7q36.1

FANCU,POF17,SPGF50

Fanconi Anemia (FA)

Germ-line bi-allelic and heterozygous mutations of XRCC2 (FANCU) are associated with different outcomes. Bi-allelic inactivating mutation of XRCC2 results in the FA-U subtype, while heterozygous inheritance of pathogenic XRCC2 mutations might increase the lifetime risk of germ-line mutation carriers for developing breast cancer.

Using whole exome sequencing, bi-allelic mutations in RAD51C were found in a 2.5 year old boy born to healthy Saudi first cousins with a spectrum of severe congenital anomalies suggestive of FA. These included microcephaly, facial palsy, a lack of thumbs, short stature and an ectopic left kidney (Shamseldin et al., 2012). Importantly, bi-allelic loss-of-function mutations were detected in two other genes, MTBP and RGS3, in addition to XRCC2 (Shamseldin et al., 2012). The authors scrutinized these genes with bi-allelic mutations, since this inheritance pattern is typical of FA. XRCC2 emerged as a candidate FA gene, in part because another RAD51 paralog, RAD51C, had previously been identified as a FA gene (Vaz et al., 2010). However, as loss-of-function mutations were present in other genes, XRCC2 could not be reliably established as causative for the clinical phenotype in this initial study. A subsequent study demonstrated that XRCC2 (FANCU) is indeed the 20^th FA gene based on the correction, following expression of XRCC2, of cellular phenotypes that are characteristic of FA cells (Park et al., 2016). These phenotypes included complementation of cellular hypersensitivity to MMC, correction of MMC-induced chromosome aberrations (total aberrations including breaks, gaps, radial and ring chromosomes, and number of radial chromosomes specifically), and MMC-induced cell cycle accumulation in G2-M (Park et al., 2016). However, the patient had normal blood counts and had not displayed cancer through age 7 (Park et al., 2016; Shamseldin et al., 2012).
To date, 22 FA or FA-like genes have been identified (Nepal et al., 2017). FA genes are autosomal recessive tumor suppressor genes, except for the X-linked FANCB and the autosomal dominant RAD51 (FANCR). The monoubiquitination of the FANCD2/ FANCI protein dimer is the central step in the FA pathway. FANC -A, -B, -C -E, -F, CC: TXT: -G ID: 295>, -L, -M, UBE2T (FANC-T) are early (or upstream) FA genes, since loss of function mutations in any of these genes results in defective monoubiquitination of FANCD2/I (Mamrak et al., 2017; Nepal et al., 2017). Late/downstream FA genes, which are associated with normal monoubiquitination of FANCD2 and FANCI when mutated include: BRCA2 (FANCD1), BRIP1 (FANCJ), PALB2 (FANCN), RAD51C (FANCO), RAD51 (FANCR), BRCA1 (FANCS), XRCC2 (FANCU), MAD2L2 (FANCV/polTheta) and RFWD3 (FANCW). Typical of late/downstream FA genes, FA-U cells display defective assembly of RAD51 foci in response to DNA damage (Park et al., 2016).
The vast majority of FA patients have bi-allelic mutations in the upstream FA genes, notably FANCA, FANCC and FANCG, and display clinical features characteristic of FA. These include progressive bone marrow failure around 7.6 years of age, a variety of congenital anomalies, and a predisposition to acute myeloid leukemia as well as various solid tumors that occur in the second and third decade of life (Kutler et al., 2003). Microcephaly, short stature, skin pigmentation defects, hypogonadism, and radial ray anomalies, many of which were observed in the single FA-U patient identified until present, are among the congenital anomalies that are often observed. Endocrine abnormalities are also seen in a significant number of FA patients (Rose et al., 2012).
Certain other FA complementation groups such as FA-O, defined by RAD51C (FANCO) mutations, and FA-R, defined by heterozygous dominant-negative RAD51 (FANCR) mutation, have been designated atypical FA (or FA-like) because of the absence of bone marrow failure and no increased incidences of cancer (Park et al., 2016). However, bone marrow failure and cancer are not always observed for every patient in each FA complementation group and the single FA-U patient identified so far has not reached the age at which these features are frequently observed (Kutler et al., 2003). Therefore it is not yet clear whether or not patients with germ-line bi-allelic defects in XRCC2 represent atypical FA.

Breast Cancer

In a collaborative exome-sequencing study involving patients from the Netherlands, Spain and Australia, Park et al. identified 6 out of 1308 patients with early-onset breast cancer as having likely deleterious mutations in XRCC2, while none of the 1,120 controls carried such a mutation. Two of the six mutations were frameshift mutations (p.Arg17^* and p.Cys217^*) and 4 of the 6 were missense alterations with single a.a. exchanges that were predicted by three algorithms to be associated with a loss-of-function (Park et al., 2012). As the frequency at which these variants occurred in breast cancer patients, as compared to controls, was statistically significant, XRCC2 was identified by the authors as a low frequency breast cancer susceptibility gene.
However, no other study so far has confirmed mutation of XRCC2 as a significant cause of inherited breast cancer (Hilbers et al., 2016; Hilbers et al., 2012; Pelttari et al., 2015). For example, Hilbers et al. analysed the coding region of 3548 non-BRCA1/2 familial breast cancer cases and 1435 healthy controls (Hilbers et al., 2012). In the patient group, they detected only one patient with a protein truncation mutation and 20 patients with missense alterations, as compared to nine controls with missense variants. Importantly, in 2016, Hilbers et al. functionally analyzed all nonsynonymous coding variants from the two major studies published in 2012 (Hilbers et al., 2012; Park et al., 2012). Out of 24 missense alterations, only four variants had a reduced functional activity with 50-75% rescue in two out of three assays (Hilbers et al., 2016).
Finally, Decker et al. sequenced the XRCC2 coding region of 13,087 breast cancer patients and 5488 healthy controls from the UK (Decker et al., 2017). Only 13 carriers of truncating XRCC2 variants were found, nine in breast cancer patients and four in controls. There were also no statistical differences for 32 rare missense variants between the patient and control groups. Thus these authors concluded that truncating/loss-of-function mutations in XRCC2 are not associated with an increased risk of breast cancer (Decker et al., 2017), and therefore germ-line mutation of XRCC2 is not a major cause of inherited breast cancer (Decker et al., 2017; Hilbers et al., 2012).

Because XRCC2 has an important role in HR, it is likely that tumors with loss-of-function of XRCC2 may be particularly responsive to poly (ADP-ribose) polymerase (PARP) inhibitors. In support of this possibility, FA cells from the FANCU complementation group, which have biallelic mutation of XRCC2, have increased sensitivity to the PARP inhibitor olaparib, as compared to complemented cells (Park et al., 2016). Thus, identification of breast cancer patients, and perhaps other cancer patients, with deleterious somatic variants of XRCC2 may be a basis for such personalized treatments.