CrossMark
ORIGINAL ARTICLE
TP53 and CDKN1A mutation analysis in families with Li-Fraumeni and Li-Fraumeni like syndromes
Raissa Coelho Andrade1 . Anna Claudia Evangelista dos Santos1 . Joaquim Caetano de Aguirre Neto2 · Julián Nevado3 · Pablo Lapunzina3 · Fernando Regla Vargas1,4,5
@ Springer Science+Business Media Dordrecht 2016
Abstract Li-Fraumeni and Li-Fraumeni like syndromes (LFS/LFL) represent rare cancer-prone conditions associ- ated mostly with sarcomas, breast cancer, brain tumors, and adrenocortical carcinomas. TP53 germline mutations are present in up to 80 % of families with classic Li-Fraumeni syndrome, and in 20-60 % of families with Li-Fraumeni like phenotypes. The frequency of LFS/LFL families with no TP53 mutations detected suggests the involvement of other genes in the syndrome. In this study, we searched for mutations in TP53 in 39 probands from families with cri- teria for LFS/LFL. We also searched for mutations in the gene encoding the main mediator of p53 in cell cycle arrest, CDKN1A/p21, in all patients with no mutations in TP53. Eight probands carried germline disease-causing mutations in TP53: six missense mutations and two partial gene deletions. No mutations in CDKN1A coding region were detected. TP53 partial deletions in our cohort repre- sented 25 % (2/8) of the mutations found, a much higher frequency than usually reported, emphasizing the need to
search for TP53 rearrangements in patients with LFS/LFL phenotypes. Two benign tumors were detected in two TP53 mutation carriers: an adrenocortical adenoma and a neu- rofibroma, which raises a question about the possible implication of TP53 mutations on the development of such lesions.
Keywords Li-Fraumeni syndrome · Li-Fraumeni like syndrome · TP53 · CDKNIA · Gene deletion
Introduction
Li-Fraumeni syndrome (LFS) represents a rare cancer- prone condition associated to germline mutations in the TP53 gene [1, 2]. The classic phenotype of this syndrome was clinically defined before the identification of germline mutations in TP53, and it consists in the following criteria: a family with one proband diagnosed with a sarcoma before age 45 years, plus one first-degree relative with any cancer before age 45 years, and another first- or second- degree relative with any cancer before age 45 years or a
Electronic supplementary material The online version of this article (doi:10.1007/s10689-016-9935-z) contains supplementary material, which is available to authorized users.
☒ Fernando Regla Vargas fernando.vargas@ioc.fiocruz.br
Raissa Coelho Andrade randrade.inca@gmail.com
Anna Claudia Evangelista dos Santos acsantos@inca.gov.br
Joaquim Caetano de Aguirre Neto caetanoaguirre@hotmail.com
Julián Nevado jnevado@salud.madrid.org
Pablo Lapunzina pablo.lapunzina@salud.madrid.org
1 Genetics Division, Instituto Nacional de Câncer, Rio de Janeiro, Brazil
2 Clínica de Oncologia Pediátrica da Santa Casa de Belo Horizonte, Belo Horizonte, Brazil
3 INGEMM, Instituto de Genética Médica y Molecular, IdiPAZ-CIBERER, Universidad Autónoma de Madrid, Madrid, Spain
4 Genetics and Molecular Biology Department, Universidade Federal do Estado do Rio de Janeiro, Rio de Janeiro, Brazil
5 Birth Defects Epidemiology Laboratory, Fundação Oswaldo Cruz, Av. Brasil 4365 - Pavilhão Leonidas Deane Sala 617, Manguinhos, Rio de Janeiro, RJ 21040-900, Brazil
sarcoma at any age [3]. Further studies revealed that, although highly specific for TP53 germline mutations, these criteria fail to include many mutation-positive fami- lies [1, 2]. Since then, less stringent criteria have been used for indicating TP53 testing, including the Li-Fraumeni like (LFL) criteria of Birch [4] and Eeles [5]. The most robust analysis of TP53 mutation carriers to this date was per- formed in France by Bougeard et al. [6], who developed the most recent version of the Chompret criteria. Table 1 shows the definitions of these criteria.
Phenotypes of families carrying TP53 mutations can be highly variable, partly due to life style and genetic back- ground of genes involved in the p53 pathway [1, 2]. Therefore, it is important to collect data from several populations to better understand the epidemiology of LFS/ LFL. Moreover, mutations in TP53 can lead to different consequences on gene function, and thereby further char- acterization of individual genotype-phenotype associations could enable a better patient stratification and cancer risk management [2].
