ALBERT

Description: Abstract 857 Background: Follicular lymphoma (FL) is an incurable B-cell non-Hodgkin's lymphoma (NHL) representing ∼20% of all NHL. Rituximab (R) is approved for the treatment of CD20+ relapsed/refractory FL, and single-agent bortezomib (Vc; VELCADE®) has shown activity in heavily pretreated patients. Vc and R show additive activity in preclinical models, and the combination was active and well tolerated in a phase 2 study. This randomized, open-label, multi-center, international, phase 3 clinical trial (LYM3001) compared the efficacy and safety of Vc plus R (Vc-R) vs R alone in patients with relapsed or refractory, R-naive or R-sensitive FL. Methods: Patients with grade 1/2 measurable FL who had relapsed/progressed following prior therapy (time to progression [TTP] ≥6 months if prior R-containing therapy), ECOG performance status ≤2, and no peripheral neuropathy grade ≥2 were randomized (1:1) to receive 5-week cycles of Vc-R (Vc 1.6 mg/m2, d 1, 8, 15, 22, cycles 1–5, plus R 375 mg/m2, d 1, 8, 15, 22 in cycle 1 and d 1 only in cycles 2–5) or R alone (administered according to the same schedule as for the Vc-R arm). In both groups, treatment was administered for five cycles or until progression or unacceptable treatment-related toxicity. Randomization was stratified by FLIPI score (0–1 vs 2 vs ≥3), prior R therapy (yes vs no), time since last dose of anti-FL therapy (≤1 vs 〉1 year), and region (US vs EU vs rest of world). Primary endpoint was progression-free survival (PFS); secondary endpoints included overall response rate (ORR), complete response (CR) rate, TTP, and safety/tolerability. Response and progression were assessed by independent radiology committee (IRC) using the modified International Workshop Response Criteria. Planned sample size was 670 patients to provide 90% power (α=0.05, 2-sided) to detect a 33% improvement in median PFS with Vc-R vs R (i.e. 13.3 vs 10 months). Results: Between April 2006 and August 2008, 676 patients (intent-to-treat [ITT] population) were enrolled from 164 centers in 29 countries across Europe, the Americas, and Asia. Baseline characteristics were well balanced between the two arms; median age was 57 years (range 21–84), 54% were female, 75% were Caucasian and 21% were Asian. The majority of patients (93%) had an ECOG performance status of 0 or 1, 51% and 48% had grade 1 and 2 FL, respectively, and 41%, 35%, and 23% had high, intermediate, and low FLIPI score, respectively; 83% had Ann Arbor Stage III or IV, and 38% had bone marrow involvement at baseline. 33% of patients had received 3 or more prior lines of therapy (range 1–6+); 44% had received prior R therapy. The most common prior regimens were CHOP (38%), CVP (25%), single-agent R (17%), R-CHOP (12%), and R-CVP (11%). At a median follow-up of 33.9 months, a total of 440 PFS events were observed by IRC in the ITT population, 212 in the Vc-R arm and 228 events in the R arm. Median PFS improved from 334 days (95% CI: 278, 365) with R alone to 389 days (95% CI: 351, 456) with Vc-R; the hazard ratio was 0.822 (95% CI: 0.681, 0.991). This improvement is statistically significant with a 2-sided P-value of 0.039. The ORR was 63% with Vc-R vs 49% with R (P6 months) was 50% with Vc-R vs 38% with R (P=0.002). The median time to subsequent antilymphoma treatment was significantly improved in the Vc-R vs R arm (700 vs 537 days, P=0.027). Median OS was not reached in either group. Patients received a median 25 weeks treatment in both the Vc-R and R groups (range 5–40 and 5–35 in the Vc-R and R groups, respectively). Adverse events (AEs) were reported for 95% of Vc-R and 78% of R patients. The most common AEs were diarrhea (52% Vc-R, 8% R), nausea (29% Vc-R, 7% R), and pyrexia (25% Vc-R, 10% R). Most AEs were grade 1 or 2. Grade ≥3 AEs were reported in 46% of Vc-R and 21% of R patients; the most common grade ≥3 AEs were neutropenia (11% vs 4%) and diarrhea (7% vs 0%). Peripheral sensory neuropathy was reported in 17% of patients in the Vc-R arm vs 1% in the R arm; 3% vs 0% grade ≥3. 