ALBERT

Description: Background: Erdheim-Chester disease (ECD) is a subtype of non-Langerhans histiocytic disorder that is shown to be driven by hyperactivation of MAPK pathway (most frequently caused by BRAFV600E mutation) and characterized by generalized organ dysfunction with infiltration of CD68-positive, CD1a-negative histiocytes. However, ECD is a rare entity and therefore very little is known about epidemiology of this disorder. Methods: We underwent a postal questionnaire-based, multi-center retrospective study to clarify the clinical features of ECD patients. We first sent 3850 questionnaires to various departments including orthopedics, respiratory medicine, dermatology, hematology, and pathology to cover as many ECD patients as possible, and identified 71 ECD patients in Japan. We further collected detailed clinical information and patients' samples if available. All cases were pathologically proven and the diagnoses of ECD were self-reported by each institute. DNA was extracted from each clinical sample and Sanger sequencing or allele-specific polymerase chain reaction (PCR) for BRAFV600E mutation were underwent with specific primers. Results: Among 71 patients with ECD, detailed clinical information about 38 patients were collected. The median age was 51 years old (range: 25-76 years old) and there was a male predominance (1.9:1). Major affected lesions were the bone (84 %), central nervous system (CNS; 50 %), cardiovascular lesion (37 %), skin (37 %), retroperitoneum (37 %), endocrines (37 %), lung (24 %), and digestive organ (12 %). C-reactive protein at onset was higher than upper normal limit in 24 of 29 patients (median 23.8 mg/L). The median time from the onset to diagnosis was 17 months and median survival from initial onset was 10.9 years. In the univariate analyses, age older than 60 years old at initial onset (hazard ratio [HR], 30.2; 95% CI, 5.7-159; p 〈 0.001), weight loss (HR, 5.4; 95% CI, 1.4-20.8; p = 0.014), CNS involvement (HR, 25.2; 95% CI, 3.1-203; p = 0.0024), cardiovascular lesion (HR, 3.6; 95% CI, 1.2-10.8; p = 0.021) and digestive organ disease (HR, 5.6; 95% CI, 1.6-20.2; p = 0.0085) were associated with poor prognosis. Interestingly, the bone involvement was associated with better outcome (HR, 0.20; 95% CI, 0.065-0.63; p = 0.0052). Multivariate analysis revealed that older age was an independent poor prognostic factor (HR, 18.9; 95% CI, 2.12-169; p = 0.0085). In addition, patients with CNS involvement were significantly older than patients without CNS disease (median, 63.7 vs 44.0 years old; p 〈 0.001). We also analyzed the BRAF V600E mutation status in seven cases. Allele-specific PCR identified that three of seven patients had BRAFV600E mutation. Conclusion: Our nationwide survey revealed that older age was associated with CNS involvement and poor prognosis in ECD patients. In addition, the bone, cardiovascular, CNS and digestive organ involvement might also affect clinical outcome. It is of note that the bone lesion was associated with better survival in the univariate analysis. However, larger and prospective studies are warranted. Regional disparity such as percentage of bone lesions should be also investigated. Disclosures Ogura: Payment for lectures including service on speakers bureaus: Speakers Bureau.

Description: Key Points Fbxl10 is a bona fide oncogene in vivo. Fbxl10 overexpression in HSCs induces mitochondrial metabolic activation and enhanced expression of Nsg2.

Description: Chronic myelogenous leukemia (CML) is a hematopoietic disorder originating from p210BCR/ABL-transformed stem cells, which begins as indolent chronic phase (CP) but progresses into fatal blast crisis (BC). To investigate molecular mechanism(s) underlying disease evolution, CML-exhibiting p210BCR/ABL transgenic mice were crossed with BXH2 mice that transmit a replication-competent retrovirus. Whereas nontransgenic mice in the BXH2 background exclusively developed acute myeloid leukemia, p210BCR/ABL transgenic littermates developed nonmyeloid leukemias, in which inverse polymerase chain reaction detected 2 common viral integration sites (CISs). Interestingly, one CIS was transgene's own promoter, which up-regulated p210BCR/ABL expression. The other was the 5′ noncoding region of a transcription factor, Zfp423, which induced aberrant Zfp423 expression. The cooperative activities of Zfp423 and p210BCR/ABL were demonstrated as follows: (1) introduction of Zfp423 in p210BCR/ABL transgenic bone marrow (BM) cells increased colony-forming ability, (2) suppression of ZNF423 (human homologue of Zfp423) in ZNF423-expressing, p210BCR/ABL-positive hematopoietic cells retarded cell growth, (3) mice that received a transplant of BM cells transduced with Zfp423 and p210BCR/ABL developed acute leukemia, and (4) expression of ZNF423 was found in human BCR/ABL-positive cell lines and CML BC samples. These results demonstrate that enhanced expression of p210BCR/ABL and deregulated expression of Zfp423/ZNF423 contribute to CML BC.

