Long-Term Efficacy, Survival, and Safety of [¹⁷⁷Lu-DOTA⁰,Tyr³]octreotate in Patients with Gastroenteropancreatic and Bronchial Neuroendocrine Tumors

Neuroendocrine tumors (NET) are a heterogeneous group of tumors. A subset of this slow-growing tumor type are the gastroenteropancreatic NETs (GEP-NETs) and bronchial NETs (1). Unfortunately, the majority of GEP-NET patients have metastatic disease at time of presentation (2). Over the past decades, the incidence of GEP-NETs is rising. After the introduction of new therapies, the 5-year survival has significantly improved (3). This improvement is partly due to first-line therapy with long-acting somatostatin analogues (Octreotide LAR, Lanreotide Autogel) and targeted therapies (e.g., everolimus and sunitinib), which are approved second-line therapies for patients with progressive inoperable GEP NETs (sunitinib only being approved in pancreatic NET) (4, 5). For GEP and bronchial NETs, peptide receptor radionuclide therapy (PRRT) with [¹⁷⁷Lu-DOTA⁰,Tyr³]octreotate (¹⁷⁷Lu-DOTATATE) has yielded very promising results. This therapy is based on the fact that the majority of these NETs express a high number of high-affinity somatostatin receptors on their cell membranes. These receptors can be used for both imaging and therapy with radiolabeled somatostatin analogues (6). For therapy, the β-emitting radionuclide Lutetium-177, with a half-life of 6.7 days and a maximum β range of 2 mm in tissue, can be used. These parameters makes Lutetium-177 the ideal radionuclide for PRRT. Treatment with β-emitting radiolabeled somatostatin analogues results in impressive percentages of tumor regression (7–10). Also, success monitored as time to progression (TTP), progression-free survival (PFS), and overall survival (OS) has been reported in uncontrolled studies (8, 11). With the recent publication of the promising results of the NETTER-1 trial (12), the first phase III trial comparing ¹⁷⁷Lu-DOTATATE to high-dose Octreotide LAR therapy in patients with inoperable metastatic midgut NETs, it may be expected that the use of this therapy will increase in the coming years. We have previously reported on the treatment with the radiolabeled somatostatin analogue ¹⁷⁷Lu-DOTATATE (8), and have compared the side-effects and the results of this treatment with other treatment modalities and historical controls. Righteously, critical commentary arose, as to the validity of the comparisons with historical controls. Also, the high number of foreign patients that we have treated, and of whom a significant number was lost to follow-up might bias the percentage of reported severe adverse events. We, therefore, present now the long-term results of efficacy, survival, and toxicity of PRRT with ¹⁷⁷Lu-DOTATATE in a cohort of more than 500 Dutch patients with metastatic GEP-NETs and bronchial NETs, who all had their follow-up at our institution.

A total of 1,214 patients were treated with ¹⁷⁷Lu-DOTATATE from January 2000 to January 2015 in our institution. We selected only Dutch patients with NETs of the midgut, foregut, hindgut, and unknown primary that were treated according to a standard protocol, because of the very limited loss in follow-up in this subgroup. For safety evaluation, all patients who received at least 100 mCi (3.7 GBq) ¹⁷⁷Lu-DOTATATE were selected. For the evaluation of efficacy, patients who received at least 600 mCi (22.2 GBq) ¹⁷⁷Lu-DOTATATE before 2013 were investigated. Patients treated after 2013 returned to their referring specialist in a different institute after therapy, and imaging was not performed according to our study protocol. For safety analysis, 610 patients were available and 443 patients for efficacy and survival analysis (Fig. 1). For analysis of efficacy, NET types were divided into bronchial, pancreatic, (other) foregut, hindgut, midgut, and NET of unknown origin. Other foregut NETs included: five NETs of the stomach, five NETs of the proximal duodenum, and two NETs of the thymus. For comparison with the NETTER-1 study (12), 106 patients with progressive midgut NETs who received a cumulative dose of ≥100 mCi (3.7 GBq) ¹⁷⁷Lu-DOTATATE were selected. Patients were treated with ¹⁷⁷Lu-DOTATATE if the tumor uptake was at least as high as the uptake in the normal parenchyma of the liver on ¹¹¹In-DTPA-octreotide scintigraphy (OctreoScan) prior to the therapy. Other inclusion criteria were serum hemoglobin ≥6.0 mmol/L (≥9.7 g/dL), total white blood cell (WBC) count ≥2 × 10⁹/L, platelet count ≥75 × 10⁹/L, serum creatinine concentration ≤150 μmol/L (≤1.7 mg/dL), or creatinine clearance ≥40 mL/minute until 2007 or 50 mL/minute from 2007, serum albumin >30 g/L, Karnofsky performance status (KPS) ≥50, and no prior treatment with other radiolabeled somatostatin analogues. Preliminary results in a subgroup of these patients with GEP-NETs were reported previously (8). All patients gave written informed consent to participate in the study, which was approved by the medical ethical committee of our hospital.

