Abstracts from the XVI International Workshop on

Leukemia & Lymphoma

ISSN: 1042-8194 (Print) 1029-2403 (Online) Journal homepage: http://www.tandfonline.com/loi/ilal20

Abstracts from the XVI International Workshop on Chronic Lymphocytic Leukemia 2015 To cite this article: (2015): Abstracts from the XVI International Workshop on Chronic Lymphocytic Leukemia 2015, Leukemia & Lymphoma To link to this article: http://dx.doi.org/10.3109/10428194.2015.1080893

Accepted online: 02 Sep 2015.

Submit your article to this journal

Article views: 605

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ilal20 Download by: [190.64.49.90]

Date: 16 October 2015, At: 11:50

Just Accepted by Leukemia & Lymphoma

LEUKEMIA & LYMPHOMA at the XVI International Workshop on Chronic Lymphocytic Leukemia

Downloaded by [190.64.49.90] at 11:50 16 October 2015

- September 2015, Sydney, Australia

Abstracts from the XVI International Workshop on Chronic Lymphocytic Leukemia 2015 6–9 September 2015 Sydney, Australia Doi:10.3109/10428194.2015.1080893

Access the abstracts from the meeting today at: www.bit.ly/IWCLL15

© 2015 Taylor & Francis. This provisional PDF corresponds to the article as it appeared upon acceptance. Fully formatted PDF and full text (HTML) versions will be made available soon. DISCLAIMER: The ideas and opinions expressed in the journal’s Just Accepted articles do not necessarily reflect those of Taylor & Francis (the Publisher), the Editors or the journal. The Publisher does not assume any responsibility for any injury and/or damage to persons or property arising from or related to any use of the material contained in these articles. The reader is advised to check the appropriate medical literature and the product information currently provided by the manufacturer of each drug to be administered to verify the dosages, the method and duration of administration, and contraindications. It is the responsibility of the treating physician or other health care professional, relying on his or her independent experience and knowledge of the patient, to determine drug dosages and the best treatment for the patient. Just Accepted articles have undergone full scientific review but none of the additional editorial preparation, such as copyediting, typesetting, and proofreading, as have articles published in the traditional manner. There may, therefore, be errors in Just Accepted articles that will be corrected in the final print and final online version of the article. Any use of the Just Accepted articles is subject to the express understanding that the papers have not yet gone through the full quality control process prior to publication.

Abstracts from the


XVI International Workshop on Chronic Lymphocytic Leukemia 2015 6 – 9 September 2015 Sydney, Australia

• • • • • • • • • • •

Cellular Biology Clinical Trials Conventional Therapy Diagnosis Epidemiology Immune disturbances Molecular Biology Novel Therapy Other Prognostication Treatment

Keywords: chronic lymphocytic leukemia, International Workshop on Chronic Lymphocytic Leukaemia (iwCLL)

Cellular Biology 1 Cellular Biology Targeting macrophages sensitizes chronic lymphocytic leukemia to apoptosis and inhibits disease progression Maria Teresa Sabrina Bertilaccio1 Giovanni Galletti1, Cristina Scielzo1, Federica Barbaglio1, Tania Veliz Rodriguez1, Michela Riba2, Dejan Lazarevic2, Davide Cittaro2, Giorgia Simonetti3, Pamela Ranghetti1, Lydia Scarfò4, Maurilio Ponzoni5, Angelo Corti6, Achille Anselmo7, Nico van Rooijen8, Christian Klein9, Carola Ries10, Paolo Ghia4, Michele De Palma11, Federico Caligaris-Cappio1


1

Unit of Lymphoid Malignancies, Division of Experimental Oncology, IRCCS San Raffaele Scientific Institute, Milan, Italy 2 Center for Translational Genomics and Bioinformatics, IRCCS San Raffaele Scientific Institute, Milan, Italy 3 Department of Experimental, Diagnostic and Specialty Medicine, Institute of Hematology "L. e A. Seràgnoli", Università di Bologna, Italy 4 Unit of B Cell Neoplasia, Division of Experimental Oncology, IRCCS San Raffaele Scientific Institute, Milan, Italy 5 Pathology Unit, IRCCS San Raffaele Scientific Institute, Milan, Italy 6 Tumor Biology and Vascular Targeting Unit, Division of Experimental Oncology, IRCCS San Raffaele Scientific Institute, Milan, Italy 7 Humanitas Clinical and Research Center, Rozzano, Milan, Italy 8 Department of Molecular Cell Biology, Vrije University Medical Center, Amsterdam, The Netherlands 9 Roche Pharma Research and Early Development, Oncology Discovery, Roche Innovation Center, Zurich, 8952 Zurich, Switzerland 10 Roche Pharmaceutical Research and Early Development, Roche Innovation Center Penzberg, Oncology Discovery 82377 Penzberg, Germany 11 The Swiss Institute for Experimental Cancer Research (ISREC), School of Life Sciences, École Polytechnique Fédérale de Lausanne, Switzerland All relevant events in CLL occur in permissive microenvironments where the activity of different cell types influences the disease natural history, accounts for its clinical heterogeneity and provides the basis for identifying novel treatment targets. A number of evidences highlight the role of the monocyte/macrophage lineage in the pathogenesis of CLL, but still therapeutic approaches aimed at targeting these specific interactions are unexplored. We thoroughly characterized the molecular repertoire of monocytes/macrophages exposed in vivo to leukemic cells and we investigated their role in the growth of CLL. We also explored whether interfering with leukemic cell-monocyte/macrophage interaction could represent a potential novel therapeutic strategy in CLL. To this end we exploited a killing approach based on the macrophage internalization of clodronate liposomes (clodrolip) and we utilized an anti-CSF-1R moAb to inhibit macrophage differentiation from monocyte precursors. We first analyzed the transcriptome (by using the Illumina whole-genome gene expression directhybridization assay, harboring probes for 45,000 transcripts) of murine TAMs and human leukemic cells isolated from the BM of Rag2-/-c-/- mice xeno-transplanted with MEC1 CLL cells at early stage of leukemia. We have found on monocytes/macrophages exposed in vivo to leukemic cells an enrichment of genes involved in inflammation and intracellular/extracellular function (e.g. FcR1, ICAM1). As regards leukemic cells,


particularly relevant was the differential expression of genes supporting the existence of monocyte/macrophage-leukemic cell cross talk, including IL10 and RNASET2, besides genes involved in CLL progression, like NOTCH1, BIRC3, and PTEN. To investigate whether CLL engraftment and progression in vivo might be affected by monocyte/macrophage populations, we transplanted Rag2-/-c-/- mice with MEC1 cells and depleted macrophages by clodrolip delivery. Macrophage targeting resulted in a drastically reduced accumulation of CLL cells in all tissues. We extended these studies to the E-TCL1 tg immunocompetent mouse model confirming a reduction of the leukemic clone in the PB and all the lymphoid tissues. To demonstrate that macrophages may represent a valuable therapeutic target in CLL, we investigated at pre-clinical level the anti-leukemic effect of the anti-mouse CSF1R moAb. In Rag2-/-c-/- mice transplanted with MEC1 cells, macrophage depletion by CSF-1R inhibition significantly reduced the number of leukemic cells in the BM and stabilized the disease severity in the spleen by inducing increasing necrosis of leukemic cells over time. Of note, anti-mouse CSF1R moAb reprogrammed the tumor microenvironment toward an antitumoral phenotype and significantly impacted on mice survival. We investigated at RNA level the cell death pathways induced in leukemic cells by macrophage targeting. We definitely demonstrated in vivo the involvement of TNF in the mechanism of leukemia cell death induced by macrophage targeting by treating xenotransplanted Rag2-/-c-/- mice with clodrolip or CSF1R moAb and etanercept, a TNF receptor fusion protein able to bind soluble TNF. Finally, we applied and confirmed the molecular and functional information gathered in mouse models to human CLL samples, by taking advantage of an anti-human CSF1R moAb. These findings let us conclude that TAMs critically support the survival and proliferation of CLL cells in vivo, therefore unveiling therapeutic strategies based upon manipulating TAM/CLL-cell interactions.

2 Cellular Biology Combining CD23 chimeric antigen receptor immunotherapy and lenalidomide as a novel therapeutic strategy for chronic lymphocytic leukemia Sarah Tettamanti1 2 1 1 Giovanni Galletti , Greta Maria Paola Giordano Attianese , Silvia Arcangeli , Tania Veliz 2 1 2 3 Rodriguez , Chiara Francesca Magnani , Federica Barbaglio , Lydia Scarfò , Maurilio 4 1 2 1* 3 Ponzoni , Andrea Biondi , Federico Caligaris-Cappio , Ettore Biagi , Paolo Ghia , Maria 2 Teresa Sabrina Bertilaccio


1

Centro Ricerca Tettamanti, Clinica Pediatrica, Università Milano Bicocca, Osp. San Gerardo/Fondazione MBBM, Monza, Italy 2 Unit of Lymphoid Malignancies, Division of Experimental Oncology, IRCCS San Raffaele Scientific Institute, Milan, Italy 3 Unit of B Cell Neoplasia, Division of Experimental Oncology, IRCCS San Raffaele Scientific Institute, Milan, Italy 4 Pathology Unit, IRCCS San Raffaele Scientific Institute, Milan, Italy Maria Teresa Sabrina Bertilaccio and Paolo Ghia equally contributed to the work * Corresponding author Chronic Lymphocytic Leukemia (CLL) is a chronic lymphoid malignancy characterized by immune dysfunction that particularly involves the T-cell compartment. Given the well known role of lenalidomide in enhancing T cell function [1] we exploited a novel therapeutic strategy by combining adoptive CD23.CAR-based immunotherapy with lenalidomide. First we used the CLL-xenograft model based on the injection of MEC1 CLL cell line into Rag2-/-/c mice [2] to verify the in vivo antitumor activity of lenalidomide on human CLL cells. We then generated CD23.CAR+ T lymphocytes from 3 different peripheral blood samples collected from CD23+ CLL patients, as already described for healthy donors [3]. We demonstrated that in pre-clinical settings, lenalidomide can be efficiently associated to CAR-/based immunotherapy. In MEC-1 xeno-transplanted Rag2-/- c mice, by combining CAR.CD23 T cells from CLL patients with low dose lenalidomide, ineffective in monotherapy, we induced a decrease of the percentage of CD19+ leukemic cells in all lymphoid and non-lymphoid tissues and an improved survival. The combination with low dose lenalidomide was more effective also when compared to human recombinant IL-2 utilized in traditional immunotherapeutic settings. In accordance with the in vivo efficacy, CAR T cells were observed in all leukemic sites suggesting an ability to migrate and home in vivo and, when purified from the bone marrow, CD23.CAR+ T cells were still able to mount a tumor-specific cytotoxic response in vitro. Surprisingly, CAR T cells exposed in vivo to daily lenalidomide were for the majority effector memory cells and maintained the expression of CD23.CAR on their surface. These results conceivably support the use in the CLL therapeutical setting of low doses lenalidomide to improve CAR cytotoxic response and to avoid the potential impairment of an effective immune response. References: [1] Ramsay AG JCI 2008 [2] Bertilaccio et al, Blood 2010 [3] (Giordano-Attianese Blood 2010).

