Investigating the Potential of Spliceosome Mutations to be Used As Biomarkers for Directing Immune Checkpoint Blockade Therapies in Myeloid Malignancies

Introduction

It is now well accepted that DNA damage repair deficiencies (DDRD) can result in the activation of the innate immune signalling pathway, cGAS-STING. This results in the up regulation of inflammatory cytokines, which promotes immune cell infiltration into the tumour site. However, chronic activation of this pathway can also promote expression of immunosuppressive checkpoint proteins, including PD-L1 and CTLA-4, resulting in an inflamed but immunosuppressed tumour microenvironment. This provides a rationale for exploring the benefits of immune checkpoint blockade (ICB) therapy for DDRD-positive tumours. Previously, we described how mutations in the premature-mRNA splicing machinery component SF3B1 conferred a DDRD highlighting a crucial role for the spliceosome in the DNA damage response. Mutations in components of the pre-mRNA splicing machinery are common in myeloid malignancies, such as myelodysplastic syndromes and acute myeloid leukaemia, and some solid tumours. This study aims to explore how common spliceosome mutations, in SRSF2 and SF3B1 can be exploited as biomarkers to guide the use of ICBs in myeloid malignancies.

Methods

Using CRISPR/Cas9-generated isogenic models harbouring the SRSF2^P95H and the SF3B1^K700E mutation, DNA damage repair kinetics, micronuclei formation and cytosolic dsDNA/dsRNA accumulation, were assessed using immunofluorescent labelling. Transcript expression and secretion of inflammatory cytokines were assessed using qRT-PCR and ELISAs, respectively, following treatment of the isogenic models with the topoisomerase II poison, daunorubicin. Moreover, immune competent in vivo validation of the benefit of ICB use with spliceosome mutated myeloid malignancies was performed using the embryonic chick chorioallantoic membrane (CAM) model.

Results

The SRSF2^P95H mutation conferred a reduced ability to repair DNA damage and enhanced sensitivity to the PARP inhibitor Olaparib, similar to our previous findings in our SF3B1^K700E model, indicating a DDRD. Additionally, post-daunorubicin treatment, SRSF2^P95H and SF3B1^K700E cell lines showed increased micronuclei and cytosolic DNA and RNA accumulation compared to WT controls. Furthermore, transcript expression of inflammatory cytokines, CXCL10, CCL5 and IFIT2, were significantly upregulated in SRSF2 and SF3B1 mutant cells compared to WT controls post-daunorubicin treatment. CCL5 secretion was also significantly increased in the SRSF2^P95H cell line, whilst CXCL10 secretion was significantly increased in the SF3B1^K700E model. No significant decreases in CCL5 transcription or secretion were observed following STING-depletion in the SRSF2^P95H model, however reductions were observed following depletion of MAVS, an adaptor protein involved in cytosolic dsRNA sensing, which also signals via STING. Furthermore, our data from the immune competent CAM model showed significant SF3B1^K700E tumour control with the ICB agent nivolumab, but interestingly made the SRSF2^P95H tumours more aggressive.

Conclusion

To date, our data suggests that the SRSF2^P95H and SF3B1^K700E mutations confer an increased ability to induce an innate immune response compared to their WT counterparts. However, this response may function independently of STING and may be reliant on the MAVS signalling pathway. Together, this data is pathing the way for use of SRSF2 and SF3B1 mutational status as a biomarker to direct the use of ICB agents in these tumour settings.