Williams, Michael E.

Abstract BACKGROUND: E4402 was a randomized phase III study comparing two different rituximab dosing strategies for patients with previously untreated, low tumor burden follicular lymphoma (FL). The primary endpoint was time to treatment failure. The initial publication (Kahl, JCO 2014) demonstrated that a retreatment strategy utilized less drug and produced comparable time to treatment failure compared to a maintenance strategy. Here we provide long term follow up results, focusing on response duration, time to first cytotoxic therapy, overall survival, and risk of histologic transformation. METHODS: Eligible patients had untreated, low tumor burden (GELF criteria) FL. Patients received R 375 mg/m 2 weekly x 4 and responders were randomized to maintenance rituximab (MR) (single dose R q 3 months) or retreatment rituximab (RR) (R weekly x 4 doses at disease progression). Each strategy was continued until treatment failure. The primary endpoint, time to treatment failure, was defined as progression within 6 months of last R, no response to R retreatment, initiation of alternative therapy, or inability to complete protocol therapy. Secondary endpoints included time to first cytotoxic therapy, quality of life and safety. The ECOG Data Monitoring Committee recommended release of study results at a planned interim analysis in September 2011 and patients and providers were notified of results. Time to treatment failure data collection halted with release of the results but limited data collection on time to first cytotoxic therapy, response duration, and risk of histologic transformation continued. INITITIAL RESULTS: From 11/03 to 9/08, 384 patients with FL were enrolled. Complete or partial response was achieved in 289 patients (71%), who were then randomized to MR (n=146) or RR (n=143). Demographic features were similar in the two arms: median age 59 years; ECOG PS 0-1 in all patients, and FLIPI low-risk (14.9 vs. 16.4%), intermediate-risk (46.3 vs. 42.9%) and high-risk (38.8 vs. 40.7%) for MR vs. RR, respectively. At initial publication, with a median follow-up of 3.8 years, the time to treatment failure was 3.9 years for MR vs. 3.6 years for RR (p=NS). LONG TERM FOLLOW UP RESULTS: Immunoglobulin levels and risks for serious infections/late complications in MR patients will be updated at the annual meeting. For the endpoint of time to first cytotoxic therapy, the median follow up is 8.7 years. At 7 years, 83% of MR and 63% of RR remained free from first cytotoxic therapy (HR 2.37; 95% CI 1.50 - 3.76) [Figure 1]. For the endpoint of response duration, the median follow up is 12.1 years. At 10 years, 66% of the MR patients remained in their 1 st remission compared to 30% of the RR patients who remained in their 1 st remission [Figure 2]. The median response duration for RR patients receiving a single 4-week course of rituximab was 3.25 years. There was no difference in the overall survival at 10 years, 84% for MR vs. 83% for RR. There was a trend towards a lower risk of histologic transformation for patients receiving MR (n = 4) compared to RR (n = 11) (p = 0.11). CONCLUSIONS: With long term follow up, the RESORT data indicates that in previously untreated, low tumor burden, follicular lymphoma, MR was superior to RR for delaying time to first cytotoxic therapy and for response duration, with a trend towards reducing the risk of histologic transformation. MR did not improve the overall survival. The original publication concluded that the time to treatment failure is similar between the two dosing strategies. Due to study design, time to treatment failure could not be analyzed in this long term follow up analysis. Compared to the historical benchmark of 3 years median time to first cytotoxic therapy when watch and wait is utilized, single agent rituximab, administered by either dosing strategy, was highly effective at delaying the time to first cytotoxic therapy. Figure 1 Figure 1. Disclosures Kahl: AbbVie, Adaptive, ADCT, AstraZeneca, Bayer, BeiGene, Bristol-Myers Squibb, Celgene, Genentech, Incyte, Janssen, Karyopharm, Kite, MEI, Pharmacyclics, Roche, TG Therapeutics, and Teva: Consultancy; AbbVie, Acerta, ADCT, AstraZeneca, BeiGene, Genentech: Research Funding. Habermann: Incyte: Other: Scientific Advisory Board; Tess