PTCH1 c.2173C>T, p.(Pro725Ser)

Reasoning

BS1 is strong benign evidence applying when the allele frequency is greater than expected for the disorder; the generic default threshold is >= 1% (generic SVI default, per-gene derivation preferred per Whiffin et al. 2017, PMID:28518168). No PTCH1 CSPEC or VCEP exists, so the generic threshold and the all-comers gnomAD defaults apply. gnomAD v4.1 reports a grpmax FAF of 0.012068 (1.2068%), driven by the Middle Eastern population at 1.4517% (88/6,062 alleles, 3 homozygotes), which exceeds the 1% threshold. This is the variant's highest population frequency, and the ancestry group is reasonably sized (6,062 alleles), so the elevation is not a small-sample artefact. The signal is version-specific: gnomAD v2.1 showed a grpmax FAF of only 0.001203 (0.12%) with a popmax of 0.2492%, but v2.1 contained no large Middle Eastern cohort, so v4.1 adds genuine ancestry coverage rather than noise. The global allele frequency remains low in both releases (0.000891 in v2.1; 0.000880 in v4.1), and the non-cancer subsets are below threshold (gnomAD v2.1 non-cancer exomes 0.0886%; gnomAD v3.1 non-cancer genomes 0.1028%), so BS1 rests entirely on the ancestry-specific maximum. Judged by the variant's highest observed population frequency against the 1% generic threshold, BS1 is met at strong. A reviewer should confirm whether the governing framework intends the ancestry-maximum (popmax/grpmax FAF) or the global allele frequency for BS1, since the two metrics give opposite answers.

Rule

BS1: strong benign, population AF >= 0.01 (generic SVI default; per-gene derivation preferred, Whiffin et al. 2017, PMID:28518168); evaluated on all-comers gnomAD v4.1 grpmax FAF / ancestry-maximum frequency.

Evidence considered

• gnomAD v4.1 grpmax FAF 0.012068 (1.2068%) >= 0.01; Middle Eastern popmax AF 1.4517% (88/6,062 alleles, 3 homozygotes).
• gnomAD v2.1 grpmax FAF 0.001203 (0.12%) < 0.01; popmax 0.2492% (Remaining individuals, 18/7,224 alleles); v2.1 lacked a large Middle Eastern cohort.
• Global allele frequency below 1% in both releases: 0.000891 (v2.1) and 0.000880 (v4.1).
• Non-cancer subsets below threshold (checked as sensitivity; not mandated by any VCEP): gnomAD v2.1 non-cancer exomes 0.0886%; gnomAD v3.1 non-cancer genomes 0.1028%.
• Threshold used verbatim: BS1 strong benign at AF >= 0.01, generic SVI default with per-gene derivation preferred (Whiffin et al. 2017, PMID:28518168); no gene-specific derivation is available for PTCH1.
• Framework check: generic ACMG/AMP 2015 fallback governs (no CSPEC/VCEP for PTCH1), so no non-cancer or exome-only population source is specified and the all-comers GNOMAD_V2_1/GNOMAD_V4_1 entries remain the default.

Still missing

• The governing framework does not specify whether BS1 uses the ancestry-maximum (popmax/grpmax FAF) or the global allele frequency. The call here rests on the ancestry-maximum (1.4517% / grpmax FAF 1.2068%); the global frequency (~0.088%) is about 11-fold below the 1% threshold.
• No PTCH1-specific CSPEC/VCEP exists to provide a gene-derived BS1 threshold or a mandated population subset, so the generic 1% default and all-comers gnomAD sources were used.

Sources

Reasoning

BS2 is strong benign evidence for a variant observed in a healthy adult for a disorder in which full penetrance is expected at an early age. gnomAD is composed of adults drawn from population-based cohorts and is not enriched for Gorlin syndrome. gnomAD v4.1 contains 1,413 heterozygous and 7 homozygous carriers (1,420/1,614,158 alleles), including 88 alleles with 3 homozygotes in Middle Eastern individuals and 1,037 alleles with 2 homozygotes in non-Finnish Europeans; gnomAD v2.1 contains 252 heterozygous carriers and no homozygotes. PTCH1-related Gorlin-Goltz syndrome is inherited autosomal dominant with complete penetrance and variable expressivity, with early-onset features such as odontogenic keratocysts and basal cell carcinomas (PMID:20485063). Adult carriers at a population frequency of roughly 0.09% (up to 1.45% in Middle Eastern individuals), and in particular 7 adult homozygotes, are incompatible with a fully penetrant early-onset dominant allele, so BS2 is met at strong. Caveat: gnomAD participants are not individually phenotype-screened, so 'healthy adult' status is inferred from cohort design rather than documented per individual; the variant has also been reported in a single Gorlin-Goltz patient (PMID:20485063) and an NBCCS index case (PMID:28733979), which does not contradict BS2 (those individuals are not healthy adults) but shows the variant is not purely a population polymorphism.

Rule

BS2: strong benign, variant observed in a healthy adult for a dominant disorder with full penetrance expected at an early age (ACMG/AMP 2015, PMID:25741868); supported by adult heterozygous and homozygous carriers in gnomAD v2.1/v4.1.

Evidence considered

• gnomAD v4.1: 1,413 heterozygotes and 7 homozygotes (1,420/1,614,158 alleles); Middle Eastern 88 alleles / 3 homozygotes; non-Finnish European 1,037 alleles / 2 homozygotes; Remaining 112 alleles / 2 homozygotes.
• gnomAD v2.1: 252 heterozygous carriers (252/282,872 alleles) and no homozygotes.
• PMID:20485063 states Gorlin-Goltz syndrome is inherited autosomal dominant with complete penetrance and variable expressivity, satisfying the 'full penetrance expected at an early age' premise of BS2.
• gnomAD cohorts are population-based adult samples not ascertained for Gorlin syndrome, so carriers are treated as apparently healthy adults for this criterion; this is an assumption from cohort design rather than per-individual phenotype data.
• PM2/BA1 screening of the same datasets shows a population frequency of ~0.088% globally (up to 1.45% Middle Eastern), i.e. far too common for a fully penetrant early-onset dominant pathogenic allele.

Still missing

• gnomAD participants are not individually phenotype-confirmed as unaffected adults; 'healthy' status is inferred from cohort design, not documented per carrier.

Sources

Reasoning

BP1 applies to a missense variant in a gene for which primarily truncating variants are known to cause disease. That is the situation for PTCH1 and this variant. PMID:28733979, in a 22-patient NBCCS cohort, reports that the majority of described PTCH1 mutations produce a truncated protein, that truncating mutations were found in 87% of the patients (11 frameshift, 5 nonsense, one splice-site among 17 novel truncating changes), and that the gene's germline disease mechanism is loss of function through premature truncation, whereas only 2 novel missense changes were identified. The gene-level germline loss-of-function/truncating mechanism for PTCH1 disease is independently recorded in the case gene-context artifact, and this variant is a missense substitution (p.Pro725Ser). Generic ACMG/AMP 2015 BP1 is therefore met at supporting strength. The usual caution applies: PTCH1 missense variants are not entirely absent from disease (rare pathogenic missense changes, e.g. in the sterol-sensing domain, and the p.R1350W predicted-pathogenic change reported in PMID:28733979), so BP1 is applied at supporting strength only and should not be counted as strong evidence of benignity.

