VHL c.541G>A, p.(Val181Ile)

Reasoning

VHL VCEP v1.1 PM1 applies at Moderate strength to putative missense variants that are known germline hotspots and/or lie in key functional domains and/or are somatic hotspots with >=10 same-AA instances in cancerhotspots.org. Codon 181 falls within the pVHL alpha (α) domain (AA 156-192), which the same VCEP specification designates a key functional domain responsible for Elongin C binding (critical for VBC complex assembly and HIF regulation). This domain placement alone qualifies for PM1_Moderate under the VCEP wording ('...AND/OR in key functional domains'). The somatic-hotspot arm of the rule is not met: cancerhotspots.org returned no statistically significant hotspot at V181, and COSMIC records only a single somatic occurrence (COSV56543552, n=1), well below the >=10-instance threshold, so the somatic-count pathway does not independently support this criterion.

Rule

VHL VCEP v1.1 PM1_Moderate: applies to missense variants in known germline hotspots and/or key functional domains and/or somatic hotspots with >=10 same-AA cancerhotspots.org instances.

Evidence considered

• VHL VCEP v1.1 PM1 rule: 'Putative missense variants that are known germline hotspots AND/OR in key functional domains AND/OR somatic variants that have >=10 instances for the same AA in cancerhotspots.org.'
• VHL VCEP v1.1 domain definitions (cited under PM4/PVS1 sections of the same specification): Alpha (α) domain spans AA 155/156-192 and mediates Elongin C binding, a function critical for VBC complex (VHL-Elongin B-Elongin C-Cullin2-RBX1) assembly; codon 181 (Val181) lies within this alpha domain.
• cancerhotspots.org query 'VHL V181' returned search_status 'no_result' / found=false, with evidence_sentence: 'This variant does not lie in a statistically significant hotspot.'
• COSMIC record COSV56543552 shows only 1 somatic occurrence for this variant, far below the VCEP's >=10-instance cancerhotspots.org threshold, so the somatic-hotspot arm of PM1 is not satisfied on its own.

Still missing

• Confirm with a human reviewer whether the VCEP intends codon-level granularity within the alpha domain (e.g., core Elongin C-binding subregion 157-172 vs the broader 156-192 domain) to qualify for PM1, since V181 lies in the broader alpha domain but outside the narrower 157-172 Elongin C-binding core cited elsewhere in the same specification.

Sources

Reasoning

VHL VCEP v1.1 restricts PS1 to variants with the identical amino acid change (p.Val181Ile) already interpreted by the VHL VCEP, or established as pathogenic using VHL VCEP specifications. ClinVar (VariationID 238111) shows p.(Val181Ile) itself is currently Uncertain significance with no expert-panel or VCEP-based pathogenic interpretation. No other source (literature, PM5 comparator search) identified a prior VCEP-based pathogenic classification of the same amino acid change. PS1 requirements are therefore not satisfied.

Rule

VHL VCEP v1.1 PS1: Pathogenic Strong when the identical amino-acid change has been interpreted by the VHL VCEP or established as pathogenic using VHL VCEP specifications.

Evidence considered

• VHL VCEP v1.1 PS1 rule: 'Applied only to variants with interpretation by the VHL VCEP or by a variant with pathogenicity established using VHL VCEP specifications.'
• ClinVar VariationID 238111 (NM_000551.4(VHL):c.541G>A / p.Val181Ile) is itself classified Uncertain significance by 6 usable submissions, with no VHL VCEP or expert-panel entry, so the same amino-acid change has not been independently established as pathogenic under VCEP specifications.

Sources

Reasoning

PM2 is not met under the VHL Expert Panel specification. PM2 Supporting requires the variant to be absent from gnomAD or to have gnomAD v4 GroupMax Filtering Allele Frequency <=0.00000156 (0.000156%); the variant is present and has GroupMax FAF 3.65e-06 (0.000365%), exceeding the permitted threshold.

Rule

Use PM2 Supporting when the variant is absent from gnomAD or gnomAD v4 GroupMax Filtering Allele Frequency is <=0.00000156 (0.000156%); if no GroupMax FAF is calculated, PM2 Supporting may also apply.

