PIK3CA c.277C>T, p.(Arg93Trp)

Reasoning

Under the ClinGen Brain Malformations VCEP PS3 specification (PMID:31892348 quality metrics; BMVCEP Supplemental Document 1), R93W was functionally characterized in two independent cell-based assays. Rudd et al. (PMID:21266528) expressed p110alpha-R93W in stably transfected U2OS cells and showed increased phospho-AKT Ser473/Thr308 relative to WT, using WT as the basic positive/negative reference and E365K and H1047R gain-of-function mutants as positive controls for AKT activation, with western blots repeated in triplicate (technical replicates documented). This is a phosphorylation-class assay, one of the 4 VCEP-endorsed 'broadly accepted historically' assay classes. Burke & Williams (PMID:22949682) independently measured basal and pY-stimulated lipid kinase activity of purified recombinant p110alpha-R93W protein and found increased basal lipid kinase activity relative to WT, with H1047R included as a canonical activating control and lipid-binding replicate assays performed in triplicate. Both assays concordantly demonstrate a gain-of-function (activating) biochemical phenotype consistent with the PIK3CA disease mechanism (gain of function per VCEP notes). Neither paper documents a validation panel of >=8 independent P/LP and B/LB control variants required by the VCEP for strong-tier evidence (Supplemental Table 1), so the moderate/supporting tier criteria for validation-control breadth are not clearly met; strong tier is not awarded. Two concordant, orthogonal, well-controlled assays (western blot pAKT signaling assay plus in vitro lipid kinase activity assay), each with WT and known-activating-mutant controls and documented technical replicates, support PS3 at the Moderate strength.

Rule

PS3_Moderate (BMVCEP: two independent, appropriately controlled cell-based/biochemical assays showing concordant gain-of-function effect; validation panel of >=8 P/LP+B/LB controls not documented, precluding Strong)

Evidence considered

• Rudd et al. 2011 (PMID:21266528): p110alpha-R93W expressed in stably transfected U2OS cells under low-serum conditions showed increased phospho-AKT Ser473 and Thr308 compared with WT; direct quote: 'we observed that the exogenous expression of six mutants, p110alpha-R93W, -G106R, -G106V, -K111E, -delP449-L455, and -E453K, led to increased levels of phospho-AKT Ser473 compared to the wild-type p110alpha'; direct quote: 'we characterized these two mutants functionally and showed that R93W mutation is a gain-of function mutant that leads to increased phosphorylation of AKT on serine 473'; E365K (C2-domain) and H1047R (kinase-domain) were included as positive controls for AKT activation; western blots were repeated in triplicate (technical replicates documented).
• Burke & Williams 2012 (PMID:22949682): recombinant p110alpha-R93W protein assessed via lipid kinase assay with 5% PIP2-containing vesicles; direct quote: 'All cancer-linked mutations, including mutants in the ABD and ABD-RBD linker, increased basal lipid kinase activity compared with WT, with H1047R acting as the most potent activator'; WT served as reference/negative comparator and H1047R as a canonical activating positive control; assays performed in triplicate.
• Both papers assess the exact R93W variant (not R93Q or other substitutions at the same codon), satisfying variant-specific grounding.
• VCEP PS3 specification (cspec, instructionsToUse for PS3 label) requires assays from one of four broadly accepted classes (phosphorylation, DEPTOR binding, cell survivability, cell proliferation) for supporting-level evidence, and >=8 P/LP + B/LB validation controls for moderate/strong tiers; the phosphorylation assay (PMID:21266528) directly matches an endorsed class. The lipid kinase assay is an orthogonal biochemical readout of the same gain-of-function mechanism and was weighed as corroborating evidence for moderate strength given the concordance of two independent, appropriately controlled assays, though it is not itself one of the four VCEP-named classes.

Still missing

• Neither functional paper documents a formal validation panel of >=8 independent P/LP and B/LB control variants as required by the VCEP for evidence above supporting/moderate tier, precluding PS3_Strong.
• Biological replicate counts are not explicitly stated in either paper beyond triplicate technical replicates.

