Accurate and comprehensive evaluation of O6‐methylguanine‐DNA methyltransferase promoter methylation by nanopore sequencing

Skarphedinn Halldorsson,Richard Mark Nagymihaly,Areeba Patel,Petter Brandal,Ioannis Panagopoulos,Henning Leske,Marius Lund‐Iversen,Felix Sahm,Einar O. Vik‐Mo
DOI: https://doi.org/10.1111/nan.12984
2024-05-25
Neuropathology and Applied Neurobiology
Abstract:We assessed the effectiveness of nanopore long‐read sequencing for evaluatingmethylation levels across the MGMT CpG‐island, crucial for prognosis in glioblastomapatients. Using 165 CNS tumour samples, nanopore sequencing was compared toestablished techniques, showing a strong correlation with pyrosequencing and effectivepatient stratification via hierarchical clustering. Nanopore sequencing provided rapid,high‐confidence results and expanded analysis to all 98 CpGs of the MGMT CpG‐island,suggesting its utility for clinically relevant patient subgroup identification and warranting further exploration in clinical settings. Aims The methylation status of the O6‐methylguanine‐DNA methyltransferase (MGMT) promoter region is essential in evaluating the prognosis and predicting the drug response in patients with glioblastoma. In this study, we evaluated the utility of using nanopore long‐read sequencing as a method for assessing methylation levels throughout the MGMT CpG‐island, compared its performance to established techniques and demonstrated its clinical applicability. Methods We analysed 165 samples from CNS tumours, focusing on the MGMT CpG‐island using nanopore sequencing. Oxford Nanopore Technologies (ONT) MinION and PromethION flow cells were employed for single sample or barcoded assays, guided by a CRISPR/Cas9 protocol, adaptive sampling or as part of a whole genome sequencing assay. Methylation data obtained through nanopore sequencing were compared to results obtained via pyrosequencing and methylation bead arrays. Hierarchical clustering was applied to nanopore sequencing data for patient stratification. Results Nanopore sequencing displayed a strong correlation (R2 = 0.91) with pyrosequencing results for the four CpGs of MGMT analysed by both methods. The MGMT‐STP27 algorithm's classification was effectively reproduced using nanopore data. Unsupervised hierarchical clustering revealed distinct patterns in methylated and unmethylated samples, providing comparable survival prediction capabilities. Nanopore sequencing yielded high‐confidence results in a rapid timeframe, typically within hours of sequencing, and extended the analysis to all 98 CpGs of the MGMT CpG‐island. Conclusions This study presents nanopore sequencing as a valid and efficient method for determining MGMT promotor methylation status. It offers a comprehensive view of the MGMT promoter methylation landscape, which enables the identification of potentially clinically relevant subgroups of patients. Further exploration of the clinical implications of patient stratification using nanopore sequencing of MGMT is warranted.
pathology,neurosciences,clinical neurology
What problem does this paper attempt to address?