3 Cellular Biology Effect of the Btk inhibitors, ibruitinb and AVL-292, and the Syk inhibitor, GS9973, on macrophage-mediated anti-tumor activity of rituximab Mercedes Borge1 1 1 1 1 María Belén Almejún , Enrique Podaza , Ana Colado , Denise Risnik , Horacio Fernandez 2 2 3 Grecco , María Cabrejo , Fernando Raimundo Bezares , Mirta Giordano1, Romina 1 Gamberale 1

IMEX-CONICET Hospital Méndez 3 Hospital Alvarez


2

In the past few years several orally administrated inhibitors targeting kinases involved in BCR signaling have been developed and entered the clinical stage showing promising efficacy and excellent tolerability. The Btk inhibitor, ibrutinib, has been approved in USA last year for CLL treatment and the combination of ibrutinib and other kinases inhibitors with targeted therapies such as the anti-CD20 antibody rituximab, is now being explored in clinical trials. Anti-CD20 antibodies are thought to act through different mechanisms but there is strong evidence suggesting that macrophages, acting through Fcγ-receptors, are the key immune effector cell for anti-CD20 therapeutic effect in vivo [1,2]. Given that kinase inhibitors might affect Fcγ-receptor signaling, the aims of our study were: i) to evaluate the effect of ibrutinib on macrophage-mediated phagocytosis of rituximab-coated leukemic cells, ii) to assess the effect AVL-292 and GS9973 on macrophage phagocytosis, and iii) to study if AVL-292 and GS9973 modulate CD20 expression on leukemic cells. Human monocytesderived macrophages were treated with or without ibrutinib for 30 minutes and then cultured for 1 hour with CFSE-labeled CLL cells coated or not with rituximab. Then, the proportion of + macrophages that have taken up CFSE-labeled CLL cells (CFSE macrophages) were scored by flow cytometry and confocal microscopy. As shown in Figure 1A and 1B we observed that ibrutinib inhibited macrophage-mediated phagocytosis of rituximab-coated CLL cells at 0.5 and 5M (n=17, p0.01). We confirmed that ibrutinib impairs the phagocytosis but not the binding of rituximab-coated CLL cells to macrophages by performing the same experiment at o 4 C (Figure 1C). Interestingly we found that the inhibition of phagocytosis was no longer present when ibrutinib was washed out and macrophages were cultured in absence of the drug for 7 or 24 hours (Figure 1D). Given that ibrutinib binds to Btk in an irreversible manner this observation suggests the existence of a reversible-target of ibrutinib in macrophages other than Btk [3]. Our next aim was to evaluate if AVL-292 or GS9973, two kinase inhibitors currently being tested in clinical trials for CLL, could also affect macrophage phagocytosis. We found that GS9973, a Syk inhibitor, impaired phagocytosis at 0.5M while the Btk inhibitor, AVL-292, only impaired the phagocytosis at 5M (Figure 1E). Finally, we found that the modulation of CD20 expression on CLL cells by ibrutinib, AVL-292 and GS9973 was heterogeneous among our cohort of patients, and that while CD20 expression was downregulated by ibrutinib in accordance with a previous report [4], it was not significantly modify by AVL-292 and GS9973 (Figure 1F). Our results show that kinases inhibitors interfere with the macrophage-mediated anti-tumour activity of rituximab suggesting that the sequential administration of kinases inhibitors and rituximab, and not the concurrent treatment with these agents, might enhance their anti-tumor activity in CLL patients. Interestingly we found that the Btk inhibitor AVL-292 at clinical relevant doses (0.5 M) did not significantly impair macrophage-phagocytosis suggesting that this inhibitor might have less interference with the anti-tumor effect of rituximab in vivo.


Figure 1. Effect of Btk and Syk inhibitors on macrophage phagocytosis of rituximab-coated CLL cells and on CD20 expression. (A) Phagocytosis assay was performed as described in the text. Shown are the percentage of macrophages that have taken up CLL cells (CFSE* macrophages) scored by flow cytometry (p=0.0001, Kruskal-Wallis test followed by Dunn’s Multiple Comparison Test *p=0.01 to 0.05 and +++p,0.001). (B) Macrophages were stained with anti-CD14PE mAb after the phagocytosis assay. Immunofluorescence images were acquired with a FluoView FV1000 confocal microscope (Olympus, Tokyo, Japan) using a Plapon 60X 1.42. Images from a representative experiment are shown. Macrophages that have phagocyte CFSE*CLL cells are indicated with white arrows. (C) Binding assay was performed as described in the text. The graph shows the percentage of macrophages that bind CFSE* cells scored by flow cytometry. The bars represent the mean ± SEM of 7 samples evaluated (NS: not statistically significant Kruskal-Wallis test). (D) Phagocytosis was performed in the presence of ibrutinib (without wash out) or after 7 or 24 hours after drug was removed (T7hs wash out, and T24hs washout, see explanation in the text). Shown are the mean ± SEM of 12 samples evaluated. *p10% were neutropenia (51%), thrombocytopenia (16%), and anemia (14%). One treatment-emergent AE (TLS) led to death; no other fatal TLS events occurred after protocol modification. Two deaths occurred after PD. Conclusions:


Venetoclax plus rituximab has a tolerable safety profile and induces deep and durable responses, with 41% achieving CR/CRi and 49% achieving MRD negativity in BM. A phase 3 trial comparing venetoclax and rituximab versus bendamustine and rituximab in patients with previously treated CLL is underway.

30 Clinical Trials Health related quality of life and patient reported outcomes of ofatumumab plus chlorambucil versus chlorambucil monotherapy in the COMPLEMENT1 trial


Stephanie Manson Astrid McKeown, Chai-Ni Chang, Ira Gupta, Peter Hillmen Affiliations Introduction: The COMPLEMENT1 study has demonstrated a statistically significant improvement in PFS in patients who are treated with ofatumumab plus chlorambucil in the first line setting for CLL over chlorambucil monotherapy and the combination therapy was well tolerated. Given that ofatumumab was added to chlorambucil, it is important to consider the impact of ofatumumab on health related quality of life (HRQoL). Methods: During the COMPLEMENT1 trial, the QLQ-C30 and the QLQ-CLL16 patient questionnaires were administered in patients during treatment, during the follow up stage and at the time that progression was identified. The primary specified patient reported outcomes were health related quality of life and fatigue from the QLQ-C30 questionnaire. Results: In HRQoL, there was no significant difference (p=0.67) or clinically relevant difference between the arms at any time point measured during treatment. In the fatigue scale, there was similarly no significant differences noted between the arms (p=0.17) during treatment. This trend of no clinically significant changes in HRQoL and fatigue continued during follow-up (p=0.49 and p=0.10 respectively). A summary of HRQoL scores measured on a scale of 0 to 100 with 100 being the maximum HRQoL can be seen in Figure 1. Discussion: These results demonstrate that adding ofatumumab to frontline chlorambucil therapy do not result in any perceptible changes in HRQoL or fatigue, with any adverse events having minimum impact. This is consistent with the adverse event profile of ofatumumab in the COMPLEMENT1 study, which demonstrated high tolerability irrespective of age and fitness. Figure 1. HRQoL scores over time in the ofatumumab+chlorambucil and chlorambucil monotherapy arms.

31 Clinical Trials Health related quality of life and patient reported outcomes in patients receiving ofatumumab maintenance versus observation in the PROLONG trial 1 2 3 4 5 Marinus van Oers , Kazimierz Kuliczkowski , Lukas Smolej , Mario Petrini , Fritz Offner , 6 7 8 8 9 Sebastian Grosicki , Mark-David Levin , Ira Gupta , Jennifer Phillips , Vanessa Williams , Stephanie Manson10, Steen Lisby11, and Christian Geisler12


1

Academisch Medisch Centrum and HOVON , Amsterdam,Netherlands 2SPSK 1Wroclaw,Poland 3 University Hospital and Faculty of Medicine -Hradec Kralove,Czech Republic 4Azienda Ospedaliero Universitaria Pisana-Pisa,Italy 5Universitair Ziekenhuis GentGent,Belgium 6 Department of Cancer Prevention, Faculty of Public Health, Silesian Medical University, Katowice, Poland 7Albert Schweitzer Ziekenhuis Dordrecht and HOVON,Netherlands 8GlaxoSmithKline, Collegeville, PA, USA 9GlaxoSmithKline, Research Triangle Park, NC, USA 10GlaxoSmithKline, Uxbridge, UK 11Genmab, Copenhagen, Denmark 12Rigshospitalet-Koebenhavn, Denmark

Introduction: The randomized phase III PROLONG study in 474 relapsed CLL patients in nd rd remission after 2 or 3 line induction treatment has demonstrated a statistically significant improvement in PFS by ofatumumab maintenance treatment as compared to observation: 29.4 months versus 15.2 months respectively (HR=0.50, p2 weeks. QoL assessments using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) were measured at 8 time points: before commencement of therapy, at the end of the 3rd and 6th cycles of treatment, at the final staging and thereafter every 3 months for 12 months. Results: Of the 116 evaluable in the study, 2 patients had no QoL data, 6 patients had only baseline QoL data, and 4 patients had no baseline QoL data, therefore the number of patients included in the QoL analysis was 104. There appeared to be a trend towards improved global health [Figure 1], physical functioning [Figure 2] and role functioning [Figure 3] with full dose (FCR5) towards the end of the followup period compared to commencement of therapy. Comparison of cognitive functioning [Figure 4] between the three arms showed a statistically significant improvement with FCR5. However, on multivariate analysis there was no statistical difference.


Figure 1. Mean change from baseline global health status scores with 95% confidence intervals.

Figure 2. Mean change from baseline physical functioning scores with 95% confidence intervals.


Figure 3. Mean change from baseline role functioning scores with 95% confidence intervals.

Figure 4. Mean change from baseline cognitive functioning scores with 95% confidence intervals.

Conclusions: Despite a trend favouring full-dose FCR, there were no definitively significant differences in the QoL domains following treatment with FCR-based immunochemotherapy between the three different groups. QoL continues to be a neglected issue, particularly for the typical elderly CLL patient. This study highlights the need for objective QoL assessment as part of all CLL trials, especially those targeted in the typical elderly patient. References [1] Mulligan, SP et al. Leuk Lymphoma 2014. Cladribine prolongs PFS and time to second treatment compared to fludarabine and high-dose chlorambucil in CLL.


[2] Mulligan SP et al. iwCLL Abstract, 2015. A randomised dose de-escalation study of FCR in the elderly.

36 Clinical Trials Detailed long-term follow-up of treatment-naïve chronic lymphocytic leukaemia (CLL) patients in the Australasian Leukaemia and Lymphoma Group (ALLG) CLL5 Trial; data on 17 (15% of total cohort) patients from a single-institution Suneet Sandhu1*, Naomi Mackinlay1, Luke Coyle1, Giles Best2, Stephen Mulligan1 1

Royal North Shore Hospital, St Leonards, Sydney Kolling Institute, University of Sydney, Sydney


2

Background: The ALLG CLL5 randomised dose de-escalation study examined the tolerability, safety and efficacy of oral FCR therapy as first-line treatment in fit elderly CLL patients [1]. The study incorporated an early stopping rule for prolonged grade 3/4 toxicity and documented the regimen to be safe, generally well tolerated, and highly effective. There is a relative paucity of published data on the long term outcomes of such patients, particularly in relation to the development of second malignancies as well as disease relapse rates, with its associated potential long-term complications. We aimed to review the long term outcomes in fit elderly patients that were recruited to ALLG CLL5 trial from Royal North Shore Hospital where 17 of the total cohort of 116 evaluable patients (15%) were recruited, treated and managed. Methods: The treatment schedule for the open-label, multi-centre, phase 2 ALLG CLL5 study has been described elsewhere, whereby fit (CIRS 6) elderly (65 years) old patients, with previously untreated CLL were randomly assigned to receive one of three different chemoimmunotherapies: FR5, FCR3, FCR5 [1]. Treatment was repeated every 28 days with a planned total of 6 courses. A total of 116 eligible patients were recruited during the period of November 2008 and July 2012. After the completion of treatment, patients were followed every 3 months for 15 months. Ongoing review was subsequently at the treating physician's discretion. We reviewed the clinical progress from the hospital medical records of the 17 patients that were recruited from Royal North Shore Hospital for this study. Results: Long term follow-up clinical data to cut-off date 3/7/2015 are described in the table below. Of this elderly CLL patient cohort of 17 patients, 5 have died (30%), the remainder (70%) are alive 4 to 7 years from therapy. About half (n=8) the cohort at this single hospital remain very well up to 7 years following FCR-based treatment, 2 with documented MRD-negativity, and 2 with a tiny small clone of >0.05x109/L.

