Oncofusion

start / graph / ana-q199-menin-and-cdk9-each-carry-one-genome-wide-significant-common-variant-so-gwas-is-silent-on-on-target-liability-while-the-revumenib-label-lists-hyperparathyroidism

Analysismoderate confidence 3 of 5: One strong study, or several smaller consistent ones.

MEN1 and CDK9 each carry exactly one genome-wide-significant in-gene common variant and PSIP1 two, so GWAS says almost nothing about on-target liability - yet the revumenib label lists hyperparathyroidism, the organ MEN1 loss famously affects

Inherited variation in the genes these drugs block is almost silent, so population genetics cannot warn us about on-target side effects. The revumenib label nonetheless lists raised parathyroid hormone, which fits what losing menin does in people born without it.

The same thing, for a scientist

In-gene genome-wide-significant SNPs: MEN1 1 (trait: Urate levels), CDK9 1 (Height), PSIP1 2 (Body mass index, Whole body fat mass (UKB data field 23100)); length-matched percentiles 42.4, 42.6 and 24.2. The corpus already records MEN1's strong loss-of-function constraint (gnomAD pLI ~1.0), the usual reason a gene shows no common-variant signal, and the revumenib prescribing information lists hyperparathyroidism.

Go deeper
What a quiet locus does and does not mean

Common variants are small regulatory nudges; they say nothing about what happens when a drug removes most of a protein's function. Genes under strong constraint show almost no common variation precisely because large changes are harmful, so an empty GWAS locus is the expected signature of an important gene, not of a safe target.

Facts and figures
scriptanalyses/q199-germline-genetics-of-the-fusion-pathway/run.py
men1
{
  "gene": "MEN1",
  "len_kb": 7.8,
  "snps": 1,
  "rows": 1,
  "traits": 1,
  "n_length_matched": 2294,
  "pct_length_matched": 42.4,
  "context": {
    "intron_variant": 1
  }
}
cdk9
{
  "gene": "CDK9",
  "len_kb": 7.6,
  "snps": 1,
  "rows": 1,
  "traits": 1,
  "n_length_matched": 2262,
  "pct_length_matched": 42.6,
  "context": {
    "3_prime_UTR_variant": 1
  }
}
psip1
{
  "gene": "PSIP1",
  "len_kb": 49.2,
  "snps": 2,
  "rows": 3,
  "traits": 2,
  "n_length_matched": 3608,
  "pct_length_matched": 24.2,
  "context": {
    "intron_variant": 3
  }
}
top traits
{
  "MEN1": {
    "Urate levels": 1
  },
  "CDK9": {
    "Height": 1
  },
  "PSIP1": {
    "Body mass index": 2,
    "Whole body fat mass (UKB data field 23100)": 1
  }
}
How this could bite us back
  • Do not treat a silent germline locus as reassurance. The label's hyperparathyroidism and the MEN1 syndrome both point to endocrine monitoring, especially in children exposed for years rather than months.
Still open
  1. Is parathyroid hormone or calcium tracked at all in the paediatric and infant menin-inhibitor protocols? The corpus records that nobody has measured the parathyroid under a menin inhibitor in tissue.
Every source behind this entry (2)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    In the GWAS Catalog, MEN1 has 1 genome-wide-significant in-gene SNP, CDK9 1 and PSIP1 2, at the 42.4th, 42.6th and 24.2th percentiles of genes of similar length.

    What this evidence cannot show

    MEN1's single hit is Urate levels; MEN1 sits next to other loci on 11q13, and the Catalog maps by proximity, so even that one hit may belong to a neighbour (not tested). MEN1 and CDK9 are both ~7.7 kb genes, giving little target for variants.

    Evidence that argues against this (1)
    • The revumenib US prescribing information lists hyperparathyroidism as an adverse reaction (intact PTH increased in 34% and corrected calcium increased in 15% of 241 patients), so the absence of a GWAS signal at MEN1 coexists with a real on-target endocrine effect. GWAS silence must not be read as safety. US Food and Drug Administration 2026
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    MEN1 shows strong loss-of-function constraint (gnomAD pLI about 1.0, LOEUF 0.204), the usual explanation for a gene with almost no common-variant association signal.

    What this evidence cannot show

    Taken from this corpus's existing gene-men1 entry (DepMap/gnomAD/HPA table), not recomputed here.

    Evidence that argues against this (1)
    • Constraint explains the quiet GWAS locus but does not predict a drug's tolerability: the drug effect is partial and reversible, whereas constraint reflects lifelong heterozygous loss. Heterozygous germline MEN1 loss causes multiple endocrine neoplasia type 1 with penetrance above 50% by age 20, which is the on-target concern for long exposure in children. Giusti F et al. 2022
Connected to