Andrei Braester

Rare coexistence of Rosai - Dorfman disease and nodal marginal zone lymphoma complicated by severe life-threatening autoimmune hemolytic anemia

Rosai – Dorfman disease (RDD), otherwise known as sinus histiocytosis with massive lymphadenopathy (SHML), is a rare macrophage-related histiocytic disorder of unknown etiology, dominated by histiocytic and lymphocytic infi ltration of enlarged lymph nodes [1]. Patients with RDD may rarely also have an associated malignant lymphoma, possibly involving diff erent anatomical sites and developing at diff erent time intervals. Until now only 22 cases of RDD in association with lymphomas have been reported in the world literature [2 – 18]. Th e concomitant occurrence of RDD and nodal marginal zone lymphoma is exceptionally rare, and we were able to fi nd only one documented case of this kind [17]. Here we report another rare case of nodal marginal zone lymphoma with coexistent RDD, which also was accompanied by life-threatening autoimmune hemolytic anemia. A previously healthy 50-year-old man was admitted to the intensive care unit with severe weakness and headache, palpitations, chest pain, nausea and vomiting. Th e symptoms began 5 days before admission and progressed rapidly. On admission to hospital, the patient was pale with sinus tachycardia and moderate splenomegaly, without palpable peripheral lymphadenopathy. Th e blood tests revealed very severe anemia with hemoglobin (Hb) 2.8 g/dL, mean corpuscular volume (MCV) 105 fL, reticulocytosis 16%, mild leukocytosis, with a white blood cell count (WBC) of 14 500, with normal diff erential count, neutrophils 57%, lymphocytes 23%, normal absolute lymphocyte count 3500 and normal platelet count 193 000. Th e blood chemistry revealed high levels of lactate dehydrogenase (LDH) 703 U/L (normal 125 – 243 U/L), with total bilirubin 2.6 mg/dL (normal 0.2 – 1.2) and indirect bilirubin 1.5 mg/dL (normal up to 0.7 mg/dL). Th e direct antiglobulin test was strongly positive (recorded as DAT ) for immunoglobulin G (IgG) and C3. Flow cytometry analysis from peripheral blood detected a monoclonal population of B lymphocytes (56%), positive for CD20, kappa light chain and FMC7. CD23 and CD38 were mildly positive (10% and 11%). Cells were negative for CD5, CD43, CD79b, CD11c, CD10, CD103 and CD25. Th e phenotype was compatible with a CD5 x03 B cell lymphoproliferative disease (LPD), other than CLL. Th e bone marrow biopsy revealed a markedly hypercellular bone marrow showing both a nodular and interstitial pattern of infi ltration by small intermediate size lymphocytes, which occupied up to 30% of the bone marrow cellular compartment. Th e immunohistochemical study revealed these cells to be strongly positive for CD20 and leukocyte common antigen (LCA). CD79a and CD10 were positive only in a few cells; CD23, CD43, BCL-1, BCL-6, DBA-44 and tartrate resistant acid phosphatase (TRAP) were all negative. Th e fi ndings were compatible with involvement by B-cell LPD, probably marginal zone lymphoma (MZL). Molecular analysis by polymerase chain reaction (PCR) from the bone marrow showed positive immunoglobulin heavy chain (IgH) rearrangement. A total body computed tomography (CT) scan showed abdominal lymphadenopathy up to 4 x04 3 cm (celiac, hepatic hilus and retroperitoneal nodes) and moderate spleno megaly (16.7 8 18.4 cm), with a negative fl uorodeoxyglucose positron emission tomography (FDG PET) scan, showing no pathological uptake. A lymph node biopsy was performed in order to establish a more accurate diagnosis of the LPD and the fi ndings described below were compatible with marginal zone lymphoma and, unexpectedly, also concomitant Rosai – Dorfman disease. Th e architecture of the lymph node was completely altered. Th ere was marked dilatation of medullary and cortical sinuses (Figure 1), containing large irregular mononuclear histiocytes with abundant pale vacuolated eosinophilic cytoplasm. Some of the histiocytes clearly had lymphocytes in their cytoplasm (emperipolesis) (Figure 2). L eu k L ym ph om a D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y M ic hi ga n U ni ve rs ity o n 10 /3 0/ 14

