Mutational and splicing landscape in a cohort of 43,000 patients tested for hereditary cancer

Spectrum and genotyping strategies of "dark" genetic matter in germline susceptibility genes of tumor syndromes

PURPOSE

Despite the widespread use of high-throughput genotyping strategies, certain mutation types remain understudied. We provide an overview of these often overlooked mutation types, with representative examples from common hereditary cancer syndromes.

METHODS

We conducted a comprehensive review of the literature and locus-specific variant databases to summarize the germline pathogenic variants discovered through non-routine genotyping methods. We evaluated appropriate detection and analysis methods tailored for these specific genetic aberrations. Additionally, we performed in silico splice predictions on deep intronic variants registered in the ClinVar database.

RESULTS

Our study suggests that, aside from founder mutations, most cases are sporadic. However, we anticipate a relatively high likelihood of splice effects for deep intronic variants. The findings underscore the significant clinical utility of genome sequencing techniques and the importance of applying relevant analysis methods.

Clinical Variant Reclassification in Hereditary Disease Genetic Testing

Importance

Because accurate and consistent classification of DNA sequence variants is fundamental to germline genetic testing, understanding patterns of initial variant classification (VC) and subsequent reclassification from large-scale, empirical data can help improve VC methods, promote equity among race, ethnicity, and ancestry (REA) groups, and provide insights to inform clinical practice.

Objectives

To measure the degree to which initial VCs met certainty thresholds set by professional guidelines and quantify the rates of, the factors associated with, and the impact of reclassification among more than 2 million variants.

Design, Setting, and Participants

This cohort study used clinical multigene panel and exome sequencing data from diagnostic testing for hereditary disorders, carrier screening, or preventive genetic screening from individuals for whom genetic testing was performed between January 1, 2015, and June 30, 2023.

Exposure

DNA variants were classified into 1 of 5 categories: benign, likely benign, variant of uncertain significance (VUS), likely pathogenic, or pathogenic.

Main Outcomes and Measures

The main outcomes were accuracy of classifications, rates and directions of reclassifications, evidence contributing to reclassifications, and their impact across different clinical areas and REA groups. One-way analysis of variance followed by post hoc pairwise Tukey honest significant difference tests were used to analyze differences among means, and pairwise Pearson χ2 tests with Bonferroni corrections were used to compare categorical variables among groups.

Results

The cohort comprised 3 272 035 individuals (median [range] age, 44 [0-89] years; 2 240 506 female [68.47%] and 1 030 729 male [31.50%]; 216 752 Black [6.62%]; 336 414 Hispanic [10.28%]; 1 804 273 White [55.14%]). Among 2 051 736 variants observed over 8 years in this cohort, 94 453 (4.60%) were reclassified. Some variants were reclassified more than once, resulting in 105 172 total reclassification events. The majority (64 752 events [61.65%]) were changes from VUS to either likely benign, benign, likely pathogenic, or pathogenic categories. An additional 37.66% of reclassifications (39 608 events) were gains in classification certainty to terminal categories (ie, likely benign to benign and likely pathogenic to pathogenic). Only a small fraction (663 events [0.63%]) moved toward less certainty, or very rarely (61 events [0.06%]) were classification reversals. When normalized by the number of individuals tested, VUS reclassification rates were higher among specific underrepresented REA populations (Ashkenazi Jewish, Asian, Black, Hispanic, Pacific Islander, and Sephardic Jewish). Approximately one-half of VUS reclassifications (37 074 of 64 840 events [57.18%]) resulted from improved use of data from computational modeling.

Conclusions and Relevance

In this cohort study of individuals undergoing genetic testing, the empirically estimated accuracy of pathogenic, likely pathogenic, benign, and likely benign classifications exceeded the certainty thresholds set by current VC guidelines, suggesting the need to reevaluate definitions of these classifications. The relative contribution of various strategies to resolve VUS, including emerging machine learning-based computational methods, RNA analysis, and cascade family testing, provides useful insights that can be applied toward further improving VC methods, reducing the rate of VUS, and generating more definitive results for patients.

