Efficacy and Safety of [177Lu]Lu-DOTA-TATE in Adults with Inoperable or Metastatic Somatostatin Receptor-Positive Pheochromocytomas/Paragangliomas, Bronchial and Unknown Origin Neuroendocrine Tumors, and Medullary Thyroid Carcinoma: A Systematic Literature Review

Background: We have performed a systematic review to evaluate the efficacy and safety of [177Lu]Lu-DOTA-TATE, a radioligand therapy, in advanced somatostatin receptor-positive pheochromocytoma/paraganglioma (PPGL), thymic neuroendocrine tumor (NET), bronchial NET, unknown primary NET, or medullary thyroid carcinoma (MTC). Methods: Studies identified in PubMed from inception to 13 May 2021 must have assessed [177Lu]Lu-DOTA-TATE as a single agent and reported outcome data for the specific NET types of interest. Results: Two independent reviewers performed the screening and data extraction, resulting in 16 publications: PPGL (n = 7), bronchial NETs (n = 6; one also included NETs of unknown origin), and MTC (n = 3). Overall, [177Lu]Lu-DOTA-TATE offers encouraging antitumor activity (overall tumor response rates and disease control rates) across NET types. Safety was favorable with most adverse events mild to moderate in severity, transient, and consistent with those seen in patients with gastroenteropancreatic (GEP)-NETs. Conclusions: [177Lu]Lu-DOTA-TATE has been used effectively in clinical practice to treat NETs of non-GEP origin.


Introduction
Neuroendocrine neoplasms (NENs) are rare tumors that are derived from sensory and secretory neuroendocrine cells and can occur at almost any anatomical site [1,2]. Neuroendocrine tumors (NETs) are a subset of well-differentiated NENs that widely express the somatostatin receptor (SSTR), particularly Subtypes 2 and 5 [1][2][3]. The most common location of NETs is the gastrointestinal tract, including the pancreas [4]. Many NETs have malignant potential and commonly metastasize before diagnosis, complicating management and limiting the potential for curative surgery [3].
Although most published evidence and experience with [ 177 Lu]Lu-DOTA-TATE is in GEP-NETs, [ 177 Lu]Lu-DOTA-TATE may be of benefit to patients with SSTR-expressing NETs that arise in other locations [9]. Well-differentiated bronchial NETs, the second most common type of NET, are classified as typical (low-grade) or atypical (intermediategrade) carcinoids, with the same terminology used for the much less common thymic NETs [1,2]. Up to 70% of bronchial NETs express SSTRs [10]. For some patients (9-14%) with metastasized NETs, the primary tumor is unknown [11,12]. The single-arm ERASMUS Phase I/II study evaluated [ 177 Lu]Lu-DOTA-TATE in patients with a variety of NET types, including midgut, hindgut, pancreatic, bronchial, and unknown origin [13]. A retrospective analysis of Dutch patients with bronchial and GEP-NETs enrolled in ERASMUS and treated with [ 177 Lu]Lu-DOTA-TATE (n = 443) reported an overall response rate (ORR) of 39% and a median PFS of 29 months [13].
There is currently a high unmet medical need for bronchial NETs, thymic NETs, NETs of unknown origin, and PPGL, with limited approved therapeutic treatment options [16][17][18][19]. The National Comprehensive Cancer Network ® (NCCN) includes treatment with [ 177 Lu]Lu-DOTA-TATE as an option for patients with SSTR-positive bronchial or thymic NETs who have progressed on standard-of-care regimens and as a primary treatment for SSTR-positive PPGL with distant metastases. As for NETs of unknown origin, NCCN recommends that they are treated similar to GEP-NETs [19]. Guidelines from the European Society for Medical Oncology (ESMO) also include peptide receptor radionuclide therapy (PRRT) as a potential therapy for patients with metastatic bronchial or thymic NETs (third-line) [16], advanced PPGL (after progression on standard of care or for those with high tumor burden) [17], and in select cases of MTC, such as when multikinase inhibitor therapy is contraindicated [18]. To further understand evolving clinical practice and describe the use of [ 177 Lu]Lu-DOTA-TATE as a treatment for NETs of non-GEP origin and unknown origin, we performed a systematic literature review to identify and summarize published evidence for the efficacy and safety of [ 177 Lu]Lu-DOTA-TATE in adult patients with inoperable or advanced SSTR-positive PPGL, thymic NET, bronchial NET, unknown primary NET, or MTC (collectively known as non-GEP-NETs for the purpose of this review).