Bortezomib-induced peripheral neuropathy is related to altered levels of brain-derived neurotrophic factor in the peripheral blood of patients with multiple myeloma

Loss of nerve growth factor-mediated neuronal survival has recently been proposed as a candidate mechanism underlying bortezomib-induced peripheral neuropathy (BIPN) (Broyl et al, 2010). However the literature does not reveal any data from patients that can support this hypothesis. Brain-derived neurotrophic factor (BDNF) is a neuronal growth factor that is crucial for neuronal survival and repair (Hofer & Barde, 1988). We report on alterations in BDNF peripheral blood levels and the development of BIPN in patients with multiple myeloma (MM). Following approval of the Ethics Committee and obtaining patient’s informed consent, peripheral neuropathy was assessed and graded in 25 MM patients using the abbreviated version of the Total Neuropathy Score (TNS) tool (Argyriou et al, 2008). These patients were examined at diagnosis and after receiving four courses of a bortezomib-based regimen given intravenously either weekly or biweekly (Table I). Patients that developed TNS ≥ 2 during treatment were determined to have BIPN and three patients with TNS ≥ 2 at diagnosis were excluded from the study. At the time of neurological examination, a venous blood sample was obtained and soluble BDNF (sBDNF) level was determined by a commercial enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN, USA) in platelet-poor-plasma. One of our patients passed away after two courses of bortezomib. Eight of the remaining patients (36%) developed BIPN. We found lower levels of sBDNF in the plasma of patients that developed BIPN as compared to patients that did not develop BIPN [BDNF level standard deviation (SD) in BIPN 2 164 0 721 vs. 4 62 0 61 ng/ml in non-BIPN; P = 0 007; Fig 1A]. By analysing the difference between BDNF levels at diagnosis and after four courses of treatment of each patient, we demonstrated that sBDNF level was reduced in patients developing BIPN while it remained stable or even minimally elevated in patients that did not develop BIPN (difference in BDNF SD in BIPN 1 668 0 670 vs. 0 405 0 708 ng/ml in non-BIPN P = 0 02) (Fig 1B). Platelets are thought to be responsible for the homeostasis of BDNF in the blood as they take up, store and secrete BDNF (Fujimura et al, 2002). As we did not observe differences in the platelet counts of patients that developed BIPN compared to patients that did not develop BIPN (data not shown), we reasoned that decreased sBDNF levels in the plasma of patients with BIPN may result from lack of secretion of BDNF from the platelets. To test this hypothesis, we determined the content of BDNF in the platelets of patients that developed BIPN relative to those who did not develop BIPN by two different methods. First, platelets that were collected and stored in 20°C at the time of the patients’ examination were lysed and the BDNF immunoreactive band was determined by gel electrophoresis using specific rabbit polyclonal anti-human BDNF mAb (Abcam, Cambridge, MA, USA). Second, BDNF and the platelet activation marker CD62p were determined in freshly isolated platelets of patients with BIPN and patients without BIPN using flow cytometry. Briefly, 20 ll of platelet-rich-plasma from each patient was stained with phycoerythrin-cyanin 5 (PE-CY5) Anti human CD41a monclonal antobody (mAb) (Beckmann Coulter, Brea, CA, USA) and PE anti human CD62p mAb (Biolegends, San Diego, CA, USA) or fixed permeabilized (Invitrogen, Frederick, MD, USA) and stained with ATTO-488 Anti human proBDNF pAb (Alomone Labs Ltd., Jerusalem, Israel). The intensity of surface CD62p or intracellular BDNF signal was determined in 10 000 platelets. Analysis of BDNF in platelets by gel electrophoresis suggested that BDNF content was increased in platelets of patients who developed BIPN as compared to platelets of patients that did not (mean BDNF expression level post-treatment relative to pre-treatment SD in BIPN 4 33 1 15-fold vs. 1 67 0 57-fold in non-BIPN P = 0 02) (Fig 1C). Flow cytometric analysis confirmed the increase of BDNF content in platelets of patients with BIPN, as detected by higher intensity of proBDNF compared to platelets of patients without BIPN [mean fluorescence intensity (MFI) of BDNF SD in BIPN 43 12 6 04 vs. 18 267 10 162 ng/ml in non-BIPN P = 0 009]. As expected, the increased BDNF signal in platelets of patients with BIPN coincided with lower