Pt

Sex, Rx age, date at enrollmt

RNS1

F 82, 3/2010

RNS2

M 74, 2/2009

Final Staging

Long term follow-up

FCR3 CR, MRD positive

Very well in 2015

FCR3 CR, MRD positive

Relapsed CLL (10/2010): then given full dose FCR IV) achieved CR


E-antigen+ chronic Hepatitis B: Tenofovir long term Global cognitive decline suggestive of an Alzeihmer's type dementia associated with intractable dizziness and anti-neuronal antibody (1/2011): no improvement with plasma exchange or IVIG. 2015 MRD+ in PB for CLL (0.2x109/L), mild pancytopenia. RNS3

F 65, 7/2009

FCR3 SD post C3, Commenced on trial due to progressive severe withdrew after C3 pancytopenia and progressive CLL - CLL due to cytopenias eradicated and remained MRD-negative in PB and BM, but severe pancytopenia never recovered. Subsequently developed Nephrotic syndrome and then fatal Stage 4 NSCLC (10/2012): DIED 8/2013

RNS4

F 74,

FCR3 nPR

11/2008 RNS5

F 65,

Small BCC treated (2/2011) Progressive multifocal leukoencephalopathy, JC virus positive: DIED 8/2011

FR5

nPR

12/2009

Progressive CLL 6/2011: Given full dose FCR, IV) Recurrent severe AIHA: steroids, splenectomy (11/2011) Listeria meningitis (7/2012) Severe bilateral visual impairment, unclear diagnosis/aetiology (10/2012) Disseminated Aspergillosis and progressive CLL: DIED 12/2013

RNS6

F 68, 11/2008

RNS7

M 81, 11/2011

FCR5 CR, MRD negative, withdrew after C5D1 due to thrombocytopenia

Remains very well with mild persistent neutropenia and thrombocytopenia and remains MRD-negative in PB in 2015. BM 3/2013 potentially consistent with therapy-related MDS but no chromosomal abnormality, and MRDnegative for CLL. Cytopenia stable for 8 years.

FR5

Well

CR

RNS8

M 74,

FR5

3/2009

CR, MRD positive

Developed AIHA while CLL MRDnegative(2/2011): treated with steroids Several small BCCs treated. B12 deficiency. No further CLL treatment. 2015 CLL MRD-positive at 0.9x109/L. Remains well on low dose prednisone.

RNS9

F 69, 3/2009

FCR5 CR, MRD negative

Developed late-onset-neutropenia (LON) requiring G-CSF for 6 to 9 months. Bell's Palsy (6/2013): incomplete recovery. Few treated BCCs, and SCCs. MRD-negative since therapy, became MRD+ in PB at 0.02x109/L on 8.8.14 and now 0.07x109/L 24 June 2015. FBC, IgG normal. Remains very well.


RNS10

F 75, 3/2009

RNS11

M 67, 1/2010


CLL remained MRD-negative. Developed Philadelphia positive ALL (11/2011), with relapse in 4/2013 and in 2/2014 treated dasatinib: DIED 5/2014

FCR5 CR, MRD negative. Stopped after C5D2 due to thrombocytopenia

AIHA prior to CLL treatment. CR of both CLL and AIHA post-therapy. CRi (moderate thrombocytopenia) persists. Bone marrow in tremed MDS with t(4:12) (8/2012). Few treated BCCs, SCCs. Platelet count stable at ~80x109/L since cessation of therapy. CLL remains MRD-negative in PB at 8 April, 2105. Very well.

RNS12

F 74,

FCR3 nPR

Some memory problems (8/2013). Otherwise very 9 well. Remains MRD-positive at 0.56x10 /L in PB at 12 June 2105.

11/2011


Prostate cancer (12/2012). Several treated BCCs and SCCs. MRD-negative until 13.4.2013 at 30% and currently only the most frequently occurring point mutations in NOTCH1, SF3B1 and MYD88, routine MLPA analysis will facilitate early identification of clinically relevant mutations and potential genomic evolution during the course of the disease, which in turn will provide valuable information for felicitous patient management.

44 Diagnosis Title differential diagnosis of LEF1 expression in B-cell chronic lymphoproliferative disorders Lei Fan1 1 1 1 1 1 1 1 Tingmei Shen , Yi Miao , Rong Wang , Yu-Jie Wu , Hui Yang , Li Wang , Wei Xu , Jian1 Yong Li 1


Department of Hematology, the First Affiliated Hospital of Nanjing Medical University, Jiangsu Province Hospital

Background and Objective: LEF1, lymphoid enhancer factor 1, is a key transcription factor of Wnt signaling pathways. Previous studies have explored the relationship between LEF1 gene and tumorogenesis as well as tumor development, and there is little report on the diagnostic significance of LEF1. Recent studies have shown that the LEF1 protein expression in chronic lymphocytic leukemia (CLL) is significantly higher than other B-cell chronic lymphoproliferative disorders (B-CLPD) and normal controls. Our study aimed to compare the LEF1 gene mRNA expression and the LEF1 protein expression between different subtypes of patients with B-CLPD to establish the differential diagnostic value of LEF1 in BCLPD. Methods: We determined the expression level of LEF1 protein by immunohistochemistry in bone marrow samples from 143 patients with leukemic phase BCLPD including 78 cases of CLL, 17 cases of mantle cell lymphoma (MCL), 20 cases of Waldenström macroglobulinemia (WM), 20 cases of follicular lymphoma (FL), 5 cases of marginal zone lymphoma (MZL), 3 cases of B-cell chronic lymphoproliferative disordersunclassified (B-CLPD-U). We used real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) to compare the level of LEF1 mRNA expression in malignant cells from 54 B-CLPD patients, including 44 cases of CLL, 3 cases of MCL, 2 cases of MZL, 3 cases of FL, one case of hairy cell leukemia (HCL), one case of B-cell prolymphocytic leukemia (B-PLL) and 5 cases of healthy donors. All statistical analysis of the data were performed using SPASS version 20.0. Mann-Whitney U test was used to compare the difference between the cohorts. P10 x 109/L circulating CLL cells. Using detailed flow-cytometric monitoring, MBL cases with a CLL-cell count 1-5 x 109/L maintained a CLL-cell count below 10x109/L in 90/109 (83%) over 3-yr follow-up. In cases with a B-cell count below 1 x 109/L the CLL-cell count decreased in 7/25 (28%) and the remainder the median doubling time was 3.8 years. Only 1/25 cases had a doubling time 12 months and other features stable

47 Diagnosis Reproducible diagnosis of CLL by flow cytometry: an ERIC & ESCCA harmonisation project Andy Rawstron1 1 1 1 1 Karl-Anton Kreuzer , Asha Soosapilla , Martin Spacek , Peter Gambell , Neil McIver1 1 1 1 Brown , Katherina Psarra , Maria Arroz , Raffaella Milani , Javier de la Serna1, M. Teresa 1 1 1 1 1 Cedena , Ozren Jaksic , Josep Nomdedeu , Carol Moreno , Gian Matteo Rigolin , Antonio 1 1 1 1 Cuneo , Preben Johansen , Hans Erik Johnsen , Richard Rosenquist , Carsten Niemann1, David Westerman1, Marek Trneny1, Stephen Mulligan1, Peter Hillmen1, David Oscier1, Michael Hallek1, Paolo Ghia1, Emili Montserrat1 1


on behalf of ERIC and ESCCA

Background: The WHO and iwCLL diagnostic criteria for CLL rely on morphology and immunophenotype based on the co-expression of CD19/CD5/CD23 on B-cells with weak CD20 and monoclonal sIg expression. These diagnostic criteria are likely to persist in the near future because there is no specific pathognomic molecular abnormality for CLL. The current criteria have some limitations affecting reproducibility, particularly flexibility in marker expression with many centres using a scoring system that permits absence of CD5 or CD23. Potentially informative new markers have been identified but there is no consensus yet on which should be routinely assessed. Aim: To identify reproducible criteria and to achieve a consensus on markers recommended for the diagnosis of CLL. Methods: ERIC/ESCCA members were invited to classify 35 flow-cytometry markers as being required or recommended for the diagnosis of CLL. Consensus was considered to be achieved if >75% of participants agreed on the marker classification. A diagnostic panel was identified by the steering committee and characteristics of component markers that could be reproducibly validated within an individual laboratory were identified. The proposed panel was assessed in 13 different centres. Results: Responses were received from 154 members (100 laboratory staff, 14 clinicians and 36 from both laboratory and clinic) with a diagnostic workload >20 cases per week in 23/154 (15%), 5-20 in 82/154 (53%) and 95%)

CD20+

CD3+

B-cells

T-cells

CD5

Positive (>20%)

CD3+ T-cells

CD16/56+ NK-cells

CD23

Positive (>20%)

Nave

Memory

B-cells

B-cells

CD20

Weak

CD19+

CD3+

B-cells

T-cells

Ig

Weak & restricted

CD20+

CD3+

B-cells

T-cells

CD43

Positive (>20%)

CD3+

CD20+

T-cells

B-cells

CD79b

Weak

CD20+

CD3+

B-cells

T-cells

CD81

Weak

CD22

Weak

CD200

Positive (>20%)

CD10

Negative (20* >14 (>18) >5* >5 (>20) >10* >7 (>50) >11 (>30)

Granulocytes >5 (>8)

CD20+

CD3+

B-cells

T-cells

CD19+

CD3+

B-cells

T-cells

Granulocytes

Minimum Relative fluorescence intensity (preferred)

Memory B-cells

>10* >5* >10*

Table: required and recommended markers for use in the diagnosis of CLL with reagent specification based on expression patterns in normal peripheral blood. Weak expression = median fluorescence intensity at least 20% lower than median for normal peripheral blood B-cells, reference range determined within each laboratory, based on ICSH/ISLH/CLIA guidelines for reproducibility *consensus, not specifically validated 48 Diagnosis

An 18F-FDG-PET maximum standardized uptake value >10 represents a novel valid marker for discerning Richter’s Syndrome Pierre Sesques Anne Sophie Michallet