The possibility of a major second locus involved in LFS is an additional issue in the etiology of the syndrome, since approximately 20 % of LFS and up to 80 % of LFL families do not exhibit TP53 mutations [1, 2]. The underlying genetic defects in these families remain to be uncovered. The most obvious candidate genes would be p53 partners in tumor suppressor pathways. Therefore, different studies investigated BAX [7], CDKN2A [8], TP63 [9], CHEK2 [9], BCL10 [10], and PTEN [11] in TP53-negative families, but no association was evi- denced. Although a few studies have occasionally linked other loci to some LFS kindreds [12, 13], TP53 remains the only gene conclusively associated to the syndrome [1]. CDKN1A/p21 (Cyclin-dependent kinase inhibitor 1A) is a
direct transcriptional target of p53 and the main regulator of p53 activity in cell cycle arrest [14]. A previous analysis of CDKN1A/p21 variants has shown that they may have a direct effect on tumor development, and tend to be less prevalent in tumors with TP53 mutations [15]. In addition, knockout mice for CDKN1A have shown to be susceptible to early cancer development, including the development of sarcomas, the main feature of LFS [14]. It has been reported that p53 mutants that downregulate the transcription of CDKN1A are highly frequent in sporadic tumors, suggesting these mutations might undergo posi- tive selection in human cancers [16]. Even so, to the best of our knowledge, germline mutations in CDKN1A in patients with LFS/LFL criteria have never been investigated.
In the present study, we characterized a cohort of Brazilian families with criteria for LFS/LFL. First, we searched for point mutations and gene rearrangements in the TP53 gene. Subsequently, we searched for germline mutations in the CDKN1A gene, as a possible second locus for TP53-negative LFS/LFL patients.
Materials and methods
Patients
Thirty-nine unrelated probands whose families fulfilled at least one of the diagnostic criteria for LFS/LFL were included in the study. The probands were referred for genetic counseling appointments both in the pediatric and adult genetic counseling outpatient clinics. Available histopathologic data and/or medical reports were used to confirm the reported data.
| Criteria | Definition |
|---|---|
| Classical [3] | Proband with a sarcoma diagnosed before age 45, and a first-degree relative with any cancer before age 45, and a first- or second-degree relative with any cancer before age 45 or a sarcoma at any age |
| Birch [4] | Proband with any childhood cancer or sarcoma, brain tumor or adrenocortical carcinoma diagnosed before age 45, and a first- or second-degree relative with a typical LFS malignancy (sarcoma, leukemia, breast, brain, adrenal cortex) regardless of age at diagnosis, and a first- or second-degree relative with any cancer diagnosed before age 60 |
| Eeles [5] | Two different tumors that are part of extended LFS in first- or second-degree relatives at any age (sarcoma, breast cancer, brain tumor, leukemia, adrenocortical tumor, melanoma, prostate cancer, pancreatic cancer) |
| Chompret (2015 version) [6] | Proband with a tumor of the LFS spectrum (premenopausal breast cancer, soft tissue sarcoma, osteosarcoma, brain tumor, adrenocortical carcinoma) before age 46 and at least one first- or second-degree relative with any LFS tumor (except BC if the proband has BC) before age 56 or with multiple tumors; OR proband with multiple tumors (except multiple BC), two of which belong to the LFS spectrum, the first before age 46; OR proband with adrenocortical carcinoma, choroid plexus tumor or rhabdomyosarcoma of embryonalanaplastic subtype, regardless of family history; OR breast cancer before age 31 years |
Age in years BC breast cancer
Sanger sequencing and multiplex ligation probe amplification (MLPA) analysis
Genomic DNA was extracted from peripheral blood sam- ples according to standard procedures. Amplification and direct sequencing of all coding and intronic flanking regions of TP53 were performed using primer sequences described at the IARC TP53 database (http://p53.iarc.fr/) [17]. The pathogenicity of the detected mutations was determined by consulting the IARC TP53 database. The mutations found were confirmed by a second, independent analysis. Mutation-negative patients were then tested for copy number variation (CNV) using a MLPA kit designed for gliomas (SALSA kit P105-D1, MRC-Holland®, Ams- terdam, Netherlands) that contains eight probes for theTP53 gene, in addition to probes for eight other glioma- related genes (CDKN2A, CDK4, EGFR, MDM2, MIR26A2, NFKBIA, PDGFRA, PTEN). The copy number variations found were confirmed by a second reaction, using a dif- ferent MLPA kit (SALSA kit P056-C1, MRC-Holland®, Amsterdam, Netherlands). Patients with no pathogenic mutations in TP53 were submitted to sequencing of CDKN1A coding regions, performed as described else- where [18].
