KRAS c.35G>C, p.Gly12Ala

Reasoning

PS3 is met at Supporting strength. PMID:26037647 directly studied KRAS G12A, corresponding to NM_004985.5:c.35G>C (p.Gly12Ala), using purified-protein intrinsic and GAP-stimulated GTP hydrolysis assays and an RAF RBD-binding assay. G12A showed approximately 40- to 80-fold lower intrinsic GTP hydrolysis than wild type, supporting abnormal KRAS function. These biochemical assays are not listed among the VCEP's specifically approved RAS, MEK, or ERK activation assays, so the VCEP supplemental text limits unlisted assays to Supporting-level PS3/BS3 and does not permit the two-approved-assay Moderate upgrade.

VCEP rule

Well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product. Approved assays are available in the supplemental materials. ... One approved assay. ... Two or more different approved assays. ... Assays not listed here are presumed to lack sufficient historical evidence to evaluate the functional impact of a given variant and may only be sufficient for the application of PS3_Supporting or BS3_Supporting.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 PS3 rule: well-established functional studies supportive of a damaging effect are applicable; one approved assay supports PS3 at Supporting strength and two or more different approved assays support Moderate strength.
• SVI-RASopathy-VCEP-V2-Approved-Functional-Studies.xlsx, READ ME: assays not listed in the approved set may only be sufficient for PS3_Supporting or BS3_Supporting; two or more unique approved assay types can upgrade PS3 to Moderate.
• PMID:26037647, Results, intrinsic KRAS mutant GTPase activity: KRAS G12A had an approximately 40- to 80-fold decrease in GTP hydrolysis rate versus wild type; the study used purified KRAS mutants with intrinsic and GAP-stimulated hydrolysis assays and RAF RBD-binding measurements.
• The required variant searches in SVI-RASopathy-VCEP-V2-Approved-Functional-Studies.xlsx for c.35G>C, p.Gly12Ala, and p.G12A found no exact entry; the table does list G12A as a pathogenic validation control for the KRAS RAS activation assay, which is not a result for this tested variant.

Still missing

• The VCEP functional-studies file has no exact entry for c.35G>C, p.Gly12Ala, or p.G12A; its KRAS G12A occurrence is only listed as a pathogenic validation control in the RAS activation assay table, not as a variant-specific test result.
• No variant-specific result from two different VCEP-approved assay types was identified; therefore PS3 cannot be upgraded to Moderate.

Sources

Reasoning

The KRAS VCEP limits PM1 to the approved critical functional domains P-loop amino acids 10-17, Switch I amino acids 25-40, Switch II amino acids 57-64, and SAK amino acids 145-156, and excludes specific amino-acid residues as a PM1 route. The case variant is KRAS p.Gly12Ala, so residue 12 falls within the authoritative P-loop interval 10-17. The VCEP alignment file independently transcribes the same P-loop interval and states that KRAS shares the HRAS numbering for these regions. Therefore PM1 applies at Moderate strength.

VCEP rule

Applicable only to critical and well-established functional domains available in the supplementary table (P-loop [AA 10-17], SW1 [AA 25-40], SW2 [AA 57-64], SAK [AA 145-156]). Not applicable to specific amino acid residues (see PM5).

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 PM1 rule specifies P-loop [AA 10-17], Switch I [AA 25-40], Switch II [AA 57-64], and SAK [AA 145-156], and states that PM1 is not applicable to specific amino-acid residues.
• The case normalization is NP_004976.2:p.(Gly12Ala)/p.(G12A), placing the altered residue at amino acid 12.
• Alignment-with-PM1-domains.pptx.pdf transcribes P-loop (HRAS 10-17), states that KRAS shares identical residue numbering across the shaded regions, and gives the same P-loop 10-17 interval for KRAS.

Still missing

• No direct c.35G>C, p.Gly12Ala, or p.G12A entry was found in Alignment-with-PM1-domains.pptx.pdf; the variant was mapped by residue number to the transcribed P-loop interval.