Pathogenesis, prevalence, and prognostic significance of cytopenias in chronic lymphocytic leukemia (CLL): a retrospective comparative study of 213 patients from a national CLL database of 1,518 cases

Utilizing the database of the Israeli CLL Study Group, we investigated the prevalence and prognostic significance of anemia and thrombocytopenia in patients with chronic lymphocytic leukemia (CLL). Of 1,477 patients, 113 had anemia and thrombocytopenia associated with “infiltrative” marrow failure, median survival of 41 and 86xa0months, respectively. Autoimmune cytopenias were diagnosed in 100 patients, autoimmune hemolytic anemia (AIHA) in 80, and immune thrombocytopenia (ITP) in 31, while 11 had both co-existent. Median survival of patients with AIHA and ITP, from CLL diagnosis, was 96 and 137xa0months, respectively, but 29 and 75xa0months from onset of cytopenia. Patients with AIHA from the time of CLL diagnosis had a significantly shorter survival than those without anemia (pu2009<u2009.0001). Survival was similar for patients with AIHA or anemia due to “infiltrative” bone marrow failure (pu2009=u2009.44). The presence of positive antiglobulin test even without hemolysis was associated with worse outcome. Overall survival of patients with ITP and those without cytopenias (pu2009=u20090.94) were similar. In conclusion, laboratory or clinical evidence of AIHA has a significant negative impact on the survival of patients with CLL. Outcome for cases with ITP and patients without cytopenias was similar.

Venous thromboembolism in patients with glioblastoma multiforme: Findings of the RIETE registry

BACKGROUNDnThere is uncertainty about the optimal therapy of venous thromboembolism (VTE) in patients with glioblastoma multiforme (GBM).nnnMETHODSnWe used the RIETE (Registro Informatizado Enfermedad TromboEmbólica) database to compare the rate of VTE recurrences and major bleeding during the course of anticoagulation in patients with GBM, other cancers and in patients without cancer.nnnRESULTSnAs of September 2014, 53,546 patients have been recruited in RIETE. Of these, 72 (0.13%) had GBM and 11,811 (22%) had other cancers. Most patients in all 3 subgroups received initial therapy with low-molecular-weight heparin (LMWH), but those with GBM received slightly lower doses than those with other cancers or without cancer. Then, most patients with GBM continued on LMWH for long-term therapy, at similar doses than those in the other subgroups. During the course of anticoagulation (mean, 202 days), 3 patients with GBM presented VTE recurrences (10.9 per 100 patient-years; 95% CI: 2.76-29.5) and 4 suffered major bleeding (one intracranial) (14.5 bleeds per 100 patient-years; 95%CI: 4.60-34.9). Compared with patients with other cancers, those with GBM had a similar rate of VTE recurrences and major bleeds, but had a higher rate of extracranial hematoma (p<0.05). Compared with VTE patients without cancer, those with GBM had a higher rate of PE recurrences (p<0.01) and major bleeding (p<0.001), particularly extracranial hematoma (p<0.001).nnnCONCLUSIONSnPatients with GBM and VTE had a similar rate of VTE recurrences or major bleeds during the course of anticoagulant therapy than those with other cancers.

Anemia and bleeding in patients receiving anticoagulant therapy for venous thromboembolism

In patients receiving anticoagulant therapy for venous thromboembolism (VTE), the important issue of anemia influence on the risk of bleeding has not been consistently studied. We used the large registry data RIETE (Registro Informatizado Enfermedad Tromboembólica) to compare the rate of major bleeding in patients receiving anticoagulant therapy for VTE according to the presence or absence of anemia at baseline. Patients with or without cancer were separately studied. Until August 2016, 63492 patients had been enrolled. Of these, 21652 (34%) had anemia and 14312 (23%) had cancer. Anemia was found in 57% of the patients with cancer and in 28% without (odds ratio 3.46; 95% CI 3.33–3.60). During the course of anticoagulant therapy, 680 patients with cancer had a major bleeding event (gastrointestinal tract 43%, intracranial 14%, hematoma 12%). Cancer patients with anemia had a higher rate of major bleeding (rate ratio [RR]: 2.52; 95% CI 2.14–2.97) and fatal bleeding (RR 2.73; 95% CI 1.95–3.86) than those without anemia. During the course of anticoagulation, 1133 patients without cancer had major bleeding (gastrointestinal tract 32%, hematoma 24%, intracranial 21%). Patients with anemia had a higher rate of major bleeding (RR 2.84; 95% CI 2.52–2.39) and fatal bleeding (RR 2.76; 95% CI 2.07–3.67) than those without. On a multivariable analysis, anemia independently predicted the risk for major bleeding in patients with and without cancer (hazard ratios: 1.66; 95% CI 1.40–1.96 and 1.95; 95% CI 1.72–2.20, respectively). During anticoagulation for VTE, both cancer- and non-cancer anemic patients had a higher risk for major bleeding than those without anemia. In anemic patients (with or without cancer), the rate of major bleeding during the course of anticoagulant therapy exceeded the rate of VTE recurrences. In patients without anemia the rate of major bleeding was lower than the rate of VTE recurrences.

