PMS2 c.137G>T, p.Ser46Ile

Reasoning

The variant is a missense substitution, NM_000535.7:c.137G>T (p.S46I). The governing PMS2 VCEP rule permits PP3_Moderate when the HCI MAPP/PP2 Prior P score is >0.81. The governing HCI prior table contains the exact c.137G>T / p.S46I row with prior probability 0.9387, so PP3 is met at Moderate strength. The SpliceAI result (maximum delta 0.025) is not used because the VCEP's alternative PP3 path applies to predicted splice defects at non-canonical splicing nucleotides, not this missense substitution, and the HCI pre-assigned result is the governing path.

VCEP rule

Missense variant with “MAPP/PP2 Prior P” score  >0.81 from [http://hci-lovd.hci.utah.edu/variants.php?select_db=PMS2_priors&action=search_unique](http://hci-lovd.hci.utah.edu/variants.php?select_db=PMS2_priors&action=search_unique)

Evidence considered

• HCI-PRIORS-PMS2.txt, exact row for c.137G>T / p.S46I: HCI prior probability of pathogenicity = 0.9387, MAPP score = 29.33, custom PP2 score = 1.000.
• Case variant normalization identifies NM_000535.7:c.137G>T as a missense variant encoding NP_000526.2:p.(Ser46Ile) / p.(S46I).
• SpliceAI maximum delta score = 0.025; this was not counted because the VCEP PP3 SpliceAI alternative is limited to predicted splice defects for non-canonical splicing nucleotides.

Sources

Reasoning

The VCEP Supporting rule is met by variant-specific tumor evidence. PMID:23709753 reports c.137G>T (p.S46I) in an individual with colorectal cancer and a PMS2-negative tumor phenotype; the extracted full-text finding states that the tumor was PMS2-negative and MLH1-positive. Loss of PMS2 expression is consistent with a PMS2 variant under the VCEP instructions. PMID:16144131 additionally reports the same p.Ser46Ile substitution in a Turcot syndrome family with tumors showing absent PMS2 expression. This supports at least one qualifying CRC/tumor with MMR protein loss consistent with the variant location. The declared VCEP reference files were searched for c.137G>T, p.Ser46Ile, and p.S46I; neither file contained a matching entry, so no pre-assigned PP4 code was available from those files.

VCEP rule

1 CRC/Endometrial MSI-H tumor using a standard panel of 5-10 markers^e or tumor genome **and/or** loss of MMR protein expression consistent with the variant location. MSI-H tumor with inconsistent protein expression does not meet PP4.

Evidence considered

• PMID:23709753, Results/Table 1-3: c.137G>T (p.S46I) was observed in a family including an individual with colorectal cancer and a PMS2-negative, MLH1-positive tumor phenotype; the paper's extracted finding reports reduced MMR activity but the PP4-relevant evidence is the tumor phenotype.
• PMID:16144131, Results/Table 1/Figure 3: the same p.Ser46Ile substitution occurred in a Turcot syndrome family, and PMS2 expression was absent in the proband's tumors and normal cells.
• VCEP PP4 protein-expression instruction: PMS2 loss is consistent with a PMS2 pathogenic variant, so the reported PMS2-negative tumor is consistent with the variant location.

Still missing

• No matching c.137G>T, p.Ser46Ile, or p.S46I entry was found in /Users/gs_agent/LYFESCI_HERMES_v11/vcep_db/PMS2/fulltext/Table-for-CMMRD-diagnosis.pdf.txt.
• No matching c.137G>T, p.Ser46Ile, or p.S46I entry was found in /Users/gs_agent/LYFESCI_HERMES_v11/vcep_db/PMS2/fulltext/VCEP-pilot-variants---MMR.xlsx.txt.

Sources

Reasoning

Expert panel International Society for Gastrointestinal Hereditary Tumours (InSiGHT) classified as Likely pathogenic.

VCEP rule

This criterion is not for use as recommended by the ClinGen Sequence Variant Interpretation VCEP Review Committee.