Sources

Reasoning

The variant is a missense change, so the VCEP's missense PP3 path applies. The governing rule names REVEL >= 0.7 for missense variants; the available REVEL score is 0.841, meeting that threshold. The rule presents the missense REVEL path and the splicing-impact path separately; it does not require both a protein-impact predictor and a splice prediction together. SpliceAI is therefore not needed to establish PP3 on the applicable missense path, and its max delta score of 0.005 is not double-counted as PP3 evidence.

VCEP rule

For missense variants: REVEL >= 0.7. For splicing impact, predicted outcome must match disease mechanism.

Evidence considered

• Variant consequence: missense, NP_004976.2:p.(Gly12Ala), from the case variant assessment.
• Local REVEL predictor lookup: REVEL score 0.841 for KRAS c.35G>C, which is >=0.7.
• The VCEP rule has separate missense and splicing-impact clauses; both predictors are not required together for the missense clause.

Sources

Reasoning

The KRAS VCEP defines PS1 as the same amino acid change as a previously established pathogenic variant regardless of nucleotide change, applicable to observed analogous residue positions in HRAS, KRAS, MRAS, NRAS, RIT1, and RRAS2. The case identifies KRAS c.35G>C as p.Gly12Ala/p.G12A and ClinVar reports that same variant as pathogenic, but no qualifying previously established pathogenic variant with the same Gly12Ala amino-acid change and a different nucleotide change was identified. The searched VCEP alignment file contains PM1 domain mappings but no direct entry for c.35G>C, p.Gly12Ala, or p.G12A.

VCEP rule

Same amino acid change as a previously established pathogenic variant regardless of nucleotide change. Applicable for observed analogous residue positions in _HRAS, KRAS, MRAS, NRAS, RIT1,_ and _RRAS2._

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 PS1 rule: same amino acid change as a previously established pathogenic variant regardless of nucleotide change, with analogous residue positions in HRAS, KRAS, MRAS, NRAS, RIT1, and RRAS2.
• ClinVar variation 45122 identifies the case variant as NM_004985.5:c.35G>C (p.Gly12Ala) and reports Pathogenic/Likely pathogenic submissions; this is the same nucleotide change, not a separately nucleotide-different p.Gly12Ala variant.
• Alignment-with-PM1-domains.pptx.pdf was searched for c.35G>C, p.Gly12Ala, and p.G12A; it provides domain alignment text but no direct variant entry.

Still missing

• No direct c.35G>C, p.Gly12Ala, or p.G12A entry was found in Alignment-with-PM1-domains.pptx.pdf; this file is a PM1 alignment resource rather than a PS1 variant table.

Sources

Reasoning

The KRAS VCEP requires applicable autosomal-dominant RASopathy phenotype point scoring. Its governing rule states: "Strength adjustment using point-based scoring for autosomal dominant cases with phenotypic specifications: full points awarded with RASopathy phenotypes, reduced points applicable to cases with minimal phenotypic information (i.e. prenatal cases, cases with a clinical order of a RASopathy panel without clinical information, and cases with limited clinical information in other global tests (such as WES))." It further states: "The prevalence of the variant in affected individuals is significantly increased compared to the prevalence in controls. Note 1: Relative risk (RR) or odds ratio (OR), as obtained from case-control studies, is >5.0 and the confidence interval around the estimate of RR or OR does not include 1.0. See manuscript for detailed guidance. Note 2: In instances of very rare variants where case-control studies may not reach statistical significance, the prior observation of the variant in multiple unrelated patients with the same phenotype, and its absence in controls, may be used as moderate level of evidence." The rule thresholds are "≥5 points." for Strong, "≥3 points." for Moderate, and "≥1 points." for Supporting; with the available evidence, the assigned score is 0: 0 ≥ 1? no, 0 ≥ 3? no, and 0 ≥ 5? no. PMID:15696205 reports KRAS exon 2 mutations in 9/38 refractory versus 0/21 drug-sensitive lung adenocarcinomas, but this is a treatment-response comparison, not a RASopathy affected-versus-unaffected case-control study, and the paper does not identify c.35G>C or p.G12A. PMID:23182985 and PMID:26662311 report G12A in metastatic colorectal-cancer cohorts and treatment/outcome associations, not RASopathy phenotype prevalence or unaffected controls. No qualifying RASopathy case-control enrichment, multiple unrelated RASopathy cases with the same phenotype, or VCEP point assignment is documented; therefore PS4 is not met.

