RE : frameshift variant FANCL *c.1096_1099dupATTA is not associated with high breast cancer risk

To the Editor, In our recent publication, we noted a borderline association of the truncating variant c.1096_1099dupATTA in the FANCL gene with increased breast cancer (BC) risk in high-risk BRCA1/BRCA2/PALB2-negative BC patients (1). However, the subsequent analysis by Pfeifer et al. (2) genotyping the c.1096_1099dupATTA variant in 2370 samples from German and Macedonian BC patients and controls failed to confirm association of this variant with BC risk. We agree with Pfeifer et al. that FANCL*c.1096_ 1099dupATTA is unlikely a high-risk BC susceptibility allele with an immediate clinical utility. There are several lines of evidence that do not support strong involvement of this variant in BC susceptibility including: (i) the relative high frequency of this variant especially in European populations, (ii) functional characteristics demonstrating that cells expressing FANCL isoform coded by c.1096_1099dupATTA variant retain the residual FANCL functional capacity in vitro (3), and (iii) phenotypic characteristics that show only a mild Fanconi anemia (FA) complementation group L phenotype in the compound heterozygote carrying FANCL*c.1096_1099dupATTA (alongside an another truncating FANCL variant) (3). Moreover, we identified a male c.1096_1099dupATTA homozygote (in controls) who had no signs of FA at his age of 57 years (Table 1). We also agree that FANCL*c.1096_1099dupATTA may confer a low (or lower) risk variant. We hypothesized that this variant may represent a modifying factor because its carriers were overrepresented only in a subgroup of high-risk BC patients in our study and also four out of six c.1096_1099dupATTA carriers analyzed by a panel next-gene sequencing (NGS) carried truncating variant(s) in other known or candidate cancer-susceptibility gene(s). After publication of our study reporting 15 carriers of c.1096_1099dupATTA in 2126 analyzed samples of Czech BC patients and controls, we identified another eight carriers using the CZECANCA multicancer panel NGS (4). The individual characteristics of c.1096_1099dupATTA carriers (Table 1) indicate a relatively low mean age at diagnose [47.2 years (range 28–76 years)] in 14 carriers with BC. We also recently identified three c.1096_1099dupATTA carries with ovarian cancer diagnosed at early age. Contrary to Pfeifer et al. who reported that only one out of 10 identified c.1096_1099dupATTA carriers had a family BC history, we have noticed a known familial history of BC (in a first or second degree relative) in nine out of 23 carriers (39%) and a familial history of some cancer in 15 carriers (65%). The c.1096_1099dupATTA variant was accompanied by another truncating variant(s) in nine out of 14 cancer patients analyzed by a panel NGS (Table 1). We suppose that these characteristics indicate that c.1096_1099dupATTA may (perhaps mildly) modify the breast (or other) cancer risk or cancer onset. However, further studies are required to estimate the risk of cancer development in c.1096_1099dupATTA carriers precisely. The segregation analyses and NGS analyses in families of carriers would be also required to evaluate the involvement of this hypomorphic variant in the risk of other cancer development or in modification of cancer onset.