Evidence considered

• ClinGen VHL Expert Panel Specification v1.1: PM2 Supporting applies when the variant is absent from gnomAD or has gnomAD v4 GroupMax Filtering Allele Frequency <=0.00000156 (0.000156%); if no GroupMax FAF is calculated, PM2 Supporting may also apply.
• gnomAD v4.1 detects the variant in 5/1,614,186 alleles, with zero homozygotes, and reports GroupMax FAF 3.65e-06 (0.000365%). A GroupMax FAF is available and is greater than the VCEP PM2 threshold.
• The variant is therefore neither absent nor sufficiently rare under the VCEP-specific PM2 rule; the absence of homozygotes does not rescue PM2.

Sources

Reasoning

VHL VCEP (cspec v1.1) specifies two independent PP3 paths. Missense path: REVEL >=0.664 required; this variant (NM_000551.4:c.541G>A, p.Val181Ile) scores REVEL=0.431 (local REVEL v1.3 lookup), below threshold, so PP3 is not met via the missense path. Splice path: requires concordance of SpliceAI (>0.5) AND VarSeak (class 4 or 5); SpliceAI max delta score for this variant is 0.014 (SpliceAI Lookup), far below the 0.5 threshold, so the splice path also fails regardless of VarSeak availability. Both VCEP-specified paths for PP3 are unmet.

Rule

PP3_supporting (VHL VCEP cspec v1.1)

Evidence considered

• VHL VCEP cspec v1.1 PP3 rule: 'For missense variants, use REVEL score >=0.664. For splice variants, concordance of Splice AI (>0.5) and VarSeak (class 4 or class 5).'
• Local REVEL v1.3 predictor lookup: REVEL score = 0.431 for NM_000551.4:c.541G>A (p.Val181Ile), below the VCEP's 0.664 missense threshold.
• SpliceAI Lookup: max delta score = 0.014 for this variant, far below the VCEP's 0.5 splice-path threshold for PP3.

Sources

Reasoning

BA1 is not met. The VHL Expert Panel specifies BA1 at a gnomAD v4 GroupMax Filtering Allele Frequency of >=0.000156 (0.0156%). The observed gnomAD v4 GroupMax FAF is 3.65e-06 (0.000365%), well below the threshold.

Rule

Use BA1 when gnomAD v4 GroupMax Filtering Allele Frequency is >=0.000156 (0.0156%).

Evidence considered

• ClinGen VHL Expert Panel Specification v1.1: BA1 requires gnomAD v4 GroupMax Filtering Allele Frequency >=0.000156 (0.0156%).
• gnomAD v4.1 reports GroupMax FAF 3.65e-06 (0.000365%), with total AF 3.09754e-06 from 5/1,614,186 alleles and no homozygotes; the maximum observed population AF is 2.19597e-05 in South Asians, also below BA1.
• The gnomAD v4 value is far below the BA1 cutoff; no threshold or ancestry artifact supports applying BA1.

Sources

Reasoning

BS1 is not met. The VHL Expert Panel specifies BS1 at a gnomAD v4 GroupMax Filtering Allele Frequency of >=0.0000156 (0.00156%). The observed gnomAD v4 GroupMax FAF is 3.65e-06 (0.000365%), below the threshold.

Rule

Use BS1 when gnomAD v4 GroupMax Filtering Allele Frequency is >=0.0000156 (0.00156%).

Evidence considered

• ClinGen VHL Expert Panel Specification v1.1: BS1 requires gnomAD v4 GroupMax Filtering Allele Frequency >=0.0000156 (0.00156%).
• gnomAD v4.1 reports GroupMax FAF 3.65e-06 (0.000365%), below the BS1 cutoff; the highest observed population AF is 2.19597e-05 in South Asians, but the VCEP rule uses GroupMax FAF rather than an unfiltered subgroup AF.
• The variant is observed at low frequency in multiple ancestry groups and has no homozygotes, but its GroupMax FAF does not reach the VCEP BS1 threshold.

Sources

Reasoning

No proband-level de novo observation is documented in the case-evidence bundle. Maternity and paternity confirmation, phenotype classification, and absence of a VHL family history are not available, so PS2 cannot be assigned.

Rule

Use the VHL VCEP PS2 scoring system: 0.5 supporting, 1 moderate, 2 strong, and 4 very strong points; a single proband cannot receive very strong strength, and PS2 requires confirmed maternity and paternity.