Evidence considered

• PMS2 VCEP Version 2.0 PP5 entry states that this criterion is not for use as recommended by the ClinGen Sequence Variant Interpretation VCEP Review Committee.
• ClinVar expert panel classification

Sources

Reasoning

PS2 is not met. The proband carried constitutional p.Ser46Ile with PMS2 Q643X on the other allele; the mother and brother carried Q643X alone, supporting paternal inheritance of p.Ser46Ile rather than a de novo origin. No de novo point assignment is available, so no VCEP PS2 threshold is reached.

VCEP rule

≥ 4 _de novo_ points ... 2 - 3.5 _de novo_ points ... 1 - 1.5 _de novo_ points ... 0.5 _de novo_ points

Evidence considered

• PMID:16144131, Results/Discussion, Table 1 and Figure 3: the proband carried constitutional PMS2 S46I with Q643X on the other allele; the mother and brother carried Q643X alone, supporting paternal inheritance of S46I.
• The VCEP PS2 criteria provide thresholds of ≥ 4, 2 - 3.5, 1 - 1.5, and 0.5 de novo points; this family observation is inheritance evidence and has no assigned de novo point value.

Sources

Reasoning

The governing PMS2 VCEP PM2 rule requires an allele frequency below 0.00002 (less than 1 in 50,000 alleles) in gnomAD v4. The case gnomAD v4.1 Grpmax FAF is 0.00032437 and the total allele frequency is 0.000281853 (454/1,610,770 alleles), both above 0.00002; therefore PM2 is not met. No matching c.137G>T, p.Ser46Ile, or p.S46I row was found in VCEP-pilot-variants---MMR.xlsx.txt.

VCEP rule

Absent/extremely rare allele frequency <0.00002 (<1 in 50,000 alleles ) in gnomAD v4 dataset

Evidence considered

• ClinGen InSiGHT PMS2 VCEP specification v2.0, PM2 rule: absent/extremely rare allele frequency <0.00002 (<1 in 50,000 alleles) in gnomAD v4 dataset.
• gnomAD v4.1: total AF 0.000281853, 454/1,610,770 alleles; Grpmax FAF 0.00032437; homozygotes = 0.
• VCEP-pilot-variants---MMR.xlsx.txt was searched under c.137G>T, p.Ser46Ile, and p.S46I; no matching row was found.

Still missing

• VCEP-pilot-variants---MMR.xlsx.txt was searched for c.137G>T, p.Ser46Ile, and p.S46I; no matching entry was found.

Sources

Reasoning

The governing PMS2 VCEP BA1 rule requires a gnomAD v4 Grpmax filtering allele frequency of at least 0.0028 (0.28%) and exclusion as a founder pathogenic variant. The case gnomAD v4.1 Grpmax FAF is 0.00032437 (0.032437%), far below 0.0028; therefore BA1 is not met. No matching c.137G>T, p.Ser46Ile, or p.S46I row was found in VCEP-pilot-variants---MMR.xlsx.txt.

VCEP rule

GnomAD v4 Grpmax filtering allele frequency >= 0.0028 (0.28%) and variant is excluded as founder pathogenic variant.

Evidence considered

• ClinGen InSiGHT PMS2 VCEP specification v2.0, BA1 rule: gnomAD v4 Grpmax filtering allele frequency >= 0.0028 (0.28%) and variant is excluded as founder pathogenic variant.
• gnomAD v4.1: total AF 0.000281853; Grpmax FAF 0.00032437; highest observed population AF 0.000352582 in European (non-Finnish); homozygotes = 0.
• VCEP-pilot-variants---MMR.xlsx.txt was searched under c.137G>T, p.Ser46Ile, and p.S46I; no matching row was found.

Still missing

• VCEP-pilot-variants---MMR.xlsx.txt was searched for c.137G>T, p.Ser46Ile, and p.S46I; no matching entry was found.

Sources

Reasoning

The variant is a missense substitution and therefore falls within the missense branch of the governing BP4 rule. That branch requires an HCI MAPP/PP2 Prior P score <0.11. The exact HCI-PRIORS-PMS2 row for c.137G>T / p.S46I has a prior probability of 0.9387, which does not satisfy the BP4 threshold. The SpliceAI no-impact result is not an applicable BP4 path for this missense variant because the VCEP's alternative is limited to intronic and synonymous variants.