VCEP rule

Strength adjustment using point-based scoring for autosomal dominant cases with phenotypic specifications: full points awarded with RASopathy phenotypes, reduced points applicable to cases with minimal phenotypic information (i.e. prenatal cases, cases with a clinical order of a RASopathy panel without clinical information, and cases with limited clinical information in other global tests (such as WES)). ... The prevalence of the variant in affected individuals is significantly increased compared to the prevalence in controls. Note 1: Relative risk (RR) or odds ratio (OR), as obtained from case-control studies, is >5.0 and the confidence interval around the estimate of RR or OR does not include 1.0. See manuscript for detailed guidance. Note 2: In instances of very rare variants where case-control studies may not reach statistical significance, the prior observation of the variant in multiple unrelated patients with the same phenotype, and its absence in controls, may be used as moderate level of evidence. ... ≥5 points. ... ≥3 points. ... ≥1 points.

Evidence considered

• VCEP PS4 rule: point-based scoring is for autosomal-dominant cases with phenotypic specifications, with full points for RASopathy phenotypes and reduced points for minimal phenotypic information.
• VCEP PS4 rule: affected-variant prevalence must be significantly increased over controls; OR/RR >5.0 with a confidence interval excluding 1.0, or for very rare variants multiple unrelated patients with the same phenotype and absence in controls may provide moderate evidence.
• VCEP PS4 strength rows: ≥5 points Strong, ≥3 points Moderate, and ≥1 points Supporting; explicit evaluations are 0 ≥ 1? no, 0 ≥ 3? no, and 0 ≥ 5? no.
• PMID:15696205, Results paragraph 1: 9/38 refractory versus 0/21 drug-sensitive lung adenocarcinoma tumors had KRAS mutations; the comparison concerns drug response, not RASopathy case-control prevalence, and the exact G12A variant was not mentioned.
• PMID:23182985, Results/Table 1: G12A was observed in metastatic colorectal-cancer patients and analyzed for panitumumab treatment effects; no RASopathy control comparison was reported.
• PMID:26662311, Table 3 and Results: 11 metastatic colorectal-cancer tumors carried G12A; the paper reports survival outcomes, not RASopathy phenotype prevalence against unaffected controls.
• literature_pass.json packet.ai_classified_papers was empty; no already-read AI-classified PS4 paper required verification.

Still missing

• A RASopathy-specific affected-versus-control comparison with variant counts and/or VCEP phenotypic point assignments is not available.

Sources

Reasoning

The governing KRAS VCEP applies PM2 at supporting strength only when the variant is absent from controls in gnomAD. The VCEP does not specify a non-cancer or exome-only source, so the default all-comers data apply; gnomAD v4.1 contains the variant at 2 of 1,612,866 alleles (AF 1.24003e-06, 0.00012%), with 0 homozygotes. Because the variant is present rather than absent in the governing default control dataset, PM2 is not met, despite absence from the older gnomAD v2.1 non-cancer exome subset.

VCEP rule

The variant must be absent from controls (gnomAD).

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 CSPEC PM2 rule: the variant must be absent from controls (gnomAD).
• gnomAD v4.1 all-comers: variant present at AF 1.24003e-06 (0.00012%), 2/1,612,866 alleles, homozygotes = 0.
• GNOMAD_V2_1_NON_CANCER: variant absent from non-cancer exomes, 0/234,992 alleles, homozygotes = 0; this subset is not the governing default because the VCEP PM2 text does not specify non-cancer or exome-only data.

Sources

Reasoning

The VCEP limits PM4 to protein-length changes caused by in-frame deletions or insertions in a non-repeat region or to stop-loss variants. NM_004985.5:c.35G>C is a single-nucleotide missense substitution producing p.(Gly12Ala), with no protein-length change and no stop-loss consequence; therefore PM4 is not met.