AFFILIATION Introduction: Richter's syndrome (RS) constitutes a primary complication in chronic lymphocytic leukemia (CLL). This study sought to validate recent findings demonstrating a correlation between positron emission tomography (PET) using fludeoxyglucose (FDG/PET) and CLL histological features. We aimed to demonstrate that a maximum standardized tumor uptake >10 on 18F-FDG-PET constitutes a valid marker for discriminating RS. Patients and Methods: From June 2006 to December 2012, 240 CLL patients were analyzed. The diagnoses were classified into three groups: stable CLL disease; rapid progression and RS. SUVmax, clinical features, laboratory findings, and gold standard histological findings were also evaluated. Results: This multicenter study evaluated 240 patients aged 62 (21-91) years on average. A statistically significant difference was observed between maximum SUV (SUVmax) of CLL patients with stable disease and that of RS patients (2.2 vs. 12.9; p 10 years from the time of initial presentation. We identified 797 patients seen at our institution between 1957 and 2003, and defined them as CLL long-term survivors (LTS). Among these LTS, the cumulative frequency of OC was 36% and it was similar between the 570 patients (72%) who required treatment (TR) for CLL, and the 227 (28%) who remained untreated (UT). Overall, the most common OC in both groups were non-melanoma skin cancers (NMSC, 86 patients, 10.8%). Aside from NMSC, prostate cancer, breast cancer, melanoma, lung cancer, and leukemia were the most frequent OC in TR patients, whereas prostate cancer, breast cancer, melanoma, lung cancer and gastrointestinal tumors were the most common OC in UT patients. To compare the incidence of OC in our group of LTS to the general population, we calculated standardized incidence ratios (SIR). This is the number of patients who developed invasive cancers in our patient cohort compared with the number of cases expected to occur if the US population rates were applied to the same cohort (ie, observed-to-expected ratio). We found that the SIR for all OC was increased to 1.2 (p=0.034). It was higher in males (SIR 1.31; p=0.013) and in patients 25 million commercially insured lives annually; 2000-2013) based on diagnoses and treatment patterns. Pts were required to be 18 years in age, continuously enrolled in their health plan for 6 months prior to and 1 month after the first indicator of R/R CLL. The index date was defined as the date occurring 30 days prior to the first indicator of R/R CLL. The incremental economic burden of R/R CLL was estimated by comparing the healthcare resource utilization (HRU) and costs of pts between the pre- R/R period (up to 12 months before the index date) and post-R/R period (up to 12 months after the index date). HRU and costs were estimated per-pt-per-month (PPPM) and compared between the pre- and post-R/R periods using generalized linear models and Wilcoxon signed-rank tests, respectively. Results: 5,066 R/R CLL pts were identified; median age of 68 years, 60.3% male, 58.3% with Medicare Supplement Plan coverage. Median time from first observed CLL diagnosis to R/R CLL was 18 months. During the post R/R period, pts had significantly higher HRU: 58% more inpatient (IP) admissions, 77% more IP days, 39% more emergency room visits, and 42% more outpatient (OP) visits compared to the pre-R/R period (all p0.58 and P-value 0.05 M Binimetinib was significantly (P < 0.05) more cytotoxic against CLL cells stimulated with either anti-IgM or CpG/IL2 than against cells cultured in media alone. Binimetinib also had a significant cytostatic effect on primary CLL cells stimulated with CpG/IL2 and against the OSU-CLL cell line. However, Binimetinib had no cytotoxic or cytostatic effect against primary CLL cells co-cultured with a CD40Lexpressing fibroblast cell line. Since CD40L co-culture also decreases sensitivity towards ABT737 we hypothesised that Binimetinib may induce down-regulation of Bim and Mcl-1, two known targets of ERK1/2-MAPK, may increase CLL-cell sensitivity towards the BH3mimetic. Consistent with this hypothesis, Binimetinib significantly reduced the IC50 of ABT737 against CLL cells co-cultured with CD40L fibroblasts. When cells were treated in combination with Binimetinib the sensitivity to ABT737 was restored to levels observed against CLL cells cultured in media alone, and that this was concomitant with a downregulation of the activities of Bim and Mcl-1. Conclusions: Our data illustrate the important role of MAPK-ERK1/2 activity in BCR-mediated CLL cell survival and on CLL-cell cycle progression. However, the cytotoxic and cytostatic effects of Binimetinib as a single agent may be abrogated by the interaction between CLL cells and the stroma within the tumour microenvironment. Combinations of Binimetinib and ABT737 may represent a novel treatment strategy for CLL, increasing the sensitivity of CLL cells in the tumour microenvironment to the BH3-mimetic.


Figure 1. Stimulation with anti-IgM or CpG/IL2 sensitises B-CLL cells to the MEK1/2 inhibitor Binimetinib. Figure 2. Binimetinib increases sensitivity of CLL cells in CD40L fibroblast co-culture to ABT737.

115 Novel Therapy Targeting NEDD8-activating enzyme abrogates NF- appaB and induces DNA damage and checkpoint activation in chronic lymphocytic leukemia B-cells Cody Paiva Claire Godbersen, Allison Berger, Jennifer Brown, Alexey Danilov1 1


Oregon Health and Science University

Microenvironment-mediated upregulation of the B-cell receptor and NFB signaling protect CLL cells from apoptosis. We recently reported that stromal-mediated NFB activation and CLL cell survival may be abrogated in vitro using pevonedistat (MLN4924; Millennium Pharmaceuticals Inc., a subsidiary of Takeda Pharmaceutical Company Limited). Pevonedistat is an investigational agent that inhibits the NEDD8-activating enzyme (NAE) and thereby prevents neddylation of Cullin-RING ubiquitin ligases, resulting in stabilization of their protein substrates, including inhibitor of NFB (IB). However, in adherent solid tumor cell lines, targeting NAE induced Cdt1, a DNA replication licensing factor, followed by DNA damage and cell cycle arrest. The importance of this mechanism in primary cancer cells, including neoplastic B-cells, has not been determined. We obtained B-cells from patients with CLL and used in vitro co-cultures with CD40L-expressing stroma to mimic the prosurvival conditions present in lymphoid tissue, and used 25 ng/mL IL-21 to facilitate CLL cell proliferation. Stromal co-cultures induced canonical and non-canonical NFB pathways and upregulated anti-apoptotic BCL2 family proteins MCL1 and BCLX in peripheral blood CLL cells, leading to survival and chemoresistance. Upon 48 hour exposure to IL-21, 15-20% of the CLL cells progressed through cell cycle. Under those conditions, CLL cells were sensitized to pevonedistat. Treatment with 1 M pevonedistat for 24 hours induced apoptosis in 63.7±2.6% (with IL-21) vs. 44.6±5.4% (without IL-21) cells (p1). Patients with del17p (>20% of cells) were excluded. Primary endpoint was IRC-assessed progression-free survival (PFS). Secondary endpoints included overall survival. Hazard ratios (HR) and 95% confidence intervals (CIs) for lowand high-risk patient subgroups were calculated. Survival was adjusted for crossover using an


inverse probability of censoring weighting method. Results: At a median follow-up of 17 months, IRC-assessed PFS favored ibrutinib+BR versus placebo+BR across all patient subgroups assessed, including high-risk groups. In the ibrutinib+BR and placebo+BR arms, respectively, 49% and 48% had received 1 prior therapy, 52% and 52% had received >1 prior therapy, 30% and 23% had del11q, 81% and 80% had unmutated IgVH, 26% and 26% were purine analogue refractory, and 58% and 54% had bulky disease (>5 cm). The reduced risk of PD or death (HR; 95% CI) in patients who received 1 prior therapy was 81% (0.19; 0.120.31); in patients who received >1 prior therapy was 78% (0.22; 0.15-0.33); in patients with del11q was 92% (0.08; 0.04-0.16); in patients with unmutated IgVH was 84% (0.16; 0.110.23); in patients refractory to purine analogue was 76% (0.24; 0.14-0.41); and in patients with bulky disease was 81% (0.19; 0.13-0.27). In patients with either a long (36 and 24 months) or a short (0.18). Of the 7 patients with CLL progression, 5 remain alive on next-line ibrutinib (n=4) or corticosteroids (1). Of the 16 patients with RS, 15 received salvage therapy; R-CHOP (n=6) and R-ICE (3) being the most commonly used regimens. 4 received consolidation with stem cell transplantation. One allogeneic stem cell recipient died at 12 months from transplantrelated complications, and the other remains alive and free of disease at 34 months. Both autologous transplant recipients have relapsed with CLL and were successful salvaged with BTK inhibitors. Conclusion: Failure of venetoclax does not automatically portend a poor prognosis. The survival of patients who progress with CLL or RS on venetoclax therapy may be superior to that reported for BTK inhibitors.

130 Novel Therapy Novel small molecule IKK inhibitors inhibit non-canonical NF- B signaling and survival of primary CLL cells Elaine Willmore1 1 1 1 1 1 Aaron Gardner , Susan Tudhope , Belinda Murtani , Jessica Caffry , Evan Mulligan , Jill 1 2 2 3 3 Hunter , Jonathan Wallis , Helen Marr , Giacomo Beretta , David Breen , Andrew Paul, Darren Edwards, Louise Young, Colin Suckling, Robin Plevin, Marie Boyd, Simon Mackay, Neil Perkins 1

Newcastle University Freeman Hospital, Newcastle upon Tyne 3 Strathclyde University


2

Constitutive activation of the stress-inducible transcription factor, NF-B, plays an important role in chemo-resistance and disease progression in CLL. NF-B exists as complexes of five protein subunits: p65 (or Rel A), p50, and c-Rel (which function by the 'canonical' pathway involving IKK (inhibitor of NF-B kinase ) and p52 and RelB which function by the 'noncanonical' pathway regulated via IKK. Most studies have focused on canonical signaling; high p65 levels confer shorter survival in CLL, and result in increased expression of prosurvival and anti-apoptotic genes. NF-B is also important in other B-cell malignancies: in multiple myeloma, mutations in genes that are key NF-B regulators occur in up to 20% of cases. These lead to e.g. increased activation of the CD40 receptor or NIK (NF-B-inducing kinase), resulting in constitutive non-canonical NF-B signaling. In CLL, the mechanisms underlying altered NF-B activity are not clearly defined. However, inducers including BAFF (B-cell activating factor of the tumour necrosis factor family) and CD40, and the negative regulator of NIK, BIRC3, have been implicated. CD40L engagement stimulates noncanonical NF-B signaling and proliferation of CLL cells and importantly, targeting B-cell receptor signaling in CLL (which also activates non-canonical NF-B activity), is proving highly effective in clinical trials. We hypothesised that receptor-activated proliferation in CLL cells is reliant on non-canonical NF-B-regulated gene transcription, and that targeting this pathway would decrease CLL cell survival. Since non-canonical NF-B signaling is driven by IKK, we evaluated potent, first-in-class IKK inhibitors. We examined compounds of known selectivity over IKK, reasoning that targeting IKK may prove cytotoxic in a more global fashion. Protein expression and DNA binding activity (ELISA) of NF-B subunits was examined in nuclear extracts from patient-derived CLL cells (n = 52). RelA and p50 were frequently high, whereas p52 and RelB varied, with high RelB in cases where RelA was low (p = 0.06). High levels of p52 were associated with high RelB expression, indicating that noncanonical signaling is important in CLL. A series of 17 IKK inhibitors were tested in viability and apoptosis assays, revealing LC50 values ranging from 0.5 - 5M in CLL cells and cell line models (48 hr exposure). Decreasing cell survival correlated with increased caspase 3/7 activation. We then used CD40L-expressing fibroblasts in co-culture with primary CLL cells to stimulate NF-B signaling and CLL cell proliferation, to more closely model the tumour microenvironment. Nuclear RelB and p52 levels increased after 2-8 hrs, but this activation (and phosphorylation of p100 (the precursor of p52, and the target of IKK) was significantly reduced following treatment with 1M IKK inhibitor (compound S, cell-free IC50 26nM). IKK inhibition had little or no effect on accumulation of nuclear p65, indicating selectivity of these inhibitors for IKK over IKK. Future studies will examine the gene expression profile following IKK inhibition in CLL, and also interrogate the underlying genotypes among patients that may drive NF-B signaling. These data provide proof of concept that targeting non-canonical NF-B signaling is a valid therapeutic strategy in CLL.