VCEP rule

Missense variant with “MAPP/PP2 Prior P” score  <0.11 from [http://hci-lovd.hci.utah.edu/variants.php?select_db=PMS2_priors&action=search_unique](http://hci-lovd.hci.utah.edu/variants.php?select_db=PMS2_priors&action=search_unique) OR For intronic and synonymous variants: SpliceAI predicts no splicing impact with delta score <= 0.1 as per Walker et al 2023.

Evidence considered

• HCI-PRIORS-PMS2.txt, exact row for c.137G>T / p.S46I: HCI prior probability of pathogenicity = 0.9387.
• Case variant normalization identifies NM_000535.7:c.137G>T as a missense variant; therefore the VCEP BP4 alternative for intronic and synonymous variants does not apply.

Sources

Reasoning

The PMS2 VCEP PS1 rule requires a different nucleotide change producing the same amino-acid change that was previously established by this VCEP as Pathogenic and is not a splice defect. Searches of the governing PMS2 full-text material for c.137G>T, p.Ser46Ile, and p.S46I found the requested variant's HCI prior row but no separate p.Ser46Ile nucleotide change established by this VCEP as Pathogenic. The available literature reports the requested c.137G>T/p.S46I variant itself, not a qualifying alternate nucleotide change.

VCEP rule

A predicted missense substitution that encodes the same amino acid change with a different underlying nucleotide change previously established by this VCEP as Pathogenic (not a predicted or confirmed splice defect).

Evidence considered

• The governing PMS2 VCEP full-text search found no separate p.Ser46Ile alternate-nucleotide entry established as Pathogenic; the searched notation forms were c.137G>T, p.Ser46Ile, and p.S46I.
• PMID:23709753 explicitly studied c.137G>T (p.S46I), so it documents the target variant rather than a different nucleotide change producing the same amino-acid change.

Still missing

• A qualifying alternate-nucleotide PMS2 variant encoding p.Ser46Ile and established as Pathogenic by this VCEP was not found in the searched governing materials.

Sources

Reasoning

PMID:23709753 explicitly evaluated PMS2 c.137G>T (p.S46I) in a transient-expression in vitro mismatch-repair assay and reported repair activity of 12.75% ± 4.68% relative to wild type, with preserved protein expression and nuclear localization. The VCEP functional-assay documentation identifies the approved mammalian MMR activity assay as abnormal at <25% in two independent assays and lists validated assay PMIDs 24362816 and 31332305, not PMID:23709753. The VCEP flowchart requires two independent concordant assays when a calibrated functional assay has not been conducted; the available variant-specific paper documents one MMR activity assay, so the VCEP PS3 rule cannot be completed and no calibrated functional odds are available.

VCEP rule

MMR function defect following functional assay flowchart*

Evidence considered

• PMID:23709753, Results, “In silico and in vitro protein analyses of PMS2 missense variants,” p. 557; Table 1, Family I; Table 2; Table 3: PMS2 c.137G>T (p.S46I) showed significantly reduced MMR activity of 12.75% ± 4.68% relative to wild type, while protein expression and nuclear localization were not different from wild type.
• Functional-assay-SVI-documentation-MMR.xlsx, sheet “MMR assays,” mammalian MMR activity assay row: abnormal readout is <25% in two independent assays; 7 P/LP and 5 B/LB validation controls; proposed strength PS3_moderate/BS3_supporting.
• Functional-assay-flowchart.pdf, page 1, Branch 1 for in-frame indel or missense variants: when no calibrated functional assay is conducted, two independent assays with concordant results are required; repair levels similar to a deficient cell line or pathogenic controls indicate deficient function.

Still missing

• Functional-assay-SVI-documentation-MMR.xlsx: no entry for c.137G>T, p.Ser46Ile, or p.S46I; the file's validated assay list does not include PMID:23709753.
• Functional-assay-flowchart.pdf: no variant-specific entry; a second independent concordant functional assay for c.137G>T (p.S46I) is not available.
• No calibrated functional odds for pathogenicity are reported for this variant.

