Li–Fraumeni syndrome (LFS) is a rare autosomal dominant disorder characterized by sarcoma, breast cancer, brain Tumours, leukemia/lymphoma, and adrenocortical carcinoma (ACC) (Li and Fraumeni 1969; Li et al. 1988). Germline mutations in the TP53 Tumours suppressor gene on 17p13.1 have been found in approximately 70 % of LFS (Malkin et al. 1990; Srivastava et al. 1990; Varley et al. 1997). Deleterious TP53 mutations leading to LFS occur in between 1 in 5,000 and 1 in 20,000 births.
The major component malignancies in LFS include sarcomas, breast cancer, brain Tumours, ACC, and acute leukemia (Li and Fraumeni 1969; Garber et al. 1991; Li et al. 1991). Other associated cancers may include Wilms Tumours; cancers of the colon, stomach, lung, and pancreas; as well as melanoma and gonadal germ cell Tumours (Garber et al. 1991; Varley et al. 1997; Birch et al. 2001), although some of these are isolated observations in a single family, and so, their exact frequencies in mutation-positive individuals are unknown. The operational diagnostic criteria for LFS, the so-called classic criteria, are as follows (Li and Fraumeni 1969):
* An individual (index case) with a sarcoma diagnosed before age of 45 years
* A first-degree relative with any cancer before age of 45 years
* A third family member who is a first- or second-degree relative with either a sarcoma diagnosed at any age or any cancer diagnosed before age of 45 years.
There are several sets of operational criteria for the diagnosis of LFS-like (LFL) families, such as the Eeles criteria (Eeles 1995) and the Manchester criteria (Varley 2003). The most recent set of criteria are the Chompret criteria, which have been recently modified (Tinat et al. 2009):
*Proband with Tumours belonging to LFS Tumours spectrum (e.g., soft tissue sarcoma, osteosarcoma, brain Tumours, premenopausal breast cancer, adrenocortical carcinoma, leukemia, lung bronchoalveolar cancer) before age 46 years and at least one first- or second-degree relative with LFS Tumours (except breast cancer if proband has breast cancer) before age 56 years or with multiple Tumours.
*Proband with multiple Tumours (except multiple breast Tumours), two of which belong to LFS Tumours spectrum and first of which occurred before age 46 years
*Patient with adrenocortical carcinoma or choroid plexus Tumours, irrespective of family history. At least one of these three separate parts need to be fulfilled for the Chompret criteria to be fulfilled.
Germline TP53 mutations cause LFS. In contrast to other cancer syndromes, missense mutations are the most common variety in LFS (Varley 2003; Ribeiro et al. 2001). Earlier studies had found that the mutation frequency in families meeting the standard clinical criteria for LFS is 70–80 % in most clinical laboratories (Varley et al. 1997; Friedl et al. 1999; Frebourg et al. 1995; Varley 2003). In families meeting the LFL criteria, the frequency is lower – 8 % using the Eeles criteria and 22–40 % using the Manchester criteria (Eng et al. 1997; Varley et al. 1997; Varley 2003; Birch et al. 1990, 1994). In a recent study, the classic criteria had high specificity (91 %) but low sensitivity (40 %), whereas the relaxed LFL criteria of Eeles were very sensitive (97 %) but not specific (16 %). The Chompret criteria offer the best compromise – 95 % sensitivity and 52 % specificity (Gonzalez et al. 2009).
The French LFS working group studied 474 French families suggestive of LFS, of which 232 fulfilled the Chompret criteria. Overall, they identified a germline mutation in 82 families (17 %). The percentage of positive results was much higher in those that met Chompret criteria (29 %) than in those that did not (6 %) (Bougeard et al. 2008). A more recent study added important details – among 525 patients sent to one US laboratory for testing, 17 % were positive for a germline TP53 mutation. Notably, all positive cases had at least one family member with a sarcoma or breast, brain or adrenocortical carcinoma (ACC). All eight persons tested who had a choroid plexus Tumours were positive for TP53 mutations (Gonzalez et al. 2009). In contrast, others have found that about 50 % of children with choroid plexus carcinomas carry germline TP53 mutations, and positive cases meet the classic criteria and had a very poor outcome. No cases that failed to meet the criteria carried germline mutations, and most had better outcomes. None of six children with a choroid plexus papilloma had a mutation (Tabori et al. 2010). As for ACC, 14 of 21 with a childhood ACC had a germline mutation, regardless of family history of cancer (Gonzalez et al. 2009).
Based on a segregation analysis of relatives of children with childhood sarcomas, the risk of cancer in LFS was estimated to be 50 % by age 30 years and 90 % by age 60 years. Molecular data have broadly confirmed these findings – the penetrance of TP53 mutations for cancer is greater than 90 % for women (~75 % for men, the difference is mainly accounted for by high incidence of breast cancer diagnoses in women with LFS) (Chompret et al. 2000). Before age 10 years, the most commonly seen cancers are soft tissue sarcomas, brain Tumours, and ACC. For those 11–20 years, osteosarcomas predominate, and for those above 20 years of age, breast cancer and brain Tumours are the most common manifestations, and all other epithelial Tumours listed above tend to occur in those over age 20 years.
