Simon Schnaiter

The identification of a pathogenic variant (PV) in one of the mismatch repair (MMR) genes confirms the diagnosis of Lynch syndrome (LS). Hence, the correct classification of MMR gene variants is of utmost importance for appropriate counselling, surveillance, and treatment of LS patients and their families. In 7/200 unrelated Tyrolean-suspected LS patients, we identified the rare variant MLH1:c.836T>G. Clinical and tumor data strongly indicate that this founder variant is associated with an increased risk for early-onset LS-associated tumors. We also demonstrate that the variant leads to aberrant mRNA splicing. However, the splice effect’s leakiness together with the small effect of the amino acid change p.(Val297Gly) encoded by the residual full-length transcripts in a functional assay preclude its formal classification as (likely) PV according to internationally accepted variant interpretation guidelines. The family histories of the carriers suggest that the obstacles to classify the variant as (likely) PV may be related with a reduced penetrance. Nonetheless, and despite the formal classification of MLH1:c.836T>G as a variant of uncertain significance, we show that carriers should undergo cancer surveillance and predictive testing should be offered to relatives. This variant illustrates the need for an improved classification framework for appropriate categorization of lower-penetrance alleles.

Secondary findings (SF) identified in massive parallel sequencing raise important clinical and healthcare related questions. To get an overview on current practices of European healthcare providers (HCP), we conducted a cross-sectional survey study among 39 stakeholders—predominantly senior medical and laboratory geneticists—from 15 European countries participating in the European Reference Network for Genetic Tumour Risk Syndromes (ERN GENTURIS). Respondents reported considerable heterogeneity in SF management and reporting, even within countries. While 31% of responding HCP return findings from all 81 genes on the American College of Medical Genetics and Genomics (ACMG) recommended SF list version 3.2, 41% restrict SF disclosure, often excluding genes associated with cardiological or metabolic disorders or with limited clinical actionability. A further 26% do not report ACMG-listed SF at all. Notably, 70% of HCP also assess additional cancer-predisposition genes beyond the ACMG list, using in-house gene lists or national guidelines. Most HCP restrict reporting to (likely) pathogenic variants (90%) and find SF in less than 5% of genetic analyses (59%). Consent procedures and patient information practices varied, with most HCP employing opt-in consent models and genetic counselling primarily delivered by medical geneticists and genetic counsellors. Major institutional challenges raised by participants, include lack of harmonised guidelines, concerns about patient anxiety, and insufficient resources for follow-up care. The findings of this study highlight the need for robust, evidence-based European guidelines to ensure clinically relevant and patient-centred SF management.

Background/Objectives: The implementation of polygenic scores (PGSs) and multifactorial risk assessments (MFRAs) has the potential to enhance breast cancer risk stratification, particularly in carriers of moderate-penetrance pathogenic variants (PVs), whose risk profiles often remain unclear if testing is limited to monogenic risk factors. Methods: To enhance breast cancer risk stratification, we included the BCAC313 polygenic score, together with MFRA, for carriers of moderate-penetrance pathogenic variants (PVs) during routine diagnostics and assessed its effect on the classification of patients’ risk categories in a real-world cohort at our center in its first year of implementation. Seventeen carriers with PVs in moderate-risk breast cancer genes were included in this study. Thirteen of them qualified for analysis for a full MFRA, including PGS, according to ancestry estimation and clinical criteria. The MFRA was performed using the CanRisk tool, which incorporates clinical, lifestyle, familial, and genetic data, including the BCAC313 score. Results: PGS z-scores were significantly higher in breast cancer patients compared to the unaffected control cohort (p = 0.016). The MFRA, including PGS, increased risk estimates for contralateral breast cancer in seven of eight patients with breast cancer and for primary breast cancer in three of five healthy carriers, compared to the risk conferred by the MFRA and moderate-penetrance pathogenic variant alone. Risk estimates varied widely, demonstrating the value of MFRA in personalized care. In five cases, one with a CHEK2-PV and four with an ATM-PV, the modified risk assessment contributed to the surgical decision for a prophylactic mastectomy. Conclusions: The MFRA, including PGS, provides the clinically meaningful refinement of breast cancer risk estimates in individuals with moderate-risk PVs. Personalized risk predictions can inform clinical management and support decision-making, which highlights the utility of this approach in clinical practice.

