These two authors contributed equally to the report.
Clinical implications of SOCS1 methylation in myelodysplastic syndrome
Article first published online: 15 SEP 2006
British Journal of Haematology
Volume 135, Issue 3, pages 317–323, November 2006
How to Cite
Wu, S.-J., Yao, M., Chou, W.-C., Tang, J.-L., Chen, C.-Y., Ko, B.-S., Huang, S.-Y., Tsay, W., Chen, Y.-C., Shen, M.-C., Wang, C.-H., Yeh, Y.-C. and Tien, H.-F. (2006), Clinical implications of SOCS1 methylation in myelodysplastic syndrome. British Journal of Haematology, 135: 317–323. doi: 10.1111/j.1365-2141.2006.06293.x
- Issue published online: 28 SEP 2006
- Article first published online: 15 SEP 2006
- Received 30 May 2006; accepted for publication 17 July 2006
- myelodysplastic syndrome;
The suppressor of cytokine signalling-1 (SOCS1) protein is a tumour suppressor. Hypermethylation of SOCS1 gene, resulting in transcriptional silencing, is suggested to play an important role in cancer development. We sought to characterise SOCS1 methylation in primary myelodysplastic syndrome (MDS) and clarify its clinical implications. The methylation status of SOCS1 was analysed by methylation-specific polymerase chain reaction in 114 patients with primary MDS and serial studies were performed in 29 of them. SOCS1 methylation occurred in 54 patients (47·4%), and was more frequent in patients with high-risk MDS than in those with low-risk (52·6% vs. 25·8%, P = 0·011). SOCS1 methylation was closely associated with NRAS mutation (P = 0·010) and inversely associated with good-risk karyotype (P = 0·021). With a median follow-up of 17 months (range: 1–231 months), two patients acquired SOCS1 methylation during disease progression. In two patients, SOCS1 methylation present at diagnosis, disappeared after haematopoietic stem cell transplantation. Patients with SOCS1 methylation had a higher cumulative risk of leukaemic transformation than those without (55·8% vs. 27·7% at 3 years, P = 0·004). This difference remained significant within the subgroup of patients with high-risk MDS (67·3% vs. 45·1% at 3 years, P = 0·045). This is the first report to demonstrate the clinical relevance of SOCS1 methylation in MDS. It may play an important role in the pathogenesis of MDS, especially among patients with high-risk subtypes.
Myelodysplastic syndrome (MDS) is a clonal stem cell disorder characterised by dysplastic changes of one or more blood cell lineages and ineffective haematopoiesis, which result in cytopenia in the peripheral blood, but usually hypercellularity in the bone marrow (BM) (Galton, 1986). About 10–35% of MDS cases eventually transform to acute myeloid leukaemia (AML) (Geddes et al, 1990; Tien et al, 1994). Although about 50% of MDS patients harbour chromosomal abnormalities (Knapp et al, 1985; Jacobs et al, 1986; Tien et al, 1994), the pathogenesis of the development and progression of the disease remain to be defined.
The proliferation and differentiation of haematopoietic precursor cells are regulated by various cytokines (Lotem & Sachs, 2002). These cytokines act partly through activation of the Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathway (Coffer et al, 2000; Ravandi et al, 2002). Constitutive activation of JAK and STAT proteins is recurrently associated with oncogenic transformation (Coffer et al, 2000). The members of the suppressor of cytokine signalling (SOCS) family (SOCS1 to SOCS7 and CIS) are composed of a poorly conserved amino-terminal region, a central SH2 domain, and a SOCS box (Hilton et al, 1998). SOCS1 is a negative regulator of the JAK/STAT pathway (Yoshikawa et al, 2001). SOCS1-deficient mice died from a myeloproliferative disorder within the first 3 weeks of life (Naka et al, 1997; Starr et al, 1998). The expression of inducible SOCS1 is associated with tumour suppression activity (Rottapel et al, 2002). Aberrant methylation of the SOCS1 gene, which results in transcriptional silencing, was first demonstrated in 17 of 26 human hepatocellular carcinomas, and restoration of SOCS1 expression suppressed growth of tumour cells (Yoshikawa et al, 2001). Recently, inactivation of SOCS1 by hypermethylation was also documented in haematological malignancies, such as multiple myeloma (Galm et al, 2003; Chim et al, 2004a), non-Hodgkin lymphoma (Chim et al, 2004b), chronic myeloid leukaemia (Liu et al, 2003) and AML (Chen et al, 2003; Chim et al, 2004c; Watanabe et al, 2004a). Studies of SOCS1 methylation in MDS are limited (Brakensiek et al, 2005; Johan et al, 2005), and its clinical significance is unclear.
