- Email: [email protected]
SF3B1-mutated myelodysplastic syndrome with ring sideroblasts harbors more severe iron overload and corresponding over-erythropoiesis
Accepted Manuscript Title: SF3B1-mutated myelodysplastic syndrome with ring sideroblasts harbors more severe iron overload and corresponding over-erythropoiesis Author: Yang Zhu Xiao Li Chunkang Chang Feng Xu Qi He Juan Guo Ying Tao Yizhi Liu Li Liu Wenhui Shi PII: DOI: Reference:
S0145-2126(16)30026-1 http://dx.doi.org/doi:10.1016/j.leukres.2016.02.011 LR 5546
To appear in:
Leukemia Research
Received date: Revised date: Accepted date:
29-10-2015 15-1-2016 25-2-2016
Please cite this article as: {http://dx.doi.org/ This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
SF3B1-mutated myelodysplastic syndrome with ring sideroblasts harbors more severe iron overload and corresponding over-erythropoiesis Yang Zhu1, Xiao Li1*, Chunkang Chang1, Feng Xu1, Qi He1, Juan Guo1, Ying Tao1, Yizhi Liu1, Li Liu1, Wenhui Shi1 Department of Haematology, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai, China 1
*
Correspondence: Xiao Li, M.D., PhD., Department of Haematology, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, 600 Yishan Road, Shanghai 200233, China. E-mail： [email protected]
Graphical abstract
Highlights: SF3B1-mutated MDS-RS harbored more severe iron overload and over-erythropoiesis. The presence of the SF3B1 mutation was associated with a better overall survival. IPSS risk groups primarily accounted for prognostic value of the SF3B1 mutation.
Abstract Objective: To clarify the possible biological differences and implication of the SF3B1 gene for patients with MDS-RS (myelodysplastic syndromes with ring sideroblasts). Methods: Sanger sequencing was performed on mutation hotspots of the SF3B1 gene in MDS-RS patients. The differences between the SF3B1 mutated and wild-type subsets, including the ultrastructure of erythroid precursors, iron profile parameters, erythropoiesis-related measurements, as well as clinical features, were analyzed. Results: SF3B1 mutations were detected in 33 out of fifty-two MDS-RS patients (63%). The vast majority of patients with mutations (94%) were categorized in the lower risk group according to the IPSS (International Prognostic Scoring System), in contrast to only fifty-eight percent of the wild-type cases. In addition to the notably higher percentages of erythroblasts and ring sideroblasts in patients with mutations, abundant electron-dense granules in the mitochondria of the erythroid precursors were clearly observed. Moreover, patients with mutations presented both improper iron uptake and distribution (lower serum hepcidin-25 concentration, P=0.028) and enhanced erythropoietic activity (higher soluble transferrin receptor level, P=0.132; higher growth differentiation factor 15 concentration, P<0.001). Finally, MDS-RS patients carrying SF3B1 mutations had a better overall survival (median 38 vs. 18 months, P=0.001) compared to those without mutations. By multivariable analysis, the prognostic significance of the SF3B1 mutation was primarily accounted for by IPSS risk categorization. Conclusion: MDS-RS patients carrying SF3B1 mutations harbored a more severe iron overload and corresponding over-erythropoiesis. The better overall survival of SF3B1-mutated MDS-RS patients may be mainly due to the clustering of patients with lower risk disease in this group. Keywords: myelodysplastic syndrome; ring sideroblasts; SF3B1 mutation; iron overload; erythropoiesis. 1. Introduction The myelodysplastic syndromes (MDSs) are a group of heterogeneous diseases characterized by dysplasia, ineffective hematopoiesis, and a variable risk of progression to acute myeloid leukemia [1,2]. In addition to marrow dysplasia, increased ring sideroblasts (RS) are a key pathological feature of MDS [2,3]. The terminology MDS with ring
sideroblasts (MDS-RS) was introduced as an entity that is morphologically characterized by dysplasia and the presence of 15% or higher RS in marrow aspirate smears. Though MDS-RS only accounts for 20-25% of all MDS patients, most are accompanied by severe anemia and iron overload, which exacerbate the difficulties in clinical treatment [4]. Until now, there has been only limited research on the pathogenesis of MDS-RS. With the application of high-throughput whole genome sequencing, somatic mutations of SF3B1 (splicing factor 3b subunit 1) had recently been identified in 60%～80% of MDS-RS patients, compared to a frequency of mutations of less than 10% in other myeloid neoplasms [5-7]. In addition, recent studies indicated that the SF3B1 mutation is associated with decreased expression of the mitochondrial iron transporter ABCB7 [8,9]. Although the SF3B1 mutations were considered to be the primary cause of the distinctive phenotype of MDS-RS, there are still approximately 30%-40% of MDS-RS patients who are SF3B1 wild type. The presence of RS arises from a defect of mitochondrial iron metabolism in erythroid precursors. Iron homeostasis in erythroblasts is damaged in MDS-RS patients, which result in an alteration of systemic iron metabolism. It has been hypothesized that differences in the mitochondrial iron phenotype and systemic iron profile may exist between SF3B1 mutated MDS-RS and their mutation-negative counterparts. Moreover, there are inconsistencies in the prognostic value of SF3B1 mutations. Some studies showed that SF3B1 mutations were associated with a benign phenotype, whereas others reported that there was no additional prognostic value of SF3B1 mutations [6,10-13]. In the present study, we investigated the biological differences (iron profile, erythropoiesis and clinical features) of MDS-RS patients carrying the SF3B1 mutation and the wild-type allele. This research lays the foundation for clarifying the biological differences between two subgroups of MDS-RS.
2. Patients and Methods 2.1. Subjects and samples The study was approved by the institutional ethics committee, and the patient-relevant research strictly abided by the ethical standards laid down in the Declaration of Helsinki. Between November 2008 and March 2014, fifty-two patients with MDS-RS who were initially
diagnosed at the institute were enrolled. At the time of diagnosis, serum and bone marrow samples were collected after obtaining informed written consent. We excluded persons with secondary acquired sideroblastic anemias including alcohol abuse, lead toxicity, copper or pyridoxine deficiency, isoniazid therapy and hereditary sideroblastic anemias. On aspirate smears stained with Prussian blue, at least 15% of erythroid precursors were ring sideroblasts. Ring sideroblasts were defined as erythroblasts in which a minimum of five siderotic granules covered at least one-third of the nuclear circumference [14]. According to the 2001 and 2008 World Health Organization (WHO) classification criteria, 23 patients had refractory anemia with ring sideroblasts (RARS), 20 patients had refractory cytopenia with multilineage dysplasia and ring sideroblasts (RCMD-RS), 6 patients had refractory anemia with excess blasts-1 (RAEB-1), and 3 patients had refractory anemia with excess blasts-2 (RAEB-2). The patients were categorized into four risk groups (low, 0; INT-1, 0.5-1.0; INT-2, 1.5-2.0; and high, ≥2.5) defined according to the International Prognostic Scoring System (IPSS) [15]. The treatment details were obtained from the patients’ medical records. Fifty-nine percent of MDS-RS patients (31/52) had received RBC transfusion support (Table 1). Lenalidomide, iron chelators, hypomethylating agents and chemotherapy drugs were not administered. Follow-up for survival analysis ended in December 2014.
2.2. DNA extraction and SF3B1 mutational analysis The mononuclear cells were isolated from bone marrow aspirates of MDS-RS patients by density-gradient centrifugation using Ficoll (Lymphosep, Biowest, France). Genomic DNA was extracted using a DNA Extraction Kit (TianGen Biotech, Beijing, China) according to the manufacturer’s instructions. SF3B1-coding exon sequencing was performed using a DNA aliquot derived from bone marrow mononuclear cells. Genomic DNA corresponding to the mutation hotspots of SF3B1 (exons 13–16) were amplified by PCR. The primers are as follows: exon13-F, 5'-TTCGTCCCTTGATTAACAAAAGTC-3', exon 13-R, 5'-TCAGGTCTCATGGTAGAGATCA-3', exon14-F, 5'-TCTGTTTATGGAATTGATTAT G-3', exon14-R, 5'- TCTAGTCCCAACTACTAAGGAGGC-3', exon15.16-F, 5'-TTATCTGC TGACAGGCTATGGTTC-3', and exon15.16-R, 5'-ATCAACTGACCTGAAATGAAGAG-3'. All products were confirmed by 1% agarose gel electrophoresis and sequenced using an
ABI 3730xl DNA analyzer (Applied Biosystems) according to standard techniques. Sequencing data were analyzed using Mutation Surveyor (DNA Variant Analysis Software, Softgenetics) followed by visual inspection of sequencing traces.
2.3. Cytogenetic analysis Chromosomal specimens were prepared by an instant manipulation and a short-term culture (24–48h) of bone marrow cells. Analysis of the karyotype was performed in at least 20 metaphase cells by the G-banding technique. Karyotypes were described according to the International System of Human Cytogenetic Nomenclature (ISCN) [16]. Karyotype anomalies were defined in three major risk categories (good risk, intermediate risk, poor risk) according to the International Myelodysplastic Syndrome Working Group [15].
2.4. Transmission electron microscopy Ten million BM mononuclear cells were first fixed in 2% glutaraldehyde for more than 2 hours. Fixed samples were washed twice in 0.1M phosphate buffered saline (PBS) for 20 minutes each. Specimens were post-fixed in 1% osmium tetroxide for 2 hours. After a second fixation step, specimens were washed three times in water solution for 20 minutes each. After the final wash, specimens were dehydrated in ascending graded alcohols (30% alcohol, ten minutes→50% alcohol, ten minutes→70% alcohol and 3% uranyl acetate, overnight→80% alcohol, ten minutes→95% alcohol, ten minutes→100% alcohol, ten minutes), followed by immersion in propylene oxide, passage through ascending graded mixtures of propylene oxide and epoxy resin 618, and finally embedding in pure epoxy resin. Specimen blocks were polymerized at 60°C for a period of 48 hours. Ultrathin sections were cut at 60-80 nm by a LKB v-shaped ultra-thin slicing machine and supported on 200 mesh copper grids. Sections were stained in lead citrate and examined using a PHILIPS CM-120 transmission electron microscope. Electron micrographs were taken using a Gatan Orius CCD digital camera.