Evidence considered

• PMID:28733979 direct quote: 'The majority of the described mutations result in premature protein truncation.'
• PMID:28733979 cohort counts: 17 novel truncating mutations (11 frameshift, 5 nonsense, one splice-site) versus 2 novel missense mutations; 'truncating mutations in 87% of the patients'.
• PMID:28733979 mechanism statement: PTCH1 truncating mutations produce a protein 'which is unable to supress Smoothened protein and continuously activates the downstream pathway', i.e. disease is driven by loss of function.
• Gene-level germline loss-of-function disease mechanism for PTCH1 (pvs1_gene_context: lof_mechanism_supported=true, PVS1 gene gate 'eligible'), consistent with truncating variants being the primary disease mechanism.
• Variant consequence class: missense substitution NP_000255.2:p.(Pro725Ser) from the normalized case description.

Still missing

• No VCEP/gene-specific PTCH1 BP1 rule was available; generic ACMG/AMP 2015 BP1 was applied at supporting strength, with the standard caution that PTCH1 missense pathogenic variants are rare but not absent.
• No quantitative gene-wide truncating-versus-missense pathogenic variant ratio was available for PTCH1; the qualitative cohort and mechanism evidence in PMID:28733979 was used.

Sources

Reasoning

BP5 is met when the variant is found in a case that has an alternate molecular basis for the disease. PMID:20485063 reports a 52-year-old man with Gorlin-Goltz (naevoid basal cell carcinoma) syndrome in whom genetic evaluation identified the pathogenic PTCH1 c.2726dupA (exon 17) causing the syndrome, and additionally the separate PTCH1 p.P725S (exon 14), described by the authors as 'of uncertain functional meaning.' The patient's disease is fully explained by the truncating c.2726dupA allele, an alternate molecular basis; p.P725S contributes nothing to the phenotype and is documented in that affected case only as an incidental secondary variant. This is the ACMG BP5 scenario, applied at supporting (default) strength. Caveat: the co-occurring causal allele is in the same gene rather than a different gene, which some frameworks restrict; even so, it is a distinct allele providing the alternate molecular basis in an affected case, so supporting-level BP5 is warranted. The independent benign corroboration is that the exact variant is a low-frequency population variant (gnomAD v4.1 AF 0.000879716, 1420/1614158 alleles, 7 homozygotes) with a benign ClinVar aggregate.

Evidence considered

• PMID:20485063 (single-patient case report, 52-year-old male with Gorlin-Goltz syndrome): genetic analysis found the pathogenic PTCH1 c.2726dupA (exon 17) causing the syndrome plus the separate PTCH1 p.P725S (exon 14), described as 'of uncertain functional meaning' - i.e. the affected case has an alternate molecular basis (c.2726dupA) explaining the phenotype.
• PMID:25741868 defines BP5 ('variant found in a case with an alternate molecular basis for disease') and its default supporting strength.

Still missing

• BP5 rests on a single case report (PMID:20485063); the co-occurring causal variant is in the same gene (PTCH1), which some frameworks restrict to a different gene.

Sources

Reasoning

PS1 requires the same amino-acid change (p.Pro725Ser) to have been previously established as pathogenic. No such established pathogenic p.Pro725Ser variant exists. The two germline reports of this exact change do not establish pathogenicity: in PMID:20485063 the p.P725S variant is an incidental secondary finding in a Gorlin-Goltz patient whose syndrome is fully explained by the pathogenic PTCH1 c.2726dupA frameshift, and the authors explicitly call p.P725S 'of uncertain functional meaning'; in PMID:28733979 the variant is found in one Spanish NBCCS index case (IC11) but the authors themselves report that in silico analysis with eight algorithms predicted p.P725S to be a neutral change, and they call for further studies. The only pathogenic claim attached to p.P725S is from somatic medulloblastoma tumour data (COSMIC, a single somatic record in this case), which cannot establish a germline pathogenic role for a variant that is also present in 1000 Genomes (rs149258400). ClinVar gives no support: 20 submissions, no expert-panel review, all classifications Benign or Likely benign (aggregate Benign/Likely benign, 2 stars, no conflicts). Therefore the PS1 prerequisite (a previously established pathogenic variant with this amino-acid change) is absent.

Evidence considered

• ClinVar record for this variant (variation 41654, NM_000264.5(PTCH1):c.2173C>T p.Pro725Ser): 19 submitters plus one non-classifying submission, no expert panel, aggregate classification Benign/Likely benign with review status 'criteria provided, multiple submitters, no conflicts' (2 stars); individual laboratory classifications are Benign, Likely benign, or Uncertain significance (single Biesecker Lab submission) - no Pathogenic or Likely pathogenic assertion.
• PMID:20485063 direct quote: 'Moreover, the evaluation revealed the variant p.P725S in the 14th exon, which was of uncertain functional meaning.' The pathogenic change in that patient is PTCH1 c.2726dupA (exon 17), not p.P725S.
• PMID:28733979 direct quote: 'Missense p.P725S was predicted to be a neutral change; however, it has already been reported as pathogenic in medulloblastoma tumours 56.' The same paragraph notes the variant is in the 1000 Genomes database (rs149258400) with unclear clinical significance, and that the NBCCS patient carrying it did not have medulloblastoma.
• Case evidence for attempted re-use of the same amino-acid change: the only somatic report is COSMIC COSV59501874 (single somatic occurrence); this is tumour data, not a germline pathogenic assertion.
• Search performed for an alternate nucleotide change encoding p.Pro725Ser: none is recorded; the ClinVar residue-725 series (c.2173C>T p.Pro725Ser, c.2174C>G p.Pro725Arg, c.2174C>A p.Pro725His) contains no second established pathogenic p.Pro725Ser.

Still missing

• No VCEP/gene-specific PTCH1 specification was available, so generic ACMG/AMP 2015 PS1 semantics were applied.

Sources

Reasoning

PS2 (Richards et al. 2015, PMID:25741868) requires the variant to have arisen de novo with both maternity and paternity confirmed, in a patient with the disease and no family history. No source in this case reports parental testing or a confirmed de novo event for PTCH1 c.2173C>T (NP_000255.2:p.(Pro725Ser)). The two published probands carrying this variant leave de novo status unconfirmed: in the Spanish NBCCS cohort (PMID:28733979) p.P725S was detected in index patient IC11 (65-year-old male meeting Kimonis criteria for NBCCS) with no relatives genotyped and no parental testing reported, and in the Gorlin-Goltz case report (PMID:20485063) p.P725S was an incidental second PTCH1 variant of unclear functional meaning alongside the pathogenic c.2726dupA, again with no family or parental analysis. De novo origin is additionally contradicted by the variant's recurrence as an established polymorphism in unrelated individuals and in unaffected reference populations: gnomAD v2.1 carries it on 252/282,872 alleles (AF 0.00089; 0 homozygotes), gnomAD v4.1 on 1,420/1,614,158 alleles (AF 0.00088; 7 homozygotes), and it is listed in the 1000 Genomes Project as rs149258400 (PMID:28733979). PS2 is therefore not met.

Rule

ACMG/AMP 2015 PS2: de novo (both maternity and paternity confirmed) in a patient with the disease and no family history (PMID:25741868)

Evidence considered

• PS2 definition and requirement for confirmed maternity and paternity taken from the ACMG/AMP 2015 standards (Richards et al., PMID:25741868); no VCEP/gene-specific specification for PTCH1 exists (ClinGen CSPEC not found, framework falls back to generic ACMG).
• Searched the full-text of PMID:28733979 (22 unrelated Spanish NBCCS patients plus 5 relatives; Sanger sequencing of PTCH1 exons 2-23) for proband-level parental testing or a de novo statement concerning p.P725S: the variant is reported only as present in index case IC11, and no parents of a p.P725S carrier were tested; the paper's only genotype-in-relatives analysis (Suppl. Table S2) concerns other index cases.
• PMID:28733979 states p.P725S had previously been detected in medulloblastoma tumours and listed in COSMIC, and is also present in the 1000 Genomes database as rs149258400, i.e. the variant is a known recurrent population polymorphism rather than a private de novo change.
• PMID:20485063 (52-year-old male with Gorlin-Goltz syndrome) describes p.P725S only as an incidental second variant 'of uncertain functional meaning'; the phenotype is attributed to the pathogenic PTCH1 c.2726dupA, and no family members or parents of that patient were genotyped.
• gnomAD v2.1: 252/282,872 alleles (AF 0.000890862), 0 homozygotes, highest population AF 0.249% in Remaining individuals - independent, unrelated carriers make a de novo origin untenable.
• gnomAD v4.1: 1,420/1,614,158 alleles (AF 0.000879716), 7 homozygotes, highest population AF 1.45% in Middle Eastern individuals - independent, unrelated carriers and homozygotes make a de novo origin untenable.

