Long-term toxicity of tandem radioligand therapy using [90Y]Y/[177Lu]Lu-DOTA-TATE in neuroendocrine tumors: a polish multicenter experience with 20-year follow-up Authors: Maciej Kołodziej, Anna Zemczak, Kacper Pełka, Dariusz Pawlak, Katarzyna Jóźwik-Plebanek, Grzegorz Kamiński, Beata Kos-Kudła, Leszek Królicki, Marek Saracyn, Jolanta Kunikowska Published: 2026 ABSTRACT PurposeRadioligand therapy (RLT) targeting somatostatin receptors is an established treatment for advanced neuroendocrine tumors (NETs). While long-term safety data for [¹⁷⁷Lu]Lu-DOTA-TATE monotherapy are increasingly available, evidence regarding late toxicity of tandem RLT combining ⁹⁰Y- and ¹⁷⁷Lu-labeled somatostatin analogues remains limited. We evaluated the long-term safety profile of tandem [⁹⁰Y]Y/[¹⁷⁷Lu]Lu-DOTA-TATE RLT in a large real-world cohort with extended follow-up.MethodsThis retrospective study included 89 patients with NETs treated between 2002 and 2016 and followed for 9–20 years. Long-term renal, hematological, and hepatic toxicities were assessed. Renal function was analyzed longitudinally using patient-level annual estimated eGFR decline and time-to-event analysis for chronic kidney disease (CKD). Patients undergoing repeated RLT were analyzed separately.ResultsGrade ≥ 3 nephrotoxicity was observed in 3.4%. Median annual eGFR decline was − 2.1 mL/min/1.73 m² and was significantly steeper in patients with nephrotoxicity (− 4.2 vs. −1.6 mL/min/1.73 m²/year; p < 0.001). The cumulative incidence of CKD reached 51% at 10 years, while no cases of end-stage renal disease requiring dialysis or transplantation were observed. Grade ≥ 3 thrombocytopenia occurred in 5.6%, and one case of myelodysplastic syndrome was observed after supra-standard cumulative activity. Among patients undergoing repeated tandem RLT (16.9%), no statistically significant excess long-term renal or hematological toxicity was detected.ConclusionTandem RLT was associated with low rates of severe late toxicity, even after very prolonged follow-up and repeated treatment courses, although gradual long-term decline in renal function was common. These data provide a rare long-term real-world benchmark for late toxicity of high-energy β-emitter–based RLT and support the long-term feasibility of tandem RLT in selected patients with NETs.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00259-026-07941-z. Introduction The overexpression of somatostatin receptors characteristic of neuroendocrine tumors (NETs) [1] enables not only effective treatment with somatostatin analogues exerting antisecretory and antiproliferative effects [2, 3, 4, 5, 6, 7, 8], but also allows the use of highly sensitive diagnostic imaging techniques [9, 10, 11, 12, 13, 14, 15] and targeted radionuclide therapies [16, 17, 18, 19, 20] in patients with progressive, advanced NETs. Radioligand therapy (RLT), historically referred to as peptide receptor radionuclide therapy (PRRT), has been used in the treatment of NETs for more than 30 years. Numerous retrospective non-randomized reports have confirmed the efficacy of RLT in disease control while maintaining an acceptable safety profile [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]. Results of NETTER-1 [32, 33, 34, 35] and NETTER-2 [36] studies led to the incorporation of RLT as a standard therapeutic option and its inclusion in national and international diagnostic and treatment guidelines for NETs [37–45]. Nevertheless, RLT based on somatostatin analogues labeled with 177Lu, a radionuclide emitting β-particles with relatively low maximum energy (Emax 0.497 MeV) and a short tissue penetration range (approximately 2–4 mm), is not the only available option. Earlier clinical studies explored the use of somatostatin analogues labeled with yttrium-90 (90Y), a radionuclide known from the treatment of lymphoma and liver metastases of solid tumors [46, 47]. Yttrium-90 emits high-energy β-particles (Emax 2.27 MeV) with a substantially longer tissue penetration range (up to 10 mm). Despite its favorable dosimetric properties for larger lesions and satisfactory therapeutic efficacy, it has been associated with an increased risk of renal toxicity, particularly in long-term follow-up [24, 48–52]. The concept of tandem RLT, combining radionuclides with complementary physical properties, emerged as a strategy to optimize treatment of tumors of varying size while potentially limiting toxicity [25, 53]. Preclinical studies by de Jong et al. demonstrated superior therapeutic efficacy of combined long- and short-range β-emitters compared with single-isotope therapy [54]. In a rat tumor model, the simultaneous administration of a mixture of [177Lu]Lu-DOTA-TATE and [90Y]Y-DOTA-TOC at a 1:1 activity ratio resulted in a threefold increase in survival compared with monotherapy. These findings provided the biological rationale for the clinical implementation of tandem RLT protocols. Subsequent multicenter clinical studies conducted in Poland and across Europe demonstrated the safety and efficacy of tandem RLT based on 90Y- and 177Lu-labeled somatostatin analogues in patients with advanced NETs [25, 31, 53, 55–58]. While randomized trials such as NETTER-1 and NETTER-2 primarily focused on efficacy and reported relatively limited short-term toxicity, data on adverse effects in the long-term perspective remain scarce. Importantly, available long-term safety data largely concern [177Lu]Lu-DOTA-TATE monotherapy, including long-term follow-up from NETTER-1 and large retrospective series [24, 35, 59], whereas evidence regarding the late toxicity profile of tandem RLT remains limited. Given the increasing life expectancy of patients with NETs and the growing use of RLT in earlier lines of therapy, including the concept of RLT as a first-line treatment [60], comprehensive evaluation of long-term treatment-related toxicity is of particular clinical relevance. The aim of the present retrospective, multicenter study was to analyze the incidence of the most common known late adverse effects associated with tandem [90Y]Y/[177Lu]Lu-DOTA-TATE therapy over an extended follow-up period. Materials and methods Study design This was a retrospective, multi-institutional study conducted at Medical University of Warsaw and the Military Institute of Medicine in Warsaw, approved by the Ethics Committee of Medical University of Warsaw (consent no KB/16/2005 and KB/21/K/2011). The study