Therapeutics: Other: Data Monitoring Committee; Seagen: Other: Data Monitoring Committee; Morphosys: Other: Scientific Advisory Board; Loxo Oncology: Other: Scientific Advisory Board; Eli Lilly & Co.,: Other: Scientific Advisor. Schuster: Abbvie: Consultancy, Research Funding; Acerta Pharma: Consultancy; AstraZeneca: Consultancy; Adaptive Biotechnologies: Research Funding; BeiGene: Consultancy; Celgene: Consultancy, Honoraria, Research Funding; DTRM: Research Funding; Genetech: Consultancy, Research Funding; Roche: Consultancy, Research Funding; Incyte: Research Funding; Juno Theraputics: Consultancy, Research Funding; Loxo Oncology: Consultancy; Merck: Research Funding; Nordic Nanovector: Consultancy; Novartis: Consultancy, Honoraria, Patents & Royalties, Research Funding; Pharmaclcyclics: Research Funding; Tessa Theraputics: Consultancy; TG Theraputics: Research Funding. Fenske: TG Therapeutics: Consultancy, Speakers Bureau; Servier Pharmaceuticals: Consultancy; Seattle Genetics: Speakers Bureau; Sanofi: Speakers Bureau; Pharmacyclics: Consultancy; MorphoSys: Consultancy; Kite (Gilead): Speakers Bureau; KaryoPharm: Consultancy; CSL Therapeutics: Consultancy; Bristol-Myers Squibb: Speakers Bureau; Biogen: Consultancy; Beigene: Consultancy; AstraZeneca: Speakers Bureau; ADC Therapeutics: Consultancy; Adaptive Biotechnologies: Consultancy; AbbVie: Consultancy.

8534 Background: Lenalidomide, an immunomodulatory agent with antitumor and antiproliferative effects, demonstrated rapid and durable efficacy in patients with relapsed/refractory mantle cell lymphoma (MCL) post-bortezomib in the phase II multicenter MCL-001 study. The objective of this analysis was to explore the efficacy of lenalidomide across patient subgroups. Methods: Single-agent lenalidomide was administered at 25 mg/d PO on days 1-21 of a 28-day cycle until disease progression or intolerability; primary endpoints were overall response rate (ORR) and duration of response (DOR) determined by an independent central review committee per modified IWG criteria. Exploratory analyses of ORR and DOR for subgroups were predefined and prospectively conducted. Results: Patients with relapsed/refractory MCL (N=134) had a median age of 67 y, with a median of 4 prior therapies (range, 2-10). The ORR was 28% (7.5% CR/CRu) and DOR was 16.6 months (95% CI, 7.7-26.7). Lenalidomide treatment provided consistent ORR and DOR across all subgroups analyzed by demographics, baseline disease status and prior therapy (Table). High vs normal baseline LDH was the only significant factor by univariate and multivariate logistic regression analysis of ORR (odds ratio=0.193; p=0.002). Conclusions: Single-agent lenalidomide provided consistent clinical benefit in patients with relapsed/refractory MCL post-bortezomib irrespective of patient subgroups at baseline with the exception of LDH. Clinical trial information: NCT00737529. [Table: see text]

8533 Background: Mantle cell lymphoma (MCL) is an aggressive form of non-Hodgkin lymphoma (NHL) with poor prognosis. The immunomodulatory agent lenalidomide shows consistent activity with tolerable safety in multiple phase II studies of relapsed/refractory aggressive NHL (NHL-002 and NHL-003) and MCL post-bortezomib (MCL-001). This pooled analysis further examined the efficacy and safety of single-agent lenalidomide in patients with relapsed/refractory MCL. Methods: Single-agent lenalidomide was given 25 mg/d PO on days 1-21 of 28-day cycles as tolerated for 52 weeks (NHL-002) or until disease progression (NHL-003 and MCL-001). All MCL patients received ≥1 prior treatment, including bortezomib in MCL-001. Efficacy data were examined by independent central review for MCL-001 and NHL-003 and by investigators for NHL-002. Results: 206 patients with relapsed/refractory MCL were studied. The median age was 67 y (range 33-84; 63% ≥65 y), 91% stage III/IV disease and 51% had received ≥4 prior regimens (76% prior bortezomib). Overall response rate (ORR) with lenalidomide was 32% (10% CR/CRu), with a median time to response of 2.1 months and median duration of response (DOR) of 16.6 months (not yet reached in patients with CR/CRu; Table). Kaplan-Meier estimates for median PFS and OS were 5.4 and 23.9 months, respectively. Mean daily dose of lenalidomide was 21 mg. Grade 3/4 AEs included neutropenia (44%), thrombocytopenia (29%), anemia (11%), and fatigue (7%). Other any-grade AEs included tumor flare reaction (7%), venous thromboembolic events (7%), and invasive second primary malignancies (3%). Conclusions: Lenalidomide produced rapid and durable responses in patients with relapsed/refractory MCL, and exhibited a predictable safety profile among 3 phase II studies of lenalidomide in heavily pretreated patients, including prior treatment with bortezomib. Clinical trial information: MCL-001: NCT00737529; NHL-002: NCT00179660; NHL-003: NCT00413036. [Table: see text]