To investigate the role of SOCS1 methylation in the pathogenesis of MDS, the methylation status of the gene was analysed in patients with primary MDS. Serial studies were performed in some patients during the follow-up period.
Materials and methods
The methylation status of SOCS1 was studied in BM cells from 114 patients with primary MDS. The diagnosis and classification of MDS were made according to French–American–British (FAB) criteria (Bennett et al, 1982). There were 81 males and 33 females with a median age of 64 years (range: 7–86 years). Three of them were children younger than 18 years. Twenty-one patients had refractory anaemia (RA), 10 had RA with ring sideroblasts (RARS), 33 had RA with excess blasts (RAEB), 20 had RAEB in transformation (RAEB-T), 25 had chronic myelomonocytic leukaemia (CMML), and five had AML preceded by MDS (MDS–AML). Patients with antecedent haematological disease or therapy-related MDS were excluded. Twenty-nine patients had serial studies during clinical follow-up. This study was approved by the Institutional Ethics Review Board of our hospital.
Chromosome analyses were performed as described previously (Tien et al, 1994). BM cells were harvested directly or after 1–3 d of non-stimulated culture. Metaphase chromosomes were banded by using the trypsin-Giemsa technique and karyotyped according to the International System for Human Cytogenetic Nomenclature (ISCN, 1985).
Analysis of SOCS1 methylation
The methylation status of SOCS1 was analysed by methylation-specific polymerase chain reaction (PCR) as previously described (Herman et al, 1996; Chen et al, 2003). Mononuclear cells were isolated from BM aspirates by Ficoll-Hypaque gradient centrifugation without further purification. High-molecular-weight DNA was extracted. DNA (4 μg) in a volume of 40 μl was denatured by addition of 10 μl of 1 mol/l NaOH (final concentration 0·2 mol/l) for 10 min at 37°C. Hydroquinone (30 μl of 10 mmol/l) (Sigma, St Louis, MO, USA) and 520 μl of 1·5 mol/l sodium bisulphite (Sigma) at pH 5·0 were added and mixed, and the samples were covered with mineral oil and incubated at 50°C for 16 h. Modified DNA was purified using the Wizard DNA purification resin and Vacuum Manifold, according to the manufacturer's instructions (Promega, Madison, WI, USA), and then eluted into 100 μl of water. Final desulphonation was achieved by treatment with 50 μl of 1 mol/l NaOH (final concentration 0·3 mol/l) at room temperature for 5 min, followed by ethanol precipitation. DNA was resuspended in 45 μl of water and used immediately or stored at −20°C before use.
The bisulphite-modified DNA was amplified by PCR using either a methylation-specific or unmethylation-specific primer set as described previously (Yoshikawa et al, 2001; Chen et al, 2003). The methylation-specific primers were 5′-TTC GCG TGT ATT TTT AGG TCG GTC-3′ (sense) and 5′-CGA CAC AAC TCC TAC AAC GAC CG-3′ (antisense). The unmethylation-specific primers were 5′-TTA TGA GTA TTT GTG TGT ATT TTT AGG TTG GTT-3′ (sense) and 5′-CAC TAA CAA CAC AAC TCC TAC AAC AAC CA-3′ (antisense). PCR reactions were run in a final volume of 25 μl containing 300 ng bisulphite-treated DNA, 200 nmol/l deoxynucleotide triphosphate, 200 nmol/l of each primer, 1 U of AmpliTaq Gold polymerase and buffer (Applied Biosystems, Foster City, CA, USA). PCR was carried out by heating at 95°C for 10 min, followed by 40 cycles of 95°C for 30 s, 65°C for 30 s, and 72°C for 30 s, with a final step for 10 min at 72°C. Normal DNA and distilled water were used as controls in each experiment. The hepatoma cell lines Hep3B and SNU423 were used as positive controls; the former had an amplified band on PCR with methylation-specific primers but not with unmethylation-specific primers, and the latter had positive bands in both conditions. The results were confirmed by repeating the analysis at least twice. With this method, SOCS1 methylation was not detected in BM cells from five transplantation donors and peripheral blood cells from 10 healthy persons.