18% and 11% of Vc-R and R patients, respectively, had serious AEs, only 4% and 1% of patients discontinued due to drug-related AEs, and there were 9 and 4 on-treatment deaths, respectively. Conclusion: The addition of weekly Vc to R therapy in patients with relapsed FL was associated with statistically significant improvements in PFS, response rate, and time to next antilymphoma treatment, with acceptable additional toxicity. Disclosures: Coiffier: Johnson & Johnson: Honoraria. Off Label Use: Discussion of Velcade in NHL subtypes other than mantle cell lymphoma is included. Mayer:Roche: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Rule:Johnson & Johnson: Consultancy, Speakers Bureau; Roche: Consultancy. Walewski:Roche: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen-Cilag: Investigators fee. Crump:Millennium Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees; Johnson & Johnson: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Roche: Consultancy. Shpilberg:Johnson & Johnson: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. Hermann:Millennium Pharmaceuticals: Research Funding. Parasuraman:Millennium Pharmaceuticals: Employment, Equity Ownership. Zhu:Johnson & Johnson: Employment. Enny:Johnson & Johnson: Employment, Equity Ownership. Theocharous:Johnson & Johnson: Employment. van de Velde:Johnson & Johson: Employment, Equity Ownership. Elsayed:Johnson & Johnson: Employment, Equity Ownership.

Description: Background: Bortezomib has been shown to be effective in the treatment of multiple myeloma (MM). However, the data on response to re-treatment with bortezomib is very limited, mostly coming from small retrospective studies. Response to re-treatment is particularly relevant in patients with relapsing incurable conditions such as MM. The primary objective of this first international prospective phase II trial was to determine the best response to bortezomib re-treatment in patients with MM. Secondary objectives included safety profile, duration of response, time to progression, best M protein response and the evaluation of the best investigator-assessed response compared to the best response reported to the previous bortezomib treatment. Methods: Patients with secretory MM were eligible if they had responded (CR or PR) to a bortezomib-based most recent treatment and had a treatment-free interval of at least 6 months since the last bortezomib dose. They needed to meet the PD criteria as defined by European Group for Blood and Marrow Transplantation (EBMT). Patients received bortezomib starting at the last tolerated dose in the previous bortezomib regimen (1.3 or 1.0 mg/m2). Bortezomib was administered alone or in combination with dexamethasone at investigator discretion. Bortezomib was administered as an intravenous injection on days 1, 4, 8 and 11 of a 21-day cycle up to a maximum of 8 cycles. Response was evaluated every 6 weeks according to EBMT criteria. Adverse events (AEs) were assessed from informed consent until at least 30 days after treatment and were graded by the NCI-CTCAE. This clinical trial closed to recruitment on 30th June 2008 after reaching the planned sample size of 128 patients, with 57 patients still receiving treatment. The results presented include the efficacy and safety data available for 100 patients who had completed at least 2 treatment cycles at the time of data cut-off (16 July 2008) or who had withdrawn from therapy for any reason. Results: The 100 patients (M: 59, F: 41; median age 66 years) who received at least 1 dose of bortezomib are included in the current safety analysis. The median time from diagnosis of MM was 4.5 years (range 0–14 years) and 59% had received 3 or more lines of prior therapy (including the previous bortezomib therapy). Karnofsky performance status was £ 70% in 16% of patients. Twenty-six percent of patients had achieved a complete response (CR) with the last bortezomib regimen. The median treatment free interval since the previous bortezomib treatment was 14.3 months. Eighty-seven percent of the patients received at least 3 cycles and the median number of completed cycles was 5 cycles. Dexamethasone was added to the bortezomib treatment in 69% of patients. In the 97 patients eligible for the current efficacy analysis, the overall response rate (ORR) by EBMT criteria [CR+PR] was 26.8% (CI=18.3– 36.8), with 3.1% complete responses, whilst ORR + minimal response (MR) was 46.4%. The ORR did not differ in patients treated with bortezomib plus dexamethasone (27.9%, CI=17.7– 40.1) versus those treated with bortezomib alone (24.1%, CI=10.3–43.5). The median time to at least PR was 43 days, with a median time to best response of 64 days. The most commonly reported related grade 3/4 adverse events were thrombocytopenia (24%), neutropenia (5%) and diarrhea (5%). Peripheral neuropathy (PN) was observed in 20% of patients, with grade 3 PN reported in 6% of patients only. Serious TEAE’s were reported in 23% of patients and 10% of patients discontinued bortezomib due to related TEAE’s including 6 cases of PN. TEAE’s leading to death were reported in three patients. Conclusions: Response after re-treatment with bortezomib alone or in combination with dexamethasone in heavily pre-treated multiple myeloma patients was still high, and the adverse event profile was consistent with the already known safety profile. Updated results will be presented at time of the meeting.

Description: Abstract 3866 Poster Board III-802 Bortezomib (Velcade®) retreatment has been shown to be active and well tolerated in patients with relapsed multiple myeloma (MM) in a number of retrospective studies and a small prospective phase 4 study (EVEREST). This large, prospective, international, multi-center, open-label phase 2 study was conducted to confirm the efficacy and safety of retreatment with bortezomib in MM patients who had previously responded (at least partial response [PR]) to bortezomib-based therapy as their most recent prior treatment. Patients had to have previously tolerated bortezomib 1.0 or 1.3 mg/m2 alone or in combination and have had a treatment-free interval (TFI; time from last dose of initial bortezomib treatment to first dose of bortezomib retreatment) of ≥6 months. Additional eligibility criteria included progressive disease or relapse from complete response (CR) by EBMT criteria, no MM therapy (except maintenance with dexamethasone, thalidomide, or interferon) since the last dose of initial bortezomib treatment, KPS ≥60, and adequate renal, hepatic, and hematologic function; patients with grade ≥2 peripheral neuropathy or neuropathic pain (as defined by NCI CTCAE v3.0) were excluded. Patients received bortezomib at the last tolerated dose (1.0 or 1.3 mg/m2) during initial treatment on days 1, 4, 8, and 11 for up to eight 21-day cycles, either alone or in combination with dexamethasone at the investigator's discretion. Response was assessed by EBMT criteria every 6 weeks during treatment and then every 2 months until disease progression. Adverse events (AEs) were graded according to NCI CTCAE v3.0. A total of 130 patients received at least 1 dose of bortezomib retreatment and were included in the safety population. Patients had a median age of 67 years, 57% were male, and 16% had KPS '70%. Median time from diagnosis of MM was 4.5 years (range 0–14 years); median number of prior therapies was 2; 15, 80, 23, and 12 patients had received 1, 2, 3, and ≥4 prior lines of therapy (excluding initial bortezomib therapy). Best response by EBMT criteria to initial bortezomib treatment was CR in 26% and PR in 74% of patients; median time to progression and TFI after initial bortezomib treatment were 17.9 months and 14.3 months, respectively. Last tolerated dose of previous bortezomib therapy was 1.3 mg/m2 and 1.0 mg/m2 for 62% and 29% of patients, respectively; 9% received another dose. Patients received a median 7.0 (range 1–8) cycles of bortezomib retreatment (23% of patients completed all 8 cycles); 72% of patients received concomitant dexamethasone. A total of 126 patients were evaluable for response. In the 126 response-evaluable patients, the overall response rate (ORR; CR+PR) by best confirmed response (EBMT criteria) was 40%; in addition, 18% of patients achieved minimal response (MR), to give a CR+PR+MR rate of 58%. After a planned secondary efficacy analysis, the ORR (CR+PR) by single best response was 55% (75% ≥MR). Median time to best confirmed response (≥MR) was 2.9 months; time to first response was 1.5 months. Analysis of ORR by patient subgroups showed comparable results in patients who did versus did not receive concomitant dexamethasone (42% vs 32%), in those who received ≤1.0 mg/m2 vs 1.3 mg/m2 bortezomib (35% vs 41%), and in those aged ≤65 years vs 〉65 years (45% vs 36%). ORR was 67%, 39%, 33%, and 25% in patients who had received 1, 2, 3, and ≥4 prior lines