Norepinephrine Transporter as a Target for Imaging and Therapy

¹²³I/¹³¹I-MIBG for Imaging of Neuroblastoma

Neuroblastoma is a tumor that arises from primordial neural crest cells and is predominantly seen in infants and young children. Neuroblastoma cells show high NET expression, and most neuroblastomas are MIBG-avid (Fig. 2) (8). Most of the iodine localizes intracellularly within the mitochondria, and lesser amounts are seen in the vesicles and nuclear membrane. The decay characteristics of ¹³¹I—with emissions of energy that allow for penetration in the nucleus—make it suitable for therapy (9), whereas the lower γ-energy of ¹²³I makes it a preferred isotope for imaging.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Imaging with ¹²³I-MIBG and ¹⁸F-MFBG. ¹²³I-MIBG whole-body anterior (A) and posterior (B) images show minimal uptake in bilateral femora (arrows). These lesions are better visualized on ¹⁸F-MFBG PET/CT imaging (C), which also shows additional lesions (blue arrows) not seen on ¹²³I-MIBG imaging.

¹³¹I/¹²³I-MIBG is widely used for imaging neuroblastoma and is considered the gold standard for staging and follow-up of neuroblastoma patients (10,11). Although initially ¹³¹I-MIBG was used for imaging, it has been replaced to a large extent by ¹²³I-MIBG, which, even though more expensive, is more favorable due to its shorter half-life and lower radiation dose that allows for higher injected activity and for superior quality of both planar and SPECT imaging.

MIBG scans have a high sensitivity for disease localization, with detection of disease in about 93%–97% of neuroblastoma patients, and have been recommended by the International Neuroblastoma Risk Group for evaluation of bone and bone marrow disease at initial staging and follow-up. It is more specific than MRI and has replaced bone scanning in routine evaluation (11–13). The reported sensitivity of ¹²³I- or ¹³¹I-MIBG for staging or recurrent neuroblastoma ranges from 76% to 100% (14–16). A prospective multicenter trial reported sensitivity of 88% and specificity of 83% for detection at diagnosis or follow-up (17). ¹²³I-MIBG is reported to detect 82% of unsuspected relapses, including 27% that were otherwise undetected by other modalities (18).

Imaging with MIBG is also prognostic and influences management in neuroblastoma. Disease extent and involvement of bone and marrow as determined by MIBG is associated with unfavorable outcome (19,20). For evaluating treatment response, an early response to chemotherapy noted on MIBG scanning denotes good prognosis and correlates with both event-free survival as well as overall survival in high-risk and advanced/stage IV neuroblastoma (21). On the other hand, a lack of or incomplete response on MIBG scanning is an indicator of unfavorable outcome (22,23), thereby guiding management decisions, especially in those receiving dose-intensive therapy at diagnosis. After therapy, false-negative MIBG scans may be seen in 24%–41% of patients probably related to changes in the NET system and trapping mechanisms caused by maturation and differentiation of tumor cells or due to the effect of chemotherapeutic agents (24–26). Imaging with ¹²³I-MIBG is required to document uptake by tumor tissue and select patients for ¹³¹I-MIBG therapy. Generally a positive scan for lesion detection and good uptake in lesions would denote suitability for therapy; liver uptake serves as a good index to compare and assess suitability for therapy. Although planar imaging is easy and requires less time, SPECT allowed for improved sensitivity of detection (27) and also localization of lesions, especially with SPECT/CT. In 75% of patients, a higher number of lesions were detected with both ¹³¹I-MIBG and ¹²³I-MIBG (28), and more soft-tissue lesions were seen in 25% of patients (29) by SPECT/CT than by planar imaging. It is recommended that stage 3 and 4 neuroblastoma patients be routinely imaged with SPECT or SPECT/CT, though it does require longer time for acquiring the scans, which may be a disadvantage in younger children, especially those requiring anesthesia for imaging.