VCEP rule

Protein length changes due to in-frame deletions/insertions in a non-repeat region or stop-loss variants. ... No known repetitive areas in gene. Use as described.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS specification v2.3 PM4 instruction: protein length changes are limited to in-frame deletions/insertions in a non-repeat region or stop-loss variants; the applicable strength rule requires no known repetitive areas in the gene.
• Variant assessment: NM_004985.5:c.35G>C is a missense substitution producing NP_004976.2:p.(Gly12Ala), with predicted protein positions 11-12 unchanged in length.

Sources

Reasoning

The governing KRAS VCEP applies BA1 when the gnomAD filtering allele frequency is at least 0.05%. The default all-comers gnomAD v4.1 data show 2 of 1,612,866 alleles, AF 1.24003e-06 (0.00012%), with 0 homozygotes, which is far below 0.05%; therefore BA1 is not met. The VCEP does not specify a non-cancer or exome-only source for this criterion, so the all-comers total is used.

VCEP rule

GnomAD filtering allele frequency ≥0.05%.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 CSPEC BA1 rule: gnomAD filtering allele frequency ≥0.05%.
• gnomAD v4.1 all-comers, NM_004985.5:c.35G>C / 12-25245350-C-G: total AF 1.24003e-06 (0.00012%), 2/1,612,866 alleles, homozygotes = 0; highest subpopulation AF 1.69592e-06 (0.00017%) in European (non-Finnish).

Sources

Reasoning

The governing KRAS VCEP applies BS1 at strong strength when the gnomAD filtering allele frequency is at least 0.025%. The default all-comers gnomAD v4.1 data show 2 of 1,612,866 alleles, AF 1.24003e-06 (0.00012%), with 0 homozygotes, which is below 0.025%; therefore BS1 is not met. The VCEP does not specify a non-cancer or exome-only source for this criterion, so the all-comers total is used.

VCEP rule

GnomAD filtering allele frequency ≥0.025%.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 CSPEC BS1 rule: gnomAD filtering allele frequency ≥0.025%.
• gnomAD v4.1 all-comers, NM_004985.5:c.35G>C / 12-25245350-C-G: total AF 1.24003e-06 (0.00012%), 2/1,612,866 alleles, homozygotes = 0; highest subpopulation AF 1.69592e-06 (0.00017%) in European (non-Finnish).

Sources

Reasoning

The variant is a missense change, so the VCEP's BP4 missense path applies. The governing rule names REVEL <=0.3; the available REVEL score is 0.841, which is above the cutoff and therefore does not satisfy BP4. The VCEP rule does not require a splice prediction together with the missense REVEL assessment. The available SpliceAI max delta score of 0.005 is not used to rescue BP4 or double-counted across criteria.

VCEP rule

For missense variants: REVEL <=0.3.

Evidence considered

• Variant consequence: missense, NP_004976.2:p.(Gly12Ala), from the case variant assessment.
• Local REVEL predictor lookup: REVEL score 0.841 for KRAS c.35G>C, which is >0.3 and fails the VCEP BP4 threshold.
• The VCEP BP4 rule names only REVEL for missense variants; a splice prediction is not required in combination.

Sources

Reasoning

The KRAS VCEP permits BP5 when points are awarded for an alternative molecular cause of a RASopathy in a different gene, with a phenotype consistent with expected RASopathy severity, or when a phenotype inconsistent with RASopathy is fully explained by a different causative variant. The VCEP strength rules are "≥ (-4) Points." for Strong, "≥ (-2) Points." for Moderate, and "≥ (-1) Point." for Supporting. No alternative causative variant, different-gene molecular cause, or phenotype fully explained by another variant is documented for this case; the assigned BP5 score is 0. Explicit threshold checks are 0 ≥ (-1)? no, 0 ≥ (-2)? no, and 0 ≥ (-4)? no. Therefore BP5 is not met.

VCEP rule

Points are awarded for an alternative molecular cause of a RASopathy in a different gene (and/or in conjunction with BP2) and the phenotype is consistent with expected severity of the RASopathy. Points are also awarded for phenotypes inconsistent with a RASopathy and fully explained by a different causative variant (e.g. WES testing). ... ≥ (-4) Points. ... ≥ (-2) Points. ... ≥ (-1) Point.