Evidence considered

• The VHL VCEP specification requires de novo scoring using 0.5/1/2/4 points according to phenotype specificity and requires both maternity and paternity confirmation for PS2.
• The consolidated case-evidence bundle contains no documented proband, parental testing, maternity/paternity confirmation, or de novo observation for NM_000551.4:c.541G>A (p.Val181Ile).

Still missing

• Proband-level clinical phenotype and VHL Danish-criteria assessment
• Molecular confirmation that the variant is absent in both parents
• Independent confirmation of maternity and paternity
• Documented absence of VHL family history and required negative testing where applicable

Sources

Reasoning

The VHL VCEP specifies PS3 requires acceptable functional assays (HIF1/2a degradation assay, VBC complex stability assay, or ECM formation/fibronectin binding assay), evaluated per the Brnich et al. control workflow (PMID:31892348) to reach PS3_Supporting/Moderate. No such wet-lab functional assay data for VHL c.541G>A (NP_000542.1:p.(V181I)) was found anywhere in the case bundle (evidence.json, literature_pass.json, vcep_materials.json, paper_extracts.json). The only variant-specific literature hit is PMID:18195360, which reports 10-ns molecular-dynamics simulations of unbound pVHL stability for a computationally designed V181I mutant -- this is an in-silico structural stability prediction, not one of the VCEP-accepted experimental assay types (HIF1/2a degradation, VBC complex stability, or ECM/fibronectin binding), and it reports no experimental functional readout, biological replicates, or positive/negative controls as required by the Brnich workflow. It therefore cannot be used to support PS3.

Rule

VHL VCEP cspec v1.1 PS3: Acceptable assays are HIF1/2a degradation, VBC/VCB complex stability, and ECM/fibronectin binding assays, applied via the Brnich et al. (PMID:31892348) control-scoring workflow (10 controls for Supporting, 11 for Moderate).

Evidence considered

• vcep_materials.json summarizes VBC-complex-form-stability and HIF1/2a-degradation assay tables from the VHL_Functional_Assay_SVI_documentation.xlsx but these catalog general assay/PMID pairs (e.g. Rechsteiner 2011 PMID:21715564, Clifford 2001 PMID:11331613) and do not list p.(V181I)/c.541G>A as a tested variant
• PMID:18195360 (paper_extracts.json): molecular-dynamics simulation found V181I 'more stable than wild type' pVHL among five designed mutants, but this is a computational stability prediction with no experimental HIF1/2a, VBC, or fibronectin/ECM readout and no Brnich-style controls, so it does not meet the VCEP's accepted PS3 assay definition

Still missing

• No HIF1/2a degradation assay data for p.(V181I)
• No VBC/VCB complex stability assay data for p.(V181I)
• No ECM formation/fibronectin binding assay data for p.(V181I)
• PMID:18195360 is a computational MD simulation, not an accepted VCEP functional assay type, and lacks documented controls/replicates

Sources

Reasoning

PS4 cannot be assigned because the VHL Expert Panel requires cumulative proband-table points, and the available evidence supplies no scored affected probands, phenotype details, or case-control enrichment data for p.(Val181Ile).

Rule

VHL VCEP v1.1 PS4: apply only from the total in the VHL Proband Scoring Tables (Supporting 1 point; Moderate 2-4; Strong 5-15; Very Strong at least 16).

Evidence considered

• The VHL Expert Panel specifies PS4 by total proband points: 1 point (Supporting), 2-4 points (Moderate), 5-15 points (Strong), and at least 16 points (Very Strong).
• PMID:18195360 explicitly modeled V181I in molecular-dynamics simulations but reports no patient-level observation; it therefore contributes no VHL PS4 proband points or case-control enrichment data.

Still missing

• Variant-specific affected-proband data with phenotype assignments sufficient for the VHL PS4 Proband Scoring Tables.
• A case-control enrichment analysis or other denominator-based enrichment evidence for p.(Val181Ile).

Sources

Reasoning

VHL VCEP v1.1 PM5 requires a different missense change at the same residue (V181) with pathogenicity established by the VHL VCEP or under VCEP specifications, with the observed change (V181I) at an equal-or-greater Grantham distance than the classified comparator, and splice metapredictors excluding a splicing effect. The automated same-residue comparator search (pm5_candidates.json) targeted this exact logic but failed to complete: it returned 'total_candidates_seen: 5' with 'same_residue_candidates: 0' and an explicit tool error 'HTTP Error 429: Too Many Requests', so the candidate harvesting from ClinVar for alternate V181 missense changes did not run to completion. No same-residue comparator variant (e.g., p.Val181X other than Ile) with an established VCEP pathogenic classification was identified in the bundle. Given the search failure rather than a confirmed negative result, this criterion cannot be reliably called not_met and is left not_assessed pending a manual same-residue ClinVar/literature search.