e19517 Background: Investigators cite complicated test (T) schedules as barriers to the conduct of clinical trials, especially T that do not contribute to outcome or pt safety. In MM protocols, some T are prone to being done inaccurately (e.g. UPEP, 24 hour urine collections), or are not useful (e.g. uric acid, AST/ALT) in determination of response (R) or safety. These T lead to increased cost, protocol violations, and do not enhance outcome. Methods: We analyzed data submitted for pts on E4A03 to examine frequency of missing data for R T: serum and urine (U) PEP, immunofixation (IFE) bone surveys (BS), and safety labs for C1-4. Results: Table outlines % pts with completed T. Most had T for serum M protein, as well as heme T but serum and U PEP, and IFE dropped off by C4. Other T e.g. uric acid were completed in 77%. Only 79% of pts had a BS repeated at C4. Incorrect or missing U data led to categorization of 16/445 (3.6%) pts as ineligible and was most frequent reason for ineligibility (69%). Missing U data led to 13/26 pts not evaluable. Analysis of SAE during the C1-4 revealed no SAE for hyperuricemia and < 5 SAE for AST/ALT. Only 5% of pts had lab-based SAE filed in 4 C. Conclusions: Required U T and BS to assess R in E4A03 were missing or completed inadequately and caused pts to be ineligible and or not evaluable. Such T need to be eliminated or replaced. Strategies to monitor data submission in real time may reduce subject loss due to T error. The requirement for routine T should be reassessed to create simpler trial design. Clinical trial information: NCT00098475. [Table: see text]

7046 Background: In CML-CP there is perceived complexity in risk stratification at diagnosis, and in testing for BCR-ABL kinase domain mutation (KD mut, including T315I) at relapse. We sought to understand AHOP practices in the context of NCCN and European LeukemiaNet guidelines. Methods: Between 2010 and 2012, we studied time dependent practice preferences of n=1,335 AHOP’s using a proprietary, live, case-based market research tool. A core case scenario and variations based on available diagnostic & treatment options were constructed. Preference data were acquired using blinded audience response technology. All responses for each scenario were obtained contemporaneously prior to any display of summary respondent selections. All sources of research support were double blinded. AHOP's were presented with a scenario of 59-year old CML-CP patient, then indicated preferred management from up to 9 relevant available/emerging diagnostic or therapeutic options following relapse after 1st-line imatinib. Results: The majority of AHOP’s in 2012 either did not calculate (52%), or were unfamiliar (21%) with Sokal or Hasford risk stratification score. Diagnostic preferences following relapse from 2010-12 are noted in Table (distribution over time, p<0.0001). If T315I was then detected in scenario, we noted marked differences in subsequent treatment preference in 2012 vs. 2011, respectively: decreased AlloBMT (16% vs. 32%); & increased 2nd-line dasatinib (26% vs. 14%) or bosutinib (12% vs. 8%) vs. nilotinib (8% vs. 20%) despite lack of efficacy for each in T315I. Preference increased slightly for ponatinib (16% vs. 14%), but not for omacetaxine (2% vs. 6%). Conclusions: Most AHOP’s do not apply risk stratification. Increasing preference for KD mut panel testing at relapse conforms to guidelines, although not all AHOP’s test. Heterogeneity in T315I treatment suggests uncertainty or unfamiliarity with newer agents, and evidence-based efforts are needed to improve awareness. [Table: see text]

