Ricardo Helman

Granulocyte whole exome sequencing and endothelial JAK2V617F in patients with JAK2V617F positive Budd‐Chiari Syndrome without myeloproliferative neoplasm

Budd-Chiari Syndrome (BCS) is characterized by hepatic venous outflow obstruction. JAK2-positive myeloproliferative neoplasms (JAK2 MPN) are one of the most frequent thrombotic conditions underlying a diagnosis of BCS (Smalberg et al, 2012). A subset of BCS patients harbour the JAK2 mutation in peripheral blood (PB) granulocytes without evidence of overt MPN (JAK2 MPN-neg). While the occurrence of other myeloid-associated mutations in genes are common in patients with JAK2 MPN, it is unknown if such mutations are present in patients with JAK2 MPN-neg. In addition, it remains to be determined if there are differences in JAK2 allele burden between JAK2 MPN and JAK2 MPN-neg. Recently, it has been demonstrated that a subset of patients with JAK2 MPN have endothelial cells (ECs) that are positive for JAK2(Teofili et al, 2011) and that these cells may possibly contribute to the prothrombotic state. However, the presence of JAK2-positive endothelial colony-forming cells (E-CFC) in JAK2 MPN-neg patients with BCS has not been evaluated. The objective of this study was to assess the presence of JAK2 in BM E-CFC from JAK2 MPN-neg patients with BCS and to evaluate the JAK2 allele burden and the presence of additional mutations. We selected patients diagnosed with BCS and the presence of JAK2 mutation in the absence of other thrombophilic conditions at Hospital Israelita Albert Einstein, Sao Paulo, Brazil, between March 2013 and June 2015. Patients were investigated for the presence of an associated MPN and diagnosed according to the 2001 World Health Organization criteria. This study was approved by the Institutional Review Board and conducted according to the Principles of the Declarations of Helsinki. Paired DNA (sorted CD66b-granulocytes/skin biopsy) was subjected to whole exome sequencing (WES) on the HiSeq 2000 platform (Illumina Inc., San Diego, CA) using the SureSelect library preparation kit (Agilent, Santa Clara, CA). Somatic variant calls were generated by combining the output of SomaticSniper (http://gmt.genome.wustl.edu/packages/somatic-sniper/), MuTect (https://www.broadinstitute.org/cancer/cga/mutec) and Pindel (www.sanger.ac.uk/ science/tools/pindel) plus additional in-house criteria (minimum coverage at both tumour/germline ≥8 reads; ratio of allele fraction tumour:germline >2). Bone marrow (BM) mononuclear cells were obtained by density centrifugation using Ficoll-Paque PLUS (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). Approximately 20 ml of BM was washed twice with phosphate-buffered saline and 2% fetal bovine serum and then suspended in Endocult liquid medium (STEMCELL Technologies Canada Inc., Vancouver, BC, Canada), as previously described (Wu et al, 2012), but extending the culture time by 7 days. ECs grown in culture were then evaluated by flow cytometry (ECs were considered CD146+, CD144+, CD34 , CD117 ) (Ozdogu et al, 2007). ECs were isolated by fluorescence-activated cell sorting (FACS) in a FACSAria (BD Biosciences, San Jose, CA). DNA was extracted from sorted ECs and tested for the presence of the JAK2 mutation by an allele-specific polymerase chain reaction (PCR). Data on JAK2 allele burden was extracted from WES results obtained from a cohort of 77 patients with JAK2-positive MPN (including 4 patients with JAK2 MPN-neg) (Santos et al, 2014) and from 110 patients previously reported by Nangalia et al (2013). We identified 32 cases of BCS, 10 (31 2%) of who were positive for the JAK2 mutation. Three of these patients passed away before inclusion in the study and 7 patients were included; their baseline features are summarized in Table I. Two patients were diagnosed with primary myelofibrosis (PMF) (Patients 1 and 7) during evaluation and the remaining 5 cases were JAK2 MPN-neg. Samples from Patients 1–6) were analysed by WES. In the 5 JAK2MPN-neg patients (Patients 2–6), the only known pathogenic somatic mutation found was JAK2. We found no additional myeloid-associated mutations in these patients. The median JAK2 variant allele frequency Table I. Demographic features of BCS patients and JAK2 alleleburden.