Results
Three families (3/39; 8 %) exhibited the classic LFS cri- teria. Among the remaining 36 families (92 %), four met only the Chompret criteria (2015 version), two met only the Birch criteria, and 18 met only the Eeles criteria. Fif- teen families fulfilled more than one of the four criteria. These data are shown in Table 2 and Online Resource 1.
Sequencing of the coding and intronic flanking regions of TP53 in 39 probands identified six (6/39) carriers of pathogenic missense TP53 mutations in heterozygosis (Table 2). Among the remaining samples (n = 33), MLPA analysis was reliable for 31 samples, and revealed partial deletion of TP53 in heterozygosis in two (2/31) probands (Table 2; Online Resource 2).
The 31 probands negative forTP53 mutations are shown in Online Resource 1. None of them exhibited mutations in CDKN1A. The pedigrees of all TP53 mutation carriers are arranged in Fig. 1. Partial deletions of TP53 represented 25 % (2/8) of the TP53 mutations found in our cohort. The total prevalence of large deletions in TP53 in our cohort was 6.5 % (2/31 patients).
The mutational rate according to the phenotypic criteria was: LFS-classic: 67 % (2/3 families); LFL-Birch: 30 % (3/10 families); LFL-Chompret (2015): 35 % (6/17fami- lies); LFL-Eeles: 15 % (5/33 families). Sensitivity, speci- ficity, and predictive values of these criteria are shown in
Online Resource 3. The Chompret criteria (2015 version) exhibited the highest sensitivity (75 %). The classic crite- ria, on the other hand, were the most specific (97 %) but the less sensitive (25 %).
All eight probands with pathogenic TP53 mutations developed pediatric neoplasias: four developed osteosar- coma, two developed rhabdomyosarcoma, one had adreno- cortical carcinoma, and one developed multiple primary cancers. Considering the tumors developed by the probands with mutations and the tumors developed by their relatives, a total of 33 tumors were reported. The most common was osteosarcoma (7/33; 21.2 %), followed by colorectal cancer (5/33; 15.2 %) and soft tissue sarcomas (4/33; 12.1 %). Breast and prostate cancers represented each 9.1 % of these tumors (3/33), and the hematological cancers, lung cancer, and adrenocortical tumors represented each 6.1 % (2/33). One case (3 %) of each of the following tumors occurred: central nervous system, stomach, skin (melanoma), testicle, and neurofibroma. The comparison of tumors in families with a TP53 mutation versus families with no TP53 muta- tion revealed that the following cancers were more prevalent in TP53-positive families (Fisher’s exact test): osteosarco- mas (p = 0.00003), soft tissue sarcomas (p = 0.0191), and adrenocortical tumors (p = 0.0219). The other core LFS cancers (breast cancer and central nervous system tumors) did not reveal the same association (p > 0.05).
Two benign tumors were observed in TP53 mutation carriers: an adrenocortical adenoma (proband #36) and a neurofibroma (proband #12). Proband #36 developed an adrenal adenoma at age 1 and an osteosarcoma at age 11, and was found to be a carrier of the TP53 R248W muta- tion. This patient was included in the study because the initial information was that she developed an “adrenal tumor”. Subsequent review of the histopathologic data revealed that this proband developed an adenoma of the right adrenal cortex, treated with surgery alone, weighing 20 g and with low mitotic index. No foci of carcinoma were observed in the tumor. Subsequently, this proband developed an osteosarcoma at age 11. Proband #12 had developed a rhabdomyosarcoma at age 2, and at age 9 he was diagnosed with an intercostal non-plexiform neurofi- broma. His family history is compatible with the four clinical diagnostic criteria for LFS/LFL. This proband and his father, who developed osteosarcoma at age 29, were found to carry the TP53 Y220C mutation.
