Peptide Receptor Radionuclide Therapy (PRRT), Explained

Nuclear medicine

Peptide receptor radionuclide therapy — PRRT — is the clearest demonstration anywhere in medicine that a peptide can be used as a delivery vehicle rather than as a drug. The peptide itself does almost nothing therapeutic. Its job is to find a receptor that a tumour has too many of, sit on it, and carry a radioactive metal close enough to break DNA. Because the same peptide can carry either an imaging isotope or a therapeutic one, the scan that decides whether a patient is eligible uses the same molecule as the treatment. That pairing has a name — theranostics — and PRRT is where it was proved.

Peptide receptor radionuclide therapy: three modules bolted together

Every PRRT agent is built from three interchangeable parts, and it is worth learning them separately because almost all the innovation in the field is a swap at one of the three positions.

• The targeting peptide. For neuroendocrine tumours this is dotatate, described in the FDA label as “a cyclic 8 amino acid peptide,” sequence H-D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr-OH, closed by one disulfide bond between the two cysteines, molecular weight 1435.6 Da. It is a close relative of octreotide, itself a synthetic analogue of somatostatin built to resist the peptidases that destroy the natural hormone in minutes.
• The chelator. DOTA is a twelve-membered macrocyclic cage — 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid — covalently attached to the peptide. Its function is to grip a metal ion so tightly that it does not fall off in circulation, because a free radiometal goes wherever its element goes: bone, liver, kidney. The historical progression makes the point neatly. The first agents used an acyclic chelator, DTPA, with indium-111; the field moved to the DOTA cage when it started carrying yttrium-90 and lutetium-177.
• The metal. This is the part that is swapped to change the job. Gallium-68 or copper-64 make the construct visible on a PET scanner. Lutetium-177, yttrium-90 or actinium-225 make it therapeutic. The peptide and the cage are unchanged.

Why somatostatin receptor 2 is the target

Somatostatin signals through five receptor subtypes, SSTR1 to SSTR5. Dotatate is not evenly distributed among them: the FDA labels for Lutathera, NETSPOT and Detectnet all state that the compound binds “with highest affinity for subtype 2 somatostatin receptors (SSTR2),” and that once bound it is internalised.

Neuroendocrine tumours happen to over-express exactly that subtype. In an immunohistochemical study of 412 samples from 165 patients with gastroenteropancreatic neuroendocrine neoplasms, SSTR2A was present in 85% of primary tumours and at high intensity, well ahead of SSTR5 (23%), SSTR3 (10%), SSTR1 (9%) and SSTR4 (4%). Primary tumours expressed more than metastases, and loss of SSTR2A was associated with worse outcomes.

The receptor is not exclusive to tumours, which is why a scan looks the way it does. Gallium-68 dotatate distributes to every SSTR2-expressing organ — pituitary, thyroid, spleen, adrenals, kidneys, pancreas, prostate, liver and salivary glands — with no uptake in cerebral cortex or heart. The spleen is one of the brightest normal structures on the image.

How eligibility is actually judged

The working tool is the Krenning score, a five-point visual scale originally developed for indium-111 pentetreotide scintigraphy and carried over to PET: 0 = no uptake, 1 = very low, 2 = uptake no more than the liver, 3 = greater than the liver, 4 = greater than the spleen. For lesions larger than 2 cm, guidance is that PRRT should be considered at a score of 3 or 4.

Quantitative SUV cut-offs exist but are less robust than they look. SUV correlates with receptor density only at low values before becoming non-linear, which means it underestimates expression in the most receptor-rich tumours — possibly through receptor saturation. Proposed thresholds of SUVmax 17.9 and 16.5 have both been published, but they do not transfer cleanly between scanners; a tumour-to-liver ratio of 2.2 has been suggested as an alternative.

Choosing the metal: half-life, emission type and range

The radiometal decides three things at once: how long the agent stays hot, what kind of particle it emits, and how far that particle travels in tissue. Range is the one with the biggest clinical consequence.