was designed to analyze long-term treatment-related adverse effects in a consecutive cohort of patients with well-differentiated neuroendocrine tumors (NETs) and a Ki-67 index ≤ 20% who underwent tandem [90Y]Y/[177Lu]Lu-DOTA-TATE therapy and were followed for at least 9 years. All patients provided written informed consent prior to treatment. The study was conducted in accordance with local legal requirements, including radiation protection regulations, and complied with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Patients Patients treated between 2002 and 2016 at two tertiary referral centers (Medical University of Warsaw and the Military Institute of Medicine in Warsaw) were included. A total of 89 patients (33 males and 56 females) with a mean age (± SD) of 56.6 ± 11.1 years were enrolled. All patients had disseminated, inoperable, histologically confirmed NETs G1 or G2 according to WHO. Tumors were classified according to the current TNM staging and grading system for neuroendocrine tumors [61, 62]. Inclusion criteria All patients met the following inclusion criteria: Histological confirmation of neuroendocrine tumor G1 or G2 with metastatic, inoperable disease.Adequate hematological, hepatic, and renal function, defined as: hemoglobin ≥ 10 g/dL, white blood cell count (WBC) ≥ 3 × 10⁹/L, platelet count (PLT) ≥ 90 × 10⁹/L, bilirubin ≤ 1.5 × upper limit of normal (ULN), alanine aminotransferase (ALT) < 2.5 × ULN, and estimated creatinine clearance (CrCl) > 40 mL/min.Positive somatostatin receptor imaging (SRI), defined as PET/CT using [68Ga]Ga-DOTA-TATE or scintigraphy using [99mTc]Tc-HYNIC-TOC (Tektrotyd®, POLATOM, Poland), with a Krenning score ≥ 2.Karnofsky performance status ≥ 70 or ECOG performance status ≤ 2.Age ≥ 18 years.Life expectancy > 3 months.Absence of pregnancy or lactation. Histological confirmation of neuroendocrine tumor G1 or G2 with metastatic, inoperable disease. Histological confirmation of neuroendocrine tumor G1 or G2 with metastatic, inoperable disease. Adequate hematological, hepatic, and renal function, defined as: hemoglobin ≥ 10 g/dL, white blood cell count (WBC) ≥ 3 × 10⁹/L, platelet count (PLT) ≥ 90 × 10⁹/L, bilirubin ≤ 1.5 × upper limit of normal (ULN), alanine aminotransferase (ALT) < 2.5 × ULN, and estimated creatinine clearance (CrCl) > 40 mL/min. Adequate hematological, hepatic, and renal function, defined as: hemoglobin ≥ 10 g/dL, white blood cell count (WBC) ≥ 3 × 10⁹/L, platelet count (PLT) ≥ 90 × 10⁹/L, bilirubin ≤ 1.5 × upper limit of normal (ULN), alanine aminotransferase (ALT) < 2.5 × ULN, and estimated creatinine clearance (CrCl) > 40 mL/min. Positive somatostatin receptor imaging (SRI), defined as PET/CT using [68Ga]Ga-DOTA-TATE or scintigraphy using [99mTc]Tc-HYNIC-TOC (Tektrotyd®, POLATOM, Poland), with a Krenning score ≥ 2. Positive somatostatin receptor imaging (SRI), defined as PET/CT using [68Ga]Ga-DOTA-TATE or scintigraphy using [99mTc]Tc-HYNIC-TOC (Tektrotyd®, POLATOM, Poland), with a Krenning score ≥ 2. Karnofsky performance status ≥ 70 or ECOG performance status ≤ 2. Karnofsky performance status ≥ 70 or ECOG performance status ≤ 2. Age ≥ 18 years. Age ≥ 18 years. Life expectancy > 3 months. Life expectancy > 3 months. Absence of pregnancy or lactation. Absence of pregnancy or lactation. Study treatment and radioisotope administration Therapeutic activities of [90Y]Y/[177Lu]Lu-DOTA-TATE were prepared by radiolabeling DOTA-TATE (NCBJ POLATOM, Poland) with ⁹⁰YCl3 and ¹⁷⁷LuCl3 (ItraPol and LutaPol, NCBJ POLATOM, Poland, respectively) [23, 63]. Radiolabeling was carried out in sodium ascorbate buffer (pH 4.5) at 90°C for 30 min, followed by sterilization through a 0.22 μm filter. The final preparations were diluted with buffer to a concentration of 1.0 GBq/mL ± 10% and dispensed into sterile vials under aseptic conditions. All patients received RLT in an inpatient setting. Antiemetic prophylaxis consisted of 8 mg of ondansetron administered 30 min before the start of the amino acid infusion. Because commercially available arginine/lysine solutions were not accessible at the time of study, nephroprotection was provided using a mixed amino acid infusion; an amino acid infusion (1 L of Vamin 18 (Fresenius Kabi) containing 18 g of lysine and 22.6 g of arginine or 1 L of Nephrotec 10% (Fresenius Kabi), containing 16.9 g of lysine and 8.2 g of arginine) was initiated 60 min before RLT administration and continued over 4 h. On the following day, an additional 0.5 L infusion of Vamin 18 or Nephrotec 10% was administered following premedication with 8 mg of ondansetron. The RLT infusion was administered intravenously, at least 60 min after amino acid initiation, over approximately 30 min. The administered activity per treatment cycle was 3.7 GBq (1.85 GBq [90Y]Y-DOTA-TATE + 1.85 GBq [177Lu]Lu-DOTA-TATE) and planned number of cycles was 4, to reach cumulative activity of 14.8 GBq. In all patients receiving long-acting somatostatin analogues (SSA), the interval between the last SSA administration and the first cycle of tandem RLT was at least 4 weeks. During the course of RLT, long-acting SSA therapy was continued in all patients previously treated with SSA, while maintaining a minimum interval of 4 weeks between SSA administration and RLT [37, 38]. In patients with increased secretory activity of NETs, short-acting SSA was used as rescue therapy around the time of RLT, with a minimum interval of 24 h between the last administration of short-acting SSA and the administration of RLT. The interval between completion of chemotherapy and initiation of RLT exceeded 3 months in all patients. Assessment of bone marrow, liver, and kidney function The primary endpoint of the study was the assessment of long-term treatment-related adverse effects. Clinical and laboratory data evaluating bone marrow, liver, and kidney function were prospectively collected and entered into a dedicated database. The database included hematological parameters (white blood cell count, red blood cell count, platelet count, hemoglobin concentration), renal function parameters (serum creatinine and estimated glomerular filtration rate), liver enzymes (aspartate aminotransferase, alanine aminotransferase), and electrolyte levels. Renal function was estimated using the Modification of Diet in Renal Disease (MDRD) formula. Laboratory parameters were assessed during the course of therapy and