ClinVar Genomic variation as it relates to human health

It is observed in the gnomAD v2.1.1 dataset at total allele frequency of 0.583%. Same nucleotide change resulting in same amino acid change has been previously reported as pathogenic/likely pathogenic/VUS (ClinVar ID: VCV000002547). Different missense changes at the same codon (p.Lys695Asn, p.Lys695Met) have been reported to be associated with MEFV-related disorder (ClinVar ID: VCV000097490 / PMID: 16730661). The evidence of pathogenicity is insufficient at this time. Therefore, this variant is classified as VUS according to the recommendation of ACMG/AMP guideline.

The MEFV c.2084A>G; p.Lys695Arg variant (rs104895094) is reported in the medical literature in individuals with familial Mediterranean fever (FMF) as well as in individuals with other autoinflammatory diseases and has been implicated as a reduced penetrance allele (Altug 2013, Bernot 1998, Comak 2013, Feng 2009, Gershoni-Baruch 2002). The variant is listed in the ClinVar database (Variation ID: 2547). This variant is found in the general population with an overall allele frequency of 0.5% (1648/282878 alleles, including 11 homozygotes) in the Genome Aggregation Database. The lysine at codon 695 is moderately conserved, and computational analyses are uncertain whether this variant is neutral or deleterious (REVEL: 0.353). Considering available information, this variant is classified as pathogenic, but may exhibit reduced penetrance. References: Altug U et al. MEFV gene mutations in Henoch-Schonlein purpura. Int J Rheum Dis. 2013 Jun;16(3):347-51. PMID: 23981758 Bernot et al. Non-founder mutations in the MEFV gene establish this gene as the cause of familial Mediterranean fever (FMF). Hum Mol Genet. 1998 7(8):1317-25. PMID: 9668175 Comak et al. MEFV gene mutations in Turkish children with juvenile idiopathic arthritis. Eur J Pediatr. 2013 172(8):1061-7. PMID: 23588594 Feng J et al. Missense mutations in the MEFV gene are associated with fibromyalgia syndrome and correlate with elevated IL-1beta plasma levels. PLoS One. 2009 Dec 30;4(12):e8480. PMID: 20041150 Gershoni-Baruch et al. Familial Mediterranean fever: the segregation of four different mutations in 13 individuals from one inbred family: genotype-phenotype correlation and intrafamilial variability. Am J Med Genet. 2002 109(3):198-201. PMID: 11977178

Disclaimer: This variant has not undergone full assessment. The following are preliminary notes: Personal communication by Dr. Serge Amselem (who has been working on FMF for many years now): ...the issue of the pathogenicity of MEFV variants can be summarized as follows: no functional assay, conservation cannot be used as an argument because the domain that contains the mutations that are accepted by the scientific community as such is not conserved in the mouse; in addition, in other species where the domain is conserved the mutation corresponds to the reference sequence in these species. So, the only argument can be the frequency in the patient population versus cohort populations of the same origin. For K695R there is no proof that the variant is indeed deleterious. There are more than 30 papers reporting this variant since the original paper by Bernot-1998. It is considered as a variant with reduced penetrance. In the paper by Gershoni-2002 it is worth noting the high number of healthy individuals that carry the variant in the family (and none of the affecteds). In the paper by Aksientevitch-1999 the K695 is found more often in healthy controls than in patients [....The K695R mutation was also overrepresented in the general Ashkenazi sample, accounting for 12% of the carrier chromosomes, but was only observed on 2 (5%) of 36 Ashkenazi FMF patient chromosomes.. ] According to Pr. Amselem it would most likely be a polymorphism but we would need a serious population study to demonstrate it and in clinic the answer they have been giving is VUS.