2.5. Iron metabolism-related assays Hepcidin-25 (25 amino acid peptide) is the active form of the peptide hormone hepcidin, a
key player in the regulation of iron metabolism. To investigate the status of iron metabolism, the levels of serum hepcidin-25 in MDS-RS patients were measured. Reliable quantitative (second-generation) hepcidin assays have been developed in recent years [17,18]. Of the quantitative hepcidin assays, the enzyme immunoassay kit of Bachem was a competitive immunoassay that had been evaluated to differentiate the hepcidin concentrations in serum samples of patients and controls with various iron loads [17,19-21]. Serum hepcidin-25 concentrations were determined using a commercial enzyme immunoassay kit following manufacturer’s protocol (Bachem, San Carlos, USA). Control serum samples were obtained from forty healthy individuals with a median age of 58 years (range, 26-81 years). The reference range of serum hepcidin-25 is from 16.9 to 52.7 ng/ml (median, 31.2 ng/ml). The serum iron and total iron binding capacity were determined using the ferrous-oxazine chromogenic method. Serum ferritin (SF) was determined by nephelometric assay. 2.6. Measurement of serum growth differentiation factor 15 and soluble transferrin receptor The concentration of growth differentiation factor 15 (GDF15) in serum samples was evaluated using an enzyme-linked immunosorbent assay (ELISA) for human GDF15 (R&D Systems, Minneapolis, MN, USA) [22]. Soluble transferrin receptor (sTfR) was quantified using an immunonephelometric method (Tina-quant Soluble Transferrin Receptor, Roche Diagnostics GmbH, Germany) [23].
2.7. Statistical analysis Numerical variables are presented as the median and range; categorical variables are described as the count and relative frequency (%) of subjects in each category. Comparison of numerical variables between groups was performed using a nonparametric approach (Mann–Whitney test or Kruskal–Wallis ANOVA). The distribution of categorical variables in different groups was compared with either the Fisher exact test (when computationally feasible) or the χ2 test. Survival was calculated from the date of diagnosis to death from any cause or last follow-up. Survival analyses were carried out with the Kaplan–Meier method. Multivariable survival analyses were performed by Cox proportional hazards regression. P
values less than 0.05 (two-sided) were considered statistically significant. Analyses were performed with the SPSS software package, version 18.0.
3. Results 3.1. Frequencies and distribution of SF3B1 mutations A total of 52 patients were diagnosed with MDS-RS (≥15% of ring sideroblasts were identified in bone marrow smear). Heterozygous point mutations of SF3B1 were detected in 33 subjects (63%) including the K700E point mutation in 18, K666T/N in 6, H662Q in 4, E622D in 3, R625L in 1 and D781G in 1 case (Figure 1). All missense mutations leading to amino acid substitutions clustered in exons 14 and 15. The frequencies of the SF3B1 mutation were 19 out of 23 (83%) for RARS, 11 out of 20 (55%) for RCMD-RS, 2 out of 6 (33%) for RAEB-1, and 1 out of 3 (33%) for RAEB-2. The proportion of subjects with SF3B1 mutations was significantly higher in the RARS and RCMD-RS than that in other MDS-RS subgroups (Figure 2A). According to the IPSS risk groups, the frequencies of the SF3B1 mutation were 12 out of 14 (86%) for the low group, 19 out of 28 (68%) for the INT-1 group, 2 out of 8 (25%) for the INT-2 group, and 0 out of 2 (0%) for the high group (Figure 2B).
3.2. Clinical features Subjects with the SF3B1 mutation had a higher percentage of bone marrow erythroblasts (median, 54.0% vs. 36.5%; P=0.036) and ring sideroblasts (median, 43% vs. 26%; P=0.026). Patients carrying SF3B1 mutations had a higher platelet counts (median, 150×109/L vs. 46×109/L; P<0.001) and a lower percentage of BM blast cells (median, 1.8% vs. 4.8%; P=0.003). In addition, the proportion of good risk IPSS cytogenetics in subjects with the SF3B1 mutation was significantly higher than in wild-type patients (73% vs. 21%). The vast majority of SF3B1 mutated patients were categorized into the lower IPSS risk groups (94% cases with IPSS ≤1.0 vs. 58% in wild-type patients) (Table 1). Survival analysis was conducted for all enrolled MDS-RS subjects with a median follow-up of 18 months (range, 6–54 months). Subjects with SF3B1 mutations had a better overall survival (median, 38 months (95% CI, 26–48 months) vs. 18 months (95% CI, 16–20 months); P=0.001).
In univariate analysis, survival was predicted according to the WHO morphologic category (P<0.001), IPSS (P<0.001), percentage of BM blasts (P<0.001), karyotype (P=0.005), and SF3B1 mutational status (P=0.004). In multivariable analysis, only the IPSS and percentage of BM blasts remained significant (Table 2). Moreover, the prognostic significance of SF3B1 mutation was lost during multivariable analysis that included, as a covariate, the percentage of BM blasts, WHO morphologic categorization or IPSS risk categorization; the P values were 0.313, 0.116 and 0.685, respectively.
3.3. Ultrastructure of erythroid precursors Although Prussian blue staining is traditionally used to identify RS in a bone marrow smear, this procedure does not distinguish microstructural changes of mitochondria. Using transmission electron microscopy, ultrastructure observations were performed on BM cells derived from six patients carrying the SF3B1 mutation (4 RARS, 2 RCMD-RS) and three patients wild-type for SF3B1 (1 RARS, 2 RCMD-RS). The predominant ultrastructural alterations in SF3B1-mutated patients were the abundant electron-dense deposits in the mitochondria around the nucleus of the erythroblasts. A great number of these mitochondria cristae were completely destroyed. In addition, numerous cytoplasmic vacuoles and membrane blebbing were observed. In contrast, sparse deposits and fewer cytoplasmic vacuoles were shown in wild-type patients (Figure 3). Quantitative analyses of iron-laden granules in mitochondria were digitalized using Image-Pro Plus software. The density values and area size of electron-dense deposits in cases with mutations were significantly higher than that of wild-type patients (Table 3). 3.4. Serum hepcidin-25 and ferritin levels Because of the different mitochondrial iron pattern in SF3B1 mutated patients, it is worth exploring whether there are differences in the iron profile parameters of SF3B1 mutated and wild-type MDS-RS cases. As illustrated in Figure 4, a higher serum ferritin level was observed in SF3B1-mutated patients (median, 1088.9 ng/ml vs. 575.7 ng/ml; P=0.107). However, patients carrying SF3B1 mutations have a notably lower serum hepcidin-25 concentration than wild-type patients (median, 17.3 ng/ml vs. 77.8 ng/ml, P=0.028). The hepcidin-25 to ferritin ratio
represents a measure of the adequacy of hepcidin levels relative to body iron stores. Significantly lower values of the hepcidin-25 to ferritin ratio were found in patients with SF3B1 mutation (median, 0.017 vs. 0.091; P=0.006). The linear correlation between hepcidin-25 and serum ferritin according to the SF3B1 mutational status in MDS-RS patients suggests that the hepcidin level did not increase proportionally compared to the body iron stores in SF3B1-mutated patients. These data implied that the degree of iron load is more severe in MDS-RS patients carrying the SF3B1 mutation.
3.5. Erythropoiesis profile measurements The proportion of marrow erythroblasts is associated with the level of erythropoietic activity. The level of serum sTfR is closely related to cellular iron demands and the erythroid proliferation rate [24]. GDF15 is released at sites of active erythropoiesis. The expression of GDF-15 is at low levels during normal erythropoiesis, but at high levels in patients with ineffective erythropoiesis [25,26]. The bone marrow differential cell count showed that the proportion of erythroblasts in SF3B1-mutated patients was much higher than that of wild-type patients (median, 54.0% vs. 36.5%; P=0.036; Figure 5A). Moreover, SF3B1-mutated patients had a higher serum sTfR level (median, 5.4 mg/L vs. 3.3 mg/L; P=0.132; Figure 5B) and GDF15 concentration (median, 5856.3 pg/ml vs. 848.6 pg/ml; P<0.001; Figure 5C). These data indicated that the erythroid marrow function in SF3B1-mutated cases showed both more active activity and ineffective erythropoiesis. 3.6. General trend in iron profile and erythropoietic activity The amount of RBC transfusions in SF3B1-mutated MDS-RS patients was not statistically more than that in wild-type cases (Table 1), which indicated that there was no significant difference in transfusional iron burden between these subgroups. As shown in Figure 6, SF3B1-mutated patients had higher iron profile measurements, including ferritin，serum iron and transferrin saturation. Moreover, the measurements related to erythropoietic activity were much higher in SF3B1-mutated cases. However, the serum hepcidin-25 concentration and hepcidin-25 to ferritin ratio of SF3B1-mutated patients were markedly lower than that of
wild-type cases.