Solving Missing Heritability in Patients With Familial Adenomatous Polyposis With DNA-RNA Paired Testing

PURPOSE

Patients with germline pathogenic variants (PVs) in APC develop tens (attenuated familial adenomatous polyposis [AFAP]) to innumerable (classic FAP) adenomatous polyps in their colon and are at significantly increased lifetime risk of colorectal cancer. Up to 10% of FAP and up to 50% of patients with AFAP who have undergone DNA-only multigene panel testing (MGPT) do not have an identified PV in APC. We seek to demonstrate how the addition of RNA sequencing run concurrently with DNA can improve detection of germline PVs in individuals with a clinical presentation of AFAP/FAP.

METHODS

We performed a retrospective query of individuals tested with paired DNA-RNA MGPT from 2021 to 2022 at a single laboratory and included those with a novel APC PV located in intronic regions infrequently covered by MGPT, a personal history of polyposis, and family medical history provided. All clinical data were deidentified in this institutional review board-exempt study.

RESULTS

Three novel APC variants were identified in six families and were shown to cause aberrant splicing because of the creation of a deep intronic cryptic splice site that leads to an RNA transcript subject nonsense-mediated decay. Several carriers had previously undergone DNA-only genetic testing and had received a negative result.

CONCLUSION

Here, we describe how paired DNA-RNA MGPT can be used to solve missing heritability in FAP families, which can have important implications in family planning and treatment decisions for patients and their families.

Diagnostic Outcomes of Concurrent DNA and RNA Sequencing in Individuals Undergoing Hereditary Cancer Testing

Importance

Personalized surveillance, prophylaxis, and cancer treatment options for individuals with hereditary cancer predisposition are informed by results of germline genetic testing. Improvements to genomic technology, such as the availability of RNA sequencing, may increase identification of individuals eligible for personalized interventions by improving the accuracy and yield of germline testing.

Objective

To assess the cumulative association of paired DNA and RNA testing with detection of disease-causing germline genetic variants and resolution of variants of uncertain significance (VUS).

Design, Setting, and Participants

Paired DNA and RNA sequencing was performed on individuals undergoing germline testing for hereditary cancer indication at a single diagnostic laboratory from March 2019 through April 2020. Demographic characteristics, clinical data, and test results were curated as samples were received, and changes to variant classification were assessed over time. Data analysis was performed from May 2020 to June 2023.

Main Outcomes and Measures

Main outcomes were increase in diagnostic yield, decrease in VUS rate, the overall results by variant type, the association of RNA evidence with variant classification, and the corresponding predicted effect on cancer risk management.

Results

A total of 43 524 individuals were included (median [range] age at testing, 54 [2-101] years; 37 373 female individuals [85.7%], 6224 male individuals [14.3%], and 2 individuals of unknown sex [<0.1%]), with 43 599 tests. A total of 2197 (5.0%) were Ashkenazi Jewish, 1539 (3.5%) were Asian, 3077 (7.1%) were Black, 2437 (5.6%) were Hispanic, 27 793 (63.7%) were White, and 2049 (4.7%) were other race, and for 4507 individuals (10.3%), race and ethnicity were unknown. Variant classification was impacted in 549 individuals (1.3%). Medically significant upgrades were made in 97 individuals, including 70 individuals who had a variant reclassified from VUS to pathogenic/likely pathogenic (P/LP) and 27 individuals who had a novel deep intronic P/LP variant that would not have been detected using DNA sequencing alone. A total of 93 of 545 P/LP splicing variants (17.1%) were dependent on RNA evidence for classification, and 312 of 439 existing splicing VUS (71.1%) were resolved by RNA evidence. Notably, the increase in positive rate (3.1%) and decrease in VUS rate (-3.9%) was higher in Asian, Black, and Hispanic individuals combined compared to White individuals (1.6%; P = .02; and -2.5%; P < .001).

Conclusions and Relevance

Findings of this diagnostic study demonstrate that the ability to perform RNA sequencing concurrently with DNA sequencing represents an important advancement in germline genetic testing by improving detection of novel variants and classification of existing variants. This expands the identification of individuals with hereditary cancer predisposition and increases opportunities for personalization of therapeutics and surveillance.