Proband #25 was diagnosed at age 12 with an adreno- cortical carcinoma, and was a carrier of the TP53R337H mutation. His mother was not a carrier of the mutation, and information regarding familial history of cancer of the pro- band’s father was not available. Two other probands diag- nosed with osteosarcoma were found to be carriers of pathogenic TP53 mutations (proband #50, mutation R175H; proband #35, mutation R337C). Proband #28 developed a
| Patient gender | Tumor(s)ª | TP53 mutation | Protein | Exon (s) | Famhxb | Clinical criteria | Eeles | ||
|---|---|---|---|---|---|---|---|---|---|
| LFS | Birch | Chompret | |||||||
| #49 M | NHL (3), OS (17), OS (25), MEL(27), LMS (33) | Deletion E2-4 | 2-4 | CRC (M-31) | 4 ☒ | ||||
| #50 M | OS (14) | c.524G > A | R175H | 5 | PC (M-65) | 4 ☒ | |||
| #28 F | RMS (3) | c.584T > A | I195 N | 6 | PC (M-58), BC (F-34) | ☒ | ☒ | 4 ☒ | |
| #12 M | RMS (2), NF (9) | c.659A > G | Y220C | 6 | OS (M-29*), BT (M-14), BC (F- 23) | ☒ | ☒ | 4 ☒ | 4 ☒ |
| #36 F | ACA (<1), OS (11) | c.742C > T | R248W | 7 | CRC (F-46, F- < 30, F-20, M-25) | 4 ☒ | |||
| #53 F | OS (14) | Deletion E8 | 8 | PC (M- >60), TC (M-25), GC (M-35), LK (F-22), LC (F-40) | 4 ☒ | ||||
| #35 M | OS (16) | c.1009C > T | R337C | 10 | BC (F-29), RMS (F-6), LC (M- N/A) | ☒ | ☒ | 4 ☒ | 4 ☒ |
| #25 M | ACC (12) | c.1010G > A | R337H | 10 | N/A | 4 ☒ | |||
Tumors in bold occurred in the same individual. Reference sequences: Genbank NP_000537.3 and NM_000546.5
M male, F female, N/A not available, ACA adrenocortical adenoma, ACC adrenocortical carcinoma, BC breast cancer, BT brain tumor, CRC colorectal cancer, GC gastric cancer, LC lung cancer, LK leukemia, LMS leiomyosarcoma, NF neurofibroma, NHL non-Hodgkin’s lymphoma, OS osteosarcoma, PC prostate cancer, RMS rhabdomyosarcoma, TC testicular cancer
*Relative tested and also a carrier of the mutation
a Age at diagnosis
b Gender-age at diagnosis
rhabdomyosarcoma at age 3, and was found to carry the TP53 I195N mutation, which has not been previously reported in the germline [17].
Two probands showed partial deletion of the TP53 gene. Proband #49 developed five primary neoplasias (non- Hodgkin lymphoma at age 3, osteosarcomas at ages 17 and 25, melanoma at age 27, and a leiomyosarcoma at age 33). MLPA probes for exons 2-4 of TP53 evidenced only one copy of this region in this patient’s DNA. Proband #53 developed an osteosarcoma at age 14, and was found to carry only one copy of exon 8 of TP53. Sequencing of this exon in this patient did not disclose polymorphisms that could result in non-hybridization of the probe. MLPA of both probands, and sequencing of exon 8 of proband #53 are shown in Online Resource 2.
Discussion
We have searched for mutations in TP53 and CDKN1Ain 39 probands with LFS/LFL phenotypes. Eight probands with pathogenic mutations in TP53 were observed. Two mutations were partial deletions of TP53, accounting for 25 % of the TP53 mutations in our cohort. This prevalence is much higher than the one reported in the IARC TP53 database (http://p53.iarc.fr/-version R18), where this type of mutation is shown to represent only 0.7 % of the
germline mutations reported in this gene. Our results emphasize the importance of performing copy number variation analysis in addition to direct sequencing to avoid the underestimation of germline TP53 mutation carriers.
An unusual feature in our cohort was the development of a non-plexiform neurofibroma in proband #12, carrier of the TP53 Y220C mutation. These benign tumors are hall- mark features of neurofibromatosis 1 (OMIM #16220), and are not associated to LFS/LFL. This patient does not exhibit clinical features and/or family history of neurofi- bromatosis 1. The neurofibroma developed by the proband might have been the result of a genomic instability in the context of an inherited TP53 pathogenic mutation.