Rule

VHL VCEP v1.1 PM5_Moderate: same-residue alternate missense change with VCEP-established pathogenicity and Grantham distance >= the comparator variant; splice metapredictors must exclude a splicing effect.

Evidence considered

• VHL VCEP v1.1 PM5 rule: pathogenicity of the prior variant at the same residue must be established by VHL VCEP interpretation or VCEP specifications; Grantham distance of the variant under consideration must be equal or larger; splice metapredictors should exclude a splicing effect.
• pm5_candidates.json automated harvest: pm5_mode 'classic_same_residue_missense', eligible_for_classic_pm5_search=true, total_candidates_seen=5, same_residue_candidates=0, recommendation level 'no_useful_candidates_found', with explicit tool error 'HTTP Error 429: Too Many Requests' indicating the search did not fully complete rather than confirming a true absence of comparators.

Still missing

• PM5 same-residue comparator search (pm5_candidates.json) failed with HTTP 429 (Too Many Requests) before completing; a manual ClinVar/literature query for other pathogenic missense variants at VHL codon 181 (e.g., p.Val181Gly/Glu/Ala/etc.) is needed to properly adjudicate PM5.
• If a qualifying comparator is found, Grantham distance between the comparator's substituted residue and Val, versus Ile and Val, must be compared per the VCEP rule (Grantham 1974, Table 2) and splice metapredictor results must confirm no splicing effect for c.541G>A.

Sources

Reasoning

No assumed de novo proband or parental testing result is documented in the case-evidence bundle. Therefore, there is no basis to calculate the VHL-specific half-point de novo score for unconfirmed maternity and/or paternity.

Rule

See VHL VCEP PS2 scoring; assumed de novo receives half the points of maternity- and paternity-confirmed de novo evidence, and PM6 strength is determined from the resulting aggregate points.

Evidence considered

• The VHL VCEP states that assumed de novo observations without confirmed maternity and paternity are scored through PM6 using half the corresponding PS2 points.
• The consolidated case-evidence bundle contains no documented assumed de novo observation, parental testing, or proband phenotype for this variant.

Still missing

• A documented de novo proband observation
• Parental genotypes and whether maternity and paternity were confirmed
• Proband phenotype specificity under the VHL Danish-criteria framework
• Number of independent probands if aggregating PM6 evidence

Sources

Reasoning

No affected relatives carrying the variant, informative meioses, pedigrees, or phenotype-confirmed VHL segregation data are documented. PP1 cannot be assigned or strength-calibrated.

Rule

VHL VCEP PP1: supporting for 3–4 meioses across at least 1 family, moderate for 5–6 meioses across at least 1 family, and strong for more than 7 meioses across at least 2 families; phenotype must be highly consistent with VHL Danish Criteria.

Evidence considered

• The VHL VCEP permits PP1 only when the phenotype is highly consistent with VHL and specifies 3–4 meioses for supporting, 5–6 meioses for moderate, and more than 7 meioses across at least 2 families for strong evidence.
• The consolidated case-evidence bundle contains no pedigree, relative genotypes, segregation observations, or meiosis count for this variant.

Still missing

• Pedigree with variant status for affected and unaffected relatives
• VHL Danish-criteria phenotype documentation for affected relatives
• Count of informative meioses
• Evidence that segregation observations are independent and not duplicated

Sources

Reasoning

BS2 cannot be assessed from the available case evidence. The VHL Expert Panel requires at least 3 individuals, all >=65 years old, unaffected, carrying the same variant, with full phenotyping and screening for absence of VHL-related cancers for BS2 Strong; a Supporting application requires the same age and count criteria without full screening but with no noted VHL-related cancers. The bundle provides gnomAD homozygote counts of zero but does not identify three eligible older carriers or provide their age, phenotype, screening, or cancer histories.