subsequently at regular 6-month intervals until death or last follow-up. The duration of follow-up ranged from 9 to 20 years. Treatment-related adverse events were classified according to the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. Toxicities of grade ≥ 3 were considered severe adverse events. The occurrence of laboratory abnormalities resolving without permanent deficit during follow-up was defined as transient. Given the long observation period, longitudinal renal function analysis was additionally performed on a per-patient basis, allowing estimation of individual eGFR slopes over time. Radiological response assessment Best radiological response after RLT was assessed on computed tomography (CT) and/or magnetic resonance imaging (MRI) performed during routine follow-up at approximately 6 and 12 months after completion of treatment, according to the Response Evaluation Criteria in Solid Tumors (RECIST), version 1.0 until 2009 and version 1.1 thereafter. Response was categorized as complete response, partial response, stable disease, or progressive disease. Patients who died before scheduled imaging assessment were classified separately as “death before assessment”. Statistical analysis Continuous variables are presented as mean values with standard deviations or medians with interquartile ranges, depending on data distribution. Categorical variables are presented as frequencies and percentages. Statistical analyses were performed using IBM SPSS Statistics (version 23; IBM Corp., Armonk, NY, USA) and the R statistical environment (version 4.3.1; R Core Team, 2023). A p-value < 0.05 was considered statistically significant. Exploratory analyses of selected baseline renal risk factors for nephrotoxicity were also performed. Continuous variables were compared using the Mann–Whitney U test, whereas categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. In addition, exploratory multivariable logistic regression analyses were performed for any-grade renal toxicity and any-grade hematological toxicity. The covariates included age, baseline estimated glomerular filtration rate, cumulative administered activity, prior chemotherapy, and retreatment status. These analyses were considered exploratory because of the retrospective design, limited sample size, and heterogeneity of the historical cohort. Results Patient characteristics A total of 89 patients with well-differentiated neuroendocrine tumors were included in the analysis. Baseline characteristics are summarized in Table 1. Seventy-four patients (83.1%) received a single course of treatment. Of these, 61 (68.5%) received 4 infusions of 3.7 GBq (i.e. 1.85 + 1.85 GBq) of [90Y]Y/[177Lu]Lu-DOTA-TATE per injection at 6–12 weeks intervals, depending on clinical status, laboratory results and radiopharmaceutical availability, whereas 10 patients (11.2%) received 3–4 reduced-activity infusions of 2.96 GBq (i.e. 1.48 + 1.48 GBq) [90Y]Y/[177Lu]Lu-DOTA-TATE due to hematological or renal deterioration during treatment. Three patients received only 2 cycles of RLT, including 2 (2.2%) because of progression during therapy and 1 (1.1%) because of patient choice. The median cumulative administered activity of tandem [90Y]Y/[177Lu]Lu-DOTA-TATE was 14.8 GBq (IQR: 14.06–15.54), corresponding to a median of 4 treatment cycles (range 2–4). Additionally, 15 of 89 patients (16.9%) received repeated [90Y]Y/[177Lu]Lu-DOTA-TATE therapy due to radiological and/or clinical disease progression following the initial treatment course, reaching 17.0-30.7 GBq total injected activity and a total of 5–9 treatment cycles (Supplementary Table S1). The median interval between courses was 34.3 months (IQR: 18.7–80). Early treatment response was assessed at 6 and 12 months after completion of RLT (Table 2). Table 1Baseline patient and disease characteristics (N = 89)CharacteristicValueSex, n (%) Male33 (37.1) Female56 (62.9)Age at RLT initiation, years Median (IQR)58 (48.5–67.0)BMI, n (%) <25 kg/m241 (46.1) 25–30 kg/m231 (34.8) >30 kg/m217 (19.1)Smoking, n (%) yes32 (36.0) no57 (64.0)Tumor grade, n (%) G135 (39.3) G254 (60.7)Primary tumor site, n (%) Pancreas30 (33.7) Small intestine (midgut)25 (28.1) Colon (hindgut)16 (18.0) Unknown primary12 (13.5) Lung6 (6.7)[18F]FDG PET status prior to RLT, n/N (%) negative39/66 (59.1)* positive27/66 (40.9)*Prior treatments, n (%) SSA65 (73.0) Chemotherapy20 (22.5) Primary tumor resection64 (71.9)Baseline eGFR Median (IQR), mL/min/1.73m290.0 (74.0–108.5)Selected baseline renal risk factors, n (%) Hypertension59 (66.3) Diabetes31 (34.8)Cumulative administered activity Median (IQR), mCi400 (380–420) Median (IQR), GBq14.8 (14.06–15.54)Median number of RLT cycles4 (range 2–4)Repeated RLT (retreatment), n (%)15 (16.9)RLT radioligand therapy, BMI body mass index; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 66) Table 1Baseline patient and disease characteristics (N = 89)CharacteristicValueSex, n (%) Male33 (37.1) Female56 (62.9)Age at RLT initiation, years Median (IQR)58 (48.5–67.0)BMI, n (%) <25 kg/m241 (46.1) 25–30 kg/m231 (34.8) >30 kg/m217 (19.1)Smoking, n (%) yes32 (36.0) no57 (64.0)Tumor grade, n (%) G135 (39.3) G254 (60.7)Primary tumor site, n (%) Pancreas30 (33.7) Small intestine (midgut)25 (28.1) Colon (hindgut)16 (18.0) Unknown primary12 (13.5) Lung6 (6.7)[18F]FDG PET status prior to RLT, n/N (%) negative39/66 (59.1)* positive27/66 (40.9)*Prior treatments, n (%) SSA65 (73.0) Chemotherapy20 (22.5) Primary tumor resection64 (71.9)Baseline eGFR Median (IQR), mL/min/1.73m290.0 (74.0–108.5)Selected baseline renal risk factors, n (%) Hypertension59 (66.3) Diabetes31 (34.8)Cumulative administered activity Median (IQR), mCi400 (380–420) Median (IQR), GBq14.8 (14.06–15.54)Median number of RLT cycles4 (range 2–4)Repeated RLT (retreatment), n (%)15 (16.9)RLT radioligand therapy, BMI body mass index; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 66) Baseline patient and disease characteristics (N = 89) RLT radioligand therapy, BMI body mass index; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 66) Table 2Best radiological response after RLTResponse category6 months n (%)12 months n (%)Stable disease (SD)67 (75.3)53 (59.6)Partial response (PR)18 (20.2)16 (18.0)Complete response (CR)02 (2.2)Progressive disease (PD)4 (4.5)11 (12.4)Death before assessment07 (7.9)Total89 (100)89 (100)RLT radioligand therapy Table 2Best radiological response