A more comprehensive assessment of disease burden and of number and sites of disease involvement helps assess suitability for MIBG therapy. A solitary large lesion or the presence of only a few lesions involving soft tissue may be more appropriately treated with surgery, if feasible, or with localized therapy, whereas a larger number of lesions with wider distribution would benefit from treatment with ¹³¹I-MIBG. A baseline and follow-up assessment with ¹²³I MIBG allows for assessment of the response to MIBG therapy. However, assessment of lesion changes and response by MIBG scanning is subjective and many times challenging due to the lack of quantitative measures. Semiquantitative scoring systems have been developed to provide an objective and uniform way of reporting lesions between readers (12,30,31) for evaluation of disease burden and of response. The Curie scoring system, proposed by Ady et al. in 1995, assigns a score of 0–3 to 9 designated regions of the whole body; a relative score compares the posttreatment score with the pretreatment value. This method is highly reproducible, and the end of the induction score is predictive of overall response with high accuracy (30). A modified Curie scoring system also includes soft-tissue lesions and is currently widely used in multicenter trials conducted by a children’s oncology group (31). A similar scoring system, proposed by the International Society of Paediatric Oncology Europe Neuroblastoma, uses a 0–6 grading scale for skeletal and soft-tissue disease (32) and is recommended by an international panel for universal routine evaluation of MIBG scanning in high-risk neuroblastoma patients (33). MIBG scoring has high prognostic value at diagnosis and is predictive of overall survival and event-free survival in high-risk and stage IV neuroblastoma after other systemic treatments and in assessment of disease before myeloablative therapy (34). It has shown utility in assessment of disease response to ¹³¹I-MIBG therapy and provides an objective criterion for assessing response and choosing patients for repeat therapies (31,35,36). MIBG scoring is recommended by the International Neuroblastoma Risk Group Task Force as essential for evaluation of efficacy of therapies (12).

¹²³I/¹³¹I-MIBG for Imaging of Pheochromocytoma (PHEO) and Paraganglioma (PGL)

PHEO and PGL originate in chromaffin tissue of the adrenal medulla or sympathetic nervous system and, if functional, secrete catecholamines that lead to various symptoms. In these tumors also, MIBG uptake in cells is facilitated by NET and stored in vesicular granules facilitated by vesicular monoamine transporters. Targeted imaging of PHEO/PGL with ¹³¹I-MIBG was first evaluated in 1987 (37). Over the years, a few studies have reported variable results, with many of these studies in a small number of patients. Detection of lesions may be lower in those with low-affinity binding of MIBG to NET, lack of storage granules, or fewer NET or vesicular monoamine transporters due to cell differentiation (38). The overall sensitivity of MIBG imaging is lower for nodal disease and liver lesions, probably related to size resolution and physiologic uptake. Additionally, lower sensitivity is reported in those with familial PGL and extraadrenal disease (39,40). False-negative scans are more frequent in those with the SDHB mutation or von Hippel–Lindau syndrome–associated PHEO (41), probably due to lower expression of NET. In a prospective study in 140 patients, ¹²³I-MIBG showed specificity of 82% and sensitivity of 88% for PHEO and 67% for PGL (42). Patients with multiple endocrine neoplasia syndrome may also be evaluated with ¹²³I-MIBG to detect concurrent lesions in adrenal and thyroid glands. The detection accuracy may be improved using a lesion uptake scoring system that grades the uptake in lesion to that of liver on a 1–4 scale (43). However, this has not been systematically evaluated for assessment of tumor burden in patients, and the role in follow-up of patients for therapy has not been studied.

¹²³I/¹³¹I-MIBG for Imaging of Carcinoids

The ability of ¹³¹I-MIBG to target and image carcinoid tumor was initially shown in 1984 (44). The overall sensitivity for detection is low (45), with midgut tumors visualized more frequently than foregut tumors. Higher MIBG uptake is seen with higher serotonin levels (46), suggesting an uptake and storage mechanism likely similar to serotonin. Somatostatin receptor–targeted agents are more sensitive in targeting carcinoids and are preferred agents.