A novel mechanism of NPM1 cytoplasmic localization in acute myeloid leukemia: the recurrent gene fusion NPM1–HAUS1

NPM1 heterozygous mutations are present in roughly a third of patients with acute myeloid leukemia (AML), making it one of the most frequent genomic alterations in these patients.[1][1] The mutations are characterized by frameshift insertions in the region encoding the C-terminus of the protein,

Induction of platelet refractoriness after myeloablative unrelated allogeneic hematopoietic peripheral blood progenitor cell transplant from HLA-sensitized female donor.

enterocolitis: a meta-analysis of observational data. Pediatrics 2012;129:529-40. 3. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297-316. 4. van Imhoff DE, Dijk PH, Hulzebos CV; BARTrial study group, Netherlands Neonatal Research Network. Uniform treatment thresholds for hyperbilirubinemia in preterm infants: background and synopsis of a national guideline. Early Hum Dev 2011;87:521-5.

Identification of ANLN as ETV6 partner gene in recurrent t(7;12)(p15;p13): a possible role of deregulated ANLN expression in leukemogenesis

The ETV6 gene encodes an ETS family transcription factor that is involved in a myriad of chromosomal rearrangements found in hematological malignancies and other neoplasms. A recurrent ETV6 translocation, previously described in patients with acute myeloid leukemia (AML) (Genes Chromosomes Cancer 51:328–337,2012, Leuk Res 35:e212-214, 2011), whose partner has not been identified is t(7;12)(p15;p13). We herein report that the t(7;12)(p15;p13) fuses ETV6 to ANLN, a gene not previously implicated in the pathogenesis of hematological malignancies, and we demonstrate that this translocation leads to high expression of the fusion transcript in the myeloid and lymphoid lineages.

The role of magnetic resonance imaging-T2* in the evaluation of iron overload early in hereditary hemochromatosis. A cross-sectional study with 159 patients