after RLTResponse category6 months n (%)12 months n (%)Stable disease (SD)67 (75.3)53 (59.6)Partial response (PR)18 (20.2)16 (18.0)Complete response (CR)02 (2.2)Progressive disease (PD)4 (4.5)11 (12.4)Death before assessment07 (7.9)Total89 (100)89 (100)RLT radioligand therapy Best radiological response after RLT RLT radioligand therapy Long-term safety and toxicity Hematological toxicity Any grade hematological toxicity was observed in 40 patients (44.9%). Anemia occurred in 31 patients (34.8%), leukopenia in 22 patients (24.7%), and thrombocytopenia in 15 patients (16.9%). All hematological toxicities are in Table 3. The majority of hematological adverse events were mild to moderate (CTCAE grade 1–2). Grade ≥ 3 thrombocytopenia was observed in 5 patients (5.6%), while grade ≥ 3 anemia or leukopenia was rare. Hematological toxicity was transient in most cases; reversible abnormalities without permanent bone marrow impairment were documented in 10.1% of patients with anemia and leukopenia and in 3.4% of patients with thrombocytopenia. One patient (1.1%) developed myelodysplastic syndrome during long-term follow-up. No significant correlation was observed between cumulative administered activity and the occurrence of overall hematological toxicity (Pearson r = 0.161, p = 0.132), nor with individual cytopenias. Prior chemotherapy was not significantly associated with the occurrence of any-grade hematological toxicity in exploratory analysis. Likewise, in the exploratory multivariable logistic regression model including age, baseline eGFR, cumulative administered activity, prior chemotherapy, and retreatment status, no variable reached statistical significance for overall hematological toxicity (Supplementary Table S2). Table 3Long-term treatment-related toxicity (CTCAE v5.0)ToxicityAny grade n (%)Grade 1Grade 2Grade ≥ 3Transient n (%)Hematological toxicity (any)40 (44.9)————Anemia (Hb)31 (34.8)161509 (10.1)Leukopenia (WBC)22 (24.7)14809 (10.1)Thrombocytopenia (PLT)15 (16.9)10253 (3.4)MDS1 (1.1)——1—Renal toxicity35 (39.3)201231 (1.1)Hepatic toxicity14 (15.7)8336 (6.7)CTCAE Common Terminology Criteria for Adverse Events, Hb hemoglobin, WBC white blood cell count, PLT platelet count, MDS myelodysplastic syndrome. Table 3Long-term treatment-related toxicity (CTCAE v5.0)ToxicityAny grade n (%)Grade 1Grade 2Grade ≥ 3Transient n (%)Hematological toxicity (any)40 (44.9)————Anemia (Hb)31 (34.8)161509 (10.1)Leukopenia (WBC)22 (24.7)14809 (10.1)Thrombocytopenia (PLT)15 (16.9)10253 (3.4)MDS1 (1.1)——1—Renal toxicity35 (39.3)201231 (1.1)Hepatic toxicity14 (15.7)8336 (6.7)CTCAE Common Terminology Criteria for Adverse Events, Hb hemoglobin, WBC white blood cell count, PLT platelet count, MDS myelodysplastic syndrome. Long-term treatment-related toxicity (CTCAE v5.0) CTCAE Common Terminology Criteria for Adverse Events, Hb hemoglobin, WBC white blood cell count, PLT platelet count, MDS myelodysplastic syndrome. Renal toxicity The baseline eGFR was preserved, with a median value of 90.0 mL/min/1.73 m² (IQR, 74.0–108.5). Any grade nephrotoxicity was observed in 35 patients (39.3%). All renal toxicities are in Table 3. Most cases were classified as CTCAE grade 1–2. Grade 3 renal toxicity occurred in 3 patients (3.4%). Renal impairment was transient in one patient (1.1%), while in the remaining cases renal function changes were persistent but clinically manageable. Longitudinal analysis demonstrated a gradual decline in estimated glomerular filtration rate (eGFR) over time (Fig. 1). Fig. 1Distribution of estimated glomerular filtration rate (eGFR) during the first 10 years of follow-up. Box-and-whisker plots show the distribution of eGFR values at consecutive follow-up years. Boxes represent the median and interquartile range; whiskers indicate the range. The dashed horizontal line denotes the threshold of eGFR 60 mL/min/1.73 m² Distribution of estimated glomerular filtration rate (eGFR) during the first 10 years of follow-up. Box-and-whisker plots show the distribution of eGFR values at consecutive follow-up years. Boxes represent the median and interquartile range; whiskers indicate the range. The dashed horizontal line denotes the threshold of eGFR 60 mL/min/1.73 m² On a per-patient basis, the median annual eGFR decline was − 2.1 mL/min/1.73 m²; the most pronounced decline in eGFR occurred within the first two years after initiation of tandem RLT, followed by a relatively stable rate of decline during subsequent long-term follow-up. In exploratory analysis, prior chemotherapy was not significantly associated with the annual rate of eGFR decline (median − 1.83 vs. −2.74 mL/min/1.73 m²/year in patients with and without prior chemotherapy, respectively; p = 0.311). Among patients evaluable for slope analysis, patients without nephrotoxicity (n = 54) showed a median annual eGFR decline of − 1.6 mL/min/1.73 m², whereas patients with nephrotoxicity (n = 35) demonstrated a steeper decline of − 4.2 mL/min/1.73 m² per year (Mann–Whitney U test, p < 0.001). No significant association was observed between cumulative administered activity and the annual rate of eGFR decline in either Pearson or Spearman correlation analyses. To explore potential confounders of renal toxicity, we performed an exploratory analysis of selected baseline renal risk factors. Patients who developed nephrotoxicity had lower baseline eGFR than those without nephrotoxicity (median 82.0 (IQR 62.5–97.0) vs. 93.0 (IQR 83.3–111.0) mL/min/1.73 m², p = 0.008). In contrast, age, sex, smoking status, hypertension, diabetes, obesity, and prior chemotherapy were not significantly associated with nephrotoxicity in this cohort. In exploratory multivariable logistic regression analysis, lower baseline eGFR (OR 0.78 per 10 mL/min/1.73 m² increase, 95% CI 0.62–0.97, p = 0.026) and higher cumulative administered activity (OR 1.25 per 1 GBq, 95% CI 1.04–1.51, p = 0.018) were independently associated with renal toxicity, whereas age, prior chemotherapy, and retreatment status were not (Supplementary Table S3). Time-to-event analysis using Kaplan–Meier methodology, with the first occurrence of eGFR < 60 mL/min/1.73 m² defined as the event, showed that 10.1% of patients met the criterion for chronic kidney disease (CKD) at baseline [64]. In the overall cohort, the cumulative incidence of eGFR < 60 mL/min/1.73 m² was 18% at 1 year, 23% at 2 years, 33% at 5 years, and 51% at 10 years, with