patient cohort
Among consecutive patients who underwent BRCA tumor testing via ION Torrent-based sequencing between August 2017 and February 2022, we retrospectively selected 222 high-grade ovarian cancer (HGOC) patients with the following histologic subtypes : 203 serous (HGSOC), seven endometrioid, five clear cell, and seven with mixed histotypes.
From NGS BRCA1/2 Tumor testing was not available prior to 2017 at our institution, 19 of 222 subjects underwent germline testing prior to tumor sequencing based on personal history (very young age at diagnosis/breast cancer prior) or relatives. According to the workflow used by our Molecular Tumor Board (MTB), 73 of 203 patients with initial tumor evidence subsequently received genetic counseling, either for specific germline sequencing of a tumor-detected PV or for a large genomic rearrangement analysis14.
The ethics committee of the Fondazione IRCCS Istituto Nazionale dei Tumori in Milan approved the use of clinical and molecular data collected by the MTB for clinical studies and granted an exemption from requiring written consent for tumor genetic testing of patients, since these analyzes were carried out in the context of a diagnosis and care setting (INT Approval Number 227/20). All probands who underwent germline testing were over 18 years of age and signed an informed consent for the use of their samples and biological data for diagnostic and research purposes. All methods were carried out in accordance with the relevant guidelines and with the ethical principles of the Declaration of Helsinki.
Tumor tests
He BRCA1 and BRCA2 genes were evaluated by in-house NGS testing using the Oncomine BRCA Research Assay (Thermo Fisher Scientific, Inc). This trial provided 100% coverage of all BRCA1 and BRCA2 exons, with an average of 64 bases of intronic flanking sequences upstream and downstream of each exon. Five ฮผm sections of formalin-fixed paraffin-embedded (FFPE) specimens were manually microdissected to isolate the highest percentage of neoplastic cells. Genomic DNA was extracted with protease K (ON incubation at 55 ยฐC) and quantified with the Qubit dsDNA BR kit (Thermo Fisher Scientific, Inc). Libraries were prepared with the IonAmpliSeq Library 2.0 Kit (Thermo Fisher Scientific, Inc) and quantified with the Qubit dsDNA HS Kit (Thermo Fisher Scientific, Inc) following the manufacturer's instructions. Libraries are diluted to 25 pm, pooled, and loaded into the Ion Chef to perform emulsion PCR and chip loading on 318 v2 chips. Sequencing was performed on the ION PGM, using the HI-Q view Chef kit, according to the manufacturer's instructions. The data was processed using Torrent Suite 5.12.3 (TS). The quality of the sequencing output was first assessed through the Coverage Analysis plugin in the TS. Only samples whose library uniformity and on-target values โโwere at least 80% and with a mean coverage of 1500X were considered valid. SNV analysis was performed in duplicate: the TS Variant Caller plugin generated the first variant call and the resulting VCF file was loaded into the Variant Effect Predictor Tool (Ensembl, version GRCh37) for variant annotation. To eliminate erroneous base calling, we set coverage of each variant > 40X, variant frequency in each sample > 2%, and quality value > 30. Variants within the homopolymer (HP) of more than eight bases were not reported. and with chain bias โฅ 80%. . In the second analysis, the BAM files were automatically uploaded from the TS to the Ion Reporter software (IR, version 5.6 to 5.16) and the variant call was integrated into the "Oncomine BRCA Research Somaticโ318" analysis pipeline. The results of both analyzes were manually compared. Each variant was shown in IGV (ver. 2.3.97). Synonymous variants were filtered, while the remaining variants were classified into pathogenicity classes according to the Evidence-Based Network for Interpretation of Mutant Alleles (ENIGMA) consortium guidelines (https://enigmaconsortium.org/). Our assay could not reliably detect large intragenic rearrangements.
