PIK3CA c.946C>T, p.(Pro316Ser)

Reasoning

BMVCEP v1.1 specifies PM2 as Supporting when the variant is 'Absent/rare from controls in an ethnically-matched cohort population sample (<=1)' (index guidance: 'PM2 is Supporting with one person maximum'). PIK3CA NM_006218.3:c.946C>T (p.Pro316Ser) has 0 observed alleles across gnomAD v2.1 (exomes), gnomAD v4.1 (exomes+genomes, ~800k alleles), and gnomAD-Canada v1.0 (HostSeq genomes), i.e. well under the <=1 occurrence ceiling across multiple population cohorts spanning diverse ancestries. PM2_Supporting is met. Per the VCEP the strength is Supporting (not the generic ACMG moderate), and per VCEP rules PM2 is a prerequisite gate for PS4 phenotype-point criteria.

Rule

VCEP PM2 (Supporting): absent/rare from controls in an ethnically-matched cohort population sample, <=1 occurrence (Brain Malformations VCEP v1.1, CSPEC doc 467907148)

Evidence considered

• gnomAD v2.1 variant page: 3-178921464-C-T absent, 0 alleles (evidence.json GNOMAD_V2_1 search_status=absent)
• gnomAD v4.1 variant page: chr3-179203676-C-T absent, 0 alleles (evidence.json GNOMAD_V4_1 search_status=absent)
• gnomAD-Canada v1.0 API: 3-179203676-C-T absent, 0 alleles (evidence.json GNOMAD_CANADA_V1 search_status=absent)
• BMVCEP v1.1 PM2 strength rule: absent/rare from controls in ethnically-matched cohort (<=1), Supporting (cspec)

Still missing

• Per-population allele counts/coverage were not captured (absence-status scrape only), so a formal ethnically-stratified AF check per ancestry group rests on the global 'absent' status across three datasets; residual risk of a rare allele in an unqueried population cohort is low but not zero.

Sources

Reasoning

PS1 (same amino-acid change as an established pathogenic variant, regardless of nucleotide change) requires a previously curated pathogenic p.Pro316Ser change arising from a different nucleotide substitution. No such record exists in any collected source: the variant is absent from ClinVar (variation_ids: [], 'This variant is absent from ClinVar'), the literature pass returned 0 PMIDs across all evidence sources, and the PM5 candidate harvest identified no same-residue comparators. With no established pathogenic p.Pro316Ser change on record, PS1 cannot be met.

Rule

No change. (Brain Malformations VCEP PS1, default Pathogenic Strong)

Evidence considered

• clinvar
• pm5_candidates
• cspec

Still missing

• No ClinVar or literature record of any p.Pro316Ser change (same amino acid from an alternative nucleotide) was found; if such a record exists in uncurated literature it would need to be surfaced before PS1 could be reconsidered

Sources

Reasoning

Brain Malformations VCEP PM1 is awarded at Supporting strength only for 'Residues affecting critical functional domains provided in Table 4'. For PIK3CA, Table 4 lists kinase domain AA 322-483 and kinase domain AA 797-1068. The variant residue is Pro316 (NP_006209.2:p.(Pro316Ser)), which falls outside both approved intervals (316 < 322 and 316 < 797). The index guidance explicitly states that PM1_Supporting is domain-based and does not require hotspot recurrence only when the variant falls INSIDE a Table 4 approved domain; since P316 does not, domain-based PM1 is not met. The variant is additionally not a statistically significant Cancer Hotspots residue (evidence.json hotspots result: found=false).

Rule

Residues affecting critical functional domains provided in Table 4 for each gene. (PM1_Supporting; PIK3CA Table 4 domains: kinase domain AA 322-483 and kinase domain AA 797-1068)

Evidence considered

• cspec
• vcep_clingen_brainmalform_acmg_specifications_v1_1
• final_classification_framework

Still missing

• No alternative VCEP-approved functional domain or motif annotation covering residue 316 was retrieved; only the two Table 4 intervals (AA 322-483, AA 797-1068) govern PM1 in this framework

Sources

Reasoning

PM5 (Moderate, no VCEP change) requires a different missense change at the same amino-acid residue (Pro316) that has been determined pathogenic. The pm5_candidates artifact (classic same-residue missense mode) harvested 0 same-residue candidates out of 2 total candidates seen (0 expert-panel P/LP, 0 non-expert majority P/LP, 0 conflicted), and recommends no PM5 application. ClinVar is empty for this variant and the literature pass found no PMIDs. No alternate missense change at Pro316 with a pathogenic classification was identified, so PM5 is not met.