a median time to eGFR < 60 mL/min/1.73 m² of approximately 10 years (Fig. 2). In patients with preserved baseline renal function (eGFR ≥ 60 mL/min/1.73 m²), the cumulative incidence of newly developed CKD was 9% at 1 year, 15% at 2 years, 26% at 5 years, and 45% at 10 years; the median time to CKD was not reached within the first 10 years of follow-up (Fig. 2). Importantly, although CKD incidence increased over time, progression to end-stage renal disease was not observed. Fig. 2Cumulative incidence of chronic kidney disease (CKD), defined as estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m². Cumulative incidence curves (1–Kaplan–Meier) are shown for the overall cohort (n = 89) and for patients with preserved baseline renal function (eGFR ≥ 60 mL/min/1.73 m²). Time to CKD was defined as the first documented eGFR < 60 mL/min/1.73 m² during follow-up Cumulative incidence of chronic kidney disease (CKD), defined as estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m². Cumulative incidence curves (1–Kaplan–Meier) are shown for the overall cohort (n = 89) and for patients with preserved baseline renal function (eGFR ≥ 60 mL/min/1.73 m²). Time to CKD was defined as the first documented eGFR < 60 mL/min/1.73 m² during follow-up Hepatic toxicity Hepatotoxicity was documented in 14 patients (15.7%), predominantly grade 1–2. Grade 3 liver enzyme elevation occurred in 3 patients (3.4%). Transient hepatic toxicity was observed in 6 patients (6.7%). No cases of treatment-related liver failure were recorded. Repeated tandem RLT Among the 89 patients included in the overall cohort, 15 patients (16.9%) underwent repeated radioligand therapy (Table 4), receiving a second course of tandem [90Y]Y/[177Lu]Lu-DOTA-TATE after achieving clinical benefit from the initial treatment. In this subgroup, renal toxicity was observed in 6 of 15 patients, hematological toxicity in 5 of 15 patients, and hepatotoxicity in 3 of 15 patients (Table 5). Compared with single-course treatment, repeated tandem RLT was associated with higher cumulative administered activity by design (median 20.72 GBq (IQR 17.39–28.31) vs. 14.80 GBq (IQR 13.04–14.80), p < 0.001). Table 4Baseline repeated RLT subgroup characteristics (N = 15)CharacteristicValueSex, n (%) Male4 (26.7) Female11 (73.3)Age at re-RLT initiation, years Median (IQR)53.0 (43.0–62.0)Tumor grade, n (%) G14 (26.7) G211 (73.3)Primary tumor site, n (%) Pancreas7 (46.7) Small intestine (midgut)1 (6.7) Colon (hindgut)5 (33.3) Unknown primary1 (6.7) Lung1 (6.7)[18F]FDG PET status prior to re-RLT, n/N (%) negative5/11 (45.5)* positive6/11 (54.5)*Prior treatments, n (%) SSA9 (60.0) Chemotherapy4 (26.7) Primary tumor resection14 (93.3)Baseline eGFR Median (IQR), mL/min/1.73m293.0 (73.0–101.0)Selected baseline renal risk factors, n (%) Hypertension7 (46.7) Diabetes5 (33.3)Cumulative administered activity Median (IQR), mCi560 (470–765) Median (IQR), GBq20.7 (17.4–28.3)RLT radioligand therapy; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 11) Table 4Baseline repeated RLT subgroup characteristics (N = 15)CharacteristicValueSex, n (%) Male4 (26.7) Female11 (73.3)Age at re-RLT initiation, years Median (IQR)53.0 (43.0–62.0)Tumor grade, n (%) G14 (26.7) G211 (73.3)Primary tumor site, n (%) Pancreas7 (46.7) Small intestine (midgut)1 (6.7) Colon (hindgut)5 (33.3) Unknown primary1 (6.7) Lung1 (6.7)[18F]FDG PET status prior to re-RLT, n/N (%) negative5/11 (45.5)* positive6/11 (54.5)*Prior treatments, n (%) SSA9 (60.0) Chemotherapy4 (26.7) Primary tumor resection14 (93.3)Baseline eGFR Median (IQR), mL/min/1.73m293.0 (73.0–101.0)Selected baseline renal risk factors, n (%) Hypertension7 (46.7) Diabetes5 (33.3)Cumulative administered activity Median (IQR), mCi560 (470–765) Median (IQR), GBq20.7 (17.4–28.3)RLT radioligand therapy; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 11) Baseline repeated RLT subgroup characteristics (N = 15) RLT radioligand therapy; [18F]FDG − 2-[18F]fluoro-2-deoxy-D-glucose, PET positron emission tomography, SSA somatostatin analogues, IQR interquartile range, eGFR estimated glomerular filtration rate; *Percentages calculated among patients with available [18F]FDG PET/CT data (N = 11) Table 5The safety in patients undergoing repeated tandem RLTParameterRepeated RLT subgroup (n = 15)Renal toxicity any grade, n (%)6 (40.0%)Hematological toxicity any grade, n (%)5 (33.3%)Hepatotoxicity any grade, n (%)3 (20.0%)End-stage renal failure (dialysis/transplant), n (%)0 (0%)Therapy-related hematological malignancy, n (%)0 (0%)MDS, n (%)1 (6.7%)RLT radioligand therapy, MDS myelodysplastic syndrome Table 5The safety in patients undergoing repeated tandem RLTParameterRepeated RLT subgroup (n = 15)Renal toxicity any grade, n (%)6 (40.0%)Hematological toxicity any grade, n (%)5 (33.3%)Hepatotoxicity any grade, n (%)3 (20.0%)End-stage renal failure (dialysis/transplant), n (%)0 (0%)Therapy-related hematological malignancy, n (%)0 (0%)MDS, n (%)1 (6.7%)RLT radioligand therapy, MDS myelodysplastic syndrome The safety in patients undergoing repeated tandem RLT RLT radioligand therapy, MDS myelodysplastic syndrome Notably, despite repeated exposure to tandem RLT, no cases of end-stage renal failure requiring dialysis or transplantation were observed. Additionally, no therapy-related hematological malignancies (except one case of myelodysplastic syndrome after supra-standard activities of repeated RLT and multiple chemotherapy lines) were observed following repeated tandem RLT. Longitudinal assessment of renal function in patients undergoing repeated tandem RLT demonstrated a gradual decline in eGFR over time. On a per-patient basis, the median annual eGFR decline was − 2.5 mL/min/1.73 m² per year (Fig. 3). Overall, any-grade renal toxicity occurred in 40.0% of patients in the repeated RLT subgroup and in 35.1% of those treated with a single course (p = 0.772), whereas hematological (33.3% vs. 43.2%, p = 0.573) and hepatic toxicity (20.0% vs. 14.9%, p = 0.698) also did not differ significantly. Fig. 3Individual trajectories of estimated glomerular filtration rate (eGFR) in patients undergoing repeated tandem [90Y]Y/[177Lu]Lu-DOTA-TATE therapy. Each line represents a single patient, illustrating interindividual variability in long-term renal function following repeated tandem radioligand therapy Individual trajectories of estimated glomerular filtration rate (eGFR) in patients