Evidence considered

• VCEP BP5 applicability: points are awarded for an alternative molecular cause of RASopathy in a different gene with phenotype consistent with expected severity, or for phenotypes inconsistent with RASopathy fully explained by a different causative variant.
• VCEP BP5 strength rows: ≥(-4) points Strong, ≥(-2) points Moderate, and ≥(-1) point Supporting; explicit evaluations are 0 ≥ (-1)? no, 0 ≥ (-2)? no, and 0 ≥ (-4)? no.
• The available case evidence documents KRAS c.35G>C/p.G12A and population/literature records but no alternative causative variant or different-gene molecular explanation.

Still missing

• No documented alternative molecular cause, causative variant in another gene, or phenotype explanation from WES/family testing is available.

Sources

Reasoning

The KRAS VCEP provides point-based PS2 strengths (Very Strong at 4 points, Strong at 2 points, and Moderate at 1 point), but the available case evidence contains no documented affected proband, parental testing, confirmed de novo status, or PS2 point assignment. The criterion therefore cannot be evaluated without family-level de novo evidence.

VCEP rule

4 Points. ... 2 Points. ... 1 Point.

Evidence considered

• The case evidence contains no family-level de novo or parental-testing result for KRAS c.35G>C (p.Gly12Ala).
• ClinGen RASopathy Expert Panel KRAS v2.3 PS2 rule: Very Strong requires 4 points, Strong requires 2 points, and Moderate requires 1 point.

Still missing

• Affected proband documentation and phenotype consistency
• Validated parental genotypes showing absence or presence of the variant
• Confirmed de novo status and any VCEP PS2 point assignment

Sources

Reasoning

The KRAS VCEP assigns PM5 Moderate for one likely pathogenic residue change at the same codon and Strong for at least two different likely pathogenic residue changes at the same codon observed in at least five probands. The available PM5 comparator collection contains 11 total candidates but zero same-residue candidates and reports a soft retrieval failure (HTTP Error 429). No qualifying alternate amino-acid change at codon 12 with the required likely pathogenic evidence was therefore available to assess either PM5 strength.

VCEP rule

≥2 different [likely] pathogenic residues changes at the same codon observed in ≥5 probands. ... 1 [likely] pathogenic residue change at the same codon.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 PM5 rules: ≥2 different [likely] pathogenic residue changes at the same codon observed in ≥5 probands gives Strong; 1 [likely] pathogenic residue change at the same codon gives Moderate.
• pm5_candidates.json reports total_candidates_seen=11, same_residue_candidates=0, expert_panel_plp_candidates=0, non_expert_majority_plp_candidates=0, and a soft retrieval error of HTTP Error 429; it provides no qualifying codon-12 comparator row.

Still missing

• The PM5 comparator collection contains zero same-residue candidates and records HTTP Error 429; qualifying codon-12 alternate-residue classifications require human or repeat retrieval review.

Sources

Reasoning

The KRAS VCEP provides point-based PM6 strengths (Strong at 2 points, Moderate at 1 point, and Supporting at 0.5 points), but the available case evidence contains no documented apparently de novo occurrence without parental testing and no PM6 point assignment. The criterion therefore cannot be evaluated without qualifying de novo case evidence.

VCEP rule

2 Points. ... 1 Point. ... 0.5 Points.

Evidence considered

• The case evidence contains no family-level apparently de novo observation, parental-testing limitation, or PM6 point assignment for KRAS c.35G>C (p.Gly12Ala).
• ClinGen RASopathy Expert Panel KRAS v2.3 PM6 rule: Strong requires 2 points, Moderate requires 1 point, and Supporting requires 0.5 points.

Still missing

• A documented affected proband with phenotype consistent with KRAS-related disease
• Evidence that the variant is apparently de novo despite unavailable or incomplete parental testing
• Any VCEP PM6 point assignment

Sources

Reasoning

The KRAS VCEP assigns PP1 Supporting for at least 3 informative meioses, Moderate for at least 5, and Strong for at least 7. No informative meioses, affected relatives, phenotype segregation, or variant genotypes in relatives are documented in the available case evidence.

