ALBERT

Description: Abstract 2663 Background: Given the recent identification of the mTOR pathway as a possible target for cancer therapy, the expression pattern of phosphorylated mTOR (p-mTOR) pathway components has attracted vivid scientific interest in several neoplasms, including primary cutaneous T cell lymphomas. However, published information examining crosstalk between AKT/mTOR cascade and interconnected signaling pathways in mycosis fungoides (MF) is limited. Materials: We analysed immunohistochemically 54 skin biopsies (21 tumour and 33 plaque stage) from 50 patients with MF, for whom clinical information was available. Paraffin embedded tissue was stained for p-mTOR, its upstream p-AKT, its downstream effectors p-p70S6K and p-4E-BP1, as well as for p-ERK1/2 (extracellular signal-related kinase 1/2), the transcription factor Notch-1 and a member of the mammalian family of signal transducer activator of transcription, namely p-STAT-3. For statistical analysis only the predominant pattern i.e. cytoplasmic for p-mTOR, p-AKT, and NOTCH1 and nuclear for p-STAT3, p-p70S6K, p-4E-BP1 and p-ERK1/2 was taken into account. A Histo-score (H-score) based on the percentage of stained neoplastic cells (labelling index-LI) multiplied by staining intensity for each protein was calculated and was correlated with clinicopathological features and cancer-free (CSS), disease-free (DFS) and progression-free (PFS) survival. Results: p-p70S6K expression was recorded in all cases, whereas the rate of positivity for the remaining molecules ranged from 52.8% to 67.3%. p-mTOR was coexpressed with p-p70S6K in 67.3% of cases, but coexpression with other molecules was less common (27.4–38.5%). Notch-1 displayed higher H-scores in tumours than in plaques (p=0.0475), as previously demonstrated. The same applied to p-p70S6K, but this correlation was of borderline significance. Significant correlations were recorded between p-4E-BP1 and p-p70S6K (p=0.0019), p-ERK and p4E-BP1 (p=0.0588), as well as between Notch-1 and p-p70S6K (p=0.0001) and p-4E-BP1 (p=0.0003). The latter two correlations remained when plaques and tumours were analysed separately. Interestingly, p-STAT3 showed a weak positive correlation with p-AKT in the entire cohort, but when analysis was restricted to plaques this relationship became statistically significant (p=0.0419). Notch1, p-4E-BP1 and p-p70S6K expression were associated with advanced stage. In univariate survival analysis increased p-p70S6K (p=0.0174) and p-AKT (p=0.0198) H-scores as well as p-4E-BP-1 immunopositivity adversely affected CSS but not DFS in plaques, along with CD30 expression (p=0.0004). In particular, blood clonality affected both DFS and CSS, as well as time to disease progression (p=0.0009, p=0.0031 and p=0.0216 respectively). The CSS time was also substantially shortened for subjects with simultaneous overexpression of p-AKT and p-p70S6K along with p-4E-BP1 immunopositivity (p= 0.0001). On the contrary in the subgroup of tumours decreased p-STAT3 H-score was only parameter that adversely affected CSS in tumour stage (p=0.0394), with the presence of blood clonality attaining marginal significance in this regard. Conclusions: Activation of AKT-mTOR components appears to have an intimate liaison with NOTCH1, p-ERK and p-STAT-3 and seems to be implicated in disease progression and in the acquisition of a more aggressive phenotype. Phosphorylation of several key components of this pathway, namely AKT, p70S6K and 4E-BP1 in MF, single or in combinations emerge as potential markers of prognostic value in plaque stage, whereas in tumours p-STAT-3 is brought forward as a favourable prognostic marker. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 2927 Background and Aims: Multiple Myeloma (MM) is characterized by bone marrow (BM) plasma cell infiltration and the presence of serum/urine monoclonal immunoglobulin (Ig). The depth of response has been associated with longer PFS in MM causing subsequent prolonged survival. Recently novel M-based biomarker immunoassays have been developed (Freelite™, Hevylite™) and their significance in MM diagnosis and prognosis has been demonstrated.1,2 Furthermore serum Free Light Chains (sFLC) are used for better assessment of treatment response, thus patients are considered to achieve stringed Complete Response (sCR) by having CR criteria plus normal serum Free Light Chains Ratio (sFLCR) and absent