NM_016507.4:c.3052G>A

CDK12 · NP_057591.2:p.(Asp1018Asn) · NM_016507.4

GRCh37: chr17:37676297 G>A · GRCh38: chr17:39520044 G>A

Gene: CDK12 Transcript: NM_016507.4

Final call

VUS

PM2 supporting BP4 supporting

Variant details

Gene

CDK12

Transcript

NM_016507.4

Protein

NP_057591.2:p.(Asp1018Asn)

gnomAD AF

2.4165335507800446e-05 (v4.1)

ClinVar

OncoKB

Unknown Oncogenic Effect

Classification rationale

Interpretation summary

Generated evidence synthesis

NM_016507.4:c.3052G>A (p.Asp1018Asn) is a rare missense variant in CDK12 observed in gnomAD at extremely low frequency (v2.1: AF=0.00389%, 11/282,794 alleles; v4.1: AF=0.00242%, 39/1,613,882 alleles), supporting PM2 at supporting strength.

Multiple in silico tools predict no deleterious effect: REVEL score 0.24, BayesDel score -0.448, and SpliceAI max delta 0.00, meeting BP4 at supporting strength.

The variant is absent from ClinVar and no publications mention this specific variant, limiting clinical interpretation to population and computational data.

One supporting pathogenic criterion (PM2) and one supporting benign criterion (BP4) are in conflict. Under generic ACMG/AMP 2015 combination rules (PMID:25741868), this pattern does not reach the threshold for Likely Pathogenic or Likely Benign classification. The variant is interpreted as a Variant of Uncertain Significance (VUS).

Final determination: Generic ACMG/AMP 2015 fallback rules do not meet benign, likely benign, likely pathogenic, or pathogenic combination thresholds, so the variant is classified as Variant of Uncertain Significance.

Criteria assessment

ACMG/AMP criteria review

Criteria shown when status is available

All criteria require review: For research and educational purposes only.

