STK11 c.160C>T, p.(Leu54=)

Reasoning

PM2 (supporting) applies when the variant is absent or extremely rare in population databases, using the SVI-recommended supporting threshold of <= 0.0001 (0.01%). The variant is far below that ceiling: gnomAD v4.1 total AF 4.34e-06 (7/1,613,928 alleles), gnomAD v2.1 total AF 8.05e-06 (2/248,308 alleles), and grpmax filtering AF 4.42e-06 (v4.1). The highest ancestry-specific frequency is 2.67e-05 in African/African American individuals (gnomAD v4.1) and 1.78e-05 in European (non-Finnish) individuals (gnomAD v2.1) -- both still more than three-fold below the 0.0001 threshold. Supporting evidence also comes from the non-cancer subsets: gnomAD v2.1 non-cancer exomes AF 8.53e-06 (2/234,470 alleles) and complete absence from gnomAD v3.1 non-cancer genomes. No homozygotes were observed in any dataset. PM2 is therefore met at supporting strength. This is a rare population observation only and must not be read as evidence of pathogenicity in isolation; it establishes rarity, which is all PM2 asserts.

Rule

PM2 supporting: allele frequency <= 0.0001 in population databases (SVI 2020).

Evidence considered

• gnomAD v4.1 (all-comers joint): total AF 4.33724e-06 (7/1,613,928 alleles), 0 homozygotes; highest ancestry group African/African American AF 2.66574e-05 (2/75,026 alleles); grpmax FAF 4.42e-06.
• gnomAD v2.1 (all-comers exomes): total AF 8.05451e-06 (2/248,308 alleles), 0 homozygotes; highest ancestry group European (non-Finnish) AF 1.7804e-05 (2/112,334 alleles); grpmax FAF 2.96e-06.
• PM2 supporting rarity threshold <= 0.0001 (0.01%), SVI recommendation to apply PM2 at supporting strength (PMID:25741868).
• gnomAD v2.1 non-cancer exomes AF 8.52988e-06 (2/234,470 alleles, 0 homozygotes) and absent from gnomAD v3.1 non-cancer genomes, confirming rarity in non-cancer cohorts.
• No VCEP/CSPEC exists for STK11, so the all-comers gnomAD v2.1/v4.1 totals were used as the default population sources rather than a specified non-cancer or exome-only source.

Still missing

• gnomAD-Canada v1.0 returned no result (query timed out), so that additional population source could not be checked.
• PM2 is applied at supporting strength per SVI guidance; no STK11-specific VCEP rarity threshold exists to alter this.

Sources

Reasoning

BP7 (Richards et al. 2015, PMID:25741868) applies to a synonymous (silent) variant for which splicing-prediction algorithms show no impact on the splice consensus sequence and no creation of a new splice site. NM_000455.5:c.160C>T in STK11 is exactly such a variant: it is synonymous, predicted to encode NP_000446.1:p.(Leu54=) with the leucine at residue 54 unchanged, so there is no amino-acid effect to assess. SpliceAI was queried for this transcript and returned DS_AG 0.001, DS_AL 0.003, DS_DG 0.002 and DS_DL 0.000, giving a maximum delta score of 0.003, with the affected sites 237 nt upstream (DP_AG), 244 nt downstream (DP_AL), 21 nt upstream (DP_DG) and 134 nt downstream (DP_DL) and SpliceAI's own consequence call of 'sequence_variant'. The 0.003 maximum delta is far below the SpliceAI no-impact cutoff of 0.1 (Jaganathan et al. 2019, PMID:30661751), and the independent Pangolin scores likewise show no splice gain or loss (SG 0.006, SL -0.004). Thus no disruption of the splice consensus sequence and no creation of a cryptic splice site is predicted, which satisfies the splicing element of BP7. Protein-level in-silico tools are not relevant here: REVEL was not found (score null) and BayesDel was not found (score null); they were not used, and no splice prediction was used as support for any missense-type criterion. BP7 is applied at supporting strength, the generic ACMG/AMP 2015 default, because no gene-specific ClinGen/VCEP specification exists for STK11 (final_classification_framework reports generic ACMG fallback, and no CSPEC is available for this gene), so there is no gene-specific BP7 weighting or upgrade rule to apply. One element of the original 2015 BP7 wording was not independently verifiable: nucleotide-level conservation at c.160 was not present in the case evidence, so the 'not highly conserved' element is recorded as a gap while the fully satisfied splice-impact element carries the criterion at supporting strength.

Evidence considered

• Variant is synonymous: NM_000455.5:c.160C>T, NP_000446.1:p.(Leu54=), leucine 54 unchanged (variant_validator / prefetch normalization).
• SpliceAI (spliceai) result available for NM_000455.5:c.160C>T (GRCh37 g.1207072C>T): DS_AG 0.001, DS_AL 0.003, DS_DG 0.002, DS_DL 0.000; maximum delta score 0.003; affected-site offsets DP_AG -237, DP_AL 244, DP_DG -21, DP_DL 134; variant_consequence 'sequence_variant'; spliceai_available = true.
• The 0.003 maximum delta is below the SpliceAI no-impact cutoff of 0.1 (Jaganathan et al. 2019, Cell, PMID:30661751), so no impact on the splice consensus sequence and no new splice site are predicted.
• Pangolin, run alongside SpliceAI, likewise shows no splice gain or loss (pangolin_SG 0.006, pangolin_SL -0.004), corroborating the absence of splice impact.
• BP7 definition source: Richards et al. 2015 ACMG/AMP standards and guidelines (PMID:25741868); the variant appears only as a synonymous change with no reported variant-specific data in the retrieved literature, so BP7 rests on the in-silico splice prediction.
• Protein-level predictors were not used: REVEL score null (found=false) and BayesDel score null (found=false); they are missense protein-impact tools and would not be applicable to a synonymous change.
• Strength: supporting, the generic ACMG/AMP 2015 default for BP7; no STK11-specific ClinGen/VCEP specification exists for this case (final_classification_framework reports generic ACMG fallback), so there is no gene-specific weighting to apply.