Rule

BS2 Strong: at least 3 unaffected carriers, all >=65 years old, with full phenotyping and screening showing no VHL-related cancers. BS2 Supporting: at least 3 carriers, all >=65 years old, with no noted VHL-related cancers but without full phenotyping and screening.

Evidence considered

• ClinGen VHL Expert Panel Specification v1.1: BS2 Strong requires at least 3 unaffected individuals all >=65 years old with full phenotyping and screening for absence of VHL-related cancers; BS2 Supporting requires at least 3 individuals all >=65 years old with no noted VHL-related cancers but lacking full phenotyping and screening.
• gnomAD v4.1 reports zero homozygotes for the variant, but this is not equivalent to three unaffected carriers aged >=65 years and provides no required individual-level clinical screening data.

Still missing

• No documented set of at least 3 carriers aged >=65 years with the variant.
• No individual-level phenotype, VHL-related cancer screening, or cancer-history information sufficient to distinguish BS2 Strong from BS2 Supporting.
• Absence of homozygotes in gnomAD does not establish BS2, which concerns unaffected older individuals carrying the variant.

Sources

Reasoning

The VHL VCEP specifies BS3_Supporting when an HIF1/2a degradation assay replicates WT function, VBC complex stability is unaffected, and/or ECM formation/fibronectin binding is unaffected, following the same Brnich et al. controls framework as PS3. No such wet-lab functional assay data for VHL c.541G>A (NP_000542.1:p.(V181I)) was found in the case bundle. The only variant-specific study, PMID:18195360, is a computational MD stability simulation of a designed V181I mutant reporting increased structural stability relative to wild-type pVHL; this is not one of the VCEP-accepted assay readouts (HIF degradation, VBC stability, ECM/fibronectin binding) and reports no experimental normal-function replication, so it cannot be used to support BS3.

Rule

VHL VCEP cspec v1.1 BS3: BS3_Supporting when HIF1/2a assay replicates WT function and/or VBC complex stability is unaffected and/or ECM formation/fibronectin binding is unaffected, per the Brnich et al. (PMID:31892348) controls framework; for splicing variants RNA assays must show no splicing impact.

Evidence considered

• vcep_materials.json summarizes VBC-complex-form-stability and HIF1/2a-degradation assay tables from the VHL_Functional_Assay_SVI_documentation.xlsx (e.g. Rechsteiner 2011 PMID:21715564, Clifford 2001 PMID:11331613) but none list p.(V181I)/c.541G>A as a tested variant
• PMID:18195360 (paper_extracts.json): molecular-dynamics simulation found V181I 'more stable than wild type' pVHL, a computational prediction with no experimental HIF1/2a, VBC, or fibronectin/ECM readout, so it does not meet the VCEP's accepted BS3 assay definition

Still missing

• No HIF1/2a degradation assay showing WT-like function for p.(V181I)
• No VBC/VCB complex stability assay showing unaffected stability for p.(V181I)
• No ECM formation/fibronectin binding assay showing unaffected binding for p.(V181I)
• PMID:18195360 is a computational MD simulation, not an accepted VCEP functional assay type, and does not report loss-of-function evidence controls

Sources

Reasoning

No affected relatives who lack the variant, or fully phenotyped unaffected relatives who carry it, are documented. The VHL-specific requirements for informative non-segregation are therefore not met with available evidence and BS4 cannot be assigned.

Rule

VHL VCEP BS4: supporting for lack of segregation in 1 family and strong for lack of segregation in affected members of at least 2 families; affected members should fulfill Danish Criteria and unaffected relatives require complete phenotyping with appropriate age consideration.

Evidence considered

• The VHL VCEP requires affected pedigree members to fulfill Danish Criteria and states that BS4 supporting applies to one family with lack of segregation, while BS4 strong requires lack of segregation in affected members of at least two families.
• The consolidated case-evidence bundle contains no non-segregation observation, affected-relative testing, unaffected-relative phenotyping, or family count for this variant.