Parallel DNA/RNA NGS Using an Identical Target Enrichment Panel in the Analysis of Hereditary Cancer Predisposition

Germline DNA testing using the next-gene-ration sequencing (NGS) technology has become the analytical standard for the diagnostics of hereditary diseases, including cancer. Its increasing use places high demands on correct sample identification, independent confirmation of prioritized variants, and their functional and clinical interpretation. To streamline these processes, we introduced parallel DNA and RNA capture-based NGS using identical capture panel CZECANCA, which is routinely used for DNA analysis of hereditary cancer predisposition. Here, we present the analytical workflow for RNA sample processing and its analytical and diagnostic performance. Parallel DNA/RNA analysis allowed credible sample identification by calculating the kinship coefficient. The RNA capture-based approach enriched transcriptional targets for the majority of clinically relevant cancer predisposition genes to a degree that allowed analysis of the effect of identified DNA variants on mRNA processing. By comparing the panel and whole-exome RNA enrichment, we demonstrated that the tissue-specific gene expression pattern is independent of the capture panel. Moreover, technical replicates confirmed high reproducibility of the tested RNA analysis. We concluded that parallel DNA/RNA NGS using the identical gene panel is a robust and cost-effective diagnostic strategy. In our setting, it allows routine analysis of 48 DNA/RNA pairs using NextSeq 500/550 Mid Output Kit v2.5 (150 cycles) in a single run with sufficient coverage to analyse 226 cancer predisposition and candidate ge-nes. This approach can replace laborious Sanger confirmatory sequencing, increase testing turnaround, reduce analysis costs, and improve interpretation of the impact of variants by analysing their effect on mRNA processing.

Benchmarking splice variant prediction algorithms using massively parallel splicing assays

BACKGROUND

Variants that disrupt mRNA splicing account for a sizable fraction of the pathogenic burden in many genetic disorders, but identifying splice-disruptive variants (SDVs) beyond the essential splice site dinucleotides remains difficult. Computational predictors are often discordant, compounding the challenge of variant interpretation. Because they are primarily validated using clinical variant sets heavily biased to known canonical splice site mutations, it remains unclear how well their performance generalizes.

RESULTS

We benchmark eight widely used splicing effect prediction algorithms, leveraging massively parallel splicing assays (MPSAs) as a source of experimentally determined ground-truth. MPSAs simultaneously assay many variants to nominate candidate SDVs. We compare experimentally measured splicing outcomes with bioinformatic predictions for 3,616 variants in five genes. Algorithms' concordance with MPSA measurements, and with each other, is lower for exonic than intronic variants, underscoring the difficulty of identifying missense or synonymous SDVs. Deep learning-based predictors trained on gene model annotations achieve the best overall performance at distinguishing disruptive and neutral variants, and controlling for overall call rate genome-wide, SpliceAI and Pangolin have superior sensitivity. Finally, our results highlight two practical considerations when scoring variants genome-wide: finding an optimal score cutoff, and the substantial variability introduced by differences in gene model annotation, and we suggest strategies for optimal splice effect prediction in the face of these issues.

CONCLUSION

SpliceAI and Pangolin show the best overall performance among predictors tested, however, improvements in splice effect prediction are still needed especially within exons.

Combined Bioinformatic and Splicing Analysis of Likely Benign Intronic and Synonymous Variants Reveals Evidence for Pathogenicity

Background

Current clinical variant analysis pipelines focus on coding variants and intronic variants within 10-20 bases of an exon-intron boundary that may affect splicing. The impact of newer splicing prediction algorithms combined with in vitro splicing assays on rare variants currently considered Benign/Likely Benign (B/LB) is unknown.

Methods

Exome sequencing data from 576 pediatric cancer patients enrolled in the Texas KidsCanSeq study were filtered for intronic or synonymous variants absent from population databases, predicted to alter splicing via SpliceAI (>0.20), and scored as potentially deleterious by CADD (>10.0). Total cellular RNA was extracted from monocytes and RT-PCR products analyzed. Subsequently, rare synonymous or intronic B/LB variants in a subset of genes submitted to ClinVar were similarly evaluated. Variants predicted to lead to a frameshifted splicing product were functionally assessed using an in vitro splicing reporter assay in HEK-293T cells.