Still missing

• Nucleotide-level conservation (e.g. phyloP/GERP) at STK11 c.160 was not available in the case evidence, so the 'nucleotide is not highly conserved' element of the ACMG/AMP 2015 BP7 wording could not be independently verified; the splice-impact element is fully satisfied and BP7 is applied at supporting strength without upgrade.

Sources

Reasoning

PS3 (Richards et al. 2015, PMID:25741868) requires that a well-established functional assay demonstrate a deleterious effect of the variant on gene or protein function. NM_000455.5:c.160C>T in STK11 is a synonymous substitution, NP_000446.1:p.(Leu54=), so the only plausible mechanism by which it could be functionally deleterious is a direct RNA-level effect such as altered splicing or transcript stability, which would require a validated RNA or splicing assay of the variant itself. No such assay exists for this variant in the retrieved evidence: the literature set assembled for this case contains no functional study of c.160C>T (the papers retrieved cover a clinical ACG hereditary GI cancer guideline, the ACMG SF v2.0 secondary-findings policy statement, a CAPS pancreatic-surveillance consensus, a background review of the Sherloc criteria, and the ACMG/AMP 2015 guideline itself), and the ClinVar submission audit records five submissions, zero expert-panel submissions, and zero criterion leads that supply functional data. The only variant-level molecular result available is a computational SpliceAI maximum delta score of 0.00, and computational prediction is not a functional assay, so it cannot satisfy PS3. Because no well-established functional assay result exists, there is nothing to calibrate for a deleterious effect, and PS3 is not met. This assessment uses the generic ACMG/AMP 2015 fallback because no ClinGen CSPEC or local vcep-specific framework is available for STK11 in this case.

Rule

PS3: well-established functional studies show a deleterious effect

Evidence considered

• Criterion definition: PS3 requires a well-established functional study showing a deleterious effect on gene or protein function (Richards et al. 2015 ACMG/AMP, PMID:25741868).
• Variant consequence: NM_000455.5:c.160C>T in STK11 is synonymous, NP_000446.1:p.(Leu54=), so any functional effect would have to act at the RNA level (splicing/transcript processing) rather than through an amino-acid change.
• The ACMG/AMP 2015 guideline notes that a variant may act directly through the specific DNA change, such as through splicing disruption, rather than through the amino acid change, so a synonymous change must not be assumed functionally neutral without evidence - but that evidence is explicitly computational at minimum in the guideline text, and PS3 additionally requires experimental functional data.
• Literature pass for this case returned no functional-assay publication on this variant; the 'functional'-themed paper retrieved was the ACMG/AMP 2015 guideline itself, which contains no STK11 or variant-specific assay data.
• ClinVar submission audit for this variant records 5 submissions, 0 expert-panel submissions, 0 criterion-lead submissions and 0 validated functional PMIDs.
• SpliceAI lookup returned a maximum delta score of 0.00 for this variant; this is a computational in-silico prediction and is therefore not admissible as PS3 functional evidence.
• No ClinGen CSPEC or local VCEP framework for STK11 was found (vcep_materials: 'no cspec url available'); the generic ACMG/AMP 2015 fallback is used, per the case's final classification framework record (generic_acmg_fallback, precedence_rank 3).

Still missing

• No well-established functional assay (e.g. validated RNA/splicing assay, minigene, or transcript analysis) has been performed on STK11 c.160C>T, so PS3 cannot be evaluated positively; an assay of this specific variant would be required before PS3 could ever be met.

Sources

Reasoning

PS4 (Richards et al. 2015, PMID:25741868) requires the prevalence of the variant in affected individuals to be significantly increased compared with controls, or a case-control/enrichment statistic at or above a gene-specific threshold. No numeric enrichment statistic of any kind (odds ratio, relative risk, or likelihood ratio) exists for NM_000455.5:c.160C>T in any source assembled for this case, so no operator-direction comparison against a threshold could be or was performed. Because this group's numeric rules require an explicit operator read from the governing source, and no governing source supplies a value or cutoff here, PS4 is decided on the presence/absence of case data rather than on a numeric comparison. The variant is a synonymous change, NP_000446.1:p.(Leu54=), present in gnomAD v2.1 (2/248308 alleles, AF 8.05e-06) and v4.1 (7/1613928 alleles, AF 4.34e-06) with no homozygotes, and no publication consulted reports any affected individual carrying it. The three ClinVar candidate sources flagged for PS4 in the literature pass (PMID:26389258 PDQ Cancer Genetics Risk Assessment and Counseling, PMID:26389505 PDQ Genetics of Colorectal Cancer, PMID:27854360 ACMG SF v2.0) are policy/guideline documents: their extracted content and available text contain no case observations or case-control data for this variant, and STK11 appears at most as a gene-list entry. No affected proband with this variant was found in the consulted literature, so the requirement for increased prevalence in cases is not satisfied.

Rule

PS4: prevalence of the variant in affected individuals is significantly increased compared with the prevalence in controls; requires case-control/enrichment data with a defined threshold.