[DOTA⁰,Tyr³]octreotate was obtained from BioSynthema. ¹⁷⁷LuCl₃ was distributed by IDB-Holland. ¹⁷⁷Lu-DOTATATE was locally prepared as described previously (13). Before the infusion of the radiopharmaceutical, Granisetron 3 mg or Ondansetron 8 mg was injected intravenously. To reduce the radiation dose to the kidneys, an intravenous infusion of amino acids (2.5% arginine and 2.5% lysine in 1 L 0.9% NaCl) was started 30 minutes before the administration of the radiopharmaceutical and lasted for 4 hours. The radiopharmaceutical was coadministered intravenously using a second pump system over 30 minutes. The intended interval between treatments was 6 to 10 weeks. The treatment interval could be extended to 16 weeks in patients with longer continuing subacute hematologic toxicity. Patients were treated up to a cumulative intended dose of 750 to 800 mCi (27.8–29.6 GBq) ¹⁷⁷Lu-DOTATATE.

Routine hematology, liver, and kidney function tests were performed after each therapy cycle and at follow-up visits at 6 weeks, 3 months, and 6 months after the last treatment cycle, and thereafter at 6-month intervals. CT or MRI was performed within 3 months before the first therapy, and at every follow-up visit. Every patient was seen at the outpatient clinic by one of the investigators and a research nurse. A case report form (CRF) was completed at every follow-up visit.

Tumor response was assessed on CT or MRI according to the Response Evaluation Criteria in Solid Tumors 1.1 (RECIST 1.1) criteria (14). Radiologic disease control was defined as all patients achieving objective response (OR) or stable disease (SD) in patients with progressive disease (PD) at baseline. Uptake on the OctreoScan was scored on planar images using a 4-point scale; grade 1: less than the uptake in the normal parenchyma of the liver, grade 2: equal to the liver, grade 3: greater uptake than the liver, grade 4: higher than the uptake in the normal spleen or kidneys. The whole-body extent of disease was scored by experienced nuclear medicine physicians as: limited, moderate or extensive on OctreoScan as described previously (15).

PFS was defined as the time from first day of treatment until day of ascertainment of objective progression or death from any cause. The TTP was calculated from the first day of treatment to the day of documented progression. Deaths were censored in the TTP analyses. OS was calculated from the first day of treatment until the day of death, or until the last date of follow-up for patients who were lost to follow-up (16).

Toxicity was scored according to the Common Terminology Criteria for Adverse Events 4.0 (CTCAE) scoring system (17).

For survival analysis, log-rank tests and Cox regression models were used. PFS and OS analyzes were performed using the Kaplan–Meier method. Comparison between subgroups was made using χ²-test (or, if applicable, Fisher exact test), ANOVA, or paired t tests. P values lower than 0.05 were considered to be significant.

For safety outcome, 610 patients who received ≥100 mCi (3.7 GBq) ¹⁷⁷Lu-DOTATATE were analyzed. Median follow-up time was 64 months [95% confidence interval (CI), 58–70 months]. Mean age was 60.4 years (range 23–88). The median TTP was 35 months (95% CI, 32–39 months) and the median OS was 58 months (95% CI, 52–64 months).

CTCAE grade 3/4 toxicity of the combination of the following hematologic parameters occurred in 61 of 582 patients (10%; 95% CI, 8%–13%), grade 3/4 toxicity of platelets occurred in 30 of 582 patients (5%; 95% CI, 4%–7%), total WBC count grade 3/4 toxicity in 32 of 582 patients (5%; 95% CI, 4%–8%), hemoglobin grade 3 toxicity in 22 of 582 patients (4%; 95% CI, 2%–6%), and no grade 4 hemoglobin toxicity was observed. No data are available on the remaining 28 patients. At first follow-up visit after 3 months, these values normalized in 77% of patients. CTCAE grade 3/4 of the lymphocytes was observed in 288 of 581 patients (50%; 95% CI, 46%–54%). At 3 months follow-up, 74 of 287 patients and at 30 months follow-up 6 of 108 patients had persistent grade 3/4 lymphocyte toxicity and 53 of 108 patients had grade 1/2 lymphocyte toxicity.