Results

KidsCanSeq exome data analysis revealed a rare, heterozygous, intronic variant (NM_177438.3(DICER1):c.574-26A>G) predicted by SpliceAI to result in gain of a secondary splice acceptor site. The proband had a personal and family history of pleuropulmonary blastoma consistent with DICER1 syndrome but negative clinical sequencing reports. Proband RNA analysis revealed alternative DICER1 transcripts including the SpliceAI-predicted transcript.Similar bioinformatic analysis of synonymous or intronic B/LB variants (n=31,715) in ClinVar from 61 Mendelian disease genes yielded 18 variants, none of which could be scored by MaxEntScan. Eight of these variants were assessed (DICER1 n=4, CDH1 n=2, PALB2 n=2) using in vitro splice reporter assay and demonstrated abnormal splice products (mean 66%; range 6% to 100%). Available phenotypic information from submitting laboratories demonstrated DICER1 phenotypes in 2 families (1 variant) and breast cancer phenotypes for PALB2 in 3 families (2 variants).

Conclusions

Our results demonstrate the power of newer predictive splicing algorithms to highlight rare variants previously considered B/LB in patients with features of hereditary conditions. Incorporation of SpliceAI annotation of existing variant data combined with either direct RNA analysis or in vitro assays has the potential to identify disease-associated variants in patients without a molecular diagnosis.

Combining clinical and molecular characterization of CDH1: a multidisciplinary approach to reclassification of a splicing variant

Pathogenic germline variants (PGVs) in the CDH1 gene are associated with diffuse gastric and lobular breast cancer syndrome (DGLBC) and can increase the lifetime risk for both diffuse gastric cancer and lobular breast cancer. Given the risk for diffuse gastric cancer among individuals with CDH1 PGVs is up to 30-40%, prophylactic total gastrectomy is often recommended to affected individuals. Therefore, accurate interpretation of CDH1 variants is of the utmost importance for proper clinical decision-making. Herein we present a 45-year-old female, with lobular breast cancer and a father with gastric cancer of unknown pathology at age 48, who was identified to have an intronic variant of uncertain significance in the CDH1 gene, specifically c.833-9 C > G. Although the proband did not meet the International Gastric Cancer Linkage Consortium (IGCLC) criteria for gastric surveillance, she elected to pursue an upper endoscopy where non-targeted gastric biopsies identified a focus of signet ring cell carcinoma (SRCC). The proband then underwent a total gastrectomy, revealing numerous SRCC foci, but no invasive diffuse gastric cancer. Simultaneously, a genetic testing laboratory performed RNA sequencing to further analyze the CDH1 intronic variant, identifying an abnormal transcript from a novel acceptor splice site. The RNA analysis in conjunction with the patient's gastric foci of SRCC and family history was sufficient evidence for reclassification of the variant from uncertain significance to likely pathogenic. In conclusion, we report the first case of the CDH1 c.833-9 C > G intronic variant being associated with DGLBC and illustrate how collaboration among clinicians, laboratory personnel, and patients is crucial for variant resolution.

Intronic Germline DICER1 Variants in Patients With Sertoli-Leydig Cell Tumor

Germline pathogenic loss-of-function (pLOF) variants in DICER1 are associated with a predisposition for a variety of solid neoplasms, including pleuropulmonary blastoma and Sertoli-Leydig cell tumor (SLCT). The most common DICER1 pLOF variants include small insertions or deletions leading to frameshifts, and base substitutions leading to nonsense codons or altered splice sites. Larger deletions and pathogenic missense variants occur less frequently. Identifying these variants can trigger surveillance algorithms with potential for early detection of DICER1-related cancers and cascade testing of family members. However, some patients with DICER1-associated tumors have no pLOF variants detected by germline or tumor testing. Here, we present two patients with SLCT whose tumor sequencing showed only a somatic missense DICER1 RNase IIIb variant. Conventional exon-directed germline sequencing revealed no pLOF variants. Using a custom capture panel, we discovered novel intronic variants, ENST00000343455.7: c.1752+213A>G and c.1509+16A>G, that appear to interfere with normal splicing. We suggest that when no DICER1 pLOF variants or large deletions are discovered in exonic regions despite strong clinical suspicion, intron sequencing and splicing analysis should be performed.