clonal cells on BM.3 The significance of Hevylite™ on response has not been assessed so far. Patients in nCR or better do not automatically restore their ratio of intact monoclonal Ig/intact polyclonal Ig of the same class (Hevylite™ or HLCR). We therefore investigated the importance of sFLCR and HLCR normalisation at plateau on PFS, in a series of patients with intact Ig MM. Patients and Methods: 50 intact immunoglobulin MM patients were studied from diagnosis to last follow up. Immunofixation was IgG (26 -kappa and 12 –lamdba) and IgA (6 –kappa and 6 -lambda). All patients were symptomatic at diagnosis. Sera samples (n=312) were analyzed for sFLC-kappa and sFLC-lambda with Freelite™ and sFLCR were calculated, and for IgGkappa, IgGlambda IgAkappa, IgAlambda with Hevylite™ and ratios IgGkappa/IgGlambda, IgGlambda/IgGkappa, IgAkappa,/IgAlambda and IgAlambda/IgAkappa (HLCRs) were calculated. sFLCRs and HLCRs values above the 95%-ile of normal individuals were considered abnormal. Statistical analysis was performed using SPSS ver 15.0. File data were reviewed. Results: At diagnosis sFLCR was abnormal in 86% of patients while HLCR was abnormal in all. All treatment lines were initiated according to standard criteria and median lines of therapy were 2 (range 1–11). Median follow up was 33 months (7–145). During patients' cumulative follow-up, 145 lines of therapy were studied and the subsequent responses were estimated. Thirty eight percent of responses were sCR, CR and nCR, 20% PR, 18% MR and 24% refractory and progressive disease. HLCR normalized in 44% of patients with sCR, CR and nCR. The depth of response correlated to PFS and patients in sCR, CR and nCR had longer PFS than the others (p

Description: The negative prognostic consequences of extramedullary plasmacytomas in the context of symptomatic MM is well known. On the contrary, the clinical impact of bone plasmacytomas (BP) has not been extensively studied, although their presence is one of Duries’ major MM diagnostic criteria. The aim of the present study was to study eventual differences between MM patients presenting at diagnosis with or without BP. Patients and Methods Two hundred and twelve symptomatic MM patients were studied from diagnosis to last follow-up. Their median age was 66 years and 60% were men. Twenty-three percent, 27% and 50% of patients were in ISS stage 1, 2 and 3 respectively; MM type was IgG in 59% , IgA in 26%, light chain only in 12%, IgD in 1,5% and non- or oligo-secretory in 1,5%. All patients required treatment and BP ones were additionally given radiotherapy. The median follow-up time of the whole cohort was 46,4 months. Results Forty patients (19%) had a BP at the time of their diagnosis. BP was located in the spine (BP-S) in 19 out of 40 and in other sites (BP-O: skull, pelvis, long bones, and ribs) in 21 patients. BP patients had equivalent demographics, stage and MM type as the others. Non-BP patients tended to have a better survival than BP ones but the difference was not statistically significant (p=0.1). However, BP-S patients had a significantly worse outcome than BP-O patients (24±6 versus 48±11 months, p

Description: Abstract 3862 Subgroups of CLL cases defined on the basis of the immunogenetic features of their B cell receptors (BcRs), especially those cases belonging to subsets with stereotyped BcRs, exhibit differential responses to immune stimulation through either the BcR or the Toll-Like Receptors (TLRs). This indicates distinct, subset-biased modalities of BcR collaboration with specific TLRs that may extend to the control of cell proliferation or apoptosis, B cell anergy and/or TLR tolerance. In both normal and CLL B cells, the final signaling outcome can be controlled by microRNAs (miRNAs) targeting critical signaling molecules downstream of the BcR and/or the TLRs. With this in mind, here we investigated the impact of immune signaling on miRNA expression and function. We used two paradigmatic stereotyped subsets with distinct molecular and clinicobiological figures: (1) subset #1: unmutated (U) IGHV1/5/7-IGKV1(D)-39 IgM BcRs, aggressive disease, the largest U-CLL subset; and, (2) subset #4: mutated (M) IGHV4–34/IGKV2–30 IgG BcRs, indolent disease, the largest M-CLL subset. We have recently shown that these two subsets show differential functional outcomes after TLR stimulation with subset #4 selectively responsive to TLR1/2 but not to TLR7 stimulation, thus