Description: Key Points Acquired expression of CblQ367P induces sustained proliferation of myelomonocytes, multilineage dysplasia, and splenomegaly resembling CMML. Combined inhibition of PI3K and JAK2 efficiently suppressed the growth of CblQ367P-induced CMML cells.

Description: Background: Kindlin-3 which is expressed mainly in hematopoietic cells, is essential for platelet fibrinogen receptor integrin αIIbβ3 (GPIIb-IIIa) activation and kindlin-3 deficiency causes severe bleeding problems. αIIbβ3 activation is tightly regulated by inside-out signaling, and the direct interaction of talin and kindlin-3 with β3-cytoplasmic tail following CalDAG-GEFI-induced Rap1 activation is critical for αIIbβ3 activation. We have reported that immediate and sustained αIIbβ3 activation by inside-out signaling is important for thrombus formation under physiological conditions (Blood 2016). However, the details of inside-out signaling are not still fully understood. Recently we identified patient with kindlin-3 deficiency as the first Japanese patient with kindlin-3 deficiency caused by novel p.W277X nonsense mutation. To clarify the role of kindlin-3 and the molecular mechanism of inside-out signaling, we analyzed single platelet behavior adhered on collagen under physiological flow conditions, and established kindlin-3 deficient human erythroleukemia HEL cell line. Case: The patient was 8-month old female, born to Japanese consanguineous parents. She has been suffering from bleeding tendency shortly after birth. Her peripheral blood showed slightly decreased platelet count (95x103/L), anemia (Hb 6.9 g/L), and elevated leukocyte count of 37.2x106/L with 1.0% blast. Although the surface expressions of glycoproteins were comparable to healthy control, her platelet aggregations and αIIbβ3 activation were strongly impaired in all agonist stimulations due to defective kindlin-3 expression. The sequencing analysis revealed the homozygous novel nonsense mutation, p.W277X (c.918G〉A) in kindlin-3. Methods: Human platelets were obtained from healthy control subjects (CT) and patients with kindlin-3 deficiency and Glanzmann thrombasthenia (GT). Whole blood was perfused at a shear rate 1,250s-1 on collagen surface and shear-induced in vitro thrombus formation during 10 minutes was observed. To evaluate the single platelet behavior after the initial attachment on collagen, each single platelet was analyzed by CellTracker software (Pccinini F. et al. Bioinformatics 2015). To further determine the role of kindlin-3 in inside-out signaling, Kindlin-3 was knocked out by CRISPR-Cas9 system and established kindlin-3 deficient HEL cell line. Results: First, we compared shear-induced thrombus formation between CT, GT, and kindlin-3 deficiency. In contrast to the stable and large platelet aggregate formation in CT after 10 minutes blood perfusion, almost no aggregate was observed in both GT and Kindlin-3 deficiency. In kindlin-3 deficiency, the initial platelet attachment on collagen seemed comparable to that of CT and GT. However, the single adherent platelets looked unstable. Next, we determined the behavior of initially attached platelets. Between CT, GT, and kindlin-3 deficiency, the numbers of platelets attached on collagen during 10 seconds were comparable. Between the initially attached platelets, 31.25% of CT platelets formed stable adhesion followed by platelet aggregation. Similar to CT, 33% of GT platelets adhered stably. However, these GT platelets did not proceed to aggregate formation due to αIIbβ3 deficiency. In contrast to GT, kindlin-3 deficient platelets showed increased detachment, only 9% of initially attached platelets stably adhered. These results suggest that kindlin-3 is indispensable for platelet initial adhesion to collagen by integrin α2β1 activation and explains severe bleeding symptoms in kindlin-3 deficiency than GT. To further investigate how kindlin-3 contributes in initial platelet adhesion to collagen, we established kindlin-3 deficient HEL cell line. In contrast to parental HEL cells, PMA stimulation did not induce αIIbβ3 activation in kindlin-3 deficient HEL cells suggesting impaired inside-out signaling. The introduction of wild type kindlin-3 cDNA, but not W227X mutant, rescued αIIbβ3 activation. Conclusion: The detailed analyses by tracking single adherent platelet on collagen confirmed the role of kindlin-3 in initial step of physiological thrombus formation. Newly established kindlin-3 deficient HEL cell line is useful for further exploration of the role of kindlin-3 and inside-out signaling in platelets and expected contribution for development of new antiplatelet therapy. Disclosures Tomiyama: Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Chugai: Honoraria; Kyowa-Kirin: Honoraria.