Einleitung Hochgradige Ovarialkarzinome (HGOC) weisen häufig genomische Instabilität auf, was zu multiplen Kopienzahlveränderungen (CNA) sowie Veränderungen in der Zygotizität führt. Die genomische Instabilität in HGOC hat verschiedene Ursachen, wobei die homologe Rekombinationsdefizienz (HRD) klinisch die wichtigste, aber nicht die einzige ist. CNAs gelten als zufällige Ereignisse. Während der Tumorentwicklung erfolgt eine klonale Selektion „vorteilhafter“ Kombinationen von CNAs. Ovcanalysis ist ein Tool in der Entwicklungsphase zur kumulativen qualitativen Analyse von CNAs und klinischen Parametern in ganzen HGOC-Kohorten, um die Tumorentwicklung und daraus resultierende klinisch-pathologisch unterschiedliche Patientensubgruppen besser zu verstehen. Material und Methode CNAs von 161 HGOC-Tumoren (aufgeteilt in eine Trainings- und eine Kontrollkohorte) mit detaillierten klinischen Informationen (26 Parameter je Tumor) wurden mittels DNA-Array-Analyse untersucht. Die resultierenden Daten wurden standardisiert und anschließend mit Ovcanalysis verarbeitet. Ovcanalysis ermöglicht eine integrierte Analyse der gesamten Kohorte hinsichtlich CNA- und klinischer Daten mittels Dimensionsreduktionsalgorithmen (z. B. PCA, LLE), gefolgt von einer automatisierten Clusteranalyse (k-means). Die mittleren CNA-Profile (CNAs inkl. B-Allel-Frequenz) der Cluster werden visualisiert, und eine Tabelle mit den gruppierten Tumoren und klinischen Parametern kann exportiert werden. Ergebnisse Eine funktionale Beta-Version von Ovcanalysis wurde entwickelt und wird derzeit getestet. Erste maschinelle Lernanalysen der CNAs ermöglichten eine Clusterbildung von Patienten mit unterschiedlichen klinisch-pathologischen Merkmalen über die HRD-Bestimmung hinaus. Schlussfolgerung Unsere Arbeit zeigt die Machbarkeit einer automatisierten CNA-Analyse zur Identifikation charakteristischer CNA-Ereignisse, die die Abgrenzung klinisch-pathologisch unterschiedlicher Patientensubgruppen (z. B. Alter bei Diagnose) ermöglichen. Article published online: 16 April 2025 © 2025. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Background: Homologous recombination deficiency (HRD) has evolved into a major diagnostic marker in high-grade ovarian cancer (HGOC), predicting the response to poly (adenosine diphosphate-ribose) polymerase inhibitors (PARPi) and also platinum-based therapy. In addition to HRD, the type of peritoneal tumor spread influences the treatment response and patient survival; miliary type tumor spread has a poorer predicted outcome than non-miliary type tumor spread. Methods: Known methods for HRD assessment were adapted for our technical requirements and the predictive-value integrated genomic instability score (PIGIS) for HRD assessment evolved as an outcome. PIGIS was validated in HGOC samples from 122 patients. We used PIGIS to analyze whether the type of tumor spread correlated with HRD status and whether this had an impact on survival. Results: We demonstrated that PIGIS can discriminate HRD-positive from HRD-negative samples. Tumors with a miliary tumor spread are HRD-negative and have a very bad prognosis with a progression-free survival (PFS) of 15.6 months and an overall survival (OS) of 3.9 years. However, HRD-negative non-miliary spreading tumors in our cohort had a much better prognosis (PFS 35.4 months, OS 8.9 years); similar to HRD-positive tumors (PFS 34.7 months, OS 8.9 years). Conclusions: Our results indicate that in a predominantly PARPi naïve cohort, the type of tumor spread and concomitant cytoreduction efficiency is a better predictor of survival than HRD and that HRD may be an accidental surrogate marker for tumor spread and concomitant cytoreduction efficiency. It remains to be determined whether this also applies for sensitivity to PARPi.

Background Homologous recombination deficiency (HRD) evolved into a major diagnostic marker in high grade ovarian cancer (HGOC) predicting the response to poly (Adenosine diphosphate-ribose) polymerase inhibitors (PARPi) and also platinum-based therapy. In addition to HRD, the type of peritoneal tumor spread influences treatment response and patient survival; miliary type tumor spread has poorer predicted outcome than non-miliary type tumor spread. Methods Known methods for HRD assessment were adapted for our technical requirements and the Predictive-value Integrated Genomic Instability Score (PIGIS) for HRD assessment evolved as an outcome. PIGIS was validated in HGOC samples from 122 patients. We used PIGIS to analyze whether the type of tumor spread correlated with HRD status and whether this had an impact on survival. Results We demonstrated that PIGIS can discriminate HRD positive from HRD negative samples. Tumors with miliary tumor spread are HRD negative and have a very bad prognosis with a progression free survival (PFS) of 15.6 months and an overall survival (OS) of 3.9 years. However, HRD negative non-miliary spreading tumors in our cohort had a much better prognosis (PFS 35.4 months, OS 8.9 years), similar to HRD positive tumors (PFS 34.7 months, OS 8.9 years). Conclusions Our results indicate that in a predominantly PARPi naïve cohort the type of tumor spread might be a better predictor of response to platinum sensitivity than HRD. It remains to be determined, whether this also applies for sensitivity to PARPi.

Colorectal cancer (CRC) in adolescents and young adults (AYA) is very rare. Known predisposition syndromes include Lynch syndrome (LS) due to highly penetrant MLH1 and MSH2 alleles, familial adenomatous polyposis (FAP), constitutional mismatch-repair deficiency (CMMRD), and polymerase proofreading-associated polyposis (PPAP). Yet, 60% of AYA-CRC cases remain unexplained. In two teenage siblings with multiple adenomas and CRC, we identified a maternally inherited heterozygous PMS2 exon 12 deletion, NM_000535.7:c.2007-786_2174+493del1447, and a paternally inherited POLD1 variant, NP_002682.2:p.Asp316Asn. Comprehensive molecular tumor analysis revealed ultra-mutation (>100 Mut/Mb) and a large contribution of COSMIC signature SBS20 in both siblings’ CRCs, confirming their predisposition to AYA-CRC results from a high propensity for somatic MMR deficiency (MMRd) compounded by a constitutional Pol δ proofreading defect. COSMIC signature SBS20 as well as SBS26 in the index patient’s CRC were associated with an early mutation burst, suggesting MMRd was an early event in tumorigenesis. The somatic second hits in PMS2 were through loss of heterozygosity (LOH) in both tumors, suggesting PPd-independent acquisition of MMRd. Taken together, these patients represent the first cases of cancer predisposition due to heterozygous variants in PMS2 and POLD1. Analysis of their CRCs supports that POLD1-mutated tumors acquire hypermutation only with concurrent MMRd.