Variant summary: MEFV c.2084A>G (p.Lys695Arg) results in a conservative amino acid change located in the B30.2/SPRY domain (IPR001870) of the encoded protein sequence. Four of five in-silico tools predict a benign effect of the variant on protein function. The variant allele was found at a frequency of 0.004 in 1614186 control chromosomes in the gnomAD database (v4), including 54 homozygotes. This frequency is not significantly higher than estimated for a pathogenic variant in MEFV causing Familial Mediterranean Fever (0.004 vs 0.022). The variant c.2084A>G has been reported in the literature in several compound heterozygous and homozygous patients, as well as in non-informative genotypes among individuals affected with FMF (e.g. Bernot_1998, Giaglis_2007, Berdeli_2011, Lainka_2012, Papa_2017, Balta_2020, Richard_2023), but has also been reported as a compound heterozygous genotype in asymptomatic individuals (e.g. Bernot_1998, Gershoni-Baruch_2002). Several studies suggested the variant may have a mild effect or be of reduced penetrance (e.g. Bernot_1998, Aksentijevich_1999, Gershoni-Baruch_2002). At least one of these studies reported a lack of co-segregation with disease in a family harboring two different MEFV variants in compound heterozygosity, namely p.M694V and p.M681I, as the segregating cause of Mediterranean Fever disease presentation (Gershoni-Baruch_2002). A recent large study analyzing the clinical findings of 27,504 FMF patients from Turkey and Northern Cyprus reported that K695R was found in 2.1% of this cohort, and 18% of the patients with K695R showed response to colchicine, although ~60% of the K695R carrying patients had no fever, and 66.4% had no joint pain and/or inflammation (Dundar_2022). On the other hand, an ex vivo functional study using patient derived cells with this variant demonstrated that the cytokine release in response to Clostridium difficile toxin A (TcdA) in the presence versus absence of colchicine fully coincided with controls (Van Gorp_2020). Further experimental evidence evaluating an impact on protein function through cell-based assays, detected no substantial increase in spontaneous cell death and TcdA/UCN-01-induced cell death for the variant compared to wild-type (Honda_2021); however, no unequivocal conclusions about correlation of these findings to actual disease manifestation in humans can be drawn. In 2023, the expert international study group for systemic autoinflammatory diseases (INSAID) reported an updated classification of uncertain significance for the variant (Infevers database; Van Gijn_2018). Since the possibility that environmental and genetic factors might contribute to the pathogenicity of this variant cannot be excluded (Gershoni-Baruch_2002), it is unknown whether this variant represents a low-penetrance, mild, common, pathogenic variant, a modifier, or a risk allele. The following publications have been ascertained in the context of this evaluation (PMID: 20534143, 10090880, 9668175, 10612841, 17489852, 10842288, 15024744, 19253030, 21413889, 22614345, 15018633, 22903357, 19934083, 23907647, 10787450, 25615955, 11977178, 29047407, 29543225, 23505242, 29599418, 26078663, 21978701, 31989427, 32312770, 33733382, 34426522, 35098403,37277124,37072232). ClinVar contains an entry for this variant (Variation ID: 2547). Based on the evidence outlined above, the variant was classified as uncertain significance.

MEFV NM_000243.2 exon10 p.Lys695Arg (c.2084A>G): This variant is a well reported but controversial variant in the literature, with several individuals described with features of Familial Mediterranean Fever (FMF) in the homozygous, heterozygous, compound and double heterozygous state, including an entry in GeneReviews (Aksentijevich 1999 PMID:10090880, Feng 2009 PMID:20041150, Attug, 2013 PMID:23981758, Oztuzcu 2014 PMID:24469716, Sediva 2014 PMID:24251727, Shohat 2016 PMID:20301405). This variant is present in 1.6% (422/25790) of European (Finnish) alleles, including 3 homozygotes in the Genome Aggregation Database (http://gnomad.broadinstitute.org/rs104895094). This variant is present in ClinVar, with several discrepant classifications from Pathogenic to Likely Benign (Variation ID:2547). Evolutionary conservation and computational predictive tools suggest that this variant may not impact the protein. Of note, at least 1 article in the literature suggests that this variant may have a mild effect or reduced penetrance (Aksentijevich 1999 PMID:10090880), but further information is required for accurate classification. In summary, data on this variant is too unclear for definitive disease classification; therefore, the clinical significance of this variant is uncertain.