Abstract Introduction Patients with mantle cell lymphoma (MCL) typically respond to initial therapy and almost inevitably relapse with frequent chemoresistance over time and poor outcome. Multiple phase II studies have established the efficacy and safety of lenalidomide, an immunomodulatory agent with tumoricidal and antiproliferative properties, in relapsed/refractory MCL. The prospective phase II multicenter MCL-001 “EMERGE” study led to FDA approval of lenalidomide for patients with relapsed/refractory MCL after 2 prior treatments, that included bortezomib. The activity of lenalidomide was seen regardless of MIPI, number of prior therapies, prior high dose therapy, bulky disease or high tumor burden. One of the most established prognostic factors in MCL is the proliferation index Ki67 (MIB1), now confirmed both in standard and dose-intensive high-dose therapy strategies. We present here longer follow-up of efficacy and safety from the MCL-001 study in patients relapsed/refractory to bortezomib and the potential relationship between Ki-67 and efficacy outcomes. Methods Patients with heavily pretreated MCL, that included prior bortezomib, received lenalidomide 25 mg/day PO, days 1-21 in 28-day cycles until disease progression or intolerability. Primary study endpoints were overall response rate (ORR) and duration of response (DOR); secondary endpoints included complete response (CR), time to response (TTR), progression-free survival (PFS), overall survival (OS), and safety. Response rates and time-to-event data were analyzed by independent central reviewers per modified IWG criteria and Kaplan-Meier estimates respectively (data cut-off March 20, 2013). Ki-67 was examined as an exploratory endpoint by immunohistochemistry for 81/134 patients (60%) either performed on biopsy samples for 24 patients, or based on the Ki-67 scores reported in local pathology reports for 57 patients. Results Median age for the enrolled intent-to-treat patient population (N=134) was 67 years (range, 43-83; 63% ≥65 years). The median number of previous therapies was 4 (range, 2-10; 78% received ≥3), 93% stage III/IV, and 72% were <6 months from last prior treatment. At a median follow-up of 13.2 months, the ORR was 28% (CR/CRu 8%), with a median DOR of 16.6 months (95% CI, 9.2-26.7; median not reached in patients with CR/CRu) by central review. Median TTR was 2.3 months (95% CI, 1.7-13.1), with a median time to CR/CRu of 4.1 months (95% CI, 1.9-13.2). Median PFS was 4.0 months (95% CI, 3.6-6.9), and median OS was 20.9 months (95% CI, 13.7-24.4). The average dose intensity of lenalidomide was 20 mg/day, for a median duration of 94.5 days (range, 1-1,256). Dose reductions or interruptions due to adverse events (AEs) occurred in 40% and 58% of patients, respectively. Neutropenia (44%), thrombocytopenia (28%), and anemia (11%) were the most common treatment-related grade 3/4 AEs. Ki-67 results were available in 81/134 patients, and efficacy data were categorized using 30% and 50% cut-off thresholds for Ki-67 expression (Table 1). Although patient numbers were limited, the ORR was similar in both lower and higher Ki-67 group, but those with lower Ki-67 levels demonstrated better CR rates, DOR and survival outcomes compared with patients with elevated Ki-67. Conclusions Single-agent lenalidomide in heavily pretreated patients with relapsed/refractory MCL post-bortezomib showed durable long-term efficacy with a consistent safety profile. Consistent with what is reported in the literature, high Ki-67 is associated with poor outcome in our cohort with shorter OS. Though based on retrospective evaluation and subsets of patients, the ORR to lenalidomide was similar in both low and high Ki-67 groups, suggesting lenalidomide can be active in patients expressing high levels of Ki67. Prospective studies are needed to confirm these findings. Disclosures: Goy: Celgene: Consultancy, Research Funding, Speakers Bureau. Off Label Use: This is a phase 2 clinical study of safety and efficacy for lenalidomide in patients with MCL. Williams:Celgene: Consultancy, Membership on an entity’s Board of Directors or advisory committees, Research Funding. Drach:Celgene: Honoraria. Ramchandren:Celgene: Research Funding. Zhang:Celgene: Employment. Cicero:Celgene: Employment. Fu:Celgene: Employment. Heise:Celgene: Employment, Equity Ownership. Witzig:Celgene: Research Funding.