Using the ACMG/AMP framework to capture evidence related to predicted and observed impact on splicing: Recommendations from the ClinGen SVI Splicing Subgroup

The American College of Medical Genetics and Genomics (ACMG)/Association for Molecular Pathology (AMP) framework for classifying variants uses six evidence categories related to the splicing potential of variants: PVS1, PS3, PP3, BS3, BP4, and BP7. However, the lack of guidance on how to apply such codes has contributed to variation in the specifications developed by different Clinical Genome Resource (ClinGen) Variant Curation Expert Panels. The ClinGen Sequence Variant Interpretation Splicing Subgroup was established to refine recommendations for applying ACMG/AMP codes relating to splicing data and computational predictions. We utilized empirically derived splicing evidence to (1) determine the evidence weighting of splicing-related data and appropriate criteria code selection for general use, (2) outline a process for integrating splicing-related considerations when developing a gene-specific PVS1 decision tree, and (3) exemplify methodology to calibrate splice prediction tools. We propose repurposing the PVS1_Strength code to capture splicing assay data that provide experimental evidence for variants resulting in RNA transcript(s) with loss of function. Conversely, BP7 may be used to capture RNA results demonstrating no splicing impact for intronic and synonymous variants. We propose that the PS3/BS3 codes are applied only for well-established assays that measure functional impact not directly captured by RNA-splicing assays. We recommend the application of PS1 based on similarity of predicted RNA-splicing effects for a variant under assessment in comparison with a known pathogenic variant. The recommendations and approaches for consideration and evaluation of RNA-assay evidence described aim to help standardize variant pathogenicity classification processes when interpreting splicing-based evidence.

Benchmarking splice variant prediction algorithms using massively parallel splicing assays

Background

Variants that disrupt mRNA splicing account for a sizable fraction of the pathogenic burden in many genetic disorders, but identifying splice-disruptive variants (SDVs) beyond the essential splice site dinucleotides remains difficult. Computational predictors are often discordant, compounding the challenge of variant interpretation. Because they are primarily validated using clinical variant sets heavily biased to known canonical splice site mutations, it remains unclear how well their performance generalizes.

Results

We benchmarked eight widely used splicing effect prediction algorithms, leveraging massively parallel splicing assays (MPSAs) as a source of experimentally determined ground-truth. MPSAs simultaneously assay many variants to nominate candidate SDVs. We compared experimentally measured splicing outcomes with bioinformatic predictions for 3,616 variants in five genes. Algorithms' concordance with MPSA measurements, and with each other, was lower for exonic than intronic variants, underscoring the difficulty of identifying missense or synonymous SDVs. Deep learning-based predictors trained on gene model annotations achieved the best overall performance at distinguishing disruptive and neutral variants. Controlling for overall call rate genome-wide, SpliceAI and Pangolin also showed superior overall sensitivity for identifying SDVs. Finally, our results highlight two practical considerations when scoring variants genome-wide: finding an optimal score cutoff, and the substantial variability introduced by differences in gene model annotation, and we suggest strategies for optimal splice effect prediction in the face of these issues.

Conclusion

SpliceAI and Pangolin showed the best overall performance among predictors tested, however, improvements in splice effect prediction are still needed especially within exons.

A Systematic Method for Detecting Abnormal mRNA Splicing and Assessing Its Clinical Impact in Individuals Undergoing Genetic Testing for Hereditary Cancer Syndromes

Nearly 14% of disease-causing germline variants result from the disruption of mRNA splicing. Most (67%) DNA variants predicted in silico to disrupt splicing are classified as variants of uncertain significance. An analytic workflow-splice effect event resolver (SPEER)-was developed and validated to use mRNA sequencing to reveal significant deviations in splicing, pinpoint the DNA variants potentially involved, and measure the deleterious effects of the altered splicing on mRNA transcripts, providing evidence for assessing the pathogenicity of the variant. SPEER was used to analyze leukocyte RNA encoding 63 hereditary cancer syndrome-related genes in 20,317 patients. Among 3563 patients (17.5%) with at least one DNA variant predicted to affect splicing, 971 (4.8%) had altered splicing with a deleterious effect on the transcript, and 40 had altered splicing due to a DNA variant located outside of the reportable range of the test. Integrating SPEER results into the interpretation of variants allowed variants of uncertain significance to be reclassified as pathogenic or likely pathogenic in 0.4%, and as benign or likely benign in 5.9%, of the 20,317 patients. SPEER-based evidence was associated with a significantly greater effect on classifications of pathogenic or likely pathogenic and benign or likely benign in nonwhite versus non-Hispanic white patients, illustrating that evidence derived from mRNA splicing analysis may help to reduce ethnic/ancestral disparities in genetic testing.