Identification of RAS mutations
Point mutations at codons 12, 13 (exon 1) and 61 (exon 2) of NRAS and KRAS were analysed by PCR using genomic DNA and direct sequencing as previously described (Chou et al, 2006).
Chi-squared or Fisher's exact test was used to compare the categorical variables. The Mann–Whitney non-parametric test was used in comparisons of scale measures. Survival curves were plotted by using the Kaplan–Meier method; differences between curves were analysed by using the log-rank test. The Cox proportional hazard regression model was used for adjustment of interaction between factors. All statistical analyses were performed by using the Statistical Package for the Social Sciences, version 13.0 for Windows (SPSS, Chicago, IL, USA). P-values of less than 0·05 were considered significant.
Methylation of SOCS1 and correlation with clinical features
The representative results of methylation-specific PCR are shown in Fig 1A. Fifty-four patients (47·4%) showed SOCS1 hypermethylation. As summarised in Table I, patients with SOCS1 methylation had a higher percentage of BM blasts than those without methylation (P = 0·033). Those who had SOCS1 methylation included four (19·0%) of the 21 patients with RA, four (40·0%) of the 10 patients with RARS, 20 (60·6%) of the 33 patients with RAEB, 11 (55·0%) of the 20 patients with RAEB-T, 10 (40·0%) of the 25 patients with CMML and all five patients with MDS–AML. The patients with high-risk subtypes of MDS (RAEB, RAEB-T and CMML) had a significantly higher incidence of SOCS1 methylation (52·6%) than those with low-risk subtypes (RA and RARS, 25·8%, P = 0·011). The same was also true if the International Prognostic Scoring System (IPSS), which weighted the effect of blast quantity, karyotype and cytopenia concurrently, was used for risk stratification (Greenberg et al, 1997) [36·8% for low and intermediate (Int)-1. vs. 63·0% for Int-2 and high risk, P = 0·008]. There were no significant differences in other clinical features, such as age, sex, initial hemoglobin levels, white blood cell counts, and platelet counts between the patients with and without SOCS1 methylation.
|Patients’ parameter||Status of SOCS1 methylation||Total (n = 114)||P|
|Methylated (n = 54)||Unmethylated (n = 60)|
|Age (years)*||64 (7–84)||65 (19–86)||64 (7–86)||0·231|
|WBC (×109/l)*||4710 (440–227200)||4640 (1570–87420)||4675 (220–227200)||0·913|
|Haemoglobin (g/dl)*||8·9 (4·4–12·9)||7·8 (3·9–14·4)||8·5 (3·9–14·4)||0·348|
|Platelet count (×109/l)*||102 (2–400)||74 (3–607)||88·5 (2–607)||0·453|
|Bone marrow blast (%)*||7·2 (0–54·8)||4·4 (0–25·4)||5·8 (0–54·8)||0·033|
|Poor||13 (68·4%)||6 (31·6%)||19|
|−7/7q-||8 (72·7%)||3 (27·3%)||11|
|Complex||6 (60·0%)||4 (40·0%)||10|
|Intermediate||10 (66·7%)||5 (33·3%)||15|
|Good||27 (38·6%)||43 (61·4%)||70|
|Low-risk||8 (25·8%)||23 (74·2%)||31|
|Refractory anaemia||4 (19·0%)||17 (81·0%)||21|
|RA with ring sideroblasts||4 (40·0%)||6 (60·0%)||10|
|High-risk||41 (52·6%)||37 (47·4%)||78|
|RA with excess blasts (RAEB)||20 (60·6%)||13 (39·4%)||33|
|RAEB-T||11 (55·0%)||9 (45·0%)||20|
|Chronic myelomonocytic leukaemia||10 (40·0%)||15 (60·0%)||25|
|Acute myeloid leukaemia||5 (100%)||0||5|
|Low-risk||21 (36·8%)||36 (63·2%)||57|
|Low||4 (23·5%)||13 (76·5%)||17|
|Int-1||17 (42·5%)||23 (57·5%)||40|
|High-risk||29 (63·0%)||17 (37·0%)||46|
|Int-2||21 (63·6%)||12 (36·4%)||33|
|High||8 (61·5%)||5 (38·5%)||13|
|Mutated||9 (81·8%)||2 (18·2%)||11|
|Wild||38 (39·6%)||58 (60·4%)||96|
|Mutated||2 (50·0%)||2 (50·0%)||4|
|Wild||44 (43·6%)||57 (56·4%)||101|
Correlation of SOCS1 methylation status with cytogenetics and RAS mutations
Cytogenetic data were available for 104 patients. Thirty-eight patients (36·5%) demonstrated chromosomal abnormalities, including 22 with simple and 10 with complex aberrations (≥3 abnormalities). Eight patients (7·7%) showed trisomy 8 (+8), seven had monosomy 7 (−7), four had deletion of 7q (7q-), two each had −5 and 5q- respectively, and four had 20q-. SOCS1 methylation was inversely associated with good-risk cytogenetics (normal, -Y, 5q- or 20q- as the sole abnormality, P = 0·021, Table I).
