Independent real-world application of a clinical-grade automated prostate cancer detection system.

Artificial intelligence (AI)-based systems applied to histopathology whole-slide images have the potential to improve patient care through mitigation of challenges posed by diagnostic variability, histopathology caseload, and shortage of pathologists. We sought to define the performance of an AI-based automated prostate cancer detection system, Paige Prostate, when applied to independent real-world data. The algorithm was employed to classify slides into two categories: benign (no further review needed) or suspicious (additional histologic and/or immunohistochemical analysis required). We assessed the sensitivity, specificity, positive predictive values (PPVs), and negative predictive values (NPVs) of a local pathologist, two central pathologists, and Paige Prostate in the diagnosis of 600 transrectal ultrasound-guided prostate needle core biopsy regions ('part-specimens') from 100 consecutive patients, and to ascertain the impact of Paige Prostate on diagnostic accuracy and efficiency. Paige Prostate displayed high sensitivity (0.99; CI 0.96-1.0), NPV (1.0; CI 0.98-1.0), and specificity (0.93; CI 0.90-0.96) at the part-specimen level. At the patient level, Paige Prostate displayed optimal sensitivity (1.0; CI 0.93-1.0) and NPV (1.0; CI 0.91-1.0) at a specificity of 0.78 (CI 0.64-0.89). The 27 part-specimens considered by Paige Prostate as suspicious, whose final diagnosis was benign, were found to comprise atrophy (n = 14), atrophy and apical prostate tissue (n = 1), apical/benign prostate tissue (n = 9), adenosis (n = 2), and post-atrophic hyperplasia (n = 1). Paige Prostate resulted in the identification of four additional patients whose diagnoses were upgraded from benign/suspicious to malignant. Additionally, this AI-based test provided an estimated 65.5% reduction of the diagnostic time for the material analyzed. Given its optimal sensitivity and NPV, Paige Prostate has the potential to be employed for the automated identification of patients whose histologic slides could forgo full histopathologic review. In addition to providing incremental improvements in diagnostic accuracy and efficiency, this AI-based system identified patients whose prostate cancers were not initially diagnosed by three experienced histopathologists. © 2021 The Authors. The Journal of Pathology published by John Wiley & Sons, Ltd. on behalf of The Pathological Society of Great Britain and Ireland.

Splicing and multifactorial analysis of intronic BRCA1 and BRCA2 sequence variants identifies clinically significant splicing aberrations up to 12 nucleotides from the intron/exon boundary.

Clinical management of breast cancer families is complicated by identification of BRCA1 and BRCA2 sequence alterations of unknown significance. Molecular assays evaluating the effect of intronic variants on native splicing can help determine their clinical relevance. Twenty-six intronic BRCA1/2 variants ranging from the consensus dinucleotides in the splice acceptor or donor to 53 nucleotides into the intron were identified in multiple-case families. The effect of the variants on splicing was assessed using HSF matrices, MaxEntScan and NNsplice, followed by analysis of mRNA from lymphoblastoid cell lines. A total of 12 variants were associated with splicing aberrations predicted to result in production of truncated proteins, including a variant located 12 nucleotides into the intron. The posterior probability of pathogenicity was estimated using a multifactorial likelihood approach, and provided a pathogenic or likely pathogenic classification for seven of the 12 spliceogenic variants. The apparent disparity between experimental evidence and the multifactorial predictions is likely due to several factors, including a paucity of likelihood information and a nonspecific prior probability applied for intronic variants outside the consensus dinucleotides. Development of prior probabilities of pathogenicity incorporating bioinformatic prediction of splicing aberrations should improve identification of functionally relevant variants and enhance multifactorial likelihood analysis of intronic variants.

Detection of splicing aberrations caused by BRCA1 and BRCA2 sequence variants encoding missense substitutions: implications for prediction of pathogenicity.

Missense substitutions in high-risk cancer susceptibility genes create clinical uncertainty in the genetic counseling process. Multifactorial likelihood classification approaches and in vitro assays are useful for the classification of exonic sequence variants in BRCA1 and BRCA2, but these currently rely on the assumption that changes in protein function are the major biological mechanism of pathogenicity. This study investigates the potentially pathogenic role of aberrant splicing for exonic variants predicted to encode missense substitutions using patient-derived RNA. No splicing aberrations were identified for BRCA1c.5054C>T and BRCA2c.7336A>G, c.8839G>A, and c.9154C>T. However, RT-PCR analysis identified a major splicing aberration for BRCA1c.4868C>G(p.Ala1623Gly), a variant encoding a missense substitution considered likely to be neutral. Splicing aberrations were also observed for BRCA2c.7988A>T(p.Glu2663Val) and c.8168A>G(p.Asp2723Gly), but both variant and wildtype alleles were shown to be present in full-length mRNA transcripts, suggesting that variant protein may be translated. BRCA2 protein function assays indicated that BRCA2p.Glu2663Val, p.Asp2723Gly and p.Arg3052Trp missense proteins have abrogated function consistent with pathogenicity. Multifactorial likelihood analysis provided evidence for pathogenicity for BRCA1 c.5054C>T(p.Thr1685Ile) and BRCA2c.7988A>T(p.Glu2663Val), c.8168A>G(p.Asp2723Gly) and c.9154C>T(p.Arg3052Trp), supporting experimentally derived evidence. These findings highlight the need for improved bioinformatic prediction of splicing aberrations and to refine multifactorial likelihood models used to assess clinical significance.