Rule

No change. (Brain Malformations VCEP PM5, default Pathogenic Moderate; classic same-residue missense comparator logic)

Evidence considered

• pm5_candidates
• clinvar
• cspec

Still missing

• PM5 candidate harvesting soft-failed with HTTP Error 429 (Too Many Requests), so ClinVar same-residue queries (PIK3CA[GENE] AND P316 / Pro316) were only partially executed; a re-run of the same-residue harvest would strengthen confidence
• No alternate amino-acid changes at Pro316 (e.g., p.Pro316Leu, p.Pro316Ala) with pathogenic/likely-pathogenic classifications were found in any collected source

Sources

Reasoning

BMVCEP v1.1 (CSPEC doc 467907148) sets the BA1 stand-alone benign threshold at allele frequency >0.0926% (adjusted from generic ACMG to maintain the 5x benign threshold). PIK3CA NM_006218.3:c.946C>T (p.Pro316Ser) is absent from all queried population datasets: gnomAD v2.1 (GRCh37 3-178921464-C-T, search_status 'absent'), gnomAD v4.1 (GRCh38 chr3-179203676-C-T, search_status 'absent'), and gnomAD-Canada v1.0 (HostSeq genomes, search_status 'absent'). Observed allele frequency is 0, far below the 0.0926% threshold, so BA1 is not met.

Rule

VCEP BA1: allele frequency >0.0926% (Brain Malformations VCEP v1.1, CSPEC doc 467907148)

Evidence considered

• gnomAD v2.1 variant page: 3-178921464-C-T absent (evidence.json GNOMAD_V2_1 search_status=absent)
• gnomAD v4.1 variant page: chr3-179203676-C-T absent (evidence.json GNOMAD_V4_1 search_status=absent)
• gnomAD-Canada v1.0 API: 3-179203676-C-T absent (evidence.json GNOMAD_CANADA_V1 search_status=absent)
• BMVCEP v1.1 BA1 strength rule: AF >0.0926% (cspec)

Still missing

• Population frequency derived from absence-status scrapes with empty source_data payloads; no explicit allele-count/coverage fields were captured for per-population AF confirmation, though all three datasets returned 'absent'.

Sources

Reasoning

BMVCEP v1.1 sets the BS1 benign-strong threshold at allele frequency >0.0185% (Supplemental Table 3). The variant is absent (AF = 0) from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, so the frequency is far below the 0.0185% threshold and BS1 is not met. Absence of the variant in large population cohorts is consistent with rarity, not with a benign population frequency.

Rule

VCEP BS1: allele frequency >0.0185% (Brain Malformations VCEP v1.1, CSPEC doc 467907148)

Evidence considered

• gnomAD v2.1 variant page: 3-178921464-C-T absent (evidence.json GNOMAD_V2_1 search_status=absent)
• gnomAD v4.1 variant page: chr3-179203676-C-T absent (evidence.json GNOMAD_V4_1 search_status=absent)
• gnomAD-Canada v1.0 API: 3-179203676-C-T absent (evidence.json GNOMAD_CANADA_V1 search_status=absent)
• BMVCEP v1.1 BS1 strength rule: AF >0.0185% (cspec)

Sources

Reasoning

BMVCEP v1.1 awards BS2 if >=3 homozygotes are present in gnomAD OR >=3 heterozygotes are seen in well-phenotyped family members. Homozygous arm: the variant is entirely absent from gnomAD v2.1, v4.1, and gnomAD-Canada, so there are zero homozygotes and this arm is definitively not met. Family arm: no well-phenotyped family-member observations of this variant are available (ClinVar has no entry; literature_pass found no PMIDs; no case reports in the evidence bundle), so this arm cannot contribute evidence. With neither arm satisfied, BS2 is not met on available evidence. Confidence is medium only because the family-member arm is unevaluated, not because the gnomAD homozygote arm is uncertain.

