Discussion

This Delphi consensus provides a comprehensive framework for the clinical use of PRRT in well-differentiated NETs, integrating evidence from randomized trials, real-world studies, and expert opinion. Importantly, this consensus should be interpreted in light of its intended scope. By organizing the recommendations into key domains—patient selection, imaging strategies, treatment administration, response assessment, and follow-up—this work seeks to harmonize practice in an area of considerable clinical heterogeneity. The Delphi process was conceived as a pragmatic, real-world, practice-oriented initiative aimed at harmonizing day-to-day clinical management of PRRT across centers, rather than redefining formal indications or addressing non-standard or investigational strategies. The consensus builds on pivotal randomized studies, recent phase II and III trials, and registry-based cohorts, providing guidance for established indications as well as for emerging scenarios, such as pheochromocytomas and paragangliomas.

Rather than establishing new formal indications, these recommendations should be regarded as supportive guidance in selected situations, particularly in bronchopulmonary NETs, pheochromocytomas/paragangliomas, and well-differentiated grade 3 tumors (Ki-67 ≤ 55%). This positioning reflects routine clinical practice in specialized centers rather than an expansion of approved indications. In the latter group, although the NETTER-2 trial demonstrated significant benefit of [177Lu]Lu-DOTA-TATE in untreated G2–3 GEP-NETs with Ki-67 up to 55% [9], regulatory approval by the EMA and FDA is still pending, and treatment should only be considered after individualized multidisciplinary discussion.

Accurate patient selection is fundamental to maximize the benefit–risk ratio of PRRT. Patients with well-differentiated GEP-NETs (grades 1–3, Ki-67 ≤ 55%), unresectable or metastatic, and with confirmed SSTR expression are considered potential candidates. The NETTER-1 trial demonstrated a median PFS of 28.4 months with [177Lu]Lu-DOTA-TATE compared with 8.4 months with high-dose octreotide, with an ORR of 18% vs 3% and a clinically relevant OS advantage despite crossover [8, 17]. More recently, the NETTER-2 trial showed that in previously untreated G2–3 GEP-NETs (Ki-67 10–55%), first-line PRRT achieved a median PFS of 22.8 vs 8.5 months and an ORR of 43% vs 9% [9]. In the COMPETE phase III trial, [177Lu]Lu-edotreotide demonstrated superiority over everolimus in G1–2 GEP-NETs, with a median PFS of 21.3 vs 9.4 months and higher disease control rates [10]. In pancreatic NETs, the OCLURANDOM trial reported a median PFS of 20.7 vs 11.0 months with PRRT compared with sunitinib, with an ORR of 28% vs 10% and a more favorable safety profile [11]. Real-world cohorts, including the Erasmus MC experience, confirmed durable disease control, with a median PFS of 29 months, OS exceeding 60 months, and long-term safety characterized mainly by manageable hematologic and renal toxicity [12].

Beyond digestive primary NETs, PRRT has shown clinically relevant activity in other NET subtypes. In the Erasmus MC series, patients with bronchopulmonary NETs achieved a median PFS of 21.6 months and OS of 52.3 months, confirming durable disease control beyond the GEP setting [12]. Similarly, the SEPTRALU registry broadened real-world evidence across multiple NET subtypes: in this cohort, bronchopulmonary tumors reached a median PFS of 17.6 months, while patients with pheochromocytomas and paragangliomas achieved a median PFS of 30.6 months and partial responses in more than 30% of cases, with favorable tolerability [13]. Prospective confirmation was provided by the phase II trial of Lin et al., where [177Lu]Lu-DOTA-TATE induced disease control in over 80% of progressive metastatic pheochromocytomas and paragangliomas [14]. These data support the cautious use of PRRT in selected non-GEP-NETs within a multidisciplinary framework, despite the absence of randomized phase III evidence in these populations. In parallel, the ongoing LEVEL trial is specifically evaluating PRRT in bronchopulmonary NETs, which will be instrumental to consolidate its role in these rare but clinically relevant entities.