Lutetium-177 emerged as the therapeutic workhorse because its numbers sit in a useful middle. Its 6.647-day half-life is long enough to survive shipping and to keep irradiating a tumour for days, and its beta particles have a maximum tissue range of 2.2 mm with a mean of 0.67 mm. Yttrium-90’s far more energetic beta travels around 11 mm, which delivers more dose but also more collateral irradiation — and the toxicity data reflect that. In one comparison, 90Y-DOTATOC produced transient grade 3/4 haematologic toxicity in 12.8% of patients and permanent grade 4/5 renal toxicity in 9.2%, which is why the Rotterdam group concluded lutetium-177 is “the radionuclide of first choice.”

The imaging side follows the same logic in reverse. Gallium-68’s 68-minute half-life means it must be eluted from an on-site germanium-68 generator, labelled at 95 °C for seven minutes and used within about four hours. Copper-64’s 12.7 hours is long enough to make and ship centrally — a logistical difference, not a biological one.

What the randomised evidence actually shows

NETTER-1 is the trial that established PRRT. It randomised 229 patients with progressive midgut neuroendocrine tumours to four doses of 7.4 GBq (200 mCi) lutetium-177 dotatate every eight weeks — a cumulative 29.6 GBq — plus octreotide LAR 30 mg, versus high-dose octreotide LAR 60 mg alone. The progression-free survival result was emphatic and is not in dispute.

The overall survival story is more interesting, and it is where most secondary summaries go wrong. At the 2017 interim analysis there were 14 deaths versus 26, hazard ratio 0.40, p = 0.004 — a number that circulated widely. But the O’Brien–Fleming stopping threshold for that first interim look was p = 0.000085. The signal never came close to crossing it. When the final analysis was performed in January 2021, after 142 deaths and a median follow-up of 76.3 months, median overall survival was 48.0 months versus 36.3 months, hazard ratio 0.84, two-sided p = 0.30. The authors describe the 11.7-month difference as possibly clinically relevant while stating plainly that the endpoint was not met, and note two honest caveats: 36% of control patients crossed over to PRRT, and the proportional-hazards assumption was not satisfied.

NETTER-2, reported in 2024, moved the treatment to first line and to harder disease — newly diagnosed, higher-grade tumours with Ki-67 between 10% and 55%, more than half of them pancreatic in origin. It was described as the first trial of first-line radioligand therapy in any solid tumour. Median progression-free survival was 22.8 versus 8.5 months with a stratified hazard ratio of 0.276, and the objective response rate was 43.0% versus 9.3%. Overall survival was not mature at the primary analysis and no figure should be quoted for it.

The real-world cohort behind the trials

Erasmus MC in Rotterdam treated 1,214 patients with lutetium-177 dotatate between January 2000 and January 2015, and its long follow-up is the best available picture of what routine practice looks like. In 443 evaluable patients, the objective response rate was 39% (2% complete, 37% partial), median progression-free survival 29 months, median time to progression 36 months and median overall survival 63 months — 71 months for pancreatic primaries, 60 for midgut, 52 for bronchial. For context, the same paper lists everolimus at 11.0 months progression-free survival and sunitinib at 11.4.

Protecting the kidneys and the marrow

Two organs limit PRRT, and the field has a specific fix for each.

The kidneys. Radiolabelled peptides are filtered and then partially reabsorbed by the proximal tubule, concentrating radioactivity exactly where you do not want it. The countermeasure is an infusion of positively charged amino acids that compete for that reabsorption. The FDA label states the purpose directly: the amino acid solution is given “to decrease the reabsorption of lutetium Lu 177 dotatate through the proximal tubules and decrease the radiation dose to the kidneys.” It contains 18–25 g each of L-lysine HCl and L-arginine HCl in 1–2 litres, starts 30 minutes before the radiopharmaceutical and continues for at least three hours after — and, notably, the amino acid dose is not reduced even if the radiopharmaceutical dose is. It works: in the Rotterdam cohort, grade 3/4 creatinine toxicity occurred in 2 of 581 patients (0.3%), and the authors concluded the kidney is no longer the dose-limiting organ. One side effect of the fix is worth knowing — in NETTER-1, roughly two-thirds of nausea and three-quarters of vomiting were attributed to the amino acid infusion rather than to the radiopharmaceutical, and resolved when it finished.