Evidence considered

• ClinVar VCV000485000 (NM_000455.5(STK11):c.160C>T, p.Leu54=): 5 submissions, all from clinical laboratories (Color Health, Ambry, Labcorp/Invitae, All of Us, Myriad), all Benign or Likely benign; none reports an affected case with the variant used as enrichment evidence, and no submission supplies a case-control statistic.
• No source in this case reports c.160C>T in any affected individual; the only observations of the variant are population-database allele counts (gnomAD v2.1 and v4.1, all without homozygotes).
• PMID:27854360 (ACMG SF v2.0): STK11/PJS is present only as a screening gene-list row (KP and EP variants to be reported); no variant-specific case or control data.
• PMID:26389258 and PMID:26389505 (PDQ guidelines): flagged in the literature pass as PS4/PP5 candidate sources, but no case-observation content for this variant is present and no full text was retrieved for either.
• Numeric-threshold protocol: no PS4 threshold or value exists in this case's evidence, so the operator-direction check (>= threshold) had no operand; conclusion rests on the complete absence of case-enrichment data.

Still missing

• No case-control or case-series study reporting STK11 c.160C>T in affected individuals is available in this case; a systematic case-enrichment analysis (or a gene-specific PS4 threshold with a named publication) would be required to raise PS4 above not_met.

Sources

Reasoning

PP4 (Richards et al. 2015, PMID:25741868) requires the patient's phenotype or family history to be highly specific for a disease with a single genetic etiology. This case carries no proband phenotype or family history: the record is a variant-level query (NM_000455.5:c.160C>T, STK11, p.(Leu54=), a synonymous substitution) with no reported clinical presentation, and no consulted source places this variant in an affected individual, let alone in one with the Peutz-Jeghers syndrome phenotype that would be specific to STK11. The STK11-related sources available (PMID:27854360 ACMG SF v2.0 listing STK11/PJS as a secondary-findings gene; PMID:31672839 CAPS consensus recommending pancreatic surveillance for STK11 carriers; PMID:25645574 ACG hereditary GI cancer guideline noting PJS among the syndromes it covers) discuss STK11 and Peutz-Jeghers syndrome only at the gene/syndrome level and explicitly do not mention c.160C>T or p.(Leu54=); a gene-level phenotype association is not phenotype specificity attributable to this variant. Consequently the two elements PP4 requires - a patient's specific phenotype and a variant-matched genotype - are both unavailable, so PP4 is not met rather than merely under-powered.

Rule

PP4: patient's phenotype or family history is highly specific for a disease with a single genetic etiology.

Evidence considered

• No phenotype or family-history field is present for this variant query; the only clinical context in ClinVar VCV000485000 is submitter condition labels ('Hereditary cancer-predisposing syndrome', 'Peutz-Jeghers syndrome'), which are gene-level panel labels applied by clinical laboratories and carry no patient-level phenotype data.
• PMID:31672839 (CAPS Consortium): STK11 carriers are recommended for pancreatic surveillance irrespective of family history because of high lifetime risk; this is a gene-level surveillance statement and the paper does not mention c.160C>T, so it argues no phenotype specificity for this specific allele.
• PMID:25645574 (ACG hereditary GI cancer guideline): PJS is listed among the syndromes addressed but STK11 and c.160C>T are not named in the retrieved text; no variant-level phenotype data.
• PMID:27854360 (ACMG SF v2.0): STK11 listed for Peutz-Jeghers syndrome (MIM 175200) as an actionable secondary-findings gene; gene-level only, no patient phenotype for this variant.
• PMID:25741868: consulted for the PP4 definition (phenotype/family history highly specific for a disease with a single genetic etiology) and the general requirement that criteria be met by variant-relevant, not gene-level, evidence.

Still missing

• No proband phenotype, family history, or affected-case report for c.160C>T is present in this case; PP4 cannot be satisfied without a patient whose phenotype is specific for STK11-related Peutz-Jeghers syndrome.

Sources

Reasoning

PP5 (Richards et al. 2015, PMID:25741868) requires a reputable source to report the variant as pathogenic; under the framework applied here that means an exact-variant ClinVar assertion from a recognised expert panel (ClinGen/VCEP) classified Pathogenic or Likely pathogenic. The exact-variant ClinVar record for NM_000455.5(STK11):c.160C>T (VCV000485000 / variation 485000) has 5 germline submissions, every one of them from a clinical laboratory (Color Health, Ambry Genetics, Labcorp/Invitae, All of Us Research Program, Myriad Genetics); the submission audit records expert_panel_submissions = 0 and is_expert_panel = false for all five. There is no ClinGen STK11 VCEP specification available for this case (vcep_materials: no cspec url, framework_complete = false, no gene-specific criteria returned), so no VCEP assertion exists to cite either. Furthermore, even setting the expert-panel requirement aside, the aggregate ClinVar classification is Benign/Likely benign - the opposite direction from PP5 - so no reading of this record supports PP5. The trigger condition for PP5 is therefore absent on both counts, and PP5 is not met.

Rule

PP5: reputable source reports the variant as pathogenic; applied here only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic.

Evidence considered

• ClinVar VCV000485000: api_classification 'Benign/Likely benign'; api_review_status 'criteria provided, multiple submitters, no conflicts' (2 stars, not expert-panel reviewed); 5 submissions, detailed on the ClinVar record page (clinvar.png / variation 485000), all from clinical laboratories with is_expert_panel = false.
• ClinVar submission audit for this case: submissions_total 5, expert_panel_submissions 0, usable_for_criterion_leads 0, criteria_provided true but no expert-panel tier; the ClinVar 'expert_panels' object is empty.
• No STK11 ClinGen CSPEC/VCEP specification was retrieved (cspec found = false; vcep_materials 'no cspec url available'; final_classification_framework source = generic_acmg_fallback), so there is no panel-issued classification of this variant to serve as a PP5 trigger.
• PMID:25741868: consulted for the PP5 definition and the requirement that the external assertion be variant-specific and from a reputable/expert source rather than an aggregate or ordinary laboratory label.