MEFV NM_000243.2 exon10 p.Lys695Arg (c.2084A>G): This variant is a well reported but controversial variant in the literature, with several individuals described with features of Familial Mediterranean Fever (FMF) in the homozygous, heterozygous, compound and double heterozygous state, including an entry in GeneReviews (Aksentijevich 1999 PMID:10090880, Feng 2009 PMID:20041150, Attug, 2013 PMID:23981758, Oztuzcu 2014 PMID:24469716, Sediva 2014 PMID:24251727, Shohat 2016 PMID:20301405). This variant is present in 1.6% (422/25790) of European (Finnish) alleles, including 3 homozygotes in the Genome Aggregation Database (http://gnomad.broadinstitute.org/rs104895094). This variant is present in ClinVar, with several discrepant classifications from Pathogenic to Likely Benign (Variation ID:2547). Evolutionary conservation and computational predictive tools suggest that this variant may not impact the protein. Of note, at least 1 article in the literature suggests that this variant may have a mild effect or reduced penetrance (Aksentijevich 1999 PMID:10090880), but further information is required for accurate classification. In summary, data on this variant is too unclear for definitive disease classification; therefore, the clinical significance of this variant is uncertain.

The MEFV c.2084A>G (p.Lys695Arg) variant has been identified in multiple individuals with a phenotype consistent with familial Mediterranean fever in the homozygous state, compound heterozygous state, and heterozygous state in the peer-reviewed literature (PMID: 9668175; 10612841; 17489852; 19253030; 19934083; 22207183; 23907647; 24469716; 26003477; 27364639; 27733942; 29543225). This variant shows variable expressivity and incomplete penetrance (PMID: 9668175; 10090880). Additionally, other nearby missense variants such as p.Met694Ile (ClinVar variation ID: 36507), p.Met694Val (ClinVar variation ID: 2538), and p.Met680Ile (ClinVar variation ID: 2539), are classified pathogenic for familial Mediterranean fever. The highest frequency of this allele in the Genome Aggregation Database is 0.01612 in the European (Finnish) population, which includes 3 homozygotes (version 2.1.1). This frequency is high but is consistent with disease prevalence estimates, variable severity, and incomplete penetrance. Based on the available evidence, the c.2084A>G (p.Lys695Arg) variant is classified as pathogenic for familial Mediterranean fever.

BS1

The p.K695R pathogenic mutation (also known as c.2084A>G), located in coding exon 10 of the MEFV gene, results from an A to G substitution at nucleotide position 2084. The lysine at codon 695 is replaced by arginine, an amino acid with highly similar properties. This variant was detected in the homozygous state in an individual with familial Mediterranean fever (FMF) phenotype type I (Comak E et al. Eur. J. Pediatr., 2013 Aug;172:1061-7). This mutation has been described in individuals with FMF from multiple different ethnic backgrounds (Touitou I. Eur. J. Hum. Genet., 2001 Jul;9:473-83). This alteration has a higher carrier frequency than expected in the Ashkenazi Jewish population, which is suggestive of reduced penetrance. In addition, this alteration was seen in three Jewish individuals, two of which were asymptomatic (Bernot A et al. Hum. Mol. Genet., 1998 Aug;7:1317-25). This alteration was also detected in one child with Henoch–Schönlein purpura and in one individual with very mild FMF symptoms (Altug U et al. Int J Rheum Dis, 2013 Jun;16:347-51; Sedivá A et al. Clin. Genet., 2014 Dec;86:564-9). In addition, this alteration is predicted to be tolerated by BayesDel in silico analysis. Since supporting evidence is conflicting at this time, the clinical significance of this alteration remains unclear.

MEFV: PM1, BP4, BS1:Supporting

Reported heterozygous without a second MEFV variant in individuals with FMF as well as in normal controls (PMID: 10090880, 29543225, 12461684, 20041150, 26003477); Observed on 0.14% of MEFV alleles of individuals with FMF in the Armenian population (PMID: 29543225); In silico analysis indicates that this missense variant does not alter protein structure/function; Has been considered a variant with reduced penetrance (PMID: 10090880, 11977178); Listed in ClinVar with conflicting classifications; This variant is associated with the following publications: (PMID: 16120953, 23907647, 27125729, 26690517, 26892483, 29707173, 30826945, 30513227, 22975760, 23588594, 24251727, 22953644, 19934083, 11977178, 16255051, 26399837, 26003477, 20041150, 25821352, 27364639, 25648235, 27022006, 26842301, 26400644, 24469716, 16730661, 28108907, 27582173, 27353043, 27051312, 23981758, 9668175, 21153919, 28927886, 18318646, 29977033, 26984802, 14615741, 11175300, 30783801, 28828621, 32199921, 32597225, 34426522, 32853466, 29080837, 33258288, 30476936, 35874679, 35110061, 35480407, 34819953, 10090880, 12461684, 33733382, 29543225, 37678716, 19302049)