Other NET-Targeting Tracers

PET imaging agents targeting NET may improve detection and quantitation of disease. Benzyl guanidine analogs labeled with ⁷⁶Br- and ¹²⁴I-MIBG have been developed; however, the labeling process is complex, and imaging is less optimal (47). ¹²⁴I-MIBG imaging in neuroblastoma and PHEO/PGL (48,49) is feasible; however, large studies validating its clinical use are lacking. The complex high-energy γ-emission compromises image quality and raises concern for radiation-absorbed doses; longer injection-to-imaging times and cost have limited its use. ¹⁸F-fluorine–labeled analogs as substrates of NET, such as 4-¹⁸F-fluoro-3-iodobenzylguanidine (50) and ¹⁸F-fluoropropylbenzylguanidine, have an advantage of imaging with a shorter-lived ¹⁸F isotope; however, again, there is limited clinical validation in oncology applications. In a single-patient report, ¹⁸F-fluoropropylbenzylguanidine showed biodistribution similar to that of ¹²³I-MIBG (51). Another fluorinated guanidine analog, ¹⁸F-metafluorobenzylguanidine (¹⁸F-MFBG), is being evaluated in humans in a prospective study (NCT02348749). Assessment in neuroblastoma and PHEO/PGL patients showed excellent targeting of lesions, with high contrast as early as 1 h after injection (Fig. 2) (52). Targeted imaging of NET mechanism with fluorine-labeled PET tracer offers a more convenient single-day imaging schedule; additionally imaging with PET/MRI would be highly desirable and most suitable for pediatric patients. ¹¹C-metahydroxyephedrine (HED) is a catecholamine analog (53) shown to target tumors of the sympathetic nervous system. Whole-body PET/CT with HED is feasible and safe for imaging neuroendocrine tumors (54) and provides a higher rate of detection of lesions (80/81) than ¹²³I-MIBG (75/81); however, soft-tissue lesions in neuroblastoma patients showed less uptake, including a reported large abdominal recurrence that was better visualized by ¹²³I-MIBG. Similarly, in a small number of patients with neuroblastoma (n = 7), lesion targeting with HED was seen within 5 min of administration; however, lesions in the abdomen were less well detected than with ¹²³I-MIBG. HED shows high uptake in liver that may affect detection of lesions (55). For PHEO, a high sensitivity of 92% and a specificity of 100% have been reported in only a small series of patients. Localization of both primary and metastatic lesions including those that were MIBG negative has been noted (56). Limitations of HED are the short half-life of isotope ¹¹C (20.3 min) for imaging and inferior image quality. Patient data are lacking and, at present, the role of HED for imaging NET remains investigative. The neurotransmitter dopamine is also transported by NET and is converted by dopamine β-hydroxylase to norepinephrine in noradrenergic neurons (57). An analog of dopamine, ¹⁸F-fluorodopamine (¹⁸F-FDA) has been synthesized (58) to image neuroendocrine tumors. In the neuronal vesicles, ¹⁸F-FDA also is converted to ¹⁸F-6-fluoronorepinephrine by dopamine β-hydroxylase or alternately metabolized by monoamine oxidase to ¹⁸F-6-fluorodihydroxyphenylacetic acid in the mitochondria (59). The radiolabeling process for ¹⁸F-FDA is also complex, limiting its availability and usage (60). Another PET tracer, 6-¹⁸F-fluoro-l-dihydroxyphenylalanine (¹⁸F-DOPA), used to image NE tumors, does not directly target the NET but targets the catecholamine synthesis pathway. Uptake of ¹⁸F-DOPA is regulated by l-amino acid decarboxylase that converts ¹⁸F-DOPA to ¹⁸F-FDA. Neuroendocrine cells strongly express l-amino acid decarboxylase enzyme and probably higher expression of amino acid transporters also to facilitate uptake. ¹⁸F-DOPA has been radiolabeled to yield a high-specific-activity tracer (61); however, the process is long and complex, limiting its availability. Both ¹⁸F-FDA and ¹⁸F-DOPA have been shown to be superior to MIBG imaging for PHEO/PGL (62,63), with reported sensitivity as high as 100%. There is superior detection of lesions in those with hereditary neuroendocrine tumors, the SDHB mutation, von Hippel–Lindau syndrome–related, and head and neck PGLs (64–66). Limited data exist for these tracers in neuroblastoma. In a small study, sensitivity and specificity of approximately 90% were reported in stage III/IV neuroblastoma patients (67), with visualization of more disease sites and better assessment of response than with ¹²³I-MIBG (68). ¹⁸F-DOPA lesion scores have also been evaluated, and a preliminary assessment showed a high correlation with the MIBG score (69). A higher prognostic correlation of metabolic burden detected with ¹⁸F-DOPA was noted in this study. In a comparison study in 19 neuroblastoma patients imaged at diagnosis (n = 4) or at relapse (n = 15), ¹⁸F-DOPA had higher sensitivity (90%) than MIBG (∼56%) (70). Larger prospective and clinically focused studies are needed to establish the role for using these tracers in clinical management.

¹²³I/¹³¹I-MIBG Theranostics for Neuroblastoma

Neuroblastoma has the ability to concentrate MIBG and is radiosensitive (71) and therefore amenable to therapy with radiolabeled ¹³¹I-MIBG. External-beam radiotherapy of widespread metastatic sites is not feasible, making systemic MIBG therapy for neuroblastoma a rational approach that has been used for several decades. MIBG therapy has usually been given to patients at relapse: both high-dose and multiple-dose strategies have been used (Fig. 3), and combined treatment with chemotherapy or biologic therapy has been studied (Table 1) (36,72–80).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

12-y-old with neuroblastoma. ¹²³I-MIBG pretherapy anterior (A) and posterior (B) images show multiple foci of uptake consistent with lesions in skull, right humerus, vertebrae, pelvic bones, and right tibia (arrows). These lesions are not seen in posttherapy scans (C and D) obtained 8 wk after treatment with 21.8 GBq (590 mCi) of ¹³¹I-MIBG (dose of 666 MBq [18 mCi]/kg).

