You have probably already seen this patient¶
Heterozygous SAMHD1 disease has never been formally described, so there is no pattern in the literature to match against. That is the whole problem: the presentation is multi-system, each system is individually unremarkable, and no single specialty ever sees the whole thing.
Below: the questions clinicians actually ask us, each answered in a sentence and linked straight into the mechanism model — set to the view that answers it, with the citation and evidence grade attached.
Not a diagnostic criterion
This is a research framework, not a diagnostic criterion or a treatment protocol. Nothing here is medical advice. Claims are labelled established, hypothesised or predicted throughout so you can weigh them yourself.
The pattern worth a second look¶
-
Post-viral fatigue that never resolves
ME/CFS with post-exertional malaise, often dating from one identifiable infection.
-
Metabolic disease at a normal BMI
Insulin resistance and steatohepatitis that do not track with weight or diet.
-
Inflammatory arthritis without the usual serology
Enthesitis-pattern disease that does not fit cleanly into any named diagnosis.
-
Connective-tissue failure
Tendon and fascial problems out of proportion to activity or age.
-
Recurrent pre-eclampsia and early delivery
Especially where it recurs on the same side of a family across generations.
-
A family history nobody has joined up
The single strongest signal. Three or more of the above, in relatives, treated separately.
The questions we get asked¶
Each link opens the atlas at the exact view — the route traced, the compartment entered, or the treated state rendered. No identifiers travel in the URL.
It's described as recessive. How can a heterozygote be affected at all?¶
SAMHD1 only works as a tetramer. One variant copy does not halve the enzyme — it produces tetramers of mixed composition, so anything requiring all four subunits intact falls far below what gene dosage predicts. Biallelic loss causes Aicardi-Goutières syndrome; the single-copy state is a different, milder, undescribed phenotype.
/function-atlas/#tour=tetramer&step=1
Where does chronic fatigue come from mechanistically?¶
Mitochondrial DNA escapes into the cytosol, cGAS-STING raises type I interferon, and sustained JAK-STAT signalling blocks mitophagy — so damaged mitochondria accumulate instead of being cleared, and the bioenergetic deficit becomes self-sustaining.
My patient responded dramatically to a JAK inhibitor, then plateaued. Why?¶
This is the model's sharpest prediction. Two mechanistically independent axes run from the same root cause: the interferon axis, which JAK inhibition suppresses almost completely, and a nucleotide/NLRP3 axis, which it does not touch. The residual fatigue and metabolic floor is that second axis. Relapse within 24–48 hours of stopping is consistent with a re-priming node on that timescale.
Metabolic disease at a normal weight — is that really the same thing?¶
Excess dNTPs are catabolised to uric acid and MSU crystals, an independent NLRP3 input; a 2026 Science study showed SAMHD1 loss drives insulin resistance and steatohepatitis at normal body weight through this route. Diet-resistant metabolic disease is a predicted feature, not a coincidence.
Recurrent pre-eclampsia in the same family — related?¶
Possibly, and it cuts both ways. A raised baseline interferon tone may support implantation early and harm the placenta late. The pattern recurring across generations is itself indirect evidence of elevated tone throughout the kindred — and a caution that interferon-suppressing therapy needs careful timing in anyone who might become pregnant. Hypothesised
How much of this is actually demonstrated?¶
Less than the diagram implies, and we would rather you see that yourself than take our word for it. Set the evidence filter to demonstrated-only and most of the picture disappears; what remains is the part that can be argued from directly. Every node and every arrow carries its own grade.
What the model predicts you would find¶
- A chronic, moderate-amplitude interferon signature — not the high-amplitude picture of AGS.
- Inflammatory markers that respond to JAK inhibition while fatigue and metabolic measures do not.
- Urate at the high end, or gout-pattern events, without the usual risk factors.
- On exome or genome data already in hand:
NM_015474.3:c.1693G>A— often filtered out as a heterozygous VUS in a recessive gene.
If you have seen a case¶
The single most useful thing anyone can do right now is add a second family to the record. There is no published heterozygous case series; one exists in draft, built from a five-generation kindred, and a second unrelated pedigree would change what it is possible to claim.
We are also glad to be told we are wrong. If a link above opens a view you think is unsupported, say which arrow and why — the grades exist to be challenged.