Search Strategy
PubMed was searched for studies published up to 13 May 2021. No geographic, language, or age restrictions were applied in the search, but only English-language publications reporting studies in adults were selected for inclusion in the systematic review. The search and report were completed according to the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) 2020 statement [20]. The review protocol was registered with INPLASY (INPLASY202330030).
The search strategy included both Medical Subject Headings terms and free-text terms, and included variants of the terminology for the NETs, such as cancer, carcinoma, carcinoid, tumor, tumour, and neoplasm, as well as variants for LUTATHERA ® , such as

Study Selection Process
The titles and abstracts of records identified on PubMed were screened initially to exclude any duplicate records and those that were clearly not relevant (Supplementary Materials S2). A more detailed screen of the remaining records was then performed, using the inclusion and exclusion criteria below, to select records for the further evaluation of the full text to establish eligibility. The selection process was performed by two reviewers independently and any differences were resolved by consensus. In addition, the reference lists of reviews were examined to identify additional studies that had not been detected by the initial search strategy.

Inclusion and Exclusion Criteria
This systematic review considered studies that included adults (as defined by the authors) with any of the following inoperable or metastatic SSTR-positive NETs: PPGL, thymic NET, bronchial NET, NET of unknown primary origin, or MTC. Studies that included multiple NET types were only included if the results and baseline characteristics were provided for individual NETs. Studies that included both pediatric and adult patients were retained, if it was possible to extract data for adults only.
Articles not written in English were excluded. Studies reporting the efficacy or safety of [ 177 Lu]Lu-DOTA-TATE in combination with other anticancer treatments, including other radioligand therapies, were excluded (except for SSAs, amino acids, and antiemetic agents, per the treatment regimen) [5]. In addition, studies were excluded where relevant outcome data (response rates, survival time, or safety) were not available for the specific NET types of interest or the specific radioligand treatment of interest. Individual case reports were excluded, as were case series that only reported relevant tumor types in a single patient.

Types of Intervention
Studies that evaluated the efficacy and safety of any [ 177 Lu]Lu-DOTA-TATE treatment administered as a single agent only were included.

Types of Outcome
The following efficacy or safety outcomes were analyzed: PFS, time to tumor progression (TTP), disease control rate (DCR), response rates (as reported in each study), overall survival (OS, as reported in each study), mortality, and adverse events (AEs) by organ and type.

Types of Study
Randomized controlled trials, non-randomized controlled trials, quasi-experimental studies, prospective and retrospective cohort studies, and case series (if the patients were analyzed or analyzable as a group) were included.

Data Extraction and Synthesis
Qualitative and quantitative data were extracted from the studies by two reviewers independently, and descriptive analyses performed. Efficacy data are presented by NET subtype and safety data are summarized for all NETs analyzed. If not reported, ORR was assessed by integrating the number of patients with complete and partial radiologic response by the total number of patients treated.