Sources

Reasoning

The PMS2 VCEP provides PM3 strength thresholds of ≥4 points for Very Strong, 2–3.5 points for Strong, 1–1.5 points for Moderate, and =0.5 points for Supporting, but the available VCEP criteria text does not provide the point weights or assignment rules needed to convert the observations into a point total. PMID:16144131 reports the S46I substitution with the truncating Q643X variant on the other allele in a proband, with parental testing supporting inheritance of S46I from the father. PMID:18602922 reports one biallelic carrier of c.137G>T with c.137G>A and six additional monoallelic carriers, but the available extraction does not establish the phase or a VCEP-assigned point value for each observation. Therefore PM3 cannot be assigned without inventing a point-calculation rule.

Evidence considered

• ClinGen InSiGHT PMS2 CSPEC v2.0 PM3 criteria entry: ≥4 points = Very Strong; 2–3.5 points = Strong; 1–1.5 points = Moderate; =0.5 points = Supporting.
• Two PMS2 mutations in a Turcot syndrome family with small bowel cancers, PMID:16144131: the proband had constitutional PMS2 S46I with truncating Q643X on the other allele; parental testing supported inheritance of S46I from the father.
• The clinical phenotype of Lynch syndrome due to germ-line PMS2 mutations, PMID:18602922: c.137G>T was reported in seven probands, including one biallelic carrier also carrying c.137G>A; the available extraction does not establish phase for that observation or provide VCEP PM3 points.

Still missing

• The VCEP PM3 point-assignment weights and rules for individual probands, phase, and the pathogenicity of the in-trans partner are not present in the available governing criteria entry.
• A defensible total PM3 point score cannot be calculated from the available evidence without that VCEP point-assignment scheme.

Sources

Reasoning

The PMS2 VCEP PM5 rule requires a different missense change at residue 46 classified by this VCEP as Pathogenic or Likely Pathogenic at the protein level, not because of aberrant splicing; it also requires PP3 at Supporting strength. The HCI-PRIORS-PMS2 table lists other residue-46 substitutions (p.S46C, p.S46N, p.S46T, and p.S46R) and the target p.S46I, but does not establish those comparators as VCEP Pathogenic or Likely Pathogenic classifications. The collected PM5 candidate file contains no candidates, and no qualifying VCEP comparator was found in the governing materials.

VCEP rule

Missense change at an amino acid residue where a different missense change was classified by this VCEP as Pathogenic on the protein level and not due to aberrant splicing. Only use PM5 if PP3 is supporting for the missense change. Use PM5\_Supporting if other variant is Likely Pathogenic due to a missense alteration.

Evidence considered

• HCI-PRIORS-PMS2.txt, residue-46 rows: c.136A>T p.S46C; c.137G>A p.S46N; c.137G>C p.S46T; c.138T>A and c.138T>G p.S46R; and target c.137G>T p.S46I.
• pm5_candidates.json reports zero total candidates, zero same-residue candidates, and zero expert-panel Pathogenic/Likely Pathogenic candidates.
• The HCI prior for the target c.137G>T is 0.9387; under the VCEP PP3 rule, this is above the >0.81 Moderate band rather than the >0.68 and <=0.81 Supporting band required by the PM5 rule.

Still missing

• A different PMS2 missense variant at residue 46 with a VCEP Pathogenic or Likely Pathogenic protein-level classification was not available.
• The required PP3 Supporting prerequisite is not established in the evidence reviewed.

Sources

Reasoning

PP1 cannot be assessed. PMID:16144131 reports family-member testing and supports paternal inheritance of S46I, but it does not report a combined Bayes Likelihood Ratio for co-segregation. The VCEP PP1 thresholds therefore cannot be applied.

VCEP rule

Co-segregation with disease in pedigree(s) with a combined\* Bayes Likelihood Ratio^f >18.7​ in ≥2 families. ... Co-segregation with disease in pedigree(s) with a combined\* Bayes Likelihood Ratio^​f >4.3 & ≤18.7. ... Co-segregation with disease in pedigree(s) with a combined\* Bayes Likelihood Ratio^f >2.08 & ≤4.3.