cancers, where a higher degree of tumor-infiltrating lymphocytes (TIL) predicts improved survival independent of TNM (tumor, node, metastases) staging [1]. TIL have also shown prognostic value in various lymphoid hematological malignancies. In AML, evidence of immune modulation is seen in the “graft-versus-leukemia” effects post-allogeneic SCT together with promising preclinical activity of novel immune therapies. Furthermore, higher peripheral blood lymphocyte recovery after induction chemotherapy [2] and allogeneic SCT for AML [3] have predicted improved overall survival (OS). The prognostic significance of immune status at diagnosis has not been defined in AML. We therefore investigated whether degree of T lymphocyte infiltration within CN-AML diagnostic bone marrow trephines, which we have labeled the “leukemia-infiltrating lymphocyte” (LIL) index, had prognostic value. Patients diagnosed with CN-AML at an adult tertiary center between 2006 and 2013, treated with intensive chemotherapy and who had suitable stored, diagnostic bone marrow trephines were analyzed with institutional ethics approval. Trephines were formalinfixed, decalcified in EDTA, and processed through to paraffin. Immunostaining for pan-T cell marker, CD3, and cytotoxic markers, CD8 and Granzyme B (GB), was performed at the time of analysis using routine methods. CD4 immunostaining was attempted, however high levels of non-specific staining precluded quantitative analysis. Positive cells were quantified using FijiVC image analysis software (v1.48o) from photographs taken at 3200 magnification of three representative areas of each trephine, expressed as a percentage of total cells and averaged over the three images. The cohort included 53 patients (51% male, median age 51 years (range 19–78)) followed-up over a median of 25.9 months (range 0.17–102.6 months). Most patients (96%) had de novo AML, 47 patients (89%) achieved CR (29 (62%) of whom relapsed), 21 (40%) underwent allogeneic SCT (10 in second CR) and 20 (38%) died. From a mean of 13,768 cells per trephine analyzed, median CD3% was 4.64% (range 0.56–37.24%) (Fig. 1A,B), CD8% was 5.08% (range 0.43–43.5%) and GB% was 0.34% (range 0–10.11%). In univariate Cox regression analyses, significant risk factors for death were higher aspirate blast% (P5 0.006), primary refractory disease (P5 0.03), and relapse (P5 0.04). Increasing age (P5 0.093), FLT3-ITD (P5 0.099), and CD3% (P5 0.079) approached significance. Gender, preceding myelodysplastic syndrome, allograft, NPM1 mutation (NPM1), CD8%, and GB% were not significant. In Cox regression multivariate analyses, combining T cell numbers with the relevant univariate factors (i.e., P< 0.1), higher CD3 and CD8 were independent predictors of OS (CD3: hazard ratio (HR) 0.929 for death, 95% CI 0.870–0.992, P5 0.029; CD8: HR 0.920, 95% CI 0.869–0.973, P5 0.004). Patients were divided into quartiles to determine whether incremental differences in survival were present with increasing T cell number. Kaplan–Meier survival analyses demonstrated that CD3 and CD8 numbers were able to stratify patient survival, with significantly superior OS in patients with higher CD3% (Quartile 4 (>12.39%) vs. Quartile 1 (<1.63%), P5 0.042) (Fig. 1C, first panel) and a trend toward better OS with higher CD8% (Quartile 4 (>10.66%) vs. Quartile 1 (<2.2%), P5 0.057) (Fig. 1C, second panel). GB was not able to risk-stratify patients (Fig. 1C, third panel). Higher CD3% and CD8% measured by flow cytometry showed similar prognostic capability, however there was weak correlation between CD3 and CD8 numbers obtained by immunohistochemistry and flow cytometry (data not shown; correlation coefficient, r5 0.534 for CD3 and r5 0.613 for CD8). This may reflect inherent differences in the type of samples analyzed, particularly with sampling artifact and varying degrees of hemodilution in aspirates. Further survival analyses incorporated known prognostic significance of FLT3-ITD and NPM1. Forty-four patients (83%) who were tested for both mutations formed three subgroups: FLT3-ITD (n5 18; 3 year OS 37%), NPM1 without FLT3-ITD (NPM1/ FLT3-ITD; n5 9; 3 year OS 70%), and “double-negative” (NPM1/FLT3-ITD; n5 17; 3 year OS 62%). As small numbers within each subgroup precluded analysis with quartiles, groups split by the median were compared. Although there were no statistically significant differences in OS for CD3, CD8, and GB within the FLT3-ITD (Fig. 1D) and NPM1/ FLT3-ITD (data not shown) subgroups, there was a trend for better survival in the FLT3-ITD subgroup with higher T cells. This was in spite of older age in high (median age 63 years) compared with low T cell (median age 53 years) subgroups. In contrast, CD3>median (16.06%) and CD8>median (13.8%) in the NPM1/FLT3-ITD subgroup were associated with significantly superior OS (CD3: P< 0.001; CD8: P5 0.010) (Fig. 1E, first and second panels) with comparable patient characteristics between above and below median groups. GB>median (1.38%) was not related to survival (P5 0.247) (Fig. 1E, third panel). As the NPM1/FLT3-ITD group often lacks defining prognostic factors, T cell numbers may be most relevant for further risk-stratification of this group. The association of low T cell numbers with poor prognosis in our analyses may reflect the proliferative advantage of blasts gained by immune evasion mechanisms described in AML, including suppression of production and function of T lymphocytes [4]. The corollary is that retained specific T cell immune responses against leukemia-associated antigens, including NPM1, may contribute to improved prognosis [5]. In addition to prognostic implications, the baseline immune status may modify response to immune therapies. Patients with lower T cells generally had a dismal prognosis with cytotoxic therapy and may benefit from immune augmentation by immune-based therapies. Conversely, low T cells may limit response to immune therapies. In in vitro studies of chimeric antigen receptor T cells [6] and bispecific T cell engaging antibodies [7] in AML, less cytolysis was seen with lower T cells to blasts (i.e., effector to target (E:T)) ratios in a dose-dependent relationship. Whether this consideration applies in vivo requires further investigation. To the best of our knowledge, this study is the first to correlate higher baseline T cell infiltration in bone marrow trephines in CN-AML with improved survival. A noted strength of the study was the objective quantitation of large numbers of cells using image analysis software. Although our patient cohort was small, patient outcomes are consistent with larger cohort analyses. Notwithstanding the other limitations of a retrospective design and variability in post-remission therapies, our study introduces the LIL index as a novel prognostic marker in CN-AML, with greatest potential utility in the indeterminate-risk NPM1/FLT3-ITD group. This work provides rationale for further studies to explore the function of the infiltrating T cells, identify T cell subsets that portend the survival benefit, and assess the LIL index in other AML genetic subgroups.