The bone marrow. Cytopenias are common and mostly manageable. In NETTER-1, anaemia occurred in 81% of the treated arm (grade 3/4 in none), thrombocytopenia in 53% (1% grade 3/4) and neutropenia in 26% (3% grade 3/4), with the platelet nadir at a median of 5.1 months after the first dose and recovery to baseline in 68% of affected patients over a median of two months. The serious long-term concern is secondary myeloid disease. With 76 months of follow-up in NETTER-1, myelodysplastic syndrome occurred in 2.3% of treated patients and none of the controls; in the larger ERASMUS safety dataset cited by FDA, 16 patients (2.0%) developed MDS and 4 (0.5%) acute leukaemia, with median times to onset of 29 and 55 months respectively. The label’s nonclinical section states the underlying point without hedging: carcinogenicity studies were not conducted, “however, radiation is a carcinogen and mutagen.”

Radiation safety. Elimination is renal and fast — 44% of administered activity in urine within 5 hours, 65% within 48 hours, over 99% within 14 days — but radiation remains detectable in urine for up to 30 days. Patient-release and household-contact instructions follow national regulatory guidance rather than the drug label alone.

Where the field is going

• Alpha emitters. Actinium-225 delivers four net alpha particles per decay over a range of 40–100 micrometres at roughly 400 times the linear energy transfer of a beta particle — in principle, far more lethal damage confined to the targeted cell and its immediate neighbours. ACTION-1 (NCT05477576) is a phase 1b/3 trial of 225Ac-DOTATATE in patients whose disease has progressed after lutetium-177 therapy.
• Antagonists instead of agonists. This is the field’s most counterintuitive result. Receptor antagonists do not internalise, which should make them useless for delivery — yet they bind a much larger population of sites. Scatchard analysis found more than ten times the number of binding sites for an antagonist than for an agonist, and in a four-patient pilot, 177Lu-DOTA-JR11 delivered a 1.7 to 10.6 times higher tumour dose than 177Lu-dotatate with better tumour-to-kidney and tumour-to-marrow ratios. The advantage is not free: in a 40-patient phase I/II study of 177Lu-satoreotide tetraxetan, grade 3 or worse treatment-related adverse events occurred in 42.5% and two patients developed treatment-related myeloid neoplasms, alongside a disease control rate of 94.7%.
• New targets entirely. Fibroblast activation protein is expressed by cancer-associated fibroblasts across many epithelial tumours and is largely absent from normal adult tissue. FAP-2286 is a cyclic peptide on the same DOTA cage, and the LuMIERE study uses a gallium-68-labelled version to select patients for the lutetium-177 version — the identical theranostic logic, moved from the tumour cell to the stroma around it.

The adjacent case that is not a peptide

Prostate-specific membrane antigen therapy is often filed under the same heading, and the clinical logic is identical, but the chemistry is not. Lutetium Lu 177 vipivotide tetraxetan (Pluvicto) is described in its FDA label as “a PSMA-binding ligand bound to a DOTA chelator” with molecular formula C₄₉H₆₈¹⁷⁷LuN₉O₁₆ and a mass of 1216.06 g/mol — a urea-based small molecule, not an amino-acid chain. It is a radioligand therapy but not, strictly, a peptide receptor radionuclide therapy.

The VISION trial randomised 831 patients 2:1 to 7.4 GBq every six weeks for up to six doses plus standard of care, versus standard of care alone. Median radiographic progression-free survival was 8.7 versus 3.4 months (HR 0.40) and median overall survival 15.3 versus 11.3 months (HR 0.62, p < 0.001). FDA’s own review adds a caveat rarely quoted alongside those numbers: interpretation of the rPFS effect was “limited due to a high degree of censoring from early drop out in the control arm.”

Frequently asked questions

Is PRRT chemotherapy?