OTHER 131 Other Patient value mapping in CLL Simon Fifer1 Todd Stephenson2, Nathan Walters2, Kathryn Glase2, Richard Vines3, Sharon Millman4, John 1 Rose 1

Institute for Choice Janssen Australia 3 Rare Cancers Australia 4 Lymphoma Australia


2

Background: Incorporating patient values into health outcomes is becoming more and more important (known as Patient Value Mapping). Unfortunately in most markets the patient voice is given little consideration in government or private payer decisions to reimburse medicines. In Australia, there is a consumer representative on the Pharmaceutical Benefits Advisory Committee (PBAC) and patients have the opportunity to provide written input during the assessment process, although the process of how PBAC consider and incorporate this information into their decision is not transparent. There is a need for a more formalized framework for eliciting meaningful patient input and a more transparent process for how that input is incorporated into the decision making process. Objectives: This research seeks to outline a novel methodological approach to elicit and quantify patient values in a systematic way for the purpose of treatment evaluation. Methods: This research examines a quantitative approach using discrete choice-based methods to elicit and quantify patient values for the purpose of treatment evaluation. This pilot focuses on patients with Chronic Lymphocytic Leukemia (CLL) and investigates their view of current treatments and what outcomes are most important to them. The figure below outlines the structure and stages of the research. Both personal and treatment preferences are included in the modelling process. •

Personal preferences: Measure the relative importance of personal outcomes related to CLL, focusing on quality of life aspects, impact on family and caregivers, and other important factors outlined in the exploratory stage (i.e., moving beyond the traditional focus on clinical outcomes of treatment and incorporating other outcomes into the treatment evaluation process).

•

Treatment preference: Quantify patient treatment preferences for clinical outcomes - what is most important to patients when deciding to undertake treatment (i.e., trade-offs between outcomes, benefits and risk). Results: Current and proposed treatments are entered into the resulting patient preference model and scored based on how well they align with patient values and expectations. The results of this analysis could be incorporated into the treatment evaluation process and used to guide decisions around the value of new medicines.

132 Other A mechanistic rationale for testing the combination of duvelisib (IPI-145) and venetoclax (ABT-199) in chronic lymphocytic leukemia Viralkumar Patel1 1 2 2 2 Kumudha Balakrishnan , Mark Douglas , Thomas Tibbitts , Jeffery Kutok , Renato 3 1 1 1 2 Guerrieri , William Wierda , Susan O'Brien , Nitin Jain , Howard Stern , Varsha Gandhi1 1

UT MD Anderson Cancer Center Infinity Pharmaceuticals Inc. 3 Davidson College


2

Background: Inhibition of BTK with ibrutinib, PI3K- with idelalisib, and PI3K-,- with duvelisib (IPI-145), have shown clinical activity in chronic lymphocytic leukemia (CLL). These B-cell receptor (BCR) pathway inhibitors impede proliferation of CLL cells and disrupt the ability of the tumor microenvironment to support growth and survival of malignant cells. Duvelisib, a potent, oral, small molecule dual inhibitor of PI3K-,-, is currently in phase 3 clinical development in CLL. In preclinical studies, duvelisib inhibited the PI3K pathway and promoted apoptosis in primary CLL lymphocytes (Balakrishnan et al, Leukemia 2015). Although most CLL patients respond to BCR network inhibitors, some patients who initially respond later develop resistance or have persistent lymphocytosis. Identifying and overcoming resistance mechanisms will be crucial for the most effective combinatorial use of these agents. We hypothesize that gene expression changes in circulating CLL cells during duvelisib treatment could point toward pathways or targets that may synergize with PI3K-,- inhibition. Methods/Results: To test this hypothesis, we isolated RNA from blood of CLL patients on the IPI-145-02 phase 1 duvelisib clinical trial. RNAseq data were evaluated at baseline (C1D1) and after 8 days of treatment (C1D8) in 31 patients to identify gene expression changes associated with duvelisib treatment. BCL2 and a number of BH3-only genes (PUMA, BIM, HRK, NOXA, and BMF) were upregulated. BIK, a BH3-only gene, was down-regulated. To validate these findings, PBMCs were collected at C1D1 and C2D1 from some of the same patients. Microfluidic chip qRT-PCR confirmed the gene expression changes for BCL2 (mean fold ± SEM: 3.0±0.4; p=0.0020; n=10) and most of the BH3-only genes indicated above. Real-time RT-PCR data further validated these findings. In addition, protein was isolated from PBMC samples for reverse-phase protein array (RPPA) analysis. Interestingly, of the >220 proteins analyzed, BCL2 was among the most significantly elevated proteins at C2D1 (1.7±0.2; p=0.015; n=7). These findings were corroborated by a trend toward elevated BCL2 protein by western blot (1.3±0.1; p=0.086; n=7). These data suggest a model in which upregulation of pro-apoptotic BH3-only genes may be offset by induction of the anti-apoptotic gene BCL2, providing a mechanistic rationale for combining duvelisib with the BCL2 antagonist venetoclax (ABT-199). To test for combinatorial activity, PBMCs from CLL patients on the IPI-145-02 trial were treated exvivo with 10 nM venetoclax. Venetoclax induced enhanced apoptosis in samples obtained during duvelisib treatment (79%) compared to samples obtained before treatment (58%) (n=5; p=0.041). The level of apoptosis as determined by Annexin V/PI staining in CLL cells from duvelisib-treated patients (n=15) was greatest after ex-vivo treatment with venetoclax (45%, p100g/L. PAMS/PNP is a rare but potentially devastating complication of CLL that can be extremely difficult to manage and is associated with a very high one year mortality risk. Bcell receptor signal pathway blockade with the Btk inhibitor ibrutinib achieved control of both the CLL and of its autoimmune complication PAMS. Ibrutinib may be a useful additional tool for the management of PAMS.

135 Other Prevalence and economic burden of chronic lymphocytic leukemia (CLL) in the era of oral targeted therapies Nitin Jain1 2 2 1 3 1 Qiushi Chen , Turgay Ayer , William Wierda , Susan O'Brien , Michael Keating , Hagop 1 4 Kantarjian , Jagpreet Chhatwal 1

Department of Leukemia, MD Anderson Cancer Center Georgia Institute of Technology 3 Chao Family Comprehensive Cancer Center, University of California Irvine 4 Department of Health Services Research, MD Anderson Cancer Center


2

Background: Better understanding of the disease biology has led to significant advances in the treatment of CLL. Oral targeted agents such as ibrutinib and idelalisib are currently approved for patients with relapsed CLL. Ibrutinib is also approved for patients with del(17p). Several other targeted therapies are expected to become available in the near future. These therapies (ibrutinib, idelalisib) have shown to improve survival in Phase III studies. However, their high cost, approaching more than $130,000/year for an indefinite duration of treatment, has raised concerns about their affordability and cost to the society (Shanafelt et al. JOP 2015). Our objective was to project the future prevalence and cost burden of CLL in the context of emerging therapeutic options. Methods: We developed a Markov micro-simulation model representing the CLL population to project the prevalence and total cost of CLL in the United States for each year from 2011 to 2035. Our model was calibrated to the Surveillance, Epidemiology, and End Results (SEER) data and closely predicted the CLL prevalence in 2011. The model included new incidences every year, and considered the individual patient's aging, disease progression, and treatment with the available therapies in the given year. The disease progression was estimated from published clinical trials presenting progression-free and overall survival data. For each patient, treatment was assigned based on the fitness status (determined by age) and the presence of del(17p). Cost estimation included the cost of drug, treatment administration, and management of adverse events. We simulated changes in the prevalence and the cost of CLL assuming oral targeted therapies as the standard-of-care for patients with relapsed CLL and for patients with del(17p) from 2014 onwards, and in the firstline setting from 2017 onwards (Fig 1A). For comparison, we also ran a scenario assuming chemoimmunotherapy (CIT) remains the standard-of-care from 2011 onwards (Fig 1B). Results: With targeted therapies, the prevalence of CLL is projected to increase from 120,000 in 2011 to 180,000 in 2025 (Fig 2A). Oral targeted therapies would result in additional 170,000 person-years in the next 10 years in the US. The annual cost of CLL treatment would increase from $0.9 billion in 2011 to $3.5 billion in 2025 (Fig 2B). Compared with CIT, oral targeted therapies would cost additional $15 billion over the course of next 10 years. The increase in prevalence and costs would be driven by substantially improved survival and continuous administration of the expensive oral therapies. Conclusion: Oral targeted agents represent a significant advance for the treatment of CLL. However, they will dramatically increase the cost burden of CLL. Such an economic impact could result in limited access and lower adherence to the oral therapies, which may undermine their clinical effectiveness. Given the increasing number of patients on first-line oral therapy that is continuously administered over an extended duration, a more sustainable pricing for such therapies is needed.


Figure 1. Scenarios of treatment strategies with emerging therapeutic options.

Figure 2. Estimates of prevalence and cost burden for oral targeted therapy and chemoimmunotherapy scenarios.

136 Other Blindness occurring in the setting of chronic lymphocytic leukaemia: an area of diagnostic uncertainty David Kliman1 Natasha Gerbis2, Karl Ng2, Kim Tan3, Naomi Mackinlay1, Suneet Sandhu1, Stephen 1 Mulligan 1

Department of Haematology, Royal North Shore Hospital Department of Neurology, Royal North Shore Hospital 3 Department of Ophthalmology, Royal North Shore Hospital


2

Central nervous system (CNS) involvement is a rare complication of CLL. Blindness has rarely been reported. We describe two patients and the diagnostic difficulties. Patient one was a 70-year old male with a two-year history of untreated CLL. Initially unilateral progressive vision loss occurred over a two-day period in the right eye. Apart from lymphocytosis and lymphadenopathy, the only abnormality on examination was complete loss of vision without light perception in the affected right eye, with mild optic disc oedema. MRI brain revealed a 4mm area of diffusion restriction at the distal right optic nerve, suggestive of a leukaemic infiltrate. Despite treatment with high dose methylprednisolone and intrathecal methotrexate, the vision loss rapidly progressed with complete bilateral blindness. No evidence of leptomeningeal CLL was found on cytology and flow cytometry of the cerebrospinal fluid (CSF). Positron emission tomography demonstrated low grade avidity in a small left medullary lesion. Subsequent neurological changes progressed with bilateral hand and rightsided facial numbness. The CNS lesions were not amenable for biopsy. Systemic chemotherapy was commenced for presumed CLL of the central nervous system, using fludarabine and rituximab as well as radiotherapy to the orbits bilaterally. There was no further progression of neurological abnormalities, though the patient never regained vision. Patient two, a 68-year old female, had CLL diagnosed 10 years previously. Prior therapies included fludarabine, cyclophosphamide and rituximab (FCR) chemo-immunotherapy as well as splenectomy for autoimmune haemolytic anaemia. She presented with fevers and confusion. Meningitis due to listeria monocytogenes was diagnosed by lumbar puncture. Following extended treatment with 6 weeks intravenous antibiotics, complete neurological recovery occurred. She re-presented 2 months later with similar symptoms. No infectious organisms were seen on repeat CSF analysis, though small numbers of lymphocytes with a CLL phenotype were seen on cytology and flow cytometry. Confusion and drowsiness progressed to coma despite empiric intravenous antibiotics. MRI of the brain revealed widespread gadolinium enhancement in the meninges and cerebral vessels. Surgical biopsy of the frontal lobe and meninges showed some scattered CLL cells. Due to possible CNS CLL as the cause for her deterioration, FCR chemotherapy was administered. The patient gradually became more responsive but with severe impaired visual acuity bilaterally and perception of light only. A repeat MRI showed reduced meningeal enhancement but enlargement and enhancement of both optic nerves. No infectious or autoimmune aetiologies were found. Despite continuation of systemic chemo-immunotherapy, the vision loss persisted. Causes of blindness in patients with CLL include mass effect, leptomeningeal infiltration and infection with typical or atypical organisms. Due to peripheral blood lymphocytosis, biopsy samples may be difficult to interpret. In the second case, prior infection may have triggered recruitment of CLL cells into the CNS, as has been seen as a 'bystander' effect in other infectious lesions. Treatment options include intrathecal and


systemic chemotherapy or radiotherapy. Despite this, outcomes are often poor, therefore early suspicion and treatment is essential. The cases highlight the diagnostic and therapeutic difficulties in this setting.