Sources

Reasoning

PM2 is met at Supporting strength under the PIK3CA Brain Malformations VCEP specification. The variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, providing absence from multiple reference control cohorts and satisfying the VCEP requirement for an absent/rare variant in at least one ethnically matched cohort population sample.

Rule

Absent/rare from controls in an ethnically-matched cohort population sample (≥1).

Evidence considered

• The governing Brain Malformations VCEP specification assigns PM2 Supporting when the variant is absent/rare from controls in an ethnically matched cohort population sample (≥1).
• The variant NM_006218.3:c.277C>T (p.Arg93Trp) is reported as absent from gnomAD v2.1.
• The variant is reported as absent from gnomAD v4.1.
• The variant is reported as absent from gnomAD-Canada v1.0.

Sources

Reasoning

The governing Brain Malformations VCEP specification (cspec) defines PM1_Supporting for PIK3CA solely by Table 4 critical-domain membership: kinase domain AA 322-483 and kinase domain AA 797-1068, and explicitly states this is domain-based and does not require hotspot recurrence. The affected residue in this variant, Arg93 (p.Arg93Trp), falls at protein position 93, which lies outside both approved Table 4 kinase-domain intervals (322-483 and 797-1068). Although OncoKB/Cancer Hotspots data (oncokb, hotspots) indicate R93 is a recurrently mutated somatic hotspot in cancer, the VCEP's PM1 rule for PIK3CA is domain-membership-based, not hotspot-based, so hotspot recurrence alone cannot substitute for domain membership under this specification. Therefore PM1 is not met under the applicable VCEP rule.

Rule

cspec PM1: defaultStrength Pathogenic Moderate, applied strength Supporting: 'Residues affecting critical functional domains provided in Table 4 for each gene.' Table 4 domains for PIK3CA: AA 322-483, AA 797-1068.

Evidence considered

• cspec Table 4 (gene notes) restricts PIK3CA PM1_Supporting to kinase domain intervals AA 322-483 and AA 797-1068; residue 93 falls outside both intervals.
• OncoKB and Cancer Hotspots data indicate R93 is a recognized somatic mutation hotspot, but the VCEP PM1 rule for PIK3CA is explicitly domain-based and does not permit hotspot recurrence as an independent pathway to PM1.

Sources

Reasoning

BA1 is not met. The variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, so there is no observed allele frequency exceeding the VCEP threshold of >0.0926%.

Rule

Allele frequency (>0.0926%).

Evidence considered

• The governing Brain Malformations VCEP specification sets BA1 at an allele frequency >0.0926%.
• The variant NM_006218.3:c.277C>T (p.Arg93Trp) is reported as absent from gnomAD v2.1.
• The variant is reported as absent from gnomAD v4.1.
• The variant is reported as absent from gnomAD-Canada v1.0.

Sources

Reasoning

BS1 is not met. The variant is absent from the available population datasets, with no observed allele frequency exceeding the VCEP threshold of >0.0185%.

Rule

Allele frequency (>0.0185%).

Evidence considered

• The governing Brain Malformations VCEP specification sets BS1 at an allele frequency >0.0185%.
• The variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0; therefore no population frequency supports BS1.

Sources

Reasoning

BS2 is not met. The available evidence documents absence from population databases but does not document at least 3 homozygotes in gnomAD/ExAC or at least 3 heterozygous occurrences in well-phenotyped family members as required by the VCEP specification.

Rule

Award BS2 if ≥3 homozygotes present in gnomAD or ≥3 heterozygous in well phenotyped family members.

Evidence considered

• The governing Brain Malformations VCEP specification requires at least 3 homozygotes in gnomAD or ExAC, or at least 3 heterozygous occurrences in well-phenotyped family members, for BS2.
• The population evidence reports the variant as absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0; no homozygote count or qualifying family observations are provided.

Sources

Reasoning

No functional assay data in the reviewed literature indicate a benign/neutral effect for R93W; both identified functional studies (PMID:21266528, PMID:22949682) demonstrate an activating (gain-of-function) biochemical phenotype consistent with a damaging effect, the opposite of what BS3 requires. No published assay showing normal/wild-type-like function for this variant was found.