Data Extraction and Synthesis
Qualitative and quantitative data were extracted from the studies by two reviewers independently, and descriptive analyses performed. Efficacy data are presented by NET subtype and safety data are summarized for all NETs analyzed. If not reported, ORR was assessed by integrating the number of patients with complete and partial radiologic response by the total number of patients treated.
More patients had a PGL diagnosis (n = 81 [84%]) than PCC (n = 15 [15%]), with one patient diagnosed with PGL and PCC. The age range of treated adults was 22-84 years, and most patients had received prior surgery and had metastatic disease at baseline (Table 1). In general, four cycles of [ 177 Lu]Lu-DOTA-TATE treatment (5.55-7.4 GBq per cycle) were administered, but some patients received less, and one study recorded more, with extra cycles given as salvage therapy (overall, 1-11 cycles were administered; Table 1). Some of the primary reasons for not receiving the recommended four cycles of therapy included toxicity and disease progression.
Five of the seven studies assessed the radiologic response using RECIST 1.1 (Response Evaluation Criteria In Solid Tumors), one used SWOG (Southwest Oncology Group) criteria, and one did not specify ( Table 2). Partial response (PR) was recorded in six of the seven studies, ranging from 7% (1/14) to 25% (1/4), but no complete response (CR) was reported. ORRs calculated from response data ranged from 0% to 25%. Three studies included minor response (MR) in the response criteria, and this ranged from 8% (1/12) to 29% (4/14). DCRs ranged from 67% to 100%. Overall, out of the 97 patients with PPGL treated with [ 177 Lu]Lu-DOTA-TATE, 13 achieved a PR, 7 had MR, 64 had stable disease (SD), and 12 had progressive disease (PD) ( Table 2). Response was unknown for one patient.
Two studies evaluated PGL and PCC separately [24,25]. Best response to [ 177 Lu]Lu-DOTA-TATE was evaluated in 37 patients with PGL in total and was PR (n = 7), MR (n = 4), SD (n = 24), and PD (n = 2). The best responses in five patients with PCC were PR (n = 1), SD (n = 1), and PD (n = 3). SD was observed in one patient with PCC plus concomitant pancreatic NET (pNET), and in one patient with PGL plus PCC and concomitant pNET ( Table 2).
Most of the patients had metastatic disease at baseline with liver and bone identified as common metastatic sites. Four of the six studies recorded a carcinoid subtype, and between 31% and 77% of patients were diagnosed with atypical carcinoid (AC) tumors. The age range of treated adults was 37-79 years, and most patients had received prior surgery. Patients received between 1 and 8 cycles of [ 177 Lu]Lu-DOTA-TATE (3.7-7.8 GBq per cycle; Table 1). Some studies stratified patients according to risk factors for kidney and bone marrow toxicity and reduced the dose of [ 177 Lu]Lu-DOTA-TATE per cycle; however, the cumulative administered activity was similar across all the studies.
Four of the six studies assessed radiologic response using RECIST, and two used SWOG or modified SWOG criteria ( Table 2). Of the patients assessed for tumor response, most studies reported PR (12-56%) or MR (11-17%) as best response following [ 177 Lu]Lu-DOTA-TATE treatment, while one patient achieved a CR. SD was observed in 22-83% of patients. ORRs calculated from the response data ranged from 13% to 56%, and DCRs of between 62% and 100% were observed (Table 2). One study analyzed the outcome by the histopathological subtype and reported a DCR of 80% for typical carcinoid (TC) tumors (n = 15) and 47% for AC tumors (n = 19) [28]. Overall, five studies reported analyzable response data for 119 patients treated with [ 177 Lu]Lu-DOTA-TATE, with one achieving CR, 22 PR, 9 MR, and 54 SD ( Table 2).

NETs of Unknown Primary Origin
Eight patients, aged between 54 and 80 years old, with metastatic NETs of unknown primary origin were included in a Swedish prospective study [30]. Seven of the eight patients had extensive disease and most (87.5%) had received prior chemotherapy. Patients received between two to seven cycles of 7.4 GBq [ 177 Lu]Lu-DOTA-TATE until the absorbed dose to the kidneys reached 23 Gy or until there were other reasons for stopping ( Table 1). The best response by RECIST 1.1 was PR (38%), SD (50%), and PD (13%) ( Table 2), giving an ORR of 38% and a DCR of 88%. The median PFS from treatment initiation was 17.5 (95% CI 7-34) months and the median OS was 43 months (95% CI 15 months-NR) ( Table 2) [30].