Acute myeloid leukemia: update in diagnosis and treatment in Brazil

OBJECTIVE To identify how the Brazilian hematology centers treated and diagnosed cases of acute myeloid leukemia in 2009. METHODS An epidemiological observational multicenter study of 11 listed Brazilian centers that treat acute myeloid leukemia and perform bone marrow transplantation. Data were collected from clinical charts of patients with acute myeloid leukemia treated at the said centers between 2005 and 2009. The availability for immunophenotyping and cytogenetic tests was assessed. RESULTS During 2009, a total of 345 new cases of acute myeloid leukemia were diagnosed. Differences were noted in the tests performed between patients who initiated treatment at the center and those referred for treatment. Of the participating centers, 72% conducted some type of molecular study in acute myeloid leukemia upon diagnosis. CONCLUSION Treatment for acute myeloid leukemia in Brazil shows significantly inferior results when compared to other centers worldwide.

Mutational Profiling of JAK2V617F vs. CALR mutated Primary Myelofibrosis

Background: The median survival of patients with primary myelofibrosis (PMF) is 5 to 7 years after diagnosis. In the majority of patients with PMF somatic mutations were detected either in Janus Kinase 2 (JAK2; in 60% of patients), Calreticulin (CALR; in 25% of patients) or MPL (in 5% of patients) genes. Neither mutation was detected 5% to 10% of PMF patients. Patients with mutated JAK2 are known to have a more aggressive disease compared to patients with mutated CALR. However patients with mutated JAK2 and high allele burden have a favorable outcome compared to patients with a low mutated JAK2 burden. Aim: To develop a model that uses genetic information to predict survival outcome of patients with PMF. Patients and Methods: Bone marrow samples were collected from 344 patients with PMF that were followed at MD Anderson Cancer Center between 2000 and 2013 (157 months). All samples were screened for JAK2 and for mutations in CALR. Patients who did not have a mutation in either gene were also screened for mutations in MPL. Results: In 226 patients (66%) JAK2 was detected and in 43 (13%) CALR was mutated. Of the 75 patients who did not have JAK2 or CALR mutations, 16 (21%) had mutated MPL. In 59 patients (17%), none of those mutations was detected. In the 226 patients who harbored the JAK2 mutation, a cut-point of 50% dichotomized patients into those with a high JAK2 burden and a favorable overall survival (OS; median OS: 80 months) and those with a low JAK2 burden and an adverse OS (median OS: 50 months). Age (above 65 years) and mutation status (low JAK2 burden or triple-negative) were independent risk factors. Patients with a favorable mutation status and age below 65 had a median survival of 126 months (n 1⁄4 82). Patients with either one risk factor, age above 65 (n 1⁄4 88) or adverse mutation status (n 1⁄4 87) had intermediate survival expectancy. The two risk factors were additive and patients age > 65 years and adverse mutation status (n 1⁄4 87) had a median survival of 35 months. Conclusions: Age and mutation status are independent predictors of survival in patients with PMF and stratify patients into 4 groups of equal size with very different survival outcome. 703 Mutational Profiling of JAK2V617F vs. CALR mutated Primary Myelofibrosis Fabio Santos, Renato Puga, Bianca Lisboa, Welbert Pereira, Mariana Miyagi, Evelyn Mata, Tarcila Datoguia, Isabel Bello, Michelli Diniz, Sandra Nakashima, Guilherme Perini, Ricardo Helman, Nelson Hamerschlak, Paulo Campregher Hematology/Oncology, Hospital Israelita Albert Einstein

Use of gemtuzumab ozogamycin combined with conventional chemotherapy in patients with acute myeloid leukemia

OBJECTIVE To analyze the outcome of patients treated with gemtuzumab ozogamycin combined with conventional therapy treated at Hospital Israelita Albert Einstein. METHODS 14 patients who had high risk features (secondary leukemia, unfavorable cytogenetics, and refractory disease) were treated with gemtuzumab ozogamycin combined with conventional therapy and their outcome was analysed by reviewing their medical records. RESULTS Overall response rate was 58%, with 43% achieving complete response, with a median follow-up of 11 months, event-free survival was 3 months. Eleven patients died, 6 of them due to refractory acute myeloid leukemia. Only four patients presented with grade 3 to 4 toxicities and only one patient had sinusoidal obstruction syndrome after bone marrow transplant. CONCLUSION gemtuzumab ozogamycin combined with chemotherapy is a feasible treatment regimen in acute myeloid leukemia patients. However, further studies are necessary to clarify which subgroup of patients may beneft from this treatment.