Eligibility for PRRT should always be reviewed in a multidisciplinary tumor board. This approach, endorsed by ENETS and NCCN guidelines [5, 18], ensures adequate integration of imaging, optimal sequencing with other systemic therapies, and thorough evaluation of comorbidities and patient preferences, ultimately aiming to maximize therapeutic benefit.

Confirmation of somatostatin receptor expression is an essential prerequisite for PRRT, as its efficacy relies on radionuclide uptake and internalization by tumor cells. In NETTER-1, eligibility required a Krenning score ≥ 2 on [111In]-DTPA-octreotide scintigraphy, equivalent to uptake at least comparable to the liver [8], a criterion subsequently endorsed by international guidelines [19, 20]. NETTER-2 confirmed that robust SSTR expression remains mandatory in earlier treatment lines for G2–3 GEP-NETs (Ki-67 10–55%), using either scintigraphy or SSTR-PET to document uptake above liver background [9].

Over the past decade, PET imaging with radiolabeled somatostatin analogs has largely replaced scintigraphy as the preferred tool for assessing receptor expression. SSTR-PET offers higher sensitivity and specificity, superior spatial resolution, and semi-quantitative parameters such as SUVmax or tumor-to-liver ratio, which correlate with treatment outcomes [21–23]. Beyond diagnostic accuracy, it has demonstrated a direct impact on management, with therapeutic strategies modified in up to 40% of patients when used after inconclusive conventional imaging [24]. On this basis, the consensus identified SSTR-PET as the standard for confirming eligibility, recommending referral to specialized centers whenever local access is limited. In parallel,

In parallel, [18F]-FDG-PET/CT provides complementary prognostic information in selected scenarios. The panel deliberately positioned FDG-PET as a complementary prognostic tool rather than as an exclusion criterion for PRRT eligibility, in line with the design of pivotal PRRT trials. Dual-tracer imaging enables biological phenotyping, with high FDG uptake identifying tumors with more aggressive behavior and poorer outcomes [15, 25]. This is particularly relevant in higher-grade well-differentiated NETs, rapid clinical progression, or discordant findings between morphology and receptor imaging. Its use, however, should not be universal but restricted to cases with clinical or pathological features suggestive of unfavorable biology. Given the absence of validated thresholds defining FDG dominance and the lack of prospective evidence supporting FDG-driven treatment algorithms, the consensus deliberately avoided proposing rigid operational cut-offs. The systemic management of well-differentiated GEP-NETs with SSTR expression usually begins with long-acting SSAs as standard first-line therapy [6, 7]. In patients with disease progression, PRRT with [177Lu]Lu-DOTA-TATE is an established option supported by randomized and real-world evidence; however, differences in schedules and concomitant SSA use across trials warrant caution when extrapolating to routine care [8–13]. In daily practice, [177Lu]Lu-DOTA-TATE is generally administered as four cycles every 8 weeks at a fixed activity of 7.4 GBq, with continuation of background SSA for hormonal control and as part of the therapeutic scheme. This NETTER-1–aligned regimen has been consistently reproduced in real-world cohorts, supporting feasibility and safety [8, 12, 13].

In this context, close monitoring is essential to ensure safety and guide adjustments. Laboratory testing (complete blood count, renal function, and liver profile) is recommended 2 weeks prior to the first cycle, before each subsequent cycle, and again 4–6 weeks after administration. After completing therapy, follow-up assessments at 3, 6, and 12 months and then annually are advised, with interval individualization according to tolerance, comorbidities, or unexpected abnormalities [26, 27]. Particular vigilance for hematological toxicity is warranted in high-risk patients, such as those older than 70 years, with baseline cytopenia, prior chemotherapy or radiotherapy, extensive bone metastases, or moderate renal impairment. Importantly, no comparative evidence supports prioritization of dose reduction, interval extension, or temporary treatment interruption over another; therefore, management strategies should be individualized rather than algorithmic. A careful risk–benefit assessment is mandatory in cases with severe renal dysfunction, impaired marrow reserve, or relevant hepatic dysfunction [26, 27].