Rule

VCEP BS2: >=3 homozygotes in gnomAD OR >=3 heterozygous in well-phenotyped family members (Brain Malformations VCEP v1.1, CSPEC doc 467907148)

Evidence considered

• gnomAD v2.1/v4.1/gnomAD-Canada absence: no homozygotes possible (evidence.json search_status=absent for all three)
• ClinVar: no entry for PIK3CA c.946C>T (evidence.json clinvar found=false)
• Literature pass: 0 PMIDs identified, no family/phenotype observations (literature_pass.json total_pmids=0)
• BMVCEP v1.1 BS2 strength rule: >=3 homozygotes in gnomAD or >=3 heterozygous in well-phenotyped family members (cspec)

Still missing

• No well-phenotyped family-member observation data (>=3 heterozygous unaffected relatives) are available to evaluate the second BS2 arm; would only matter if homozygotes/family observations ever emerge.

Sources

Reasoning

BP2 (Brain Malformations VCEP v1.1, applicable as originally described) requires that the variant be 'Observed in cis or trans with a known pathogenic variant in the same gene' (supporting). No such observation exists for NM_006218.3:c.946C>T (p.(Pro316Ser)): the variant is absent from ClinVar (variation_ids [], no submissions), no PMIDs were identified across any evidence source (literature_pass total_pmids 0, criteria_index empty), no affected-proband or parental/family data are present in the case bundle to establish phase, and no second PIK3CA variant in cis or trans is documented anywhere in the evidence. BP2 is a positive-evidence criterion; with no documented co-occurrence the criterion is not met.

Rule

cspec_brain_malformations_v1.1: BP2 requires observation in cis or trans with a known pathogenic variant in the same gene; no such observation available

Evidence considered

• cspec ruleset BP2 entry: rule 'Observed in cis or trans with a known pathogenic variant in the same gene' (Benign Supporting)
• ClinVar search: no entry for PIK3CA c.946C>T, therefore no ClinVar-documented co-occurrence in cis or trans
• gnomAD v2.1 / v4.1 / gnomAD-Canada v1.0: variant absent from population cohorts; no population-based co-occurrence signal
• literature_pass.json: 0 PMIDs screened, no criteria-indexed papers for BP2

Still missing

• No parental or family testing data available to determine cis/trans phase of any second PIK3CA variant
• No second PIK3CA variant (known pathogenic or otherwise) observed in the proband or family in the case evidence
• No literature PMIDs and no full-text publications available to search for reported co-occurrence with a known pathogenic PIK3CA variant

Sources

Reasoning

The governing framework is the ClinGen Brain Malformations Expert Panel VCEP specification v1.1 (cspec), which applies to PIK3CA and modifies PS2. Under this specification, PS2_Strong requires BOTH Criteria 1 (variant present at a detectable allele fraction but absent from parental samples with confirmed maternity and paternity) AND Criteria 2 (variant present at a detectable allele fraction in an affected tissue but absent from or at lower allele fraction in another tissue); PS2_Moderate requires Criteria 1 alone, or Criteria 2 alone when parents are unavailable. This case bundle contains no proband genotyping data, no parental testing results (no maternity/paternity confirmation), no tissue allele-fraction comparisons, and no family history information. The variant is absent from ClinVar (no submissions reporting de novo occurrence) and the literature pass returned 0 PMIDs. Because the proband-level and parental-level data required to apply the VCEP PS2 rules are entirely absent, de novo evidence cannot be evaluated; PS2 is therefore not assessed pending availability of parental and/or multi-tissue testing data.

Rule

Award PS2_Strong if Criteria 1 AND Criteria 2 fulfilled (variant at detectable AF absent from parental samples with confirmed maternity and paternity; AND detectable AF in affected tissue absent/lower in another tissue). Award PS2_Moderate if Criteria 1 fulfilled OR parents unavailable but Criteria 2 fulfilled. None of the required proband/parental/tissue data are present in this case, so the criterion cannot be applied.