APPLICATION OF THE ACMG/AMP FRAMEWORK TO CAPTURE EVIDENCE RELEVANT TO PREDICTED AND OBSERVED IMPACT ON SPLICING: RECOMMENDATIONS FROM THE CLINGEN SVI SPLICING SUBGROUP

The American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) framework for classifying variants uses six evidence categories related to the splicing potential of variants: PVS1 (null variant in a gene where loss-of-function is the mechanism of disease), PS3 (functional assays show damaging effect on splicing), PP3 (computational evidence supports a splicing effect), BS3 (functional assays show no damaging effect on splicing), BP4 (computational evidence suggests no splicing impact), and BP7 (silent change with no predicted impact on splicing). However, the lack of guidance on how to apply such codes has contributed to variation in the specifications developed by different Clinical Genome Resource (ClinGen) Variant Curation Expert Panels. The ClinGen Sequence Variant Interpretation (SVI) Splicing Subgroup was established to refine recommendations for applying ACMG/AMP codes relating to splicing data and computational predictions. Our study utilised empirically derived splicing evidence to: 1) determine the evidence weighting of splicing-related data and appropriate criteria code selection for general use, 2) outline a process for integrating splicing-related considerations when developing a gene-specific PVS1 decision tree, and 3) exemplify methodology to calibrate bioinformatic splice prediction tools. We propose repurposing of the PVS1_Strength code to capture splicing assay data that provide experimental evidence for variants resulting in RNA transcript(s) with loss of function. Conversely BP7 may be used to capture RNA results demonstrating no impact on splicing for both intronic and synonymous variants, and for missense variants if protein functional impact has been excluded. Furthermore, we propose that the PS3 and BS3 codes are applied only for well-established assays that measure functional impact that is not directly captured by RNA splicing assays. We recommend the application of PS1 based on similarity of predicted RNA splicing effects for a variant under assessment in comparison to a known Pathogenic variant. The recommendations and approaches for consideration and evaluation of RNA assay evidence described aim to help standardise variant pathogenicity classification processes and result in greater consistency when interpreting splicing-based evidence.

NBN Pathogenic Germline Variants are Associated with Pan-Cancer Susceptibility and In Vitro DNA Damage Response Defects

PURPOSE

To explore the role of NBN as a pan-cancer susceptibility gene.

EXPERIMENTAL DESIGN

Matched germline and somatic DNA samples from 34,046 patients were sequenced using Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets and presumed pathogenic germline variants (PGV) identified. Allele-specific and gene-centered analysis of enrichment was conducted and a validation cohort of 26,407 pan-cancer patients was analyzed. Functional studies utilized cellular models with analysis of protein expression, MRN complex formation/localization, and viability assessment following treatment with γ-irradiation.

RESULTS

We identified 83 carriers of 32 NBN PGVs (0.25% of the studied series), 40% of which (33/83) carried the Slavic founder p.K219fs. The frequency of PGVs varied across cancer types. Patients harboring NBN PGVs demonstrated increased loss of the wild-type allele in their tumors [OR = 2.7; confidence interval (CI): 1.4-5.5; P = 0.0024; pan-cancer], including lung and pancreatic tumors compared with breast and colorectal cancers. p.K219fs was enriched across all tumor types (OR = 2.22; CI: 1.3-3.6; P = 0.0018). Gene-centered analysis revealed enrichment of PGVs in cases compared with controls in the European population (OR = 1.9; CI: 1.3-2.7; P = 0.0004), a finding confirmed in the replication cohort (OR = 1.8; CI: 1.2-2.6; P = 0.003). Two novel truncating variants, p.L19* and p.N71fs, produced a 45 kDa fragment generated by alternative translation initiation that maintained binding to MRE11. Cells expressing these fragments showed higher sensitivity to γ-irradiation and lower levels of radiation-induced KAP1 phosphorylation.

CONCLUSIONS

Burden analyses, biallelic inactivation, and functional evidence support the role of NBN as contributing to a broad cancer spectrum. Further studies in large pan-cancer series and the assessment of epistatic and environmental interactions are warranted to further define these associations.