Diagnostic utility of droplet digital PCR to detect TERT promoter mutations among glioblastoma samples using 7-deaza-dGTP

Abstract

Telomerase reverse transcriptase (TERT) promoter mutations, resulting in increased TERT activity, are known to be associated with many different cancer types, including among 70%–80% of glioblastomas and 60% each of bladder carcinomas and hepatocellular carcinomas. Increased TERT activity has also been associated with several molecular features including rare rearrangements, duplications and amplifications. However, of all the molecular changes, mutations in the TERT promoter region are known to be the most frequent. They are often restricted to two hotspots, 124 and 146 base pairs upstream of the transcription start sites, often referred to as C228T and C250T. 2 The presence of these promoter mutations increases the similarity of these sites to the erythroblast transformationspecific (ETS) binding region, serving as a de novo binding site for ETS family transcription factors. Over time these mutations have assumed greater importance as markers of prognostication. In a recent study, Kikuchi et al have clearly demonstrated that glioblastomas with TERT promoter mutations had a shorter progressionfree survival (7 months vs 10 months; p=0.015) and overall survival (16 months vs 24 months; p=0.017) as compared with TERT wildtype glioblastomas. Such studies emphasise the need for accurate laboratory assays to detect these TERT promoter mutations. However, detection of TERT promoter mutation can be challenging in tissues with low tumour cellularity, high heterogeneity and poor DNA quality, especially if they are formalinfixed. The problem is further compounded by the high guaninecytosine (GC) content of the region (~80%) that can form strong secondary structures impairing strand denaturation. Validated assays that amplify GCrich regions often use additives like the Q solution and locked nucleic acids (LNA) on a droplet digital PCR (ddPCR) platform to amplify the region successfully. Recently, 7deaza-2’-deoxyguanosine -5’-triphosphate (7deazadGTP) has also been tested as an alternative to LNA, but has been validated only on cell lines. In this study, we determined the utility of 7deazadGTP in TERT promoter analysis using DNA from fresh frozen glioblastoma samples, with a known TERT and Isocitrate Dehydrogenase (IDH) status assessed by Sanger sequencing previously. Since currently available literature on TERT promoter analysis by ddPCR is mostly based on DNA obtained from fresh/frozen or other body fluids, 7–9 we decided to also extend the work to DNA from formalin fixed paraffin embedded (FFPE) to determine if the two provide comparable results. Thirtythree fresh frozen tissues obtained from histopathologically confirmed cases of glioblastoma were included for DNA extraction using Qiagen DNA Mini Kit (QIAamp, Qiagen, India). Twentysix (78%) of these 33 samples included were positive for IDH mutations, when characterised previously. Further, a subset (n=15) of these were also extracted, in parallel, from FFPE blocks using NucleoSpin DNA FFPE XS (MachereyNagel, USA) and quantitated using Nanodrop (Nanodrop, USA). Twenty nanograms of DNA were used for further analysis. The two TERT promoter mutations were detected separately using the primers and probes as described by Colebatch et al, and 200 μmol/L 7deazadGTP (New England Biolabs, USA) was added to each reaction and eventually titrated to 100 μmol/L, during optimisation. The rest of the reaction was set up as per the manufacturers’ specification for the QX200 Droplet Digital PCR System (BioRad, USA). PCR was performed using the T100 thermal cycler and the following cycling conditions were used: 95°C for 10 min, followed by 35 cycles of 95°C for 30 s, 63.5°C for 1 min and 98°C for 5 min. The data were analysed using the QuantSoft Pro software suite (V.1.0, BioRad). The amplitude thresholds were set for C228T and C250T at 2000 for channels 1 and 2, respectively. A positive, negative and a nontemplate control was included for each run and all tests were done in duplicate. Initially, dilution experiments were conducted diluting the positive sample with a wild type to determine sensitivity and the assay was found to pick up to 0.1% of mutant allele fraction, corresponding to ~0.2% tumour content. Of the 33 glioblastoma samples assayed for the TERT promoter mutations with 7deazadGTP, all yielded a result and there were no issues of amplification with either sets of probes. Twentyone tested positive for TERT promoter mutations, with 15 being positive for C228T and 6 positive for C250T. These results Correspondence