Refining prognosis capacity and implementing priciples of personalized therapies for Neurofibromatosis Type 2

Abstract

Neurofibromatosis type 2 (NF2) is an autosomal dominant condition caused by loss of function variants in the NF2 gene. The pathognomonic feature of the disease is the development of bilateral vestibular schwannomas (VS), benign tumours that result from the overgrowth of Schwann cells (SC) on the vestibulocochlear nerves. Affected individuals are at high risk of developing schwannomas in other cranial, spinal and peripheral nerves, as well meningiomas and ependymomas, tumours that are responsible for significant morbidity. The clinical management of these patients is complex due to the limited treatment options and the broad and variable clinical spectrum of the disease, strongly associated with the type of variant inherited in the NF2 gene. The general objective of this thesis was to contribute to the development of strategies for a personalized medicine for Neurofibromatosis Type 2. Specifically, the three aims were to assess and improve the prognosis capacity of the disease, to evaluate the use of antisense gene therapy approaches in vitro for pathogenic variants causing NF2 and finally, to develop an induced pluripotent stem cell-based model for further studies on the NF2 gene role and the development of targeted therapies. In 2017, the UK NF2 Reference Group established a Genetic Severity Score (GSS) to predict the severity of the disease considering the NF2 pathogenic variant. In the present work, we validated the GSS in a cohort of NF2 patients from the Spanish National Reference Centre (CSUR) of Phacomatoses. In addition, data from the identification of potential novel prognostic molecular markers were considered to suggest a revision of the GSS, called the Functional Genetic Severity Score (FGSS). The FGSS includes data from functional assays and the predicted effect of pathogenic variants on merlin, the NF2 protein. With the aim to develop a personalized therapeutic approach for NF2 patients, we evaluated the use of antisense oligonucleotides for NF2 loss of function variants in patients’ primary fibroblast cell cultures. First, we tested the use of phosphorodiamidate morpholino oligomers (PMOs) to modulate the effect of NF2 splice site variants, although the approach proved unsuccessful to restore merlin levels. The second strategy consisted of reducing the severe effect of truncating variants affecting the NF2 gene. We used PMOs to induce the skipping of an in-frame exon harbouring a nonsense or frameshift variant to generate novel, and potentially functional, merlin forms. Encouraging results emerged for germline truncating variants in exon 11 of NF2, for which it was possible to generate a hypomorphic merlin (merlin-e11) that partially rescued the NF2 phenotype in primary fibroblasts cultures. This result constitutes an in vitro proof of concept of the use of PMOs as a personalized therapy for NF2 patients.

Finally, due to the limited preclinical models available for NF2, we generated induced pluripotent stem cell (iPSC) lines with single or bi-allelic loss of function of NF2 through combining direct reprogramming of VS cells with the use of CRISPR/Cas9 genome editing. In our experience, the NF2 gene could be essential for reprogramming and maintaining pluripotency. Despite difficulties encountered in maintaining pure merlin-deficient pluripotent cultures, iPSCs lines were characterized and differentiated towards the Neural Crest – Schwann cell axis. The application of a 3D Schwann cell differentiation protocol led to the successful generation of NF2 (+/-) and NF2 (-/-) spheroids that co-expressed the classical lineage markers p75 and S100B, potentially representing a genuine VS model.