MTC
The study and patient characteristics for the three studies that analyzed the data of 62 patients (7-43 per study) with MTC are summarized in Table 1 and the results in Table 2. One of the three studies was a prospective, single-arm interventional study [34], and the other two were retrospective, single-center case series [35,36].
Overall, patients were aged between 19 and 80 years and all patients had metastases at baseline, with approximately one in three patients having extensive metastatic disease (Table 1). In general, three or four cycles of [ 177 Lu]Lu-DOTA-TATE treatment (5.55-7.4 GBq per cycle) were administered ( Table 1). The intended cumulative activity was up to 29.6 GBq. The average cumulative activity was lower at 18.5 GBq for one study, which administered an average of three cycles of the lower dose of [ 177 Lu]Lu-DOTA-TATE (5.55 GBq per cycle).
Tumor response was assessed using RECIST 1.1 in all studies. PR was recorded in two of the three studies (5% [2/43] and 43% [3/7]), but no CR was observed. DCRs ranged between 40% and 86%. Overall, out of the 60 patients with MTC that were treated with [ 177 Lu]Lu-DOTA-TATE, five achieved a PR, 32 had SD, and 23 had PD ( Table 2). The median PFS was measured in two studies and was 24 (95% CI 15.1-32.9) months and 0.7 (range 0.3-12.0) years. The relatively short PFS observed in the latter study (0.7 years, 8.4 months) was attributed to many of the patients having poor prognostic features including paraneoplastic endocrine syndrome, loss of tumor marker expression, and old age (median 62 years) [36]. The respective OS data were 26 (95% CI 16.6-35.3) months and 1.14 (range 0.4-12.0) years [35,36].

Safety of [ 177 Lu]Lu-DOTA-TATE
Three studies reported safety on populations that included patients with NETs or SSTR-positive tumors that are not the primary focus of this review. Specific data for PPGL, bronchial NETs or NETs with unknown primary origin could not be delineated, and for this reason, these three studies have been excluded from the safety analysis [22,29,30]. Safety data from the remaining 13 studies for patients treated with [ 177 Lu]Lu-DOTA-TATE (n = 269) are summarized in Table 3. One study reported that persistent thrombocytopenia limited the number of cycles of [ 177 Lu]Lu-DOTA-TATE administration in three patients, so they received only two or three cycles. When reported, AEs were generally mild and resolved without sequelae. One study described acute toxicity of nausea (34%), pain (23%), and vomiting (13%) after the administration of [ 177 Lu]Lu-DOTA-TATE.
Of the 11 studies that reported on hematologic toxicity, most events were mild; only three studies, encompassing 70 patients, recorded 14 events of Grades ≥3. Thrombocy-topenia, leukopenia, and anemia were the most frequently cited events (Table 3). Eight studies included data on renal toxicity, recording either no nephrotoxicity or no high-grade nephrotoxicity.
Few high-grade AEs were reported. In a retrospective case series (n = 30), two patients experienced reversible cardiac AEs after their first [ 177 Lu]Lu-DOTA-TATE cycle. One patient with PGL and lung and bone metastases experienced cardiac failure, possibly due to chronic catecholamine release. Another patient with PCC developed pleural effusion and delirium that may have resulted from cardiac failure or catecholamine release. Both patients made full recoveries and successfully received further cycles of [ 177 Lu]Lu-DOTA-TATE treatment [24]. In a small retrospective case series (n = 5), one patient with PGL that had metastasized to the lungs, bone, and mediastinal and supraclavicular lymph nodes received three cycles of [ 177 Lu]Lu-DOTA-TATE and stopped further treatment after experiencing suspected pneumonitis considered plausibly associated with treatment [21]. In a different retrospective study, one patient with bronchial NET and mediastinal invasion developed Grade 3 radiation pericarditis following two cycles of [ 177 Lu]Lu-DOTA-TATE. After a successful pericardiocentesis procedure; this patient went on to receive a further two cycles of [ 177 Lu]Lu-DOTA-TATE [32]. One patient with MTC developed Grade 3 hemoptysis, which was attributed to the progression of pulmonary metastases [36].
Overall, secondary hematologic malignancies were reported in one patient who was treated for PPGL. This individual developed MDS, which was considered to be possibly related to the [ 177 Lu]Lu-DOTA-TATE treatment (cumulative dose 44.4 GBq), given that they had not received prior chemotherapy and did not present with bone metastases [24].

