Retinoblastoma (RB) is a highly aggressive tumor and the most common primary intraocular malignancy in children. Metastatic RB is particularly challenging, as it requires intensive, multimodal therapies that often come with substantial toxicities, especially in very young children and even more critically in those with a constitutional Rb1 mutation. A recent study on patients with metastatic RB treated with high-dose chemotherapy followed by autologous hematopoietic stem cell transplant (HDC-ASCT) reported an overall survival rate grater then 70% for those without CNS metastases. Survival dropped less than 10% for patients with CNS involvement (1, 2). Complications associated with metastatic RB and its treatments include treatment related mortality, hearing loss, neurocognitive deficits and the development of second malignant neoplasms (SMNs), especially in patients carrying genetic predisposition (3–6). This highlights the urgent need for alternative treatment strategies that are both effective and less toxic. In this regard, RB cells, like other neuroectodermal-derived tissues—including neuroblastoma (NB)—highly express GD2, a cell surface disialoganglioside that plays a key role in malignant transformation and is a well-recognized therapeutic target (7, 8). Anti-GD2 monoclonal antibodies (mAbs) have markedly improved outcome in patients with high-risk neuroblastoma (HR-NB) (9–13). Dinutuximab (ch14.18) and Dinutuximab beta (ch14.18/CHO) had been approved by the United States Food and Drug Administration and the European Medicines Agency, respectively, for the treatment of HR-NB. Clinical trials of GD2-targeting therapy are currently ongoing for other GD2-expressing tumors apart from NB (14). According to recent literature, only seven patients with metastatic RB treated with anti-GD2 mAb have been reported (15–17). Anti-GD2 therapy was used as consolidation treatment for minimal residual disease (MRD) following HDC; however, in 2/7 patients, it was administered in the context of active disease. Here, we report a case of metastatic RB, with orbital and multifocal bulky bone disease at diagnosis, managed with anti-GD2 mAb Dinotuximab beta (Db) as consolidation after HDC-ASCT.

A 2-year-old boy was referred to our Hospital with a suspected diagnosis of RB in the left eye. Prior to admission, the patient had undergone orbital exenteration. At the clinical visit, the patient was in poor general condition, with severe malnutrition, generalized bone pain, and limb weakness. A total body CT scan revealed massive extraocular disease extension involving the left orbit and multifocal skeletal lesions ( Figures 1A–D ). A biopsy of a lesion in the right shoulder confirmed the diagnosis of RB. Molecular analysis of the tumor tissue demonstrated a homozygous pathogenic mutation in the Rb1 gene (c.1959dup). FISH analysis showed a normal copy number of the MYCN gene. The bone marrow biopsy showed infiltration by malignant cells. GD2 expression levels in the CD45-negative cells of the fresh tumor biopsy sample were quantified by flow cytometry and found to be 79.7% ( Figure 2 ). Brain and spine MRI showed no evidence of parenchymal or leptomeningeal disease. Cerebrospinal fluid (CSF) analysis was negative for tumor cells. The patient initiated systemic chemotherapy - ICE regimen (Ifosfamide 3gr/mq day 1-2-3, Carboplatin 750mg/mq day 1, Etoposide 150mg/m2 day 1-2-3)-, combined with intrathecal Topotecan (0.4 mg/dose) for CNS prophylaxis. Restaging after four cycles, including whole-body CT and brain and spinal MRI, demonstrated a good partial remission of left orbital and bone lesions. The patient’s general condition progressively improved, including regained mobility and adequate nutritional intake. Constitutional genetic testing for germline Rb1 alterations was negative. Afterwards the patient underwent HDC according to the COG ARET 0321 protocol (Carboplatin 350mg/mq day 1-2-3/Thiotepa 300mg/mq day 4-5-6/Etoposide 250mg/mq day 4-5-6) followed by ASCT (1). No moderate or severe complications during treatment. Post- HDC whole-body CT imaging showed no radiological evidence of residual disease ( Figures 1E–H ). We did not administer radiotherapy to the sites of bulky disease due to the patient’s young age. Given the high disease burden at diagnosis and the expression of GD2 on tumor cells, anti-GD2 mAb-Db was initiated 60 days after HDC, aiming to replicate the results previously achieved in MRD, following an approach analogous to HR-NB protocols. Three cycles were administered on a compassionate use basis, following informed consent for off-label therapy. The dosing schedule was 100 mg/m² over 10 days, every 5 weeks. Supportive care included morphine and gabapentin. Dinutuximab beta was well tolerated with no serious adverse effects. At the latest follow-up, the patient is in complete remission, with an overall survival of 6 years since diagnosis.

