The Final Report of a Phase I/II Study of Panobinostat in Combination with ICE (Ifosfamide, Carboplatin and Etoposide) in Patients (pts) with Relapsed or Refractory (R/R) Classical Hodgkin Lymphoma (cHL)

Background: ICE is a commonly used regimen for R/R cHL with complete response (CR) of 60-70%. Panobinostat (P) is a histone deacetylase inhibitor (HDACi) with clinical activity against R/R cHL after transplant with dose limiting toxicity (DLT) of reversible thrombocytopenia. Based on synergistic anti-lymphoma activity of chemotherapy and HDACis in vitro, we conducted a clinical trial combining P and ICE.

Methods: This was an open-label phase (Ph) I/II study conducted at MD Anderson Cancer Center. Primary objective of Ph I was to determine recommended Ph II dose. Primary objective for Ph II was to estimate CR rates in pts receiving ICE and P-ICE. Secondary objectives include rate of successful stem cell collection, percentage undergoing auto-HSCT, failure free survival (FFS, treatment failure is defined by unplanned treatment excluding chemomobilization). Major inclusion criteria include pts with histologically confirmed cHL who failed standard front-line anthracycline-containing regimen with measurable disease ≥2cm, PS ≤2, age ≥16, and adequate organ function. Original treatment schedule in Ph I study (A) was P Monday/Wednesday/Friday starting 1 week prior to cycle 1 (C1) of ICE, and during first and second week of ICE for C1-2. Cycles repeat q14 days provided resolved cytopenia (platelet≥75K). Starting dose of P was 20mg; target dose was 30mg based on 3+3 design, with planned expansion cohort enrolling 10 additional pts at the highest dose. MTD was not exceeded at 30mg, but 84% of pts developed grade (Gr) 4 thrombocytopenia and was considered excessive. Thus, for schedule B, P was given 1 week before ICE and during first week of C1-2 of ICE; 63% had Gr 4 thrombocytopenia. Schedule B was used for the subsequent randomized Ph II study (ICE vs P+ICE).

Results: We enrolled 29 evaluable pts in Ph I. At 20mg, DLT of febrile neutropenia was observed in 1 of 6 pts. At 30mg, 3 pts were enrolled without DLT. As mentioned in methods, cohort was expanded at 30mg in two schedules: A (n=10) and B (n=10). ORR was 86% (n=25) in Ph 1 cohort with 72% (n=21) CR. Median follow up duration was 41 months (mo) (range 7-65) at time of analysis of 6/30/16. Only 1 pt did not undergo auto-HSCT. Four year FFS and OS were 52% and 82%, respectively. (One and 2-year FFS were 66% and 51%, respectively. Median FFS was not reached.) Gr 3/4 non-hematologic toxicity was observed in 4 pts (14%) including syncope (Gr 3), low phosphorus (Gr 4), and infection (Gr 3). Common toxicities included fatigue (65%), nausea (100%), vomiting (38%) and diarrhea (34%). Hematologic toxicity included anemia (50% Gr 3, 3% Gr 4), neutropenia (10% Gr 3, 52% Gr 4), thrombocytopenia (14% Gr 3, 79% Gr 4) and febrile neutropenia (14% Gr 3).

In Ph II cohort, 23 were evaluable for response (12 ICE, 11 ICE+P). Patient characteristics were similar. Primary refractory disease was seen in 62% ICE and 64% ICE+P, p=0.9. CR was observed in 8/12 in ICE arm (67%) and 9/11 in ICE+P arm (82%), p=0.27. Median follow up duration was 11 mo (range 2-22). FFS was 44% in ICE and 46% in ICE+P at 18 mo, p=0.9. All pts were alive at time of analysis. Nausea (100% Gr 1/2 ICE, 82% Gr 1/2 ICE+P), vomiting (75% ICE Gr 1/2; 64% ICE+P Gr 1/2), diarrhea (25% ICE; 64% ICE+P) and hypomagnesemia (33% ICE; 45% ICE+P) were the most common toxicities. Gr 4 neutropenia was more common in ICE+P (73% vs. 8%) as was Gr 4 anemia (18% vs. 0%) and Gr 4 thrombocytopenia (100% vs. 33%). Despite Gr 4 hematologic toxicities, there was no difference in rates of delayed administration of subsequent cycles by >1 week: 33% in ICE vs. 45% in ICE+P, p=0.7.

In Ph II cohort, all pts who achieved CR but one proceeded to auto-HSCT; one pt did not undergo auto-HSCT due to insurance issues. Eighteen pts (9 ICE and 9 ICE+P) underwent successful peripheral blood stem cell collection of 2x10⁶CD34 cells and auto-HSCT, immediately after protocol treatment or after one chemomobilization.

Conclusion: ICE+P is an effective regimen with CR of 82%. However there was no difference in FFS at 18 mo compared to ICE. There seems to be no adverse impact on stem cell collection after P. Due to high rates of Gr 4 hematologic toxicities of ICE+P, emergence of other effective and less toxic agents such as brentuximab vedotin and anti-PD1 antibody, and limited additional benefit of P in combination with ICE, the investigation of P in this study was terminated. Nonetheless, given high response rate of ICE+P, investigation of HDACis in combination with other novel agents such as anti-PD1 antibody is of interest.