Germline findings in patients with advanced malignancies screened with paired blood-tumour testing for personalised treatment approaches

BACKGROUND

Sequencing of tumour tissue with comprehensive gene panels is increasingly used to guide treatment in precision oncology. Analysis of tumour-normal pairs allows in contrast to tumour-only assessment direct discrimination between somatic and germline alterations, which might have important implications not only for the patients but also their families.

METHODS

We performed tumour normal sequencing with a large gene panel in 1048 patients with advanced cancer to support treatment decision. Sequencing results were correlated with clinical and family data.

RESULTS

We identified 156 likely pathogenic or pathogenic (LP/P) germline variants in cancer predisposition genes (CPGs) in 144 cases (13.7%). Of all patients, 8.8% had a LP/P variant in autosomal-dominant cancer predisposition genes (AD-CPGs), most of them being genes with high or moderate penetrance (ATM, BRCA2, CHEK2 and BRCA1). In 48 cases, the P/LP variant matched the expected tumour spectrum. A second variant in tumour tissue was found in 31 patients with AD-CPG variants. Low frequency mutations in either TP53, ATM or DNMT3A in the normal sample indicated clonal haematopoiesis in five cases.

CONCLUSIONS

Tumour-normal testing for personalised treatment identifies germline LP/P variants in a relevant proportion of patients with cancer. The majority of them would not have been referred to genetic counselling based on family history. Indirect functional readouts of tumour-normal sequencing can provide novel links between CPGs and unexpected cancers. The interpretation of increasingly complex datasets in precision oncology is challenging and concepts of interdisciplinary personalised cancer prevention are needed to support patients and their families.

PolyQ-expanded ataxin-3 protein levels in peripheral blood mononuclear cells correlate with clinical parameters in SCA3: a pilot study

In view of upcoming clinical trials, quantitative molecular markers accessible in peripheral blood are of critical importance as prognostic or pharmacodynamic markers in genetic neurodegenerative diseases such as Spinocerebellar Ataxia Type 3 (SCA3), in particular for signaling target engagement. In this pilot study, we focused on the quantification of ataxin-3, the protein altered in SCA3, in human peripheral blood mononuclear cells (PBMCs) acquired from preataxic and ataxic SCA3 mutation carriers as well as healthy controls, as a molecular marker directly related to SCA3 pathophysiology. We established two different highly sensitive TR-FRET-based immunoassays to measure the protein levels of either total full-length, non-expanded and expanded, ataxin-3 or specifically polyQ-expanded ataxin-3. In PBMCs, a clear discrimination between SCA3 mutation carrier and controls were seen measuring polyQ-expanded ataxin-3 protein level. Additionally, polyQ-expanded ataxin-3 protein levels correlated with disease progression and clinical severity as assessed by the Scale for the Assessment and Rating of Ataxia. Total full-length ataxin-3 protein levels were directly influenced by the expression levels of the polyQ-expanded ataxin-3 protein, but were not correlated with clinical parameters. Assessment of ataxin-3 levels in fibroblasts or induced pluripotent stem cells allowed to distinguish mutation carriers from controls, thus providing proof-of-principle validation of our PBMC findings across cell lines. Total full-length or polyQ-expanded ataxin-3 protein was not detectable by TR-FRET assays in other biofluids like plasma or cerebrospinal fluid, indicating the need for ultra-sensitive assays for these biofluids. Standardization studies revealed that tube systems, blood sampling, and PBMC preparation may influence ataxin-3 protein levels indicating a high demand for standardized protocols in biomarker studies. In conclusion, the polyQ-expanded ataxin-3 protein is a promising candidate as a molecular target engagement marker in SCA3 in future clinical trials, determinable even in—easily accessible—peripheral blood biomaterials. These results, however, require validation in a larger cohort and further standardization of modifying conditions.