Still missing

• No parental testing (maternity/paternity confirmation) reported for any proband carrying p.P725S; a de novo event has never been confirmed or assumed in the literature.

Sources

Reasoning

Generic ACMG PS4 requires that the prevalence of the variant in affected individuals be significantly increased compared with controls (a case-control/cohort enrichment signal). No case-control study, cohort enrichment analysis, odds ratio or p-value for PTCH1 c.2173C>T (p.Pro725Ser) exists in this case. No PTCH1 ClinGen VCEP specification is available (cspec not found; no local framework), so there is also no gene-specific numeric PS4 threshold to test. The nearest variant-level observations are single-case/descriptive only: PMID:28733979 found p.P725S in 1 of 22 unrelated Spanish NBCCS patients (index case IC11) and absent from 400 control alleles from 200 cancer-free controls, and PMID:20485063 reported it as an incidental PTCH1 variant 'of uncertain functional meaning' alongside a pathogenic truncating variant. A single carrier versus 400 control alleles is not statistically significant enrichment and does not establish increased prevalence in cases. In the opposite direction, the variant is carried in the general population (gnomAD v2.1 AF 0.000890862, 252/282872 alleles; gnomAD v4.1 AF 0.000879716, 1420/1614158 alleles, 7 homozygotes), and the two large secondary-finding studies screened here (PMID:22703879, 572 individuals; PMID:24728327, 681 healthy adults) list no PTCH1 variant as pathogenic. PS4 is therefore not met; this is an absent-evidence not-met, not a refuted one.

Evidence considered

• No PTCH1 ClinGen VCEP/CSPEC specification was found (cspec found: false; no local gene framework), so generic ACMG/AMP PS4 applies and there is no gene-specific numeric enrichment threshold to compare against.
• PMID:28733979 reports PTCH1 p.P725S (rs149258400) in the NBCCS index patient IC11 (1 of 22 unrelated Spanish NBCCS patients) and states the variant was not found in 400 alleles from 200 healthy controls; this is a single carrier observation, not a significant case-control enrichment.
• PMID:20485063 reports p.P725S as an incidental secondary PTCH1 variant 'of uncertain functional meaning' in a single Gorlin-Goltz patient whose syndrome was attributed to a different pathogenic PTCH1 variant.
• gnomAD v2.1 and v4.1 show the variant present in the general population (v2.1 AF 0.000890862, 252/282872 alleles, 0 homozygotes; v4.1 AF 0.000879716, 1420/1614158 alleles, 7 homozygotes), consistent with absence of case-specific enrichment.
• PMID:22703879 (572 ClinSeq exomes, 37 cancer genes) and PMID:24728327 (681 healthy adults) were screened; neither approaches PTCH1 p.P725S as a case-enrichment finding.

Still missing

• No case-control or case-enrichment study (affected vs control prevalence, OR, or p-value) for PTCH1 c.2173C>T exists in the retrieved evidence.

Sources

Reasoning

PM1 requires location in a mutational hotspot and/or a critical, well-established functional domain without benign variation. No gene-specific domain table is available for PTCH1 (the VCEP materials carry no domain_tables entry and no CSPEC), so the generic ACMG/AMP 2015 definition was applied against the primary literature and population data. Hotspot: PMID:28733979 states explicitly, for a 22-patient NBCCS cohort, that no hotspots for mutations were found in PTCH1; the recurrent concentration of inactivating variants is regional (first and fourth extracellular loops, which form the SHH-binding site) rather than at a single codon, and no hotspot call was returned for PTCH1 P725. Domain: PTCH1 encodes 12 transmembrane domains with two large extracellular loops, intracellular loop 3 (ICL3) and the sterol-sensing domain; residue 725 lies in ICL3 (PMID:28733979 localizes the two missense changes to 'the intracellular loop 3 and C-terminus'), not in the SHH-binding extracellular loops where the authors report that most inactivating variants concentrate, and only ~10% of their inactivating variants fell in the sterol-sensing domain. Without benign variation: this requirement fails at this codon. gnomAD v4.1 carries 1420/1,614,158 alleles (AF 0.088%, 7 homozygotes; Middle Eastern AF 1.45%, grpmax FAF 1.21%), and gnomAD v2.1 carries 252/282,872 alleles (AF 0.089%). ClinVar lists additional, non-pathogenic missense changes at the same codon (c.2174C>G p.Pro725Arg: Ambry VUS, Labcorp Benign, St Jude VUS, Fulgent VUS; c.2174C>A p.Pro725His: two VUS) and at the adjacent codon 726 (p.Pro726Arg, p.Pro726Ser, p.Pro726Ala: Likely benign/Benign/VUS), demonstrating that this region tolerates missense variation in the population rather than being depleted of benign variation. PM1 is therefore not met.

Evidence considered

• PMID:28733979 direct quote: 'Although no hotspots for mutations were found, most of the truncating mutations were located within the first and the forth extracellular loops, both involved in the binding of SHH.'
• PMID:28733979, protein architecture: 'PTCH1 gene encodes a 1450 amino-acid glycoprotein, which has 12 transmembrane-spanning domains, intracellular amino- and carboxy-terminal regions and two large hydrophilic extracellular loops'.
• PMID:28733979 localization of this variant: 'Genetic screening also detected two missense mutations located in the intracellular loop 3 and C-terminus of PTCH1, respectively' - p.P725S is the intracellular loop 3 (ICL3) change; the companion variant is p.R1350W.
• PMID:28733979 domain distribution of inactivating variants: 65% in the SHH-binding region (extracellular loops) and only 10% within the sterol-sensing domain.
• PMID:28733979 also reports strictly conserved residues in a conserved region around these missense changes, and that p.P725S was predicted neutral by the eight in silico tools used (Suppl. Table S5); conservation alone does not establish a critical functional domain.
• gnomAD v4.1 (GRCh38 chr9-95468828-G-A): total AF 0.0008797, AC 1420/AN 1614158, 7 homozygotes; highest population Middle Eastern AF 1.45167% (88/6062, 3 homozygotes); grpmax FAF 0.0120684.
• gnomAD v2.1 (9-98231110-G-A): total AF 0.000890862, 252/282872 alleles, 0 homozygotes; highest population 'Remaining individuals' AF 0.249% (18/7224).
• ClinVar API query for PTCH1 residue 725 (searched via NCBI E-utilities for PTCH1[gene] AND Pro725): VCV000409209 c.2174C>G p.Pro725Arg (Ambry Genetics Uncertain significance, Labcorp Genetics Benign, St. Jude Uncertain significance, Fulgent Genetics Uncertain significance; conflicting classifications); VCV001013881 c.2174C>A p.Pro725His (Labcorp Genetics Uncertain significance, Ambry Genetics Uncertain significance); VCV000041654 c.2173C>T p.Pro725Ser (this variant; Benign/Likely benign). Adjacent codon 726 variants VCV000135877 p.Pro726Arg, VCV000409194 p.Pro726Ser, VCV003701216 p.Pro726Ala, VCV000486197 p.Pro726Leu are Likely benign/Benign/Uncertain significance, indicating the region tolerates missense change.
• Hotspot screenshot/query for PTCH1 P725 returned no hotspot row ('This variant does not lie in a statistically significant hotspot').

