The management of childhood malignancies has attained overall survival (OS) rates of up to 80% in many tumors [1]. This is partly due to the progress in treatment possibilities. In low- and middle-income countries (LMICs), the outcomes are less favorable mainly because treatment options are not as diverse as in high-income countries (HICs) [1]. Neuroblastoma is a tumor that, despite these advances, has only attained an OS rate of up to 60% in HICs [2]. Survival in LMICs is poor, with fewer treatment possibilities than in HICs [2]. In South Africa, the 2-year OS rate for high-risk neuroblastoma (HR-NB) is 27.1%, with long-term outcomes being even lower [3]. To improve disease outcomes, new and innovative applications of available treatments should be evaluated while being aware of resource limitations.

Neuroblastoma, a neuroendocrine tumor of the sympathetic nervous system, accounts for 7% to 8% of pediatric cancer deaths worldwide [4]. High-risk disease (HR-NB), driven by metastatic spread and adverse biology, represents roughly 50% of cases in HICs but up to 80% in LMICs [3,4]. This imbalance drives a stark survival gap: 5-year OS rates reach 60% to 70% in HICs with full multimodal therapy (chemotherapy, surgery, radiotherapy, autologous stem cell transplant, and anti-GD2 immunotherapy) but remains as low as 10% in LMICs, where these modalities are frequently unavailable [1]. In South Africa, specifically, retrospective data show that the majority of children present with HR-NB, and a substantial proportion require palliative rather than curative-intent management [3,5].

Two sequential goals define HR-NB management: achieving sufficient metastatic remission during induction chemotherapy to permit local therapies (surgery or radiotherapy) and then consolidating that remission [1,2]. In HICs, standard induction regimens (rCOJEC [rapid cisplatin, vincristine, etoposide, cyclophosphamide, and carboplatin] or doxorubicin-containing protocols) achieve metastatic complete remission (mCR) in 36% to 37% of patients [6]. In the South African cohort, mCR was achieved in only 30.7% of patients with HR-NB [3]. In HICs, mCR is not necessary to proceed to local treatment because subsequent treatment steps, such as high-dose chemotherapy and maintenance treatment with immunotherapy, may eradicate remaining tumor cells [2,7]. Yet, high-dose chemotherapy and immunotherapy are not available to the majority of South African patients [3]. Clinicians remain reluctant to proceed to surgery without at least a very good partial response, limiting progression to definitive local therapy [7]. Escalating chemotherapy intensity (eg, extended topotecan-vincristine-doxorubicin induction [8]) adds modest gains in mCR (5%‐10%) at the cost of increased toxicity, without a clear survival benefit [2]. Anti-GD2 immunotherapy improves survival by 15% to 20% in HICs [2] but is prohibitively expensive (approximately €210,000 [US $245,000] per treatment course) [7,9] and is not accessible in South African public-sector hospitals. Autologous stem cell transplant is similarly constrained by resource availability. There is, therefore, a clear, unmet need for an induction strategy that can improve mCR rates without depending on infrastructure or agents that are inaccessible in this setting.

Neuroblastoma is highly radiosensitive [10], and up to 95% to 98% of tumors express somatostatin receptors (SSTRs) amenable to [⁶⁸Ga]⁶⁸Gallium DOTA-peptide ([⁶⁸Ga]Ga-DOTA-peptide) imaging and [¹⁷⁷Lu]Lutetium DOTATATE ([¹⁷⁷Lu]Lu-DOTATATE) peptide receptor radionuclide therapy (PRRT) [11,12]. [¹³¹I]Iodine metaiodobenzylguanidine ([¹³¹I]I-mIBG) therapy, the existing radionuclide option in South Africa, is already used as an induction adjunct but carries significant hematological toxicity, requires prolonged radiation-shielded inpatient admission, excludes approximately 10% of mIBG-nonavid tumors, and depends on an inconsistent local isotope supply [2,5,13-15]. [¹⁷⁷Lu]Lu-DOTATATE PRRT, by contrast, has a more favorable early safety profile (reversible grade 2‐3 hematological toxicity, minimal renal toxicity, outpatient administration), targets a receptor expressed in a larger proportion of tumors, and is comparatively less resource-intensive to deliver [11,12,16-18]. A full comparison of radiopharmacology, administration protocols, safety profiles, and cost is provided in Table S1 of Multimedia Appendix 1. Given these features, [¹⁷⁷Lu]Lu-DOTATATE PRRT, used as an adjunct to standard induction chemotherapy rather than as monotherapy, represents a plausible, more accessible strategy to improve mCR rates in South African children with HR-NB.