No. Chemotherapy is a cytotoxic drug distributed throughout the body; PRRT is a targeting molecule carrying a radiation source that acts only where it lands. The peptide itself has essentially no anti-tumour activity — remove the radiometal and you have a somatostatin analogue.

Why does the same peptide appear in both the scan and the treatment?

Because the biology that makes a tumour visible is the biology that makes it treatable. If a gallium-68-labelled dotatate scan shows uptake greater than the liver, the same construct carrying lutetium-177 will go to the same places. That is the whole meaning of a theranostic pair, and it is why patient selection for PRRT is imaging-based rather than biopsy-based.

Did PRRT improve survival in NETTER-1?

Progression-free survival, decisively. Overall survival, not to statistical significance: 48.0 versus 36.3 months, hazard ratio 0.84, p = 0.30 at final analysis, with 36% of the control arm having crossed over to PRRT. Both statements are true simultaneously, and any summary that reports only one of them is incomplete.

Does any of this apply to research peptides handled in a laboratory?

Only conceptually. PRRT agents are radiopharmaceuticals prepared and administered under nuclear-medicine licensing, and nothing about their handling generalises to ordinary peptide work. What does transfer is the design principle: receptor affinity, chelator stability and clearance route determine where a peptide ends up, whether or not it carries a radioactive payload.

References

1. Strosberg J, et al. Phase 3 trial of 177Lu-dotatate for midgut neuroendocrine tumors (NETTER-1). N Engl J Med 2017;376:125–135. nejm.org
2. Strosberg J, et al. 177Lu-dotatate plus long-acting octreotide versus high-dose long-acting octreotide: final overall survival and long-term safety (NETTER-1). Lancet Oncol 2021;22:1752–1763. thelancet.com
3. Singh S, et al. NETTER-2 primary analysis, first-line 177Lu-DOTATATE in higher-grade GEP-NET. Abstract, 2024. endocrine-abstracts.org · registry NCT03972488
4. Brabander T, et al. Long-term efficacy, survival and safety of [177Lu-DOTA0,Tyr3]octreotate in patients with gastroenteropancreatic and bronchial neuroendocrine tumors. Clin Cancer Res 2017;23:4617–4624. aacrjournals.org
5. US FDA. LUTATHERA (lutetium Lu 177 dotatate) prescribing information, NDA 208700. accessdata.fda.gov
6. US FDA. NETSPOT (kit for gallium Ga 68 dotatate injection) prescribing information, NDA 208547. accessdata.fda.gov
7. US FDA. DETECTNET (copper Cu 64 dotatate injection) prescribing information, NDA 213227. accessdata.fda.gov
8. US FDA. PLUVICTO (lutetium Lu 177 vipivotide tetraxetan) prescribing information, NDA 215833 — includes the VISION efficacy tables. accessdata.fda.gov
9. European Medicines Agency. Lutathera: EPAR summary for the public (EMEA/H/C/004123). ema.europa.eu
10. Hope TA, Mittra E, et al. Somatostatin receptor imaging and theranostics: current practice and future prospects. J Nucl Med 2021;62:1323. jnm.snmjournals.org
11. Lupp A, et al. Different somatostatin and CXCR4 chemokine receptor expression in gastroenteropancreatic neuroendocrine neoplasms depending on their origin. Sci Rep 2019;9:4339. ncbi.nlm.nih.gov
12. Wild D, et al. Comparison of somatostatin receptor agonist and antagonist for peptide receptor radionuclide therapy: a pilot study. J Nucl Med 2014;55:1248–1252. jnm.snmjournals.org
13. Hicks RJ, et al. A phase I/II study of [177Lu]Lu-satoreotide tetraxetan in patients with progressive neuroendocrine tumours. Eur J Nucl Med Mol Imaging 2024;51:183. ncbi.nlm.nih.gov

Informational only — not medical advice. VialHelp does not sell peptides and does not recommend any product, treatment or dose. Radiopharmaceuticals are prescribed and administered only by licensed nuclear-medicine services. Discuss any medical question with a qualified healthcare professional. Intended for readers 21+.