of therapy (excluding initial bortezomib), respectively. Analysis of best confirmed responses according to response to initial bortezomib showed that 63% and 52% of patients who achieved a CR or PR, respectively, to initial bortezomib treatment responded to retreatment. Most (98%) patients experienced a treatment-emergent AE; 60% experienced a grade 3/4 AE, and 32% experienced a serious AE; there were 8 deaths, 2 of which (due to sepsis and stroke) were possibly treatment-related. The most common grade 3/4 AEs were thrombocytopenia (35%), neutropenia (7%), diarrhea (7%), and pneumonia (5%). AEs leading to dose reductions or discontinuations were reported for 22% and 12% of patients, respectively. The incidence of neuropathy was 39%, including 9% grade 3; 4% of patients discontinued treatment due to PN; 61% of neuropathy events resolved or improved within a median 1.3 months. These results confirm that bortezomib retreatment is a well-tolerated, feasible, and active therapeutic option for heavily pretreated MM patients without evidence of cumulative toxicity. Disclosures: Petrucci: Janssen-Cilag: Honoraria; Celgene: Honoraria. Dimopoulos:Ortho-Biotech: Consultancy, Honoraria; Millennium Pharmaceuticals, Inc.: Honoraria. Drach:Janssen-Cilag: Consultancy, Honoraria; Amgen: Honoraria; Celgene: Honoraria. Blade:Janssen-Cilag: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Honoraria; Johnson and Johnson: Honoraria; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Description: Abstract 4860 Backgroud: Chronic lymphocytic leukemia (CLL) is one of the most frequent chronic lymphoproliferative disorders in Europe. It is characterized by persistent monoclonal lymphocitosis with localized or generalized lymphadenopathy. Despite the initial clinical presentation, it has a heterogeneous natural history, with the majority of patients living 10–12 years, but with some patients dying rapidly, within 2–3 years of diagnosis. Beside clinical prognostic factors, novel cytogenetic markers are recognized to be useful in predicting disease free and overall survival in CLL. AIMS: In a retrospective study throughout 10 years (1999-2009), we analyzed the clinical and biological presentation and compared the evolution and survival of patients with B-CLL using different cytogenetic markers. METHODS: We identified 112 cases (63 males and 49 females) of B-CLL with cytogenetic study by fluorescence in situ hybridization (FISH). RESULTS: Amongst 112 patients, the male to female (M/F) ratio was 1.3:1 and the median age was 70 (43-96) years. At diagnosis, the median lymphocyte count was 15.5 G/L (5.4-173). Fifty five patients (49%) had lymphadenopathies and seventeen (15%) had splenomegaly and/or hepatomegaly at presentation. By the revised Rai staging system seventy (63%) patients were included in low risk group, thirty (27%) in intermediate risk group and twelve (10%) in high risk group. The expression of ZAP-70 and CD38 by flow citometry was performed in 75 patients and revealed 13 (17%) patients CD38+ and 12 (16%) ZAP70+. The study of chromosomal aberrations with FISH showed thirty six patients (32%) with no abnormality, thirty six (32%) with isolated 13q deletion, fifteen (14%) with 12 trisomy, twelve (11%) with 11q deletion and thirteen (11%) with 17p deletion. Forty (36%) patients showed progressive disease in a median time of sixteen months (0-120), thirteen with 13qdel, seven with 17pdel and five with 12 trisomy. After treatment two patients showed progressive disease, six maintain a stable disease and thirty two obtain a remission, nine in complete remission. The Overall Survival (OS) at ten years was 70%. By the revised Rai staging system the OS at ten years was 80% for low risk, 70% for intermediate risk and all the high risk patients died during follow up. The OS at five years for the del13q-, 12 trisomy, del11q- and del17p- was 90%, 88%, 58% and 60%, respectively. SUMMARY: Chronic lymphocytic leukemia is currently considered a chronic disorder with a favourable outcome, but with a variable evolution to progressive disease. This retrospective study allowed the characterization of patient with CLL in our department and the acknowledgement that our results are quite similar to the published data. Disclosures: No relevant conflicts of interest to declare.