Routine interim imaging during PRRT is not recommended as it rarely changes management [28]. Available data do not demonstrate a clear clinical benefit of systematic interim imaging, and early assessments may be confounded by delayed responses or pseudoprogression. It may be considered in cases of clinical deterioration, impaired liver function, or high-risk disease features (e.g., higher grade, large tumor burden, or rapid progression to prior therapies), balancing the need for decision-making against the risk of misinterpretation [29, 30].

After completion of PRRT, a structured follow-up is essential to evaluate treatment response, detect disease progression, and monitor late toxicities. Morphological imaging remains the cornerstone of surveillance. Multiphasic contrast-enhanced CT or MRI allows accurate assessment of tumor size and vascularity, particularly relevant in slowly growing neoplasms such as NETs [31]. Response is usually evaluated according to RECIST 1.1, which, despite limitations in indolent tumors and atypical patterns such as pseudoprogression, remains the most reproducible tool in clinical practice and trials [32, 33]. The proposed imaging schedule reflects a pragmatic approach aligned with routine NET surveillance rather than a PRRT-specific evidence-based optimal timing. Most guidelines and expert recommendations suggest imaging at 3 and 6 months after the last cycle of PRRT, followed by surveillance every 6 months, with interval adjustment according to clinical features, tumor aggressiveness, or patient-related factors [31].

Functional imaging with SSTR-targeted techniques, either PET or scintigraphy, provides complementary information. Although not recommended for routine follow-up, performing SSTR imaging 9–12 months post-therapy can establish a new baseline [28]. Thereafter, its use should be indication-driven rather than protocolized [24]. Changes in uptake intensity (SUVmax or visual grading) should not be used as stand-alone criteria, given variability in receptor expression and confounders such as concomitant SSA therapy [28]. In patients with aggressive disease biology or discordant findings, [18F]-FDG-PET can identify dedifferentiated clones with limited SSTR expression and provide prognostic information [15, 25]. Whenever possible, the same SSTR-based imaging modality should be used before and after PRRT to ensure consistency [30].

The role of circulating biomarkers is more limited. Chromogranin Ahas low sensitivity and specificity and is not recommended for routine PRRT response assessment [26]. In functioning tumors, hormone-specific markers remain clinically useful. In functioning mid-gut NETs, measurement of 5-HIAA in plasma or 24-h urine remains valuable for monitoring serotonin secretion, with decreases after therapy correlating with improved symptom control and reduced risk of carcinoid heart disease [26]. In functioning pancreatic tumors, determination of peptide hormones such as insulin, gastrin, glucagon, somatostatin, or VIP can also provide clinically meaningful information for monitoring and early relapse detection [3]. Notably, agreement among panelists was lower for statements related to follow-up imaging and non-specific biomarkers, reflecting ongoing heterogeneity in real-world practice and limitations of the available evidence.

The strengths of this consensus include its multidisciplinary composition, alignment with both randomized trials and real-world evidence, and the achievement of high agreement across domains. Limitations relate to reliance on expert opinion where data remain scarce, variability in access to specialized imaging and nuclear medicine resources, and the evolving nature of ongoing trials that may further refine indications and management strategies. Despite these challenges, the recommendations provide actionable, harmonized, and evidence-informed guidance to support the safe and effective use of PRRT in patients with well-differentiated, SSTR-positive NETs.

In summary, this Delphi consensus reflects deliberate, practice-oriented choices in areas where high-level evidence is lacking and where international guidelines appropriately remain non-prescriptive. Rather than issuing rigid or investigational recommendations, the panel sought to distinguish established standards from unresolved clinical questions and provide pragmatic guidance to support multidisciplinary decision-making in routine PRRT practice.