Initial studies demonstrated feasibility of MIBG therapy in neuroblastoma and palliation of bone pain at submyeloablative doses (81), with responses reported in about 46% of patients (82). Early phase 1 studies in chemoresistant neuroblastoma demonstrated myeloablation as a dose-limiting toxicity uniformly observed at a dose of greater than or equal to 444 MBq (12 mCi)/kg (83). Severe myelosuppression, particularly a grade 4 thrombocytopenia, with the nadir occurring at about 4–6 wk, was observed at a 666 MBq dose (18 mCi)/kg (84). High-dose therapy, with autologous hematopoietic stem cells to reverse myeloablation, resulted in few other major toxicities, The maximum dose was primarily limited by radiation safety logistics (85). Predictors of increased marrow toxicity include whole-body or marrow-absorbed dose, extensive marrow/bone disease, prior chemotherapy treatment, and prior stem cell transplant (84,86). A phase 2 study of ¹³¹I-MIBG therapy given at a 444- or 666-MBq (12 or 18 mCi)/kg dose with and without stem cell support, respectively, showed a dose–response effect; a higher response was seen with larger dose and 18% event-free survival noted at 1 y (80).

Most ¹³¹I-MIBG trials have been conducted in patients with advanced neuroblastoma who have failed first-line therapy. Improvements in bone pain and objective responses have been reported at ¹³¹I-MIBG doses as low as 37–185 MBq (1–5 mCi)/kg (81). In patients with refractory neuroblastoma, a phase 1 trial evaluated escalating doses of ¹³¹I-MIBG. Doses greater than or equal to 444 MBq (12 mCi)/kg appeared to elicit better responses than lower doses (85). In patients with refractory or relapsed neuroblastoma who were heavily prior-treated with chemotherapy, overall reported response rates to a single-agent MIBG therapy even at doses as high as 666 MBq (18 mCi)/kg were only about 30%–40% and complete response rates were much lower. Factors influencing response include site of disease, with skeletal disease demonstrating better response than soft tissue and number of prior chemotherapeutic regimens (80). The effect of MIBG therapy on long-term survival in this group of patients remains unclear because most received further therapy after MIBG therapy (80,87). In addition, patients with relapsed disease appeared to have shorter progression-free survival after MIBG therapy than those with primary refractory neuroblastoma (88).

MIBG therapy as part of induction treatment has been studied, though in fewer patients. Reasons for its limited use in this patient group include the documented chemosensitivity of the disease, logistical issues related to isolation in sick children with high disease burden at diagnosis, and prolonged thrombocytopenia after MIBG therapy that could affect further therapy. Up-front MIBG therapy appears to be safe and tolerable, again with myelosuppression being the major toxicity (89–91). With up-front treatment combined with chemotherapy, higher VGPR/CR responses were seen in those treated with doses of greater than or equal to 444 MBq/kg (12 mCi/kg) (90). In a recent pilot study, only 66% of stage 4 neuroblastoma patients could receive planned MIBG therapy (173.9–562.4 MBq [4.7–15.2 mCi]/kg). The response rate (38%) was similar to that observed in patients with relapsed disease though prior studies on smaller groups of patients have reported higher response rates when combined with chemotherapy either before or after induction chemotherapy (91).

The causes of the relatively modest response to MIBG therapy are likely to be multifactorial (92) and have motivated consideration of strategies such as high-dose or multiple-dose MIBG therapy to enhance response in refractory patients. Limitations related to radiation safety precautions and tolerability restrict the dose of MIBG that can be administered as a single dose. Thus, fractionated or split-dose therapy may be a suitable alternative. An initial study reported on the safety and feasibility of 2 MIBG doses designed to administer a whole-body dose of 4 Gy. Higher whole-body exposures were obtained when double MIBG therapy was administered with autologous stem cell rescue after each infusion. Hematopoietic reconstitution was unaffected (36). Tandem MIBG therapy given as 2 doses, with the second dose given within 2–12 wk of the first, has also been administered safely with a single stem cell rescue after the second dose administered 42–100 d after the first. However, responses were less than 40% (36,93), similar to single-dose treatments.

In patients receiving 1–5 treatments of ¹³¹I-MIBG (2,516–5,513 MBq [68–149 mCi] each), a 30% objective response rate and 2% complete response were noted. The 5-y event-free survival was 40% for stage 4 and 92% for stage 3 (94). A more recent retrospective study of 111 patients with refractory neuroblastoma noted better 3-y event-free survival for patients treated with MIBG than for those not treated with MIBG, although the results were similar to those obtained with high-dose chemotherapy and autologous stem cell transplant (ASCT) (95).