undergoing repeated tandem [90Y]Y/[177Lu]Lu-DOTA-TATE therapy. Each line represents a single patient, illustrating interindividual variability in long-term renal function following repeated tandem radioligand therapy Discussion In this retrospective, multicenter analysis, we present one of the largest and longest-followed cohorts of patients with neuroendocrine tumors treated with tandem radioligand therapy (RLT) using dual-isotope [90Y]Y/[177Lu]Lu-DOTA-TATE. Although long-term RLT toxicity has been reported previously in larger cohorts [24, 65], the present study, with follow-up extending up to 20 years, provides rare long-term real-world data on the safety of tandem RLT. It should be noted that results from shorter follow-up periods of cohorts partially overlapping with the one currently presented have been reported previously [53, 55, 56]; however, such an extended observation period has not been published to date. One of the key findings of the present analysis is the low rate of severe late toxicity of tandem RLT, despite the use of 90Y, a high-energy β-emitter historically associated with increased renal toxicity [66, 67]. In a phase I clinical dose-escalation study using [90Y]Y-DOTA-TOC, Marincek et al. demonstrated that the incidences of grade > 3 renal toxicity were 8.4%, 6.5%, and 14.0% in patients receiving median administered activities of 2.4 GBq/cycle, 3.3 GBq/cycle, and 6.7 GBq/cycle, respectively [68]. Importantly, in our cohort, the administered activity of [90Y]Y-DOTA-TATE did not exceed 1.85 GBq per cycle. In a large cohort exceeding 1,100 patients treated with RLT using [90Y]Y-DOTA-TOC, Imhof et al. reported severe (grade > 3 according to CTCAE) nephrotoxicity in more than 9% of patients (n = 102) [65]. Similarly, in a Danish cohort of 69 patients with NETs treated with [90Y]Y-DOTA-TOC-based RLT, deterioration in glomerular filtration rate 6–12 months after completion of therapy was reported in 32% of patients, although the study did not provide an annualized median decline directly comparable to the present analysis [69]. Furthermore, a retrospective analysis of 53 patients treated with RLT using [90Y]Y-DOTA-TATE, either as monotherapy or as tandem RLT, demonstrated a higher mean annual decline in GFR in the monotherapy subgroup compared with tandem RLT (− 6.8 mL/min/1.73 m² vs. −3.0 mL/min/1.73 m², respectively) [51]. In our cohort gradual decline in estimated GFR over time was also noted. When analyzed on a per-patient basis, the median annual decline in GFR was − 2.1 mL/min/1.73 m² per year. When patients were stratified according to the occurrence of nephrotoxicity, clear differences in the dynamics of renal function became apparent. Patients without nephrotoxicity exhibited a modest and relatively slow decline in renal function, with a median annual eGFR decrease of − 1.6 mL/min/1.73 m². In contrast, patients who developed nephrotoxicity showed a significantly steeper decline in renal function, with a median annual eGFR decrease of − 4.2 mL/min/1.73 m². This difference was statistically significant (p < 0.001), indicating that nephrotoxicity was associated with an accelerated loss of renal function over time. In contrast, Durma et al. reported surprisingly different findings in a cohort of 128 patients treated with RLT [70]. In their analysis, RLT was paradoxically associated with an improvement in renal function among patients with impaired baseline kidney function, with the eGFR increasing by 2.95 mL/min/1.73 m² in the subgroup with baseline eGFR < 60 mL/min/1.73 m², while a decline of − 2.77 mL/min/1.73 m² was observed in patients with baseline eGFR > 60 mL/min/1.73 m² (p = 0.043). The authors primarily attributed this effect to adequate pre-therapeutic hydration, an overall improvement in general condition, the limitation of neoplastic cells and disease. Of note, however, that the subgroup of patients with baseline eGFR below 60 mL/min/1.73 m² was relatively small (n = 20). Moreover, the overall cohort received heterogeneous treatment approaches, including both tandem RLT and 177Lu-based monotherapy, without specification of how many patients received tandem RLT or whether this modality was also applied in those with impaired baseline renal function. Importantly, the duration of follow-up was substantially shorter (median 27.8 months) than in our cohort, in which follow-up ranged from 9 to 20 years. At the same time, interpretation of long-term renal decline should remain cautious, because renal function over such prolonged follow-up may be influenced not only by RLT, but also by aging, comorbidities, and subsequent therapies. Our exploratory analysis of selected baseline renal risk factors suggested that lower baseline renal function, rather than age, sex, smoking status, hypertension, diabetes, obesity, or prior chemotherapy, was associated with subsequent nephrotoxicity in this cohort. Evidence regarding baseline clinical predictors of RLT-associated nephrotoxicity is mixed. In pure 177Lu-based cohorts, conventional renal risk factors such as hypertension and diabetes were not consistently associated with renal function decline [71], although age has been identified as a predictor in some studies [72]. In contrast, tandem RLT and other 90Y-based regimen series have more often suggested a contribution of traditional renal risk factors to long-term nephrotoxicity [21, 24, 52]. Accordingly, our negative findings for age, sex, hypertension, diabetes, smoking status, and obesity should be interpreted with caution and considered hypothesis-generating rather than definitive, particularly given the limited sample size, exploratory nature of the present analysis and low number of severe renal events. At the same time, these findings underscore the importance of baseline renal reserve when assessing long-term renal risk after tandem RLT. In our study no significant association was observed between the cumulative administered activity of tandem RLT and the annual rate of eGFR decline, suggesting that cumulative activity alone did not predict the rate of long-term renal function deterioration in this cohort. Additionally, renal function dynamics were compared between patients who underwent a single course of tandem RLT and those who underwent repeated tandem RLT; the median annual decline in GFR in patients undergoing repeated tandem RLT (− 2.5 mL/min/1.73 m² per year) was numerically