Another benign tumor observed in our cohort was an adrenocortical adenoma developed by proband #36, carrier of the TP53 R248W mutation. This case is remarkably similar to a previously reported patient with a mosaic TP53 mutation (R282W) who also developed an adrenal ade- noma at age 1, and an osteosarcoma at age seven [19]. Interestingly, our proband was also a carrier of an arginin- to-tryptophan substitution (R248W), and both mutations are located in the DNA binding domain of p53. Somatic mosaicism for TP53 R248Q mutation was recently described in a 2 year-old child with three primary neo- plasias [20]. Importantly, in spite of being a mutation carrier, the phenotype presented by our proband does not fit into any LFS/LFL criteria. Our findings emphasize the idea
PC (65)
BC (21)
CRC (31)
BC (34)
PC (58)
BT (14)
OS (29)
NHL (3), OS (17),
7
OS (25), MEL (27), LMS (33)
7
OS (14)
7
RMS (3)
7
RMS (2) NF (9)
#49
#50
#28
#12
TP53 deletion E2-4
TP53 R175H
TP53 I195N
TP53 Y220C
CRC (46)
LC (40)
PC (>60)
BC(29)
LC
CRC (<30) CRC (20) CRC (25)
LK (22)
TC (25) GC (35)
OS (16)
RVS (6)
7
ACC (12)
7
ACA (<1) OS (11)
7
7
OS (14)
#36
#53
#35
#25
TP53 R248W
TP53 deletion E8
TP53 R337C
TP53 R337H
that large and/or functioning pediatric adrenocortical ade- nomas, when in association with LFS core cancers in the proband or in close relatives, should be considered for TP53 mutation screening.
Benign tumors are occasionally reported in TP53 germline mutation carriers, e.g. leiomyomatosis [21], meningioma [22], endometrial polyp [23], trichilemmoma [24], thymoma [6]. Available data on benign lesions in TP53 mutation carriers are scarce. It is not clear whether benign tumors are incidental findings or may be causally associated to TP53 germline mutations. If TP53 mutations are implicated in the development of benign tumors, it is plausible to think these might represent premalignant lesions, and therefore the follow-up of these tumors should demand more caution.
We did not find mutations in the main mediator of p53 in cell cycle arrest, CDKN1A, although knockout mice for this gene have shown to be susceptible to early cancer
neurofibroma, NHL non-Hodgkin’s lymphoma, OS osteosarcoma, PC prostate cancer, RMS rhabdomyosarcoma, TC testicular cancer. Ages at diagnosis (in years) are presented in parenthesis
development, including core cancers of LFS [14]. Other candidate genes in LFS/LFL phenotypes have shown lack of association with these phenotypes [7-11]. Likewise, abnormal TP53 promoter methylation was not observed in LFS/LFL families negative for mutations in this gene [25]. Given the large spectrum of LFS/LFL phenotypes, the underlying genetic defect in some families may be linked to specific patterns of tumor occurrence. For example, genes involved in chromatin remodeling function were associated with excess of brain tumors in Chompret-posi- tive TP53-negative families [13]. Likewise, a pathogenic mutation in POT1, which regulates telomere length, was associated to cancer predisposition in LFL families with cardiac angiosarcoma [26].
In conclusion, our findings reinforce the need to include CNV analysis in TP53 mutation screening protocols. Moreover, large and/or functioning adrenal adenomas, when in association with LFS core cancers in the proband
and/or close relatives, should be considered for TP53 mutation analysis.
Acknowledgments F. R. V. is a recipient of the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) Grant 486599/2012-4 and Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) Grant E26/110.535/2012. RCA is recipient of a Ministério da Saúde/Instituto Nacional de Câncer Grant.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent Informed consent was obtained from all individ- ual participants included in the study.
