This meta-analysis was conducted in strict accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (8). We systematically searched the PubMed/MEDLINE, Embase, and Web of Science databases for relevant literature up to April 8, 2025. Search terms included “neuroendocrine tumor”, “pancreatic neuroendocrine tumor”, “gastroenteropancreatic neuroendocrine tumor”, “PRRT”, and “¹⁷⁷Lu-DOTATATE”, along with related variants of these terms. Additionally, we manually searched the reference lists of relevant articles to identify any further relevant studies.

Inclusion Criteria:

Exclusion Criteria:

The risk of bias (RoB) assessment was performed using the latest Cochrane Collaboration’s RoB tool for randomized controlled trials (RCTs) (9). This tool evaluates six domains: randomization process, deviations from intended interventions, missing outcome data, outcome measurement, selective reporting, and overall RoB, providing an overall judgment (9). For cohort studies and case-control studies, the Risk of Bias in Nonrandomized Studies of Interventions (ROBINS-I) tool was used for analysis (10). The overall risk of bias for each study was classified as low, moderate, serious, or critical.

Two researchers independently extracted data from the included studies, collecting the following information: first author, year of publication, study design, number of patients, tumor type, average therapeutic dose of radiopharmaceuticals, treatment duration, and follow-up period. The primary outcomes were efficacy and adverse events. Efficacy outcomes included disease control rate (DCR), ORR, pooled PFS, and overall survival (OS). DCR was defined as the percentage of patients achieving complete response, partial response, or stable disease, and ORR as the percentage achieving complete response or partial response. PFS was defined as the time from treatment initiation to tumor progression (e.g., enlargement or metastasis) or death, and OS as the time from treatment initiation to death from any cause. Imaging responses were assessed according to RECIST and mRECIST criteria (11, 12). Adverse events were recorded in accordance with the National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0, with an emphasis on hematologic and renal toxicities (13).

A meta-analysis was performed using Stata version 18.0 (StataCorp LLC, College Station, TX, USA) and R version 3.5.2 (R Foundation for Statistical Computing, Vienna, Austria). The efficacy of ¹⁷⁷Lu-DOTATATE treatment was assessed by calculating the DCR, ORR, PFS, and OS, each reported with 95% confidence intervals (CI). Treatment-related adverse events were also evaluated and reported with 95% CI. Statistical heterogeneity among studies was assessed using Cochran’s Q test and the Higgins I² statistic. Heterogeneity was considered significant if the Cochran’s Q test P-value was < 0.10 and I² was > 50%; in this case, a random-effects model was applied (14). Otherwise, a fixed-effects model was used. Sensitivity analyses were conducted by sequentially excluding each study (leave-one-out analysis) to test the robustness of the results. Statistical significance was defined as P < 0.05.

Systematic searches of electronic databases identified 1507 records (Figure 1). After removing 383 duplicate records, we screened the titles and abstracts of the remaining 1124 articles. During the title/abstract screening, 1,027 records were excluded due to being irrelevant to survival rates, meta-analyses and review articles, case reports, and non-English publications. A total of 97 articles were retained for full-text review. Full-text assessment led to the exclusion of 83 articles, primarily due to inappropriate publication type or failure to report DCR or ORR. Ultimately, 14 studies met all inclusion criteria and were included in the meta-analysis.

The 14 included studies comprised 5 retrospective and 9 prospective investigations; 12 employed a cohort design, and 2 were RCTs (Table 1). Cohort studies were assessed using the ROBINS-I, and RCTs were evaluated with the Cochrane Collaboration’s RoB tool. All included studies were judged to have a low overall RoB, and the three RCTs were considered to be of high methodological quality. We enrolled a total of 1844 patients. The baseline characteristics of the study participants are presented in Table 2.

We analyzed the DCR from 12 studies involving a total of 1279 patients (Figure 2a). The Cochran Q test yielded a P-value of <0.05 and an I² of 80.95%, indicating significant heterogeneity among the studies. Therefore, a random-effects model was employed for the meta-analysis. The DCR across studies ranged from 74.6% to 100%. The pooled data from all studies revealed a pooled DCR of 87.6% (95% CI, 82.5%-92%) for ¹⁷⁷Lu-DOTATATE treatment in NETs. The sensitivity analysis revealed consistent effect size estimates, ranging from 0.8546 to 0.9054, following the sequential exclusion of individual studies (Supplementary Figure 1) (Supplementary Table 1). The combined effect rate of the remaining studies was 86%, indicating that the results were stable and reliable.