Still missing

• No ClinGen/VCEP expert-panel assertion for STK11 c.160C>T exists in ClinVar and no STK11 VCEP specification was available for this case; if such a panel assertion is later published, PP5 must be re-adjudicated.

Sources

Reasoning

BA1 requires a stand-alone benign allele frequency of at least 0.05 (5%). The variant is extremely rare: gnomAD v4.1 total AF 4.34e-06 (7/1,613,928 alleles, 0 homozygotes), gnomAD v2.1 total AF 8.05e-06 (2/248,308 alleles, 0 homozygotes), and the highest allele frequency observed in any ancestry group is 2.67e-05 (African/African American, gnomAD v4.1). The grpmax filtering allele frequency is 4.42e-06 (v4.1). Even the single highest subpopulation figure is roughly three orders of magnitude below the 0.05 stand-alone threshold, so BA1 is not met. Because no STK11-specific ClinGen VCEP/CSPEC specification exists (generic ACMG/AMP 2015 fallback), the generic stand-alone threshold applies. The non-cancer subsets give the same picture (gnomAD v2.1 non-cancer exomes AF 8.53e-06, 2/234,470; absent from gnomAD v3.1 non-cancer genomes) and, being rarer than the all-comers totals in this case, do not alter the verdict. This criterion is not a benign-classification driver for this variant; the ClinVar benign/likely-benign submissions are not used as evidence here.

Rule

BA1 stand-alone benign: allele frequency >= 0.05 (ACMG/AMP 2015).

Evidence considered

• gnomAD v4.1 (all-comers joint): total AF 4.33724e-06 (7/1,613,928 alleles), 0 homozygotes; highest ancestry group African/African American AF 2.66574e-05 (2/75,026 alleles); grpmax FAF 4.42e-06.
• gnomAD v2.1 (all-comers exomes): total AF 8.05451e-06 (2/248,308 alleles), 0 homozygotes; highest ancestry group European (non-Finnish) AF 1.7804e-05 (2/112,334 alleles); grpmax FAF 2.96e-06.
• BA1 stand-alone benign threshold >= 0.05 (ACMG/AMP 2015, PMID:25741868).
• No governing VCEP/CSPEC specification exists for STK11, so the generic ACMG/AMP 2015 population thresholds were applied rather than any gene-specific calibration.

Still missing

• No STK11-specific VCEP/CSPEC allele-frequency thresholds exist; generic ACMG/AMP 2015 defaults were applied instead.
• gnomAD-Canada v1.0 returned no result (query timed out), so that additional population source could not be checked.

Sources

Reasoning

BS1 requires a strong benign allele frequency of at least 0.01 (1%). The variant's frequency is far below that: gnomAD v4.1 total AF 4.34e-06 and gnomAD v2.1 total AF 8.05e-06, with the single highest ancestry-specific frequency being 2.67e-05 (African/African American, gnomAD v4.1) and a grpmax filtering AF of 4.42e-06. The maximum observed frequency is therefore approximately 375-fold below the 0.01 threshold, so BS1 is not met. Because there is no STK11-specific VCEP/CSPEC, the generic (SVI/Whiffin) 0.01 default is used rather than a per-gene frequency derivation; note that a gene-specific derivation would in any case be expected to yield a lower, not higher, threshold for a rare autosomal dominant cancer-predisposition gene. The non-cancer subsets (gnomAD v2.1 non-cancer exomes AF 8.53e-06; absent from v3.1 non-cancer genomes) are likewise far below threshold and do not change the verdict.

Rule

BS1 strong benign: allele frequency >= 0.01 (generic SVI/Whiffin default).

Evidence considered

• gnomAD v4.1 (all-comers joint): total AF 4.33724e-06 (7/1,613,928 alleles), 0 homozygotes; highest ancestry group African/African American AF 2.66574e-05 (2/75,026 alleles); grpmax FAF 4.42e-06.
• gnomAD v2.1 (all-comers exomes): total AF 8.05451e-06 (2/248,308 alleles), 0 homozygotes; highest ancestry group European (non-Finnish) AF 1.7804e-05 (2/112,334 alleles); grpmax FAF 2.96e-06.
• BS1 strong benign threshold >= 0.01 (generic SVI/Whiffin default, Whiffin et al. 2017, PMID:28518168).
• gnomAD v2.1 non-cancer exomes AF 8.52988e-06 (2/234,470 alleles, 0 homozygotes) and no observation in gnomAD v3.1 non-cancer genomes, consistent with the all-comers totals.
• STK11/Peutz-Jeghers syndrome carriers have high lifetime cancer risk (CAPS consensus, PMID:31672839), so the gene is not one in which a low-frequency change can be assumed benign-tolerable; a frequency-based benign call cannot be reached at these observed rates.

Still missing

• No STK11-specific VCEP/CSPEC frequency calibration exists; the generic BS1 default of 0.01 was applied (Whiffin et al. 2017, PMID:28518168) rather than a per-gene-derived threshold.
• gnomAD-Canada v1.0 returned no result (query timed out), so that additional population source could not be checked.