4. Discussion In recent years, high-throughput sequencing has revealed that RNA splicing pathway genes were frequently mutated (～45 to ～85%) in patients with MDS and that most of the mutations occurred in a mutually exclusive manner [5]. These mutations are considered to be highly associated with the pathogenesis of these diseases and have become a hotspot of research. In this study of fifty-two patients with MDS-RS, recurrent mutations in SF3B1, a gene encoding a core component of the RNA splicing machinery, were identified in thirty-three subjects (63%). All observed mutations were missense substitutions. Mutations clustered in six codons of exon 14 and 15, and codon 700 was the most frequent mutation spot (54.5% of mutated cases). Analysis of the results of the SF3B1 mutation were consistent with the data reported previously that indicated a significant association of SF3B1 mutations with the presence of ring sideroblasts in MDS. Ringed sideroblasts are characteristic of refractory anemia with ring sideroblasts (RARS), but they are frequently observed in other subtypes of MDS. The presence of ringed sideroblasts has previously been observed in 40% of patients with refractory anemia with excess blasts (RAEB) and greater than 20% of RS in one-fifth of such cases [27]. We investigated the clinical correlates of SF3B1 mutations in patients with MDS-RS. The lower proportion of marrow blast cells (median 1.8% vs. 4.8%, P=0.003) and higher proportion of good or intermediate risk karyotypes (100% vs. 63%, P<0.001) were found in SF3B1 mutated patients. The vast majority of SF3B1-mutated patients were classified into IPSS ≤ 1.0 groups (94% vs. 58%). In addition, the median survival of SF3B1-mutated patients was significantly longer than that of wild-type patients (38 months vs. 18 months). A recent report described coarse iron deposits in erythroid mitochondria of a RARS patient carrying a SF3B1 mutation, and low hepcidin levels in SF3B1-mutated MDS patients were reported in other study [28,29]. In the present study, we confirmed the findings of earlier studies and focused on the iron profile and erythropoiesis in one population characterized by MDS and increased ring sideroblasts. As illustrated in Figure 3 and Table 3, large amounts of electron-dense granules were deposited in the mitochondria of erythroid precursors from
SF3B1-mutated patients. By image analysis with Image-pro Plus software, the density of the granules was significantly increased compared with those from patients without mutation. The ultrastructural alterations suggested that the trafficking of iron molecules between mitochondria and cytoplasm was disturbed more severely in SF3B1-mutated patients. In addition, we analyzed the biochemical indexes related to iron metabolism. The results showed that the serum ferritin level of SF3B1-mutated patients was higher than that of patients
without
the
mutation,
and
the
serum
hepcidin-25
concentration
and
hepcidin-25/ferritin ratio was statistically lower than those of wild-type patients. This indicated that there was a heavier iron load in SF3B1-mutated patients. Due to the decrease in the hepcidin-25 concentration, increased absorption of iron in the intestine is still present. As illustrated in Figure 4, when the serum ferritin level rose in wild-type patients, the hepcidin-25 concentration increased correspondingly to inhibit iron absorption. However, in SF3B1-mutated patients, even if the level of serum ferritin rose, the hepcidin-25 concentration could not increase proportionally, resulting in an iron overload in the body. The results of erythropoiesis-related indicators were also analyzed. The proportion of marrow erythroblasts in SF3B1-mutated patients was significantly higher compared to wild-type patients. This finding suggested that the erythroid precursors multiplied markedly and that erythropoiesis was exuberant in SF3B1-mutated patients. We also found statistically elevated sTfR and GDF-15 levels in patients with mutations. With regard to cytokines, there was more vigorous erythropoietic activity in SF3B1-mutated patients. Earlier studies [25,26] reported that serum GDF15 levels was markedly increased in hematological diseases with ineffective erythropoiesis and that an elevated GDF15 concentration suppressed hepcidin synthesis in primary human hepatocytes. We concluded that more active erythropoietic activity diminished the hepcidin response to increased iron burden in MDS-RS patients with SF3B1 mutations. Taken together, a possible mechanism of pathogenesis for the more active iron uptake and erythropoiesis in SF3B1-mutated MDS-RS may be a consequence of the under-utilization of iron and iron deposition in mitochondria. Abnormal iron metabolism of erythroblasts may have promoted the erythropoiesis-related cytokines through a feedback loop to enhance erythropoiesis. This futile effort (increased levels cytokines acting on under-utilized iron) finally lead to the clinical phenotype of iron overload and severe anemia.
The observation that MDS-RS patients with the SF3B1 mutation had a better overall survival compared to wild-type patients was explored. This study showed SF3B1 mutations clustered with good risk karyotypes (73% vs. 21%). Twenty-eight out of the 52 MDS-RS patients were enrolled in another research study in which an additional 10 targeted genes were sequenced (unpublished data). Additional mutations were found in 80% (8/10) of wild-type patients, compared to 39% (7/18) of SF3B1-mutated patients. The incidence rate of poor prognosis-related gene mutations in wild-type patients was higher than that in SF3B1-mutated patients (Table 4). There was a possibility of coexistence of a founding clone and subclone in hematopoietic stem cells of MDS [30]. Therefore, further study is warranted to define whether MDS-RS SF3B1 wild-type is a later stage caused by the evolution of more aggressive clone.
5. Conclusion In this study, SF3B1 mutations were detected in 33 out of fifty-two MDS-RS patients (63%). Compared to wild-type MDS-RS, patients carrying the SF3B1 mutation harbored more severe iron overload and corresponding over-erythropoiesis. The presence of the SF3B1 mutation was associated with a better overall survival. By multivariable analysis, the prognostic significance of the SF3B1 mutation was primarily accounted for by IPSS risk categorization.
Further
study
is
warranted
to
determine
whether
the
SF3B1
mutation-positive MDS-RS and mutation-negative allele are two distinct subtypes of the disease or just different stages of the same MDS subset.
References [1] Cazzola M, Malcovati L. Myelodysplastic syndromes--coping with ineffective hematopoiesis. N Engl J Med 2005; 352: 536-8. doi:10.1056/NEJMp048266 [2] Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189-99. [3] Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002; 100: 2292-302. doi:10.1182/blood-2002-04-1199 [4] Malcovati L, Porta MG, Pascutto C, Invernizzi R, Boni M, Travaglino E, et al. Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making. J Clin Oncol 2005; 23: 7594-603. doi:10.1200/jco.2005.01.7038
[5] Yoshida K, Sanada M, Shiraishi Y, Nowak D, Nagata Y, Yamamoto R, et al. Frequent pathway mutations of splicing machinery in myelodysplasia. Nature 2011; 478: 64-9. doi:10.1038/nature10496 [6] Papaemmanuil E, Cazzola M, Boultwood J, Malcovati L, Vyas P, Bowen D, et al. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med 2011; 365: 1384-95. doi:10.1056/NEJMoa1103283 [7] Visconte V, Makishima H, Jankowska A, Szpurka H, Traina F, Jerez A, et al. SF3B1, a splicing factor is frequently mutated in refractory anemia with ring sideroblasts. Leukemia 2012; 26: 542-5. doi:10.1038/leu.2011.232 [8] Boultwood J, Pellagatti A, Nikpour M, Pushkaran B, Fidler C, Cattan H, et al. The role of the iron transporter ABCB7 in refractory anemia with ring sideroblasts. PloS one 2008; 3: e1970. doi:10.1371/journal.pone.0001970 [9] Nikpour M, Scharenberg C, Liu A, Conte S, Karimi M, Mortera-Blanco T, et al. The transporter ABCB7 is a mediator of the phenotype of acquired refractory anemia with ring sideroblasts. Leukemia 2013; 27: 889-96. doi:10.1038/leu.2012.298 [10] Malcovati L, Papaemmanuil E, Bowen DT, Boultwood J, Della Porta MG, Pascutto C, et al. Clinical significance of SF3B1 mutations in myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms. Blood 2011; 118: 6239-46. doi:10.1182/blood-2011-09-377275 [11] Patnaik MM, Lasho TL, Hodnefield JM, Knudson RA, Ketterling RP, Garcia-Manero G, et al. SF3B1 mutations are prevalent in myelodysplastic syndromes with ring sideroblasts but do not hold independent prognostic value. Blood 2012; 119: 569-72. doi:10.1182/blood-2011-09-377994 [12] Bejar R, Stevenson KE, Caughey BA, Abdel-Wahab O, Steensma DP, Galili N, et al. Validation of a prognostic model and the impact of mutations in patients with lower-risk myelodysplastic syndromes. J Clin Oncol 2012; 30: 3376-82. doi:10.1200/jco.2011.40.7379 [13] Damm F, Thol F, Kosmider O, Kade S, Loffeld P, Dreyfus F, et al. SF3B1 mutations in myelodysplastic syndromes: clinical associations and prognostic implications. Leukemia 2012; 26: 1137-40. doi:10.1038/leu.2011.321 [14] Mufti GJ, Bennett JM, Goasguen J, Bain BJ, Baumann I, Brunning R, et al. Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts. Haematologica 2008; 93: 1712-7. doi:10.3324/haematol.13405 [15] Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997; 89: 2079-88. [16] Gonzalez Garcia JR, Meza-Espinoza JP. Use of the International System for Human Cytogenetic Nomenclature (ISCN). Blood 2006; 108: 3952-3. doi:10.1182/blood-2006-06-031351 [17] Kroot JJ, Tjalsma H, Fleming RE, Swinkels DW. Hepcidin in human iron disorders: diagnostic implications. Clin Chem 2011; 57: 1650-69. doi:10.1373/clinchem.2009.140053 [18] Kemna EH, Tjalsma H, Willems HL, Swinkels DW. Hepcidin: from discovery to differential diagnosis. Haematologica 2008; 93: 90-7. doi:10.3324/haematol.11705 [19] Castagna A, Campostrini N, Zaninotto F, Girelli D. Hepcidin assay in serum by SELDI-TOF-MS and other approaches. J Proteomics 2010; 73: 527-36. doi:10.1016/j.jprot.2009.08.003
