ALBERT

Description: Abstract 3696 Poster Board III-632 Platinum based regimens such as ICE with or without Rituximab (RICE) have been widely used to treat relapsed or refractory lymphoid malignancies prior to transplantation. However, a significant portion of patients do not respond to treatment, and improved therapies are needed. Vorinostat (V) is an oral HDAC inhibitor with moderate toxicity and has clinical activity against a variety of tumors including cutaneous T cell lymphoma. Preclinical data demonstrated marked anti-tumor synergism between V and platinum analogues as well as etoposide. We present data from a phase I, open-label, multicenter, dose escalation study estimating the maximally tolerated dose of V that can be combined with RICE or ICE (V-RICE or VICE) in patients with relapsed or refractory lymphoid malignancies or untreated T- or Mantle Cell Lymphoma. Other endpoints include tolerability, exploratory anti-tumor activity and impact of above regimen on stem cell reserve. Patients (aged ≥18 years, an ECOG performance status of 0-2, measurable disease, no active central nervous system involvement, adequate bone marrow, hepatic and renal function, no active arrhythmias on EKG) were sequentially enrolled on escalating doses of V combination therapy using the “two stage” design introduced by Storer with single patient cohorts until a dose limiting toxicity (DLT) is observed, followed by cohorts of 4 patients. A DLT was defined as any gastrointestinal grade 3 NCI-CTCAE adverse event (AE) lasting longer than 7 days, or any related non-hematologic grade 4 or 5 AE; any event that prevents the completion of one full cycle of therapy (5 days of V) due to toxicity from V; or any AE at the discretion of the principal investigator. Therapy consisted of V ranging from 400 mg daily to 700 mg BID days 1 to 5 in combination with standard RICE or ICE on days 3 to 5 every 21 days. A total of 18 patients have been enrolled on this study thus far (9 in stage 1, 9 in stage 2) and 14are fully evaluable to date, including: Hodgkin lymphoma (4), T-cell lymphoma (3); mantle cell lymphoma (2); diffuse large B cell lymphoma (2); follicular lymphoma (2), and chronic lymphocytic leukemia (1). Baseline characteristics (n=14) included: median age 55 (range: 33-67), male 10 (71%), stage III/IV 14 (100%), median number of prior therapies 2 (range: 0-6), elevated LDH 5 (35%), prior anthracycline 13 (93%), prior platinum 2 (14%), refractory diseases 5 (36%), relapsed diseases 8 (57%), untreated disease 1 (7%). A maximum V dose of 700 mg BID was attained in stage I (Table). The dose adjustment schema of stage II has ranged from 600 mg BID to 400 mg BID currently. Among the 14 evaluable patients, six received only1 cycle of treatment (3/5 patients declined the second cycle; 2/5 patients developed DLT), 8 completed 2 cycles. Eight of 14 (57%) patients experienced non-hematologic AEs≥ grade 3 with most common being nausea, vomiting, or diarrhea seen in 6 and grade 4 hypokalemia in 2. Twelve patients (86%) responded including 1 with complete remission (CR), 2 with unconfirmed CR, and 9 with partial responses. One patient had stable disease and one had disease progression. Nine of 12 patients (75%) who attempted peripheral blood stem cell collection following VICE/V-RICE were successful (〉5×106 CD34+/kg) Collectively, these findings indicate that the combination of vorinostat with RICE or ICE, is feasible and active in patients with lymphoid malignancies. Special attention should be given to the management of the frequent gastrointestinal AEs. Pending identification of the MTD, phase II evaluation of VICE or V-RICE regimen will be designed to formally define its efficacy. Table Summary of DLTs and responses at various dose levels Stage Dose level Dose n DLTs responses I 1 400 mg QD 2 0 PR(1); SD (1) 2 300 mg BID 1 0 PR (1) 3 400 mg BID 1 0 