A prospective study of those receiving doses of greater than 444 MBq (12 mCi)/kg showed an overall response rate of 37%, a 3% complete response, and a median survival of 6 mo (84). Multiple doses administered based on projected whole-body doses of 1, 2, or 2.5 Gy as estimated from diagnostic scans showed a 33% objective response rate (82). Compared with these low responses with larger and multiple doses, single-dose administrations based on body weight (96.2–666 MBq [2.6–18 mCi]/kg), whole-body dose (1–2.5 Gy), or empiric dose of 3,700–7,400 MBq (100–200 mCi) resulted in response rates of 25%–46% (79,82,84,96). An early phase 2 study from multiple centers that treated patients with 1,110–3,700 MBq (30–100 mCi), as a single or multiple doses, reported palliation of pain in refractory patients though no significant objective response was seen (97). Weight-based regimens have included low-activity (37–148 MBq [1–4 mCi]/kg) to high-activity (296–666 MBq [8–18 mCi]/kg) multiple dosing (Table 1). Overall, the data from many studies are heterogeneous, largely because of the retrospective nature of studies that have highly variable schema of treatments (single vs. multiple doses, low fixed dose vs. intermediate or high dose, or body-weight or whole-body dose–based treatments). Additionally, there is high variability in defining the clinical states, extent of disease, and reporting or results. The response rates and outcomes reported are highly variable across the literature (overall response, overall survival, and event-free survival). For these reasons, it is difficult to derive firm conclusions on how to best apply these therapies in routine clinical practice. Nevertheless, MIBG in refractory disease, especially with high-dose therapies and stem cell support, remains an actively pursued strategy especially for patients with chemoresistant neuroblastoma.

Most common toxicity with MIBG therapy is hematologic, which generally results in dose-dependent thrombocytopenia and is postulated to occur due to binding of MIBG to platelets (99). Other commonly observed toxicities are transient thyroid dysfunction, sialoadenitis (100), hepatic transaminitis (101), and mild acute hypertension (102). The incidence of secondary malignancy is similar to that observed in children treated with chemotherapy with a cumulative incidence of 7.6% and 14.3% at 5 and 10 y after therapy, respectively (103). Secondary malignancies including acute nonlymphoblastic leukemia, chronic myelomonocytic leukemia, angiomatoid malignant fibrous histiocytoma, malignant schwannoma, and rhabdomyosarcoma have been reported to occur at a median of 3 y (1–15 y) (104).

The role of radiosensitizers, if any, in enhancing MIBG therapy is likely to be modest but needs to be defined because radiosensitizers have additional toxicities. Such a study is currently being performed under the aegis of the Newer Advances in Neuroblastoma Therapy consortium (ClinicalTrials.gov identifier: NCT02035137). The availability of autologous hematopoietic stem cells for bone marrow rescue after high-dose MIBG therapy is limited because most patients undergo autologous bone marrow transplantation as part of the up-front therapy for high-risk neuroblastoma. The optimal use of stem cells after tandem MIBG therapy is being evaluated (ClinicalTrials.gov identifier: NCT00107289). Finally, the optimal integration of MIBG therapy into up-front treatment of high-risk neuroblastoma is unclear, but likely to be further evaluated in forthcoming studies.

¹²³I/¹³¹I-MIBG Theranostics for PHEO and PGL

The optimal therapy for PHEO/PGL consists of surgical resection when feasible and is generally curative. In contrast, there is no standard curative therapy for metastatic disease. Most PHEO/PGLs, particularly those in cluster 2 related to the tyrosine hydroxylase signaling pathway, have a high expression of the norepinephrine transporter and concentrate ¹³¹I-MIBG (105), allowing for targeted therapy of disease.

¹³¹I-MIBG is generally given to those with progressive or symptomatic disease (Fig. 4). Systematic prospective studies with dose escalation and survival analysis are lacking, and many reported studies are in a small number of patients (Table 2). The published literature is also heterogeneous in patient selection and the reported follow-up data, which may explain differences in observed response rates and survival. The reported overall complete response rates are low, ranging from 0% to 18% (106–110).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Patient with metastatic pheochromocytoma presented with progression of disease. ¹²³I-MIBG pretherapy anterior (A) and posterior (B) images show increased uptake in bilateral lung lesions and intense uptake in large sternal lesion, left adrenal lesion, and sacral lesion (arrows). Patient was eligible for ¹³¹I-MIBG and was treated with approximately 8,510 MBq (230 mCi) of ¹³¹I-MIBG. (C–G) Posttherapy planar anterior (C) and posterior (D) scans and SPECT/CT fused images at 3 levels (E–G) show good targeting of lesions (arrows). Plasma normetanephrine and total metanephrine levels decreased from greater than 20,000 and 40,000 pg/mL, respectively, at baseline to 14,295 and 14,329 pg/mL, respectively, at 13 wk after treatment.

