Correlation of myelodysplastic syndromes with i(17)(q10) and TP53 and SETBP1 mutations

Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal stem cell disorders that are highly prevalent in the elderly. Isochromosome 17(q10) [i(17q)] has a frequency of 0 4–0 8%, and cytogenetic risk stratification suggests that isolated i(17q) is of intermediate prognostic significance (Brunning et al, 2008; Greenberg et al, 2012; Schanz et al, 2012). It has been postulated that i(17q) may be associated with TP53 mutations, although a recent study found no association (Kanagal-Shamanna et al, 2012). SETBP1 is located at 18q21.1 and mutations of this gene have been reported in several haematological malignancies (Crist obal et al, 2010; Fernandez-Mercado et al, 2013; Piazza et al, 2013). We investigated the mutational status of TP53 and SETBP1 in 27 untreated MDS patients (who provided written informed consent) reported to the Spanish MDS group, as TP53 mutations and i(17q) have been shown to rarely co-exist in recent reports of concurrent i(17q) and SETBP1 mutations. Patients were classified according to the World Health Organization 2008 criteria (Brunning et al, 2008) and G-banding karyotypes were described according to the International System for Human Cytogenetic Nomenclature 2009 (Shaffer et al, 2009) (Table I). Genomic DNA was isolated from cells fixed in Carnoy’s fixative and from fresh bone marrow samples (QIAamp DNA mini-kit, Qiagen Inc, Valencia, CA, USA). TP53 exons 5–9 and SETBP1 exon 3 were amplified and sequenced as previously described (Meggendorfer et al, 2013). Patients were divided into three groups: no mutations, TP53 mutations and SETBP1 mutations. Statistical analyses (SPSS 22.0. SPSS Inc., Chicago, IL, USA) were performed for these three groups and for the overall patient population. Baseline characteristics and main clinical variables were compared by Kruskal–Wallis and Pearson or Fisher’s exact test for continuous or categorical variables, respectively. Overall survival (OS) was measured from haematological diagnosis to death from any cause or last follow-up. Survival curves were plotted according to the Kaplan–Meier method, and compared using the log-rank test. Univariate and multivariate analyses of OS were performed using the Cox proportional hazard ratio (HR) model. Two-sided P values <0 05 were considered statistically significant. TP53 mutations were analysed in all patients. Five of the 27 (18 5%) had non-synonymous point mutations (Table I). Two patients had mutations in exon 7, G245S and R175H [complex karyotype (CK) and isolated i(17q)]; two had mutations in exons 5 and 6 (H179P and Y220H, respectively, all with a CK); and one patient (with CK) had an intronic mutation (c.919 + 1G>A) at the splicing recognition site. SETBP1 was analysed in all patients. Eleven (41%) had SETBP1 non-synonymous point mutations (mainly located in residues 868–871). Five of 11 (45%) had heterozygous mutations in D868N. Four of these patients had isolated i (17q); the remaining one had i(17q) with one additional abnormality. SETBP1 heterozygous G870S mutations (27%) were found in 3/11 patients, all with isolated i(17q). The three remaining patients had single heterozygous mutations in D868Y, S869G and I871T [D868Y patient had a CK, while the others presented isolated i(17q)]. A statistically significant relationship was found between mutation group and karyotype classification (P = 0 009). Most TP53 mutated cases (80%) presented a CK, meanwhile 82% of SETBP1 mutations presented isolated i(17q). There were no statistically significant differences between the groups regarding haemoglobin, leucocytes, platelets, polymorphonuclear cells, absolute neutrophil count or blast percentage. The TP53-mutated group showed worse OS than non-mutated and SETBP1-mutated groups {median OS [95% confidence interval (CI)]: 2 9 (2 4, 3 5), 14 1 (0, 33 7) and 13 9 (7 9, 19 8), respectively, P = 0 001}. Patients with a CK showed worse OS than isolated i(17q) and i(17q) with one additional abnormality [median OS (95% CI): 3 8 (2 5, 5 1), 26 5 (7 2, 45 8) and 21 3 (8 3, 34 3), respectively, P < 0 001] (Figure S1). Univariate analysis of OS (Table II) revealed a statistically significant difference between non-mutated and TP53mutated patients [HR (95% CI): 0 07 (0 01, 0 3); P = 0 002], and between SETBP1-mutated and TP53mutated patients [HR (95% CI): 0 11 (0 02, 0 5); P = 0 006]. There were also significant differences between the cytogenetic groups: isolated i(17q) versus CK [HR (95% CI): 0 07 (0 02, 0 3); P < 0 001]and i(17q) with an additional abnormality versus CK [HR (95% CI): 0 07 (0 01, 0 3); P = 0 001]. In multivariate analysis (Table II), and because of the high correlation between karyotype and mutation type, only the karyotype remained as a prognostic factor, considering CK as the reference category: isolated i (17q) versus CK (P = 0 002), and i(17q) with an additional abnormality versus CK (P = 0 01). correspondence