Background and Aim: Deletion of 5q is the most frequent cytogenetic aberration in MDS and is associated with distinct clinical characteristics, disease course and sensitivity to lenalidomide. The serine-threonine kinase CSNK1A1 is located in the commonly deleted region at 5q32 and has been described as a tumor-suppressor gene in colon cancer and acute myeloid leukemia through regulation of ß-catenin and p53. Recently, missense mutations in CSNK1A1 have been described in individual patients with del(5q) MDS. The aim of our study was to characterize the frequency and potential prognostic impact of CSNK1A1mutations in MDS and AML following MDS.

Methods: 192 patients with MDS or AML following MDS (sAML) and deletion of chromosome 5q and 406 patients with MDS/sAML without deletion of chromosome 5q were included in the current analysis (n=598 in total). Patients with MDS (n=442) or sAML (n=156) were cytogenetically characterized by chromosome banding analysis and molecularly analyzed for mutations in exon 3 and 4 of CSNK1A1, the region critical for the kinase function, by Sanger sequencing. Patients with mutated CSNK1A1 were also analyzed for mutations in TP53 by next-generation or Sanger sequencing.

Results:CSNK1A1 mutations were found in 17 (8.9%) of 192 MDS patients with del(5q). The mutation frequency was similar between patients with isolated del(5q) (n=153) and patients with concurrent cytogenetic aberrations or missing additional cytogenetic information (n=39)(9.2% vs 7.7%, P=.7). No mutation of CSNK1A1 was found in any of 406 MDS/sAML patients without del(5q). Thirteen patients (76%) had missense mutations affecting amino acid E98 in exon 3 of CSNK1A1. Of these, the glutamic acid to lysin substitution was the most frequent amino acid substitution (n=7). All mutations of glutamic acid 98 had a high probability to be damaging to the protein based on PolyPhen2 predictions (scores 0.922 to1). One patient had an Asn86Tyr mutation concurrently with the Glu98Ala mutation. Four patients (24%) had missense mutations affecting aspartic acid 140 in exon 4 of CSNK1A1. These mutations had moderate PolyPhen2 prediction scores (0.558-0.798). Three of the 17 CSNK1A1 mutated patients had additional cytogenetic aberrations besides del(5q), i.e. one trisomy 8, one trisomy 11, and one monosomy 7. None of the CSNK1A1 patients had a concurrent TP53 mutation.

Del(5q) patients with wildtype or mutated CSNK1A1 had a similar median age (73.3 vs 77.5 years, P=.15). 70% and 59% of wildtype and mutated CSNK1A1 patients had female sex, respectively (P=.33). The WBC count was similar between wildtype and mutated CSNK1A1patients (3.9 vs 4.6, P=.47).

Survival information was available for 155 patients with del(5q) (81%) including 16 patients (94%) with mutated CSNK1A1. Median follow-up from the time of sample harvest was 2.02 years. The probability of survival at 2 years was 41% for CSNK1A1 mutated and 72% for CSNK1A1wildtype patients (P=.059, log-rank test), suggesting a potential negative prognostic impact of CSNK1A1 mutations in del(5q) MDS patients.

Conclusion:CSNK1A1 mutations are highly specific for MDS patients with del(5q) and are one of the most frequent recurrent genetic aberrations in these patients. Our survival analysis suggests that CSNK1A1 mutations have an unfavorable prognostic effect in patients with MDS and del(5q); however, the prognostic impact has to be confirmed in additional patients. Mutation analysis of exon 3 and 4 of CSNK1A1 should be included in the routine workup of MDS patients with deletion of 5q.

Disclosures

Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Kobbe:Celgene: Honoraria, Research Funding; Amgen: Honoraria, Research Funding; Medac: Other; Astellas: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; Neovii: Other. Haferlach:MLL: Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution