POT1 tumour predisposition: a broader spectrum of associated malignancies and proposal for additional screening program

Protection of Telomeres Protein 1 (POT1) protein is an essential subunit of the shelterin telomere binding complex, regulating telomere length. Some POT1 gene pathogenic variants (PV) lead to telomere elongation, genomic instability and higher risk of cancer. POT1 tumour predisposition syndrome (POT1-TPD) has autosomal dominant inheritance and unknown penetrance. It is associated with increased risk of cutaneous melanoma, chronic lymphocytic leukaemia, angiosarcoma and gliomas. In this work, we aim to describe a broader cancer phenotype related to POT1-TPD, in three families (two with a four generation pedigree, one with a five generation pedigree). The three index cases were referred to our oncogenetic centre for genetic counselling due to their personal history of cancer. Two underwent clinical exome sequencing of 4,867 genes associated with Mendelian genetic diseases, and another underwent gene panel sequencing including POT1, which identified three different POT1 PV: NC_000007.14(NM_015450.2):c.349C>T; NC_000007.14(NM_015450.2):c.233T>C and NC_000007.14(NM_015450.2):c.818G>A; already described in the literature. Referenced relatives, did a target genetic test (according to the POT1 PV identified in the family). In total, 37 individuals were tested (51.4% females), median age of 46 (22–81) years, with POT1 PV detected in 22. POT1-TPD was observed, but also a higher incidence of other cancers (other sarcomas, papillary thyroid cancer, early onset prostate cancer and leukaemia). These findings contribute to an increase in our knowledge about POT1 PV, and it can play a role in the definition of future POT1 PV screening criteria, POT1 carrier surveillance protocols (possibly considering screening for all types of sarcomas) and in genetic counselling.


INTRODUCTION
Protection of Telomeres Protein 1 [1] gene is located on chromosome 7 (7q31.33)and POT1 protein is an essential subunit of the shelterin telomere binding complex [1,2].It binds to the telomeric overhangs, preventing the activation of DNA damage response at telomeres and regulating telomere length [2,3].POT1 pathogenic variants (PV) can present with different phenotypes [3].Some variants cause telomere truncations leading to dysfunctional telomeres and causing a telomeric syndrome known as Coats Plus Syndrome, others cause telomere shortening that leads to as idiopathic pulmonary fibrosis [3].On the other hand, there are POT1 variants that cause an opposite telomere phenotype, allowing telomere elongation which leads to prolongation of cell lineage's life span, facilitating the acquisition of several somatic mutations [4,5].This loss of the tumour-suppressor mechanism of telomere shortening leads to clonal population expansion and genomic instability, predisposing to a higher risk of cancer [3,6].
The penetrance of POT1-TPD is currently unknown, since only several hundred probands have been tested [1,13].Therefore, the full phenotypic spectrum and penetrance of this syndrome is yet to be determined [1,3].
POT1-TPD diagnosis is made using a molecular genetic test, with the detection of a heterozygous germline PV in the POT1 gene [1].It should be suspected in people with multiple CM, one of POT1-TPD core cancers (CM, CLL, angiosarcoma or glioma) and a firstor second-degree relative with a confirmed POT1-PTD cancer, or a somatic POT1 PV identified on tumour tissue sequencing [1].
There is no targeted treatment available for POT1 PV.POT1-TPD is treated according to the standard of care for each type of tumour [1,2].However, longer telomeres and upregulation of POT1 were associated with resistance to radiotherapy, in cell line studies [21][22][23].Therefore, the development of POT1 and telomerase inhibitors may be a potential approach to enhance radiosensitivity in these tumours [23].
Most of the tumours related to POT1 PV are diagnosed in adulthood [1,3].The age of onset for first primary tumour described is 15 years [12], therefore all the surveillance procedures are recommended to start at age 18 or two to five years earlier than the earliest diagnosis in the family [1].There are no published guidelines for surveillance of POT1 PV carriers, the following recommendations are based on an expert opinion publication about POT1-TPD that also addresses this topic [1].For POT1 mutation carriers, a comprehensive physical examination with careful annual examination of lymph nodes, full skin examination by a dermatologist (every three to six months in individuals with multiple atypical naevi, personal history of CM, and/or family history of CM) and an annual complete blood count is recommended [1].In families with Li-Fraumeni syndrome or Li-Fraumeni-like criteria, an annual whole-body MRI is recommended [1].It can also be considered in other carriers depending on personal and family history of noncutaneous and non-brain malignancies (every one to two years) [1].In families with glioma cases, a brain MRI every one to two years is recommended [1].
With this work, we aim to describe all types of tumours diagnosed in three families carrying a POT1 PV, both malignant diseases previously described as associated with POT1-TPD and also other cancer types not described or less often reported as part of this syndrome, showing a broader phenotype of POT1 associated tumours.

MATERIALS AND METHODS
Probands were referred to our oncogenetic centre, Jules Bordet Institute, in Brussels, Belgium.Cancer diagnoses were confirmed by pathological specimen review (in index cases of families A and B), medical records (in index case of family Cpreviously diagnosed and treat in another medical centre; and in all the tested relatives from families A, B and C), detailed direct anamnesis and also based on self or family report.All patients signed an informed consent for genetic testing.
We performed for index cases of families A and B a clinical exome sequencing of 4867 genes associated with mendelian genetic diseases, filtered to analyse a panel of cancer susceptibility genes.Index case of family C did a gene panel sequencing, which included POT1.
Relatives referenced to our centre, did a target gene analysis, which was performed with PCR amplification followed by direct Sanger sequencing of exon 8/7/10 (according to the mutation found in the family) of POT1 gene.Our reference sequence is the coding sequence NM_015450 (A of ATG = 1).In all cases, a control of the result was made on an independent sample.
All tested individuals were evaluated through a genetic counselling consultation and received pre-genetic testing psychological support.

RESULTS
In total, 37 individuals were tested: 19 females and 18 males, median age of 46  years, between 2019 and 2023.
Regarding more distant relatives, there was history of a CM at age 42 (Fig. 1 In total, genetic tests were performed on twenty-seven family relatives (twenty-one of them as pre-symptomatic tests).The The individual III.9 was diagnosed with low grade glioma (at age 50) after brain MRI, prescribed as part of the screening program for POT1 carriers.
The individual III.16, besides having a glioblastoma had a negative test for the POT1 familiar variant.

Family B
The index case (Fig. 2, III.3) is a male patient, also with Ashkenazi Jewish ancestry, with history of left leg and arm liposarcomas at age 73, a PTC at age 74.Regarding first degree family members, there was history of two leukaemia at age 70 (Fig. 2, II.3) and 83 (Fig. 2, III.1), one CM at age 30 (Fig. 1, IV.7), one prostate cancer at age of 60 (Fig. 2, III.1); one relative with colon carcinoma at age 65, lung cancer at age 81 and brain tumour at an unknown age of diagnosis (Fig. 2, III.2); and another relative with breast cancer at age of 44 and lung cancer at the age of 68 (Fig. 2, III.4).The genetic testing of the index patient detected a PV in POT1 gene: NC_000007.14(NM_015450.2):c.233T>C(Table 1).
In total, four other relatives were tested (pre-symptomatic testing for three of them).The NC_000007.14(NM_015450.2):c.233T>Cvariant was detected in the two daughters of the proband, one of them having history of CM (Fig. 2 IV.7).

Family C
The index case (Fig. 3, III.8) is a male patient, with a history of prostate cancer at age 47 and an undifferentiated spindle cell sarcoma of the left arm at age 57.The patient had previously undergone genetic tests in 2009 and 2011 in other institutions with a panel of several genes (including BRCA1, BRCA2, PTEN, CHEK2, TP53 as well as Lynch syndrome genes) without any anomaly detected.Considering his personal (a second cancer diagnosisarm sarcoma) and family history, an additional genetic analysis was proposed in 2022 in order to search for mutations in other genes, in particular on POT1 whose spectrum could correspond to the history of cancers reported and was included in our oncogenetic gene panel.This third genetic test detected a POT1 pathogenic variant: NC_000007.14(NM_015450.2):c.818G>A(Table 1).
Regarding the maternal side of the family, only two cases with oncological history (first-and second-degree relatives) and in older ages were present: one CM at age 72 (Fig. 3, II.6) and one colon cancer at age 70 (Fig. 3, I.4).The mother was tested and POT1 mutation was not detected.Several members had a history of cancer on the paternal side: a gastric cancer at unknown age of diagnosis (Fig. 3, I.1), two kidney cancers (Fig. 3, II.1at age 54, and II.5at age 60), two colon cancers (Fig. 3, II.1after 65 years old, and Fig. 3, III.3unknown age), a breast cancer after 50 years old (Fig. 3, II.3), a left leg STS at age of 40 (Fig. 3, II.4), two early onset prostate cancers (Fig. 3, II.5under 60 years old, and III.6 at age 40), an oesophageal cancer at unknown age of diagnosis (Fig. 3, II.5), a lung cancer at age 70 (Fig. 3, II.5) and a gynaecological cancer at unknown age (Fig. 3, III.4).Despite this frequent cancer incidence on the paternal side of the family, only one of these relatives was pre-symptomatically tested (Fig. 3, III.1), and had a negative result.
Table 2 summarizes the types of tumours in these tree families, according to POT1 status and Table 3 specifies the age of onset of each tumour per individual.

DISCUSSION
In family A, the NC_000007.14(NM_015450.2):c.349C>Tpathogenic variant was identified, which leads to replacement of an arginine with a cysteine at amino acid 117 (p.Arg117Cys) (Table 1).This has already been reported in three Li-Fraumeni like families (with cardiac angiosarcomas and other STS) [24].These individuals  had reduced telomere-bounded POT1 levels, with longer more fragile telomeres and one mutation carrier also developed cutaneous melanoma (CM) [24].In a recent study that analysed more than 1500 cases of sarcoma probands, POT1 PV was identified in six and associated to familial melanoma pedigree in two of these [19].Further, another study reported higher prevalence of CM (13.2%) and sarcomas (3.5%) among POT1 PV carries [13].These findings are in accordance with the observations in family A: index case with STS and CM, two relatives with early onset cardiac angiosarcoma (two with other STS) and another one with CM.Nonetheless, two cases of brain tumours and a case of lymphoproliferative syndrome were identified, which have not been described before in association with this specific POT1 PV, but are known to be related to POT1-TPD, reinforcing the recognised disease spectrum of POT1 mutations [1].On the other hand, one can observe that individuals with the mutation also developed other types of cancers that have been described in association with mutation of POT1, but with less evidence, such as colon cancer, lung cancer, leukaemia and PTC [6,8,13,16,17].Also, the relative III.30 developed a bone giant cell tumour, which despite not being a malignant tumour is clinically relevant, and was not previously associated with POT1 PV. Surprisingly, one case of glioblastoma was not related with the familial POT1 mutation (Fig. 1, III.16), and was thus a phenocopy.This phenomenon is more likely to occur in large families, such as family A.
Regarding family B, we detected the NC_000007.14(NM_015450.2):c.233T>Cpathogenic variant, which results in isoleucine replaced by a threonine in POT1 protein (p.Ile78Thr) (Table 1).This PV has been involved in familial melanoma (three of these families with self-reported Jewish descent as in family B), lymphoid and myeloid neoplasms [6,[25][26][27].In family B, there is one case of CM (in a carrier) and two cases of leukaemia (in non-tested relatives), in accordance with the previous descriptions regarding this PV.This family had a broader cancer spectrum: STS, PTC, brain tumour, prostate, colon, lung, gastric and breast cancer, although mostly in non-tested individuals.Therefore, we cannot formally establish a causal relationship between these tumours and the presence of the POT1 PV.Nonetheless, by analysing the index case it is possible to suppose that this family harbours a broader spectrum of POT1-TPD, since the patient was diagnosed with two STS (other than angiosarcomas) and a PTC, already described as associated with POT1 mutation, but not related to this specific PV, neither considered in the surveillance recommendations as a POT1-TPD [1,17,19].
In family C, NC_000007.14(NM_015450.2):c.818G>Avariant in POT1 gene was detected in the proband (p.Arg273Gln) (Table 1).This variant was already described in association with CM [8] and also related to a higher risk of lymphoid and myeloid clonal haematopoiesis [6].All predictive in silico tools, that evaluate the effect of missense changes on protein structure and function, suggest that this variant is likely to be disruptive [28].These arguments, and the good concordance with the phenotype, Table 2. Tumour types of different members of the studied families according to POT1 status.
In grey background, there are the tumours considered strongly related to POT1 mutations.AS angiosarcoma, BT brain tumour, CM cutaneous melanoma, CRC colorectal cancer, GCBT giant cell tumour of bone, Haem haematological, NT non-tested, POT1 pv POT1 pathogenic variant, PTC papillary thyroid carcinoma.
indicate that the variant is likely to be pathogenic.In family C, there was only one case of CM in a non-carrier (Fig. 3, II.6).Since the mother was negative for the POT1 mutation, we can hypothesise that this is case of de novo POT1 PV or a case of paternal inheritance.This last hypothesis seems the more probable due to the wider spectrum of cancers on that side of the family, although none of those relatives were tested.The tested carrier in this family (Fig. 3, II.5) presented an early onset prostate cancer and a STS, tumours not traditionally associated with POT1-TPD, but already reported in another POT1 related study [13].
In these three families, it was possible to observe the typical POT1-TPD: two cases of CM, two cases of cardiac angiosarcoma, one case of CLL and two cases of brain tumours, in total.Nevertheless, a broader cancer spectrum related to POT1 mutations is described.It was possible to observe several sarcomas other than angiosarcomas (six in total).As referred before, there is growing evidence suggesting that POT1 PV could increase the risk of sarcoma and not just angiosarcoma [2,13,18,19,24].Regarding CRC and PTC, already mentioned as being associated to POT1-PTD, but not included in the recommendations for screening and surveillance of these patients, here we observed one CRC case and two PTC in carriers, reinforcing this possible association [1,16,17].Another POT1 PV (p.V29L) was described in a family with PTC [17].However, it is known that several other hereditary syndromes are associated with differentiated thyroid cancer [29], so further studies are needed regarding POT1 mutations and the risk of PTC.Concerning POT1-TDP and CRC risk, a study with thousands of patients affected with CRC detected three POT1 PV in affected individuals, pointing this gene as a candidate for CRC susceptibility genes [16].
Regarding haematological diseases, it is well documented that POT1 mutations increase CLL risk [8,12,30].Moreover, a recent study showed the relationship between long telomeres due to POT1 mutations and familial clonal haematopoiesis syndrome, which broadens the spectrum of predisposition to malignant haematological diseases [6].The p. Ile78Thr variant identified in family B was already described to be associated with CLL and myelodysplastic syndromes in a large study [27].Indeed, family B has two cases of leukaemia, although in non-tested individuals.A third case of leukaemia was described in family A, in an obligate carrier.Despite the absence of leukaemia reported in association to the p. Arg117Cys variant, considering the recent data on the increased risk of malignant haematological diseases associated to POT1 mutations [6], we cannot exclude the association between leukaemia and this POT1 PV.Future studies will be important to investigate the link between specific POT1 PV and the incidence of haematological diseases.
Lung cancer was observed in six family relatives (including two p.Arg117Cys variant carriers), but there is no description of its association with POT1-TPD.A study with more than 30,000 lung cancer patients showed that POT1 mutations play a role in lung cancer predisposition [15], however, without further data it is not possible to draw a clear conclusion on the association of POT1 PV and lung cancer.
Finally, early onset prostate cancer was present in three individuals (one carrier, one obligate carrier and one non-tested).A recent study reported higher prostate cancer among men with POT1 PV, but with a median age of onset of 67.5 years [13].Nonetheless, these cases can indicate a possible link that needs further investigation.
Several pre-symptomatic tests were performed on relatives of the three families, with the detection of POT1 mutations in 16 individuals.All the individuals carrying a POT1 PV underwent a surveillance program according to the current recommendations [1].The individual III.9 (Fig. 1) was diagnosed with low grade glioma at age 50, after a brain MRI performed as part of the surveillance program.No other tumours were detected in other individuals.However, itis important to note that most of them are young and they are under this surveillance program for one to four years only.
This study has some limitations that are important to mention, the small number of individuals tested in families B and C, the fact that the test is not carried out consecutively on all family members, but only on the ones with indication and willing to do it.Also, cancer history of relatives, mainly regarding the first families' generations, was based on family-reporting information, which may contribute to less accurate data.Despite these, the wide pattern of cancers observed in the three families suggests a larger POT1-TPD spectrum than the one previously described [1].The presence of a POT1 PV, its penetrance, genetic modifiers and other external factors are possible causes for the different phenotypes observed.We estimate that POT1-TPD is probably underdiagnosed with the current recommendation criteria prompting to search for germline PV.Regarding the age to start testing patients and relatives, our findings are in accordance with the current recommendations, since all individuals in these families developed cancer in their adulthood [1].
According to our findings, we consider that it would be important to discuss the inclusion of all types of sarcomas in the screening criteria for POT1 mutations.Also, the observation of early prostate cancer in these families may be a point to discuss with patients, namely the role of prostate cancer screening, on an individual basis and taking into account patient family history.Current recommendations already recommend a whole-body MRI for screening of cardiac angiosarcoma, which is the ideal screening exam for other types of sarcoma too [1].The main recommendations suggest a surveillance protocol similar to Li-Fraumeni-like patients, with whole-body MRI and in POT1 carriers not fulfilling these criteria, MRI should be considered depending on personal and family history of non-cutaneous, nonbrain malignancies [1,20].This is an important topic, since it is estimated that POT1 PV carriers have 6 times more risk of developing a sarcoma than POT1 wild type individuals [13].A comprehensive physical exam and a complete blood count are already proposed to screen for LLC and will be useful in screening other haematological malignancies as well.We also hypothesise the clinical utility of earlier start of prostate cancer screening with annual PSA dosage from age 40-45 years, or 10 years before the youngest case diagnosed in the family, on an individual basis, as discussed before, since it is a non-invasive and non-expensive test.Another point to consider in surveillance programs would be behavioural measures such us smoking cessation, avoidance of smoke and occupational carcinogens, eating a healthy diet and exercising regularly.
Despite the limitations of this study, we conclude that it suggests that POT1 germline PV are associated with a broader spectrum of hereditary cancer than the previously described POT1-TPD.Since some of the reported tumours are also common in the general population, a clear definition of POT1 tumour spectrum is hard to make.Nonetheless, our observations align with recent studies also reporting sarcoma (and cardiac angiosarcoma), CRC and PTC associated to POT1 PV [11,13,14,[16][17][18][19].Moreover, the cases of leukaemia described in these POT1 carrier families are in accordance with more recent data about the link between POT1 PV and the higher risk of B/Tcell lymphoproliferative and myeloproliferative diseases [6].These findings are important to better understand the implications of POT1 PV, its prevalence and penetrance.Furthermore, findings of a broader spectrum of diseases related to POT1-TPD should be considered for future guidelines, not only about the testing criteria for POT1 PV, but also to adapt the surveillance program to these other malignancies within the POT1-TPD (mainly sarcomas).Further studies on POT1 PV, their penetrance and associated types of cancers are needed, especially prospective trials with large cohorts of patients, in order to improve the knowledge of cancer genetic mechanisms, POT1 cancer spectrum and also the genetic counselling for these patients and their families.

Table 1 .
Germline variants shared by the affected individuals that were analysed according to their family.

Table 3 .
Tumour types of different members of the studied families according to POT1 status and age of diagnosis.