similar to that observed in the overall single-course subcohort (− 2.1 mL/min/1.73 m² per year), indicating no apparent acceleration of long-term renal function loss associated with repeated tandem RLT. Considerable interindividual variability in renal function trajectories was observed. While most patients exhibited a slow to moderate decline in GFR, a subset demonstrated a steeper decrease over time. Importantly, despite repeated exposure to tandem [90Y]Y/[177Lu]Lu-DOTA-TATE, no cases of end-stage renal failure requiring dialysis or kidney transplantation were observed in this subgroup during long-term follow-up. In the cohort presented here, nephrotoxicity was observed in 39.4% of patients (n = 35); however, the vast majority of cases (36%, n = 32) were classified as grade 1–2, while severe renal impairment (grade ≥ 3) occurred in only 3.4% (n = 3) of patients. In an ongoing multicenter, randomized study evaluating the efficacy and safety of RLT with variable, dosimetry-guided administered activities (including tandem RLT), no significant early renal complications (grade > 2) were observed, even when [90Y]Y-DOTA-TATE activities exceeding 1.85 GBq per cycle were applied [73]. However, the follow-up in that study was short and the reported data represent a preliminary safety analysis. It should also be emphasized that in earlier Polish analyses involving shorter follow-up of NET patients treated with tandem RLT [56], no cases of renal toxicity exceeding grade 2 were reported. This observation may suggest that renal impairment can emerge several years after completion of tandem RLT, underscoring the need for continuous long-term follow-up in this patient population. On the other hand, the very long follow-up period allowed us to show that the largest decrease in eGFR occurred during the first two years after initiation of tandem RLT, followed by a relatively stable rate of decline during subsequent long-term follow-up, without evidence of progressive acceleration over time. At the same time, these findings support the concept that the combination of long- and short-range β-emitters, together with amino acid–based renal protection, enables the safe administration of tandem RLT. Overall, the renal safety profile observed in our cohort compares favorably with historical data on monotherapy RLT using 90Y-labeled somatostatin analogues. The renal findings require careful contextualization. In healthy adults, the expected age-related decline in kidney function is substantially smaller than the median annual eGFR decline observed in our study. In a cohort of 12,062 patients, van der Burgh et al. reported an annual eGFR decline of approximately − 0.82 mL/min/1.73 m² [74]. In contrast, Gruppy et al., in a systematic review encompassing over 129,000 individuals from 12 studies between 1958 and 2021, estimated that the physiological rate of eGFR decline ranges from − 0.37 to − 1.07 mL/min/1.73 m² per year in healthy adults without hypertension [75]. Importantly, the authors also noted that decline rates differed between women and men, and decline rates sped up or slowed down with age. Thus, the renal changes documented in our cohort are unlikely to be explained by healthy aging alone. At the same time, over very prolonged follow-up in patients with metastatic NETs, renal function may be influenced not only by RLT, but also by baseline renal reserve, hypertension, diabetes, age, tumor-related factors, and subsequent therapies. Accordingly, the long-term renal findings should not be interpreted as being exclusively treatment-specific in a causal sense. Rather, our data suggest that tandem RLT was associated with low rates of severe renal toxicity and no end-stage renal disease, despite gradual long-term decline in kidney function. Although the present study was not designed to directly compare tandem RLT with [177Lu]Lu-DOTA-TATE monotherapy, the observed long-term safety profile suggests that tandem RLT may preserve an acceptable therapeutic window despite inclusion of the higher-energy beta emitter 90Y. This should be interpreted cautiously, as both radionuclides contribute to absorbed dose in organs at risk. Nevertheless, the shorter tissue penetration range and lower beta energy of 177Lu are generally associated with a more favorable renal safety profile than historical 90Y-based monotherapy [49, 50], while the tandem approach may offer a biological advantage in patients with lesions of heterogeneous size. It is noteworthy that tandem RLT was associated with good hematological tolerability in the present cohort. The only clinically relevant hematological toxicity (grade > 2) involved the platelet lineage and occurred in 5.6% (n = 5) of patients. Similar observations were reported in the NETTER-1 trial, in which the main hematological adverse events also affected the platelet lineage; however, it should be emphasized that NETTER-1 evaluated RLT based on [177Lu]Lu-DOTA-TATE monotherapy [32]. Likewise, in the ongoing multicenter DUONEN study, the only significant (grade > 2) hematological adverse events reported to date were cases of thrombocytopenia [73]. However, the available reports do not specify whether these events occurred in the [177Lu]Lu-DOTA-TATE monotherapy arm or in the tandem RLT arms. Previous retrospective observations consistently indicate a low rate of hematological toxicity associated with tandem RLT, predominantly limited to grade 1–2 events [24, 53, 56]. The present analysis, with its substantially longer follow-up, confirms the long-term hematological safety of tandem RLT, even many years after completion of therapy. In the cohort analyzed, a single case of myelodysplastic syndrome (1.1%, n = 1) was observed. Notably, this event occurred in a patient who underwent numerous chemotherapy lines and repeated RLT, with a cumulative tandem RLT activity of 30.7 GBq (i.e., 15.35 GBq [90Y]Y-DOTA-TATE plus 15.35 GBq [177Lu]Lu-DOTA-TATE), which exceeds the activity levels currently applied in retreatment protocols. It should be emphasized that the concept of repeated RLT is gaining increasing acceptance, and numerous reports—predominantly retrospective—have demonstrated its efficacy and safety [57, 76–78]. In the cohort presented here, 15 of 89 patients underwent repeated RLT after achieving a satisfactory response to initial treatment, without a significant deterioration in the overall safety profile of the