Abstract Abstract 3816 Background: The hypomethylating agent (HMA) azacitidine (AZA) significantly improves survival compared to conventional care regimens for patients (pts) with “higher risk” MDS (including both Int-2 & High-Risk according to IPSS) (Lancet Oncol 10:223, 2009 – first presented Dec 2007); a more recent trial of the HMA decitabine (DEC) vs. best supportive care in pts with similar IPSS failed to show a significant survival advantage (JCO 29:1987, 2011 – first presented Dec 2008). Since 2006 the immunomodulatory drug lenalidomide (LEN) has been FDA-approved for treatment of anemia in lower-risk MDS pts & deletion of chromosome 5q (Del5q). However, no phase III survival data have been presented with LEN in patients with higher risk MDS, nor is it an expert consensus 1st-line treatment recommendation in higher risk setting (NCCN Guidelines 2011). We sought to evaluate the change in 1st-line treatment preference with these 3 approved agents for higher risk MDS among AHOP's from 2006–2010 in relation to pertinent survival data as it became available. Methods: Between 2006–10 we studied time dependent prescribing preferences of n=1960 AHOP's in a prospective annual research series. Approx. 330–450 individual AHOP's participated in any given year. AHOP demographic data was available from 2007–10 (n=1510); they were predominantly male (77%), working in Community setting (79% vs. 21% Academic), & in small group practice (≤5 providers −59%, including 18% “solo practice”). All regions were represented (NE 23%; SE 22%; Central 19%; SW 13%; West 23%). Most (57%) were >10yrs out from training, and only 18% were <5yrs out from training. A total sample was studied using an extensively tested, live, case-based, market research vehicle to anonymously acquire data. AHOP's were presented with a scenario of an older pt with higher risk MDS (High-Risk 2006–09; Int-2 in 2010) & specific cytogenetic abnormalities, then indicated their preferred treatment from up to 10 relevant available/emerging 1st-line therapeutic options. In each, the scenario was of a pt age 63 yrs (2006–08) or 68–70 yrs (2009–10); with symptomatic anemia & pancytopenia; no response to darbepoietin; with RAEB-2 - excess Bone Marrow blasts 11% (2006–09) to 16% (2010); complex (2006–09) or normal cytogenetics (2010); and both without and then with Del5q in same scenario (2008–10). The distribution of treatment choices was compared over time & according to Del5q status using a chi-square test. Results are shown in Table below. In NON-Del5q scenario, a clear & significant shift toward greater AZA preference by almost 2-fold (from 40% to 78%) was apparent after 2007 (P<0.0001). For same scenario With Del5q (2008–2010), treatment preferences changed significantly over time (P<0.0001), with lower 1st-line AZA and greater DEC & LEN preference (the latter >40% after 2008). In addition, a marked and significant increase in 1st-line LEN preference was observed when AHOP's were presented with the same higher-risk scenario each yr but With-Del5q vs. NON-Del5q (p<0.0001). For “Non Del 5q” vs. “With Del 5q” scenarios, p-values result from a chi-square test, testing for any difference in the distribution of treatment choices over time. *(p<0.0001) - LEN With-Del5q vs. NON-Del5q in higher-risk scenario each year, 2008–2010. Conclusions: Significant changes in 1st-line treatment preferences of AHOPs for pts with higher-risk MDS are apparent in the findings from 2008–10, the time frame following the availability of randomized phase III data demonstrating a survival advantage for AZA therapy. When the presence of a Del5q abnormality is added to the same higher-risk scenario, treatment preferences change dramatically with a significantly increased preference for LEN (despite no available ph III survival data supporting this approach) and for DEC (despite a “negative” ph III trial). These findings suggest important educational gaps concerning the available phase III survival data supporting first line prescribing preferences for patients with higher risk MDS. Efforts to address this gap via evidence-based approaches are warranted. Disclosures: Foran: Celgene: Honoraria; Xcenda: Honoraria. Williams:Xcenda: Honoraria; Celgene: Consultancy, Honoraria, Independent DSMB member, Research Funding. Willey:Xcenda: Employment; Celgene: Business relationship; Eisai: Business relationship. Green:Xcenda: Employment; Celgene: Business relationship; independent DSMB; Eisai: Business relationship.

