Radiopharmaceuticals and theranostics

A radiopharmaceutical is three parts bolted together: a vector that recognises something on or in a tumour cell, a chelator or prosthetic group that holds a radionuclide securely, and the nuclide itself. The design insight the field is built on is that the first two can stay fixed while the third is swapped. Attach a positron emitter and the compound is an imaging agent that reports exactly where the vector goes; attach a particle emitter with the same coordination chemistry and it delivers dose to the same places. Gallium-68 with a 68-minute half-life paired to lutetium-177 with a 6.6-day half-life on a PSMA or somatostatin-receptor ligand is the canonical example — the scan is a direct measurement of the therapy’s biodistribution, which no oral drug can offer.

Two kinds of killing

Lutetium-177 emits beta particles with a maximum energy near 0.5 MeV and a mean range in tissue of well under a millimetre, crossing tens of cell diameters. That produces crossfire: a labelled cell irradiates its neighbours, so antigen-negative cells inside a positive tumour are still hit, and micro-heterogeneity of expression matters less than it would for an antibody-drug conjugate. But beta radiation is low linear energy transfer. Most of its damage is indirect, through water radiolysis, produces mainly single-strand breaks and base damage, and is repairable — and radiolysis needs oxygen, so hypoxic tumour cores are relatively protected.

Alpha emitters invert both properties. Actinium-225 and its daughters release several alpha particles with linear energy transfer around 100 keV/µm over a range of only 50–80 µm — a few cell diameters. The dense ionisation produces clustered double-strand breaks that cells repair poorly, and the effect is largely oxygen-independent. The cost is that the recoil energy of alpha decay far exceeds any chemical bond, so the daughters — francium-221, bismuth-213 — are released from the chelator and redistribute in the body, which is the origin of the salivary and renal toxicity seen with targeted alpha therapy.

What limits the dose

Not the tumour. Small targeted molecules are cleared renally and reabsorbed in the proximal tubule, so the kidney takes dose it never targeted; co-infusing basic amino acids blocks that reabsorption and is standard for peptide receptor therapy. PSMA ligands additionally concentrate in salivary and lacrimal glands, which express the target, and bone marrow sets the limit for haematological toxicity. These are on-target or physiologic, not impurity effects, and they cannot be engineered away without changing the vector.

An honest gap sits here. Approved regimens are dosed as fixed activity on a fixed schedule, not to a calculated absorbed dose in the organ at risk, even though the imaging pair makes patient-specific dosimetry technically feasible. Whether personalised dosimetry improves outcomes is still being tested; the current practice is a pragmatic compromise, not a settled scientific answer.

The decay clock

Everything else in pharmaceutical manufacturing is fixed by chemistry; here it is fixed by physics. A gallium-68 dose is made and used within hours, so production must sit near the patient. Lutetium-177 must be produced, radiolabelled, released and shipped inside days. Actinium-225 supply has been the field’s binding constraint because the historical source is decay of thorium-229 held in a handful of generators worldwide, which is why accelerator routes — proton irradiation of radium-226 among them — are being built.