References
1. Malkin D (2011) Li-Fraumeni syndrome. Genes Cancer 2:475-484. doi: 10.1177/1947601911413466
2. McBride KA, Ballinger ML, Killick E et al (2014) Li-Fraumeni syndrome: cancer risk assessment and clinical management. Nat Rev Clin Oncol 11:260-271
3. Li FP, Fraumeni JF, Mulvihill JJ et al (1988) A cancer family syndrome in twenty-four kindreds. Cancer Res 48:5358-5362
4. Birch JM, Hartley AL, Tricker KJ et al (1994) Prevalence and diversity of constitutional mutations in the p53 gene among 21 Li-Fraumeni families. Cancer Res 54:1298-1304
5. Eeles RA (1995) Germline mutations in the TP53 gene. Cancer Surv 25:101-124
6. Bougeard G, Renaux-Petel M, Flaman J-M et al (2015) Revisiting Li-Fraumeni syndrome from TP53 mutation carriers. J Clin Oncol 33:2345-2352. doi:10.1200/JCO.2014.59.5728
7. Barlow JW, Mous M, Wiley JC et al (2004) Germ line BAX alterations are infrequent in Li-Fraumeni syndrome. Cancer Epidemiol Biomarkers Prev 13:1403-1406
8. Portwine C, Lees J, Verselis S et al (2000) Absence of germline p16(INK4a) alterations in p53 wild type Li-Fraumeni syndrome families. J Med Genet 37:E13
9. Bougeard G, Limacher JM, Martin C et al (2001) Detection of 11 germline inactivating TP53 mutations and absence of TP63 and HCHK2 mutations in 17 French families with Li-Fraumeni or Li-Fraumeni-like syndrome. J Med Genet 38:253-257
10. Stone JG, Eeles RA, Sodha N et al (1999) Analysis of Li-Frau- meni syndrome and Li-Fraumeni-like families for germline mutations in Bcl10. Cancer Lett 147:181-185
11. Brown LT, Sexsmith E, Malkin D (2000) Identification of a novel PTEN intronic deletion in Li-Fraumeni syndrome and its effect on RNA processing. Cancer Genet Cytogenet 123:65-68
12. Bachinski LL, Olufemi S-E, Zhou X et al (2005) Genetic map- ping of a third Li-Fraumeni syndrome predisposition locus to human chromosome 1q23. Cancer Res 65:427-431
13. Aury-Landas J, Bougeard G, Castel H et al (2013) Germline copy number variation of genes involved in chromatin remodelling in families suggestive of Li-Fraumeni syndrome with brain tumours. Eur J Hum Genet 21:1369-1376. doi:10.1038/ejhg. 2013.68
14. Martín-Caballero J, Flores JM, García-Palencia P, Serrano M (2001) Tumor susceptibility of p21(Waf1/Cip1)-deficient mice. Cancer Res 61:6234-6238
15. Mousses S, Ozçelik H, Lee PD et al (1995) Two variants of the CIP1/WAF1 gene occur together and are associated with human cancer. Hum Mol Genet 4:1089-1092
16. Petitjean A, Achatz M, Borresen-Dale A et al (2007) TP53 mutations in human cancers: functional selection and impact on cancer prognosis and outcomes. Oncogene 26:2157-2165. doi:10. 1038/sj.onc.1210302
17. Petitjean A, Mathe E, Kato S et al (2007) Impact of mutant p53 functional properties on TP53 mutation patterns and tumor phe- notype: lessons from recent developments in the IARC TP53 database. Hum Mutat 28:622-629. doi:10.1002/humu.20495
18. Gravina S, Lescai F, Hurteau G et al (2009) Identification of single nucleotide polymorphisms in the p21 (CDKN1A) gene and correlations with longevity in the Italian population. Aging (Al- bany NY) 1:470-480
19. Prochazkova K, Pavlikova K, Minarik M et al (2009) Somatic TP53 mutation mosaicism in a patient with Li-Fraumeni syn- drome. Am J Med Genet A 149A:206-211. doi:10.1002/ajmg.a. 32574
20. Behjati S, Maschietto M, Williams RD et al (2014) A pathogenic mosaic TP53 mutation in two germ layers detected by next generation sequencing. PLoS One 9:e96531. doi:10.1371/journal. pone.0096531
21. Kazarin O, Vlodavsky E, Guralnik L et al (2013) Association between esophageal leiomyomatosis and p53 mutation. Ann Thorac Surg 95:1429-1431. doi: 10.1016/j.athoracsur.2012.08. 081
22. Rieske P, Zakrzewska M, Biernat W et al (2005) Atypical molecular background of glioblastoma and meningioma devel- oped in a patient with Li-Fraumeni syndrome. J Neurooncol 71:27-30. doi: 10.1007/s11060-004-9181-3
23. Varley JM, Thorncroft M, McGown G et al (1996) A novel deletion within exon 6 of TP53 in a family with Li-Fraumeni-like syndrome, and LOH in a benign lesion from a mutation carrier. Cancer Genet Cytogenet 90:14-16
24. Eeles RA, Warren W, Knee G et al (1993) Constitutional muta- tion in exon 8 of the p53 gene in a patient with multiple primary tumours: molecular and immunohistochemical findings. Onco- gene 8:1269-1276
25. Finkova A, Vazna A, Hrachovina O et al (2009) The TP53 gene promoter is not methylated in families suggestive of Li-Fraumeni syndrome with no germline TP53 mutations. Cancer Genet Cytogenet 193:63-66. doi: 10.1016/j.cancergencyto.2009.04.014
26. Calvete O, Martinez P, Garcia-Pavia P et al (2015) A mutation in the POT1 gene is responsible for cardiac angiosarcoma in TP53- negative Li-Fraumeni-like families. Nat Commun 6:8383. doi:10.1038/ncomms9383