Still missing

• No authoritative gene-specific domain table exists for PTCH1 in this case (vcep_materials domain_tables is empty), so critical-domain boundaries were taken from the published structural description in PMID:28733979 rather than from an approved VCEP list.
• PMID:28733979 does note that mutations concentrate in the extracellular loops, the intracellular loop and the N-terminal region at a regional level; this regional language was considered but is not a hotspot or a defined critical domain, and the codon itself shows population and ClinVar missense tolerance.

Sources

Reasoning

PM2 (supporting pathogenic after the ClinGen SVI 2020 downgrade; threshold <= 0.0001, PMID:25741868) requires the variant to be absent or at extremely low frequency in population databases. The variant is unambiguously present. gnomAD v4.1 reports a total AF of 0.000880 (1,420/1,614,158 alleles, 7 homozygotes) and gnomAD v2.1 a total AF of 0.000891 (252/282,872 alleles, 0 homozygotes), both approximately nine-fold above the 0.0001 supporting cutoff; the v4.1 grpmax FAF is 1.2068% and the highest ancestry frequency is 1.4517% (Middle Eastern, 88/6,062 alleles). The non-cancer subsets agree (gnomAD v2.1 non-cancer exomes 0.0886%, 210/236,946 alleles; gnomAD v3.1 non-cancer genomes 0.1028%, 152/147,906 alleles), so no subset choice rescues a rare-variant claim. Because the variant is common rather than rare, PM2 is not met; the moderate-strength form of PM2 is likewise not met.

Rule

PM2: supporting pathogenic, population AF <= 0.0001 (ClinGen SVI 2020 downgrade, PMID:25741868); evaluated on all-comers gnomAD v2.1/v4.1 total and ancestry-maximum frequencies.

Evidence considered

• gnomAD v4.1 total AF 0.000880 (1,420/1,614,158 alleles; 7 homozygotes; grpmax FAF 0.012068).
• gnomAD v2.1 total AF 0.000891 (252/282,872 alleles; 0 homozygotes; grpmax FAF 0.001203).
• Non-cancer subsets (checked as sensitivity; not mandated by any VCEP): gnomAD v2.1 non-cancer exomes 0.0886%; gnomAD v3.1 non-cancer genomes 0.1028%.
• gnomAD-Canada v1.0 (HostSeq genomes) inconsistent between artifacts (0.1466% in case_summary.json vs absent in evidence_brief.json); either way above the 0.0001 threshold.
• Threshold used verbatim: PM2 supporting at AF <= 0.0001 (ClinGen SVI 2020 downgrade recommendation, PMID:25741868).
• Framework check: no PTCH1 CSPEC/VCEP, so generic ACMG/AMP 2015 defaults and all-comers gnomAD sources apply.

Sources

Reasoning

PM5 requires a different missense change at the same amino-acid residue (PTCH1 codon 725) to have been determined pathogenic. Two different missense changes at residue 725 are recorded in ClinVar, and neither is pathogenic: c.2174C>G (p.Pro725Arg, VCV000409209) is classified Uncertain significance by Ambry Genetics, St. Jude Molecular Pathology and Fulgent Genetics and Benign by Labcorp Genetics (conflicting classifications overall); c.2174C>A (p.Pro725His, VCV001013881) is classified Uncertain significance by both submitters (Labcorp Genetics and Ambry Genetics, no conflicts). No submitter reports a Pathogenic or Likely pathogenic classification for any residue-725 missense variant, and no established pathogenic same-residue change appears in the literature - the only other missense reported alongside p.P725S in PMID:28733979 is p.R1350W, a different residue, and that too was classified as uncertain in the 1000 Genomes context. The automated same-residue candidate harvest returned zero candidates for this variant. PS1 (the same amino-acid change) is likewise unmet, so PM5 has no anchor and is not met.

Evidence considered

• ClinVar API query (NCBI E-utilities, PTCH1[gene] AND Pro725) returned exactly three residue-725 missense records: this variant (c.2173C>T p.Pro725Ser) and two alternative changes, c.2174C>G p.Pro725Arg and c.2174C>A p.Pro725His. All adjacent codon-726 records (p.Pro726Arg/Ser/Ala/Leu) are separate residues and therefore outside PM5 scope.
• VCV000409209 (c.2174C>G, p.Pro725Arg) per-submitter germline classifications: Ambry Genetics Uncertain significance; Labcorp Genetics Benign; St. Jude Molecular Pathology Uncertain significance; Fulgent Genetics Uncertain significance - no pathogenic assertion.
• VCV001013881 (c.2174C>A, p.Pro725His) per-submitter germline classifications: Labcorp Genetics Uncertain significance; Ambry Genetics Uncertain significance.
• PM5 comparator candidate collection for this variant: found=false, 0 candidates seen, 0 same-residue candidates, mode 'unknown'; the artifact explicitly instructs review of whether any collected same-residue comparator satisfies the governing PM5 rule and warns not to apply PM5 from the artifact alone.
• PMID:28733979 reports the residue-725 variant itself and p.R1350W at a different residue (C-terminus), with no other residue-725 change reported.
• No paper among the fetched full texts reports a pathogenic different missense change at PTCH1 residue 725.

Still missing

• No VCEP/gene-specific PTCH1 PM5 rule was available, so generic ACMG/AMP 2015 same-residue semantics were applied.
• HGMD and locus-specific PTCH1 databases were not queried for residue-725 comparators; ClinVar and the assembled literature were searched instead. Because every residue-725 comparator found is VUS or Benign, a database-only pathogenic comparator is unlikely to change the verdict.

Sources

Reasoning

PM6 (Richards et al. 2015, PMID:25741868) covers an assumed de novo occurrence for which maternity and paternity have not been confirmed. No source in this evidence base states or assumes a de novo event for PTCH1 c.2173C>T (p.Pro725S). In PMID:28733979 the variant is presented as an already-known missense change (rs149258400, listed in the 1000 Genomes Project and previously reported in medulloblastoma/COSMIC) found in one NBCCS index patient, and the authors' own in silico analysis predicted it could represent a neutral change; PMID:20485063 reports it only as an incidental second PTCH1 variant of unclear functional meaning in a proband whose Gorlin-Goltz phenotype is explained by the pathogenic c.2726dupA. The variant also recurs independently of any proband: 252 alleles in gnomAD v2.1 (AF 0.00089, 0 homozygotes) and 1,420 alleles with 7 homozygotes in gnomAD v4.1 (AF 0.00088), plus an appreciable Middle Eastern population frequency of 1.45%. An assumed de novo origin in a single patient is therefore both unasserted and contradicted by population data, so PM6 is not met.

Rule

ACMG/AMP 2015 PM6: assumed de novo, but without confirmation of maternity and paternity (PMID:25741868)

Evidence considered

• PM6 definition (assumed de novo without confirmation of maternity and paternity) taken from the ACMG/AMP 2015 standards (Richards et al., PMID:25741868); generic ACMG applies because no PTCH1 VCEP specification exists.
• Full-text review of PMID:28733979 found no de novo claim for p.P725S: the variant is described as previously detected in medulloblastoma tumours, reported in COSMIC, and present in 1000 Genomes as rs149258400 'with unclear clinical significance', detected in index patient IC11 (BCCs, cleft palate, frontal bossing, hypertelorism, mental retardation, melanoma; no 'family history' entry in Table 1).
• Full-text review of PMID:20485063 found p.P725S reported only as a secondary variant 'of uncertain functional meaning' in a patient whose Gorlin-Goltz syndrome is caused by the pathogenic truncating PTCH1 c.2726dupA; no de novo assumption is made.
• gnomAD v2.1 (252/282,872 alleles, AF 0.000890862, 0 homozygotes) and gnomAD v4.1 (1,420/1,614,158 alleles, AF 0.000879716, 7 homozygotes; Middle Eastern AF 1.45%, 3 homozygotes) both document multiple unrelated carriers, which is incompatible with treating this change as an assumed de novo event.