contrasting subset #1 which shows the opposite profile. Based on the above, we cultured negatively selected CD19+ B cells from subset #1 and subset #4 cases under the following conditions: (1) non stimulated; (2) stimulated through the BcR with anti-IgM and anti-IgG, respectively; (3) stimulated through TLR7 (Imiquimod) for subset #1, or through TLR1/2 (Pam3CSK4) for subset #4; and, (4) BcR/TLR co-stimulated. Using RQ-PCR Arrays, we measured the expression levels of a series of 33 miRNAs relevant for CLL biology with potential targets in BCR and/or TLR signaling pathways. In subset #1, none of the TLR ligands affected miRNA expression. On the contrary, TLR1/2 stimulation in subset #4 significantly (p

Description: Abstract 4976 Background and Aims: TGF-beta1 normally down-regulates B-cell proliferation. Resistance to its effects is reported in malignant cells of B-lineage such as chronic lymphocytic leukemia (CLL) and multiple myeloma (MM) cells. There are however very few studies on serum TGF-beta1 levels in patients with B-cell lymphoproliferative diseases and their eventual correlations with parameters of disease activity and survival. The purpose of our study was to assess any eventual relationship between serum TGF-beta1 levels and disease parameters, as well as time to first treatment (TFT) and overall survival (OS) in a series of MM, Waldenstrom's macroglobulinemia (WM) and CLL patients at diagnosis. Patients and Methods: 92 MM, 64 WM and 110 CLL patients were studied. In the MM group, paraprotein was IgG in 64, IgA in 18, LC in 8 and IgD in 2 patients while 32, 25 and 43% of patients were in ISS stage 1, 2 and 3 respectively. MM patients' median follow-up was 51 months. All patients required treatment a some point, 55% immediately. In the WM group, 44, 27 and 29% of patients were in WM – IPSS stage 1, 2 and 3 respectively. WM patients' median follow-up was 49 months; 75% of them required treatment during follow – up. In the CLL group, 54% and 74%, 25% and 20%, 14% και 6%, were in stage 0 and A, 1 and β, 2 and C according to Rai and Binet respectively. Median follow up was 43 months; 14 patients needed immediate treatment, 30 more required treatment during follow up while the others are still just followed. Patients' frozen sera, drawn at diagnosis, were retrospectively analyzed while sera from 20 healthy individuals (HI) were tested as controls. Serum TGF-beta1 measurements were done by ELISA according to the manufacturer's instructions. Statistical analysis was performed by SPSS software, version 15. 0. Results: Median serum TGF-beta1 levels were 36330 pg/ml (range 4. 6 – 77976) in MM patients, 33965 pg/ml (range 1665–615000 pg/ml) in WM, 12207 pg/ml in CLL (range 0 – 42970) and 32902 pg/ml in HI (range 1941 – 58123). CLL patients presented significantly lower TGF-beta1 levels than those with MM and WM (p

Description: Abstract 4568 Background and Aim: Symptomatic CLL patients need treatment immediately. For these patients, molecular-genetic factors (mutated-unmutated, ZAP 70, ATM, p53) are important prognostic factors of response and survival. Nevertheless, 2/3 of newly diagnosed patients are asymptomatic and require only of follow up that can last for months or years. For these patients overall survival (OS) depends on the time to first treatment (TFT). The most frequent paraprotein produced in CLL is serum free light chain in 50% of the patients. It has recently been shown that serum free light chains (sFLC) and their sum above 60 (κ+λ above 60) are useful prognostic factors for TFT. We therefore studied the eventual prognostic implication of sFLC and the summated FLC-kappa plus FLC-lambda in a CLL patients' series. Patients and Methods: 143 CLL patients were studied of which 18 needed immediate treatment while 37 more needed treatment during their follow up. 64% and 72%, 28% and 18%, 7.5% and 10%, were in stage 0 and A, 1 and β, 2 and C according to Rai and Binet respectively. Median patients' follow up was 32 months (range 4–228). Light chain restriction was established by flow cytometry or bone marrow biopsy immunohistochemistry. Serum free light chain values were retrospectively determined by nephelometry (Freelite™, the Binding Site Birmingham, UK) in frozen sera drawn at diagnosis. Elevated sFLC values were defined using as cut-off values the 95th percentile range of healthy individuals. Statistical analysis was performed using