[20] Kroot JJ, Laarakkers CM, Geurts-Moespot AJ, Grebenchtchikov N, Pickkers P, van Ede AE, et al. Immunochemical and mass-spectrometry-based serum hepcidin assays for iron metabolism disorders. Clin Chem 2010; 56: 1570-9. doi:10.1373/clinchem.2010.149187 [21] Kroot JJ, Kemna EH, Bansal SS, Busbridge M, Campostrini N, Girelli D, et al. Results of the first international round robin for the quantification of urinary and plasma hepcidin assays: need for standardization. Haematologica 2009; 94: 1748-52. doi:10.3324/haematol.2009.010322 [22] Ramirez JM, Schaad O, Durual S, Cossali D, Docquier M, Beris P, et al. Growth differentiation factor 15 production is necessary for normal erythroid differentiation and is increased in refractory anaemia with ring-sideroblasts. Br J Haematol 2009; 144: 251-62. doi:10.1111/j.1365-2141.2008.07441.x [23] Vernet M, Doyen C. Assessment of iron status with a new fully automated assay for transferrin receptor in human serum. Clin Chem Lab Med 2000; 38: 437-42. doi:10.1515/cclm.2000.064 [24] Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG. Clinical utility of the soluble transferrin receptor and comparison with serum ferritin in several populations. Clin Chem 1998; 44: 45-51. [25] Tanno T, Noel P, Miller JL. Growth differentiation factor 15 in erythroid health and disease. Curr Opin Hematol 2010; 17: 184-90. doi:10.1097/MOH.0b013e328337b52f [26] Tanno T, Bhanu NV, Oneal PA, Goh SH, Staker P, Lee YT, et al. High levels of GDF15 in thalassemia suppress expression of the iron regulatory protein hepcidin. Nat Med 2007; 13: 1096-101. doi:10.1038/nm1629 [27] Juneja SK, Imbert M, Sigaux F, Jouault H, Sultan C. Prevalence and distribution of ringed sideroblasts in primary myelodysplastic syndromes. J Clin Pathol 1983; 36: 566-9. [28] Visconte V, Rogers HJ, Singh J, Barnard J, Bupathi M, Traina F, et al. SF3B1 haploinsufficiency leads to formation of ring sideroblasts in myelodysplastic syndromes. Blood 2012; 120: 3173-86. doi:10.1182/blood-2012-05-430876 [29] Ambaglio I, Malcovati L, Papaemmanuil E, Laarakkers CM, Della Porta MG, Galli A, et al. Inappropriately low hepcidin levels in patients with myelodysplastic syndrome carrying a somatic mutation of SF3B1. Haematologica 2013; 98: 420-3. doi:10.3324/haematol.2012.077446 [30] Xu L, Gu ZH, Li Y, Zhang JL, Chang CK, Pan CM, et al. Genomic landscape of CD34+ hematopoietic cells in myelodysplastic syndrome and gene mutation profiles as prognostic markers. Proc Natl Acad Sci U S A 2014; 111: 8589-94. doi:10.1073/pnas.1407688111 [31] Xu F, Wu LY, Chang CK, He Q, Zhang Z, Liu L, et al. Whole-exome and targeted sequencing identify ROBO1 and ROBO2 mutations as progression-related drivers in myelodysplastic syndromes. Nat Commun 2015; 6: 8806. doi:10.1038/ncomms9806
Figure 1. Types and frequencies of SF3B1 mutations in patients with MDS-RS (n=52)
Figure 2. Distribution of the SF3B1-mutated and wild-type patients according to WHO morphologic subtypes or IPSS risk groups
A
SF3BI mutated
25
B
SF3BI mutated
30
SF3B1 wild-type
20
number of patients
number of patients
SF3BI wild-type
15 10 5 0
25 20 15 10 5 0
RARS
RCMD-RS RAEB1-RS RAEB2-RS
Low
Int-1
Int-2
High
Figure 3. Ultrastructural features of erythroid precursors from MDS-RS patients (A) Ultrastructure of erythroid precursors from SF3B1-mutated patients
A1-2: Abundant electron-dense granules in the mitochondria around the nucleus of polychromatophilic normoblast and orthochromatic normoblast
A3-4: Heavy destruction of mitochondria cristae and numerous cytoplasmic vacuoles in polychromatophilic normoblast
(B) Ultrastructure of erythroid precursors from SF3B1 wild-type patients
B1-2: Less amount of electron-dense deposits at lower intensity in the mitochondria of erythroblast, and fewer cytoplasmic vacuoles
Figure 4. Serum ferritin and hepcidin-25 levels in MDS-RS patients
A
P=0.107
B
4000 3000 2000 1000
300
200
100
0
0 SF3B1 mutated
C
SF3B1 mutated
SF3B1 wild-type
P=0.012
D serum hepcidin-25 (ng/ml)
0.8 hepcidin-25/ferritin ratio
P=0.028
400 serum hepcidin-25 (ng/ml)
serum ferritin (ng/ml)
5000
0.6
0.4
0.2
400
SF3B1 wild-type
SF3B1 wild-type SF3B1 mutated
300
200
100
0
0.0 SF3B1 mutated
SF3B1 wild-type
0
1000
2000 3000 4000 serum ferritin (ng/ml)
5000
(A) Comparison of serum ferritin according to SF3B1 mutational status (P=0.107) (B) Comparison of serum hepcidin-25 concentration according to SF3B1 mutational status (P=0.028) (C) Comparison of serum hepcidin-25/ferritin ratio according to SF3B1 mutational status (P=0.012) (D) Linear correlation between hepcidin and serum ferritin according to SF3B1 mutational status in patients with MDS-RS.
Figure 5. Indicators related with erythropoietic activity in MDS-RS patients
bone marrow erythroblasts(%)
A
P=0.036
100 80 60 40 20 0
SF3B1 mutated
B
soluble TfR (mg/L)
15
SF3B1 wild-type
P=0.132
10
5
0 SF3B1 mutated
C
SF3B1 wild-type
P<0.001
GDF 15 (pg/ml)
40000
30000
20000
10000
0 SF3B1 mutated
SF3B1 wild-type
(A) Comparison of the percentage of BM erythroblasts according to SF3B1 mutational status (P=0.036) (B) Comparison of soluble transferring receptor level according to SF3B1 mutational status (P=0.132) (C) Comparison of serum GDF-15 concentration according to SF3B1 mutational status (P<0.001)
Figure 6. Schematic comparison of erythropoietic activity and iron profile according to SF3B1 mutational status
22
Table 1. Clinical features of 52 patients with MDS-RS stratified by the presence or absence of the SF3B1 mutation 52 patients with MDS-RS
MDS-RS patients with SF3B1 mutations (N=33)
MDS-RS patients with wild-type SF3B1 (N=19)
P
63 (20-83)
64 (34-83)
61(20-81)
0.243
31 (60)
19 (58)
12 (63)
0.693
67 (32-96)
67 (44-96)
70 (32-83)
0.571
3.3 (1.1-20.8)
3.2 (1.1-20.8)
3.7 (1.7-10.8)
0.217
1.8 (0.4-16)
1.6 (0.4-16)
2 (0.4-4.9)
0.466
112 (13-365)
150 (17-365)
46 (13-334)
<0.001
BM blasts, %, median (range)
2.2 (0-17)
1.8 (0-15)
4.8 (1.5-17)
0.003
BM erythroblasts, %, median (range)
51 (16-85)
54 (16-85)
36.5 (17-80)
0.036
Ring sideroblasts, %, median (range)
41 (15.5-87)
43 (17.5-87)
26 (15.5-69)
0.026
Good
28 (54)
24 (73)
4 (21)
Intermediate
17 (33)
9 (27)
8 (42)
Poor
7 (13)
0 (0)
7 (37)
Low
14 (27)
12 (36)
2 (11)
Int-1
28 (54)
19 (58)
9 (47)
Int-2
8 (15)
2 (6)
6 (32)
High
2 (4)
0 (0)
2 (11)
RARS
23 (44)
19 (58)
4 (21)
RCMD
20 (38)
11 (33)
9 (47)
RAEB-1
6 (12)
2 (6)
4 (21)
RAEB -2
3 (6)
1 (3)
2 (10)
variable Median age, year (range) Male number (%) Hemoglobin, g/L, median (range) WBC, x109/L, median (range) 9
ANC, x10 /L, median (range) 9
Platelets, x10 /L, median (range)
Karyotype risk categories, n (%)
<0.001
IPSS risk categories, n (%)
0.002
WHO morphologic categories, n (%)
0.006
RBC transfusions at diagnosis, units (%) 0
17 (33)
9 (27)
<10
14 (27)
9 (27)
5 (26)
10-20
8 (15)
5 (15)
3 (16)
>20
9 (17)
7 (21)
2 (11)
Not available
4 (8)
3 (9)
1 (5)
23
8 (42)
0.209
Deaths, n (%)
32 (61)
18 (55)
14 (74)
0.172
WBC: white blood cell count; ANC: absolute neutrophil count; IPSS: International Prognostic Scoring System; WHO: World Health Organization; RARS: refractory anemia with ring sideroblasts; RCMD: refractory cytopenia with multilineage dysplasia; RAEB-1: refractory anemia with excess blasts-1; RAEB-2: refractory anemia with excess blasts-2.
24
Table 2. Survival analysis in 52 patients with MDS-RS Variable Age
Univariate analysis P value 0.631
Multivariable analysis P value
Gender
0.165
WHO morphologic category
<0.001
0.050
IPSS risk category
<0.001
0.041
Hemoglobin level
0.059
Bone Marrow erythroblasts %
0.219
Ring sideroblasts %
0.266
Absolute neutrophil count
0.773
Platelet count
0.168
Bone Marrow blasts %
<0.001
0.034
Karyotype category
0.005
0.408
SF3B1 mutational status
0.004
0.602
RBC transfusions at diagnosis
0.232
25
Variable, median (range)
P
Average optical density
MDS-RS patients with MDS-RS patients with SF3B1 mutations wild-type SF3B1 (N=6) (N=3) 104.7 (56.6-132.8) 81.4 (58.2-123.7)
0.154
Maximum density
184.0 (160-223)
104.5 (79-128)
<0.001
Area of electron-dense deposits
192.5 (91-479)
76.3 (21-126)
<0.001
Integrated optical density
9641.0 (6522-30826)
6054.5 (607-8321)
0.001
Table 3. Quantitative analysis of electron-dense deposits in mitochondria Note: In SF3B1-mutated or wild-type group, electron-dense deposits in thirty mitochondria were analyzed by Image-Pro Plus software
26
Table 4. Sequencing data of additional 10 targeted genes variable
MDS-RS patients with MDS-RS patients with SF3B1 mutations wild-type SF3B1 (N=18) (N=10) Patients with additional gene mutation, n (%) 7 (39%) 8 (80%) Patients (n) with additional gene mutation not related with poor outcome UPF3A/TET2/TDG 3 1 Patients (n) with additional gene mutation related with poor prognosis TP53 2 1 WT1 0 1 RUNX1 1 1 ASXL1 1 3 SRSF2 0 2 U2AF1 0 2 IDH 1 1 Note: Targeted gene sequencing was performed according to a recent report [31]
27
S0145-2126(16)30026-1 http://dx.doi.org/doi:10.1016/j.leukres.2016.02.011 LR 5546
To appear in:
Leukemia Research
Received date: Revised date: Accepted date:
29-10-2015 15-1-2016 25-2-2016
Please cite this article as: {http://dx.doi.org/ This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
SF3B1-mutated myelodysplastic syndrome with ring sideroblasts harbors more severe iron overload and corresponding over-erythropoiesis Yang Zhu1, Xiao Li1*, Chunkang Chang1, Feng Xu1, Qi He1, Juan Guo1, Ying Tao1, Yizhi Liu1, Li Liu1, Wenhui Shi1 Department of Haematology, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai, China 1
*
Correspondence: Xiao Li, M.D., PhD., Department of Haematology, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, 600 Yishan Road, Shanghai 200233, China. E-mail： [email protected]
Graphical abstract
Highlights: SF3B1-mutated MDS-RS harbored more severe iron overload and over-erythropoiesis. The presence of the SF3B1 mutation was associated with a better overall survival. IPSS risk groups primarily accounted for prognostic value of the SF3B1 mutation.