Still missing

• No source states or assumes de novo occurrence of p.P725S, and there is no parental information from which that assumption could be drawn.

Sources

Reasoning

PP2 requires that missense variation is a common mechanism of disease in this gene. PTCH1-associated disease (nevoid basal cell carcinoma syndrome / Gorlin syndrome) is caused predominantly by truncating variants: in the 22-patient NBCCS cohort of PMID:28733979, Sanger sequencing identified 17 novel truncating mutations (11 frameshift, 5 nonsense, one splice-site) plus 2 novel missense mutations, and truncating mutations were found in 87% of the patients; the paper states that 'The majority of the described mutations result in premature protein truncation', and the gene-level loss-of-function mechanism for germline PTCH1 disease is separately supported. PMID:22703879, screening PTCH1 in 572 population exomes, found 20 PTCH1 variants - 4 class 1 and 16 class 3 (uncertain significance) and none class 4 or 5 - further indicating that missense changes in this gene are not a well-established common disease mechanism. Because missense change is the minority mechanism (2/19 novel mutations in the largest assembled cohort), the PP2 precondition is not satisfied.

Evidence considered

• PMID:28733979 direct quote: 'The majority of the described mutations result in premature protein truncation.'
• PMID:28733979 cohort composition: 'Sanger sequencing of PTCH1 gene in this cohort identified 17 novel truncating mutations (11 frameshift, 5 nonsense and one mutation affecting an exon-intron splicing site) and 2 novel missense mutations that were predicted to be pathogenic' and 'truncating mutations in 87% of the patients, mainly frameshift alteration, in agreement with literature, and missense variations in the remaining individuals'.
• PMID:22703879: PTCH1 was one of 37 cancer-susceptibility genes screened in 572 ClinSeq exomes; 20 PTCH1 variants were identified (4 class 1, 16 class 3) with no class 4 or class 5 variant, i.e. missense changes in PTCH1 are predominantly of uncertain significance rather than established disease-causing.
• Gene-level germline disease mechanism context for PTCH1 (loss-of-function/truncating disease mechanism supported, with no official CSPEC/VCEP available) as recorded in the case gene-context artifact.

Still missing

• No VCEP/gene-specific PTCH1 specification was available, so generic ACMG/AMP 2015 PP2 was applied; no quantitative missense-benign-rate calibration exists for PTCH1.

Sources

Reasoning

NM_000264.5:c.2173C>T (NP_000255.2:p.(Pro725Ser)) is a missense substitution, so PP3/BP4 are evaluated exclusively through the REVEL missense path; the SpliceAI splice-impact path is out of scope for this variant class and is not inspected for PP3. No PTCH1-specific VCEP/CSPEC specification was retrieved (final_classification_framework found=false), so the generic ACMG/AMP 2015 framework applies with the published ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997): PP3 supporting >=0.644, moderate >=0.773, strong >=0.932. The local REVEL v1.3 lookup (chr9:95468828 G>A) returns 0.299, well below the 0.644 supporting cutoff and inside the 0.29-0.644 gray zone. PP3 is therefore not met. A BayesDel score (-0.283106) is also present but has no verified published PP3/BP4 calibration and was treated as not available.

Evidence considered

• Variant consequence from the normalized transcript: NM_000264.5:c.2173C>T produces NP_000255.2:p.(Pro725Ser), a missense substitution, so PP3/BP4 use the REVEL missense path only; the SpliceAI path is out of scope for a missense variant.
• REVEL v1.3 score 0.299 for chr9:95468828 G>A (source_registry key 'revel'); PP3 requires >=0.644 (supporting), >=0.773 (moderate) or >=0.932 (strong) per the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997).
• No PTCH1 VCEP/CSPEC specification is available for this case; the generic ACMG/AMP 2015 fallback framework applies (framework_complete=false).
• BayesDel score -0.283106 is present in the case evidence but has no verified published PP3/BP4 calibration, so it was treated as not_available for this criterion.

Sources

Reasoning

PP5 applies only when a reputable, independent source reports the exact variant as pathogenic, and per this group's rule it is triggered only by an exact-variant ClinVar expert-panel (ClinGen VCEP/EP) classification of Pathogenic or Likely pathogenic. The ClinVar record for the exact variant NM_000264.5(PTCH1):c.2173C>T (p.Pro725Ser) (VCV000041654 / variation 41654) has an empty expert_panels field and zero expert-panel submissions among the 20 audited submissions; its review status is 'criteria provided, multiple submitters, no conflicts' (2 stars) with an aggregate classification of Benign/Likely benign. All submitters are ordinary clinical laboratories (Ambry Genetics, GeneDx, Illumina, Quest, Sema4, LabCorp, CeGaT, TU Dresden, LUMC, Amsterdam UMC, Rigshospitalet, Dasa, ITMI, NIH Biesecker Lab), none of which is a ClinGen expert panel. There is no ClinVar expert-panel Pathogenic/LP assertion for this exact variant, and the aggregate ClinVar label and individual laboratory submissions must not trigger PP5. A guideline source (PMID:15604628) was consulted and contains no variant-level assertion for this variant. PP5 is not met.

Evidence considered

• ClinVar VCV000041654 (variation 41654) for the exact variant c.2173C>T (p.Pro725Ser): expert_panels empty, 0 expert-panel submissions of 20 audited, review status 2-star ('criteria provided, multiple submitters, no conflicts'), aggregate classification Benign/Likely benign.
• All 20 audited ClinVar submissions are from ordinary clinical laboratories; none carries an expert-panel flag, so no PP5 trigger exists.
• PMID:15604628 (NSGC cancer genetic risk assessment guideline) was triaged as a PP5/PS4 candidate; it only lists PTCH at gene level for basal cell nevus syndrome and makes no variant-level assertion, so it cannot support PP5.
• PMID:25741868 defines PP5 (reputable source reports the variant as pathogenic) and supplies the general framework; it contains no data on this variant.

Sources

Reasoning

BA1 is stand-alone benign evidence requiring a population allele frequency at or above 5% (ACMG/AMP 2015, PMID:25741868). No population source comes close. gnomAD v4.1 (GRCh38, all-comers) reports the variant at a total AF of 0.000880 (1,420/1,614,158 alleles, 7 homozygotes) and gnomAD v2.1 (GRCh37, all-comers) at 0.000891 (252/282,872 alleles, 0 homozygotes). The highest frequency observed in any ancestry is 1.4517% in Middle Eastern individuals (88/6,062 alleles, grpmax FAF 1.2068%); every other ancestry is at or below 0.249% (v2.1 Remaining individuals 18/7,224; v4.1 Remaining 112/62,512). The non-cancer subsets, checked as a sensitivity analysis, are equally low (gnomAD v2.1 non-cancer exomes 0.0886%, 210/236,946 alleles; gnomAD v3.1 non-cancer genomes 0.1028%, 152/147,906 alleles), and gnomAD-Canada/HostSeq reports either 0.147% (27/18,418 alleles) or absence depending on which artifact is read. All observed frequencies sit roughly two orders of magnitude below the 5% stand-alone threshold, so BA1 is not met. The framework in force is the generic ACMG/AMP fallback (no PTCH1 CSPEC or VCEP exists), so the generic 5% threshold applies and the all-comers gnomAD datasets are the default sources.

