A Phase I/II Trial of Combination Gemcitabine and Bortezomib (VELCADE®) for Relapsed/Refractory T-Cell Non-Hodgkin Lymphoma (NHL) and Aggressive B-Cell NHL

NF-κB has been shown to be deregulated in B-NHL and T-NHL subtypes. The proteasome inhibitor, bortezomib, has the capacity to reverse the downstream consequences of NF-κB, while gemcitabine has documented single-agent activity in relapsed/refractory NHL. Further, in vitro and murine xenograft tumor models have demonstrated synergy between these two agents. Based on these data, and the continued unmet clinical need for patients with relapsed/refractory aggressive NHL either ineligible for or relapsed after stem cell transplant (SCT), we conducted a phase I/II trial utilizing this novel combination.

This was a phase I/II investigator-initiated clinical trial conducted through two centers. The phase I design was a classic 3+3 with dose escalation of bortezomib (1.3 mg/m² to 1.6 mg/m² given day (D) 1 and D8) with static gemcitabine dosing (800 mg/m² D1 and D8) given on q 21 day cycles. The definition of dose limiting toxicity (DLT) was: a) grade 3 or 4 non-hematologic toxicity (other than grade 3 nausea or vomiting); b) grade 4 vomiting despite maximal anti-emetic support; c) grade 4 neutropenia on D1 of a treatment cycle (despite growth factor support); and d) grade 4 thrombocytopenia on D1 of a treatment cycle. Following completion of bortezomib escalation, a planned phase II expansion of the study was planned. The null hypothesis was that the true success was less than or equal to 15% and the alternate hypothesis was that the true success was 40% or higher; type I error of 5% and a power of 80% was assumed.

From April 2006 to December 2010, we enrolled 32 relapsed/refractory NHL pts onto this phase I/II clinical trial. Sixteen pts had T-NHL (n=12 peripheral T-NHL NOS and n=1 each with angioimmunoblastic T-NHL, NK-/T-NHL, transformed large cell [from pre-existing cutaneous T-cell], and hepatosplenic) and 16 had B-NHL (all relapsed/refractory DLBCL). There were 16 women and 16 men with a median age of 61 years (range, 37–85 years). The median ECOG performance status was 1 (range, 0–2), median prior therapies were 2.5 (range, 1–5), while 35% had failed prior autologous SCT. During the initial phase I dose escalation, 2 DLTs were noted (grade 3 hypertension and grade 3 elevation of liver function tests), while a maximally tolerated dose was not identified. However, among the first 18 pts treated on the D1+D8 (q21 day) dosing schedule, 67% experienced grade 3/4 neutropenia and/or grade 3/4 thrombocytopenia, primarily on D8 of treatment cycles. These recurrent D8 cytopenias resulted in repeated treatment delay(s). The median number of cycles delivered for these 18 pts were 1.0 (due to hematotoxicity), which was associated with a low (59%) median normalized dose-intensity. Thus, in early 2009, the clinical trial was amended instituting a modified treatment schedule of gemcitabine 800 mg/m² and bortezomib 1.6 mg/m² to both be administered on D1 and D15 of a 28-day schedule for an additional 22 patients. Treatment-related toxicity was markedly reduced using this modified treatment schedule; only one grade 3 event each of anemia and thrombocytopenia were recorded. However, after 14 pts had accrued to the modified treatment schedule, efficacy data were analyzed by the Northwestern University Data Monitoring Committee (DMC). Among all 32 patients, the ORR was 16% (complete remission (CR) 13%). Further, the ORR for all B-NHL pts was 6% (no CR) and 25% for T-NHL (19% CR). On the modified D1+D15 treatment schedule, the ORR for B-NHL was 0% (0/8); while among T-NHL, the ORR was 50% (3/6) with each of these latter pts remaining in continued remission at 29+, 26+, and 19+ months. Nevertheless, an analysis performed by the DMC for the overall study conduct recommended premature study closure; thus the final planned 8 pts did not enroll.

We determined in this phase I/II study for pts with relapsed/refractory, aggressive T-NHL and B-NHL that combined bortezomib/gemcitabine using a dosing schedule of D1+8 q21 days was not tolerated and is not recommended for further study. Modification of bortezomib/gemcitabine dosing to D1+15 q28 days was tolerated markedly better, allowing consistent treatment delivery. Altogether, clinical efficacy of gemcitabine plus bortezomib in aggressive B-NHL was low (with either schedule), while there was a potential signal of activity with durable responses in a small number of pts in the T-NHL population utilizing the modified treatment schedule.

Evens:Millennium: Research Funding, advisory board. Off Label Use: Velcade in T-cell and aggressive (non-MCL) B-cell NHL. Winter:Millennium: Research Funding. Smith:Millennium: Research Funding.