Abstract Objective: To clarify the possible biological differences and implication of the SF3B1 gene for patients with MDS-RS (myelodysplastic syndromes with ring sideroblasts). Methods: Sanger sequencing was performed on mutation hotspots of the SF3B1 gene in MDS-RS patients. The differences between the SF3B1 mutated and wild-type subsets, including the ultrastructure of erythroid precursors, iron profile parameters, erythropoiesis-related measurements, as well as clinical features, were analyzed. Results: SF3B1 mutations were detected in 33 out of fifty-two MDS-RS patients (63%). The vast majority of patients with mutations (94%) were categorized in the lower risk group according to the IPSS (International Prognostic Scoring System), in contrast to only fifty-eight percent of the wild-type cases. In addition to the notably higher percentages of erythroblasts and ring sideroblasts in patients with mutations, abundant electron-dense granules in the mitochondria of the erythroid precursors were clearly observed. Moreover, patients with mutations presented both improper iron uptake and distribution (lower serum hepcidin-25 concentration, P=0.028) and enhanced erythropoietic activity (higher soluble transferrin receptor level, P=0.132; higher growth differentiation factor 15 concentration, P<0.001). Finally, MDS-RS patients carrying SF3B1 mutations had a better overall survival (median 38 vs. 18 months, P=0.001) compared to those without mutations. By multivariable analysis, the prognostic significance of the SF3B1 mutation was primarily accounted for by IPSS risk categorization. Conclusion: MDS-RS patients carrying SF3B1 mutations harbored a more severe iron overload and corresponding over-erythropoiesis. The better overall survival of SF3B1-mutated MDS-RS patients may be mainly due to the clustering of patients with lower risk disease in this group. Keywords: myelodysplastic syndrome; ring sideroblasts; SF3B1 mutation; iron overload; erythropoiesis. 1. Introduction The myelodysplastic syndromes (MDSs) are a group of heterogeneous diseases characterized by dysplasia, ineffective hematopoiesis, and a variable risk of progression to acute myeloid leukemia [1,2]. In addition to marrow dysplasia, increased ring sideroblasts (RS) are a key pathological feature of MDS [2,3]. The terminology MDS with ring
sideroblasts (MDS-RS) was introduced as an entity that is morphologically characterized by dysplasia and the presence of 15% or higher RS in marrow aspirate smears. Though MDS-RS only accounts for 20-25% of all MDS patients, most are accompanied by severe anemia and iron overload, which exacerbate the difficulties in clinical treatment [4]. Until now, there has been only limited research on the pathogenesis of MDS-RS. With the application of high-throughput whole genome sequencing, somatic mutations of SF3B1 (splicing factor 3b subunit 1) had recently been identified in 60%～80% of MDS-RS patients, compared to a frequency of mutations of less than 10% in other myeloid neoplasms [5-7]. In addition, recent studies indicated that the SF3B1 mutation is associated with decreased expression of the mitochondrial iron transporter ABCB7 [8,9]. Although the SF3B1 mutations were considered to be the primary cause of the distinctive phenotype of MDS-RS, there are still approximately 30%-40% of MDS-RS patients who are SF3B1 wild type. The presence of RS arises from a defect of mitochondrial iron metabolism in erythroid precursors. Iron homeostasis in erythroblasts is damaged in MDS-RS patients, which result in an alteration of systemic iron metabolism. It has been hypothesized that differences in the mitochondrial iron phenotype and systemic iron profile may exist between SF3B1 mutated MDS-RS and their mutation-negative counterparts. Moreover, there are inconsistencies in the prognostic value of SF3B1 mutations. Some studies showed that SF3B1 mutations were associated with a benign phenotype, whereas others reported that there was no additional prognostic value of SF3B1 mutations [6,10-13]. In the present study, we investigated the biological differences (iron profile, erythropoiesis and clinical features) of MDS-RS patients carrying the SF3B1 mutation and the wild-type allele. This research lays the foundation for clarifying the biological differences between two subgroups of MDS-RS.
2. Patients and Methods 2.1. Subjects and samples The study was approved by the institutional ethics committee, and the patient-relevant research strictly abided by the ethical standards laid down in the Declaration of Helsinki. Between November 2008 and March 2014, fifty-two patients with MDS-RS who were initially
diagnosed at the institute were enrolled. At the time of diagnosis, serum and bone marrow samples were collected after obtaining informed written consent. We excluded persons with secondary acquired sideroblastic anemias including alcohol abuse, lead toxicity, copper or pyridoxine deficiency, isoniazid therapy and hereditary sideroblastic anemias. On aspirate smears stained with Prussian blue, at least 15% of erythroid precursors were ring sideroblasts. Ring sideroblasts were defined as erythroblasts in which a minimum of five siderotic granules covered at least one-third of the nuclear circumference [14]. According to the 2001 and 2008 World Health Organization (WHO) classification criteria, 23 patients had refractory anemia with ring sideroblasts (RARS), 20 patients had refractory cytopenia with multilineage dysplasia and ring sideroblasts (RCMD-RS), 6 patients had refractory anemia with excess blasts-1 (RAEB-1), and 3 patients had refractory anemia with excess blasts-2 (RAEB-2). The patients were categorized into four risk groups (low, 0; INT-1, 0.5-1.0; INT-2, 1.5-2.0; and high, ≥2.5) defined according to the International Prognostic Scoring System (IPSS) [15]. The treatment details were obtained from the patients’ medical records. Fifty-nine percent of MDS-RS patients (31/52) had received RBC transfusion support (Table 1). Lenalidomide, iron chelators, hypomethylating agents and chemotherapy drugs were not administered. Follow-up for survival analysis ended in December 2014.
2.2. DNA extraction and SF3B1 mutational analysis The mononuclear cells were isolated from bone marrow aspirates of MDS-RS patients by density-gradient centrifugation using Ficoll (Lymphosep, Biowest, France). Genomic DNA was extracted using a DNA Extraction Kit (TianGen Biotech, Beijing, China) according to the manufacturer’s instructions. SF3B1-coding exon sequencing was performed using a DNA aliquot derived from bone marrow mononuclear cells. Genomic DNA corresponding to the mutation hotspots of SF3B1 (exons 13–16) were amplified by PCR. The primers are as follows: exon13-F, 5'-TTCGTCCCTTGATTAACAAAAGTC-3', exon 13-R, 5'-TCAGGTCTCATGGTAGAGATCA-3', exon14-F, 5'-TCTGTTTATGGAATTGATTAT G-3', exon14-R, 5'- TCTAGTCCCAACTACTAAGGAGGC-3', exon15.16-F, 5'-TTATCTGC TGACAGGCTATGGTTC-3', and exon15.16-R, 5'-ATCAACTGACCTGAAATGAAGAG-3'. All products were confirmed by 1% agarose gel electrophoresis and sequenced using an
ABI 3730xl DNA analyzer (Applied Biosystems) according to standard techniques. Sequencing data were analyzed using Mutation Surveyor (DNA Variant Analysis Software, Softgenetics) followed by visual inspection of sequencing traces.
2.3. Cytogenetic analysis Chromosomal specimens were prepared by an instant manipulation and a short-term culture (24–48h) of bone marrow cells. Analysis of the karyotype was performed in at least 20 metaphase cells by the G-banding technique. Karyotypes were described according to the International System of Human Cytogenetic Nomenclature (ISCN) [16]. Karyotype anomalies were defined in three major risk categories (good risk, intermediate risk, poor risk) according to the International Myelodysplastic Syndrome Working Group [15].
2.4. Transmission electron microscopy Ten million BM mononuclear cells were first fixed in 2% glutaraldehyde for more than 2 hours. Fixed samples were washed twice in 0.1M phosphate buffered saline (PBS) for 20 minutes each. Specimens were post-fixed in 1% osmium tetroxide for 2 hours. After a second fixation step, specimens were washed three times in water solution for 20 minutes each. After the final wash, specimens were dehydrated in ascending graded alcohols (30% alcohol, ten minutes→50% alcohol, ten minutes→70% alcohol and 3% uranyl acetate, overnight→80% alcohol, ten minutes→95% alcohol, ten minutes→100% alcohol, ten minutes), followed by immersion in propylene oxide, passage through ascending graded mixtures of propylene oxide and epoxy resin 618, and finally embedding in pure epoxy resin. Specimen blocks were polymerized at 60°C for a period of 48 hours. Ultrathin sections were cut at 60-80 nm by a LKB v-shaped ultra-thin slicing machine and supported on 200 mesh copper grids. Sections were stained in lead citrate and examined using a PHILIPS CM-120 transmission electron microscope. Electron micrographs were taken using a Gatan Orius CCD digital camera.