Rule

BA1: stand-alone benign, population AF >= 0.05 (ACMG/AMP 2015, PMID:25741868); evaluated on all-comers gnomAD v2.1/v4.1 total and ancestry-maximum frequencies.

Evidence considered

• gnomAD v4.1 (GRCh38) all-comers total: AF 0.000880 (1,420/1,614,158 alleles; 7 homozygotes); grpmax FAF 0.012068; highest ancestry Middle Eastern 1.4517% (88/6,062 alleles, 3 homozygotes).
• gnomAD v2.1 (GRCh37) all-comers total: AF 0.000891 (252/282,872 alleles; 0 homozygotes); grpmax FAF 0.001203; popmax 0.2492% (Remaining individuals, 18/7,224 alleles).
• Non-cancer subsets checked as sensitivity only (no VCEP mandates them here): gnomAD v2.1 non-cancer exomes AF 0.0886% (210/236,946); gnomAD v3.1 non-cancer genomes AF 0.1028% (152/147,906).
• gnomAD-Canada v1.0 (HostSeq genomes) data conflict between artifacts: 27/18,418 alleles (AF 0.1466%) in case_summary.json vs 'absent' in evidence_brief.json; immaterial because either value is far below 5%.
• Threshold used verbatim: BA1 stand-alone benign at AF >= 0.05, ACMG/AMP 2015 (PMID:25741868).
• Framework check: no ClinGen CSPEC and no gene-specific VCEP for PTCH1, so generic ACMG/AMP 2015 defaults and the all-comers gnomAD sources are the governing defaults.

Sources

Reasoning

BP2 requires one of two observations: the variant in trans with a pathogenic variant for a fully penetrant dominant gene/disorder, or the variant in cis with a pathogenic variant in any inheritance pattern (ACMG/AMP 2015, PMID:25741868). The single allelic observation relevant to this variant is PMID:20485063, a clinically confirmed Gorlin-Goltz patient heterozygous for the pathogenic PTCH1 frameshift c.2726dupA (exon 17) who additionally carried p.(Pro725Ser) in exon 14, described by the authors as 'of uncertain functional meaning'. The report never states whether the two variants lie on the same (cis) or opposite (trans) alleles, and no parental or family phasing was performed, so neither arm of BP2 can be credited. In the Spanish NBCCS cohort (PMID:28733979) the p.P725S carrier (index case IC11) had no second PTCH1 variant identified, so there was no pathogenic allele available to test trans phase; affected relatives were genotyped only for the truncating mutations of IC1, IC3, IC14 and IC15, not for p.P725S. No ClinVar submission for VCV000041654 reports phase, zygosity or a co-occurring pathogenic PTCH1 allele. Because the phase-dependent observation that BP2 requires has not been made, the criterion is not met; the phase of p.(Pro725Ser) relative to c.2726dupA in the PMID:20485063 patient remains the one resolvable observation that could convert this to BP2 supporting, which is why human review is flagged. Penetrance language in the sources ('complete penetrance', PMID:20485063; 'high penetrance', PMID:28733979) would also have to be confirmed before relying on BP2's fully penetrant-dominant clause.

Rule

BP2 (supporting, benign): observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder, or observed in cis with a pathogenic variant in any inheritance pattern.

Evidence considered

• PMID:20485063 (Clinical Report, genetic evaluation paragraph): a 52-year-old man with Gorlin-Goltz syndrome, family history negative, was found to carry the pathogenic PTCH1 c.2726dupA in exon 17 plus 'the variant p.P725S in the 14th exon, which was of uncertain functional meaning'. No cis/trans phase, zygosity or segregation statement is made for p.P725S, so the co-occurrence cannot be assigned to either BP2 arm.
• PMID:28733979 (Table 1 / Results): in 22 unrelated Spanish NBCCS index cases, p.P725S (rs149258400, exon 14/ICL3) was the sole PTCH1 variant in index case IC11 (65-year-old male: BCCs, cleft palate, frontal bossing, hypertelorism, mental retardation, melanoma), i.e. no second pathogenic PTCH1 allele existed in that carrier to be in trans. The variant was absent from 400 control alleles and eight in-silico tools predicted it as a neutral change.
• PMID:28733979 also notes the p.P725S carrier did not have the medulloblastoma previously associated with this variant in tumours, and the authors state they could not establish a genotype-phenotype correlation; this context was reviewed but contributes no phase information.
• ClinVar VCV000041654 (variation 41654): all 19 submissions with an origin field record germline origin, but none provides zygosity, cis/trans phase, or a co-occurring pathogenic PTCH1 allele; the classification was not used as evidence for this criterion.
• Gorlin syndrome penetrance wording relevant to BP2's trans-in-dominant arm: PMID:20485063 describes inheritance as autosomal dominant with 'complete penetrance and variable expressivity'.
• Searched every fetched full text and the ClinVar submission list for a cis or trans configuration of p.(Pro725Ser) with a pathogenic PTCH1 variant, using the terms 'in cis', 'in trans', 'cis', 'trans', 'allele', 'heterozygous', 'phase', 'phasing', 'segregation', 'paternal', 'maternal'; the only hit was the unphased c.2726dupA co-occurrence above.
• No ClinGen CSPEC/VCEP or local PTCH1 framework exists in this case (vcep_materials.json: 'no cspec url available'; final_classification_framework.json: found=false, source='generic_acmg_fallback'), so the generic ACMG/AMP 2015 BP2 definition governs.

Still missing

• Cis/trans phase of p.(Pro725Ser) relative to the pathogenic PTCH1 frameshift c.2726dupA was not determined in the Gorlin-Goltz patient reported in PMID:20485063 (no allele-specific testing, no parental phasing, family history negative); resolving it would settle whether BP2 applies via the cis arm.
• No segregation or phasing data exist for p.(Pro725Ser) in any affected pedigree; the relatives genotyped in PMID:28733979 carried truncating PTCH1 variants (IC1, IC3, IC14, IC15), not p.P725S.
• PMID:29575684 (Reinders 2018, updated PTCH1 mutation database) is cited by ClinVar submitters for this variant's record but was not retrieved into the case workspace, so any allele-level or co-occurrence context it might contain could not be checked.

Sources

Reasoning

The variant is missense, so BP4 is likewise evaluated only via the REVEL path; the SpliceAI result (max delta 0.003, i.e. no predicted splice impact) describes splice effects and cannot serve as BP4 evidence for a missense substitution. Under the ClinGen SVI REVEL calibration (PMID:36413997), BP4 requires REVEL <=0.29 (supporting), <=0.183 (moderate) or <=0.016 (strong). The REVEL v1.3 score for this variant is 0.299, which sits 0.009 above the <=0.29 supporting cutoff, placing it in the gray zone where BP4 is not met (and PP3 is also not met). An independent published in silico analysis of the exact variant (PMID:28733979; eight bioinformatics tools including PolyPhen-2, SIFT and MutationTaster) predicted p.P725S as a possible neutral change, but this uncalibrated multi-tool result does not satisfy a calibrated BP4 threshold and does not change the determination. BP4 is not met.

Evidence considered

• Missense consequence (NP_000255.2:p.(Pro725Ser)) means BP4 is assessed on the REVEL path only.
• REVEL v1.3 score 0.299 for chr9:95468828 G>A (source_registry key 'revel'); BP4 requires <=0.29 (supporting), <=0.183 (moderate) or <=0.016 (strong) per the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID:36413997).
• SpliceAI max delta 0.003 reflects a splice prediction and is not usable as BP4 support for a missense variant.
• PMID:28733979 (full text read) reports an in silico bioinformatics analysis predicting p.P725S could represent a neutral change; this uncalibrated eight-tool result was considered but does not meet a calibrated BP4 threshold.