Mutations of NRAS and KRAS were previously evaluated in 107 and 105 of the patients, respectively. Eleven patients had NRAS mutations and four had KRAS mutations (Chen et al, 2006). The patients with NRAS mutations had a significantly higher incidence of SOCS1 methylation than those without methylation (81·8% vs. 39·6%, P = 0·010).
Sequential analysis of SOCS1 methylation status during disease evolution
Twenty-nine patients, 15 with SOCS1 methylation and 14 without, had sequential studies performed two to five times during their disease course. With a median follow-up time of 28 months (range: 1–231), two (14·3%) of the 14 patients who did not have SOCS1 methylation on initial study acquired it during disease progression (Fig 1B). On the other hand, SOCS1 hypermethylation detected at diagnosis disappeared in disease remission after allogeneic haematopoietic stem cell transplantation (HSCT) in two patients (Fig 1C).
Correlation of SOCS1 methylation with outcome
With a median follow-up duration of 17 months (range: 1–231), leukaemic transformation occurred in 31 patients (29·8%). The patients with SOCS1 methylation had a higher cumulative risk of leukaemic transformation at 3 years than those without (55·8% vs. 27·7%, P = 0·004, Fig 2A); this difference remained significant within the subgroup of patients with high-risk subtypes of MDS (67·3% vs. 45·1%, P = 0·045, Fig 2B). Analysis of the combined effect of SOCS1 methylation and NRAS mutation showed that the patients with both NRAS mutations and SOCS1 methylation had the highest risk of developing AML, followed by those with either one of the two alterations, and those with neither (100% vs. 49·6% vs. 25·9% at 3 years, P < 0·001, Fig 2C). The same was also true in the subgroup of patients with high-risk subtypes of disease (100% vs. 66·0% vs. 42·7% at 3 years, P < 0·001, Fig 2D). We did not detect any significant effect for SOCS1 methylation on overall survival, probably because of the small sample size.
As methylation of SOCS1 was associated with many other clinical factors that may also correlate with leukaemic transformation, Cox proportional hazard regression model was used to adjust for the possible confounding effects between individual factors. Analysis of factors including age (≤20, 21 to 60, ≥61 years), gender, risk of FAB subtype (high risk, low risk), initial haemogram (white blood cell count, haemoglobin, platelet count, and absolute neutrophil count), initial BM blast percentage, karyotype (good risk, intermediate risk and high risk), SOCS1 methylation status, and mutations of NRAS and KRAS, demonstrated that only the FAB subtype (high risk vs. low risk, relative risk 13·2, P = 0·013) and NRAS mutation (mutant vs. wild, relative risk 9·21, P < 0·001) were significant risk factors for leukaemic transformation.
This study demonstrated that a substantial portion of patients with MDS had SOCS1 methylation. It occurred more frequently in high-risk subtypes of MDS, which, in our institution, included CMML (data not shown). In most patients with SOCS1 methylation, the abnormality could be detected at diagnosis. Only a few patients who did not have SOCS1 methylation initially acquired this gene alteration during clinical follow-up, implying that the aberration might occur in the early stage of the disease rather than during disease progression. No sequential study of SOCS1 methylation in MDS has been reported in the literature so far. Our findings require confirmation by further studies on more patients.
