Investigator
Radboud University Medical Center
Evidence for pathogenicity of BRCA2 c.8351G>A p.(Arg2784Gln) and the challenges in classification of pathogenic variants with reduced penetrance
Background The BRCA2 c.8351G>A p.(Arg2784Gln) variant has long been classified as a variant of uncertain significance (VUS) due to conflicting evidence used in variant classification. This study aims to clarify its pathogenicity and associated risks for breast and ovarian cancer. Methods This study was conducted by the international Evidence-based Network for the Interpretation of Germline Mutant Alleles consortium. We collected data from 29 informative families with this variant. Co-segregation likelihood ratios (LRs) were calculated using the full-likelihood method to assess pathogenicity, and cancer risks were estimated with modified segregation analysis. Results Co-segregation analysis using a grid search across scaled penetrance levels for BRCA2 truncating variants yielded the strongest evidence in favour of pathogenicity, with LR maximised at approximately 20% of full penetrance (LR=11.026). Furthermore, estimated breast cancer risks were markedly higher for early onset breast cancer; women diagnosed at <50 years had a HR of 4.5, compared with a HR of 1.65 for women diagnosed at ≥50 years. The estimated lifetime risks were 25% for breast cancer and 6% for ovarian cancer. Evidence of pathogenicity was also supported by the presence of the variant allele in two patients with Fanconi anaemia. Conclusions Our results indicate that BRCA2 c.8351G>A p.(Arg2784Gln) has a disease-causing effect, with reduced penetrance, similar to other pathogenic variants in moderate risk breast cancer genes such as ATM and CHEK2 . We also provide risk-adapted recommendations for clinical management. Importantly, one should be aware of a reduced penetrance as the underlying reason for conflicting results among pieces of evidence used for variant classification.
The genomic landscape of breast and non-breast cancers from individuals with germline CHEK2 deficiency
Abstract CHEK2 is considered to be involved in homologous recombination repair (HRR). Individuals who have germline pathogenic variants (gPVs) in CHEK2 are at increased risk to develop breast cancer and likely other primary cancers. PARP inhibitors (PARPi) have been shown to be effective in the treatment of cancers that present with HRR deficiency—for example, caused by inactivation of BRCA1/2. However, clinical trials have shown little to no efficacy of PARPi in patients with CHEK2 gPVs. Here, we show that both breast and non-breast cancers from individuals who have biallelic gPVs in CHEK2 (germline CHEK2 deficiency) do not present with molecular profiles that fit with HRR deficiency. This finding provides a likely explanation why PARPi therapy is not successful in the treatment of CHEK2-deficient cancers.
Microsatellite instability in noncolorectal and nonendometrial malignancies in patients with Lynch syndrome
Abstract Background Individuals with Lynch syndrome are at increased hereditary risk of colorectal and endometrial carcinomas with microsatellite instability (MSI-H) and mismatch repair-deficiency (dMMR), which make these tumors vulnerable to therapy with immune checkpoint inhibitors. Our aim is to assess how often other tumor types in these individuals share these characteristics. Methods We retrieved the full tumor history of a historical clinic-based cohort of 1745 individuals with Lynch syndrome and calculated the standardized incidence ratio for all tumor types. MSI status, somatic second hit alterations, and immunohistochemistry-based MMR status were analyzed in 236 noncolorectal and nonendometrial malignant tumors. Results In individuals with Lynch syndrome MSI-H/dMMR occurred both in Lynch-spectrum and in non–Lynch-spectrum malignancies (85% vs 37%, P &lt; .01). MSI-H/dMMR malignancies were found in nearly all non–Lynch-spectrum tumor types. A high percentage (33%) of breast carcinomas with medullary features was observed, and most of them were MSI-H/dMMR. Breast carcinoma with medullary features were shown to be associated with Lynch syndrome (standardized incidence ratio = 38.8, 95% confidence interval = 16.7 to 76.5). Conclusions In individuals with Lynch syndrome, MSI-H/dMMR occurs in more than one-half of the malignancies other than colorectal and endometrial carcinomas, including tumor types without increased incidence. The Lynch-spectrum tumors should be expanded to breast carcinomas with medullary features. All malignancies in patients with Lynch syndrome, independent of subtype, should be tested for MSI-H/dMMR in case therapy with immune checkpoint inhibitors is considered. Moreover, Lynch syndrome should be considered an underlying cause of all MSI-H/dMMR malignancies other than colorectal and endometrial carcinomas.
Histopathological phenotyping of cancers in PTEN Hamartoma Tumor Syndrome for improved recognition: A single‐center study
AbstractPTEN hamartoma tumor syndrome (PHTS) has a broad clinical spectrum including various benign and malignant tumors at varying age of diagnosis. Many patients remain unrecognized, unaware of their increased cancer risk. We aimed to describe the cancer spectrum, age of onset and histopathological cancer characteristics to assess whether specific cancer characteristics could improve PHTS recognition. Genetic testing results and pathology reports were collected for patients tested for germline PTEN variants between 1997 and 2020 from the diagnostic laboratory and the Dutch nationwide pathology databank (Palga). The cancer spectrum and age of onset were assessed in patients with (PTENpos) and without (PTENneg) a germline PTEN variant. Histopathological cancer characteristics were assessed in a nested cohort. 341 PTENpos patients (56% females) and 2882 PTENneg patients (66% females) were included. PTENpos patients presented mostly with female breast (BC, 30%), endometrial (EC, 6%), thyroid (TC, 4%) or colorectal cancer (4%). PTENpos were significantly younger at cancer onset (43 vs. 47 years) and had more often (46% vs. 18%) a second BC than PTENneg. PTEN detection rates were highest for BC <40 years (9%), TC <20 years (15%) and EC <50 years (28%), and dropped to 6%, 4%, and 15% by age 60. Histopathological characteristics were similar between groups. No histopathological cancer characteristics were distinctive for PHTS. However, PTENpos were significantly younger at cancer onset. Therefore early‐onset BC, EC, or TC warrants consideration of PHTS diagnostics either through a pre‐screen for other PHTS features or direct germline testing.