Evidence considered

• PMID:16144131, Results/Discussion, Table 1 and Figure 3: family-member testing is reported, including the proband, mother, brother, and paternal inheritance of S46I, but no co-segregation Bayes Likelihood Ratio is reported.
• The VCEP PP1 rule specifies combined Bayes Likelihood Ratio intervals >18.7 in ≥2 families, >4.3 and ≤18.7, or >2.08 and ≤4.3; no value is available here.

Still missing

• A combined Bayes Likelihood Ratio for co-segregation in the relevant pedigree(s) is not available.

Sources

Reasoning

The governing PMS2 VCEP BS1 rule requires a gnomAD v4 Grpmax filtering allele frequency of at least 0.00028 and less than 0.0028, plus exclusion as a founder pathogenic variant. The gnomAD v4.1 Grpmax FAF is 0.00032437 and falls within the frequency interval, but the available literature describes c.137G>T as a possible founder mutation while stating that available numbers are insufficient to draw firm conclusions about origin or population prevalence. The required founder-exclusion condition therefore cannot be established, so BS1 is not assessed.

VCEP rule

GnomAD v4 Grpmax filtering allele frequency >= 0.00028 and < 0.0028 (0.028-0.28%) and variant is excluded as founder pathogenic variant.

Evidence considered

• ClinGen InSiGHT PMS2 VCEP specification v2.0, BS1 rule: gnomAD v4 Grpmax filtering allele frequency >= 0.00028 and < 0.0028 (0.028-0.28%) and variant is excluded as founder pathogenic variant.
• gnomAD v4.1: Grpmax FAF 0.00032437, within the VCEP BS1 frequency interval; homozygotes = 0.
• The clinical phenotype paper reports c.137G>T in seven probands and describes it as a possible additional PMS2 founder mutation, while stating that the number of affected probands and family members was insufficient to draw firm conclusions about origin or population prevalence.
• VCEP-pilot-variants---MMR.xlsx.txt was searched under c.137G>T, p.Ser46Ile, and p.S46I; no matching row was found.

Still missing

• Founder status is unresolved: PMID:18602922 calls c.137G>T a possible PMS2 founder mutation but says available data are insufficient for firm conclusions.
• VCEP-pilot-variants---MMR.xlsx.txt was searched for c.137G>T, p.Ser46Ile, and p.S46I; no matching entry was found.

Sources

Reasoning

The governing PMS2 VCEP BS2 rule requires co-occurrence in trans with a known pathogenic sequence variant in the same gene in a patient with colorectal cancer after age 45 or another Lynch-syndrome cancer above its median age of onset, with no clinical manifestations of CMMRD, and parental or offspring testing to confirm phase. The reviewed variant-specific reports describe co-occurrence with another PMS2 variant, but the available evidence does not establish all required age, cancer-type, CMMRD, and phase conditions for BS2.

VCEP rule

Co-occurrence _in trans_ with a known pathogenic sequence variant in the same gene in a patient with colorectal cancer after age 45 (or other LS cancer above the median age of onset for that cancer in LS), and who has no previous or current evidence of clinical manifestations of CMMRD as per Aronson et al 2022 (Refer to 'Table for CMMRD diagnosis.pdf'). Confirmation of phase requires testing of parents or offspring.

Evidence considered

• ClinGen InSiGHT PMS2 VCEP specification v2.0, BS2 rule requires co-occurrence in trans with a known pathogenic sequence variant in the same gene in a patient with colorectal cancer after age 45 (or another Lynch-syndrome cancer above the median age of onset), no previous or current CMMRD manifestations, and parental or offspring testing to confirm phase.
• PMID:16144131 reports S46I with PMS2 Q643X in a Turcot syndrome family and describes family testing, but the reported phenotype and available excerpt do not establish the VCEP's qualifying BS2 conditions.
• PMID:18602922 reports c.137G>T among PMS2 carriers and one biallelic case with c.137G>A and sigmoid cancer at age 14; this does not satisfy the VCEP BS2 age requirement and does not establish all phase and CMMRD conditions.
• VCEP-pilot-variants---MMR.xlsx.txt was searched under c.137G>T, p.Ser46Ile, and p.S46I; no matching row was found.