Description: Oxidative stress (OS) is recognized to be a evident feature in cancer development and progression. Indirect evidences suggest a role for oxidative stress (OS) in Chronic Myelogenous Leukemia (CML) etiology and pathogenesis. OS, resulting from an imbalance between Reactive Oxygen Species (ROS) production and antioxidant defenses, contributes to cell damage, apoptosis and ineffective hematopoiesis. The antioxidant enzymes superoxide dismutases (SOD) and catalase (CAT) are important components of cell defense against OS, and polymorphisms in the genes may contribute to differences in susceptibility of individuals to oxidative damage since it can lead to reduced protection against OS. In the present study we set to investigate the influence of polymorphisms of oxidative stress related genes, namely SOD1 (A251G), SOD2 (Ala16Val), COX2 (G-765C), CAT (C-262T) and NADPH oxidase p22 phox (C242T), as a risk factor for CML development and also as a prognostic risk marker in CML patients. Our study population consisted of 48 patients diagnosed with CML and the same number of healthy controls. Diagnosis was set according to international criteria. The genetic polymorphisms of SOD1 (A251G), SOD2 (Ala16Val), COX2 (G-765C), CAT (C-262T) and NADPH oxidase p22 phox (C242T) were assessed by RFLP-PCR. The patient group median age was 51 years (18-86), gender M/F=27/21. The strength of association between polymorphisms and CML risk was assessed by odds ratio (OR) with the corresponding 95% confidence interval (CI95%) and Kaplan-Meier survival analysis will be assessed to investigate the prognostic importance of these polymorphisms. Our preliminary results show a higher wild type allelic frequency of SOD1 (94%), COX2 (80%), CAT (82%) and NADPH oxidase p22 phox (67%) polymorphism and a higher variant allelic frequency SOD2 (52%) in CML patients, compared to controls. In these patients the predominant genotype was AA (88%), CG (52%), GG (67%), CC (73%) and CT (46%), respectively for SOD1, SOD2, COX2, CAT and NADPH oxidase p22 phox. Besides that, individuals with CC genotype of NADPH oxidase p22 phox and with TT genotype of CAT have an increase risk for CML about 2,3636-fold (CI95% 1,0346-5,3997; p=0,032) and 12,368-fold (CI95% 1,5151-100,9687; p=0,0076), respectively. These preliminary results show that CAT (C-262T) and NADPH oxidase p22 phox (C242T) genetic polymorphisms might be related with CML development and may be a novel genetic markers for CML susceptibility. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4870 Background: Dendritic cells (DC) are a heterogeneous population of lineage-negative antigen-presenting cells derived from CD34+ hematopoietic progenitors, present in tissue, blood and bone marrow (BM), where plasmacytoid DC (pDC) are a normal finding, representing 0.2 ± 0.1% of cell populations (Matarraz et al, 2010). DC neoplasms include solid tumors (such as DC sarcomas) and an entity classified by the World Health Organization (2008) as an acute myeloid leukemia (AML)-related precursor neoplasm: blastic pDC neoplasm/leukemia, an aggressive disease with poor prognosis, with no clinical trials to orient consensus regarding the most effective treatment; it is usually chemo-resistant, although some cases respond to AML-like regimens and allogeneic hematopoietic stem cell transplant. It is not clear if the presence of an increased DC population in non-DC AML confers pDC neoplasm-like biological characteristics to the former. Aims: This study aims to evaluate whether an increase in the size of DC populations in newly-diagnosed non-DC AML affects the latter's biological behavior, as represented by the overall survival (OS) of patients with the disease. Methods: We reviewed all AML diagnosed in our Hospital between January 1st 2008 and December 31st 2010, identifying 146 patients. We excluded 9 patients who had no flow cytometry immunophenotyping (IP) performed, and 7 whose first IP was performed after treatment was instituted. In that time frame, we also diagnosed 4 pDC neoplasms. Of the 130 patients included, 91 had their presenting IP performed on BM aspirate, while the remaining 39 were phenotyped on blood samples. The size of the DC populations and blastic DC maturation were determined on these samples. Patients were classified into 2 groups according to the size of the DC component; one (the Non-DC Group) had a DC component of up to 0.3% (in practice, the highest value in this group was 0.2%); the other (DC Group) had a percentage over this limit (the lowest value being 1.0%). OS data was determined for both groups; special consideration was given to age strata, separating patients under 65 years of age (Under-65) from those 65 or older (Over-65) and etiology (distinguishing de novo AML from AML secondary to therapy, myelodysplasia or myeloproliferative diseases). The percentage of DC identified by IP did not influence nor alter the type of treatment instituted. Results: We found that the presence of a DC component above the normal BM interval (as determined by Matarraz et al) was associated with a significantly decreased OS, with patients with DC components over 0.3% presenting with a median OS of 2.4 months (mean: 6.4 ± 1.6) and those with a component under 0.3% with a median OS of 8.6 months (mean: 17.0 ± 1.9) (p = 0.033). In our series, patients Over-65 had a median OS of 2.9 months (mean = 6.9 ± 1.0) and those Under-65 a median of 21.3 months (mean = 22.5 ± 2.5), p 〈 0.001. The differences in OS according to DC component were attenuated in patients Over-65 (median = 1.8 vs. 3.9 months, p = NS), whereas in patients Under-65 the median survival was 2.7 months (mean: 8.7 ± 2.9) for the DC Group and 24.4 months (mean: 24.3 ± 2.7) for the non-DC Group (p = 0.035). The differences in OS were also significant for de novo AML (median = 2.4 vs. 16.0 months, mean = 4.7 ± 1.9 vs. 20.5 ± 2.6, p = 0.017), but not statistically relevant for secondary AML (median = 4.4 vs. 5.5 months, mean = 8.4 vs. 10.8, p = NS). Discussion: In this study, we found that an increase in the size of the DC component as determined by IP at diagnosis on newly-diagnosed AML had a negative impact on prognosis, with a significant decrease in median and mean OS in patients with a percentage of DC over the upper limit of the normal interval. We also determined that the decreased survival was primarily attributed to the better-prognosis groups (patients under 65 and with de novo AML), whereas the effect of the worsened prognosis was attenuated in those patients with a bad prognosis at the outset (patients over 65 and with secondary AML). If data from DC neoplasms could be extrapolated, we could suggest that AML with increased DC components are less chemo-sensitive, which would explain the OS differences found in the Under-65 group, as well as the no-difference found in the Over-65 Group, which is frequently undertreated due to comorbidities. Conclusion: Our study suggests that the size of the DC component at diagnosis as determined by IP is a new prognostic marker predictive of decreased survival. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4620 Myelodysplastic syndrome (MDS) is a clonal hematopoietic stem cell disorder characterized by dyshematopoiesis and high susceptibility to acute myeloid leukemia. Deregulated epigenetic mechanisms are likely involved in the pathogenesis of MDS. Gene silencing through aberrant CpG island methylation is the most extensively analyzed epigenetic event in human tumorigenesis and has huge diagnostic and prognostic potential. Aberrant methylation of gene promoter region is responsible for inappropriate gene silencing, and it has been associated to initiation and progression of cancer. However, in the MDS disease process, more and more gene dysfunction has been related with the pathogenesis. FLT3 