Description: Chronic immune activation is a major cause for progressive immunodeficiency in human immunodeficiency virus type-1 (HIV) infection. The underlying trigger, however, remains largely unknown. HIV single-stranded RNA is a potent immune activator by triggering Toll-like receptor (TLR) 7/8. Thus, we hypothesized that sustained TLR7 triggering induces chronic immune activation and thereby contributes to progressive immunodeficiency. We used the synthetic compound R848 or a mixture of uridine-rich HIV single-stranded (ss) RNA oligonucleotides—both are potent TLR7/8 agonists—to explore the effects of sustained TLR7 triggering on the murine lymphoid system. Sustained TLR7 triggering induced an immunopathology reminiscent of progressive lymphoid destruction in HIV disease; we observed lymphopenia, elevated proinflammatory cytokines, splenomegaly, contracted lymphoid subsets, and lymphoid microarchitecture alteration with reduced marginal zone B-lymphocytes. Upon exposure to inactivated vesiculo-stomatitis virus, antibody production was abolished, although splenic lymphocytes were activated and total IgG was elevated. Our data imply that HIV itself may directly contribute to immune activation and dysfunction by stimulating TLR7. Thus, manipulation of TLR7 signaling may be a potential strategy to reduce chronic hyper-immune activation and, thereby, disease progression in HIV infection.

Description: Prognosis of patients with AML varies widely reflecting the heterogeneity of AML and the background of each patient. It is important for both physicians and patients to predict the response to chemotherapy to correctly select the therapeutic options. In the JALSG-AML97 study, patients were stratified using JALSG scoring system, and those who were categorized into the intermediate or adverse prognosis group were assigned to allo-HSCT during CR1 if they had an HLA-matched sibling. JALSG scoring contains several factors that have 1 or 2 points: 2 points, age 〈 50 years, WBC 50%; 1 point, PS