Existing clinical experience with [¹⁷⁷Lu]Lu-DOTATATE PRRT in neuroblastoma is confined to the relapsed or refractory setting, in small cohorts of heavily pretreated patients, and shows encouraging but inconsistent responses [12,16,17]. No published trial has evaluated [¹⁷⁷Lu]Lu-DOTATATE PRRT in previously untreated, newly diagnosed children. A UK phase 2a study of empiric-dose, 4-cycle monotherapy in relapsed or refractory disease reported no objective responses, and dosimetric follow-up work showed that tumor-absorbed dose falls sharply across repeated cycles while renal dose remains stable, which implies that low-intensity, repeated-cycle monotherapy progressively underdoses the tumor [19,20]. Where clinical benefit has been observed, it has come from combination approaches: Fathpour et al [21] reported partial or complete responses in 3 of 5 children treated with [¹⁷⁷Lu]Lu-DOTATATE PRRT alongside intensive chemotherapy in a relapse or refractory setting. A Chinese cohort using higher dose per-cycle activity (100‐200 MBq/kg) reported disease control in approximately 55% of children with no renal toxicity [22]. Ongoing trials (eg, the LuDO-N study) are testing dose-intensified, dosimetry-guided monotherapy schedules, but outcome data are not yet available [23]. Taken together, the literature supports 2 consistent points: single-cycle or combination use, rather than repeated low-dose monotherapy, is likely to be more effective, and no data yet exist on [¹⁷⁷Lu]Lu-DOTATATE PRRT administered as part of first-line induction. This represents the specific gap this study addresses. By limiting [¹⁷⁷Lu]Lu-DOTATATE PRRT to a single administration, the trial aims to prevent treatment delays but also to evaluate the toxicity in combination with OJEC (vincristine, carboplatin, etoposide, and cyclophosphamide) and OPEC (vincristine, cisplatin, etoposide, and cyclophosphamide) in previously untreated children.

We hypothesize that adding a single cycle of induction [¹⁷⁷Lu]Lu-DOTATATE PRRT to standard first-line chemotherapy is feasible and safe and may improve mCR rates in South African children with newly diagnosed stage 4 HR-NB, thereby increasing the proportion of children able to proceed to definitive local therapy. This proof-of-concept pilot study aims to evaluate the feasibility, safety, and preliminary efficacy, with mCR as the end point of a single-cycle induction [¹⁷⁷Lu]Lu-DOTATATE PRRT combined with standard chemotherapy in this setting.

The primary objectives are to evaluate the dose-limiting toxicity (DLT) profile of participants treated with induction [¹⁷⁷Lu]Lu-DOTATATE PRRT in conjunction with standard induction chemotherapy; to evaluate dosimetry in South African patients receiving induction [¹⁷⁷Lu]Lu-DOTATATE PRRT; and to evaluate the resource-related feasibility in patients treated with induction [¹⁷⁷Lu]Lu-DOTATATE PRRT in conjunction with standard induction chemotherapy.

The secondary objective (exploratory objective) is to determine the postinduction metastatic response rate in patients diagnosed with HR-NB after receiving a single cycle of induction [¹⁷⁷Lu]Lu-DOTATATE PRRT in conjunction with standard induction chemotherapy based on the International Neuroblastoma Response Criteria (INRC).

Eligible participants are enrolled across 4 participating academic hospital sites in South Africa (Tygerberg Hospital, Red Cross War Memorial Children’s Hospital, Chris Hani Baragwanath Academic Hospital, and Inkosi Albert Luthuli Central Hospital). Following confirmation of eligibility, participants are allocated in a 1:1 ratio to either the standard or experimental arm using a computer-generated permuted-block randomization sequence with randomly varying block sizes of 4 and 6. Randomization is performed centrally by the national principal investigators (PIs), with allocation concealed from site investigators and participants until after enrollment. Participants in the standard arm receive 7 cycles of OJEC and OPEC induction chemotherapy per the standard South African protocol. Participants in the experimental arm receive a single intravenous administration of [¹⁷⁷Lu]Lu-DOTATATE PRRT (3.5‐5.5 GBq; approximately 0.4 GBq/kg), administered following completion of cycle 1 of OJEC and OPEC, with subsequent cycles 2 to 7 proceeding as per the standard arm. Response assessment is conducted at the end of induction (after cycle 7) in all participants using [⁶⁸Ga]Ga-DOTATATE PET (positron emission tomography)/CT (computed tomography); CT of the thorax, abdomen, and pelvis; and bone marrow trephine biopsy, with responses based on the 2017 INRC (Multimedia Appendix 2). A schematic of the trial design is presented in Figure 1.