and c-KIT are important members of the receptor tyrosine kinase family that are overexpress or dysexpress in many malignant hematologic diseases. However, little is known about the distribution and the role of these proteins in MDS. The study is to investigate the role of receptor tyrosine kinase FLT3 and c-KIT expression in patients with myelodysplastic syndromes (MDS) and their clinical implication. We have at moment examined c-kit protein (CD117) expression by flow cytometry, in CD34 bone marrow cells collected at diagnosis of 12 patients with de novo MDS and 5 non-neoplastic patients (controls). FLT3 mutations, in particular Internal Tandem Duplications (ITD) and the D835 mutation were analysed by PCR-RFLP. The median age was 72 years (22–89), gender M/F=5/7, WHO subtypes: RCMD (n=6), RA (n=3), RARS (n=1), AREB-2 (n=1), CMML (n=1) and IPSS: low (n=6), intermediate-1 (n=5) and intermediate-2 (n=1). None of the patients evolved to acute leukemia, with a median follow up of 24 months (7–74). Our preliminary results show an increase in c-KIT expression in CD34 positive cells in MDS patients as compared with controls. However, the percentage of c-KIT protein expressing cells was also higher than in the controls in particular in CD34 negative cells. There was a correlation of the c-kIT protein expression with the CD34 antigen of the cells. Expression is correlated with the WHO MDS classification and with IPSS, being highest in RAEB-2 and INT2 MDS prognostic group. These results suggest that the elevated c-KIT expression could maintain the affected clone in MDS. Besides that we didn't find any FLT3 mutations in our population However further data and refinement of data analysis are needed to confirm our results and to predict clinical outcomes. The preliminary results suggest that c-KIT expression could be helpful to the pathogenesis and prognosis prediction of MDS patients. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 265 Background: Treatment goals in patients with relapsed FL are to prolong PFS and improve overall survival (OS). To optimize treatment for individual patients, identification of subgroups most likely to benefit from a specific therapy is important. The international, randomized, phase 3 LYM3001 study in patients with relapsed or refractory FL demonstrated improved PFS with bortezomib-rituximab vs rituximab alone (median 12.8 vs 11.0 months, HR 0.822, p=0.039), plus increased overall response rate (ORR; 63% vs 49%, p=0.0004), complete response rate (CR/CRu; 25% vs 18%, p=0.035), and durable (≥6 months) response rate (50% vs 38%, p=0.002) in an unselected patient population. Here we present exploratory biomarker analyses aimed at identifying patient subgroups deriving a longer PFS benefit with bortezomib-rituximab and showing a trend for better OS. Methods: Patients received five 5-week cycles of bortezomib-rituximab (N=336) or rituximab (N=340). Response was assessed using modified International Working Group response criteria. Archived tumor tissue was collected at baseline from 502 (74%) patients; whole blood samples for germ-line DNA were collected on day 1 of cycle 1 from 619 (92%) patients. Protocol-specified candidate biomarkers were based on associations with bortezomib (NF-κB p65, PSMA5, p27, PSMB1/5/8/9) or rituximab (CD68, FCGR2A/3A) activity. Immunohistochemistry assays were used for protein analysis. Taqman SNP assays and PCR/LDR were used for genotyping. Statistical analyses included single-marker analyses, pair-wise combination analyses (n=1140 comparisons), and multiple comparison analyses of all evaluable patients in LYM3001. Clinical covariates included in the analysis were baseline FLIPI score, prior rituximab, time since last anti-lymphoma therapy, region, age, gender, race, Ann Arbor stage, high tumor burden, and number of prior lines of therapy. Results: Single markers and biomarker pairs (n=102) highlighted patient