Sources

Reasoning

BS2 requires the variant to be observed in a healthy adult individual, i.e. homozygous for a fully penetrant recessive disorder or heterozygous for a dominant/X-linked disorder with full penetrance expected at an early age. Neither condition is satisfied here. The variant has never been observed in the homozygous state in any queried dataset: gnomAD v2.1 carries 2 heterozygous alleles (0 homozygotes), gnomAD v4.1 carries 7 heterozygous alleles (0 homozygotes), and the v2.1 non-cancer subset carries 2 alleles with 0 homozygotes. STK11-related Peutz-Jeghers syndrome is an autosomal dominant, incompletely and age-dependently penetrant cancer-predisposition condition (ACMG SF v2.0 lists STK11/PJS as autosomal dominant with child/adult onset, PMID:27854360), so the BS2 prerequisite of penetrance expected in full at an early age is not satisfied and gnomAD carriers cannot be treated as phenotypically confirmed healthy adults for this condition. BS2 is therefore not met. This assessment rests on absence of homozygotes plus the penetrance character of the gene-disease pair, not on any positive healthy-adult observation.

Rule

BS2 benign strong: observed in a healthy adult (homozygous recessive, or dominant/X-linked with full penetrance at early age).

Evidence considered

• gnomAD v2.1 and v4.1 all-comers totals show 2/248,308 and 7/1,613,928 alleles respectively, all heterozygous, with zero homozygotes in every ancestry group; gnomAD v2.1 non-cancer exomes likewise 2/234,470 with zero homozygotes.
• STK11/Peutz-Jeghers syndrome is autosomal dominant with child/adult onset and incomplete, age-dependent penetrance (ACMG SF v2.0, PMID:27854360), so BS2's requirement of a dominant disorder with full penetrance expected at an early age is not met.
• No homozygous or biallelic observation of this variant exists in any gnomAD dataset queried, so the recessive-disorder branch of BS2 is also inapplicable.

Still missing

• gnomAD provides no phenotype-confirmed healthy-adult status, so a healthy-adult observation for BS2 cannot be positively established from these population sources.
• No variant-specific case or control cohort report for this STK11 c.160C>T variant was identified in the reviewed literature.

Sources

Reasoning

BS3 (Richards et al. 2015, PMID:25741868) requires that a well-established functional assay demonstrate no damaging effect on protein function or splicing. This is a distinct requirement from a benign population or case-observation argument, and it must not be inferred from another source's classification label. For NM_000455.5:c.160C>T (NP_000446.1:p.(Leu54=)), no functional assay of any kind has been reported: the literature assembled for this case contains no in-vitro, RNA, minigene, transcript or protein-function study of this variant, the ClinVar submission audit for it yields zero criterion-level leads and zero validated functional references, and OncoKB records no variant-specific functional or oncogenicity assay data (oncogenicity label 'Unknown Oncogenic Effect', not variant-specific). The only variant-level result available is a computational SpliceAI maximum delta score of 0.00, which is an in-silico prediction rather than an assay and cannot support BS3; a low SpliceAI delta also addresses only the splicing mechanism and would not by itself establish preserved protein function. Because no well-established functional study exists showing normal function, BS3 is not met, and the variant's ClinVar 'Likely benign' label was not used as a substitute for functional evidence. The generic ACMG/AMP 2015 fallback applies (no STK11 CSPEC or VCEP framework available).

Rule

BS3: well-established functional studies show no damaging effect

Evidence considered

• Criterion definition: BS3 requires a well-established functional study showing no damaging effect on protein function or splicing (Richards et al. 2015 ACMG/AMP, PMID:25741868).
• No functional assay data exist for NM_000455.5:c.160C>T: the literature pass for this case returned no functional-assay publication on this variant (the functional-themed paper retrieved was the ACMG/AMP guideline itself, with no STK11 or variant-specific assay data).
• ClinVar submission audit for this variant: 5 submissions, 0 expert-panel submissions, 0 criterion-lead submissions, 0 validated functional PMIDs - no assay evidence in the submission set.
• OncoKB returned 'Unknown Oncogenic Effect' with no variant-specific functional data and no citations; its assessment is explicitly non-variant-specific for this entry.
• The variant-level ClinVar classification (Likely benign) was deliberately not used to satisfy BS3, since BS3 requires functional assay evidence and must not be inferred from another source's final classification.
• SpliceAI maximum delta 0.00 is an in-silico prediction, not a functional assay, and cannot establish BS3.
• No ClinGen CSPEC or local VCEP framework for STK11 was available (vcep_materials: 'no cspec url available'); the generic ACMG/AMP 2015 fallback applies per the case's final classification framework record (generic_acmg_fallback, precedence_rank 3).

Still missing

• No well-established functional assay (RNA/splicing or protein-function) has been performed on STK11 c.160C>T, so BS3 cannot be supported; a validated assay of this variant showing normal function would be required.
• The available SpliceAI result (max delta 0.00) is computational only and is handled under the in-silico splicing group, not as functional evidence.

Sources

Reasoning

BP2 (Richards et al. 2015, PMID:25741868) is met when the variant is observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder, or observed in cis with a pathogenic variant under any inheritance pattern. No STK11 CSPEC/VCEP specification exists (generic ACMG/AMP 2015 fallback applies). Neither configuration is documented in any source for NM_000455.5:c.160C>T: all five ClinVar submissions for ClinVarID 485000 (three Likely benign, one Likely Benign, one Benign; all germline clinical testing) report the variant in isolation, with no second allele, zygosity, or phase data, and the variant-level papers reviewed (PMID:25645574, PMID:25741868, PMID:31672839, PMID:27854360) contain no cis/trans genotype observation for this variant - PMID:31672839 and PMID:25645574 discuss STK11 only at the gene/syndrome level, and PMID:27854360 lists STK11 as an autosomal dominant secondary-findings gene without any variant detail. In the absence of any observed in-cis or in-trans pathogenic partner allele, BP2 is not met. The existing benign ClinVar consensus was deliberately not used as the basis for BP2, because a criterion may never be inferred from another source's final classification.