Evidence considered

• cspec PS2 VCEP rules (PS2_Strong requires Criteria 1 AND Criteria 2; PS2_Moderate requires Criteria 1 alone, or Criteria 2 alone if parents unavailable)
• clinvar: variant absent from ClinVar, no submissions documenting de novo or parental testing
• literature_pass.json: 0 PMIDs identified; no published case-level de novo/segregation data available

Still missing

• No proband genotype data for the variant (variant presence/allele fraction not documented)
• No parental testing results with confirmed maternity and paternity (required for PS2 Criteria 1)
• No affected-tissue vs other-tissue allele fraction comparison (required for PS2 Criteria 2)
• No published reports of this variant in de novo or mosaic cases (0 PMIDs; ClinVar absent)

Sources

Reasoning

PS3 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

Governing framework is the ClinGen Brain Malformations VCEP v1.1 (cspec, precedence rank 1), which specifies PS4 as a phenotype-point criterion: points are assigned per reported case according to Table 2A and strength is determined per Table 2B (Supporting 0.5-1.25, Moderate 1.5-3.25, Strong 3.5-15.75, Very Strong >=16 points), usable only if the variant meets PM2 (absent/rare from controls). The PM2 prerequisite is satisfied here (variant absent from gnomAD v2.1, v4.1, and gnomAD-Canada), but no phenotype points can be awarded because no affected individuals carrying NM_006218.3:c.946C>T are reported in any evidence source: ClinVar has no entry for this variant (no variation ID, no submissions, no classifications), COSMIC has no somatic occurrence, OncoKB reports no variant-specific reviewed evidence, and the literature pass returned 0 PMIDs. With zero reported cases there is no case-control enrichment and no case-series phenotype data to score under Table 2A, so PS4 cannot be evaluated and is returned not_assessed pending any case reports.

Rule

VCEP PS4: points assigned for phenotype per Table 2A; strength per Table 2B (Supporting 0.5-1.25, Moderate 1.5-3.25, Strong 3.5-15.75, Very Strong >=16); applicable only if variant meets PM2 (absent/rare from controls).

Evidence considered

• cspec: VCEP PS4 rule (Table 2A phenotype points, Table 2B strength thresholds, PM2 prerequisite)
• clinvar: variant absent, no reported cases or classifications
• gnomad_v2, gnomad_v4, gnomad_canada: variant absent from controls (PM2 prerequisite satisfied, confers no PS4 points by itself)

Still missing

• No reported cases or case-control data for c.946C>T in the literature (literature_pass: 0 PMIDs); Table 2A phenotype points cannot be computed
• No ClinVar submissions or classifications for this variant; no independent case counts available
• No COSMIC somatic occurrences; tumor-based points (0.25 per occurrence category under Table 2A) cannot be awarded

Sources

Reasoning

The Brain Malformations VCEP PP2 rule is quantitative: '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).' PIK3CA is an applicable gene, but the ExAC/gnomAD missense z-score for PIK3CA was not retrieved into the case bundle: hci_prior was not found (probability: null), the custom PP2 score is null, and no gene-cache/constraint file exists in the workspace. Without the numeric z-score the rule cannot be evaluated, so PP2 remains not assessed pending retrieval of the constraint metric.

Rule

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)

Evidence considered

• cspec
• final_classification_framework

Still missing

• ExAC/gnomAD missense z-score for PIK3CA not retrieved (hci_prior lookup not found; custom_pp2_score null); PP2 requires z-score > 3.09 and cannot be scored without it

Sources

Reasoning

BS3 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

The VCEP retains BP5 unchanged ('No change', Supporting): 'Variant found in a case with an alternate molecular basis for disease.' There is no case-level or literature information for this variant at all (literature_pass: 0 PMIDs, no ClinVar entry, no COSMIC occurrence), so it cannot be determined whether any carrier had a documented alternate molecular basis for disease. Evidence is insufficient; BP5 is returned not_assessed pending any case-level data describing an alternate molecular diagnosis in a carrier of c.946C>T.

Rule

VCEP BP5: 'No change' — Supporting; 'Variant found in a case with an alternate molecular basis for disease.'

Evidence considered

• cspec: BP5 rule definition ('Variant found in a case with an alternate molecular basis for disease')
• clinvar: variant absent, no case observations available

Still missing

• No case reports or literature describing this variant; cannot evaluate whether any carrier had an alternate molecular basis for disease

Sources