137 Other Establishment of a pre-clinical in vivo platform spanning low-risk to high-risk CLL Gero Knittel1 Paul Liedgens1, Christian Fritz1, Darya Korovkina1, Malte Hülsemann1, Yussor Al-Baldawi1, 1 1 1 Thorsten Persigehl , Lukas Frenzel , Christian Reinhardt 1


University Hospital Cologne

Chronic lymphocytic leukemia is extraordinarily heterogeneous in its clinical course. While some patients survive for decades without the need for therapeutic intervention, others suffer from rapidly progressing disease. Specifically, patients with deletions of the short arm of chromosome 17, which is where TP53 is located, and patients with 11q deletion, the location of ATM, have a dramatically worse prognosis compared to karyotypically normal patients (32, 79 and 111 months, respectively). Currently, the lack of genetically engineered mouse models (GEMMs) faithfully recapitulating the disease developed by these patient subgroups is a critical bottleneck for the evaluation of novel targeted therapies. On the basis of the well established ETCL1 mouse model, we generated new GEMMs with B cell-specific loss of fl/fl fl/fl either ATM (ETcl1;Cd19:Cre;Atm , abbreviated TCA) or Tp53 (ETcl1;Cd19:Cre;Tp53 , abbreviated TCP), which mimic the more aggressive course of disease and shorter life expectancy correlated with these lesions in humans. While histologically similar phenotypes are observed in TCP, TCA and control mice (ETcl1;Cd19:Cre, abbreviated TC), TCP and TCA mice show a significantly faster increase in CD5+/CD19+ leukemic cells in the blood stream, as well as accelerated thrombocytopenia, compared to TC control mice. To noninvasively assess splenomegaly, a hallmark of CLL, we established a magnetic resonance imaging protocol. In addition to these mice with a constitutive Cd19:Cre allele, we made use of the Cd19:CreERT2 allele to generate inducible models which, upon injection of tamoxifen, lose either Tp53 or Atm in a B cell-specific manner. The relatively low recombination efficiency of the Cd19:CreERT2 allele allows us to create a situation where recombined and unrecombined B cells coexist in vivo and where the Tp53- or Atm-deficient clone represents only a small fraction of leukemic cells, mimicking the appearance of a new clone in human patients. To identify the recombined clone and follow its fate, we crossed in a Cre reporter, which upon recombination switches from the expression of tdTomato to eGFP expression. Taken together, we have established a fully functional preclinical in vivo platform that allows for testing and screening of new compounds and treatment strategies in novel GEMMs that recapitulate the human disease. We can closely and longitudinally monitor the course of the disease and the impact of therapeutic interventions via the analysis of peripheral blood and MRT imaging. With our fluorescence-based reporter system, we are able to assess the fate of Tp53- or Atm-deficient subclones under treatment. For instance, we demonstrated an increased sensitivity of Atm-deficient EMyc lymphomas and primary human del11q CLL cells to DNA-PKcs inhibition has and we will now use our novel in vivo platform to validate this observation.

138 Other German patient and physician preferences for treatment characteristics in relapsed or refractory chronic lymphocytic leukaemia: a conjoint analysis Erik Landfeldt1 1 2 2 3 3 Jennifer Eriksson , Steve Ireland , Patience Musingarimi , Claire Jackson , Emma Tweats , 2 Maren Gaudig , Mark Wildgust 1

Mapi Group Janssen 3 Adelphi


2

Background: Due to the clinical heterogeneity of the disease, and the advanced age of those affected, historically, no universal treatment algorithm has been applicable to all patients with chronic lymphocytic leukaemia (CLL). Limited options have also been available regarding modalities of administration. However, a number of novel modalities recently licensed and currently in trial are expected to improve patient outcomes across a spectrum of CLL patients. Aims: The objective of this study was to estimate preferences in a conjoint analysis (CA) for treatment characteristics (so called "attributes") in relapsed/refractory (r/r) CLL among German patients and physicians (haematologists/oncologists) experienced in treating CLL. Methods: The most relevant treatment attributes for CLL to include in the CA were identified through a literature review and in qualitative interviews with six haematologists/oncologists and six r/r CLL patients in Germany. Six treatment attributes, each described in three levels, were selected by this process: (i) overall survival (OS) (Levels: 5,3,2 years), (ii) progression-free survival (PFS) (3,2,1 years), (iii) fatigue (no to mild, mild relieved by rest, mild not relived by rest), (iv) nausea (no, mild, moderate), (v) risk of serious infection (low, moderate, high), and (vi) treatment administration (oral pill daily at home, monthly infusion at hospital, weekly infusion at hospital). The final survey consisted of 13 CA tasks. Full profiles were presented and a forced-choice elicitation format was chosen. We estimated the relative importance of each attribute by fitting a hierarchical Bayesian model. Attribute importance is a measure of how much influence a specific attribute has on individual choice and expresses the respondent's willingness to trade between attributes, and range between 0% and 100% per attribute (and sum to 100% across all attributes). Patients were asked to base choices on their own preferences. Physicians were asked to make their choices as if they were patients. All participants provided informed consent and the study was approved by Freiburger Ethik Kommission International (FEKI). Results: A total of 44 patients and 50 physicians participated. In the pooled sample, OS was the most important attribute (40%; bootstrapped 95% confidence interval: 38%-43%), followed by risk of serious infection (18%; 16%-20%), treatment administration (14%; 12%-16%), fatigue (11%; 10%13%), PFS (10%; 9%-11%), and nausea (6%; 6%-7%). OS and PFS were significantly more important to physicians (43% vs. 37%, p=0.001; and 11% vs. 9%, p=0.007, respectively), whereas treatment administration was more important to patients (18% vs. 11%, p=0.001). Conclusions: We show that overall survival is the most important attribute of treatments for r/r CLL, followed by risk of serious infection, and treatment administration. Our results suggest that patients to a larger extent than physicians value convenient administration. These data could help align therapeutic decision-making in CLL with patient preferences to improve patient care satisfaction and treatment compliance. Preferences for treatment attributes should be taken into account in decisions regarding endpoints in clinical trials and in HTAs of CLL interventions to help capture their true value from a patient perspective.

139 Other Recent enhancements of minimal residual disease detection by flow-cytometry in chronic lymphocytic leukemia: a study of the French CLL-MRD group Abdelmalek Dahmani1 2 3 4 5 Christine Arnoulet , Lucile Baseggio , Lydia Campos , Bernard Chatelain , Agathe 6 6 7 8 Debliquis , Bernard Drenou , Marie-Christine Jacob , Eric Legac , Magali Le Garff9 9 10 4 11 Tavernier , Claire Quiney , Nelly Robillard , Françoise Solly , Michel Ticchioni , 12 13 14 9 Guillaume Cartron , Caroline Dartigeas , Pierre Feugier , Veronique Leblond , Stéphane Leprêtre15, Eric Van Den Neste16, Marie-Hélène Delfau-Larue17, Florence Cymbalista1, Remi Letestu1 1

Hôpital Avicenne, APHP, Université Paris 13, Bobigny CLCC IPC, Marrseille 3 CHU Pierre Bénite, Lyon 4 CHU St Priest en Jarez, St Etienne 5 UCL Mont-Godinne, Yvoir 6 CH E Muller, Mulhouse 7 EFS, Grenoble 8 CHR de la Source, Orléans 9 Hôpital Pitié-Salpêtrière, Paris 10 CHU Nantes 11 Hôpital L'Archet, Nice 12 CHU Montpellier 13 CHU Tours 14 CHU Nancy 15 Centre Henri Becquerel, Rouen 16 UCL Saint-Luc, Bruxelles 17 CHU Henri Mondor, Créteil


2

Introduction: Evaluation of minimal residual disease (MRD) is a strong predictor of the -4 outcome after FCR-based immunochemotherapy. Patients achieving MRD below 10 have longer OS/PFS compared to subjects with MRD above 0.01%. The 6-CLR strategies were found comparable to 4-CLR approaches and even as effective as new molecular techniques such as high throughput sequencing between 10-4 to 10-5 levels. As MRD became an endpoint with increasing importance in the recent clinical trial conducted by the French Intergroup, we developed and evaluated new approaches of flow-cytometry techniques. Methods and Results: Initially, a 6-CLR panel was designed comprising the main markers validated in the standardized panel of the ERIC, a harmonization effort was done by using fluorescent beads to standardize the settings. The cellular background was evaluated and the limit of detection of the technique was estimated at 0.7x10-5. To alleviate MRD testing we opted for a singletube strategy. Thus, an innovative 8-CLR combination was developed. We harmonized the instrument settings and used daily adjustments to allow the consistency of the results during the study. We tested the use of CD43 as an alternative for CD45. To improve the discrimination between CLL cells and normal T and B-cells, we evaluated the combination of CD79b and CD22 in the same fluorescence channel resulting in an 8-color 9-antibody panel. The inclusion of CD200, a marker negative on the T and NK cells but overexpressed on CLL cells, was a clear benefit for the specificity of the panel. To assess the cellular background of the panel, we tested control samples and found no false positive events among 2 million leukocytes. Therefore, we developed an acquisition procedure to increase the number of events recorded conciliating reasonable time of acquisition. By using the strategy of


fluorescence triggered acquisition, we were able to evaluate the cellular background among 10 million leukocytes. For this panel the limit of detection was established at 0.3x10-6 and the technique was cross-validated by IGH ASO RQPCR. Finally, we developed a new strategy for the analysis of the results. Our goal was to help the positioning of the gates by using a control population specific for each dot-plot. Conclusions: In our group we have developed new approaches for CLL-MRD detection by flow-cytometry, improving both the production of the data and their interpretation. These procedures were therefore validated for MRD monitoring in clinical trials. They proved to be reliable in a multicentric setting.