Physicians’ lack of knowledge - a possible reason for red blood cell transfusion overuse?

BackgroundA significant percentage of red blood cell transfusions are inappropriately overused. This study investigated physicians from the western Galilee in terms of their knowledge of transfusion medicine as a potential reason for red blood cell overuse, and assessed the influence of personal background characteristics on their knowledge.MethodsData were collected via anonymous questionnaires. The questionnaires included a personal background section and a professional section. Study participants were grouped according to field of specialty, seniority, and location of medical school graduation, in order to correlate participant characteristics with knowledge.ResultsScores were calculated on a 0–100 scale. The overall knowledge of the study population was low (mean score 47.8xa0±xa018.6). Knowledge regarding basic physiology of red blood cell transfusion was also low. Internal medicine physicians and senior physicians had significantly greater overall knowledge scores and were more familiar with a restrictive blood management policy than were surgeons and residents, respectively. Comparing knowledge scores, no difference was found regarding indications for transfusion.ConclusionGeneral and fundamental knowledge in transfusion medicine is lacking among physicians in the non-operating room setting, which may play a role in red blood cell transfusion overuse. Field of specialty and professional status influenced knowledge of transfusion medicine. Educational programs and increased physicians’ awareness might help decrease unnecessary transfusions.Trial registrationNot applicable.

Absolute lymphocyte count as a prognostic marker in newly diagnosed multiple myeloma patients.

Sir, Multiple myeloma (MM), a malignant disorder of plasma cells, is the second most common hematologic malignancy. Different prognostic strategies are used to stratify patients at the time of the diagnosis. The survival of MM patients has improved significantly in the last decade. This improvement can be attributed especially to the use of novel agents in the therapy of MM patients [1]. These include especially two classes of drugs: the immunomodulatory drugs (thalidomide, lenalidomide, and pomalidomide) and proteosome inhibitors (bortezomib, carfilzomib). Several studies suggest a prognostic significance for the absolute lymphocyte count (ALC), as a surrogate marker of the immune status [2], at diagnosis and after autologous stem cell transplantation (ASCT) [3] in patients with myeloma and that the quantitative numbers of CD19 cells at MM diagnosis are associated with superior survival [4]. Myeloma cells survival is dependent on microenvironmental interactions in their niche. Recently, the role of tumor-associated macrophages (TAM) was studied also in the pathogenesis of multiple myeloma. It was shown that macrophages support myeloma cell growth, viability, and drug resistance [5]. Increased peripheral blood absolute monocyte count (AMC) has been reported to be a poor prognostic factor in malignant lymphoproliferative disorders and also in solid tumors [6, 7]. The relationship between the number of circulating monocytes and clinical outcomes was mainly attributed to their role as precursors of TAM. A prognostic significance was also reported for the LMR (lymphocyte/monocyte ratio) in newly diagnosed myeloma patients [8] and also at the start of the conditioning regimen of autologous stem cell transplantation (ASCT) [9]. Until now, to our best knowledge, the statistical value of these measures has not been verified in patients treated with novel agents. We retrospectively evaluated these simple measurable markers, the AMC, the ALC, and the LMR at diagnosis of multiple myeloma patients in our hospital and evaluated their prognostic value in patients treated with novel agents. This study included new multiple myeloma patients who were diagnosed between 2006 and 2014 at Galilee Medical Center, Nahariya, Israel. The diagnosis was made according to the IMWG guidelines for symptomatic MM. Approval for review of the records was obtained from the Institutional Review Board of the Galilee Medical Center. AMC and ALC were obtained at diagnosis from routine complete blood count calculated by Advia 2120 hematology analyzer (Simens). We also collected demographic and clinical characteristics of the patients from their records. Quantity data were described by median and range. Qualitative data were described by frequencies and percentages. Comparisons of quantity data between groups were examined by independent sample t-test or Wilcoxon ranks sum test, as appropriate. Correlation of qualitative variables was examined by chi-squared test or Fisher’s exact test, as appropriate. Overall survival (OS) was measured from the time of diagnosis to the date of death or last follow-up. Univariate and multivariate survival analysis was performed using Cox Regression Model and Kaplan–Meier method with Log Rank (Mantel-Cox) test. All P-values represented were one-sided, and a pvalue<0.05 was considered statistically significant. Sixty-two newly diagnosed MM patients were evaluated. The median age was 69 years (range: 46–88 years); 35 (56.5%) patients were male and 27 (43.5%) female. The paraprotein type was as follows: 33 (55%) IgG, 14 (23.3%) light chain and 13 (21.7%) IgA; 15 (25%) patients were ISS I, 16 (26.7%) were ISS II, 29 (48.3%) were ISS III. Forty (64.5%) patients were treated with chemotherapy, 16 (25.8%) were treated with chemotherapy followed by ASCT, and 6 (9.7%) patients had conservative therapy. Fifty-one patients (82%) were treated with novel agents at diagnosis. The median follow-up for the whole cohort after the diagnosis was 28.2 mo (range 0.5–98.8 months). At the time of writing, 41 (66.1%) of patients were living. The optimal cutoff for the ALC and AMC, and the LMR was set at 1600/lL,