Genotype-phenotype correlations have been observed in some studies but not in others. For example, in one, families with missense mutations in the DNA-binding domain trended towards an overall higher cancer incidence, particularly of those of the breast and CNS, with an earlier age at onset, when compared to families with protein truncating or inactivating mutations or families with no mutation at all (Birch et al. 1998). Further, a systematic database study of all LFS families revealed that the mean age of breast cancer diagnosis in TP53 mutation-positive families was 34.6 years in contrast to 42.5 years in mutation-negative families (P = 0.0035) (Olivier et al. 2003). In mutation-positive families, brain Tumours were overrepresented in those families with missense mutations in the DNA-binding loop that contacts the minor groove of DNA. Development of ACCs was associated with missense mutations in the loops opposing the protein–DNA contact surface (P = 0.0003) (Olivier et al. 2003). In contrast, a study of 56 TP53 mutation- positive individuals from 107 kindreds ascertained through cases of childhood soft tissue sarcoma reported no difference in phenotype between patients with missense mutations compared to those with truncating mutations (Hwang et al. 2003). Differences in patient accrual, study design, and mutation site classification may have contributed to these disparate findings, and thus more studies or a pooled analysis are needed to clarify this issue.
The frequency of germline TP53 mutations in patients with multiple primary cancers unselected for family history and in those with apparently sporadic LFS component Tumours has been extensively studied. While early studies suggested that only 1 % of early-onset breast cancer cases will harbor a germline TP53 mutation (Sidransky et al. 1992; Lalloo et al. 2003), the frequencies may be higher in cases of sporadic osteosarcomas, 2–3 % (McIntyre et al. 1994), 9 % for rhabdomyosarcomas (Diller et al. 1995), and 2–10 % for brain Tumours (Felix et al. 1995). Perhaps the most striking association occurs in cases of childhood ACC. In a series of 14 ACC patients unselected for family history, 11 (82 %) carried a germline TP53 mutation (Gonzalez et al. 2009), and this is one reason why the Chompret criteria allow for ACC in the absence of family history (although it is fair to say that germline TP53 mutations are very rare in those diagnosed over 20 years old). Interestingly of 36 cases of childhood ACC in southern Brazil, 35 carried an identical R337H (p. Arg337His) mutation. There is a single common origin for all carriers of the Brazilian R337H mutation (Pinto et al. 2004). The cancer family history in mutation-positive cases was not striking (there was no evidence for LFS in the 30 mutation-positive kindred), suggesting a low- penetrance and possibly tissue-specific effect of this particular mutation (Figueiredo et al. 2006). More recent data, however, have shown that other cancers can occur in R337H carriers (Gomes et al. 2012; Seidinger et al. 2011). Another recent development is the finding that most LFS-related breast cancers are ER-positive, HER2-positive Tumours (Wilson et al. 2010; Masciari et al. 2012). This may help in determining when to offer TP53 testing to young women with breast cancer who do not meet classic or Chompret criteria, when generally the yield is low (Tinat et al. 2009).
Genetic and Medical Management
The management of LFS is not straightforward. Many are reluctant to offer predictive testing for minors in families with a known pathogenic TP53 mutation in view of the difficulties in screening for such a wide spectrum of cancers in childhood without proven benefit from screening. However, it may be useful to view TP53 mutation analysis as a molecular diagnostic test for LFS, especially in the setting of LFL. While breast cancer screening by MRI can be instituted, little effective surveillance is widely available for many of the other component Tumours. On the other hand, most clinical cancer geneticists will acknowledge that a predictive test that is negative in the setting of a known family-specific germline TP53 mutation can be useful.
Such an informative negative test will relieve that particular family member from LFS-directed clinical surveillance as his/her cancer risk would be no different from that of the general population. Recently, a prospective study applying an intensive screening protocol originating at the Hospital for Sick Children, Toronto (see Table. 1), has provided some hope that aggressive screening can be of benefit, particularly in detecting small, operable brain Tumours and in finding intra-abdominal Tumours at a readable stage (Villani et al. 2011). Further large-scale studies with longer follow-up are required before this, or similar screening protocols can be accepted for wide-scale use (Table. 1).
Table 1 Surveillance strategy for individuals with germline TP53 mutations
|Ultrasound of abdomen and pelvis every 3–4 months|
|Complete urinalysis every 3–4 months|
|Blood tests every 4 months: β-human chorionic gonadotropin, alpha-fetoprotein, 17-OH-progesterone, testosterone, dehydroepiandrosterone sulfate, androstenedione|
|Annual brain MRI|
|Soft tissue and bone sarcoma|
|Annual rapid total body MRI|
|Leukaemia or lymphoma|
|Blood test every 4 months: complete blood count erythrocyte sedimentation rate, lactate dehydrogenase|
|Monthly breast self-examination starting at age 18 years|
|Clinical breast examination twice a year, starting at age 20–25 years, or 5–10 years before the earliest known breast cancer in the family|
|Annual mammography and breast MRI screening starting at age 20–25 years, or at earliest age of onset in the family|
|Consider risk-reducing bilateral mastectomy|
|Annual brain MRI|
|Soft tissue and bone sarcoma|
|Annual rapid total body MRI|
|Ultrasound of abdomen and pelvis every 6 months|
|Colonoscopy every 2 years, beginning at age 40 years, or 10 years before earliest known colon cancer in the family|
|Annual dermatological examination|
|Leukaemia or lymphoma|
|Complete blood count every 4 months|
|Erythrocyte sedimentation rate, lactate dehydrogenase every 4 months|
Adapted with permission from Villani et al. (2011)