Primary NRAS-driven leptomeningeal melanoma is a rare childhood cancer most often seen in the context of mosaic disorder congenital melanocytic naevus syndrome. It is universally and rapidly fatal, although treatment with MEKi trametinib confers temporary symptomatic benefit. We sought to explore other therapeutic options, small interfering RNA (siRNA) to oncogenic NRAS, which has recently been shown to induce apoptosis in dermal melanocytic naevus cells1, and a high throughput FDA approved small molecule screen.

Antisense siRNAs to siNRASQ61K (NM_002524.5(NRAS):c.181C>A,p.(Q61K)) and siNRASQ61R (NM_002524.5(NRAS:c.182A>G,p.(Q61R)) were designed to be variant allele-specific, and transfected into the following cell lines: NRAS-variant primary leptomeningeal cells from three patients with leptomeningeal dysplasia, the precursor to leptomeningeal melanoma; NRAS-variant primary leptomeningeal cells from one patient with leptomeningeal melanoma; two control wild-type immortalised melanocyte cell lines hTERT-dermal melanocytes and HERMES-1. In parallel, an in vitro drug screen of 2529 FDA-approved drugs was performed in primary leptomeningeal cells from one patient with leptomeningeal dysplasia and the two wild-type melanocyte cell lines, with validation of potential hits in all patient and control lines. IncuCyte repeated phase-contrast imaging was used to measure confluency and a DNA-binding fluorescent caspase 3/7 dye to measure apoptosis across both experiments.

We report here that oncogenic driver knockdown with siNRASQ61K or siNRASQ61R triggers apoptosis in all leptomeningeal patient cells, dysplastic or malignant. Furthermore, addition of current clinical standard of care trametinib to siNRAS further elevates levels of apoptosis over siNRAS alone, despite a complete lack of apoptosis with trametinib alone. Mechanistic exploration demonstrates siNRAS treatment leads to downregulation of antiapoptotic genes BIRC5 and ARL6IP1, a gene protective against ER stress recently identified as downstream of NRAS. Additionally, we report here that the top four drug hits from the FDA screen epirubicin, tucidinostat, dasatinib and valemetostat suppress proliferation in patient cells significantly more than in controls, with epirubicin having the greatest effect. None of these small molecules, however, has any effect on apoptosis. Lastly, we explore the genome-wide expressional profiles of dysplastic and malignant cells pre- and post- treatment with epirubicin, trametinib and siNRAS.

Precision targeting of oncogenic NRAS driver alleles using siRNA is more promising than repurposing of small molecule inhibitors in currently untreatable leptomeningeal dysplasia and melanoma, potentially in combination with trametinib.