Rule

BP2 (ACMG/AMP 2015, PMID:25741868): 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

• BP2 requires a specific allelic configuration - the variant in trans with a pathogenic variant in a fully penetrant dominant gene/disorder, or in cis with a pathogenic variant in any inheritance pattern - so only phase/zygosity-level observations can satisfy it.
• No STK11 CSPEC/VCEP framework is available, so the generic ACMG/AMP 2015 criteria were used (rule source PMID:25741868).
• ClinVar variation 485000 (NM_000455.5(STK11):c.160C>T (p.Leu54=)) has five germline clinical-testing submissions (three Likely benign, one Likely Benign, one Benign, from Color, Ambry, Labcorp/Invitae, All of Us and Myriad); none provides a second allele, zygosity, or cis/trans phase statement.
• PMID:31672839 (CAPS consortium) discusses STK11 only as a pancreatic-surveillance eligibility gene for carriers of deleterious germline variants; it reports no variant-level or phase data for c.160C>T.
• PMID:25645574 (ACG hereditary GI cancer guideline) names Peutz-Jeghers syndrome among the syndromes covered but provides no STK11 variant-level or phase content.
• PMID:27854360 (ACMG SF v2.0) lists STK11/Peutz-Jeghers syndrome (MIM 175200) as autosomal dominant 'KP and EP' with no variant-level detail.
• Per the standing rule against inferring a criterion from another source's final classification, the multi-submitter benign ClinVar consensus was not treated as BP2 evidence; BP2 rests solely on the absence of any cis/trans observation.

Still missing

• No phase-resolved genotype (proband plus a relative carrying a pathogenic STK11 variant, or long-read/phased sequencing) is available, so an in-cis configuration with a pathogenic variant could not be affirmatively tested; the criterion remains unmet on the evidence available.

Sources

Reasoning

BP5 (Richards et al. 2015, PMID:25741868) requires the variant to have been found in a case in which an alternate molecular basis for the disease (a second, independent pathogenic cause) has also been identified. BP5 therefore needs case-level information: at least one affected individual carrying this variant together with documented alternate causation. No such case exists in the evidence assembled for NM_000455.5:c.160C>T. The ClinVar record VCV000485000 contains only aggregate clinical-laboratory classifications (Benign/Likely benign) with no deposited patient histories of alternate molecular diagnosis; the STK11-relevant publications consulted (PMID:27854360, PMID:31672839, PMID:25645574) contain no case carrying this variant; and the two sources the literature pass flagged for case-observation review (PMID:26389258, PMID:26389505) were not retrieved as full text and contain no variant-level content. Because the variant has not been reported in any affected individual at all, there is no case in which an alternate molecular basis could have been found, and BP5 is not met - not because an alternate cause was excluded in a case, but because the required case observation is entirely absent. No numeric threshold or likelihood-ratio value is attached to BP5 in the governing generic ACMG/AMP framework, so no operator-direction comparison arises for this criterion.

Rule

BP5: variant found in a case with an alternate molecular basis for disease.

Evidence considered

• ClinVar VCV000485000 submissions carry no case histories or co-occurring variant data - only classification, review status and submitter identity - so no alternate-molecular-basis case can be identified from this record.
• The variant is observed only in population cohorts (gnomAD v2.1/v4.1, no homozygotes) and in ClinVar aggregate benign-leaning laboratory classifications, not in any reported patient.
• PMID:26389258 and PMID:26389505 (PDQ guidelines, flagged as case_observation sources in the literature pass) yielded no full text in this case and no variant-level or case-level content.
• PMID:25741868: consulted for the BP5 definition (variant found in a case with an alternate molecular basis for disease); the generic framework attaches no numeric value or cutoff to BP5, so no directional comparison was required.

Still missing

• No case-level data (patient histories, co-occurring pathogenic variants) is available for this variant; BP5 could only be met if an affected carrier with an independent alternate molecular diagnosis were reported.

Sources

Reasoning

BP6 (Richards et al. 2015, PMID:25741868) requires a reputable source to report the variant as benign; under the framework applied here that requires an exact-variant ClinVar assertion from a recognised expert panel (ClinGen/VCEP) classified Benign or Likely benign. The exact-variant ClinVar record VCV000485000 for NM_000455.5(STK11):c.160C>T does carry a benign-direction aggregate label - api_classification 'Benign/Likely benign', with 5 submissions (3 Likely benign, 1 Likely Benign, 1 Benign) from Color Health, Ambry Genetics, Labcorp/Invitae, All of Us Research Program and Myriad Genetics - and this was checked directly rather than assumed. However, every one of those five submissions is from an ordinary clinical laboratory: the submission audit records expert_panel_submissions = 0 and is_expert_panel = false for each, the ClinVar expert_panels object is empty, and the review status is 'criteria provided, multiple submitters, no conflicts' (2 stars), which is not an expert-panel review. No STK11 ClinGen/VCEP specification or panel assertion was available for this case either (no cspec; framework_complete = false). Because the framework applied here permits BP6 only for an exact-variant expert-panel classification, and the aggregate benign labels in this record come solely from non-expert laboratory submissions, BP6 is deliberately not triggered; the agreement of five laboratories is not an expert-panel assertion.

Rule

BP6: reputable source reports the variant as benign; applied here only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign.