PD (1) 4 500 mg BID 2 0 PR (1); CR (1) 5 600 mg BID 1 0 PR (1) 6 700 mg BID 2 1 PR (2) II 5 600 mg BID 4 3 PR (1); CRu (3) 4 500 mg BID 1 1 PR (1) Total - - 14 5 12 (86%) Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 3327 Poster Board III-215 Mantle cell lymphoma (MCL) exhibits short remission durations and a poor prognosis. To improve outcomes, many have advocated the use of high-dose therapy (HDT) and autologous hematopoietic stem cell transplantation (ASCT). However, most data indicate that ASCT is not curative with nearly all patients eventually relapsing. Since little is known regarding the outcomes of these individuals, we retrospectively evaluated MCL patients relapsing after HDT and ASCT and described the association of factors with survival. A total of 134 consecutive MCL patients received HDT and ASCT at our center from 1996 to 2009. Among them, 51 has relapsed so far (38%). Their baseline characteristics included: median age at diagnosis = 57 years (range, 35-70), median age at transplant = 59 years (range, 35 – 70), stage III-IV at diagnosis=98%, median LDH/upper limit of normal (ULN) at diagnosis = 1.14 (range, 0.56 -5.65), median LDH/ULN at transplant = 0.92 (range, 0.58-3.00), median white blood cell (WBC) count at diagnosis = 9.70 × 109 /liter (range, 1.40 – 48.00), mean WBC count at transplant = 5.30 × 109 /liter (range, 0.10 – 17.60), median number of prior chemotherapy regimens = 2 (range, 1-5). The median time to relapse following ASCT was 0.97 years (range: 0.08 – 5.61). Seven patients (14%) received a non-ablative allogeneic transplant (NMAT) for the treatment of relapse after ASCT. Of the 51 patients who relapsed, 40 (78%) have died. Overall survival at 3 years after relapse was 13.2% (95% CI, 4.9-25.5) (Fig.1). The MCL international prognostic index (MIPI) at diagnosis (MIPI-Dx) was a predictor of OS when evaluated either as a categorical variable (0-3 vs. 4 vs. 5-9, p=0.05, global test) or continuous linear variable (p=0.005). When compared to patients with a MIPI-Dx of 0-3, those with a score of 5-9 had a 3.5 (95% CI 1.3-9.7) fold higher risk of mortality after relapse. Similarly the MIPI at autologous transplant (MIPI-Trx) was also a significant predictor of OS after relapse when evaluated as a continuous linear variable (p=0.01). A multivariable model of all factors collected at ASCT identified ECOG performance status (PS)〉0 (HR 2.1, 95% CI 1.0-4.2), TBI-based conditioning regimen (HR 4.3, 95% CI 1.8-10.0) and transplant beyond 1st remission (HR 3.5, 95% CI 1.5-8.0) as associated with a significantly higher risk of mortality after relapse; whereas time to relapse from transplant 〉1 year (HR 0.1, 95% CI 0.04 -0.5) was associated with lower risk of mortality. For example, patients relapsing one year or later after transplant had an estimated 1-year survival after relapse of 76% compared to 17% for those who relapsed earlier (Fig.2); patients transplanted in CR/PR1 had an estimated 1-year survival of 75% after relapse compared to 28% transplanted beyond first remission. These results indicate that OS following relapse of MCL after HDT and ASCT is poor, with the worst outcomes in those with early relapse, transplant later in the disease course, poor PS, TBI-based HDT, and high MIPI score. Only a small minority of patients attempted salvage with NMAT after relapse. These data can be used to better counsel patients and optimize treatment strategies based on the identified mortality risk factors, as improved outcomes are clearly needed in this population. Fig.1. OS of all MCL patients after relapse Fig.1. OS of all MCL patients after relapse Fig.2. OS of MCL patients who relapsed 〈 1 year after ASCT and those relapsed ≥ 1 year after ASCT Fig.2. OS of MCL patients who relapsed 〈 1 year after ASCT and those relapsed ≥ 1 year after ASCT Disclosures No relevant conflicts of interest to declare.