Discussion
Sixteen studies involving 304 patients were identified and provided data for the efficacy analysis of the [ 177 Lu]Lu-DOTA-TATE treatment in SSTR-positive PPGL (seven studies, 97 patients), bronchial NETs (six studies, 137 patients), NETs of unknown primary origin (one study, eight patients), and MTC (three studies, 60 patients). One publication reported data for bronchial and unknown primary NETs. The safety analysis included 269 patients; three studies did not report safety data by a NET subtype and so were excluded from the safety analysis. The dosing of [ 177 Lu]Lu-DOTA-TATE was similar for all indications with most studies recording an average of four cycles and administering 7.4 GBq per cycle.
Most of the studies assessed radiologic response using the RECIST 1.1 criteria. Based on the SWOG criteria, a CR was achieved by one patient with bronchial NET. ORRs, calculated using response data as measured by RECIST 1.1, varied from 0% (PPGL study with six patients and an MTC study with 10 patients) to 43% (MTC study with seven patients) but the majority of studies (9/11) had ORR rates of up to 27%. This is broadly in agreement with the ORR of 18% (one CR and 17 PR) observed with [ 177 Lu]Lu-DOTA-TATE treatment in patients with advanced midgut NETs enrolled in the NETTER-1 study (101 evaluable patients) [8]. The highest calculated ORR, as measured by the modified SWOG criteria, was 56% (bronchial NETs study with nine patients). Calculated ORRs assessed using RECIST 1.1 for bronchial NETs were 17% and 27%, lower than that reported in the ERASMUS study (30%) [37]. In this study, calculated DCRs (RECIST 1.1) with [ 177 Lu]Lu-DOTA-TATE were slightly higher for PPGL, bronchial NETs, and NETs of unknown origin (67-100%, 68-100%, and 88%, respectively) compared with MTC (40-86%). Recent meta-analyses of [ 177 Lu]Lu-DOTA-TATE in advanced NETs of mixed origins reported ORRs of between 25% and 33% and DCRs of between 74% and 83% [38][39][40], similar to those identified for each NET subtype analyzed in this study, suggesting that NET origin does not dramatically affect the efficacy of [ 177  Survival outcomes were not consistently reported in the studies analyzed. Median PFS, when reported, ranged from 0.7 years (8.4 months; MTC study) to 31 months (bronchial NETs), with no obvious pattern according to NET subtype. The latter is comparable to those observed in the ERASMUS clinical trial setting when the overall median PFS was 29 months, with 20 months recorded for bronchial NETs and 29 months for NETs of unknown origin [13]. The patient population in the MTC study that reported a relative short PFS of 0.7 years (8.4 months) included a number of factors associated with a poor prognosis (paraneoplastic endocrine syndrome, loss of tumor marker expression, and older age) that the authors believed may have contributed to the poor survival outcome [36]. The follow-up periods for many studies were not long enough to measure median OS, but for those studies that did report data, the values were comparable for three NET subtypes (49.6 months for PPGL, 42-48.6 months for bronchial NETs, 43 months for NETs of unknown origin) but shorter for MTC (26 months and 1.14 years [13.7 months]). NETTER-1 reported a median OS value of 48.0 months for patients with midgut NETs, which is consistent with those recorded for PPGL, bronchial NETs, and NETs of unknown origin in this analysis [7].
The survival benefits observed with [ 177 Lu]Lu-DOTA-TATE in non-GEP-NETs are encouraging when compared with the current standard of care for these indications. In PPGL, [131I]meta-iodobenzylguanidine (MIBG) and systemic chemotherapy are used to treat progressive disease [17]. In a Phase II study, the median OS with [131I]-MIBG was 36.7 months, and a retrospective analysis of patients receiving systemic chemotherapy reported a median OS of 6.4 years (76.8 months) for responders and 3.7 years (44.4 months) for non-responders [41,42]. Everolimus is often used as first-line therapy to treat patients with bronchial NETs who have progressed on SSAs [16]. In subgroup analyses of patients with progressive bronchial NETs and NETs of unknown origin from the RADIANT-4 study, the median PFS observed with everolimus were 