Indolent Phenotype of RAS-Associated Autoimmune Leukoproliferative Disorder (RALD) Is Characterized By Single Somatic Mutations in RAS Genes with Absence of Cooperating Mutations

RAS-Associated Autoimmune Leukoproliferative Disorder (RALD) is a chronic indolent condition that typically presents in childhood with monocytosis, lymphocytosis, autoimmune phenomena, splenomegaly, and variable lymphadenopathy. RALD is characterized by somatic mutations in KRAS or NRAS in hematopoietic cells (myeloid and lymphoid), resulting in missense mutations involving the 12^th or 13^thcodon. Despite the indolent course of RALD, it shares some overlapping clinical features and similar genetic defects with Juvenile Myelomonocytic Leukemia (JMML), which is characterized by canonical and non-canonical RAS-pathway mutations. All RALD patients have normal bone marrow karyotypes. While patients with JMML often harbor monosomy 7 or other karyotypic abnormalities, normal karyotypes are also not uncommon in JMML. Optimal therapeutic management of RALD and JMML differs significantly, necessitating accurate diagnostic discrimination of RALD and normal karyotype JMML.

Recent JMML studies have characterized the genomic landscape of JMML, which involves additional recurrent mutations that are thought to cooperate with RAS-pathway mutations driving the malignant phenotype of JMML. We sought to evaluate the genomic landscape of RALD for the presence of non-RAS cooperating mutations reported in JMML.

Using targeted sequencing for 54 genes associated with myeloid malignancies, amplicon libraries were constructed from peripheral blood mononuclear cells (PBMCs) collected from 11 patients diagnosed with RALD (6 with previously identified KRAS and 5 with previously identified NRAS mutations) and sequenced with an average read depth of ~3000x. Variants were identified using the Illumina miSeq reporter software and analyzed using Illumina Variant Studio. Variants were filtered based SIFT, PolyPhen, and CADD predictions, ClinVar annotations, and mutations previously identified in JMML. Over 90% of the mutated genes reported in JMML sequencing reports, were evaluated (including KRAS, NRAS, CBL, PTPN11, SH2B3, SETBP1, GATA2, RUNX1, ASXL1, EZH2, DNMT3A, and ZRSR2). Of note, only the KRAS or NRAS mutations, initially identified from Sanger sequencing, were identified as disease-contributing and no other known pathogenic mutations were detected in the RALD cohort. Variant allele frequencies (VAFs) of KRAS and NRAS mutations in RALD ranged from 32.3%-50.8% with a median VAF of 46.6%. In order to further investigate genes not included in the targeted sequencing panel that may be contributing to disease, whole exome sequencing (WES) was performed on PBMCs from 7 RALD patients (4 KRAS patients, 3 NRAS patients). Despite the expanded genomic sequencing in WES (mean coverage of 225x), only the single pathogenic KRAS or NRAS mutation, known to be disease-related, was identified. No other mutations that were previously identified in JMML were detected in the WES data set. Based on presence of autoimmune phenomena in RALD, a multiplex inflammatory cytokine immunoassay investigating GM-CSF, IFN-γ, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, and IL-12p70 was used to quantify cytokine levels in the plasma of 9 RALD patients and 17 healthy volunteers. The levels of IL-2, IL-6, and IL-10 were found significantly increased in RALD patients. IL-2 was 4.6 fold higher in RALD than controls (median 0.696 pg/mL vs. 0.149 pg/mL; p=0.049). Similarly, IL-6 was 10 fold higher in RALD than controls (median 6.42 pg/mL vs. 0.63 pg/mL; p=0.035), and IL-10 was 18 fold higher in RALD than controls (median 26.9 pg/mL vs. 1.49 pg/mL; p=0.018).

Overall, these findings provide evidence that RALD, in our patient cohort with an indolent clinical course, is characterized by single somatic KRAS or NRAS mutations with the absence of cytogenetic abnormalities and cooperating mutations that are associated with malignancy. This suggests that in the spectrum of so-called RASopathies, RALD represents a non-malignant proliferative hematopoietic disease that is clinically distinct from JMML. The altered plasma cytokine profile in RALD may play a role in the autoimmune features of the disease and be related to sustained activation of the RAS pathway in hematopoietic cells. Further investigation is imperative to distinguish these two overlapping entities characterized by RAS pathway mutations, and to better understand the spectrum of disease from indolence to aggressive malignancy.