Sources

Reasoning

BP6 applies only when a reputable, independent source reports the exact variant as benign, and per this group's rule it is triggered only by an exact-variant ClinVar expert-panel (ClinGen VCEP/EP) classification of Benign or Likely benign. The ClinVar record for the exact variant NM_000264.5(PTCH1):c.2173C>T (p.Pro725Ser) (VCV000041654 / variation 41654) has an empty expert_panels field and no expert-panel submissions among the 20 audited submissions; the aggregate classification is Benign/Likely benign at 2 stars ('criteria provided, multiple submitters, no conflicts'). Although 15 ordinary clinical laboratories submitted Benign or Likely benign interpretations, none is a ClinGen expert panel, and per the governing rule ordinary laboratory submissions and aggregate ClinVar labels must never trigger BP6. There is consequently no BP6-eligible assertion for this exact variant, and BP6 is not met (despite the benign ClinVar consensus, which is not a permitted BP6 trigger).

Evidence considered

• ClinVar VCV000041654 (variation 41654) for the exact variant c.2173C>T (p.Pro725Ser): expert_panels empty, 0 expert-panel submissions of 20, review status 2-star, aggregate Benign/Likely benign.
• 15 clinical laboratories submitted Benign/Likely benign interpretations (10 Likely benign, 5 Benign, 1 benign), but none is a ClinGen expert panel; ordinary submissions and aggregate labels do not trigger BP6.
• PMID:25741868 defines BP6 (reputable source reports the variant as benign) and the general framework; it contains no data on this variant.

Sources

Reasoning

No PTCH1-specific VCEP or ClinGen CSPEC specification was identified for this case (cspec.found=false; vcep_materials.json reports 'no cspec url available'), so PS3 is governed by the generic ACMG/AMP 2015 definition (PMID:25741868): a well-established functional assay, performed in a disease-relevant system with validated wild-type, positive and negative controls and a quantitative readout, demonstrating a damaging effect on PTCH1 function. NM_000264.5:c.2173C>T is a missense substitution, NP_000255.2:p.(Pro725Ser), in exon 14 (intracellular loop 3) of PTCH1, the Hedgehog-pathway receptor whose germline loss of function causes Gorlin (naevoid basal cell carcinoma) syndrome. No such functional assay exists for this variant. PMID:28733979 reports the exact p.P725S variant (rs149258400) in a 65-year-old male NBCCS index patient but explicitly performed no cell- or protein-based assay; it reports only in silico prediction (eight algorithms favouring a neutral change) and absence from 400 control alleles. PMID:20485063 reports the same variant in exon 14 of a Gorlin-Goltz patient and describes it as being 'of uncertain functional meaning'. OncoKB reports an 'Unknown Oncogenic Effect' and states that no variant-specific reviewed functional evidence was identified for PTCH1 P725S. PMID:26467025 and PMID:28492532 are methodology documents (weighting of published functional studies within variant-scoring frameworks) and provide no PTCH1- or p.P725S-specific assay data. Computational scores (REVEL 0.299; BayesDel -0.283; SpliceAI max delta 0.003) are not functional-assay evidence and cannot satisfy PS3. Because the required input — a validated functional assay of this exact variant — is absent rather than negative, PS3 cannot be assessed and no PS3 strength is assigned.

Rule

Generic ACMG/AMP 2015 PS3 (PMID:25741868): requires well-established functional studies showing a damaging effect; no PTCH1-specific VCEP/CSPEC specification is available for this case.

Evidence considered

• No PTCH1-specific VCEP or ClinGen CSPEC specification was identified (cspec.found=false; vcep_materials.json reports 'no cspec url available'), so the generic ACMG/AMP 2015 PS3 definition (PMID:25741868) was applied.
• PMID:28733979 reports the exact PTCH1 p.P725S variant (rs149258400; exon 14, intracellular loop 3) in a 65-year-old male NBCCS index patient (IC11), notes it was previously detected in medulloblastoma tumours and listed in COSMIC and in the 1000 Genomes database with unclear clinical significance, and reports that eight in silico algorithms predicted a neutral change; the paper states no functional assay was performed and calls for further studies.
• PMID:20485063 identified the same p.P725S variant in exon 14 of PTCH1 in a Gorlin-Goltz syndrome patient, where the pathogenic syndrome-causing change was a separate c.2726dupA, and described p.P725S as being 'of uncertain functional meaning' with no functional assay reported.
• OncoKB returned 'Unknown Oncogenic Effect' for PTCH1 P725S and stated that no variant-specific reviewed functional evidence was identified.
• PMID:26467025 (rules-based scoring system covering 11,894 variants in 338 genes) and PMID:28492532 (Sherloc refinement of the ACMG/AMP criteria) are methodology documents describing how published functional study results are weighted; neither contains PTCH1- or p.P725S-specific functional data.
• ClinVar page citation PMID:28627087 reports the exact germline PTCH1 p.Pro725Ser variant (as p.Pro725Ser, alongside FGFR3 p.Pro449Ser) in a family with familial seborrhoeic keratosis, but the full text was not fetched for this case; the available abstract describes clinical, dermoscopic and histopathological characterisation plus germline FGFR3/PTCH1 variant screening, not a functional assay.
• Computational predictor outputs for this missense variant (REVEL 0.299; BayesDel -0.283) and the SpliceAI max delta of 0.003 are in silico evidence, not functional-assay evidence, and cannot be used to meet PS3.
• A transcript-level PTCH1 assay (minigene, demonstrating exon 20 skipping) has been reported for a different PTCH1 variant, the canonical splice-donor NM_000264.5:c.3449+1G>A (PMID:41816662), confirming such assays are feasible for this gene but have not been applied to p.(Pro725Ser).

Still missing

• No validated disease-relevant functional assay for PTCH1 p.(Pro725Ser) (e.g. a Hedgehog-pathway reporter or SMO-repression assay) with wild-type, positive and negative controls and a quantitative readout has been reported.
• The full text of the ClinVar page citation PMID:28627087, which reports the exact germline p.Pro725Ser variant in a familial seborrhoeic keratosis family, was not retrieved for this case; only a truncated abstract was available, so any functional data it contains remains unchecked.
• PMID:28492532 (Sherloc) is available only as an abstract in this case, so its PS3/BS3 weighting guidance could not be read in full.

Sources

Reasoning

PP1 (Richards et al. 2015, PMID:25741868) requires co-segregation of the variant with disease in multiple affected family members. No family genotypes exist for PTCH1 c.2173C>T (p.Pro725S) in this evidence base, so zero informative meioses are available. The only segregation analysis in the case evidence comes from PMID:28733979, where relatives diagnosed with Gorlin syndrome were genotyped in families IC1, IC3, IC14 and IC15 - all of which segregate truncating PTCH1 mutations, not the missense p.P725S; the index patient who carries p.P725S (IC11) had no relatives tested, and the five relatives studied belonged to other index cases. PMID:20485063 reports a single sporadic patient with no family study, and no other source discusses family data for this variant. Because the criterion can only be satisfied by a co-segregation finding that has never been generated, PP1 cannot be evaluated and is recorded as not assessed rather than met or refuted.