Evidence considered

• Both reviewed functional papers for this variant (PMID:21266528, PMID:22949682) report increased pAKT signaling and increased basal lipid kinase activity for R93W relative to WT, i.e. an activating/damaging phenotype, which is contrary evidence to BS3 rather than supportive of it.

Still missing

• No functional study reporting a neutral/wild-type-like readout for R93W was identified during this review.

Sources

Reasoning

PS1 (per cspec Brain Malformations VCEP v1.1) requires that the same amino acid change, p.(Arg93Trp), be produced by a different nucleotide substitution that has already been established as pathogenic with no evidence of splicing/other functional impact difference. The case bundle contains no data identifying any alternate codon change (i.e., a different c. variant at codon 93 that also yields p.Arg93Trp) with a prior pathogenic classification. No such comparator was surfaced by prefetch, evidence, ClinVar records, or the PM5 comparator search (which targets the same residue, not the same amino-acid-change pathway used for PS1).

Rule

cspec PS1: defaultStrength Pathogenic Strong, rule text 'No change.' (uses unmodified generic ACMG PS1 definition: same amino acid change as an established pathogenic variant, via a different nucleotide change).

Still missing

• No search was performed/returned for an alternate-codon variant producing the identical p.(Arg93Trp) change with an established pathogenic classification; this is required to adjudicate PS1 and is absent from the bundle.

Reasoning

The applicable Brain Malformations VCEP specification requires confirmed parental absence of the variant for PS2_Moderate, and additionally requires affected-tissue detection with absence or lower allele fraction in another tissue for PS2_Strong. The case bundle provides no proband-level variant allele fraction, parental sequencing results, confirmed maternity/paternity, affected-tissue versus blood/skin testing, or technical validation details. ClinVar contains one submission with origin recorded as de novo, but it provides no auditable parental-testing evidence and cannot establish PS2 by itself.

Rule

VCEP PS2_Moderate requires criterion 1: the variant is present at a detectable allele fraction but absent from parental samples with confirmed maternity and paternity. VCEP PS2_Strong requires criterion 1 plus criterion 2: detectable variant in affected tissue and absent or at lower allele fraction in another tissue.

Evidence considered

• The Brain Malformations VCEP specification states that PS2_Moderate requires parental absence with confirmed maternity and paternity, while PS2_Strong additionally requires detectable variant in affected tissue and absence or lower allele fraction in another tissue.
• The case bundle contains no family-testing, parental-genotype, tissue-specific allele-fraction, or validation data for NM_006218.3:c.277C>T (PIK3CA p.Arg93Trp).
• ClinVar records one submission with origin listed as de novo, but the submission has no assertion criteria and does not provide the underlying parental-testing evidence needed for PS2.

Still missing

• Proband variant detection and allele fraction in the relevant affected tissue
• Confirmed maternity and paternity
• Variant absence from both parental samples with assay and coverage details
• For PS2_Strong, comparison of variant allele fraction across affected and other tissues
• Orthogonal or laboratory validation of the de novo finding

Sources

Reasoning

The governing PIK3CA Brain Malformations VCEP requires PM2 before PS4, assigns phenotype points using Table 2A, and determines strength from the total using Table 2B. PM2 is supported by absence from the available population cohorts, but the bundle contains no auditable exact-variant cerebral-malformation case series, phenotype-category assignments, or case-control enrichment statistics. COSMIC records 76 somatic occurrences, but cancer occurrence is not phenotype-specific evidence for the Brain Malformations disease context and cannot be converted into a VCEP PS4 point total from the available data.

Rule

VCEP PS4: after the PM2 prerequisite, assign each qualifying exact-variant case to one Table 2A phenotype category, sum the points, and assign strength from Table 2B.

Evidence considered

• The Brain Malformations VCEP specification requires PM2 before PS4, assigns phenotype points under Table 2A, and maps the total to PS4 strength using Table 2B.
• The exact variant NM_006218.3:c.277C>T (PIK3CA p.Arg93Trp) is reported as absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada.
• COSMIC reports COSV55874903 with 76 somatic occurrences, but provides no VCEP phenotype-category breakdown or cerebral-malformation case-control analysis.
• ClinVar contains laboratory submissions and an aggregate pathogenic/likely pathogenic label, not independently auditable phenotype case data for PS4.