9.2 and 13.6 months, respectively. OS data for RADIANT-4 were immature [43,44]. Recurrent MTC has a significant impact on survival rate, and current systemic treatments have limited effect on response and OS [18,45]. The tyrosine kinase inhibitors vandetanib and cabozantinib are considered first-line systemic options to treat progressive metastatic MTC [18,46]. The Phase III ZETA and EXAM studies in advanced MTC reported a median PFS of 30.5 months for vandetanib (predicted) and 11.2 months for cabozantinib, respectively [47,48]. The OS for cabozantinib in the EXAM study was 26.6 months [49]. The comparable survival outcomes of [ 177 Lu]Lu-DOTA-TATE and standard therapies for advanced non-GEP-NETs highlight the potential clinical importance of [ 177 Lu]Lu-DOTA-TATE as a treatment option for these patients with a poor prognosis.
In this review, safety data were not reported consistently by the studies. In some studies where efficacy was separated by NET subtype, the safety data were presented for the whole population, making it difficult to draw conclusions by NET subtype. However, in general, the mean cumulative dose of [ 177 Lu]Lu-DOTA-TATE received by patients was similar and most AEs were of mild or moderate severity. In general, renal and bone marrow toxicities, when observed, were not clinically significant and the safety profiles observed were consistent with data from the NETTER-1 and ERASMUS trials [7,8,13]. A systematic review of nephrotoxicity after PRRT in different types of NETs concluded that there was a greater risk of long-term kidney damage with 90Y-labeled SSA analogs compared with those labeled with 177 Lu [50]. Overall, in our analysis, nephrotoxicity severity, when reported, was low, but the studies varied considerably in the reporting of results and in the length of follow-up. Similarly, the reporting of secondary hematologic malignancies was infrequent, but the follow-up periods could have been too short to identify cases. A recent evaluation of treatment-related myeloid neoplasms in patients with NETs after PRRT concluded that the risk was low, but that patients should be monitored closely [51].
The initial search strategy for this review included thymic NETs. Only one retrospective case series assessing [ 177 Lu]Lu-DOTA-TATE in 27 patients with mediastinal NETs was identified, but it failed to meet the inclusion criteria for our review because pediatric cases were included and efficacy data for thymic and mediastinal tumors were presented together [52]. The scarcity of data indicates that further studies in thymic NETs may be warranted to understand the potential clinical benefit of [ 177 Lu]Lu-DOTA-TATE in these patients.
For this review, the search strategy did not have any language or time limits and was conducted by two independent reviewers to avoid selection bias. Many of the included studies were retrospective in design, with inherent possible selection and detection biases regarding the cases included in the analyses. The studies did report similar cumulative doses, and most studies used the RECIST 1.1 criteria to assess response. Limitations of this systematic literature review include the single database search, absence of data from randomized clinical trials, heterogeneous study designs, limited patient numbers per study, different follow-up periods, and the inconsistent reporting of data across the studies regarding baseline characteristics, treatment, treatment outcomes, and safety. Despite the heterogenous nature of the studies, the results indicate that the efficacy of [ 177 Lu]Lu-DOTA-TATE in adult patients with metastatic, progressive, SSTR-positive PPGL, bronchial NETs, NETs of unknown primary origin, and MTC was consistent with the NETTER-1 and ERASMUS studies. These data support the inclusion of [ 177 Lu]Lu-DOTA-TATE as a treatment option for non-GEP-NETs, as outlined in NCCN and ESMO guidelines [16][17][18][19].

Conclusions
This systematic literature review has shown that [ 177 Lu]Lu-DOTA-TATE is used in clinical practice for the treatment of patients with PPGL, bronchial NETs, NETs of unknown primary origin, and MTC. When reported, the cumulative administered dose in the studies was mostly similar to that approved for [ 177