2.5. Iron metabolism-related assays Hepcidin-25 (25 amino acid peptide) is the active form of the peptide hormone hepcidin, a
key player in the regulation of iron metabolism. To investigate the status of iron metabolism, the levels of serum hepcidin-25 in MDS-RS patients were measured. Reliable quantitative (second-generation) hepcidin assays have been developed in recent years [17,18]. Of the quantitative hepcidin assays, the enzyme immunoassay kit of Bachem was a competitive immunoassay that had been evaluated to differentiate the hepcidin concentrations in serum samples of patients and controls with various iron loads [17,19-21]. Serum hepcidin-25 concentrations were determined using a commercial enzyme immunoassay kit following manufacturer’s protocol (Bachem, San Carlos, USA). Control serum samples were obtained from forty healthy individuals with a median age of 58 years (range, 26-81 years). The reference range of serum hepcidin-25 is from 16.9 to 52.7 ng/ml (median, 31.2 ng/ml). The serum iron and total iron binding capacity were determined using the ferrous-oxazine chromogenic method. Serum ferritin (SF) was determined by nephelometric assay. 2.6. Measurement of serum growth differentiation factor 15 and soluble transferrin receptor The concentration of growth differentiation factor 15 (GDF15) in serum samples was evaluated using an enzyme-linked immunosorbent assay (ELISA) for human GDF15 (R&D Systems, Minneapolis, MN, USA) [22]. Soluble transferrin receptor (sTfR) was quantified using an immunonephelometric method (Tina-quant Soluble Transferrin Receptor, Roche Diagnostics GmbH, Germany) [23].
2.7. Statistical analysis Numerical variables are presented as the median and range; categorical variables are described as the count and relative frequency (%) of subjects in each category. Comparison of numerical variables between groups was performed using a nonparametric approach (Mann–Whitney test or Kruskal–Wallis ANOVA). The distribution of categorical variables in different groups was compared with either the Fisher exact test (when computationally feasible) or the χ2 test. Survival was calculated from the date of diagnosis to death from any cause or last follow-up. Survival analyses were carried out with the Kaplan–Meier method. Multivariable survival analyses were performed by Cox proportional hazards regression. P
values less than 0.05 (two-sided) were considered statistically significant. Analyses were performed with the SPSS software package, version 18.0.
3. Results 3.1. Frequencies and distribution of SF3B1 mutations A total of 52 patients were diagnosed with MDS-RS (≥15% of ring sideroblasts were identified in bone marrow smear). Heterozygous point mutations of SF3B1 were detected in 33 subjects (63%) including the K700E point mutation in 18, K666T/N in 6, H662Q in 4, E622D in 3, R625L in 1 and D781G in 1 case (Figure 1). All missense mutations leading to amino acid substitutions clustered in exons 14 and 15. The frequencies of the SF3B1 mutation were 19 out of 23 (83%) for RARS, 11 out of 20 (55%) for RCMD-RS, 2 out of 6 (33%) for RAEB-1, and 1 out of 3 (33%) for RAEB-2. The proportion of subjects with SF3B1 mutations was significantly higher in the RARS and RCMD-RS than that in other MDS-RS subgroups (Figure 2A). According to the IPSS risk groups, the frequencies of the SF3B1 mutation were 12 out of 14 (86%) for the low group, 19 out of 28 (68%) for the INT-1 group, 2 out of 8 (25%) for the INT-2 group, and 0 out of 2 (0%) for the high group (Figure 2B).
3.2. Clinical features Subjects with the SF3B1 mutation had a higher percentage of bone marrow erythroblasts (median, 54.0% vs. 36.5%; P=0.036) and ring sideroblasts (median, 43% vs. 26%; P=0.026). Patients carrying SF3B1 mutations had a higher platelet counts (median, 150×109/L vs. 46×109/L; P<0.001) and a lower percentage of BM blast cells (median, 1.8% vs. 4.8%; P=0.003). In addition, the proportion of good risk IPSS cytogenetics in subjects with the SF3B1 mutation was significantly higher than in wild-type patients (73% vs. 21%). The vast majority of SF3B1 mutated patients were categorized into the lower IPSS risk groups (94% cases with IPSS ≤1.0 vs. 58% in wild-type patients) (Table 1). Survival analysis was conducted for all enrolled MDS-RS subjects with a median follow-up of 18 months (range, 6–54 months). Subjects with SF3B1 mutations had a better overall survival (median, 38 months (95% CI, 26–48 months) vs. 18 months (95% CI, 16–20 months); P=0.001).
In univariate analysis, survival was predicted according to the WHO morphologic category (P<0.001), IPSS (P<0.001), percentage of BM blasts (P<0.001), karyotype (P=0.005), and SF3B1 mutational status (P=0.004). In multivariable analysis, only the IPSS and percentage of BM blasts remained significant (Table 2). Moreover, the prognostic significance of SF3B1 mutation was lost during multivariable analysis that included, as a covariate, the percentage of BM blasts, WHO morphologic categorization or IPSS risk categorization; the P values were 0.313, 0.116 and 0.685, respectively.
3.3. Ultrastructure of erythroid precursors Although Prussian blue staining is traditionally used to identify RS in a bone marrow smear, this procedure does not distinguish microstructural changes of mitochondria. Using transmission electron microscopy, ultrastructure observations were performed on BM cells derived from six patients carrying the SF3B1 mutation (4 RARS, 2 RCMD-RS) and three patients wild-type for SF3B1 (1 RARS, 2 RCMD-RS). The predominant ultrastructural alterations in SF3B1-mutated patients were the abundant electron-dense deposits in the mitochondria around the nucleus of the erythroblasts. A great number of these mitochondria cristae were completely destroyed. In addition, numerous cytoplasmic vacuoles and membrane blebbing were observed. In contrast, sparse deposits and fewer cytoplasmic vacuoles were shown in wild-type patients (Figure 3). Quantitative analyses of iron-laden granules in mitochondria were digitalized using Image-Pro Plus software. The density values and area size of electron-dense deposits in cases with mutations were significantly higher than that of wild-type patients (Table 3). 3.4. Serum hepcidin-25 and ferritin levels Because of the different mitochondrial iron pattern in SF3B1 mutated patients, it is worth exploring whether there are differences in the iron profile parameters of SF3B1 mutated and wild-type MDS-RS cases. As illustrated in Figure 4, a higher serum ferritin level was observed in SF3B1-mutated patients (median, 1088.9 ng/ml vs. 575.7 ng/ml; P=0.107). However, patients carrying SF3B1 mutations have a notably lower serum hepcidin-25 concentration than wild-type patients (median, 17.3 ng/ml vs. 77.8 ng/ml, P=0.028). The hepcidin-25 to ferritin ratio
represents a measure of the adequacy of hepcidin levels relative to body iron stores. Significantly lower values of the hepcidin-25 to ferritin ratio were found in patients with SF3B1 mutation (median, 0.017 vs. 0.091; P=0.006). The linear correlation between hepcidin-25 and serum ferritin according to the SF3B1 mutational status in MDS-RS patients suggests that the hepcidin level did not increase proportionally compared to the body iron stores in SF3B1-mutated patients. These data implied that the degree of iron load is more severe in MDS-RS patients carrying the SF3B1 mutation.
3.5. Erythropoiesis profile measurements The proportion of marrow erythroblasts is associated with the level of erythropoietic activity. The level of serum sTfR is closely related to cellular iron demands and the erythroid proliferation rate [24]. GDF15 is released at sites of active erythropoiesis. The expression of GDF-15 is at low levels during normal erythropoiesis, but at high levels in patients with ineffective erythropoiesis [25,26]. The bone marrow differential cell count showed that the proportion of erythroblasts in SF3B1-mutated patients was much higher than that of wild-type patients (median, 54.0% vs. 36.5%; P=0.036; Figure 5A). Moreover, SF3B1-mutated patients had a higher serum sTfR level (median, 5.4 mg/L vs. 3.3 mg/L; P=0.132; Figure 5B) and GDF15 concentration (median, 5856.3 pg/ml vs. 848.6 pg/ml; P<0.001; Figure 5C). These data indicated that the erythroid marrow function in SF3B1-mutated cases showed both more active activity and ineffective erythropoiesis. 3.6. General trend in iron profile and erythropoietic activity The amount of RBC transfusions in SF3B1-mutated MDS-RS patients was not statistically more than that in wild-type cases (Table 1), which indicated that there was no significant difference in transfusional iron burden between these subgroups. As shown in Figure 6, SF3B1-mutated patients had higher iron profile measurements, including ferritin，serum iron and transferrin saturation. Moreover, the measurements related to erythropoietic activity were much higher in SF3B1-mutated cases. However, the serum hepcidin-25 concentration and hepcidin-25 to ferritin ratio of SF3B1-mutated patients were markedly lower than that of
wild-type cases.
