was 3.67 for patients with MEF2B mutations.
On the other hand, increased levels of ctDNA were strongly associated with treatment failure. A significant increase in ctDNA at day 15 after treatment initiation could predict a lack of response with a sensitivity of 71.4% and a specificity of 100%.
“This trial has generated considerable data regarding methodology for processing samples from a clinical study, the genetic mutations associated with DLBCL and how they evolve over time, on ctDNA, and mechanisms of resistance to histone deacetylase inhibitors,” Dr Assouline concluded.
Novartis provided panobinostat and financial support for this trial, and Hoffman-LaRoche supplied rituximab.