Description: A randomized study had been performed between December 2001 and December 2005 to assess the optimal post remission therapy for adult AML in the first CR. The updated results are here presented, after a median follow-up of 48 months. JALSG AML201 enrolled 1064 previously untreated AML patients (pts) aged 15–64 yrs. The induction therapy consisted of cytarabine (Ara-C 100 mg/m2 day1–7) and idarubicin (IDR 12 mg/m2 day1–3) (arm A) or cytarabine (100 mg/m2 day1–7) and daunorubicin (DNR 50 mg/m2 day1–5) (arm B). If the patients did not achieve remission after the first induction therapy, then the same therapy was given once more. Pts were categorized into good, intermediate or poor risk groups by risk factors based on the criteria established in previous JALSG AML studies (Miyawaki et al. Cancer 2005). All CR pts were stratified according to the induction, the number of courses of induction, age and karyotype and were randomly assigned to the high dose Ara-C (HDAC) post remission regimen (arm C) or the conventional JALSG post remission regimen (arm D). Arm C: the three courses of HDAC which consisted of Ara-C 2.0 g/m2 q12h day1–5, arm D: the first course consisted of Ara-C 200 mg/m2 day1–5+ mitoxantrone (MIT) 7 mg/m2 day1–3, 2) Ara-C 200 mg/m2 day1–5+ DNR 50 mg/m2 day1–3, 3) Ara-C 200 mg/m2 day1–5+ aclarubicin (ACR) 20 mg/m2 day15, 4) Ara-C 200 mg/m2 day1–5+ etoposide (ETP) 100 mg/m2 day1–5 + vincristine (VCR) 0.8 mg/m2 day 8 + vindesine (VDS) 2 mg/m2 day10. Results: Of the 1064 pts registered, 1057 pts (median age: 47 years) were evaluable. 823 pts (78%) achieved CR after one or two courses of induction therapy. Of the 823 pts in CR, 781 pts were assigned to arm C (n=389) or arm D (n=392). The 5-year OS rate of arm C was 57.8% while that of arm D was 55.9% (p=0.96). The 5-year RFS rate of the CR pts was 42.7% in arm C and 38.9% in arm D (p=0.73). Among the good risk group (n=155), the 5-year OS rate of arm C was 69.9% while that of arm D was 80.5 % (p=0.11), and the 5-year RFS rate of arm C was 54.5% while that of arm D was 55.7% (p=0.53). Among the intermediate risk group (n=439), the 5-year OS rate of arm C was 50.9% while that of arm D was 48.5% (p=0.59), and the 5-year RFS rate of arm C was 41.5% while that of arm D was 36.5% (p=0.50). Among the poor risk group (n=49), the 5-year OS rate of arm C was 12.9% while that of arm D was 17.2% (p=0.58), and the 5-year RFS rate of arm C was 14.3% while that of arm D was 15.5% (p=0.78). In the CBF leukemia group (n=218), the 5-year OS rate of arm C was 75.0% while that of arm D was 65.8% (p=0.17), and the 5-year RFS rate of arm C was 56.5% while that of arm D was 38.7% (p=0.05). Among the young group (=50 yrs) (n=314), the 5-year OS rate of arm C was 51.3% while that of arm D was 40.1% (p=0.16), and the 5-year RFS rate of arm C was 40.0% while that of arm D was 28.1% (p=0.23). After all of consolidation, the lowest WBC count and the duration of neutropenia in arm C were significantly lower and longer than those in arm D. There was a higher rate of documented infection in arm C (20.9%) than in arm D (14.5%) (p〈 0.001). Conclusion: The conventional post remission therapeutic regimen established by JALSG consisting of 4 courses of consolidation was found to be as effective as the three courses of HDAC therapy. HDAC therapy produced a slightly positive effect on RFS in only the CBF leukemia group.

Description: In attempts to isolate myeloid tumor-suppressor genes responsible for 7q deletion, we identified a common microdeletion cluster in chromosome subband 7q21.2 by microarraybased CGH analyses of JMML (ASH Annual Meeting, 2006). This region was also deleted in nearly 30% of unselected adult MDS/AML patients, mostly as a part of monosomy 7 or larger 7q deletions. In this region, there are three poorly-characterized genes (Miki = LOC253012, Kasumi = Samd9, and Titan = Samd9L). Miki encoding a centrosomal protein is likely involved in myelodysplasia and chromosomal instability, which are characteristic of -7/7q- MDS/AML, as is presented in this meeting elsewhere. Kasumi (Samd9) and Titan (Samd9L) are related genes that encode 60% homologous proteins. Neither Kasumi nor Titan has homology with any other proteins or contain known functional motifs. Kasumi and Titan were ubiquitously expressed at a relatively constant level. However, in six cell lines derived from MDS/AML patients harboring monosomy 7, Kasumi protein was barely detectable, whereas Titan expression levels were roughly half of those in other AML cells. The mouse genome contains only Titan and lacks Kasumi gene, suggesting that the function of these two gene products are overlapping. We started to characterize these genes by generating mice deficient in Titan (titan−/−). titan−/− mice appear normal and no hematological abnormalities have been observed, suggesting that additional gene alterations are required for leukemia development. To address this issue, retroviral insertional mutagenesis was applied to the mice. Virus infection induced acute leukemia in homozygous (titan−/−) and heterozygous (titan+/−) mice with higher morbidity and mortality than in wild-type (titan+/+) littermates. Leukemias developed in titan+/+ mice were mainly of T-cell lineage. By contrast, those developed in titan−/− and titan+/− mice were negative for lymphoid markers but expressed various combination of cell surface markers for myeloid (Gr1), monocytic (Mac1), erythyroid (Ter119) and megakaryocytic (CD61) progenitors. Histopathology demonstrated that leukemia cells infiltrated the liver, lung, kidneys and spleen, and a portion of the infiltrated cells were maturated. These data suggests that leukemias that developed in titan-deficient mice represent stem cell malignancy rather than AML. Inverse PCR detected two common integration sites (CIS) specific for titan−/− and titan+/− mice, which induced deregulated expression of a zinc finger transcription factor, Evi1, and a histone demethylase, Fbxl10. In addition, although it was not a CIS, TGFβ was isolated as a major viral integration site in one tumor. These results demonstrated that haploinsufficiency and deficiency of Titan predispose leukemia development through inhibition of TGFβ-mediated signaling or an epigenetic change. Recently, deleterious mutations in the Titan gene were reported to be involved in Normophosphatemic Familial Tumoral Carcinosis, a rare autosomal recessive disease in five families of Jewish-Yemenite origin. Impairment of cell migration is suspected to be a cause of this disease and, indeed, wound healing test revealed that fibroblasts established from titan−/− and titan+/− mice migrate slower than those established from wild-type mice. Relevance of the impairment of cell migration to development of leukemia in titan-deficient mice is currently under investigation.