Still missing

• The available variant-specific reports do not establish all BS2 requirements, especially confirmed phase, qualifying cancer age, and no CMMRD manifestations.
• VCEP-pilot-variants---MMR.xlsx.txt was searched for c.137G>T, p.Ser46Ile, and p.S46I; no matching entry was found.

Sources

Reasoning

No variant-specific proficient-function result is available. PMID:23709753 reported markedly reduced MMR activity rather than proficient function, and the available extraction documents one functional assay rather than the two independent concordant assays required by the VCEP flowchart when no calibrated assay is used. No calibrated functional odds for pathogenicity are available to evaluate either BS3 threshold.

Evidence considered

• PMID:23709753, Results, “In silico and in vitro protein analyses of PMS2 missense variants,” p. 557; Table 2: p.S46I showed MMR activity of 12.75% ± 4.68% relative to wild type, which is damaging rather than a variant-specific proficient-function result.
• Functional-assay-SVI-documentation-MMR.xlsx, sheet “MMR assays”: BS3 requires calibrated functional odds or variant-specific proficient function according to the MMR functional assay flowchart.
• Functional-assay-flowchart.pdf, page 1, Branch 1 for in-frame indel or missense variants: proficient function requires the specified concordant assay pathway, including nuclear localization, sensitivity to DNA damage, and no mutator phenotype where applicable.

Still missing

• Functional-assay-SVI-documentation-MMR.xlsx: no entry for c.137G>T, p.Ser46Ile, or p.S46I; no calibrated functional odds for this variant.
• Functional-assay-flowchart.pdf: no variant-specific entry and no independent concordant proficient-function assay pair for c.137G>T (p.S46I).

Sources

Reasoning

BS4 cannot be assessed. The available family report does not provide a pedigree-level combined Bayes Likelihood Ratio for lack of co-segregation or a quantified non-segregation analysis meeting either VCEP BS4 interval.

VCEP rule

Lack of co-segregation with disease in pedigree(s) with a combined\* Bayes Likelihood Ratio^f \<0.05​.  ... Lack of co-segregation with disease in pedigree(s) with a combined\* Bayes Likelihood Ratio^f >0.05 & ≤0.48.

Evidence considered

• No quantified pedigree-level lack-of-co-segregation analysis or combined Bayes Likelihood Ratio is reported in the available case evidence.
• The VCEP BS4 rule requires a combined Bayes Likelihood Ratio <0.05 for Strong or >0.05 and ≤0.48 for Supporting; no value is available here.

Still missing

• A combined Bayes Likelihood Ratio for lack of co-segregation in the relevant pedigree(s) is not available.

Sources

Reasoning

BP5 is applicable under the VCEP, but the available evidence does not establish either qualifying pathway: two or more CRC/endometrial/LS-spectrum tumors with MSS or absent MMR-protein loss inconsistent with PMS2, or BRAF V600E/MLH1 methylation evidence meeting the VCEP count. The available variant-specific extracts describe PMS2 loss rather than the required benign-inconsistent tumor findings, and do not provide the required BRAF V600E or MLH1 methylation count.

VCEP rule

2 or 3 tumors: CRC/Endometrial tumors with MSS and/or no loss of MMR protein expression and/or LS spectrum tumors^d with loss of MMR protein(s) that is inconsistent with the gene demonstrating genetic variation. ... 1 BRAF V600E (Colon only)/_MLH1_ methylation (in LS spectrum tumor only) with MSI-H/_MLH1_ loss.

Evidence considered

• The VCEP BP5 rule requires 2 or 3 tumors for Supporting or at least 4 tumors for Strong when based on MSS/no MMR-protein loss or inconsistent LS-spectrum protein loss, or requires the specified BRAF V600E/MLH1 methylation pathway.
• PMID:23709753 and PMID:16144131 report PMS2-negative tumor/protein-expression findings, which do not establish the BP5 benign-inconsistent phenotype or the BRAF/methylation pathway.

Still missing

• No qualifying BP5 tumor count or BRAF V600E/MLH1 methylation result is available in the reviewed case evidence.

Sources