The standard of care for neuroblastoma management is defined as management according to the national prospective management protocol SACCSG (South African Children’s Cancer Study Group) NB-2017.

Induction [¹⁷⁷Lu]Lu-DOTATATE PRRT is defined as a single administration (including diagnostics and prophylaxis) of [¹⁷⁷Lu]Lu-DOTATATE PRRT.

Medical exclusion is defined as the exclusion of any significant sites of active disease, as determined by unequivocal, contrast-enhancing lesions on CT or magnetic resonance imaging that lack SSTR expression, which can be confirmed by a 2-[¹⁸F]fluorodeoxyglucose PET/CT, if available.

DLT is defined as any grade 3, 4, or 5 hematological or nonhematological side effect according to the Food and Drug Administration’s Common Terminology Criteria for Adverse Events (CTCAE).

The study will be a phase 1, multicenter, prospective, open-label, 2-arm safety study. This protocol was developed and is reported in accordance with the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) guidelines (Checklist 1).

The pediatric hematology and oncology units and associated nuclear medicine departments that are participating in the study are Red Cross War Memorial Children’s Hospital, Cape Town; Tygerberg Academic Hospital, Cape Town; Chris Hani Baragwanath Academic Hospital, Johannesburg; and Inkosi Albert Luthuli Academic Hospital, Durban.

All 4 sites are accredited to deliver standard-of-care chemotherapy treatment and radioisotope therapy, as well as biochemical and radiological treatment monitoring services.

Patients diagnosed with stage 4 HR-NB according to the SACCSG NB-2017 national protocol will be eligible to be recruited prospectively and centrally and electronically randomized to either the [¹⁷⁷Lu]Lu-DOTATATE PRRT arm or the standard-of-care arm. Therefore, proactive recruitment is not possible. The inclusion and exclusion criteria are given in Textbox 1 [24-26].

The study will randomize 10 patients to each of the standard and experimental (interventional) arms. Thus, a total of 20 patients will be included in the study (see the SPIRIT figure in Multimedia Appendix 3).

The 4 hospitals account for nearly a third of the 30 to 40 patients diagnosed with neuroblastoma per year in South Africa. Therefore, we expect the study duration to be 3 to 4 years, starting when all institutional ethics committees and authorities have given approval to conduct the study.

The end point of the study is the evaluation of the treatment toxicity profile. Anemia, thrombocytopenia, neutropenia, and infections during induction chemotherapy will serve as outcomes to evaluate the side-effect profile (see the data collection sheet “Toxicity” in Multimedia Appendix 5) [26].

An exploratory end point is the metastatic tumor response between the standard and [¹⁷⁷Lu]Lu-DOTATATE PRRT arms after completion of the induction chemotherapy (after cycle 7 of chemotherapy). The measure of both study arms will be the INRC, and the radioisotope response criteria will be based on modified Curie scores.

The experimental (interventional) arm is a single administration. Therefore, discontinuation will primarily be due to noninclusion based on exclusion criteria, if a participant withdraws consent before the intervention, and unacceptable DLT.

DLT is defined as any grade 3, 4, or 5 hematological or nonhematological adverse event according to the Food and Drug Administration’s CTCAE, judged by the investigator to be at least possibly related to study treatment. Safety monitoring will focus on the experimental (interventional) arm (OJEC and OPEC plus [¹⁷⁷Lu]Lu-DOTATATE) therapy. Recruitment to the experimental (interventional) arm will be suspended and reviewed by the safety committee if more than 2 DLTs occur among the first 5 patients treated in this arm, or if 3 or more DLTs occur at any time among the first 10 patients treated in this arm. In this case, the safety committee will determine whether the regimen is unacceptably toxic. If so, accrual to the [¹⁷⁷Lu]Lu-DOTATATE therapy arm will be permanently discontinued and the combination considered not suitable for further investigation at this dose and regimen.

All adverse events occurring from the administration of the study intervention until the completion of induction chemotherapy will be recorded and graded according to the CTCAE. Severity, expectedness, and relationship to the study treatment will be assessed by the treating local investigator and reported to the national PIs. Serious adverse events and suspected unexpected serious adverse reactions will be reported to SAHPRA (South African Health Products Regulatory Authority) and the institutional ethics committee according to South African Good Clinical Practice requirements. Participants experiencing an adverse event will be followed until resolution, stabilization, or determination of a clinically acceptable outcome.

Concomitant care is the same in both study arms (Multimedia Appendices 2, 5 and 7). If there is an adverse reaction to the [¹⁷⁷Lu]Lu-DOTATATE therapy, standard international care guidelines will be followed. If participants discontinue the allocated intervention but do not withdraw consent for data collection, protocol-specified safety and outcome assessments will continue wherever clinically feasible. The study has secured insurance to cover all costs and reimbursements in case of adverse events.