Description: Between 2001 and 2005, JALSG conducted a randomized study to assess the optimal post remission therapy for adult AML in the first CR. JALSG AML201 enrolled 1064 previously untreated AML patients (pts) aged 15–64 yrs. The induction therapy consisted of cytarabine (Ara-C 100mg/m2 day1–7) and idarubicin (IDR 12mg/m2 day1– 3) (arm A) or cytarabine (100mg/m2 day1–7) and daunorubicin (DNR 50mg/m2 day1– 5) (arm B). If the patients did not achieve remission after the first induction therapy, then the same therapy was given once more. Pts were categorized into good, intermediate or poor risk groups by risk factors based on the criteria established in previous JALSG AML studies. All CR pts were stratified according to the induction, the number of courses of induction, age and karyotype and were randomly assigned to the high dose Ara-C (HDAC) post remission regimen (arm C) or the conventional JALSG post remission regimen (arm D). Arm C: the three courses of HDAC which consisted of Ara-C 2.0g/m2 q12h day1–5, arm D: the first course consisted of Ara-C 200mg/m2 day1–5+ mitoxantrone (MIT) 7mg/m2 day1–3, 2) Ara-C 200mg/m2 day1–5+ DNR 50mg/m2 day1–3, 3) Ara-C 200mg/m2 day1–5+ aclarubicin (ACR) 20mg/m2 day1–5, 4) Ara-C 200mg/m2 day1–5+ etoposide (ETP) 100mg/m2 day1–5 + vincristine (VCR) 0.8mg/m2 day 8 + vindesine (VDS) 2 mg/m2 day10. Results: Of the 1064 pts registered, 1057 pts (median age: 47 years) were evaluable. 825 pts (78%) achieved CR after one or two courses of induction therapy. Of the 825 pts in CR, 781 pts were assigned to arm C or arm D. The 4-year OS rate of arm C was 61.6% while that of arm D was 62.8% (p=0.58). The 4-year RFS rate of the CR patients was 42.8% in arm C and 40.8% in arm D (p=0.65). Among the good risk group, the 4-year OS rate of arm C was 77.0% while that of arm D was 75.8 % (p=0.40), and the 4-year RFS rate of arm C was 54.6% while that of arm D was 53.1% (p=0.71). Among the intermediate risk group, the 4-year OS rate of arm C was 63.2% while that of arm D was 65.7% (p=0.78), and the 4-year RFS rate of arm C was 38.7% while that of arm D was 42.2% (p=0.63). Among the poor risk group, the 4-year OS rate of arm C was 36.3% while that of arm D was 34.1% (p=0.71), and the 4-year RFS rate of arm C was 25.9% while that of arm D was 6.6% (p=0.17). In the CBF leukemia group, the 4-year OS rate of arm C was 79.4% while that of arm D was 66.5% (p=0.09), and the 4-year RFS rate of arm C was 57.7% while that of arm D was 43.7% (p=0.14). Among the young group (=50 yrs), the 4-year OS rate of arm C was 53.4% while that of arm D was 57.7% (p=0.90), and the 4-year RFS rate of arm C was 39.1% while that of arm D was 33.8% (p=0.67). Conclusion: The conventional post remission therapeutic regimen established by JALSG consisting of 4 courses of consolidation was thus found to be as effective as the three courses of HDAC therapy. To further confirm these results, a longer follow-up is therefore needed.