Phase II trial on panobinostat monotherapy for R/R DLBCL

In April 2018, Francesco Zaja from the University of Udine, Italy, and colleagues published ahead of print in Leukemia and Lymphoma, a phase II clinical trial on panobinostat for relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL), on behalf of the Italian Lymphoma Foundation (Fondazione Italiana Linfomi).

The aim of the study was to evaluate the efficacy and safety of oral panobinostat (PAN) administration for the treatment of heavily pre-treated R/R DLBCL patients, who have received at least two previous regimens including rituximab-containing therapies and ASCT if they were eligible for it. The primary endpoint of this study was overall response rate (ORR) following induction phase completion, with secondary endpoints including progression-free survival (PFS), overall survival (OS), response duration (RD), time to response (TTR) and safety.

Patient characteristics & study design

• N = 35 patients with confirmed DLBCL R/R to one or multiple chemotherapy lines (R-CHOP, DHAP, ESHAP, ICE or similar) and autologous stem cell transplantation (ASCT), enrolled between June 2011 and May 2014 at any of the 12 participating centers of the Fondazione Italiana Linfomi
• Median age (range): 73 (65–75.5) years
• Median number of previous treatments (range): 2 (1–4) with all patients having previously received at least one rituximab-containing regimen
• Patients with advanced-stage DLBCL: n = 27 (82%); increased low-density lipoprotein (LDL): n = 24 (69%) and increased International Prognostic Index (IPI): 65%
• Treatment schedule:
  • Oral single dose of PAN (40 mg) thrice weekly
  • Induction phase (cycle 1–6) followed by maintenance phase (cycle 7–end of therapy)
  • Cycle duration: 28 days
  • Patients with grade ≥ 3 non-hematological toxicities and grade 4 neutropenia or thrombocytopenia discontinued PAN temporarily until recovery, when they restarted PAN at a lower dose (30 mg)

Key findings

Efficacy at induction phase:

• Eight patients (22.8%) completed the induction phase, while n = 27 discontinued due to: no response (NR) or progressive disease (PD), 19; treatment-related toxicities, 5; consent withdrawal, 1; poor compliance, 1.
• ORR after induction phase completion: 17.1% (95% CI [6.6–33.6])
  • Four complete responses (CR)
  • Two partial responses (PR)
  • One stable disease (SD)
• Median TTR (range): 2.6 (1.8–12) months
• IPI score before PAN administration was the only characteristic to correlate with response at the end of the induction phase (P = 0.042)
• Median RDI (range): 58% (44–100%)

Efficacy at maintenance phase:

• Six patients (17.1%) discontinued PAN up to 12 months of treatment and 2 patients discontinued due to PD
• ORR and CR at 12 months: 14% (n = 5) and 11.4% (n = 4), respectively
• Median follow-up: 51 months, at the time of analysis no patients were on PAN
• In the 5 patients responding the overall period of PAN treatment was 12, 44, 48, 50 and 62 months
• Median PFS: 2.4 months (95% CI [1.4–7.0])
• Extrapolated 12- , 24- and 36-month PFS were 26%, 11% and 11%, respectively
• Median OS: 7.6 months (95% CI [3.0–12.7])
• Calculated 12- , 24- and 36-month OS were 34%, 19% and 12%, respectively

Genomic associations:

• No significant correlation was observed between MLL2, MLL3, PIM1, BCL-2, MYD88 or CREBPP mutations found in participating patients and response to PAN
• TP53 mutation did not seem to impact clinical outcome
• Two patient cases with MEF2B mutations did not respond to PAN

Safety

• Reported deaths: n = 29 patients (83%)
  • Due to lymphoma: n = 26
  • Due to pneumonia and pancytopenia: n = 1
  • Due to recurrent infectious complications: n = 1
  • Due to complications secondary to ASCT performed after PAN
• Six patients discontinued PAN due to treatment-associated toxicities (3, prolonged hematologic toxicity; 2, atrial fibrillation and 1, poor compliance)
• Most commonly observed adverse events (AEs):
  • Thrombocytopenia (grade 3–4): 83%
  • Neutropenia (grade 3–4): 34%
  • One patient presented with febrile neutropenia and another with an infectious bacterial complication

These results demonstrate than PAN has a modest activity in R/R DLBCL patients and that treatment-related toxicities like thrombocytopenia limit the use of this regimen. The authors noted that with the development of better biological markers associated with PAN response, the identification of a certain DLBCL patient subgroup that might benefit from this treatment will be feasible.