The study will conform to the ethical considerations described in the Declaration of Helsinki and the South African Good Clinical Practice: Clinical Trial Guidelines [31,32]. The study will also be done according to the standards for research in LMICs [31]. Both of these guidelines will ensure respectful research conduct toward study participants by the research team during the study period.

The LuDO-SA trial received ethics approval from Stellenbosch University (M21/12/027), the University of Cape Town (467/2022), the University of the Witwatersrand (R14/49), and University of KwaZulu-Natal and SAHPRA regulatory approval (reference: 20220313) in 2022.

Recruitment commenced in January 2023 across 3 participating academic hospital sites, with the fourth starting recruitment in January 2025. As of July 2026, a total of 15 of the 20 planned participants have been enrolled: 7 in the standard arm and 8 in the experimental arm.

In January 2024 (version 2.0), the inclusion criteria were amended: [⁶⁸Ga]Ga-DOTATATE PET/CT alone became sufficient for eligibility (previously both [⁶⁸Ga]Ga-DOTATATE PET/CT and [¹²³I]I-mIBG were required). [¹²³I]I-mIBG avidity was removed as a criterion due to the limited and consistent availability of [¹²³I]I-mIBG isotopes in South Africa, thereby excluding potential inclusions.

From the inclusion of the first patient, the medical physicists coordinated across the four sites to adapt the timing of the dosimetry scan in order to be logistically manageable during extended weekends, as well as due to limitations of staff availability and patient transportation requirements. The first draft protocol recommended exact SPECT/CT acquisitions at 2, 24, 96, and 160 hours. The timings were amended to the following time points: 1 to 3 hours postinjection, 20 to 24 hours, 68 to 76 hours, and 144 to 196 hours (when scanning is not interrupted by a weekend or public holiday); or 1 to 3 hours postinjection, 20 to 24 hours, 90 to 100 hours, and 144 to 196 hours (when scanning is interrupted by a weekend or public holiday; see the full imaging procedure in Multimedia Appendix 8).

Recruitment is projected to close by June 2027. Data analysis is anticipated to commence in the third quarter of 2027, with results expected to be published in the fourth quarter of 2027.

A significant number of children in LMIC settings have limited treatment options for neuroblastoma and present with advanced disease. The poor metastatic tumor response during induction in South African patients has been reported by the study team under the main PhD study (SU HREC reference: S18/07/138 [PhD]; “The outcomes and management of neuroblastoma in South Africa”). Findings during this research have created the need to evaluate therapeutic options to improve the response rate of the tumor during induction. Research into neuroblastoma in South Africa is being continued under the main study as a post doc project (SU HREC reference: S18/07/138). This will be the first prospective study evaluating the role of [¹⁷⁷Lu]Lu-DOTATATE therapy added to standard first-line chemotherapy for stage 4 HR-NB in South Africa and sub-Saharan Africa to improve mCR.

We anticipate that the addition of [¹⁷⁷Lu]Lu-DOTATATE to standard OJEC and OPEC induction chemotherapy will be feasible within the constraints of a resource-limited, multisite South African setting and that its DLT profile will not exceed that of standard induction chemotherapy alone. Based on the mechanistic rationale for combining a radio-sensitizing or receptor-targeted agent with chemotherapy in a highly somatostatin receptor–avid tumor, we further anticipate a numerically higher end-of-induction mCR rate in the experimental arm than in the standard-of-care arm, although the trial is not powered to demonstrate this difference statistically. Confirmation of an acceptable DLT profile alongside a signal of improved mCR would support progression to a larger, adequately powered trial.