germ line test
Two EDTA tubes of peripheral blood samples were collected from each patient who underwent genetic counseling and was eligible for germline testing, either for targeted sequencing of tumor-detected pathogenic/probably pathogenic variants or for large-scale analysis. genomic rearrangements in patients without processable variants. detected in tumor tests. Whole blood DNA was isolated through the MagCoreยฎ Super automatic workstation with MagCoreยฎ Genomic DNA whole blood kit (Diatech LabLine SRL, Jesi, Italy). Targeted Sanger sequencing of detected tumors BRCA1/2 PVs were performed on purified PCR products using BigDye.ยฎ Terminator v.3.1 Cycle Sequencing kit (Thermo Fisher Scientific, Inc.) and run on the 3730Xl DNA Analyzer (Applied Biosystems; Thermo Fisher Scientific, Inc.), after purification with Agencourt CleanSeqยฎ-Beckman Coulter. The sequences were analyzed by Mutation Surveyorยฎ Software (v5.0.1; SoftGenetics, LLC., State College, PA, USA). Targeted sequencing results were confirmed in both aliquots of blood collected from each patient. Variants of uncertain clinical significance identified in tumor testing have not been systematically investigated at the germline level. Eligible probands, who tested negative for tumors with the Oncomine BRCA assay, were tested for large deletions and duplications of BRCA1 and BRCA2 on DNA in blood with SALSA MLPA P045 kits BRCA2/CHECK2 and P002 BRCA1 probe mix (MRC-Holland, Amsterdam, The Netherlands), following the manufacturer's instructions. MLPA products were run on the 3730Xl DNA Analyzer (Applied Biosystems; Thermo Fisher Scientific, Inc.) with the Gene Mapper module (Applied Biosystems; Thermo Fisher Scientific, Inc.). Results were analyzed using Gene Marker Software v2.7.0 (SoftGenetics, LLC, State College, PA, USA).
Evaluation of homopolymeric regions and statistical analysis
Building on previous observations on the homopolymer performance of ion semiconductor sequencing techniques, we focused our analysis on stretches of six or more identical bases, as calling precision has been shown to drop dramatically beyond this length.fifteen,17,25,28,30. Therefore, we selected the 29 homopolymeric regions that exceeded five repeats to be analyzed within the coding regions of both genes, including five in BRCA1 and 24 in BRCA2. Since truncating variants beyond codon 3326 of BRCA2 are not classified as high risk variants, the homopolymers downstream of residue c.9976 of BRCA2 were not included in the analysis (Table 2).
To overcome the limitations of the ION reporter software, which filters out most of the indels in the homopolymeric regions, we manually visualized the BAM alignment files of the 222 patients in the 29 regions with the IGV software (ver. 2.3.97). The median depth of coverage of the regions of interest ranged from 1045 to 6989X. Each sample showed a variable frequency of sequence alterations (both insertions and deletions) in each region. We estimated the variant allele frequency (VAF) of insertions and deletions (indels) by calculating the ratios of the maximum inserted or deleted reads over the total reads in each homopolymer (Supplementary Table 1).
Since the VAF of indels in homopolymeric regions is, in general, skewed to the left, we employed a modified version of the Cancer Outlier Profile Analysis (COPA) approach.31which consists of scaling the above to a normal distribution and later calculating the outliers that exceeded the threshold of the mean + 3 adjusted deviations to the median (ฮผ + 3ฯ).
To validate outliers, for each of the 29 regions, we further defined a threshold based on the normalized distributions (either by percentage insertion or deletion) of a control population. Since in both ovarian cancer and other BRCA-associated cancers, the predominant second hit is usually represented by loss of heterozygosity (LOH), whereas a second point mutation is an extremely rare event.32,33,3. 4We used as a control population a cohort of 46 patients in whom a non-homopolymeric PV had already been identified (either somatic or germinal).
To avoid possible selection bias, which would affect the estimated frequency of pathogenic variants occurring in homopolymeric regions in our cohort, we excluded from the analysis the 19 patients who underwent germline testing prior to tumor testing. This group also included two patients with confirmed germline homopolymeric PV who tested negative for tumors.
Therefore, we applied the thresholds (ฮผ + 3ฯ) estimated in the control population to the normalized distributions of the study cohort, composed of 157 individuals with no evidence of pathogenic variants in ION Torrent tumor tests. Also, consistent with the filtering criteria used in a previous study, which focused on germline variants29, we consider only homopolymeric indels with an absolute VAF greater than 15% and, in any case, greater than the maximum value of the control population in each homopolymeric region. Finally, regions with a total read count less than 100 were excluded from the analysis.
Targeted Sanger sequencing was performed on tumor DNA to confirm the occurrence of selected atypical homopolymeric indels using defined thresholds.