Bayes analysis provides evidence of pathogenicity for the BRCA1 c.135-1G>T (IVS3-1) and BRCA2 c.7977-1G>C (IVS17-1) variants displaying in vitro splicing results of equivocal clinical significance.

Although in vitro splicing assays can provide useful information about the clinical interpretation of sequence variants in high-risk cancer genes such as BRCA1 and BRCA2, results can sometimes be difficult to interpret. The BRCA1 c.135-1G>T (IVS3-1G>T) variant has been shown to give rise to an in-frame deletion of exon 5 (BRCA1 c.135_212del) that is predicted to encode 26 amino acids. BRCA2 c.7977-1G>C (IVS17-1G>C) was shown to increase the expression of two naturally occurring transcripts that contain frameshifts (BRCA2, c.7977_8311del (exon 18 deletion); BRCA2, c.7806_8331del (exon 17&18 deletion)). In this study we conducted multifactorial likelihood analysis to evaluate the clinical significance of these two variants, including assessing variant segregation in families by Bayes analysis, and breast tumor pathology features suggestive of positive mutation status. Multifactorial analysis provided strong evidence for causality for both of these variants. The Bayes scores from a single family with BRCA1 c.135-1G>T was 9528:1, and incorporation of pathology features gave an overall likelihood of causality of 28108:1. The Bayes scores from five informative families with BRCA2 c.7977-1G>C was 47401:1, and the combined Bayes-pathology odds of causality was 29389:1. Multifactorial likelihood analysis indicates that the BRCA1 c.135-1G>T and BRCA2 c.7977-1G>C variants are disease-associated mutations which should be managed clinically in the same fashion as classical truncating mutations.

Clinical classification of BRCA1 and BRCA2 DNA sequence variants: the value of cytokeratin profiles and evolutionary analysis--a report from the kConFab Investigators.

PURPOSE: Rare missense substitutions and in-frame deletions of BRCA1 and BRCA2 genes present a challenge for genetic counseling of individuals carrying such unclassified variants. We assessed the value of tumor immunohistochemical markers in conjunction with genetic and evolutionary approaches for investigating the clinical significance of unclassified variants. PATIENTS AND METHODS: We studied 10 BRCA1 and 12 BRCA2 variants identified in Australian families with breast cancer. Analyses assumed a prior probability based on revised cross-species sequence alignment methods assessing amino acid evolutionary conservation and position, combined with likelihoods from data on co-occurrence with pathogenic mutations in the same gene, segregation analysis, and immunohistochemistry. We specifically explored the value of estrogen receptor, cytokeratin 5/6, and cytokeratin 14 as tumor markers of BRCA1 mutation status. RESULTS: Posterior probabilities classified 72% of variants. BRCA1 variants IVS18+1 G>T (del exon 18) and 5632 T >A (V1838E) were classified as pathogenic, with >99% posterior probability of being deleterious, and tumor histopathology was particularly important for their classification. BRCA2 variant classification was improved over previous studies, largely by incorporating the prior probability of pathogenicity based on amino acid cross-species sequence alignments. CONCLUSION: Variant classification was considerably improved by analysis of estrogen receptor, cytokeratin 5/6, and cytokeratin 14 tumor expression, and use of updated methods estimating the clinical relevance of amino acid evolutionary conservation and position. These methodologies may assist genetic counseling of individuals with unclassified sequence variants.

In situ carcinoma - can we predict which patient will come back with a recurrence?

The frequency of in situ carcinomas has been rising since the introduction of mammographic screening. The management of patients with preinvasive disease remains difficult due to our lack of ability to accurately predict which patients will recur and progress to invasive carcinoma. Although some factors, such as lesion size and extent of margin clearance, are strong predictors of recurrence, many patients are still under- or overtreated. In this issue of Cancer Cell, Gauthier and colleagues suggest that abrogated response to cell stress measured by analysis of p16 and the proliferation marker Ki67 accurately predicts recurrence in ductal carcinoma in situ.