Rule

ACMG/AMP 2015 PP1: co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease (PMID:25741868)

Evidence considered

• PP1 definition (co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease) taken from the ACMG/AMP 2015 standards (Richards et al., PMID:25741868); no PTCH1 VCEP specification exists, so generic ACMG applies.
• PMID:28733979 reports that in five cases (IC1, IC3, IC14 and IC15) relatives diagnosed with Gorlin syndrome were genotyped and carried the same mutation as the index case, 'showing that these variations segregate with the disease (Suppl. Table S2)'; these families carry truncating PTCH1 mutations, and the statement does not extend to the p.P725S-carrying index case IC11.
• PMID:28733979 Table 1: IC11 (Male, 65; BCCs, cleft palate, frontal bossing, hypertelorism, mental retardation, melanoma) carries rs149258400C>T / p.P725S in exon 14 (ICL3); no family history is listed for this case, consistent with an uninformative pedigree.
• PMID:20485063 reports one 52-year-old male with Gorlin-Goltz syndrome and therefore contributes a single affected carrier, not a segregation pedigree.
• grep of every constituent case file for 'segregat', 'de novo', 'proband', 'parental' and 'meios' returned no further family-level evidence for this variant.

Still missing

• No relatives of any p.P725S carrier have been genotyped; co-segregation of this variant with Gorlin syndrome/NBCCS has never been tested (0 informative meioses).

Sources

Reasoning

PP4 requires that the patient's phenotype or family history be highly specific for a disease with a single genetic etiology. The case evidence bundle contains no proband clinical phenotype, indication for testing, or family history for the individual in whom this variant was found; the compact evidence carries only variant-level annotations and the ClinVar condition labels (Gorlin syndrome, Hereditary cancer-predisposing syndrome, Holoprosencephaly 7, PTCH1-related disorder, 'not provided', 'not specified'), which are non-specific submitter assertions rather than an evaluated proband phenotype. Because the phenotype input that PP4 is defined against is absent, the criterion cannot be evaluated and is recorded as not_assessed rather than not_met.

Evidence considered

• The consolidated case evidence contains no proband phenotype or family-history field; only variant-level source annotations and non-specific ClinVar condition labels are available.
• PMID:28733979 does describe clinical features in an NBCCS patient carrying p.P725S (BCCs, cleft palate, frontal bossing, hypertelorism, mental retardation, melanoma), but this is a published case observation and not the phenotype of the individual being adjudicated in this case, so it cannot ground PP4 here.

Still missing

• No proband clinical phenotype, testing indication, or family history is provided in the case evidence; PP4 cannot be evaluated without it.

Sources

Reasoning

No PTCH1-specific VCEP or ClinGen CSPEC specification was identified for this case (cspec.found=false; vcep_materials.json reports 'no cspec url available'), so BS3 is governed by the generic ACMG/AMP 2015 definition (PMID:25741868): a well-established functional assay using appropriate controls that demonstrates no damaging effect on protein function. NM_000264.5:c.2173C>T (NP_000255.2:p.(Pro725Ser), exon 14/intracellular loop 3 of PTCH1) has not been characterised by any such assay. PMID:28733979, the only study reporting the exact variant in a patient cohort, performed no functional assay; its evidence is limited to in silico prediction (eight algorithms favouring a neutral change) and absence of the allele from 400 healthy-control alleles, neither of which is a functional assay demonstrating normal protein function. PMID:20485063 reports the variant in a Gorlin-Goltz patient and calls it 'of uncertain functional meaning'. OncoKB identified no variant-specific reviewed functional evidence. PMID:26467025 and PMID:28492532 are methodology documents with no variant-specific assay data. In addition, a benign functional read-out is not biologically supported here: the variant has been observed in affected NBCCS individuals (PMID:28733979; PMID:20485063), so there is no assay-based demonstration of preserved PTCH1 function. Because the required input — a validated normal-function assay of this exact variant — is absent rather than negative, BS3 cannot be assessed and no BS3 strength is assigned.

Rule

Generic ACMG/AMP 2015 BS3 (PMID:25741868): requires well-established functional studies showing no damaging effect; no PTCH1-specific VCEP/CSPEC specification is available for this case.

Evidence considered

• No PTCH1-specific VCEP or ClinGen CSPEC specification was identified (cspec.found=false; vcep_materials.json reports 'no cspec url available'), so the generic ACMG/AMP 2015 BS3 definition (PMID:25741868) was applied.
• PMID:28733979 reports the exact PTCH1 p.P725S variant in an NBCCS index patient and provides only in silico neutral predictions (eight tools) and absence from 400 control alleles; no assay measuring normal PTCH1 function or protein stability was performed.
• PMID:20485063 reports the same p.P725S variant as an incidental finding of 'uncertain functional meaning' in a Gorlin-Goltz patient, with no functional assay.
• OncoKB reported 'Unknown Oncogenic Effect' and stated that no variant-specific reviewed functional evidence was identified for PTCH1 P725S.
• PMID:26467025 and PMID:28492532 are methodology/weighting documents and contain no PTCH1- or p.P725S-specific normal-function assay data.
• The variant has been observed in individuals affected with NBCCS (PMID:28733979; PMID:20485063), so there is no functional-assay basis for a benign (no damaging effect) interpretation.
• The SpliceAI max delta of 0.003 and the REVEL 0.299 / BayesDel -0.283 outputs are computational predictions; for this missense variant they are PP3/BP4-path evidence and are not functional-assay evidence that can meet BS3.

Still missing

• No validated disease-relevant assay demonstrating normal PTCH1 function for p.(Pro725Ser) with appropriate controls and a quantitative readout is available.
• The full text of the ClinVar page citation PMID:28627087, which reports the exact germline p.Pro725Ser variant in a familial seborrhoeic keratosis family, was not retrieved for this case; only a truncated abstract was available, so any functional data it contains remains unchecked.
• PMID:28492532 (Sherloc) is available only as an abstract in this case, so its BS3 weighting guidance could not be read in full.

Sources

Reasoning

BS4 (Richards et al. 2015, PMID:25741868) requires a documented lack of segregation - an informative family in which the variant is absent from affected members or present in unaffected members. No family has been genotyped for PTCH1 c.2173C>T (p.Pro725S): the segregation data in PMID:28733979 concern other, truncating variants, and PMID:20485063 contributes only a single patient without relatives. Absence of segregation data is not evidence of non-segregation, so BS4 cannot be met and is recorded as not assessed. Related non-family context exists but does not satisfy BS4 and belongs to the population-frequency criteria: the variant is common in unaffected reference cohorts (gnomAD v2.1 AF 0.00089, 252/282,872 alleles, 0 homozygotes; gnomAD v4.1 AF 0.00088, 1,420/1,614,158 alleles with 7 homozygotes; gnomAD v2.1 non-cancer 210/236,946 alleles), and in PMID:20485063 it co-occurs with a pathogenic truncating variant that explains the phenotype.

Rule

ACMG/AMP 2015 BS4: lack of segregation in affected members of a family (PMID:25741868)

Evidence considered

• BS4 definition (lack of segregation in affected members of a family) taken from the ACMG/AMP 2015 standards (Richards et al., PMID:25741868); no PTCH1 VCEP specification exists, so generic ACMG applies.
• PMID:28733979 supplies the case's only genotype-in-relatives analysis, and it is restricted to families segregating truncating mutations (IC1, IC3, IC14, IC15; Suppl. Table S2) rather than p.P725S, so it yields neither segregation nor non-segregation evidence for the variant under review.
• PMID:20485063 provides a single male Gorlin-Goltz patient in whom p.P725S is a secondary finding to the pathogenic c.2726dupA; no obligate carriers, unaffected or affected relatives are described.
• Population-based context recorded for completeness: gnomAD v2.1 252/282,872 alleles (AF 0.000890862, 0 homozygotes), gnomAD v4.1 1,420/1,614,158 alleles (AF 0.000879716, 7 homozygotes), gnomAD v2.1 non-cancer exomes 210/236,946 alleles - these are frequency observations, not segregation observations, and do not constitute BS4.

Still missing

• No family-based segregation or non-segregation study exists for p.P725S; BS4 requires an informative pedigree, which is absent from the evidence base.

Sources