The LuDO-SA trial differs from existing [¹⁷⁷Lu]Lu-DOTATATE neuroblastoma trials in 3 respects that are worth situating explicitly. First, published experience with [¹⁷⁷Lu]Lu-DOTATATE in neuroblastoma has so far been confined to the relapsed or refractory setting in small, heavily pretreated cohorts [12,16,17]. The UK phase 2a trial of empiric-dose, 4-cycle [¹⁷⁷Lu]Lu-DOTATATE monotherapy in relapsed or refractory disease reported no objective responses, and companion dosimetric work showed that tumor-absorbed dose falls sharply across repeated cycles while renal dose remained stable, suggesting that repeated low-dose monotherapy progressively underdoses the tumor rather than reflecting a lack of radiosensitivity [19,20]. Where clinical benefit has been reported, it has come from combination approaches: Fathpour et al [21] observed partial or complete responses in 3 of 5 children treated with [¹⁷⁷Lu]Lu-DOTATATE alongside intensive chemotherapy. The LuDO-SA trial builds directly on this observation by testing a single, higher-activity dose delivered during first-line induction chemotherapy, rather than as sequential monotherapy after relapse. Second, the ongoing LuDO-N trial is evaluating dose-intensified, dosimetry-guided [¹⁷⁷Lu]Lu-DOTATATE monotherapy schedules but in the relapsed or refractory rather than first-line setting, and outcome data are not yet available [23]. The LuDO-SA trial is, to our knowledge, the only trial evaluating [¹⁷⁷Lu]Lu-DOTATATE as part of first-line induction in previously untreated children. Third, and specific to the South African and LMIC context, the LuDO-SA trial responds to a documented gap in first-line outcomes: 2-year OS rates for HR-NB in South Africa is markedly lower than in high-income settings (27.1% vs up to 60%) [2,3], at a time when stem cell transplantation and immunotherapy, both used to intensify first-line treatment in HICs, remain largely unavailable or unaffordable locally. [¹⁷⁷Lu]Lu-DOTATATE’s more favorable early safety profile and outpatient deliverability relative to the existing local radionuclide option, [¹³¹I]I-mIBG therapy [2,5,13-15], make it a plausible intensification strategy that does not presuppose access to transplantation infrastructure. The LuDO-SA trial is therefore best understood not as a version of a HIC trial but as a first-line intensification strategy purpose-built for a setting where the comparator arm already represents the ceiling of currently available standard care.

The trial has several methodological strengths: a prospective, randomized, multisite design across 4 accredited South African academic centers; standardized dosimetry performed and reconstructed centrally (Tygerberg Hospital, using a single calculation algorithm) to ensure comparability of absorbed-dose estimates across sites; blinded, adjudicated response scoring despite the trial being open-label; and a genuinely novel first-line application of [¹⁷⁷Lu]Lu-DOTATATE PRRT, addressing a treatment setting with no existing published data. The protocol is also explicit about its feasibility-driven design, providing a transparent sample-size rationale rather than presenting an underpowered comparison as if it were confirmatory.

Several limitations should be acknowledged. The sample size (10 patients per arm) is small, reflecting both the rarity of stage 4 HR-NB in South Africa (approximately 30 new cases nationally per year) and the trial’s feasibility and safety objective; as a result, efficacy estimates will carry wide CIs and cannot support definitive conclusions about comparative benefit. Blinding of participants, families, and treating clinicians is not possible, given the nature of the intervention (a radioisotope infusion with a distinct admission and monitoring pathway), which could introduce ascertainment bias into nonblinded aspects of care, though the primary response assessment itself is blinded and adjudicated. The trial is conducted in a single country, which limits the generalizability of feasibility findings (staffing, isotope logistics, and accommodation support) to other LMIC settings with different infrastructures. Finally, the design does not include a head-to-head comparison with [¹³¹I]I-mIBG therapy, the existing local radionuclide standard for induction intensification: the standard arm reflects chemotherapy alone, so the LuDO-SA trial can establish whether [¹⁷⁷Lu]Lu-DOTATATE improves on chemotherapy alone but not how it compares with the current alternative radionuclide approach.

In the event the LuDO-SA trial confirms both an acceptable DLT profile in the experimental arm per the prespecified stopping rule and a numerically favorable trend in end-of-induction mCR relative to the standard arm, the findings would support progression to a larger trial. A larger trial should be an adequately powered randomized controlled trial with metastatic response rate as the formally powered primary end point rather than toxicity and feasibility. Such a trial should ideally be multinational to achieve adequate accrual given the rarity of the disease. Conversely, if the DLT stopping rule is triggered or no meaningful trend in mCR is observed, it would argue against further first-line development of single-dose [¹⁷⁷Lu]Lu-DOTATATE at this activity and scheduling. A new randomized trial with adjustments to dose, timing, or patient-selection refinements should be evaluated before any further first-line trial is considered. In settings where more robust molecular diagnostics are available, randomizations can be planned between patients with positive and negative molecular diagnostics, for example, MYCN and ALK-gene mutations at diagnosis.

Trial results will be disseminated to participants, health care professionals, the public, and other relevant stakeholders through reporting in the trial registry; publication in an open-access, peer-reviewed journal; presentation at relevant academic congresses and meetings; and inclusion as part of a doctoral thesis at the University of Antwerp, Belgium. Annual progress reports will be submitted to the SAHPRA. Authorship of resulting publications will follow standard authorship eligibility guidelines. Data will be shared upon reasonable request, within the terms and regulations set by the participating ethics committees.