We showed for the first time that patients with SOCS1 methylation had a higher cumulative risk of leukaemic transformation than did those without this abnormality (55·8% vs. 27·7%, P = 0·004). Only two papers concerning SOCS1 methylation in MDS have been published (Brakensiek et al, 2005; Johan et al, 2005), and the correlation of SOCS1 methylation to leukaemic transformation was not mentioned in either report. In the study by Brakensiek et al (2005), patients with RAEB had a higher frequency of SOCS1 methylation than those with RA and RARS, but it was not known whether the patients with SOCS1 methylation had a higher chance of leukaemic transformation. SOCS1 methylation was correlated with several other factors associated with leukaemic transformation of MDS, including FAB subtype, karyotype and NRAS mutations. Multivariate analysis showed that it was not an independent risk factor for leukaemic transformation. These results indicated that methylation of SOCS1 might not be a major factor on its own to predict MDS progression. As SOCS1 protein itself is a negative regulator of signalling pathways, rather than a factor that directly controls transcription and proliferation, its role is probably a tumour suppressor (Rottapel et al, 2002) and its silencing may be phenotypically recessive in MDS. However, in the presence of other genetic abnormalities, such as chromosomal aberrations or NRAS mutations, loss of SOCS1 activity may exacerbate their effects and lead to MDS progression as well as leukaemic transformation. This is probably why methylation of SOCS1 was associated with leukaemic transformation, even within the subgroup of patients with high-risk MDS, and the patients with both NRAS mutation and SOCS1 methylation had a higher risk of leukaemic transformation than those with either one or neither of the changes.
Previous reports raised the debate regarding the significance of hypermethylation of SOCS1 in exon 2 or in the 5′ promoter area (Chen et al, 2003; Galm et al, 2003, 2004; Chim et al, 2004a–c; Fujitake et al, 2004, Yoshikawa et al, 2001; Guo et al, 2004; To et al, 2004; Watanabe et al, 2004b). The present study analysed the region located in exon 2, as was that first studied by Yoshikawa et al (2001). Although no direct functional validation was performed in this study, several studies have convincingly shown transcriptional silencing of SOCS1 by exon 2 hypermethylation (Okochi et al, 2003; Lin et al, 2004; Oshimo et al, 2004; Sutherland et al, 2004; To et al, 2004; Yoshida et al, 2004). Additionally, Brakensiek et al (2005) demonstrated that demethylation of exon 2 resulted in increased SOCS1 mRNA in KGIa cells. The present study showed that hypermethylation of SOCS1 exon 2 was correlated with the clinical and biological characteristics of MDS. Furthermore, the hypermethylation status was no longer detected when complete remission was achieved after HSCT in two patients who had SOCS1 exon 2 methylation at diagnosis. All of these findings strongly suggest that exon 2 methylation plays an important role in the pathogenesis of MDS.
Brakensiek et al (2005) found SOCS1 exon 2 methylation in 21% of the patients with RA, 29% of patients with RARS, and 48% of the patients with RAEB, incidences that were quite similar to those in this study. In contrast, we did not detect SOCS1 hypermethylation in the 5′ promoter region in any of the 50 MDS patients studied (data not shown). Johan et al (2005) found that 11% of MDS patients had SOCS1 methylation in the 5′ promoter region. Taken together, it seems that methylation of the SOCS1 promoter region is not as important as that of exon 2 for the development of MDS. However, the studies conducted to date are too limited to reach any conclusion. Although SOCS1 exon 2 methylation was demonstrated in some normal controls (Chim et al, 2004a–c; Johan et al, 2005), it was not detected in peripheral blood or BM cells of healthy controls in the current study. Furthermore, hypermethylation was no longer detected after HSCT in two patients who had SOCS1 exon 2 methylation at diagnosis in our series (Fig 1C), suggesting that donor-derived normal haematopoietic cells showed no SOCS1 exon 2 methylation.
Because gene methylation is an important mechanism of oncogenic transformation, hypomethylation or demethylation therapy is, thus, a rational direction of treatment. The methylation status of many genes were studied as candidates that influence the pathogenesis and progression of MDS, including SOCS1 in the current study (Tien et al, 2001; Au et al, 2003; Voso et al, 2004; Iwai et al, 2005; Johan et al, 2005). It is hoped that the demethylation agent, decitabine or azacytidine, which achieved early success in the therapy of MDS (Issa et al, 2004; Lubbert & Wijermans, 2005), will be a treatment for MDS in the future.