Abstract Abstract 905 Introduction: Though dose-intensive strategies using chemoimmunotherapy have significantly improved mantle cell lymphoma (MCL) outcomes with prolonged progression-free survival (PFS), most patients still relapse over time. In the relapsed setting, MCL patients often develop chemoresistance and have a poor overall prognosis. The immunomodulatory agent lenalidomide has demonstrated tumoricidal and antiproliferative effects in MCL and clinical activity and safety in multiple phase II studies in aggressive non-Hodgkin's lymphoma. The objective of the MCL-001 “EMERGE” study was to examine the safety and efficacy of single-agent lenalidomide in subjects with MCL who relapsed or were refractory to bortezomib. Methods: This phase II, multicenter, single-arm, open-label study examined single-agent lenalidomide administered at 25 mg/d PO on days 1–21 of a 28-day cycle until disease progression, unacceptable toxicity, or voluntary withdrawal. The subjects were required to have had prior treatment with rituximab, cyclophosphamide and anthracycline (or mitoxantrone), and to have relapsed or progressed (<12 months) after or were refractory to bortezomib. The primary endpoints were overall response rate (ORR) and duration of response (DOR); secondary endpoints included complete response (CR), PFS, time to progression (TTP), overall survival (OS) and safety. Efficacy data were measured by investigators and an independent central review committee according to modified International Working Group criteria and analyzed by SAS. Results: 134 subjects with relapsed or refractory MCL who were heavily pretreated (no limitation in number of prior therapies) were enrolled. The median age was 67 y (range, 43–83), two-thirds of them being 65 y or older and 93% with advanced stage disease (stage III-IV). The median number of prior therapies was 4 (range, 2– 10) with 78% of subjects having received ≥ 3 prior lines of treatment. The ORR to single-agent lenalidomide was 28% (CR/CRu 8%; Table 1) by independent central review, with a median DOR of 16.6 mo (95% CI, 7.7–26.7; Figure 1). The ORR was 32% (CR/CRu 16%) by investigator assessment, with a median DOR of 18.5 mo. Median time to response (central review) was 2.2 mo (3.7 mo to achieve CR). The median PFS was 4.0 mo (95% CI, 3.6–5.6); median OS was 19.0 mo (95% CI, 12.5–23.9). Subjects received an average dose of 20 mg/d of lenalidomide. Lenalidomide was dose reduced in 38% of subjects; treatment discontinuation due to an adverse event (AE; primarily myelosuppression) was reported in 19% of subjects. The most common grade 3/4 AEs were neutropenia (43%), thrombocytopenia (28%), anemia (11%), pneumonia (8%) and fatigue (7%). Other AEs (any grade) included tumor flare reaction (13 subjects, 10%), deep vein thrombosis (5 subjects, 4%), pulmonary embolism (3 subjects, 2%) and invasive second primary malignancies (3 subjects, 2%). Conclusions: The EMERGE study demonstrated rapid and durable efficacy of lenalidomide in MCL subjects who relapsed or progressed after or were refractory to bortezomib. These results in heavily pretreated MCL subjects (median of 4 prior treatments), and with an expected toxicity profile support single-agent lenalidomide in subjects with relapsed or refractory MCL after bortezomib. Disclosures: Goy: Pfizer: Advisory board member, Advisory board member Other; Seattle Genetics: Advisory board member Other; J&J: Advisory board member, Advisory board member Other; Pharmacyclics: Advisory board member, Advisory board member Other; Millenium: Advisory board member, Advisory board member Other, Speakers Bureau; Celgene: Advisory board member Other. Off Label Use: This is a phase 2 clinical study of safety and efficacy for lenalidomide in patients with MCL. Sinha:Celgene: Research Funding. Williams:Celgene: Clincial Trial Research Support, Advisory Boards, Data Safety Committee Member, Consultant Other, Consultancy. Drach:Celgene: Speakers Bureau; Janssen: Speakers Bureau; Roche: Research Funding. Ramchandren:Seattle Genetics: Speakers Bureau. Herbrecht:Pfizer: Advisory board member Other. Zhang:Celgene: Employment. Cicero:Celgene: Employment. Fu:Celgene: Employment. Witzig:Celgene : Research Funding.