Criterion	Status	Rationale	Evidence used
PVS1	N/A	PVS1 is reserved for null variants (nonsense, frameshift, canonical ±1,2 splice, initiation codon, exon-level deletions). NM_016507.4:c.3052G>A is a missense variant (p.Asp1018Asn) and does not fall into any of the generic PVS1 null-variant buckets per ClinGen SVI recommendations (PMC6185798).	pvs1_generic_framework pvs1_variant_assessment
PS1	Not met	PS1 requires a different nucleotide change resulting in the same amino acid substitution (D1018N) with a known pathogenic classification. No such comparator variant exists in ClinVar or the literature for this residue.	clinvar
PS2	Not met	PS2 requires a de novo observation (maternity and paternity confirmed). No de novo data are available for this variant in any source.
PS3	Not met	PS3 requires well-established in vitro or in vivo functional studies supporting a damaging effect. No variant-specific functional data exist for NM_016507.4:c.3052G>A. OncoKB reports Unknown Oncogenic Effect with no reviewed functional evidence. No publications mention this variant. In silico predictions (REVEL, BayesDel, SpliceAI) do not constitute functional evidence for PS3.	oncokb
PS4	Not met	PS4 requires a significantly increased prevalence of the variant in affected individuals compared to controls. No case-control or prevalence data are available for this variant. The variant is absent from ClinVar and no publications report its observation in affected individuals.
PS5	Not met	PS5 requires a different nucleotide change at the same codon predicted to produce the same missense change, classified as pathogenic by a reputable source. No pathogenic variant at codon 1018 (alternate nucleotide) has been identified in ClinVar.	clinvar pm5_candidates
PM1	Not met	PM1 requires the variant to be located in a mutational hotspot or critical well-established functional domain. Residue D1018 lies within the CDK12 kinase domain but is not flagged as a statistically significant hotspot by cancerhotspots.org, and no variant-specific literature evidence establishes this exact residue as a critical functional domain. General kinase domain membership alone, without residue-level hotspot significance, is insufficient for PM1 under generic ACMG.
PM2	Met	This variant is present in gnomAD population databases at extremely low frequency: v2.1 AF=0.00389% (11/282,794 alleles) and v4.1 AF=0.00242% (39/1,613,882 alleles), with no homozygotes observed. Both frequencies fall well below the 0.1% threshold for PM2 under generic ACMG. Absent from ClinVar, consistent with a rare variant.	gnomad_v2 gnomad_v4 clinvar
PM5	Not met	PM5 requires a different pathogenic missense variant at the same codon (Asp1018). PM5 candidate harvesting returned no same-residue candidates from ClinVar. No pathogenic variant at codon 1018 with a different amino acid change has been identified.	pm5_candidates clinvar
PM6	Not met	PM6 requires a de novo observation without confirmed maternity and paternity. No de novo data are available for this variant in any source.
PP1	Not met	PP1 requires cosegregation of the variant with disease in multiple affected family members. No segregation data are available for this variant.
PP2	Not met	PP2 requires a missense variant in a gene with a low rate of benign missense variation and where missense variants are a common disease mechanism. While CDK12 has some evidence for a germline disease role in prostate cancer, no missense constraint metric (e.g., Z-score, o/e) is available to establish a low rate of benign missense variation, and the disease mechanism has not been clearly defined as missense-driven.
PP3	Not met	PP3 requires multiple lines of computational evidence supporting a deleterious effect. REVEL score 0.24 is below the typical pathogenic threshold (0.5), BayesDel score -0.448 is negative (benign prediction), and SpliceAI max delta score is 0.00 (no splice impact). No computational tool predicts a damaging effect.	revel bayesdel spliceai
PP4	Not met	PP4 requires the variant to be observed in a patient with a phenotype highly specific for the gene or with a well-defined syndrome. No patient phenotype data are available for this variant.
PP5	Not met	PP5 requires the variant to be classified as pathogenic by a reputable source (e.g., ClinVar expert panel, clinical diagnostic laboratory). This variant is absent from ClinVar entirely; no reputable source has classified it.	clinvar
BA1	Not met	BA1 requires an allele frequency >1% in a general population. The highest observed frequency in gnomAD is 0.033% (v4.1 Middle Eastern subpopulation), well below the 1% stand-alone benign threshold. The variant is too rare for BA1.	gnomad_v2 gnomad_v4
BS1	Not met	BS1 requires an allele frequency >0.3% in a general population. The highest observed frequency in gnomAD is 0.033% (v4.1 Middle Eastern subpopulation), below the 0.3% strong benign threshold.	gnomad_v2 gnomad_v4
BS2	Not met	BS2 requires observation in a healthy adult for a disorder with full penetrance expected at an early age. CDK12 germline variants are associated with adult-onset prostate cancer with reduced penetrance, so observation in healthy adults does not constitute strong evidence of benignity under the BS2 framework. Furthermore, no specific clinical data for the observed carriers are available.
BS3	Not met	BS3 requires well-established in vitro or in vivo functional studies showing no damaging effect. No functional data exist for this variant. In silico predictions (REVEL 0.24, BayesDel -0.448) are computational, not functional, and do not constitute BS3 evidence.
BS4	Not met	BS4 requires lack of segregation of the variant with disease in affected family members. No segregation data are available for this variant.
BP1	Not met	BP1 requires a missense variant in a gene where only truncating variants are the primary known disease mechanism. While CDK12 has evidence for loss-of-function as a disease mechanism, missense variants are also reported in cancer contexts (COSMIC: n=1), and there is insufficient evidence that only truncating variants drive CDK12-related germline disease.
BP2	Not met	BP2 requires observation of the variant in trans with a known pathogenic variant for a fully penetrant dominant disorder, or in cis with a pathogenic variant. No such data are available.
BP4	Met	BP4 requires multiple lines of computational evidence suggesting no impact on the gene product. REVEL score 0.24 (below pathogenic threshold), BayesDel score -0.448 (negative/benign prediction), and SpliceAI max delta 0.00 (no predicted splicing impact) collectively indicate no deleterious effect at the protein or splicing level.	revel bayesdel spliceai
BP5	Not met	BP5 requires an alternative molecular basis for disease to have been identified in the affected individual. No such data are available for this case.
BP6	Not met	BP6 requires the variant to be classified as benign by a reputable source (e.g., ClinVar expert panel). This variant is absent from ClinVar entirely; no reputable source has classified it as benign.	clinvar
BP7	N/A	BP7 applies to synonymous (silent) variants with no predicted splicing impact and low conservation. NM_016507.4:c.3052G>A is a missense variant (p.Asp1018Asn), not a synonymous variant.
BP3	N/A	BP3 applies to in-frame insertions/deletions in repetitive regions without a known function. This is a single-nucleotide substitution.
PM3	N/A	PM3 applies to recessive disorders with a pathogenic variant observed in trans. No trans-phase data are available.
PM4	N/A	PM4 applies to protein length changes from in-frame indels or stop-loss variants. This is a single-nucleotide missense substitution.

Disclaimer: The content and results provided by LYFE Sciences are for research and educational purposes only and must not be used as a substitute for professional medical judgment, diagnosis, or treatment. Always consult a qualified healthcare professional before making any clinical decisions.