In summary, this is the first report to demonstrate associations of SOCS1 exon 2 methylation with clinical and biological features of MDS and transformation to AML. Methylation of SOCS1 occurred more frequently in high-risk subtypes of MDS than in low-risk ones and was correlated with karyotype and NRAS mutations. Although it was not an independent risk factor for leukaemic transformation of MDS in multivariate analysis, its presence would potentiate the effect of other risk factors on evolution of the disease. The methylation of SOCS1 usually occurred in the early stage of MDS, but a few patients acquired this gene alteration during disease progression. It disappeared after HSCT. All of these findings suggest that SOCS1 exon 2 methylation plays an important role in the pathogenesis of MDS, especially those of high-risk subtypes. SOCS1 methylation may be a potential marker for monitoring the treatment response in MDS patients with this gene alteration.
This work was supported by grants from National Science Council of the Republic of China, NSC 93-2314-B002-038 and 94-2314-B002-143, and Department of Medical Research in National Taiwan University Hospital.
- 2003) Aberrant p15 gene promoter methylation in therapy-related myelodysplastic syndrome and acute myeloid leukaemia: clinicopathological and karyotypic associations. British Journal of Haematology, 120, 1062–1065. , , , , , & (
- 1982) Proposals for the classification of the myelodysplastic syndromes. British Journal of Haematology, 51, 189–199. , , , , , & (
- 2005) Hypermethylation of the suppressor of cytokine signalling-1 (SOCS-1) in myelodysplastic syndrome. British Journal of Haematology, 130, 209–217. , , , & (
- 2003) SOCS1 methylation in patients with newly diagnosed acute myeloid leukemia. Genes, Chromosomes and Cancer, 37, 300–305. , , , , , , , , , & (
- 2006) Acquisition of JAK2, PTPN11, and RAS mutations during disease progression in primary myelodysplastic syndrome. Leukemia, 20, 1155–1158. , , , , , , , , , , , , & (
- 2004a) SOCS1 and SHP1 hypermethylation in multiple myeloma: implications for epigenetic activation of the Jak/STAT pathway. Blood, 103, 4630–4635. , , , & (
- 2004b) SOCS1 and SHP1 hypermethylation in mantle cell lymphoma and follicular lymphoma: implications for epigenetic activation of the Jak/STAT pathway. Leukemia, 18, 356–358. , , & (
- 2004c) Epigenetic dysregulation of the Jak/STAT pathway by frequent aberrant methylation of SHP1 but not SOCS1 in acute leukaemias. Annals of Hematology, 83, 527–532. , & (
- 2006) Nucleophosmin mutations in de novo acute myeloid leukemia: the age-dependent incidences and the stability during disease evolution. Cancer Research, 66, 3310–3316. , , , , , , , , , , , , & (
- 2000) The role of STATs in myeloid differentiation and leukemia. Oncogene, 19, 2511–2522. , & (
- 2004) Aberrant methylation of SOCS-1 was observed in younger colorectal cancer patients. Journal of Gastroenterology, 39, 120–124. , , , , , & (
- 2003) SOCS-1, a negative regulator of cytokine signaling, is frequently silenced by methylation in multiple myeloma. Blood, 101, 2784–2788. , , , & (
- 2004) DNA methylation changes in multiple myeloma. Leukemia, 18, 1687–1692. , , , , , & (
- 1986) The myelodysplastic syndromes. Part I. What are they? Part II. Classification. Scandinavian Journal of Haematology. Supplementum, 45, 11–20. (
- 1990) Clonal karyotype abnormalities and clinical progress in the myelodysplastic syndrome. British Journal of Haematology, 76, 194–202. , & (
- 1997) International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood, 89, 2079–2088. , , , , , , , , , , , , , & (
- 2004) Promoter hypermethylation of resected bronchial margins: a field defect of changes? Clinical Cancer Research, 10, 5131–5136. , , , , , , , , & (
- 1996) Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proceedings of the National Academy of Sciences of the United States of America, 93, 9821–9826. , , , & (
- 1998) Twenty proteins containing a C-terminal SOCS box form five structural classes. Proceedings of the National Academy of Sciences of the United States of America, 95, 114–119. , , , , , , , , & (
- ISCN (1985) An International System for Human Cytogenetic Nomenclature. F.Mitelman (ed.). S Karger, Basel.