In addition to evaluating disease control, we further analyzed the ORR. A total of 12 studies involving 1279 patients were included in the analysis (Figure 2b). The results, derived from a random-effects model, demonstrated that the pooled ORR of ¹⁷⁷Lu-DOTATATE in NET-treated patients was 32.2% (95% CI, 25.4%-39.3%). The study included 30 patients, all with G1 or G2 grade NETs (15).This study reported a notably high ORR of 58.1%, which may be attributable to the small sample size and the specific NETs grading, which differed from other studies.

PFS is a critical measure of the long-term benefits of treatment. We analyzed the PFS data from seven studies involving 665 patients (Figure 3a). Seven additional studies were excluded due to missing data. The meta-analysis using the random-effects model showed a pooled PFS of 30.87 months (95% CI, 22.71-39.04 months), with individual study PFS ranging from 18.9 to 59.8 months. The study reported the longest PFS at 59.8 months, though this study had the smallest sample size and the widest 95% confidence interval, indicating greater uncertainty in the result (24). The results of the sensitivity analysis confirmed the robustness of the combined PFS estimate (Supplementary Figure 2) (Supplementary Table 2). Sequential exclusion of individual studies maintained the PFS of the remaining studies at approximately 28.34 months, with no significant changes observed. This demonstrates the robustness and reliability of the combined PFS results.

In evaluating long-term efficacy, we examined OS after ¹⁷⁷Lu-DOTATATE therapy in NETs. The figure illustrates OS data from seven clinical studies involving 676 patients (Figure 3b). Using a random-effects model, the pooled OS was 51.85 months (95% CI, 39.99–63.71), with a range of 34 to 80.8 months.

Due to observed heterogeneity in efficacy outcomes, subgroup analysis were performed to explore potential factors. First, DCR and ORR were compared between G1–2 and G3 NETs; next, PFS was compared across tumor locations. We first analyzed DCR by grade (Figure 4a). For G1–2 NETs (2 studies, 58 patients), the pooled DCR was 97.7% (95% CI, 91.4%–100%). For G3 NETs (152 patients), the pooled DCR was 90.8% (95% CI, 85.1%–94.4%). The DCR for G1–2 tumors was slightly higher than for G3 tumors. However, the between-group P-value was 0.119 (P > 0.05), indicating no significant difference in DCR between the grades. Given heterogeneity in DCR and ORR across studies, we further analyzed ORR by grade (Figure 4b). The between-group P-value was 0.002 (P < 0.05), indicating significant heterogeneity. For G1–2 NETs (3 studies, 160 patients), the pooled ORR was 45.4% (95% CI, 35.3%–55.6%). For G3 NETs (2 studies, 209 patients), the pooled ORR was 27.1% (95% CI, 21.2%–33.4%). Thus, the ORR for G3 tumors was significantly lower than that for G1–2 tumors.

We also performed a subgroup analysis by tumor location (Figure 5). Four studies involved gastrointestinal NETs; the pooled PFS was 66.32 months (95% CI, 41.78–90.87). One study involved pancreatic NETs, reporting a PFS of 93.9 months (95% CI, 39.45–148.35). Patients with pancreatic NETs had significantly better PFS than those with gastrointestinal NETs.

We also evaluated the safety profile of ¹⁷⁷Lu-DOTATATE therapy. The adverse reactions were evaluated in 677 patients across 9 studies (Figure 6). The Cochran Q test indicated statistically significant heterogeneity in the overall incidence of adverse reactions across the included studies (P = 0.03), with an I² statistic of 58.925%, suggesting high heterogeneity. Thus, a random-effects model was applied for the meta-analysis. The overall incidence of adverse reactions was low 4.1% (95% CI, 2.6%–5.8%). The pooled incidence of grade ≥3 adverse reactions was 4.3% (95% CI, 1.6%–7.9%). Hematologic adverse events occurred in 6.1% of patients (95% CI, 3.7%–8.9%). Thrombocytopenia had a pooled incidence of 3.4% (95% CI, 0.3%–9.1%). Anemia was the most common adverse event 5.4% (95% CI, 0.3%–14.1%). Leukocytopenia occurred in 7.5% (95% CI, 0.8%–18.6%). Myelodysplastic syndrome (MDS) had an incidence of 2.7% (95% CI, 0.4%–6.5%). Acute Myeloid Leukemia (AML) is the least common adverse event. Among the 9 studies included, only one study from 2018 reported AML cases. The incidence rate was 1.1% (95% CI: 0.3%-3.9%) (27).