Azacitidine-lenalidomide (ViLen) combination yields a high response rate in higher risk myelodysplastic syndromes (MDS)—ViLen-01 protocol

Azacitidine treatment is effective in higher risk MDS (HR-MDS), with less than 50xa0% response, lasting 2xa0years. Aza and lenalidomide (Len) have a potential synergistic effect. ViLen-01 phase IIa trial includes 6-month induction (Aza 75xa0mg/m2/day, days 1–5, Len 10xa0mg/day, days 6–21, every 28xa0days), 6-month consolidation (Aza 75xa0mg/m2/day, days 1–5, every 28xa0days), and 12-month maintenance (Len 10xa0mg/day, days 1–21, every 28xa0days). Response was evaluated according to IWG criteria. Totally, 25 patients enrolled, with an average of 76.3xa0years old (60–87), and 88xa0% with major comorbidities. Thirteen patients completed induction, 7xa0proceeded for consolidation, and 2 for maintenance. The overall response rate (ORR) was 72xa0% (18/25), with 6 (24xa0%) for CR, 3 (12xa0%) for marrow CR, and 9 (36xa0%) for hematologic improvement (HI). The 7 non-responding patients were on the study 3xa0days to 4.1xa0months. At 6xa0months, 4 of 6 evaluable patients achieved complete cytogenetic response and 2 with del (5q) at diagnosis. Adverse events (AEs) were as expected in these patients: grades III–IV, mainly hematologic—thrombocytopenia (20 patients) and neutropenia (13 patients). The common non-hematologic AEs were infections (14 patients), nausea (7), vomiting (7), diarrhea (7), and skin reactions (5). The median progression-free survival (PFS) was 12u2009±u20091.36xa0months, with median overall survival (OS) of 12u2009±u20091.7xa0months. Quality of life (FACT questionnaire) data were available for 12 patients with a tendency towards improved QoL. This trial with elderly HR-MDS patients with an expected poor prognosis demonstrates a high (72xa0%) response rate and a reasonable expected safety profile but a relatively short PFS and OS.

