Heritable TP53-related cancer syndrome in Sweden : characterisation of genotype-phenotype correlation and surveillance

Abstract

Around 25% of all cancers are considered as familial and are caused by an inherited susceptibility to develop certain tumours. But only 5-10% are hereditary and caused by known high risk cancer genes associated with specific cancer risk syndromes. One of the most pronounced is the heritable TP53 related cancer (hTP53rc) syndrome, commonly referred to as the Li-Fraumeni syndrome.

In 1969, two physicians, Li and Fraumeni, identified and described a new hereditary cancer predisposition syndrome. Twenty years later, germline pathological variants of the TP53 gene – also referred to as “the guardian of the genome”, was identified to be the cause. The Li-Fraumeni syndrome (LFS) is characterised by an extreme life-time risk of cancer in germline TP53 (gTP53) carriers, up to a 100% by the age of 70. The most commonly occurring cancers are (early onset) breast cancer, CNS tumours, adrenocortical tumours, and sarcomas. In addition, individuals with a gTP53 variant are also more prone to develop other types of cancers in comparison with the general population. Since the first descriptions of LFS with childhood tumours, a broader phenotype has emerged, with families prone to mainly breast cancer, or mainly adult-onset cancers. Therefore, the term “heritable TP53 related cancer syndrome (hTP53rc) has been proposed to include the wide phenotypic range, not always explained by the genotype.

This thesis aims to explore the genotype-phenotype correlations in the Swedish gTP53 cohort (Paper I), and to evaluate an extended clinical handling including surveillance within a prospective study, The Swedish TP53 Study (SWEP53, Papers II-IV).

Within Paper I, we identified all gTP53 carriers in Sweden since testing for the gene started. This retrospective national characterisation identified 188 carriers in 91 families harbouring 47 different gTP53 variants. After reclassification according to the latest TP53specific criteria from 2021, 42 of the gTP53 variants were clinically actionable (class 4 or 5) and identified in the remaining 83 families and 176 carriers. These families fulfilled one of four different phenotypic characteristics; classical LFS (13 families), Chompret criteria (37 families), hereditary breast cancer (29 families), or other (4 families). The most commonly occurring gTP53 variant c.542G>A/p.R181H was identified in 18 families and were considered as a potential Swedish founder variant, mainly associated with hereditary breast cancer. However, this variant need further evaluations to explore its founder potential.

The prospective SWEP53 study is the base for Papers II-IV. In Paper II, we describe the outline of the Swedish surveillance program including (for adults) yearly whole-body MRI (WB-MRI), surveillance for children including abdominal ultrasound, the collection of cell-free DNA, and the psychosocial evaluation of surveillance participation. In this study, comparisons are done with the pivotal first protocol of WB-MRI surveillance, the ‘Toronto protocol’ from 2011. A notable difference is the surveillance protocol for children, , as children below the age of 15 years were neither carrier tested nor offered annual WB-MRI in SWEP53. Paper II also adds data to the first Swedish registry on gTP53 carriers established in Paper I, collecting information on the different gTP53 variants, pedigree data, ages of tumour onset, and tumour types.

An evaluation of the consequences of WB-MRI surveillance in terms of radiological findings, anatomical distribution, and the need for further work-up, was performed within Paper III. A total of 61 asymptomatic gTP53 carriers performed a baseline WB-MRI and 19 (19/61=31%) individuals had in total 30 radiological lesions that needed further workup. Three of the 19 (3/19=16%) participants were diagnosed with a malignant disease, which is considered to be a high proportion. Notably, all three were asymptomatic at baseline imaging. Because of the complexity to correctly handle all the findings and incidental findings in this high-risk group, we recommend a multidisciplinary approach to ensure the best possible care.

With Paper IV, we also wanted to evaluate the psychosocial aspects of a surveillance strategy with WB-MRI. There are few available quantitative studies within this area. We measured the perceived benefits and barriers to surveillance, cancer worry scale (CWS) and perceived overall health (The 36-Item Short Form Health Survey, SF-36) by comparing participants with and without previous cancer. They were evaluated both at baseline WBMRI and at the one-year surveillance. In general, gTP53 carriers with previous cancer reported higher levels of cancer worry than those without previous cancer, but, notably, at one year of surveillance the cancer worry was not increased. Both groups reported few barriers and scored high on the benefits. In conclusion, surveillance with WB-MRI did not increase cancer-specific worry, and was perceived as feasible by the participants regardless of previous cancer.

To summarise, the work presented in this thesis adds to the knowledge of the rare hTP53rc syndrome, both in the understanding of genotype and phenotype presentations, and in clinical handling in terms of genetic counselling and surveillance. It has also contributed to the implementation of new work-flows for gTP53 carriers within the health care system to improve the overall care of this families.