The MEFV c.2084A>G variant is predicted to result in the amino acid substitution p.Lys695Arg. The c.2084A>G variant has been reported in both control populations and in patients with Familial Mediterranean Fever (FMF) in the presence of a second pathogenic allele (Tunca et al. 2002. PubMed ID: 12461684; Caglayan et al. 2010. PubMed ID: 19934083; Bernot et al. 1998. PubMed ID: 9668175; Aksentijevich et al. 1999. PubMed ID: 10090880). This variant is reported in 1.6% of alleles in individuals of European (Finnish) descent in gnomAD V2 (as displayed in the table above) and in 7 homozygotes globally. The c.2084A>G variant has conflicting interpretations from different laboratories ranging from likely benign to pathogenic. While the c.2084A>G change may be a low penetrant pathogenic variant, at this time the clinical significance of the c.2084A>G variant is classified as uncertain due to the absence of conclusive functional and genetic information.

The MEFV p.Lys695Arg variant was identified in multiple individuals with familial Mediterranean fever (FMF) or with suspected FMF (Cekin_2017_PMID:28483595; Bernot_1998_PMID:9668175; Debeljak_2015_PMID:26399837; Caglayan_2009_PMID:19934083). However, this variant was also identified in several phenotypically healthy heterozygous individuals (Bernot_1998_PMID:9668175; Milenkovic_2016_PMID:27364639; Debeljak_2015_PMID:26399837; Salehzadeh_2013_PMID:25793047). The variant was identified in dbSNP (ID: rs104895094) and ClinVar (classified as pathogenic by eight submitters including Ambry Genetics, GeneDx, Illumina, and ARUP Laboratories; as likely pathogenic by Counsyl and EGL Genetics; as likely benign by Invitae; and as uncertain significance by Laboratory for Molecular Medicine and four other submitters). The variant was identified in control databases in 1648 of 282878 chromosomes (11 homozygous) at a frequency of 0.005826 increasing the likelihood this could be a low frequency benign variant (Genome Aggregation Database March 6, 2019, v2.1.1). The variant was observed in the following populations: European (Finnish) in 405 of 25120 chromosomes (freq: 0.01612), Ashkenazi Jewish in 95 of 10370 chromosomes (freq: 0.009161), European (non-Finnish) in 1026 of 129188 chromosomes (freq: 0.007942), Other in 57 of 7224 chromosomes (freq: 0.00789), Latino in 55 of 35440 chromosomes (freq: 0.001552), African in 7 of 24966 chromosomes (freq: 0.00028) and South Asian in 3 of 30616 chromosomes (freq: 0.000098), but was not observed in the East Asian population. The p.Lys695 residue is not conserved in mammals and four out of five computational analyses (PolyPhen-2, SIFT, AlignGVGD, BLOSUM, MutationTaster) do not suggest a high likelihood of impact to the protein; however, this information is not predictive enough to rule out pathogenicity. The variant occurs outside of the splicing consensus sequence and in silico or computational prediction software programs (SpliceSiteFinder, MaxEntScan, NNSPLICE, GeneSplicer) do not predict a difference in splicing. In summary, based on the above information the clinical significance of this variant cannot be determined with certainty at this time. This variant is classified as a variant of uncertain significance.

Variant interpreted as Likely benign and reported on 04-27-2020 by Lab or GTR ID 500031. GenomeConnect assertions are reported exactly as they appear on the patient-provided report from the testing laboratory. GenomeConnect staff make no attempt to reinterpret the clinical significance of the variant. This variant was reported in an individual referred for clinical diagnostic genetic testing.