4. Discussion In recent years, high-throughput sequencing has revealed that RNA splicing pathway genes were frequently mutated (～45 to ～85%) in patients with MDS and that most of the mutations occurred in a mutually exclusive manner [5]. These mutations are considered to be highly associated with the pathogenesis of these diseases and have become a hotspot of research. In this study of fifty-two patients with MDS-RS, recurrent mutations in SF3B1, a gene encoding a core component of the RNA splicing machinery, were identified in thirty-three subjects (63%). All observed mutations were missense substitutions. Mutations clustered in six codons of exon 14 and 15, and codon 700 was the most frequent mutation spot (54.5% of mutated cases). Analysis of the results of the SF3B1 mutation were consistent with the data reported previously that indicated a significant association of SF3B1 mutations with the presence of ring sideroblasts in MDS. Ringed sideroblasts are characteristic of refractory anemia with ring sideroblasts (RARS), but they are frequently observed in other subtypes of MDS. The presence of ringed sideroblasts has previously been observed in 40% of patients with refractory anemia with excess blasts (RAEB) and greater than 20% of RS in one-fifth of such cases [27]. We investigated the clinical correlates of SF3B1 mutations in patients with MDS-RS. The lower proportion of marrow blast cells (median 1.8% vs. 4.8%, P=0.003) and higher proportion of good or intermediate risk karyotypes (100% vs. 63%, P<0.001) were found in SF3B1 mutated patients. The vast majority of SF3B1-mutated patients were classified into IPSS ≤ 1.0 groups (94% vs. 58%). In addition, the median survival of SF3B1-mutated patients was significantly longer than that of wild-type patients (38 months vs. 18 months). A recent report described coarse iron deposits in erythroid mitochondria of a RARS patient carrying a SF3B1 mutation, and low hepcidin levels in SF3B1-mutated MDS patients were reported in other study [28,29]. In the present study, we confirmed the findings of earlier studies and focused on the iron profile and erythropoiesis in one population characterized by MDS and increased ring sideroblasts. As illustrated in Figure 3 and Table 3, large amounts of electron-dense granules were deposited in the mitochondria of erythroid precursors from
SF3B1-mutated patients. By image analysis with Image-pro Plus software, the density of the granules was significantly increased compared with those from patients without mutation. The ultrastructural alterations suggested that the trafficking of iron molecules between mitochondria and cytoplasm was disturbed more severely in SF3B1-mutated patients. In addition, we analyzed the biochemical indexes related to iron metabolism. The results showed that the serum ferritin level of SF3B1-mutated patients was higher than that of patients
without
the
mutation,
and
the
serum
hepcidin-25
concentration
and
hepcidin-25/ferritin ratio was statistically lower than those of wild-type patients. This indicated that there was a heavier iron load in SF3B1-mutated patients. Due to the decrease in the hepcidin-25 concentration, increased absorption of iron in the intestine is still present. As illustrated in Figure 4, when the serum ferritin level rose in wild-type patients, the hepcidin-25 concentration increased correspondingly to inhibit iron absorption. However, in SF3B1-mutated patients, even if the level of serum ferritin rose, the hepcidin-25 concentration could not increase proportionally, resulting in an iron overload in the body. The results of erythropoiesis-related indicators were also analyzed. The proportion of marrow erythroblasts in SF3B1-mutated patients was significantly higher compared to wild-type patients. This finding suggested that the erythroid precursors multiplied markedly and that erythropoiesis was exuberant in SF3B1-mutated patients. We also found statistically elevated sTfR and GDF-15 levels in patients with mutations. With regard to cytokines, there was more vigorous erythropoietic activity in SF3B1-mutated patients. Earlier studies [25,26] reported that serum GDF15 levels was markedly increased in hematological diseases with ineffective erythropoiesis and that an elevated GDF15 concentration suppressed hepcidin synthesis in primary human hepatocytes. We concluded that more active erythropoietic activity diminished the hepcidin response to increased iron burden in MDS-RS patients with SF3B1 mutations. Taken together, a possible mechanism of pathogenesis for the more active iron uptake and erythropoiesis in SF3B1-mutated MDS-RS may be a consequence of the under-utilization of iron and iron deposition in mitochondria. Abnormal iron metabolism of erythroblasts may have promoted the erythropoiesis-related cytokines through a feedback loop to enhance erythropoiesis. This futile effort (increased levels cytokines acting on under-utilized iron) finally lead to the clinical phenotype of iron overload and severe anemia.
The observation that MDS-RS patients with the SF3B1 mutation had a better overall survival compared to wild-type patients was explored. This study showed SF3B1 mutations clustered with good risk karyotypes (73% vs. 21%). Twenty-eight out of the 52 MDS-RS patients were enrolled in another research study in which an additional 10 targeted genes were sequenced (unpublished data). Additional mutations were found in 80% (8/10) of wild-type patients, compared to 39% (7/18) of SF3B1-mutated patients. The incidence rate of poor prognosis-related gene mutations in wild-type patients was higher than that in SF3B1-mutated patients (Table 4). There was a possibility of coexistence of a founding clone and subclone in hematopoietic stem cells of MDS [30]. Therefore, further study is warranted to define whether MDS-RS SF3B1 wild-type is a later stage caused by the evolution of more aggressive clone.
5. Conclusion In this study, SF3B1 mutations were detected in 33 out of fifty-two MDS-RS patients (63%). Compared to wild-type MDS-RS, patients carrying the SF3B1 mutation harbored more severe iron overload and corresponding over-erythropoiesis. The presence of the SF3B1 mutation was associated with a better overall survival. By multivariable analysis, the prognostic significance of the SF3B1 mutation was primarily accounted for by IPSS risk categorization.
Further
study
is
warranted
to
determine
whether
the
SF3B1
mutation-positive MDS-RS and mutation-negative allele are two distinct subtypes of the disease or just different stages of the same MDS subset.
References [1] Cazzola M, Malcovati L. Myelodysplastic syndromes--coping with ineffective hematopoiesis. N Engl J Med 2005; 352: 536-8. doi:10.1056/NEJMp048266 [2] Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189-99. [3] Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002; 100: 2292-302. doi:10.1182/blood-2002-04-1199 [4] Malcovati L, Porta MG, Pascutto C, Invernizzi R, Boni M, Travaglino E, et al. Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making. J Clin Oncol 2005; 23: 7594-603. doi:10.1200/jco.2005.01.7038
[5] Yoshida K, Sanada M, Shiraishi Y, Nowak D, Nagata Y, Yamamoto R, et al. Frequent pathway mutations of splicing machinery in myelodysplasia. Nature 2011; 478: 64-9. doi:10.1038/nature10496 [6] Papaemmanuil E, Cazzola M, Boultwood J, Malcovati L, Vyas P, Bowen D, et al. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med 2011; 365: 1384-95. doi:10.1056/NEJMoa1103283 [7] Visconte V, Makishima H, Jankowska A, Szpurka H, Traina F, Jerez A, et al. SF3B1, a splicing factor is frequently mutated in refractory anemia with ring sideroblasts. Leukemia 2012; 26: 542-5. doi:10.1038/leu.2011.232 [8] Boultwood J, Pellagatti A, Nikpour M, Pushkaran B, Fidler C, Cattan H, et al. The role of the iron transporter ABCB7 in refractory anemia with ring sideroblasts. PloS one 2008; 3: e1970. doi:10.1371/journal.pone.0001970 [9] Nikpour M, Scharenberg C, Liu A, Conte S, Karimi M, Mortera-Blanco T, et al. The transporter ABCB7 is a mediator of the phenotype of acquired refractory anemia with ring sideroblasts. Leukemia 2013; 27: 889-96. doi:10.1038/leu.2012.298 [10] Malcovati L, Papaemmanuil E, Bowen DT, Boultwood J, Della Porta MG, Pascutto C, et al. Clinical significance of SF3B1 mutations in myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms. Blood 2011; 118: 6239-46. doi:10.1182/blood-2011-09-377275 [11] Patnaik MM, Lasho TL, Hodnefield JM, Knudson RA, Ketterling RP, Garcia-Manero G, et al. SF3B1 mutations are prevalent in myelodysplastic syndromes with ring sideroblasts but do not hold independent prognostic value. Blood 2012; 119: 569-72. doi:10.1182/blood-2011-09-377994 [12] Bejar R, Stevenson KE, Caughey BA, Abdel-Wahab O, Steensma DP, Galili N, et al. Validation of a prognostic model and the impact of mutations in patients with lower-risk myelodysplastic syndromes. J Clin Oncol 2012; 30: 3376-82. doi:10.1200/jco.2011.40.7379 [13] Damm F, Thol F, Kosmider O, Kade S, Loffeld P, Dreyfus F, et al. SF3B1 mutations in myelodysplastic syndromes: clinical associations and prognostic implications. Leukemia 2012; 26: 1137-40. doi:10.1038/leu.2011.321 [14] Mufti GJ, Bennett JM, Goasguen J, Bain BJ, Baumann I, Brunning R, et al. Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts. Haematologica 2008; 93: 1712-7. doi:10.3324/haematol.13405 [15] Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997; 89: 2079-88. [16] Gonzalez Garcia JR, Meza-Espinoza JP. Use of the International System for Human Cytogenetic Nomenclature (ISCN). Blood 2006; 108: 3952-3. doi:10.1182/blood-2006-06-031351 [17] Kroot JJ, Tjalsma H, Fleming RE, Swinkels DW. Hepcidin in human iron disorders: diagnostic implications. Clin Chem 2011; 57: 1650-69. doi:10.1373/clinchem.2009.140053 [18] Kemna EH, Tjalsma H, Willems HL, Swinkels DW. Hepcidin: from discovery to differential diagnosis. Haematologica 2008; 93: 90-7. doi:10.3324/haematol.11705 [19] Castagna A, Campostrini N, Zaninotto F, Girelli D. Hepcidin assay in serum by SELDI-TOF-MS and other approaches. J Proteomics 2010; 73: 527-36. doi:10.1016/j.jprot.2009.08.003