An increase of aminotransferases (aspartate transaminase and/or alanine transaminase) grade 3/4 was observed in 20 of 581 patients (3%; 95% CI, 2%–5%). After 3 months follow-up, there were three patients with persistent grade 3/4 toxicity of the aminotransferases. Creatinine grade 3/4 toxicity occurred in 2 of 581 patients (0.3%; 95% CI, <1%–1%). Serum creatinine levels normalized in all patients at 3 months follow-up.

Acute leukemia occurred during follow-up in 4 patients (0.7%) after a median follow-up of 55 months after first therapy (range 32–125 months). Three of these patients died after a median follow-up of 7 months after diagnosis of acute leukemia. Myelodysplastic syndrome (MDS) occurred in nine patients (1.5%) after a median follow-up of 28 months (range 9–41 months) after first therapy. Five of these patients died after a median follow-up of 7 months after diagnosis of MDS. Of the remaining patients, no date of death is known, because these patients were referred back to their own local hospital after the diagnosis acute leukemia/MDS was made. None of these patients were treated with alkylating chemotherapy prior to PRRT (Supplementary Table S5).

Renal failure occurred in six patients (1%) during follow-up. The most probable cause of the renal failure was postrenal in one patient (hydronephrosis) and prerenal in the other patients (hypotension after gastrointestinal bleeding in one patient and in four patients dehydration caused by severe vomiting and diarrhea). No hepatic failure was observed during or after therapy.

Patient characteristics of the 443 patients who received ≥600 mCi (22.2 GBq) ¹⁷⁷Lu-DOTATATE and who were evaluated for efficacy and survival are presented in Table 1. Median follow-up was 78 months from the first day of treatment. The majority of all patients (61%) were treated with somatostatin analogues at referral.

Best objective response rate (ORR) was defined as the proportion of patients achieving complete response (CR) and partial response (PR) at follow-up according to the RECIST 1.1 criteria. Best response rates are presented in Table 2. The ORR in the entire patient group was 39%. SD was found in 43% of patients. PD as treatment outcome was observed in 12% of patients and 5% of patients had nonevaluable treatment outcome. In patients with midgut NETs and pancreatic NETs with PD at baseline, radiologic disease control was observed in 84% and 81%, respectively.

For the entire group of 443 NET patients, the median OS was 63 months (95% CI, 55–72 months). The median PFS was 29 months (95% CI, 26–33 months). The median TTP was 36 months (95% CI, 32–40 months; Table 2). Patients with a primary NET in the pancreas had the longest OS (Fig. 2). We identified several risk factors associated with shorter OS (Table 3). The OS was significantly shorter in patients with liver or bone metastases at baseline. Also, patients with increased alkaline phosphatase (ALP) levels (>120 IU/L) and patients with extensive disease as scored with the OctreoScan had a worse prognosis (Table 3).

For the group of 239 NET patients with objective PD at baseline, the median PFS was 30 months (95% CI, 27–33 months), the median TTP 36 months (95% CI, 32–41 months), and the median OS was 58 months (95% CI, 52–64 months).

Characteristics of patients with NETs of the midgut and PD at baseline are presented in Table 4. For comparison with the NETTER-1 study, patients that received a cumulative dose of ≥100 mCi (3.7 GBq) ¹⁷⁷Lu-DOTATATE were selected. The PFS of patients with progressive midgut NETs was 24 months (95% CI, 18–30 months). OS was 46 months (95% CI, 32–60 months). The OS was significantly worse for patients with PD at baseline as compared with patients without PD at baseline (Table 2; Supplementary Fig. 3).

The results of this study demonstrate that PRRT with ¹⁷⁷Lu-DOTATATE is an excellent therapeutic option for patients with advanced grade 1 to 2 GEP and bronchial NETs. This treatment has limited side-effects and is relatively safe.