Evidence considered

• ClinVar VCV000485000: api_classification 'Benign/Likely benign'; api_review_status 'criteria provided, multiple submitters, no conflicts' (review_stars 2, not expert-panel reviewed); 5 submissions from clinical laboratories only, each with is_expert_panel = false (SCV000910030 Color Health LB; SCV000672328 Ambry LB; SCV001090454 Labcorp/Invitae LB; SCV005425393 All of Us LB; SCV006092736 Myriad Benign).
• ClinVar submission audit for this case: expert_panel_submissions 0 of 5; 'expert_panels' object empty; no criterion-lead submissions.
• No ClinGen STK11 VCEP/CSPEC assertion exists for this variant (cspec found = false; vcep_materials 'no cspec url available'; generic_acmg_fallback), so the benign consensus has no expert-panel backing.
• PMID:25741868: consulted for the BP6 definition and the requirement that the benign external assertion be variant-specific and from a reputable expert source, not an aggregate or ordinary laboratory label.

Still missing

• BP6 is held at not_met solely because no ClinGen/VCEP expert-panel assertion exists for this variant; if an expert panel later classifies it Benign/Likely benign, BP6 should be re-adjudicated (the underlying five-laboratory benign consensus would then be usable).

Sources

Reasoning

Governing framework is generic ACMG/AMP 2015 (Richards et al., PMID:25741868): the case manifest reports framework_mode='generic_acmg', no STK11 ClinGen CSPEC was found (cspec.found=false) and no gene VCEP files exist, so no gene-specific specification overrides the generic criteria. Generic PS2 requires a de novo occurrence in a patient with the disease and no family history, with BOTH maternity and paternity confirmed by parental testing. This case is a variant-level assessment of STK11 NM_000455.5:c.160C>T (NP_000446.1:p.(Leu54=)), a synonymous change; no proband, no parental or trio genotypes, no pedigree, and no meioses are present in any case source. The ClinVar record (ClinVar ID 485000) contains five germline submissions (3x Likely benign, 1x Likely Benign, 1x Benign; 2-star, 'criteria provided, multiple submitters, no conflicts'), but the submission audit records zero expert-panel submissions and zero criterion-lead submissions, i.e. no submitter documented a de novo event, and an aggregate classification is not a de novo observation regardless. None of the reviewed full-text papers mentions this variant, and the only STK11 de novo report retrieved in the case literature (PMID:22118009) concerns a different variant (c.698_699insG, p.F234LfsX3) in a different patient, so it cannot be transferred to c.160C>T. PS2 is therefore not assessed; the missing evidence is a proband carrying this variant with confirmed maternity and paternity testing.

Rule

PS2 - De novo (both maternity and paternity confirmed) in a patient with the disease and no family history (ACMG/AMP 2015, strong).

Evidence considered

• No STK11 ClinGen CSPEC or VCEP specification was available (cspec.found=false, gene_vcep_dir=null), so the ACMG/AMP 2015 generic criteria (PMID:25741868) govern; PS2 is defined as a de novo occurrence in a patient with the disease and no family history, with both maternity and paternity confirmed.
• The case is a variant-level assessment of NM_000455.5:c.160C>T (p.Leu54=); no proband record, no parental testing result, no pedigree and no meioses exist in the case bundle or in prefetch.json/vcep_materials.json.
• ClinVar (ClinVar ID 485000) shows five germline submissions (3 Likely benign, 1 Likely Benign, 1 Benign; 2-star, criteria provided, multiple submitters, no conflicts); the submission audit reports 0 expert-panel submissions and 0 criterion-lead submissions, so no submitter provided a de novo observation or parental testing for this variant.
• PMID:22118009 ('Identification of a novel de novo STK11 mutation in a Chinese child with Peutz-Jeghers syndrome') reports a de novo STK11 c.698_699insG (p.F234LfsX3) in an unrelated patient; it is a different variant and therefore does not support PS2 for c.160C>T (variant not mentioned in that report).
• Pre-computed full-text extractions for PMID:25741868, PMID:25645574, PMID:31672839 and PMID:27854360 all record variant_mentioned=false; none supplies de novo data for this variant.

Still missing

• No proband carrying STK11 c.160C>T (p.Leu54=) with parental (trio) genotypes confirming both maternity and paternity; de novo status is therefore undeterminable.
• No family history documentation (phenotype, affected/unaffected relatives) accompanying this variant.

Sources

Reasoning

Governing framework is generic ACMG/AMP 2015 (PMID:25741868); no STK11 CSPEC was found and no gene VCEP materials exist, so no gene-specific PM6 modification applies. PM6 covers an assumed de novo occurrence in a patient with the disease but without confirmation of paternity and maternity. As with PS2, this case contains no proband and no clinical/pedigree record for STK11 NM_000455.5:c.160C>T (p.Leu54=): the case is a variant-level assessment, and neither the bundle, prefetch.json nor vcep_materials.json contains a patient, phenotype, parental testing or family structure. ClinVar (ID 485000) aggregates five germline submissions (3x Likely benign, 1x Likely Benign, 1x Benign; 2-star, no conflicts) but the audit shows zero criterion-lead and zero expert-panel submissions, so no submitter reported even an unconfirmed de novo event; an aggregate classification cannot be converted into a de novo assumption. No reviewed paper mentions the variant. PM6 is not assessed; the missing evidence is a proband carrying the variant with parent testing that is either absent or unconfirmed.

Rule

PM6 - Assumed de novo, but without confirmation of paternity and maternity (ACMG/AMP 2015, moderate).