Individual doses of ¹³¹I-MIBG used for therapy in published studies are highly variable, ranging from low (2,960 MBq [80 mCi] to less than or equal to 7,400 MBq [200 mCi]), intermediate (up to 18,500 MBq [500 mCi]), to high doses of ¹³¹I-MIBG that are often weight-based (444–666 MBq [12–18 mCi]/kg) (107–109,111). Most common treatment schemas used are multiple, low doses of MIBG therapy typically administered every 3–6 mo, primarily with a palliative rather than curative intent. Formal phase 1 or optimal dose-finding studies are lacking in PHEO/PGL patients and most reports are from retrospective studies. In general, doses below 11.1 GBq (300 mCi) are well tolerated without severe myelosuppression. Acute adverse events are transient and consist of mild to moderate nausea, vomiting, and hypertension (112). The most common subacute toxicity is myelosuppression, with thrombocytopenia being more common than neutropenia. Myelosuppression is much less common as patients may have received little to no prior chemotherapy. ¹³¹I-MIBG doses with a weight-based dose of greater than 2 Gy caused a higher number of grade 3 or 4 hematologic toxicity (107). Biochemical hypothyroidism has been reported in 11%–20% despite thyroid blockade with potassium iodide but symptomatic hypothyroidism is rare (113). Those receiving high single doses of MIBG therapy have a higher incidence of toxicities. In a phase 2 study in 50 patients with PHEO/PGL receiving ¹³¹I-MIBG doses ranging from 18.2 to 42.9 GBq (492–1,160 mCi), greater than or equal to grade 2 neutropenia and thrombocytopenia were observed in 87% and 83% of patients, respectively (108). Rare pulmonary toxicities, such as adult respiratory distress syndrome, can occur, and an increased risk is noted in those who have proteinuria.

Overall, MIBG therapy has demonstrated clinical responses in patients with PHEO/PGL. However, although objective responses are common, complete response rates are exceedingly low (<5% in a large retrospective review) (106). Furthermore, there is a wide variation in response rates in reported studies. These differences are probably related to patient selection, biologic factors such as NET expression, dose and frequency of MIBG administrations, and length of follow-up. Reported objective response rates are typically less than 50%. A metaanalysis of 17 studies involving 243 patients reporting a response rate of 27%. Rates of symptom control and hormonal responses are typically higher than objective responses (114). There does not appear to be a clear dose-response effect of MIBG therapy. The quality of responses has been associated with the subtype of disease (better responses in PGL than in PHEO) (114) and sites of disease (better responses in soft tissues vs. bone). An objective response to the initial dose of MIBG therapy appeared to be associated with a favorable prognosis in patients receiving multiple doses (115).

The impact of MIBG therapy on overall survival is hard to assess because PHEO/PGL even when malignant is an indolent disease, and patients often survive for more than 10 y without any anticancer therapy. In a retrospective study of 33 patients, responders and patients receiving doses of greater than 18.5 GBq (500 mCi) appeared to have better survival rates than nonresponders and those receiving doses of less than 18.5 GBq (500 mCi) (111).

At present, the general approach to administering MIBG therapy is giving multiple fractionated low or intermediate doses without the need for autologous stem cell rescue. Multiple doses at intervals of 8–12 wk may be administered with the intention of stabilization of disease or symptoms. MIBG has an established role in symptom relief and palliation; improvement in disease-associated pain and symptoms related to catecholamine excess has been reported in 50%–85% of patients (111,116). However, robust and durable objective and clinical responses are rare, with few complete responses. ¹³¹I-MIBG therapy is therefore not curative. Although symptomatic relief and partial responses may be achieved, the survival benefit and efficacy have not been conclusively proven. There is no universal consensus on ¹³¹I-MIBG treatment schemas or doses for maximizing benefit. ¹³¹I-MIBG therapy has no significant role in certain specific biologic subtypes of malignant PHEO/PGL such as those with the SDHB mutation due to lack of significant uptake and false-negative results (110). The role of the combination of MIBG with other targeted therapies has not been studied and needs to be evaluated.