VCEP rule

≥7 informative meioses. ... ≥5 informative meioses. ... ≥3 informative meioses.

Evidence considered

• The case evidence contains no documented informative meioses or family segregation data for KRAS c.35G>C (p.Gly12Ala).
• ClinGen RASopathy Expert Panel KRAS v2.3 PP1 thresholds are Supporting at at least 3 informative meioses, Moderate at at least 5, and Strong at at least 7.

Still missing

• Number and identity of informative meioses
• Variant genotypes and phenotypes of affected and unaffected relatives
• Segregation assessment across the family

Sources

Reasoning

The governing KRAS VCEP requires point-based phenotypic evidence from healthy homozygous or heterozygous individuals, with strength determined by the applicable negative point score. The available population data report allele counts and homozygote counts but do not establish that any carrier was a healthy adult, provide individual phenotype information, or provide the VCEP point assignment; therefore BS2 cannot be evaluated.

VCEP rule

Strength adjustment using point-based scoring based on phenotypic specifications. Phenotypic specifications: based on healthy homozygote or heterozygote individuals, reduced points for apparently unaffected heterozygous individuals, applicable to parent or sibling samples during clinical family evaluations.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 CSPEC BS2 instructions require strength adjustment using point-based scoring based on phenotypic specifications for healthy homozygote or heterozygote individuals.
• gnomAD v4.1 reports 0 homozygotes and does not provide phenotype or healthy-adult status for the 2 observed alleles.

Still missing

• Healthy adult heterozygote or homozygote observations with sufficient phenotype information and the corresponding VCEP point assignment are unavailable.

Sources

Reasoning

The KRAS VCEP requires one informative meiosis for BS4. The available case evidence contains no documented unaffected relative who was tested and does not establish non-segregation of KRAS c.35G>C (p.Gly12Ala), so BS4 cannot be assessed.

VCEP rule

Requires only one informative meiosis.

Evidence considered

• The case evidence contains no documented non-segregation result or tested unaffected relative for KRAS c.35G>C (p.Gly12Ala).
• ClinGen RASopathy Expert Panel KRAS v2.3 BS4 rule requires one informative meiosis.

Still missing

• An informative meiosis involving an affected proband and an unaffected relative
• Validated genotype of the unaffected relative
• Evidence that the variant does not segregate with disease

Sources

Reasoning

BP2 is applicable under the KRAS VCEP when an alternative molecular cause of a RASopathy is present in the same gene, the phenotype is consistent with the expected severity, and the associated point threshold is reached. The available case evidence documents KRAS c.35G>C (p.Gly12Ala) and literature observations of the variant in cancer cohorts, but does not document a second pathogenic KRAS variant, cis/trans phase, an affected proband with both variants, or a RASopathy phenotype and severity assessment. Consequently, no BP2 points can be assigned and the criterion remains not assessed rather than not met.

VCEP rule

Points are awarded for an alternative molecular cause of a RASopathy in the same gene (and/or in conjunction with BP5) and the phenotype is consistent with expected severity of the RASopathy. ... ≥ (-1) Point.

Evidence considered

• ClinGen RASopathy Expert Panel KRAS v2.3 CSPEC criterion entry for BP2: points require an alternative molecular cause of a RASopathy in the same gene (and/or in conjunction with BP5) and a phenotype consistent with expected severity; the supporting threshold is ≥ (-1) Point.
• Case evidence identifies NM_004985.5:c.35G>C (NP_004976.2:p.(Gly12Ala)) but contains no documented affected-proband cis/trans phase result or second pathogenic KRAS variant for BP2.
• The reviewed variant-specific publications concern cancer cohorts or biochemical/clinical KRAS G12A observations and do not provide the required same-individual alternative molecular cause, phase, and RASopathy-severity evidence for BP2.

Still missing

• Document whether an alternative pathogenic KRAS variant is present in the same individual.
• Determine cis/trans phase if two KRAS variants are present.
• Document the affected-proband RASopathy phenotype and whether its severity is consistent with the VCEP requirement.
• Obtain the VCEP point assignment needed to evaluate the ≥ (-1) Point threshold.

Sources