ID #692 The clinicopathologic characteristics and outcomes of KRAS-altered low-grade glioma
Julie Bennett, Rawan Hammad, Liana Nobre, Anthony Arnoldo, Mansuba Rana, Robert Siddaway, Adrian Levine, Julia Keith, Lananh Nguyen, Andrew Gao, Garth Nicholas, Sumit Das, Chris Dunham, Lucie Lafay-Cousin, Mary Jane Lim-Fat, Jane Barron, Craig Erker, Ronald Ramos, Gerard Jansen, Viktor Zherebitskiy, Magimairajan Issai Vanan, Nirav Thacker, Sébastien Perreault, Colleen D’Arcy, David Munoz, Eric Bouffet, James Rutka, Plavo Plavskyi, Naureen Mushtaq, Kristopher Langson, Michal Zapotocky, Uri Tabori, Cynthia HawkinsAbstract
Background
KRAS alterations are known genetic drivers across different types of cancer. Low-grade gliomas (LGG) are rarely reported to harbor this alteration with limited study of this cohort.
Methods
An international retrospective cohort of KRAS-altered LGG was identified from pathology databases. Clinical data was collected with molecular analysis including targeted RNA sequencing and pathway analysis using Nanostring nCounter system.
Results
Thirty-two pediatric, adolescent and adult patients were identified to have KRAS-altered LGG. Median age of diagnosis was 12.7 years (range 0.4-69.8). Most patients (n = 31) had localized disease with tumors predominantly found in cerebral hemispheres (56%) and diencephalon (38%). Tectal tumors were rare (n = 1). Thirty-one patients had single nucleotide variants (18 at KRAS p.Q61 and 8 at KRAS p.G12 locus) with one deletion. Seven patients had co-occurring RAS/MAPK alterations (including FGFR1, NF1 and BRAF alterations) and one patient had co-occurring MYBL1 fusion. Outcome was excellent with 5-year overall survival and progression-free survival (PFS) of 100% and 69%, respectively. One patient with gross total resection (n = 11) had subsequent recurrence. For those with subtotal resection (n = 21), 9 had further post-operative therapy. Outcomes were comparable to that of BRAF-fused LGG (5 year PFS 65% vs 69% for KRAS-altered, p = 0.9), with a comparable number of patients with KRAS-altered LGG requiring therapy beyond surgery (31% vs 30%). There was no difference in PFS based on age, tumor location or KRAS genotype. Upregulation of the PI3K and NFκB pathway were seen compared to normal brain (p < 0.05), and upregulation of the PI3K pathway when compared to BRAF-mutant LGG (p = 0.03), suggesting these tumors may have different therapeutic vulnerability compared to other RAS/MAPK-driven LGG.
Conclusions
This suggests that conservative management to avoid long-term toxicity is imperative in KRAS-altered LGG and further data on response to targeted therapy is important to collect.