In carcinoids, good symptomatic relief and control of disease at cumulative doses ranging from 1,850 MBq (50 mCi) to more than 74 GBq (2 Ci) as well as in single or multiple doses of less than 7,400 MBq (200 mCi) have been noted with ¹³¹I-MIBG therapy (117). The role of high-dose therapy has not been evaluated. Objective response rates are reported in up to 30% of patients, symptomatic responses are reported in 38%–92%, and biochemical responses are reported in about 13%–73% patients; complete responses are rare (118–120). Again, due to the high variation in the performance of studies and reported outcomes, the assessment of survival benefit is limited. Median 5-y survival is reported as 42%–78% (119,120). Longer survival has been noted in those treated with larger single treatment with greater than 14.8 GBq (400 mCi) (4.69 y vs. 1.86 y) and in those with symptomatic response and biochemical improvement as against those without. At present, because of the higher avidity of somatostatin receptor–directed agents, MIBG therapy is generally not used.

High-Specific-Activity ^123/131I-MIBG

The process of uptake of MIBG is saturable, hence conventional MIBG preparation containing high proportions of noniodinated MIBG molecules could saturate binding sites and reduce the antitumor effect of MIBG therapy. High-specific-activity non–carrier-added ¹²³I/¹³¹I-MIBG concentrates in higher amounts in tumor cells (121) and thus has the potential to be more effective as a targeting agent. Specific activities of greater than 44.4 GBq (1,200 mCi)/μmol for ¹²³I-MIBG and more than 59.2 GBq (1,600 mCi)/ μmol for ¹³¹I-MIBG can be obtained using an efficient method of radiolabeling (Ultratrace; Molecular Insight Pharmaceuticals, Inc.) (122). A phase 1 study of non–carrier-added ¹²³I-MIBG showed favorable dosimetry and safety. A higher contrast was noted for uptake in the heart and slightly larger activity noted in the gastrointestinal tract compared with conventional ¹²³I-MIBG (123). At present, limited data exist regarding efficacy, and distinct clinical advantage of high-specific-activity MIBG has not been proven, though a lower incidence of side effects, such as nausea, vomiting, and hypertension, has been reported for both neuroblastoma and PHEO/PGL patients for imaging and therapy (109,124). In a phase 1 clinical trial with non–carrier-added MIBG in patients with high-risk neuroblastoma at doses of 444, 555, or 666 MBq (12, 15, or 18 mCi, respectively)/kg, no dose-limiting toxicities occurred; however, response rates were similar to those observed with conventional MIBG therapy (79). More data are needed to establish its advantage over regular MIBG for therapeutic use. A phase 2 trial of Ultratrace ¹³¹I-Iobenguane (Molecular Insight Pharmaceuticals, Inc.) in PHEO/PGL patients is ongoing (NCT00874614).

Use of α-Emitter for NET Targeting

To increase the cytocidal effect, an α-emitter with high linear energy transfer within a short range of 50–80 μm would potentially be advantageous in smaller tumors. It can also cause bystander effect in regions of specific uptake with limited radiation to adjacent normal tissues, due to its short path length, resulting in lower side effects. ²¹¹At is a long lived α-emitter with a half-life of 7.2 h that has been used to label MIBG analogs: ²¹¹At-metaastatobenzylguanidine (MABG) and ²¹¹At-astatofluorobenzylguanidine have been developed (138). A kit method for labeling has also been developed that allows for more convenient and easy utilization (139). MABG showed high specific binding in vitro (140) and high uptake in tumor cells in mice (141). In cell cultures, MABG showed dose-dependent toxicity at low level radiation but elimination of cytotoxicity at high levels (142). Given the physical characteristics, it is expected to be effective treatment, however, clinical studies with MABG as a therapeutic are lacking. Feasibility of use in patients, dosage, and effects are yet to be evaluated and established.

Combination Therapy with NET-Targeting Agents with Radiosensitizers or Chemotherapeutic Agents

A further approach to enhance NET-directed MIBG therapy involves its combination with radiosensitizers. Preclinical mechanistic justifications for such a combination include enhancement of NET expression or activity, for example, by vorinostat (143); synergistic effect on inhibition of radiation-induced DNA repair, for example, by topoisomerase inhibitors (144); and synergistic elicitation of autophagy and apoptosis, for example, by arsenic trioxide (145). Cisplatin pretreatment may also enhance expression of NET and MIBG uptake (146). Coadministration with poly(adenosine diphosphate) ribose polymerase induced increased DNA damage in xenografts (147). Histone deacetylase inhibitors act by derepressing gene-encoded proteins, such as caspaze-8, essential for apoptosis. Vorinostat, a histone deacetylase inhibitor, has been noted to increase NET expression and uptake of ¹²³I-MIBG in xenografts (143). These observations led to the clinical testing of MIBG therapy in combination with radiosensitizers.