SPSS v15.0. Hazard ratios and prognostic significance of abnormal sFLC, HLC and ratios were determined by univariate Cox regression analysis. Kaplan Meier method was used for pictorial representation of survival and time to treatment. Results: Increased sFLC were found in 45% of the patients while the summated FLC-kappa plus FLC-lambda was higher than 60 mg/dl in 14%. Increased sFLC values as well as values of FLC κ+λ〉60 were related to shorter TFT (p=0,0005 and p=0,000003 respectively). In addition, high levels of sFLC and FLC κ+λ 〉60 correlated with β2-microglobulin (r=0.2, p=0.009 και r=0.2, p=0.03 respectively), serum albumin (r=0.2, p=0.009 only for FLC κ+λ 〉 60), negatively with hemoglobin (r=-0.3, p=0.000003 και r=-0.2, p=0.0002 respectively), increased LDH (p=0.037 και 0.001 respectively), Rai stage (p=0.03 και 0.003 respectively) and Binet stage (p=0.02 only for FLC κ+λ 〉 60) and with the presence of beta-symptoms (p=0.004 only for FLC κ+λ 〉 60). Finally, increased sFLC and FLC κ+λ〉60 values correlated with shorter OS (p=0.05 and p=0.003 respectively). Conclusion: The results of our study confirmed the significance of sFLC in CLL with regard to TFT and their relationship with adverse prognostic clinical and laboratory parameters but also demonstrated for the first time their impact on OS. Disclosures: No relevant conflicts of interest to declare.

Description: Changes in the production of monoclonal intact immunoglobulin (M-Ig) or of serum free light chains (sFLC) during symptomatic MM patients’ relapse eventually reflect molecular myeloma cells changes and / or reveal clonal cell competition. We aimed to study changes in Ig production (CIg) in symptomatic MM patients and to evaluate related frequency and corresponding specific disease characteristics. Patients and Methods 232 symptomatic MM patients with available follow-up sFLC measurements, were retrospectively studied. Light chain escape (LCE) was defined as sFLC increase with stable or falling M-Ig concentrations, M-Ig escape (MCE) as decreasing sFLC with increasing M-Ig, de-differentiation (DD) as clinical relapse with normal or decreased M-Ig and sFLC and Clonal Domination (CD) as normalization of formerly increased IgG, IgA or FLC in relapsing patients presenting increase of another component. Survival was calculated from diagnosis or from CIg date to last follow-up or death, survival curves were plotted by Kaplan-Meyer Method and assessed by the log-rank test. Results There were 94 women and 128 men, median aged 66 years; 29%, 42%, 29% and 22%, 30%, 48% were in Durie-Salmon and ISS stages I, II, III and 1, 2, 3 respectively. MM type was IgG MM in 59%, IgA in 21%, light chain only in 17%, IgD in 2% and non-secretory in1%. Median survival of the whole cohort was 46,4 months. CIg was observed in 39 patients (17%), consisting in LCE in 15 patients (6%), MCE in 7 (3%), DD in 5 (2%) and CD in 10 (4%); two additional patients (1 IgD and 1 LC) transiently produced another monoclonal component that was IgG in both cases, while in stringent complete remission. In CD patients, the dominated clone was IgG in 9 out of 10 patients, while the dominating one was LC in 8 and IgA in 2. LCE and MCE were more frequent in IgG patients. The median number of treatment lines received prior CIg was 5 for LCE, 4 for MCE, 2 for DD and 1 for CD. LCE and MCE patients had all received novel agents and/or ASCT. The median time from CIg to last follow-up or death was 2,6 months (2,2-3) for LCE, 3,3 months (2,2-4,4) for MCE, 6,3 months (1,1-11,6) for DD and 31,1 months (23,6-38,6) for CD. Patients presenting LCE, MCE and DD had a considerably shorter survival after CIg compared to patients presenting CD (p=0,0002). However because CIg was usually a late event in the course of the disease the overall survival of CIg patients was 60,6 months. In conclusion, LCE, MCE and DD are late events in the course of MM, mainly observed in patients whose previous treatments included with novel drugs. They reflect a very aggressive disease behavior with shortened survival thereafter, probably due to the emergence of a new resistant clone. CD was mainly observed in patients secreting low IgG levels and FLCs, and possibly reflect IgG clone remission in biclonal patients, given that thereafter, the disease behaves as a usual multiple myeloma, secreting however the other clone. Disclosures: No relevant conflicts of interest to declare.