Association between Met-BDNF allele and vulnerability to paclitaxel-induced peripheral neuropathy

Paclitaxel interference with microtubule axonal transport in normal neuronal cells induces peripheral neuropathy that limits its effectiveness [1]. Themechanism of enhanced vulnerability to chemotherapy-induced peripheral neuropathy (CIPN) is still unknown. Our previous report that showed dysregulated blood levels of brain-derived neurotrophic factor (BDNF) in patient developingCIPN[2] raised theneed for investigating apossible link between genetic alterations in BDNF and vulnerability to CIPN. The human BDNF gene carries a frequent non-conservative single nucleotide polymorphism (Val66Met). The MetBDNF allele was reported to cause a deficit in the cellular distribution and regulated secretion of neuronal BDNF [3].We, therefore, wanted to check whether Met-BDNF allele is associated with vulnerability to CIPN. Peripheral neuropathy was assessed and graded in 21 women, 20 with breast cancer, and 1 with ovarian cancer, using the reduced version of total neuropathy score (TNSr) and the neurotoxicity subscale Fact/GOG-NTx. Patients with TNSr C2 were diagnosed as patients with peripheral neuropathy. Allelic discrimination of BDNF polymorphism was done in the patient’s peripheral blood by DNA gene sequencing. Seven patients (33.33 %) were diagnosed with TNSr score C2 before initiation of paclitaxel treatment and determined to have pre-existing peripheral neuropathy. The remaining 14 patients (66.66 %) were diagnosed with TNSr score B1 before initiation of paclitaxel treatment and identified as non-pre-existing peripheral neuropathy. As regards BDNF genotype, we found Val/Val in 13 patients (62 %), Val/Met in eight patients (38 %), and Met/Met in none of the patients (0 %). We found higher incidence of the Met-BDNF allele in the pre-existing peripheral neuropathy group than in the group of patients defined as nonpre-existing peripheral neuropathy (5/7 (71.4 %) versus 3/14 (21.4 %) respectively, prob[V 0.05). This observation was further confirmed by showing that at baseline, patients carrying the Met-BDNF allele had higher mean TNSr and Fact-GOG scores than the Val/Val patients (Mean TNSr ± SEM in Val/Val 0.46 ± 0.21 vs. 3.5 ± 1.78 in Val/Met, prob[ t 0.02. Mean FactGOG ± SEM in Val/Val 3.23 ± 1.05 vs. 8.62 ± 1.05 in Val/Met, prob[ t 0.02) (Fig. 1a). Furthermore, patients carrying the Met-BDNF allele reached higher maximal TNSr and Fact-GOG scores in response to paclitaxel compared to Val/Val patients (Mean TNSr ± SEM 4.69 ± 0.74 vs. 8.87 ± 2 respectively, prob[ t 0.02. Mean Fact-GOG ± SEM 8.69 ± 1.96 vs. 20.37 ± 3.24 respectively, prob[ t 0.002) (Fig. 1b). Pre-existing neuropathy is suggested to be the strongest clinical risk factor for CIPN development [1]. However, the mechanism of enhanced vulnerability to neurotoxic agents, such as paclitaxel, is still unknown. BDNF is a neuronal growth factor that is crucial for the survival and repair of injured neuronal cells. The genetic data presented hereby suggest that dysfunctional neuronal BDNF-induced repair mechanism due to BDNF-SNP may be a predisposing condition that exacerbates paclitaxel-induced neurotoxicity. We Andrei Braester and Hadassah Goldberg are equally contributed for this article.

Elevated serum BDNF levels are associated with favorable outcome in CLL patients: Possible link to CXCR4 downregulation

Increased chemokine C-X-C receptor 4 (CXCR4) expression is related to unfavorable outcome in chronic lymphocytic leukemia (CLL). Brain-derived neurotrophic factor (BDNF) is a neuronal growth factor that has been shown previously to interact with CXCR4 in neuronal cells. Here, we studied the in vitro effect of BDNF on CXCR4 expression and chemotaxis toward stromal derived factor-1 (SDF-1) in freshly isolated CLL cells. We also explored the correlations between serum BDNF levels in CLL patients and disease characteristics and clinical course. Incubation of CLL cells with recombinant BDNF (50 ng/mL) resulted in a downregulation of CXCR4 surface expression and atenuated chemotaxis toward SDF-1. Higher serum BDNF levels were associated with a mutated immunoglobulin heavy chain variable (IGHV) gene, an early clinical stage, and a stable clinical course. Our findings suggest that increased circulating blood BDNF may be associated with a favorable effect in CLL. However, the exact mechanism of this favorable effect should be investigated further.