A Phase I Study of Gemcitabine and Bendamustine in Relapsed/Refractory Hodgkin Lymphoma

Background: Current salvage regimens for Hodgkin lymphoma (HL) often incorporate novel approaches, but there remains a role for combination chemotherapy in patients who require salvage therapy prior to autologous stem cell transplant (ASCT) or who have failed less intensive approaches.  Both gemcitabine (gem), with a single agent overall response rate (ORR) of  39% (Santoro et al, JCO 2000) and bendamustine (benda), with a single agent ORR of 53% and median progression-free survival (PFS) of 5.2 months (Moskowitz et al, JCO 2013) are active in relapsed/refractory HL. However, the combination of gem and benda has not been previously evaluated. We conducted a phase I study of benda and gem in patients with relapsed/refractory classical HL who failed at least 1 prior therapy to determine the maximum tolerated dose.  

Methods:  Utilizing a 3+3 design, patients received gem dosed at 1000 mg/m² on day 1 and benda doses of 60mg/m² to 120 mg/m²on days 1 and 2 of each cycle for up to 6 cycles, with cycle lengths of 28 (dose levels [DL] 1-3) and 21 (DL 4-5) days.  Dose limiting toxicity (DLT) was determined during cycle 1 and included grade 3 febrile neutropenia > 7 days, grade 4 infection or febrile neutropenia, treatment delay > 14 days for grade 3-4 neutropenia or thrombocytopenia, and most grade 3-4 non-hematologic toxicities.  Patients could discontinue study therapy after at least 2 cycles to pursue potentially curative autologous or allogeneic stem cell transplantation.  Efficacy was assessed using the ORR, and we evaluated PFS using the Kaplan-Meier method.

Results:  Sixteen patients with a median age of 36 years (range: 23-60), including 9 males have initiated study therapy.  Patients receiving a median of 4 (range: 1-8) prior lines of therapy have completed a median of 4 cycles at DL 1 (n=3), 2 (n=3), 3 (n=3), 4 (n=4), and 5 (n=3). Nine patients had a prior ASCT, including one patient who had a prior ASCT and allogeneic transplant. The benda dose was 60mg/m² for 3 patients, 90mg/m² for 7 patients, and 120mg/m² for 6 patients. One patient receiving benda 90mg/m²(DL4; 21 day cycle) did not return for the day 2 benda dose, and as a result this patient was excluded from dose escalation determination. No DLTs have been observed through the first 16 enrolled patients. Four patients have required hospitalization, including 2 heavily pretreated patients with pulmonary symptoms after cycle 2.  One patient underwent a biopsy that was consistent with a drug-induced pneumonitis that was likely gem-related.  This patient fully recovered after a course of corticosteroids and was removed from study therapy but died of sepsis during subsequent treatment off study. A second patient died from multi-factorial respiratory failure that was deemed secondary to infection although a relationship with gem could not be ruled out. Additional grade 3-4 toxicities include: lymphopenia (n=10) thrombocytopenia (n=2), atrial fibrillation (n=1), fever (n=1), renal failure (n=1), hypotension (n=1), pneumonia (n=1), hypoalbuminemia (n=1), hypokalemia (n=1), and rash (n=1). In 15 evaluable patients, the response rate was 80% (CR, n=5; PR, n=7; SD, n=2; PD, n=1).  Among 3 patients who attempted subsequent stem cell collection for ASCT, 2 were successful and proceeded to ASCT while one failed collection and proceeded to allogeneic transplant. Two additional patients discontinued study therapy to proceed with ASCT with previously collected stem cells.  In addition to the two patient deaths above, one patient has died of unknown causes (suspected progression), while two additional patients have died from progressive disease.  The remaining 11 patients are alive at last follow-up, including 3 with documented progression at 5.1, 7, and 9 months, and 8 patients who are progression-free with response durations of 1 – 30 months after a median follow-up of 10 months (range: 1-32 months). Post-study management for the 8 patients who remain progression-free includes: observation (n=2), ASCT/planned ASCT (n=4), allogeneic transplant (n=1), and umbilical cord transplant (n=1).  The estimated median PFS is 10.1 months, and the estimated 1–year PFS is 41.9%. 

Conclusion:  The combination of benda 120 mg/m2 with gem 1000 mg/m2 is tolerable and appears efficacious.  We continue to monitor closely for additional pulmonary toxicity as we complete accrual to our final DL using a 21-day cycle and plan to present toxicity and dose-escalation outcomes for the entire study at the Annual Meeting.