The off switch

Most hormones say a specific thing to a specific tissue. Somatostatin says one thing to almost everything: stop. It shuts down growth hormone and TSH from the pituitary, insulin and glucagon from the pancreatic islet, and a long roster of gut hormones and digestive secretions besides. It was discovered in 1973 as the factor in hypothalamic extract that blocked growth-hormone release — first named somatotropin-release-inhibiting factor — and the surprise was how far its reach extended beyond that one job (Brazeau, Guillemin et al., Science, 1973).

In the growth-hormone axis, somatostatin is the brake. The companion to this piece walked the two accelerators that raise GH — GHRH and ghrelin — and noted that a GH pulse is what happens when those rise while somatostatin falls. This is the brake’s own story, and it is an unusual one, because somatostatin is the rare hormone whose entire drug strategy is inverted: you do not supply it to boost a signal, you engineer a longer-lasting version to hold a system down. The whole art is in choosing how broadly it presses.

Native somatostatin is a hopeless drug, for the same reason GLP-1 is: it vanishes. Its half-life in blood is about two minutes, so an infusion works only while it runs, and cutting it off can provoke a rebound surge of the very hormones it was suppressing. Add to that its indiscriminate reach — a molecule that silences GH, insulin, glucagon and gut secretion all at once is a blunt instrument — and the native peptide is useful for little but a short hospital infusion.

The fix was the same one the incretin field would later reach for in the half-life engineering of semaglutide: don’t administer the fragile native hormone, build an analog that lasts. Octreotide and lanreotide are small cyclic peptides that reproduce somatostatin’s inhibition but survive long enough to matter, and in depot form they release over roughly a month — a single injection standing in for a hormone that natively lasts minutes. The drug is the brake, made durable.

Here is the key move. Native somatostatin binds all five receptor subtypes, SSTR1 through SSTR5, which is exactly why it shuts down everything — and exactly why its breadth is a liability. The analogs narrow it. Octreotide and lanreotide preferentially grip SSTR2 (with some SSTR5), and that narrower focus is what turns a universal off-switch into a targeted one: enough to silence growth hormone in acromegaly and the hormone oversecretion of neuroendocrine tumors, without flattening every other secretion in the body.

So the question that decides what a somatostatin analog does is not “how strong” but “which subtypes.” SSTR2 governs the growth-hormone and most of the neuroendocrine-tumor story. SSTR5 matters in places SSTR2 does not reach — and, as the pancreas will show in a moment, carries a cost. Reading these drugs means reading their receptor profile, the same way reading an incretin co-agonist means reading which receptors it touches and in which direction.

The first job was secretion. In acromegaly — a pituitary tumor pouring out growth hormone — an SSTR2-preferring analog suppresses GH and brings IGF-1 down toward normal. In carcinoid syndrome, the flushing and diarrhea driven by a hormone-secreting tumor are calmed by the same mechanism. For years that was the whole indication: these were drugs that turned the volume down on a gland that would not stop talking.

Then came the second discovery, which is more interesting. The analogs do not merely quiet secretion; they slow the tumor itself. In the CLARINET trial, lanreotide roughly doubled progression-free survival against placebo in metastatic enteropancreatic neuroendocrine tumors — the median was not even reached on the drug, versus 18 months on placebo (Caplin et al., NEJM, 2014). The same SSTR2 that works as a secretion valve also acts as a growth brake on the tumor cell. That dual role is why SSTR2 is now used not only to treat these tumors but to find them — the receptor is the target for both imaging and radionuclide therapy.

If narrowing to SSTR2 made the analogs useful, the obvious next idea is to cover more subtypes — and pasireotide does, binding SSTR1, 2, 3 and 5. The payoff is real and specific. Cushing’s disease — a corticotroph tumor driving excess ACTH and cortisol — runs largely through SSTR5, which octreotide barely touches, so octreotide mostly fails there. Pasireotide’s strong SSTR5 engagement lets it succeed: in its phase 3 trial it lowered cortisol and controlled disease in a meaningful fraction of patients (Colao et al., NEJM, 2012).

But breadth is never free, and pasireotide pays for it in the pancreas. SSTR5 also sits on the insulin-secreting beta cell, so the same receptor activity that controls Cushing’s also throttles insulin — and in that trial the majority of patients developed hyperglycemia, many frankly diabetic. It is the cleanest lesson in the whole family: the extra receptor that makes the drug work in one disease is the extra receptor that disturbs glucose in every patient. More coverage is a different drug, not a strictly better one — the same truth the GIP paradox teaches from the agonist side.

Somatostatin is the inverse of almost everything else on this shelf. The incretins and the growth secretagogues are accelerators you supply; the natriuretic peptides are signals you protect rather than replace. Somatostatin is a brake you cannot usefully supply in native form, so medicine rebuilds it to last and then machines down which receptors it grips. Read that receptor fingerprint and the family resolves: why octreotide and lanreotide own acromegaly and neuroendocrine tumors, why pasireotide reaches Cushing’s at the price of blood sugar, and why the most inhibitory hormone in the body became a precision tool only once it was made less universal.

• Somatostatin is the universal inhibitor. It suppresses GH, TSH, insulin, glucagon and gut secretion through five receptor subtypes — the brake that opposes GHRH and ghrelin in the GH axis.
• The native hormone can't be a drug. A ~2-minute half-life and rebound hypersecretion make it useless outside a short infusion. The analogs are engineered to last weeks — the brake made durable.
• Receptor selectivity is the whole story. Native somatostatin hits all five SSTRs. Octreotide and lanreotide narrow to SSTR2 (+some 5), turning a universal off-switch into a targeted one.
• Anti-secretory and antiproliferative. They quiet the hormone (acromegaly GH, carcinoid flushing) and slow the tumor — CLARINET showed lanreotide roughly doubling PFS in neuroendocrine tumors. SSTR2 is both valve and growth brake.
• Pasireotide shows breadth has a price. Its broad profile (SSTR1/2/3/5) unlocks Cushing's via SSTR5 — but the same SSTR5 on the beta cell drives hyperglycemia in most patients. Broader is a different drug, not a better one.