We present the case of a patient diagnosed with metastatic RB who underwent intensive, multimodal therapy. Given the initially inappropriate management and the massive widespread disease, we opted to consolidate remission following myeloablative therapy with Db. This decision was supported by existing clinical experience with GD2-positive tumors, particularly NB. Notably, our patient’s tumor cells, analyzed from a fresh biopsy, showed high GD2 expression—exceeding 75%—which further supported our rationale for using anti-GD2 mAb in this context. Importantly, published studies indicate that optimal response to dinutuximab is observed in cases where more than 75% of tumor cells express GD2, with a threshold of at least 100,000 GD2 molecules per cell considered significant for clinical benefit (18). The multimodal treatment of extraocular RB has significantly improved prognosis, particularly in cases with extra-CNS metastasis. However, notable treatment-related toxicities have been reported. Among the 60 patients enrolled in the COG ARET 0321 trial (1) there were two therapy-related deaths due to septicemia during induction therapy. Additional relevant complications included: encephalopathy in one patient, hearing impairment in four cases, gastric hemorrhage in one. Furthermore, the incidence of SMNs was high: six patients developed second cancers (2/6 received radiotherapy). Radiotherapy is generally reserved for bulky disease not in complete remission after induction chemotherapy. Its use is associated with an increased risk of SMNs within the irradiated field, particularly in patients with an underlying genetic predisposition (3–6). Moreover, due to the typically young age of these patients, radiotherapy is often not recommended. In this context, there is an increasing need for new therapeutic strategies that enhance treatment efficacy while reducing side effects. Dinutuximab is a chimeric monoclonal antibody that targets the disialoganglioside GD2, a glycolipid highly expressed on the surface of several tumor types, including neuroblastoma, small cell lung cancer, and certain sarcomas. By binding to GD2, dinutuximab induces tumor cell lysis through antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC). Clinically, it has proven effective primarily in high-risk neuroblastoma, where it is used as maintenance therapy post-consolidation to improve event-free and overall survival. Emerging evidence supports its potential utility in other GD2-expressing malignancies, such as osteosarcoma and triple-negative breast cancer, although these indications are still under active investigation (18, 19). Dinutuximab is administered intravenously, typically as a continuous infusion over 4 consecutive days per treatment cycle, with a standard dose of 17.5 mg/m²/day. Dinutuximab beta, a similar formulation, can be given as a continuous infusion over up to 10 days, which has been associated with reduced pain toxicity compared to shorter infusions. Administration is usually through a central intravenous line to minimize extravasation risk and facilitate management of adverse events. Common side effects include neuropathic pain, fever, hypersensitivity reactions, capillary leak syndrome, and hypotension. Pain management with analgesics, such as opioids and gabapentin, is critical during treatment. Close monitoring for allergic reactions, including anaphylaxis, is mandatory during infusion (20). Recently, seven patients with recurrent extraocular RB treated with anti-GD2 mAb have been reported in the literature (15–17). ( Table 1 ) Unlike our case, these patients initially presented with intraocular disease and later developed distant metastases following the failure of conservative treatment. All patients received HDC-ASCT. Patients 1 and 2 were treated with a reduced-intensity myeloablative regimen excluding Carboplatin, due to previous ototoxicity from platinum-based compounds, aiming to preserve hearing in light of their significant visual impairment. Patients 3, 4, 5, and 6 received radiotherapy to sites of bulky disease. Patient 4, developed an alveolar rhabdomyosarcoma of the maxillary bone five years later, located at the periphery of the previous radiation field. The expression of GD2 on tumor sample was assessed only in patients 3, 4, 5, and 6, while for the remaining cases this information was not reported. Patients 3 and 5 had active disease prior to anti-GD2 therapy, allowing for assessment of treatment response: patient 3 showed a complete response of a left mandibular lesion, while patient 5 had a partial response of a left orbital lesion. Our patient also received induction chemotherapy followed by HDC-ASCT, as well as CNS prophylaxis with intrathecal Topotecan. Anti-GD2 therapy was administered during remission after HDC with the intent to eliminate MRD and mitigate the risk of distant relapse, consistent with protocols for HR-NB. Notably, in our patient, the sites of bulky disease identified on imaging after induction chemotherapy were not irradiated, given the patient’s very young age (<3 years). The ability to avoid radiotherapy—along with its well-known long-term toxicities, particularly in very young children—is a significant advantage, highlighting the potential of intensive systemic therapy and targeted immunotherapy to achieve disease control while sparing patients from the harmful effects of radiation. None of the patients reported in the literature, including our case, experienced severe toxicity during anti-GD2 treatment. All adverse effects were effectively managed with appropriate supportive care. Six out of eight patients are alive and in complete remission. Patients 5 and 6 died due to disease relapse in the CNS, occurring 12 and 10 months after anti-GD2 therapy, respectively. This may be attributed to the limited ability of anti-GD2 antibodies to cross the blood-brain barrier, suggesting that CNS disease may require additional therapeutic strategies. These could include cellular immunotherapy employing anti-GD2 Chimeric Antigen Receptor (CAR) T cells (PMID: 37018492) (PMID: 33898999), and/or intrathecal delivery of radiolabeled antibodies, which have already been employed in other solid tumors with CNS metastases (21). All previously reported cases developed distant metastases following the failure of conservative treatments for intraocular tumors. In this context, integrating anti-GD2 mAb into the treatment arsenal for intraocular disease could be a promising strategy. This approach may not only enhance the effectiveness of conservative therapies and improve ocular outcomes, but also help control minimal disseminated disease (MDD) and reduce the risk of extraocular spread. The rationale for such integration is further strengthened by the well-established safety and tolerability profile.

Our case is unique compared with those previously reported in the literature due to the extensive disease burden at diagnosis and the successful use of dinutuximab beta as consolidation therapy without the need for radiotherapy. The challenges we encountered included managing a very young patient with severe initial clinical conditions and balancing treatment intensity with long-term toxicity risks. In conclusion, anti-GD2 therapy appears to be a promising option for the control of both MDD and MRD outside the CNS, contributing to a reduced incidence of distant relapse while maintaining a favorable safety profile. This approach may allow for the avoidance of more toxic treatments in very young patients, particularly those with an underlying genetic predisposition. For optimal use of this therapeutic strategy, preliminary assessment of GD2 expression is advisable. Importantly, anti-GD2 mAb should also be considered as a potential component in the management of advanced and/or refractory intraocular disease, with the aim of improving ocular outcomes and minimizing the risk of extraocular dissemination. These encouraging observations strongly support the rationale for designing prospective multinational collaborative trials that incorporate anti-GD2 therapy into the multimodal treatment of retinoblastoma to validate these findings and standardize future therapeutic approaches.