[20] Kroot JJ, Laarakkers CM, Geurts-Moespot AJ, Grebenchtchikov N, Pickkers P, van Ede AE, et al. Immunochemical and mass-spectrometry-based serum hepcidin assays for iron metabolism disorders. Clin Chem 2010; 56: 1570-9. doi:10.1373/clinchem.2010.149187 [21] Kroot JJ, Kemna EH, Bansal SS, Busbridge M, Campostrini N, Girelli D, et al. Results of the first international round robin for the quantification of urinary and plasma hepcidin assays: need for standardization. Haematologica 2009; 94: 1748-52. doi:10.3324/haematol.2009.010322 [22] Ramirez JM, Schaad O, Durual S, Cossali D, Docquier M, Beris P, et al. Growth differentiation factor 15 production is necessary for normal erythroid differentiation and is increased in refractory anaemia with ring-sideroblasts. Br J Haematol 2009; 144: 251-62. doi:10.1111/j.1365-2141.2008.07441.x [23] Vernet M, Doyen C. Assessment of iron status with a new fully automated assay for transferrin receptor in human serum. Clin Chem Lab Med 2000; 38: 437-42. doi:10.1515/cclm.2000.064 [24] Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG. Clinical utility of the soluble transferrin receptor and comparison with serum ferritin in several populations. Clin Chem 1998; 44: 45-51. [25] Tanno T, Noel P, Miller JL. Growth differentiation factor 15 in erythroid health and disease. Curr Opin Hematol 2010; 17: 184-90. doi:10.1097/MOH.0b013e328337b52f [26] Tanno T, Bhanu NV, Oneal PA, Goh SH, Staker P, Lee YT, et al. High levels of GDF15 in thalassemia suppress expression of the iron regulatory protein hepcidin. Nat Med 2007; 13: 1096-101. doi:10.1038/nm1629 [27] Juneja SK, Imbert M, Sigaux F, Jouault H, Sultan C. Prevalence and distribution of ringed sideroblasts in primary myelodysplastic syndromes. J Clin Pathol 1983; 36: 566-9. [28] Visconte V, Rogers HJ, Singh J, Barnard J, Bupathi M, Traina F, et al. SF3B1 haploinsufficiency leads to formation of ring sideroblasts in myelodysplastic syndromes. Blood 2012; 120: 3173-86. doi:10.1182/blood-2012-05-430876 [29] Ambaglio I, Malcovati L, Papaemmanuil E, Laarakkers CM, Della Porta MG, Galli A, et al. Inappropriately low hepcidin levels in patients with myelodysplastic syndrome carrying a somatic mutation of SF3B1. Haematologica 2013; 98: 420-3. doi:10.3324/haematol.2012.077446 [30] Xu L, Gu ZH, Li Y, Zhang JL, Chang CK, Pan CM, et al. Genomic landscape of CD34+ hematopoietic cells in myelodysplastic syndrome and gene mutation profiles as prognostic markers. Proc Natl Acad Sci U S A 2014; 111: 8589-94. doi:10.1073/pnas.1407688111 [31] Xu F, Wu LY, Chang CK, He Q, Zhang Z, Liu L, et al. Whole-exome and targeted sequencing identify ROBO1 and ROBO2 mutations as progression-related drivers in myelodysplastic syndromes. Nat Commun 2015; 6: 8806. doi:10.1038/ncomms9806
Figure 1. Types and frequencies of SF3B1 mutations in patients with MDS-RS (n=52)
Figure 2. Distribution of the SF3B1-mutated and wild-type patients according to WHO morphologic subtypes or IPSS risk groups
A
SF3BI mutated
25
B
SF3BI mutated
30
SF3B1 wild-type
20
number of patients
number of patients
SF3BI wild-type
15 10 5 0
25 20 15 10 5 0
RARS
RCMD-RS RAEB1-RS RAEB2-RS
Low
Int-1
Int-2
High
Figure 3. Ultrastructural features of erythroid precursors from MDS-RS patients (A) Ultrastructure of erythroid precursors from SF3B1-mutated patients
A1-2: Abundant electron-dense granules in the mitochondria around the nucleus of polychromatophilic normoblast and orthochromatic normoblast
A3-4: Heavy destruction of mitochondria cristae and numerous cytoplasmic vacuoles in polychromatophilic normoblast
(B) Ultrastructure of erythroid precursors from SF3B1 wild-type patients
B1-2: Less amount of electron-dense deposits at lower intensity in the mitochondria of erythroblast, and fewer cytoplasmic vacuoles
Figure 4. Serum ferritin and hepcidin-25 levels in MDS-RS patients
A
P=0.107
B
4000 3000 2000 1000
300
200
100
0
0 SF3B1 mutated
C
SF3B1 mutated
SF3B1 wild-type
P=0.012
D serum hepcidin-25 (ng/ml)
0.8 hepcidin-25/ferritin ratio
P=0.028
400 serum hepcidin-25 (ng/ml)
serum ferritin (ng/ml)
5000
0.6
0.4
0.2
400
SF3B1 wild-type
SF3B1 wild-type SF3B1 mutated
300
200
100
0
0.0 SF3B1 mutated
SF3B1 wild-type
0
1000
2000 3000 4000 serum ferritin (ng/ml)
5000
(A) Comparison of serum ferritin according to SF3B1 mutational status (P=0.107) (B) Comparison of serum hepcidin-25 concentration according to SF3B1 mutational status (P=0.028) (C) Comparison of serum hepcidin-25/ferritin ratio according to SF3B1 mutational status (P=0.012) (D) Linear correlation between hepcidin and serum ferritin according to SF3B1 mutational status in patients with MDS-RS.
Figure 5. Indicators related with erythropoietic activity in MDS-RS patients
bone marrow erythroblasts(%)
A
P=0.036
100 80 60 40 20 0
SF3B1 mutated
B
soluble TfR (mg/L)
15
SF3B1 wild-type
P=0.132
10
5
0 SF3B1 mutated
C
SF3B1 wild-type
P<0.001
GDF 15 (pg/ml)
40000
30000
20000
10000
0 SF3B1 mutated
SF3B1 wild-type
(A) Comparison of the percentage of BM erythroblasts according to SF3B1 mutational status (P=0.036) (B) Comparison of soluble transferring receptor level according to SF3B1 mutational status (P=0.132) (C) Comparison of serum GDF-15 concentration according to SF3B1 mutational status (P<0.001)
Figure 6. Schematic comparison of erythropoietic activity and iron profile according to SF3B1 mutational status
22
Table 1. Clinical features of 52 patients with MDS-RS stratified by the presence or absence of the SF3B1 mutation 52 patients with MDS-RS
MDS-RS patients with SF3B1 mutations (N=33)
MDS-RS patients with wild-type SF3B1 (N=19)
P
63 (20-83)
64 (34-83)
61(20-81)
0.243
31 (60)
19 (58)
12 (63)
0.693
67 (32-96)
67 (44-96)
70 (32-83)
0.571
3.3 (1.1-20.8)
3.2 (1.1-20.8)
3.7 (1.7-10.8)
0.217
1.8 (0.4-16)
1.6 (0.4-16)
2 (0.4-4.9)
0.466
112 (13-365)
150 (17-365)
46 (13-334)
<0.001
BM blasts, %, median (range)
2.2 (0-17)
1.8 (0-15)
4.8 (1.5-17)
0.003
BM erythroblasts, %, median (range)
51 (16-85)
54 (16-85)
36.5 (17-80)
0.036
Ring sideroblasts, %, median (range)
41 (15.5-87)
43 (17.5-87)
26 (15.5-69)
0.026
Good
28 (54)
24 (73)
4 (21)
Intermediate
17 (33)
9 (27)
8 (42)
Poor
7 (13)
0 (0)
7 (37)
Low
14 (27)
12 (36)
2 (11)
Int-1
28 (54)
19 (58)
9 (47)
Int-2
8 (15)
2 (6)
6 (32)
High
2 (4)
0 (0)
2 (11)
RARS
23 (44)
19 (58)
4 (21)
RCMD
20 (38)
11 (33)
9 (47)
RAEB-1
6 (12)
2 (6)
4 (21)
RAEB -2
3 (6)
1 (3)
2 (10)
variable Median age, year (range) Male number (%) Hemoglobin, g/L, median (range) WBC, x109/L, median (range) 9
ANC, x10 /L, median (range) 9
Platelets, x10 /L, median (range)
Karyotype risk categories, n (%)
<0.001
IPSS risk categories, n (%)
0.002
WHO morphologic categories, n (%)
0.006
RBC transfusions at diagnosis, units (%) 0
17 (33)
9 (27)
<10
14 (27)
9 (27)
5 (26)
10-20
8 (15)
5 (15)
3 (16)
>20
9 (17)
7 (21)
2 (11)
Not available
4 (8)
3 (9)
1 (5)
23
8 (42)
0.209
Deaths, n (%)
32 (61)
18 (55)
14 (74)
0.172
WBC: white blood cell count; ANC: absolute neutrophil count; IPSS: International Prognostic Scoring System; WHO: World Health Organization; RARS: refractory anemia with ring sideroblasts; RCMD: refractory cytopenia with multilineage dysplasia; RAEB-1: refractory anemia with excess blasts-1; RAEB-2: refractory anemia with excess blasts-2.
24
Table 2. Survival analysis in 52 patients with MDS-RS Variable Age
Univariate analysis P value 0.631
Multivariable analysis P value
Gender
0.165
WHO morphologic category
<0.001
0.050
IPSS risk category
<0.001
0.041
Hemoglobin level
0.059
Bone Marrow erythroblasts %
0.219
Ring sideroblasts %
0.266
Absolute neutrophil count
0.773
Platelet count
0.168
Bone Marrow blasts %
<0.001
0.034
Karyotype category
0.005
0.408
SF3B1 mutational status
0.004
0.602
RBC transfusions at diagnosis
0.232
25
Variable, median (range)
P
Average optical density
MDS-RS patients with MDS-RS patients with SF3B1 mutations wild-type SF3B1 (N=6) (N=3) 104.7 (56.6-132.8) 81.4 (58.2-123.7)
0.154
Maximum density
184.0 (160-223)
104.5 (79-128)
<0.001
Area of electron-dense deposits
192.5 (91-479)
76.3 (21-126)
<0.001
Integrated optical density
9641.0 (6522-30826)
6054.5 (607-8321)
0.001
Table 3. Quantitative analysis of electron-dense deposits in mitochondria Note: In SF3B1-mutated or wild-type group, electron-dense deposits in thirty mitochondria were analyzed by Image-Pro Plus software
26
Table 4. Sequencing data of additional 10 targeted genes variable
MDS-RS patients with MDS-RS patients with SF3B1 mutations wild-type SF3B1 (N=18) (N=10) Patients with additional gene mutation, n (%) 7 (39%) 8 (80%) Patients (n) with additional gene mutation not related with poor outcome UPF3A/TET2/TDG 3 1 Patients (n) with additional gene mutation related with poor prognosis TP53 2 1 WT1 0 1 RUNX1 1 1 ASXL1 1 3 SRSF2 0 2 U2AF1 0 2 IDH 1 1 Note: Targeted gene sequencing was performed according to a recent report [31]
27