Early-phase studies have demonstrated that high-dose MIBG therapy administered in conjunction with the topoisomerase inhibitors topotecan and irinotecan (126), the histone deacetylase inhibitor vorinostat (74), and arsenic trioxide (72) is safe and tolerable in neuroblastoma with a modest increase in additional toxicities. However, in the heterogeneous groups of patients studied, there did not appear to be a marked improvement in response rates. ¹³¹I-MIBG plus arsenic trioxide in neuroblastoma was well tolerated, with an adverse event profile similar to that of ¹³¹I-MIBG therapy alone. The addition of arsenic trioxide to ¹³¹I-MIBG did not significantly improve response rates compared with historical data therapy only (72). For high-risk neuroblastoma, a New Approaches to Neuroblastoma Therapy study showed feasibility of treatment with ¹³¹I-MIBG in combination with CEM regimen (carboplatin, etoposide, and melphalan) and ASCT in 50 patients. A 10% response was noted (complete response and partial response) in those with primary refractory or progressive disease (148). A comparative study of MIBG therapy alone, MIBG combined with vincristine and irinotecan, and MIBG with only vorinostat is ongoing (NCT02035137); MIBG with vorinostat in resistant or relapsed neuroblastoma has been evaluated (NCT010119850). Further investigations are under way to determine whether radiation sensitizers can indeed enhance the therapeutic effect of radioiodinated MIBG; this would be most applicable to refractory and advanced neuroblastoma.

Experience with combined chemotherapy and ¹³¹I-MIBG therapy is limited in PHEO but feasible. In a small study, whole-body dosimetry–based treatment to deliver 85–90 cGy per administration was given with cyclophosphamide–vincristine–dacarbazine chemotherapy (149). Hematologic toxicity limits dosing and requires reduction of chemotherapy in some patients (150). In hereditary PGL/PHEO syndrome with SDHB mutation patients, combination of ¹³¹I-MIBG and sunitinib resulted in a significant response (151).

Combination Therapy with NET-Targeting Agents and Other Radiopharmaceuticals

Use of ¹³¹I-MIBG with somatostatin-targeting agents in nonneuroblastoma neuroendocrine tumors is being evaluated (NCT01099228). Patient groups are receiving either ¹³¹I-MIBG and ⁹⁰Y-DOTATOC or ¹³¹I-MIBG and ¹⁷⁷Lu-DOTATATE. The different targets for the 2 radiopharmaceuticals and organ distribution may allow for more tumor dosing while limiting organ toxicity. Theoretic modeling showed increased tumor doses and tumor–to–critical-organ dose ratio; maximum combined therapy tumor dose increased 68% for a tumor dose ratio of 2.57, using 92% of the maximum tolerated ⁹⁰Y-DOTATOC activity supplemented with 76% of the maximum tolerated activity of ¹³¹I-MIBG. Variations in organ radiation-absorbed dose/MBq and dose-limiting values altered both the magnitude of the increase and the range at which combined-agent therapy was advantageous (152).

Localized Therapy

Injection of the radiopharmaceutical to deliver the agent directly or selectively to the lesions from a localized delivery technique enables administration of higher radiation dose to the site of tumor while minimizing systemic side effects. Intraarterial injection of ¹³¹I-MIBG has been used to treat liver metastases or primary tumors from neuroendocrine tumors (153). In a small study (n = 17), ¹³¹I-MIBG injected via the hepatic artery or superior mesenteric artery/ileocolic artery/lumbar branch delivered a 69% higher dose to tumor (154). No systematic studies have been conducted to further evaluate the approach for effectiveness or efficacy. The invasive nature of this approach and availability of other localized techniques of therapy for liver metastasis have probably limited wider use. Similar localized approaches with MABG or combined therapeutic agents have not been evaluated.

Novel Techniques for Targeting NET: Reporter Systems

NET may serve as a reporter gene for noninvasive imaging of genetically modified cells using radiolabeled targeting agents. Availability of existing radiolabeled agents for imaging the specific expression of NET makes this an attractive probe. Retroviral transfection of the human NET gene can enhance uptake of MIBG (155). ¹³¹I-MIBG imaging of transduced cells showed 66- to 120-times-higher uptake, compared with parental cells, that was highly specific and NET driven (156). ¹¹C-m-hydroxyephedrine as a PET imaging probe for imaging gene therapy has been explored; the short life of the isotope is, however, limiting (157). Subsequent modifications to this approach have included linking to fluorescent dye by constructing an internal ribosomal entry site linked hybrid NET and green fluorescent protein reporter gene that allowed for imaging with ¹²⁴I-MIBG PET as well as fluorescence optical imaging (158). Transfected viruses with NET reporter genes may allow for imaging and targeting of a variety of tumors and are currently being explored in vitro and in xenografts (159). Highly sensitive imaging of T cells has been performed in xenografts using a PET tracer ¹⁸F-MFBG–human NET reporter probe (160).