therapy. Because cycle number closely paralleled cumulative administered activity, our exploratory analyses focused primarily on cumulative activity and retreatment status. Despite the fact that patients undergoing repeated tandem RLT received higher cumulative activities by design, this was not accompanied by a statistically significant increase in late hematological or hepatic toxicity. Although any-grade renal toxicity was numerically more frequent in the repeated RLT subgroup, this was not accompanied by evidence of excess severe renal toxicity or end-stage renal failure during long-term follow-up, and cumulative administered activity was not significantly associated with the annual rate of eGFR decline. These findings might support the long-term feasibility of carefully selected retreatment, but should be interpreted cautiously, given the limited size of the repeated RLT subgroup and the resulting restricted power to detect smaller but potentially clinically relevant differences. Hepatic toxicity should also be interpreted with caution in patients with metastatic NETs, as abnormalities in liver biochemistry may reflect not only treatment-related effects but also underlying hepatic tumor involvement. In our cohort, hepatotoxicity was documented in 14 patients (15.7%; predominantly grade 1–2) and no cases of treatment-related liver failure were observed. In a post hoc analysis of the NETTER-1 trial, Strosberg et al. reported that grade 3–4 liver function abnormalities were rare and did not appear to be associated with high baseline liver tumour burden [34]; however, these data were derived from patients treated with [177Lu]Lu-DOTA-TATE monotherapy only. Similarly, Duan et al. found no grade 3–4 hepatic toxicity during a 2-year follow-up after RLT, further supporting the view that clinically significant hepatotoxicity is uncommon, although that study also evaluated only [177Lu]Lu-DOTA-TATE monotherapy [79]. Additional support comes from the small prospective cohort reported by Bober et al., which specifically focused on hepatic complications and included both patients treated with [177Lu]Lu-DOTA-TATE monotherapy and a minor subgroup receiving [90Y]Y/[177Lu]Lu-DOTA-TATE tandem RLT; importantly, no hepatic complications above grade 1 were observed in that analysis [80]. Nevertheless, that cohort was limited in size, and, similarly to our study, liver tumor burden was not incorporated into the analysis in a standardized quantitative manner. Taken together, the available evidence suggests that clinically relevant hepatotoxicity after RLT is rare; however, the specific contribution of liver metastatic burden remains difficult to define, particularly in retrospective tandem RLT cohorts lacking standardized quantitative assessment of hepatic tumor load. The strengths of this study include the large cohort size, the multicenter nature of the data, and the exceptionally long follow-up period—likely among the longest reported to date—which enables a robust assessment of both late toxicities associated with tandem RLT and long-term survival outcomes. It should also be highlighted that the present cohort represents a historical treatment paradigm that is no longer routinely applied, as current standards favor 177Lu-based monotherapy. In this context, our findings provide a unique benchmark for long-term safety, particularly relevant for contemporary discussions on retreatment strategies, and the potential reintroduction of higher-energy β-emitters in selected clinical scenarios. Nevertheless, several limitations of the present analysis should be acknowledged, although they do not diminish the relevance of the reported findings. First, the retrospective design, which is a common feature of studies evaluating RLT, inherently limits causal inference. Second, the uniquely long data collection and follow-up period, while a major strength, is also associated with an evolution of diagnostic and therapeutic standards over nearly two decades, introducing heterogeneity into the analyzed population. The heterogeneity of prior management should also be emphasized. Although prior treatment exposure was captured at a broad level, including somatostatin analogue therapy, chemotherapy, and surgery, the historical database did not uniformly record detailed chemotherapy regimens or the exact number of prior systemic treatment lines. Moreover, quantitative tumor burden and the full impact of subsequent post-RLT treatments were not retrospectively available in a sufficiently standardized form for formal modeling. These factors limit the extent to which long-term renal and hematological outcomes can be attributed to tandem RLT alone. Nevertheless, exploratory analyses did not identify prior chemotherapy as a clear driver of the observed hematological or renal findings in this cohort. Likewise, liver metastatic burden was not retrospectively captured in a standardized quantitative manner, which precluded formal analysis of its relationship with hepatic toxicity. These limitations reflect real-world clinical practice and should be taken into account when interpreting the results. Finally, it should be emphasized that the administered activities of tandem RLT in the present cohort were determined empirically, and dosimetry was not routinely performed. Contemporary approaches increasingly advocate the use of dosimetry [81, 82] and individualized adjustment of administered activities [73] or number of cycles [83]; however, it should be recognized that such strategies are time-consuming and are not yet widely implemented in routine clinical practice. Conclusion In conclusion, tandem [90Y]Y/[177Lu]Lu-DOTA-TATE therapy was associated with low rates of severe late toxicity in patients with well-differentiated neuroendocrine tumors. Although gradual long-term decline in renal function was common, severe renal outcomes, including end-stage renal failure requiring dialysis or transplantation, were not observed. Likewise, no secondary hematological malignancies were observed, apart from a single case of myelodysplastic syndrome in a patient treated with above-standard cumulative tandem RLT activities. These findings provide important real-world evidence supporting the long-term feasibility of tandem RLT and contribute to the ongoing discussion on individualized radionuclide therapy strategies in NETs. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1