Still missing

• Informative pedigree and genotypes of affected relatives
• Documentation that affected relatives fulfill VHL Danish Criteria rather than isolated pheochromocytoma or renal cell carcinoma
• Full VHL phenotyping and age of unaffected relatives, including age at least 65 years where relevant
• Number of families showing lack of segregation

Sources

Reasoning

The VHL VCEP BP2 rule requires either a phase-confirmed trans observation with a known pathogenic variant and absence of congenital polycythemia, a homozygous observation without personal or family history of VHL disease or congenital polycythemia, or cis/unknown phase with three or more different pathogenic VHL variants for Strong BP2; Supporting BP2 requires cis or unknown phase with a pathogenic VHL variant. The case evidence bundle contains no affected-proband or family-level observation documenting a pathogenic variant in cis or trans with NM_000551.4:c.541G>A (p.Val181Ile), no confirmed homozygous observation, and no qualifying phase or congenital-polycythemia clinical context. BP2 therefore cannot be assigned from the available evidence.

Rule

VHL VCEP Version 1.1 BP2: Strong for phase-confirmed trans with a known pathogenic variant plus absence of congenital polycythemia, or qualifying homozygous/cis-or-unknown-phase observations; Supporting for cis or unknown phase with a pathogenic VHL variant.

Evidence considered

• The VHL VCEP Version 1.1 BP2 rule requires specified cis/trans or homozygous observations and, for the trans pathway, absence of congenital polycythemia.
• The case bundle provides population, ClinVar, computational, and gene-level information but does not provide a qualifying affected-proband or family-level allelic observation with phase for this variant.

Still missing

• No documented cis/trans phase with a known pathogenic variant.
• No documented homozygous observation with the required negative personal and family history.
• No documented congenital-polycythemia assessment for a potential trans observation.
• No documented observation in cis or unknown phase with a pathogenic VHL variant, including the number of distinct pathogenic VHL variants.

Sources

Reasoning

VHL VCEP (cspec v1.1) explicitly prohibits use of missense in-silico predictors (e.g. REVEL, BayesDel) for BP4 in this gene, citing lack of benign controls and poor accuracy: 'missense predictors should not be used to assign the BP4 evidence code.' The only permitted BP4 path is splicing: concordance of SpliceAI (<=0.1) AND VarSeak (Class 1 or Class 2). SpliceAI max delta = 0.014 satisfies the SpliceAI <=0.1 arm, but no VarSeak result is present anywhere in the case bundle, so the required two-tool concordance cannot be verified. BayesDel score (0.021) exists in source data but has no verified published calibration threshold available to this pipeline and cannot be used for BP4 regardless of the VCEP's missense-predictor prohibition, which independently rules out its use here.

Rule

BP4_supporting (VHL VCEP cspec v1.1)

Evidence considered

• VHL VCEP cspec v1.1 BP4 rule: 'Due to the lack of benign variants, and the drop in classification accuracy for benign VHL variants, missense predictors should not be used to assign the BP4 evidence code. BP4 can be applied to assess lack of splicing impact, with concordance of Splice AI (<=0.1) and VarSeak (Class 1 or Class 2).'
• SpliceAI Lookup: max delta score = 0.014, satisfying the VCEP's <=0.1 SpliceAI arm of the BP4 splice-path concordance requirement.
• No VarSeak record found anywhere in review_manifest.json, source_registry.json, prefetch.json, evidence.json, evidence_brief.json, or vcep_materials.json for this case.

Still missing

• VarSeak splicing classification is not present in this case's evidence bundle; the VCEP requires concordance of SpliceAI (<=0.1) AND VarSeak (Class 1 or Class 2) to apply BP4 via the splice path, and only the SpliceAI half of that pair is available.
• Missense predictor path (REVEL/BayesDel) is explicitly disallowed for BP4 by the VHL VCEP specification itself, independent of predictor calibration status.

Sources

Reasoning

No qualifying co-occurrence evidence is available. The VHL Expert Panel requires at least two co-occurrences with pathogenic variants in another highly penetrant gene that fully explains the individual's phenotype, with age, tumor type, sex, and personal and family history showing no overlap with VHL features.

Rule

VHL VCEP v1.1 BP5: at least two qualifying co-occurrences with pathogenic variants in a different gene that fully explain the phenotype, with no overlapping VHL features after clinical-context review.

Evidence considered

• The case bundle contains no co-occurrence, alternate molecular diagnosis, phenotype, or family-history dataset that meets the VHL-specific BP5 requirements.

Still missing

• At least two independently documented co-occurrences with a pathogenic variant in a different highly penetrant gene.
• For each co-occurrence, phenotype, age, tumor type, sex, and personal/family history demonstrating that the alternate molecular diagnosis fully explains the phenotype without overlap with VHL-associated features.

Sources