Evidence considered

• Generic ACMG/AMP 2015 (PMID:25741868) defines PM6 as an assumed de novo occurrence in a patient with the disease without confirmation of paternity and maternity; no STK11 CSPEC/VCEP exists to modify it (cspec.found=false).
• No proband, phenotype, parental testing (confirmed or otherwise), pedigree or meioses for NM_000455.5:c.160C>T are present in the case sources.
• ClinVar ClinVar ID 485000: five germline submissions (3 Likely benign, 1 Likely Benign, 1 Benign), 2-star review status, 0 expert-panel submissions, 0 criterion-lead submissions - no submitter documented a de novo occurrence for this variant.
• Reviewed full-text/abstract papers (PMID:25741868, PMID:25645574, PMID:31672839, PMID:27854360) do not mention this variant and contain no de novo or family data for it.
• The one STK11 de novo report in the case literature (PMID:22118009) concerns a different variant (c.698_699insG) and is not applicable to c.160C>T.

Still missing

• No proband carrying STK11 c.160C>T (p.Leu54=) with an observed/assumed de novo occurrence; no parental genotypes of any kind (confirmed or unconfirmed).
• No clinical phenotype or family-history documentation for any carrier of this variant.

Sources

Reasoning

Governing framework is generic ACMG/AMP 2015 (PMID:25741868); STK11 has no ClinGen CSPEC or VCEP specification (cspec.found=false; gene_vcep_files empty), so no gene-specific segregation thresholds or meioses-scoring rules apply. PP1 requires co-segregation of the variant with disease across multiple affected family members, which depends on a pedigree and genotyped relatives: it is scored from the number of informative meioses showing co-segregation. No pedigree, no affected or unaffected relative genotypes, and no segregation analysis for STK11 NM_000455.5:c.160C>T (p.Leu54=) exist anywhere in the case sources; this is a variant-level assessment with no family attached. ClinVar (ID 485000) offers five germline submissions with zero criterion leads and zero expert-panel submissions, so no segregation evidence was submitted either, and a classification is not a segregation observation. No reviewed paper reports this variant in a family. PP1 is not assessed; the missing evidence is a family pedigree with genotyped affected and unaffected relatives.

Rule

PP1 - Co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease (ACMG/AMP 2015, supporting; strength can increase with the number of co-segregating meioses).

Evidence considered

• Generic ACMG/AMP 2015 (PMID:25741868) defines PP1 as co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease, scored by informative meioses; no STK11-specific specification exists to supersede it.
• No pedigree, relative genotypes, or segregation analysis for NM_000455.5:c.160C>T exists in the case bundle or in the additional case files (prefetch.json, vcep_materials.json).
• ClinVar ClinVar ID 485000 contains five germline submissions (2-star, no conflicts) but the audit reports 0 criterion-lead submissions and 0 expert-panel submissions, so no co-segregation evidence was provided by any submitter.
• Reviewed papers (PMID:25741868, PMID:25645574, PMID:31672839, PMID:27854360) do not mention c.160C>T and report no family segregation for it; the case literature only contains gene-level PJS family reports for other STK11 variants.

Still missing

• No pedigree and no genotyped affected/unaffected relatives for STK11 c.160C>T (p.Leu54=); zero informative co-segregating meioses are available.
• No segregation analysis reported by any ClinVar submitter (0 criterion leads) or in any retrieved publication.

Sources

Reasoning

Governing framework is generic ACMG/AMP 2015 (PMID:25741868); no STK11 ClinGen CSPEC or VCEP exists (cspec.found=false), so generic BS4 applies without modification. BS4 requires a demonstrated lack of segregation: the variant is present in an affected individual yet absent from other affected family members (or otherwise fails to track with disease), which demands a pedigree with genotyped relatives. No pedigree, relative genotypes, or non-segregation analysis for STK11 NM_000455.5:c.160C>T (p.Leu54=) exists in any case source, so non-segregation cannot be demonstrated. ClinVar (ID 485000) holding an aggregate Benign/Likely benign classification across five germline submissions (3x Likely benign, 1x Likely Benign, 1x Benign; 2-star, no conflicts) was deliberately NOT converted into BS4 evidence: a classification is not a segregation observation, and the submission audit shows zero criterion-lead submissions, so no submitter supplied the family data BS4 needs. BS4 is therefore not assessed; the missing evidence is a family in which affected members do not carry the variant.

Rule

BS4 - Lack of segregation in affected members of a family (ACMG/AMP 2015, strong).

Evidence considered

• Generic ACMG/AMP 2015 (PMID:25741868) defines BS4 as lack of segregation in affected members of a family; no STK11-specific specification exists to override it.
• No family, pedigree, relative genotypes or non-segregation analysis for NM_000455.5:c.160C>T exists in the case bundle or in the additional case files (prefetch.json, vcep_materials.json).
• ClinVar ClinVar ID 485000: five germline submissions (3 Likely benign, 1 Likely Benign, 1 Benign), review status 2-star 'criteria provided, multiple submitters, no conflicts'; the submission audit reports 0 criterion-lead submissions and 0 expert-panel submissions. This aggregate classification was not used as BS4 evidence, since another source's final classification cannot be converted into a segregation observation and no submitter supplied family data.
• Reviewed papers (PMID:25741868, PMID:25645574, PMID:31672839, PMID:27854360) do not mention this variant and contain no non-segregation data for it.

Still missing

• No pedigree with genotyped affected and unaffected relatives, so lack of segregation for STK11 c.160C>T cannot be established.
• ClinVar's aggregate Benign/Likely benign classification cannot substitute for BS4 because no submitter provided segregation/family data (0 criterion leads) and a classification is not a segregation observation.

Sources