In the last decade, various types of targeted therapies have been introduced. After the presentation of the PROMID study (18) in 2009 and the CLARINET study in 2014 (19), the use of somatostatin analogues for progressive, inoperable grade 1–2 GEP-NETs was registered. Rinke and colleagues (18) demonstrated that the median TTP in patients with metastatic midgut tumors is longer (14.3 months) with the use of 30 mg Octreotide LAR per 4 weeks as compared to placebo (6.0 months), although there seemed to be no significant effect of Octreotide LAR on the OS (20). Also for Lanreotide Autogel 120 mg per 4 weeks, Caplin and colleagues have demonstrated that the PFS in GEP-NETs with a Ki67 index < 10% was better (32.8 months) as compared with placebo (18 months; refs. 19, 21).

Recently, the first results of the NETTER-1 study were published (12). This is the first randomized phase III study comparing ¹⁷⁷Lu-DOTATATE to high dose octreotide LAR therapy in patients with inoperable, progressive midgut NETs. The dosing protocol was comparable with our protocol with four administrations of 200 mCi (7.4 GBq) ¹⁷⁷Lu-DOTATATE every 8 weeks. The PFS for the control group receiving 60 mg Octreotide LAR per 4 weeks was 8.4 months and the PFS was not reached for the patients receiving ¹⁷⁷Lu-DOTATATE plus 30 mg Octreotide LAR per 4 weeks. For comparison with the NETTER-1 study, we selected patients with progressive midgut NETs, who received at least the same minimum cumulative dose of ≥100 mCi (3.7 GBq) of ¹⁷⁷Lu-DOTATATE. Most patient characteristics were not significantly different from patients in the NETTER-1 study (Table 4). However, tumor regional distribution seemed to be higher and tumor uptake on OctreoScan lower in our patients. The latter may be affected in the NETTER-1 study by the high dose Octreotide LAR pretreatment (22).

The proliferation marker Ki67 is used for the grading of NETs (23–25). In our study, the results of the Ki67 index labeling was not available in all tumors, because we started to routinely use the Ki67 index for grading of NETs in 2007. To have an estimation of the presence of high-grade NETs in our patient cohort, we took a sample of 230 GEP-NET patients who were treated after 2006. Of these 230 patients, 88 patients (38%) had WHO grade 1 NETs, 131 patients (57%) had grade 2 NETs, and 11 patients (5%) had grade 3 NETs (23–25). Because mainly grade 1–2 NETs were treated, we can conclude that PRRT is especially a therapeutic option for these tumors. Moreover, the majority of grade 3 tumors are negative on OctreoScan (26) and therefore cannot be treated with PRRT with ¹⁷⁷Lu-DOTATATE.

For progressive, metastatic grade 1–2 pancreatic NETs, both everolimus and sunitinib are approved therapies. In the RADIANT-3 study (27), comparing 10 mg everolimus orally daily to placebo, the PFS was 11.0 months for everolimus. In the study comparing sunitinib 37.5 mg orally daily to placebo (28), the PFS for sunitinib was 11.4 months. The PFS for pancreatic NETs with radiologic progression at baseline (Table 2) after therapy with ¹⁷⁷Lu-DOTATATE was 31 months and suggests being longer than with the currently available targeted therapies.

Recently, the results of the RADIANT-4 study (29) have also been published. This study compared everolimus 10 mg orally daily to placebo in patients with advanced grade 1–2 NETs of the lung and gastrointestinal origin. The PFS for everolimus was 11.0 months and thus comparable with the PFS for everolimus in the RADIANT-3 study. We found a PFS after PRRT with ¹⁷⁷Lu-DOTATATE of 29 months in the total group of patients with primary NETs in the lung or gastrointestinal tract. Although our patients were not randomized at inclusion, the current results can be compared with other studies due to the large number of patients and very long follow-up period. Therefore, not only in patients with pancreatic, but also in patients with bronchial and gastrointestinal NETs, the PFS seems to be better after ¹⁷⁷Lu-DOTATATE than the current targeted therapies which are registered for these indications.

A disadvantage of PRRT in general is the potential side-effects, especially on the bone marrow and kidneys. Because of coinfusion of lysine and arginine starting just before therapy, the radiation dose to the kidneys can be lowered and, therefore, the kidney is no longer the dose-limiting organ (13) with ¹⁷⁷Lu-DOTATATE. We have observed renal failure in six patients during follow-up after this therapy. The cause of this renal failure was probably not related to PRRT, as we found other more plausible causes in all patients. Acute leukemia and MDS are severe complications related to PRRT and occurred after a median of 28 months after the first cycle of PRRT with ¹⁷⁷Lu-DOTATATE for MDS and after a median of 55 months for acute leukemia. Although none of our patients who were diagnosed with acute leukemia/MDS received prior chemotherapy, recent reports suggest that there might be a higher risk of MDS/acute leukemia after alkylating chemotherapy (30–32).