subsets that had significantly improved outcomes with bortezomib-rituximab vs rituximab. For 14 of the pairs, the PFS benefit was ≥6 months. Using false discovery rate (FDR) to control for multiple comparison corrections, one biomarker pair was significant. This pair (presence of the PSMB1 P11A C/G heterozygote, and low CD68 expression [0–50 CD68-positive macrophages in the follicular space]) was associated with significantly improved PFS in patients receiving bortezomib-rituximab vs rituximab (median 16.6 vs 9.1 months, HR 0.407, p2 prior lines of therapy). There was also a trend towards an OS benefit (medians not reached, HR 0.426, p=0.0550), as well as a significantly higher ORR (73.7% vs 47.5%, p=0.0077), a higher CR rate (33.3% vs 23%, p=0.3044), and a significantly longer time to next therapy (median 33.1 vs 14.8 months, p=0.0013). In patients lacking this biomarker pair (N=238) no significant efficacy differences were seen. No other similar studies were available to confirm the reproducibility of these analyses. Therefore, we split the LYM3001 dataset into discovery and confirmation cohorts (7:3 ratio of biomarker-evaluable patients) to enable evaluation and confirmation in independent cohorts of patients The significant biomarker pair of PSMB1 P11A C/G heterozygote and low CD68 was identified in the discovery cohort (N=198) with a PFS advantage with bortezomib-rituximab vs rituximab of 5.7 months (median 14.2 vs 8.4 months, p=0.0003) and an indication of longer OS (HR 0.47, p=0.1291). This biomarker pair also showed a clear PFS advantage in the confirmation cohort (N=108, 8.7-month PFS benefit; median 18.2 vs 9.5 months, HR 0.44, p=0.0817). Other significant biomarker combinations, including combinations of molecular and clinical variables (e.g. high tumor burden) were identified and will be presented. Conclusions: Analyses of the phase 3 LYM3001 trial identified biomarker combinations present in a third of patients offering a significant PFS benefit with bortezomib-rituximab vs rituximab. Use of such biomarker assays in patients with relapsed or refractory FL may aid identification of subgroups deriving maximal benefit from the addition of bortezomib to rituximab therapy. Disclosures: Coiffier: Janssen-Cilag: Consultancy; Roche: Consultancy; Amgen: Consultancy; Sanofi: Consultancy; Pfizer: Consultancy; Millennium Pharmaceuticals, Inc.: Consultancy; Celgene: Consultancy; Pharmacyclics: Consultancy; MedImmune: Consultancy; CTI: Consultancy. Off Label Use: Bortezomib used in combination with rituximab in patients with relapsed/refractory follicular lymphoma. Li:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Henitz:Janssen Research & Development: Employment. Karkera:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Favis:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Gaffney:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Shapiro:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Theocharous:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Elsayed:Janssen Research & Development: Employment; Johson & Johnson: Equity Ownership. de Velde:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Rule:Johnson & Johnson: Advisory Board, Institutional grant, meeting attendance expenses, Honoraria. Walewski:Janssen-Cilag: Institutional/personal grants, advisory board; Hoffman La Roche: Honoraria, Institutional/personal grants, travel/accommodation expenses; Mundipharma: Honoraria; Celgene: Honoraria. de Vos:Millennium Pharmaceuticals, Inc: Consultancy. Crump:Janssen/Ortho-Biotech: Consultancy. Shpilberg:Janssen-Cilag: Consultancy, Honoraria. Cakana:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership. Esseltine:Millennium Pharmaceuticals, Inc: Employment; Johnson & Johnson: Equity Ownership. Mulligan:Millennium Pharmaceuticals, Inc.: Employment. Ricci:Janssen Research & Development: Employment; Johnson & Johnson: Equity Ownership.