Still missing

• Exact-variant primary-source cases or cohort data with documented cerebral-malformation phenotypes
• A single applicable VCEP Table 2A phenotype category and point assignment for each qualifying case
• Case-control enrichment statistics or sufficient auditable unrelated affected cases to calculate the PS4 point total
• Confirmation that reported cases are relevant to the Brain Malformations disease context rather than cancer-only somatic occurrence

Sources

Reasoning

PM5 requires identification of a different, previously established pathogenic missense change at the same residue (Arg93) with a different resulting amino acid (i.e., not Trp). The pm5_candidates artifact records that the same-residue comparator search failed (HTTP 429 Too Many Requests) rather than returning a definitive zero-candidate result; it explicitly instructs that this failure should not be used to adjudicate PM5. No usable comparator dataset for residue 93 was otherwise retrieved in this bundle.

Rule

cspec PM5: defaultStrength Pathogenic Moderate, rule text 'No change.' (uses unmodified generic ACMG PM5 definition: different missense change at same residue as an established pathogenic variant).

Evidence considered

• pm5_candidates.json reports total_candidates_seen=4 but same_residue_candidates=0, with an explicit error 'HTTP Error 429: Too Many Requests' and a warning that PM5 candidate harvesting failed softly and should not be used to adjudicate PM5 automatically.

Still missing

• Same-residue (Arg93) alternate-amino-acid comparator search failed with HTTP 429 (rate limited) per pm5_candidates artifact; a successful ClinVar/literature search for other pathogenic missense variants at codon 93 is needed before PM5 can be adjudicated.

Sources

Reasoning

The cspec Brain Malformations VCEP rule for PP2 applies specifically to PIK3CA (along with MTOR and AKT3, but not PIK3R2) and requires the ExAC/gnomAD missense constraint z-score to exceed 3.09. No gnomAD/ExAC missense constraint z-score for PIK3CA was retrieved or included anywhere in this case's prefetch, evidence, or source_registry data. Without the actual computed z-score value from a cited source, PP2 cannot be adjudicated.

Rule

cspec PP2: defaultStrength Pathogenic Supporting, rule text 'Missense constraint computed in ExAC/gnomAD was utilized. Award PP2 if the z-score > 3.09. (applicable to MTOR, PIK3CA and AKT3 but not PIK3R2).'

Still missing

• gnomAD/ExAC missense Z constraint score for PIK3CA was not retrieved in this case bundle; required to compare against the VCEP's >3.09 threshold before PP2 can be applied.

Reasoning

The VCEP permits BP2 at Supporting strength when the variant is observed in cis or trans with a known pathogenic variant in the same gene. The case evidence bundle contains no affected-proband observation, parental testing, haplotyping, or other phase-resolved evidence showing NM_006218.3:c.277C>T (p.Arg93Trp) in cis or trans with another known pathogenic PIK3CA variant.

Rule

Observed in cis or trans with a known pathogenic variant in the same gene.

Evidence considered

• The VCEP BP2 rule is: observed in cis or trans with a known pathogenic variant in the same gene.
• The consolidated case-evidence bundle reports the target variant and general ClinVar origins/classifications but does not report a second PIK3CA pathogenic variant with confirmed cis/trans phase in an affected proband or family.
• General variant-level pathogenic classifications, somatic cancer recurrence, and functional or population observations do not establish BP2 without a second pathogenic PIK3CA allele and phase information.

Still missing

• Phase-resolved evidence is needed to establish whether the variant is in cis or trans with another known pathogenic PIK3CA variant.
• The identity and pathogenicity evidence for any alleged second PIK3CA variant are not provided.

Sources

Reasoning

The VCEP retains BP5 without modification, but the bundle does not document a specific exact-variant case in which an independently established alternate molecular basis explains the disease. Somatic cancer occurrence, ClinVar classifications, and disease labels alone do not establish BP5.

Rule

BP5 Supporting: the variant is found in a case with an alternate molecular basis for disease; case-level alternate-basis evidence is required.

Still missing

• A patient-level exact-variant case with an independently established alternate molecular basis, including the alternate causal variant or diagnosis and relevant phenotype