Other therapy-related side-effects are hormone related problems. In this study, we did not focus on hormonal crises after PRRT. In a previous study, we have reported on 6 of 479 patients (1%) with GEP-NETs and pheochromocytomas who developed severe symptoms after PRRT due to a release of bioactive substances (33). When treating patients with hormone-producing tumors, these side-effects should be taken into account as well.

Various centers in Europe use [⁹⁰Y-DOTA⁰,Tyr³]-octreotide (⁹⁰Y-DOTATOC) for PRRT. The results of therapy with this radiopharmaceutical have been extensively reported in the last decade (7, 11, 34–36). Because of the higher energy of Yttrium-90, as compared with Lutetium-177, more side-effects have been reported. Imhof and colleagues (36) reported transient grade 3/4 hematologic toxicity in 12.8% of patients and permanent grade 4/5 renal toxicity in 9.2% with ⁹⁰Y-DOTATOC. In our opinion, ¹⁷⁷Lu is therefore the radionuclide of first choice for treating patients with NETs with PRRT because of less renal toxicity.

In 2008, we reported on 310 GEP-NET patients who were treated with ¹⁷⁷Lu-DOTATATE (8). Although part of the patients reported in the current article was included in the 2008 study, we believe that the current results are more distinctive and more unique. In contrast to the former report, in this cohort only Dutch patients with a very long follow-up in our institution were included and results of various subtypes of GEP-NETs are reported. The relative low loss to follow-up makes the current results more solid.

For this study no intention-to-treat analysis was made. We believe that an analysis of all 1,214 patients is not reliable. This group includes patients from different countries who were only treated at our institute and all follow-up was conducted in their own country. No data about this follow-up are available. This group also includes patients with different kind of malignancies, for example, thyroid cancers, paragangliomas and lymphomas. This heterogeneity will make an analysis not reliable and not representative for the investigated group of bronchial and GEPNET patients.

The main limitation of this study is the fact that it is a nonrandomized study. However, with the selection of patients that were all treated strictly according to the inclusion criteria and an active follow-up during many years make the results very reliable. Also the inclusion of patients with stable disease at baseline is a limitation compared with other studies reporting on NETs. Patients with a very large tumor load or many symptoms of the tumor or its produced hormones were treated, because PRRT was the best available treatment option at that point in time. However, analysis of the patients with PD at baseline demonstrated only small differences in PFS, TTP, and OS compared with all GEPNET patients.

PRRT with ¹⁷⁷Lu-DOTATATE produces good tumor response rates for patients with grade 1–2 GEP-NETs and bronchial NETs. The side-effects are limited and in the vast majority of patients reversible on short term. Severe long-term toxicities are acute leukemia or MDS, occurring in 2% of patients. No therapy-related long-term renal or hepatic failure was observed. Although this was a nonrandomized study, the PFS and TTP for ¹⁷⁷Lu-DOTATATE are favorable to all other registered medical options that are currently available for patients with advanced or metastasized G₁–G₂ GEP-NETs and bronchial NETs.

E.P. Krenning holds ownership interest (including patents) in and is a consultant/advisory board member for Advanced Accelerator Applications. D.J. Kwekkeboom held ownership interest (including patents) in Advanced Accelerator Applications. No potential conflicts of interest were disclosed by the other authors.

Conception and design: T. Brabander, W.A. van der Zwan, W.W. de Herder, E.P. Krenning, D.J. Kwekkeboom

Development of methodology: T. Brabander, W.A. van der Zwan, W.W. de Herder, E.P. Krenning, D.J. Kwekkeboom

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T. Brabander, J.J.M. Teunissen, B.L.R. Kam, R.A. Feelders, W.W. de Herder, C.H.J. van Eijck, G.J.H. Franssen, E.P. Krenning, D.J. Kwekkeboom

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T. Brabander, W.A. van der Zwan, W.W. de Herder, E.P. Krenning, D.J. Kwekkeboom

Writing, review, and/or revision of the manuscript: T. Brabander, W.A. van der Zwan, J.J.M. Teunissen, B.L.R. Kam, W.W. de Herder, C.H.J. van Eijck, E.P. Krenning, D.J. Kwekkeboom

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T. Brabander, G.J.H. Franssen, D.J. Kwekkeboom

Study supervision: D.J. Kwekkeboom

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.