Aplicaciones traslacionales en neoplasias mieloides de la secuenciación de nueva generación de ARNnuevas dianas de letalidad sintética e inmunoterapia

Abstract

Background: Recent therapeutic and survival advances in the lymphoid malignancies setting have not been paralled in the myeloid setting, where there is an urgent need for new targets. Despite the absence of mutations in the DNA repair machinery in myeloid malignancies, the advent of high-throughput sequencing and discovery of splicing and epigenetics defects in chronic myelomonocytic leukaemia (CMML) prompted us, in the first chapter of our work, to revisit a pathogenic role for genes involved in DNA damage response. In the second chapter, we focused on the cancer testis antigens (CTA), a group of molecules whose expression has been silenced by hypermethylation in the somatic cells of healthy individuals. However, cancer cells re-express some of them, specific for the type of tumor, with immunogenic properties. Even when re-expressed, these molecules show a low number of transcripts, and global transcriptome RNA-seq approaches could be not sensitive enough for their characterisation. A customized targeted RNA-seq approach has shown to be more reliable for these purposes. Aims: to take advantage of recent RNA-seq approaches to identify new molecules to be targeted in CMML and related disorders, focusing in a synthetic lethality strategy for DNA repair genes, and in CTAs change of expression after azacitidine for immunotherapeutic purposes. Methods: First, we screened for misregulated DNA repair genes by enhanced RNA-sequencing on bone marrow from a discovery cohort of 27 CMML patients and 9 controls. We validated differentially expressed candidates in CMML CD34+bone marrow selected cells and in an independent cohort of 74 CMML patients, mutationally contextualized by targeted sequencing. Second, we performed a customized RNA-seq experiment targeting 210 CTAs at diagnosis and after one cycle of azacitidine in 19 cases, validated at the protein level the behaviour of our main candidates, and compared the results with global RNA-seq results from 11 patients. Resultados: First, we found BAP1 and PARP1 down-regulation to be specific to CMML compared with other related disorders. Chromatin-regulator mutated cases showed decreased BAP1 dosage. We validated a significant over-expression of the double strand break-fidelity genes CDKN1A and ERCC1, independent of promoter methylation and associated with chemorefractoriness. In addition, patients bearing mutations in the splicing component SRSF2 displayed numerous aberrant splicing events in DNA repair genes, with a quantitative predominance in the single strand break pathway. Second, we performed a RNA-seq study targeting 210 CTAs pre and post (+28) first cycle of azacitidine in 19 patients with MDS (n=13) and CMML (n=6). We found that 137 out of 210 CTAs showed a significant expression (CPM>1) at diagnosis. Some of the CTAs not expressed at baseline showed a re-expression after azacitidine; highlighting ADAM29 as a potential candidate based on the highest level of expression after a cycle of hypomethylating agent. In addition, TFDP3 y DDX53, both X-linked CTAs, emerged as main candidates as therapeutic targets, as they fulfilled two criteria: i) significant icrease I expression after 1 cycle of azacitidine in those patients who achieved complete remission after 4-6 cycles, and ii) a parallel dynamic was validated at the protein level. As we had the results from the global transcritptome pre and post azacitines in 11 of these targeted sequenced patients, we compared both datasets. We could identify 203 of CTA genes common in both designs. Of those, 181, at baseline and/or after treatment, were found to be significantly expressed in the targeted approach and only 53, at baseline and/or after treatment, in the global transcriptome experiment. Conclusion: Our results highlight potential targets in these group of diseases, selected with clinical and biological rationale, which currently has few therapeutic options.