Background: Aberrant mRNA splicing occurs in myeloid malignancies and affects genes involved in tumor suppression, heme biosynthesis and mitochondrial iron metabolism. Functional studies demonstrated impaired cellular differentiation upon targeting of aberrant splice variants. Hypomethylating agents (HMA) constitute the backbone of therapy of myeloid malignancies. Whether HMA treatment in myeloid malignancies alters the novel splicing transcriptional landscape and whether it correlates with responses remain largely unexplored.

Methods: Total RNA sequencing was done on CD34+ cells from 79 patients bone marrow samples involved by acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), chronic myelomonocytic leukemia (CMML), TF1 cell lines and CD34+ murine bone marrow cells. Novel alternatively spliced transcripts were detected using SplAdder and included the following splicing events: alterative 3' splice junction, alternative 5' splice junction, exon skipping, intron retention, multiple exon skipping and mutually exclusive exons. All alternatively splicing events were normalized to total transcript count in order to correct for total transcript levels. A false discovery rate of <0.1 was used to identify significant events.

Results:

A total of 79 myeloid disease patients (27.8% females, 72.1% males) with a median age of 70 years (range, 31-87 years) were included in this study. In aggregate analysis of all 79 myeloid malignancies (39.2% (n=31) pre-treatment and 60.8% (n=48) post-treatment), there were 160 versus 37 (4.3 folds), 112 versus 40 (2.8 folds), 292 versus 51 (5.7 folds), 172 versus 80 (2.1 folds) and 29 versus 9 (3.2 folds) and 2 versus 0 novel splicing events occurring in pre- versus post- HMA treatment, respectively, in alterative 3' splice junction, alternative 5' splice junction, exon skipping, intron retention, multiple exon skipping and mutually exclusive exons, respectively. This suggested that treatment with HMA led to downregulation of novel alternative splicing events after normalization to total transcripts. However, upon excluding AML patients from the analysis, there were no significant events associated with treatment suggesting that the findings could be due to random events. To further explore whether HMA therapy influenced novel splicing events, we examined the novel splicing pattern in 7 MDS patients with paired BM samples at pre- and post-HMA and found no significant differences in alternative splicing events before and after the treatment. We then examined TF1 (human erythroleukemia) cell lines at pre- and post- HMA time points, but did not identify notable differences in the novel alternative splicing events with respect to HMA treatment. To assess whether CD34+ bone marrow cells from mice treated with hypomethylating agents have differential novel alternatively spliced events, we conducted similar analysis and did not find any discernible differences pre- and post- HMA treatment. These findings suggest that HMA does not influence novel alternative splicing events.

Conclusions:

Aberrant splicing has been linked to myeloid neoplasms especially myelodysplastic syndrome with mutations in splice variant genes. Our findings suggest that HMA does not influence novel alternative splicing events in myeloid malignancies. Therefore, the alternative splicing in myeloid disease is inherent to the disease and not affected by treatment.

Disclosures

Garcia-Manero:H3 Biomedicine: Research Funding; Novartis: Research Funding; Bristol-Myers Squibb: Consultancy, Research Funding; Merck: Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Onconova: Research Funding; Acceleron Pharmaceuticals: Consultancy, Honoraria; Amphivena Therapeutics: Research Funding; Genentech: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Helsinn Therapeutics: Consultancy, Honoraria, Research Funding; Jazz Pharmaceuticals: Consultancy; Astex Pharmaceuticals: Consultancy, Honoraria, Research Funding; AbbVie: Honoraria, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.