140 Other A case of long term survival in a patient with advanced stage chronic lymphocytic leukemia with central nervous system involvement


Charles Li A 56-year-old woman was referred for assessment of cervical lymphadenopathy and lymphocytosis. She felt well but had noted lumps around her neck for two months prior to presentation. She had lost 8 lb but denied fever and night sweats. Flow cytometry confirmed Chronic Lymphocytic Leukemia (CLL), CD 38 negative. Abdominal ultrasound showed an enlarged spleen at 14 cm. Rai Stage 2 CLL was diagnosed and no specific treatment was instituted with the patient feeling completely well. Florescent in-situ hybridization (FISH) for high risk disease (including 17p- and 11q-) was negative. 9 months after diagnosis, the patient presented with severe anemia with no evidence of hemolysis or blood loss. Significant worsening of cervical lymphadenopathy was noted. Rai Stage 3 CLL was diagnosed and systemic treatment with chemotherapy with fludarabine and rituximab (FR) was offered but the patient declined intervention. The next month the patient developed severe numbness and tingling in her feet. Detailed neurologic assessment showed findings suggestive of myelopathy. MRI of the spinal cord showed diffuse abnormalities, consistent with leptomeningeal and intramedullary involvement by CLL. The same imaging also showed extensive lymphadenopathy in the neck , chest, abdomen and pelvis. Blood lymphocytosis was marked and bone marrow examination revealed extensive infiltration of marrow space by small mature lymphocytes, representing 90% of bone marrow cellularity. Bone marrow FISH again did not reveal any high risk features. CSF showed small mature lymphocytes that were clonal. Advanced stage CLL with leptomeningeal and intramedullary involvement was diagnosed. Lumbar puncture with intrathecal methotrexate (IT MTX) was given and the same continued for a total of 12 treatments via an Ommaya reservoir (IT MTX 12 mg in 6 mL preservative free normal saline twice weekly X 4, then once weekly X 4, then once monthly X 4). CSF was cleared of clonal lymphocytes after 4 IT MTX treatments. Systemic chemotherapy with FR was given concurrently for a total of 6 months cycles. The patient tolerated treatments well and her CLL went into complete remission both systemically and from the standpoint of Central Nervous System (CNS) involvement. Repeat MRI showed the spinal cord returning to normal at the end of treatments. 6 years after the initial diagnosis of CLL (5 years after the last chemotherapy), the patient was completely well with a normal hematology panel (Leukocytes 6.0 Giga/L , Hemoglobin 131 g/L, Platelets 162 Giga/L, Neutrophils 2.8 Giga/L, Lymphocytes 2.8 Giga/L, Monocytes 0.2 Giga/L). Repeat detailed assessment by the patient's neurologist showed normal results. 6 months later, the patient noted cervical lymphadenopathy and blood lymphocytosis re-emerged. These worsened over time and CT scan showed extensive lymphadenopathy. Systemic chemotherapy with FR was started and cervical lymphadenopathy resolved after the first cycle of treatment. To date the patient has done well having completed 4 of 6 intended cycles. CLL uncommonly involves the CNS. This presentation can be dramatic but does not necessarily portend a poor prognosis. With prompt recognition and treatment with a combination of systemic and intrathecal chemotherapy, patients can do very well with long-term event free survival.

141 Other Cost-effectiveness analysis of idelalisib-rituximab in relapsed or refractory CLL Monia Marchetti1 Antonio Cuneo2, Marco Montillo3, Mauro Francesca Romana4, Elisa Martelli5, Maria Paola 5 Pedone 1

Oncology-Hematology Unit, Hospital C. Massaia, Asti Hematology Unit, Dept of Medical Sciences, University of Ferrara 3 Service of Haematology, Niguarda Cancer Center, Milan 4 Hematology Dpt, University La Sapienza, Rome 5 Health Economics and Outcomes Research, Gilead Italia


2

Background: Idelalisib is a first-in-class PI3k inhibitor recently reported to prolong survival of relapsed/refractory CLL patients (R/R CLL) in combination with rituximab (IR), as compared to rituximab single-agent (R). The economic impact of this new oral drug, however, is still to be assessed. Aims: To investigate the comparative costs and benefits of IR versus R, the control arm in the reported randomized trial, and versus bendamustinerituximab (BR) and fludarabine-cyclophosphamide-rituximab (FCR), the most widely used second-line treatments for R/RCLL. Methods: We simulated six 2nd3rd line treatment sequences considering that the patients incurring progression on IR crossed-over to one of the comparator treatments, while patients progressing on R, BR or FCR crossed-over to IR. We therefore built by TreeAgePro2015 a Markov model including 5 health states: i) progressionfree on therapy, ii) progression-free on extended-therapy (for IR only), iii) progression-free off-therapy, iv) progressed disease, v) death. The model was run for 360 monthly cycles, approximating a lifelong horizon. The typical patient analyzed at baseline was a 68-year old patient with a male to female ratio of 1.8 (HEMACARE). Probabilities of progression were obtained from published randomized studies (Furman et al. 2014, Awan et al. 2014) or phase II studies (Fisher et al. 2011) if randomized trials were not available for the specific RR CLL setting. Data were adapted to a second-line setting according to fixed ratios: 0.67 for 3rd to 4th nd th line ratios, 0.44 for 2 to 4 line ratios. Age- and sex-adjusted general mortality was calculated from Italian 2012 life tables. Quality of life of progression-free patients was 0.85 while it was 0.65 for those with progressive disease (Beusterien et al. 2010). The analysis was performed from the perspective of the Italian health-care system, therefore administration costs for endovenous drugs (288/day), ex-factory drug costs (idelalisib 66.67/150 mg tablet; rituximab 277/100 mg vial) and costs of adverse events (febrile neutropenia 2,956, grade 3-4 thrombocytopenia 1,994, grade 3-4 diarrhea 416, grade 3-4 anemia 1,322) were considered. Furthermore, a yearly discount rate of 3% was applied to future costs and benefits according to international guidelines. Beta distributions were used for probabilities and gamma distributions for costs. First- and second-order sensitivity analysis and Monte-Carlo analysis were performed to test the robustness of the results. Results: Second-line treatment with IR was expected to improve quality-adjusted life expectancy (QALE) of R/R CLL patients by 0.93-2.08 quality-adjusted years (QALY) at an incremental cost ranging from 19,564 to 29,817, as compared with R, BR and FCR. The incremental cost-utility ratio (ICUR) ranged from 14,350/QALY versus R to 29,397/ QALY versus BR. IR had a reasonable economic profile as compared with FCR as well. At Monte-Carlo analysis IR resulted to gain an ICUR lower than 40,000/QALY with a probability of 76% versus FCR and 93% versus BR. Results were moderately sensitive to the time horizon of the analysis, idelalisib unit cost and the duration of treatment of idelalisib after cross-over. Conclusions: IR is an effective therapy and can be considered a sustainable therapeutic option for R/R CLL patients.



QALE INCREMENTAL THERAPY COSTS (QALY) COST RIR

143,812 2.41

IRR

173,630 4.48

BRIR

128,713 2.95

IRBR

169,188 4.48

FCRIR

147,749 3.55

IRFCR

167,313 4.48

INCREMENTAL ICUR EFFICACY (QALY) (/QALY)

29,817

2.08

14,350

40,475

1.53

26,397

19,564

0.93

20,993

142 Other CD200 improves the diagnosis and prognostification of chronic lymphocytic leukemia Lei Fan1 Yi Miao1, Yu-Jie Wu1, Yan Wang1, Rui Guo1, An-Li Shen1, Yao-Yu Chen1, Wei Xu1, Jian1 Yong Li 1


Department of Hematology, the First Affiliated Hospital of Nanjing Medical University, Jiangsu Province Hospital

Objective: CD200, formerly known as OX-2, is a type I glycoprotein that is expressed on a variety of cell types. CD200 has both diagnostic and prognostic significance in hematological neoplasms. A few studies have confirmed that CD200 immunophenotyping was useful in differentiating chronic lymphocytic leukemia (CLL) from other leukemic mature B cell neoplasms, especially mantle cell lymphoma (MCL). Additionally, CD200 expression has been identified as a prognostic factor in acute myeloid leukemia and multiple myeloma. However, the diagnostic value of CD200 has not been validated in a relatively large cohort and the prognostic role of CD200 expression in CLL remains to be determined. We therefore sought to investigate the value of CD200 in the diagnosis and prognostification of CLL. Methods: Between November 2009 and January 2015, four hundred and seventy five fresh peripheral blood (PB)/bone marrow (BM) samples prior to therapy were collected from patients with leukemic phase of B-cell non-Hodgkin lymphomas including 307 cases of CLL, 44 cases of MCL, 32 cases of spleen marginal zone lymphoma, 6 cases of nodal marginal zone lymphoma, 4 cases of mucosal-associated lymphoid tissue lymphoma, 28 cases of follicular lymphoma, 15 cases of B-cell prolymphocytic leukemia, 4 cases of hairy cell leukemia (HCL), and 33 cases of lymphoplasmacytic lymphoma. Flow cytometric analysis was performed, after subgating on CD19 positive tumor cells, CD200 expression positivity and mean fluorescence intensity (MFI) were calculated. ROC (receiver operating characteristic curve) and AUC (area under the ROC curve) were established to evaluate diagnostic value of CD200 in differentiating among leukemic mature B cell neoplasms. Survival curves were plotted using Kaplan-Meier method and log-rank test was used for comparison. Results: CD200 was variably expressed on tumor cells among leukemic mature B cell neoplasms. The CD200 MFI of tumors cells for patients with CLL was significantly higher than that of any other mature B cell neoplasms except HCL, which displayed bright CD200 expression. The median CD200 MFI was 192.6 and 10.5 for CLL and MCL, respectively. ROC plot reflected a strong separation between CLL and MCL with an AUC of 0.9540 (p3.5 mg/L were found in 103/138 patients (74%) and 85/128 (66%) patients had unmutated IGVH genes. At a median follow-up of 45.8 months (range 0 - 77), 83 (52%) events occurred in PFS and 33 (21.7%) events in OS. EGR2 mutations were found in 2/152 cases (1.3%), BIRC3 mutations in 3/81 cases (3.7%). Among the 152 cases analyzed, TP53 was mutated in 16 (10.5%), NOTCH1 in 26 (17%) and SF3B1 in 25 (16.4%). Because of the low frequencies of mutations in BIRC3 and EGR2, these genes were excluded from further analyses. NOTCH1 and SF3B1 mutations were almost non-overlapping and coincided in only 1 patient. NOTCH1 mutations correlated significantly with trisomy 12 (p=0.0025). TP53 mutations were associated with del(17p) (p=0.0001), high levels of 2microglobulin (p=0.019) and refractoriness to FCR (p=0.01). No associations were found for SF3B1 mutations. In univariate analyses, PFS was significantly inferior for mutated TP53 (8.8 vs 46.4 mo, p6 and/or CrCl 3, and CD38 was positive in 45% of cases. Performance status (97% 2) and nutritional status (median Corporal Mass Index of 25.25 kg/m) were preserved. The Cumulative Index Rating Scale (CIRS) comorbidity score was 5.8 and median creatinine clearance was 50.5 ml/min (Cockroft formula). Majority of patients lived at home (90.4%) but with familial or professional help. A complete geriatric assessment has been performed for 12% of them. At treatment initiation, Binet stage was either A (28.5%), B (27.3%) or C (42.4%). Therapy consisted mainly in Chlorambucil (65%), Bendamustine (10%) and Rituximab (44%). Indeed, therapy regimens were composed of Chlorambucil alone (41%) or chemo-immunotherapy (44.7%) including Rituximab+Chlorambucil (17.41%), Rituximab+Bendamustine (8.42%), Rituximab+Cyclophosphamide+Dexamethasone (5%). In terms of tolerance, 20% of the patients required hospitalization, in 11% of the cases for febrile neutropenia. Finally, 32% required a dose reduction of chemotherapy. The Overall Response Rate was 65% with 30% of clinical complete remission. The median OS and PFS (from treatment initiation) were 49 and 17.6 months, respectively. Very few data are available about this very old population. Bairey et al reported a series of 214 patients (80 years or older) diagnosed in Israel between 1979 and 2009 with a mean age of 84. However, in this cohort, 56% had Rai stage 0 disease and only 53 patients received treatment. Median survival was 56 months. Patients diagnosed and treated with CLL over 80 years of age have several years of life expectancy. We report a large series of such patients who received first line treatment at a median age of 83. It suggests that treatment is feasible, even with the use of immunochemotherapy. Prospective trials should target this population. Oncogeriatric evaluation and new targeted therapies should be part of such future trials.


Reference

175 Treatment Comparison of outcomes in chronic lymphocytic leukemia (CLL) with the addition of rituximab to initial treatment: a comparative effectiveness analysis in the province of British Columbia, Canada Alina Gerrie1 Lauren J. Lee2, Steven J.T. Huang1, Cynthia L. Toze1, Tanya L. Gillan3, Joseph M. Connors4, 4 3 5 Laurie H. Sehn , Helene Bruyere , Khaled M. Ramadan 1


Leukemia/BMT Program of BC, British Columbia Cancer Agency, University of British Columbia 2 Division of Hematology, University of British Columbia 3 Department of Pathology and Laboratory Medicine, University of British Columbia 4 Centre for Lymphoid Cancer Research, British Columbia Cancer Agency, University of British Columbia 5 Division of Hematology, St. Paul's Hospital, University of British Columbia Background: Rituximab (R) has been available for first-line CLL treatment in BC since 2004. We compared clinical outcomes with and without addition of R to chemotherapy in a large unselected provincial cohort of patients treated for CLL to determine the real-world effectiveness of addition of R to standard chemotherapy. Methods: Three large provincial databases were used to identify eligible patients: the BC Provincial CLL, BCCA Lymphoid Cancer, and Providence Hematology CLL databases. All patients who received minimum 1 cycle of first-line treatment for confirmed CLL/SLL were included. Overall survival (OS) was calculated from date of initial treatment to death; treatment-free survival (TFS) from date of initial treatment to next treatment/death. Multivariate analysis (MVA) was performed using Cox proportional hazard models. Results: A total of 3328 patients diagnosed with CLL from 1973-2014 were identified, of which 1345 patients (40%) received treatment in followup. Of treated patients, 694 (52%) received chemotherapy alone (NoR) and 651 (48%) received R (+R). Treatment included purine-analogs in 34% and 78% in the NoR and +R groups respectively (P20 mg/day) were excluded per protocol. Other patients with AIHA/ITP including those meeting IWCLL 2008 criteria for indication for treatment were eligible for enrollment. Results: AIHA/ITP status at study entry is shown in Table 1. Approximately twice as many patients randomized to ibrutinib (10.8%) had AIHA at baseline compared to ofatumumab (4.7%). Patients on ibrutinib with ongoing AIHA had median hemoglobin of 10.4 g/dL (range, 8.1 - 13.7 g/dL), which increased to 12.4 g/dL (range, 10.3 - 14.6 g/dL) at 24 weeks (Figure 1A). Patients on ibrutinib with ongoing ITP had 9 9 9 median platelet count 48.510 /L (range, 20 - 13810 /L) at baseline compared to 94 10 /L 9 (range, 64 - 24810 /L) at 24 weeks (Figure 1B). Baseline hemoglobin for AIHA patients was similar between treatment arms, whereas baseline platelet count in ITP patients was lower with ibrutinib than ofatumumab (median 66 x 109/L; range, 23-126x109/L). Two ibrutinib patients with AIHA were receiving concomitant corticosteroids for autoimmune complications at baseline and both discontinued corticosteroids during therapy, one on Day 42 and the other on an unspecified date. Only 1 patient not receiving steroids at baseline

began corticosteroids for autoimmune-related hematologic reasons during ibrutinib treatment, versus 4 on ofatumumab. Median treatment duration (adverse event follow-up) was longer for patients on ibrutinib (18.3 months) than ofatumumab (5.3 months). In all treated patients (ibrutinib n=195; ofatumumab n=191), 2 patients on ofatumumab developed 3 episodes of AIHA (n=1 Grade 2, n=2 Grade 3); 2 other patients on ofatumumab developed 3 episodes of ITP (all Grade 4). All 6 episodes occurred either prior to crossover or in the absence of crossover to the ibrutinib arm. No new cases of autoimmune cytopenias were observed throughout study follow-up in patients randomized to ibrutinib. Conclusions: A role for ibrutinib in controlling AIHA/ITP in patients via BTK inhibition has not been widely explored. These data corroborate findings that ibrutinib does not precipitate recurrence of AIHA/ITP in patients with CLL and can be administered in patients with a history of these complications. Table I. AIHA and ITP status in patients receiving Ibrutinib or Ofatumumab


Figure 1. Patients with autoimmune cytopenias on Ibrutinib


181 Treatment Bendamustine with rituximab (BR) is safe treatment option with high response rate for chronic lymphocytic leukemia in elderly patients with comorbidities Petra Obrtlikova1 1 2 2 2 Martin Spacek , Michael Doubek , Marketa Hadrabova , Anna Panovska , Katerina 1 1 Svackova , Marek Trneny 1


First Department of Medicine - Dept. of Hematology, General University Hospital, Charles University in Prague 2 Department of Internal Medicine - Hematology and Oncology, University Hospital Brno Background: Fludarabine, cyclophosphamide, and rituximab (FCR) is the standard frontline therapy for medically fit patients with CLL (CLL8 study). However, most patients with CLL are elderly and have pre-existing comorbidities. For unfit patients, there is a combination of chlorambucil with a monoclonal anti-CD20 antibody, preferably obinutuzumab, the newly defined standard of care in first-line treatment of CLL patients with relevant comorbidities (CLL11 study). Recently, bendamustine plus rituximab (BR) demonstrated promising efficacy and safety in previously untreated elderly fit CLL patients (CLL10 study). There is, however, no data with the BR regimen in comorbid patients. The purpose of our pilot study was to evaluate safety and efficacy of bendamustine combined with rituximab as first-line therapy in patients with CLL and significant comorbidities. Patients and Methods: Thirtysix consecutive patients with a Cumulative Illness Rating Scale (CIRS) 6, who were treated with BR regimen in first line setting, were included in the study. Bendamustine was administered at a dose of 90 mg/m2 on Days 1 and 2 combined with rituximab 375 mg/m2 on Day 0 of the first course and 500 mg/m2 on Day 1 during subsequent courses every 28 days for up to six courses. Response evaluations were based on peripheral blood counts and physical examination findings. Results: The median age at the start of therapy was 71 years (range 57-79), 53% were male, 58% patients were at Rai stage III or IV, 78% of cases showed a WHO performance status (PS) of 1(range 0-2) and median CIRS (CLL not included) was 9 (range 6-17), FISH data, available in 34/36 cases, identified a del (17p) in 6% of the patients and del (11q) in 29%. Unmutated IGHV was detected in 77 % of the patients. The median number of BR cycles was 4 (range 1-6). In 32 pts data for response assessment were available. Overall response rate (ORR) was 97% with clinical complete response (CR/CRi) in 62% (20 pts) and partial response (PR) in 34% (11 pts), progressive disease 4% (1pt). Response evaluation by CT scan was available in 12 pts: CR/PR: 5/7 pts. Grade 3 or 4 infections occurred in 22 % of patients. Grade 3 or 4 adverse events for neutropenia, thrombocytopenia, and anemia were documented in 53 %, 14 %, and 11 % of patients, respectively. G-CSF as a secondary prophylaxis of neutropenia was administrated in 53% patients. Treatment was discontinued before the sixth cycle in 13 pts (36 %) because of serious infections (n=4), hematological toxicity (n=6), injury (n=1), planned operation (n=1), pts decision (n=1).One patient died due to sepsis after second cycle of therapy. No unexpected toxicities were observed. Conclusions: Chemoimmunotherapy with BR is an effective therapeutic option with manageable toxicity for the initial treatment of CLL in elderly patients with significant comorbidities. These are preliminary results. More data from the study will be presented.

182 Treatment Updated efficacy including genetic subgroup analysis and overall safety in the phase 3 RESONATE trial of ibrutinib versus ofatumumab in previously-treated CLL/SLL John Pagel1 2 3 4 5 6 Jennifer Brown , Peter Hillmen , Susan O'Brien , Jacqueline Barrientos , Nishitha Reddy , 7 8 9 10 11 Steven Coutre , Constantine Tam , Stephen Mulligan , Ulrich Jaeger , Paul Barr , Richard 12 13 14 15 Furman , Thomas Kipps , Florence Cymbalista , Patrick Thornton , Federico Caligaris16 17 18 19 Cappio , Julio Delgado , Marco Montillo , Sven DeVos , Carol Moreno20, Talha Munir3, Jan Burger4, Devon Chung21, Jennifer Lin21, Linda Gau21, Betty Chang21, George Cole21, Emily Hsu21, Danelle James21, John Byrd22 1

Swedish Cancer Institute Dana-Farber Cancer Institute 3 The Leeds Teaching Hospitals, St. James Institute of Oncology 4 MD Anderson Cancer Center 5 North Shore Long Island Jewish Health System 6 Vanderbilt-Ingram Cancer Center 7 Stanford University School of Medicine 8 Peter MacCallum Cancer Centre and St. Vincent's Hospital 9 Royal North Shore Hospital 10 Medical University of Vienna 11 University of Rochester Cancer Center 12 Weill Cornell Medical College 13 Moores UCSD Cancer Center 14 Hôpital Avicenne 15 Beaumont Hospital 16 Universita Vita-Salute San Raffaele 17 Hospital Clinic 18 Niguarda Ca’ Granda Hospital 19 David Geffen School of Medicine at UCLA 20 Hospital de la Santa Creu Sant Pau 21 Pharmacyclics LLC, an Abbvie Company 22 The Ohio State University Medical Center


2

Background: Ibrutinib, a first-in-class, once-daily, oral covalent inhibitor of Bruton's tyrosine kinase, is indicated by EMA for treatment of adult patients with chronic lymphocytic leukemia (CLL) with 1 prior therapy, or first-line in the presence of 17p deletion or TP53 mutation in patients unsuitable for chemoimmunotherapy. This analysis reports updated efficacy including genetic features and overall safety data from the phase 3 RESONATE study of ibrutinib vs. ofatumumab. Methods: Patients were randomized to receive ibrutinib 420 mg once daily or intravenous ofatumumab for up to 24 weeks. At interim analysis, improvement in the primary endpoint of PFS assessed by independent review committee in addition to OS and ORR relative to ofatumumab prompted the data monitoring committee to recommend crossover to the ibrutinib arm. Here we report updated investigator-assessed efficacy outcomes since interim analysis inclusive of patient subgroups with baseline gene mutations and other high-risk factors. Results: Among 391 randomized patients (ibrutinib [n=195], ofatumumab [n=196]), 142 (ibrutinib) and 149 (ofatumumab) patients had samples evaluable for gene mutations and chromosomal abnormality analysis at baseline. For patients randomized to ibrutinib and ofatumumab, 32% and 33%, respectively, had del17p, 33% and 31% del11q, and 73% and 63% unmutated IGHV. At baseline, 20% and 22% had ATM


mutations, 28% and 30% NOTCH1 mutations, and 15% and 10% BIRC3 mutations. Median time on study was 19 months for each treatment arm. Investigator-assessed PFS was not reached with ibrutinib vs. 8.1 months with ofatumumab (HR=0.106, P