- 2004) Phase 1 study of low-dose prolonged exposure schedules of the hypomethylating agent 5-aza-2′-deoxycytidine (decitabine) in hematopoietic malignancies. Blood, 103, 1635–1640. , , , , , , , , , & (
- 2005) Expression and methylation status of the FHIT gene in acute myeloid leukemia and myelodysplastic syndrome. Leukemia, 19, 1367–1375. , , , , , & (
- 1986) Prognostic implications of morphology and karyotype in primary myelodysplastic syndromes. Blood, 67, 1765–1772. , , , , & (
- 2005) Aberrant methylation of the negative regulators RASSFIA, SHP-1 and SOCS-1 in myelodysplastic syndromes and acute myeloid leukaemia. British Journal of Haematology, 129, 60–65. , , , & (
- 1985) Cytogenetic studies in 174 consecutive patients with preleukemic or myelodysplastic syndromes. Mayo Clinic Proceedings, 60, 507–516. , & (
- 2004) Promoter CpG methylation of tumor suppressor genes in colorectal cancer and its relationship to clinical features. Oncology Reports, 11, 341–348. , , , , , & (
- 2003) Epigenetic alteration of the SOCS1 gene in chronic myeloid leukaemia. British Journal of Haematology, 123, 654–661. , , , , & (
- 2002) Cytokine control of developmental programs in normal hematopoiesis and leukemia. Oncogene, 21, 3284–3294. & (
- 2005) Epigenetic therapy in MDS and acute AML: focus on decitabine. Annals of Hematology, 84(Suppl. 13), 1–2. & (
- 1997) Structure and function of a new STAT-induced STAT inhibitor. Nature, 387, 924–929. , , , , , , , , , , & (
- 2003) Methylation-mediated silencing of SOCS-1 gene in hepatocellular carcinoma derived from cirrhosis. Clinical Cancer Research, 9, 5295–5298. , , , , , & (
- 2004) Epigenetic inactivation of SOCS-1 by CpG island hypermethylation in human gastric carcinoma. International Journal of Cancer, 112, 1003–1009. , , , , , & (
- 2002) Cellular signalling pathways: new targets in leukaemia therapy. British Journal of Haematology, 116, 57–77. , , & (
- 2002) The tumor suppressor activity of SOCS-1. Oncogene, 21, 4351–4362. , , , , , , , & (
- 1998) Liver degeneration and lymphoid deficiencies in mice lacking suppressor of cytokine signaling-1. Proceedings of the National Academy of Sciences of the United States of America, 95, 14395–14399. , , , , , , & (
- 2004) Differential hypermethylation of SOCS genes in ovarian and breast carcinomas. Oncogene, 23, 7726–7733. , , , , , , & (
- 1994) Cytogenetic studies, ras mutation, and clinical characteristics in primary myelodysplastic syndrome. A study on 68 Chinese patients in Taiwan. Cancer Genetics and Cytogenetics, 74, 40–49. , , , , , , , , & (
- 2001) Methylation of the p15(INK4B) gene in myelodysplastic syndrome: it can be detected early at diagnosis or during disease progression and is highly associated with leukaemic transformation. British Journal of Haematology, 112, 148–154. , , , , , , & (
- 2004) Constitutional activation of IL-6-mediated JAK/STAT pathway through hypermethylation of SOCS-1 in human gastric cancer cell line. British Journal of Cancer, 91, 1335–1341. , , , , , , , , , , & (
- 2004) Aberrant methylation of DAP-kinase in therapy-related acute myeloid leukemia and myelodysplastic syndromes. Blood, 103, 698–700. , , , , , , & (
- 2004a) Suppressor of cytokine signalling-1 gene silencing in acute myeloid leukaemia and human haematopoietic cell lines. British Journal of Haematology, 126, 726–735. , , , , , , , , , , , , , , & (
- 2004b) Implication of SOCS-1 gene methylation in acute myeloid leukaemia. British Journal of Haematology, 127, 608–609. , & (
- 2004) SOCS1 is a suppressor of liver fibrosis and hepatitis-induced carcinogenesis. Journal of Experimental Medicine, 199, 1701–1707. , , , , , , , & (
- 2001) SOCS-1, a negative regulator of the JAK/STAT pathway, is silenced by methylation in human hepatocellular carcinoma and shows growth-suppression activity. Nature Genetics, 28, 29–35. , , , , , , & (