Radiotheranostics - Precision Medicine in Nuclear Medicine and Molecular Imaging

'See what you treat and treat what you see, at a molecular level', could be the motto of theranostics. The concept implies diagnosis (imaging) and treatment of cells (usually cancer) using the same molecule, thus guaranteeing a targeted cytotoxic approach of the imaged tumor cells while sparing healthy tissues. As the brilliant late Sam Gambhir would say, the imaging agent acts like a 'molecular spy' and reveals where the tumoral cells are located and the extent of disease burden (diagnosis). For treatment, the same 'molecular spy' docks to the same tumor cells, this time delivering cytotoxic doses of radiation (treatment). This duality represents the concept of a 'theranostic pair', which follows the scope and fundamental principles of targeted precision and personalized medicine. Although the term theranostic was noted in medical literature in the early 2000s, the principle is not at all new to nuclear medicine. The first example of theranostic dates back to 1941 when Dr. Saul Hertz first applied radioiodine for radionuclide treatment of thyroid cells in patients with hyperthyroidism. Ever since, theranostics has been an integral element of nuclear medicine and molecular imaging. The more we understand tumor biology and molecular pathology of carcinogenesis, including specific mutations and receptor expression profiles, the more specific these 'molecular spies' can be developed for diagnostic molecular imaging and subsequent radionuclide targeted therapy (radiotheranostics). The appropriate selection of the diagnostic and therapeutic radionuclide for the 'theranostic pair' is critical and takes into account not only the type of cytotoxic radiation emission, but also the linear energy transfer (LET), and the physical half-lives. Advances in radiochemistry and radiopharmacy with new radiolabeling techniques and chelators are revolutionizing the field. The landscape of cytotoxic systemic radionuclide treatments has dramatically expanded through the past decades thanks to all these advancements. This article discusses present and promising future theranostic applications for various types of diseases such as thyroid disorders, neuroendocrine tumors (NET), pediatric malignancies, and prostate cancer (PC), and provides an outlook for future perspectives.


Introduction
The term theranostic became increasingly popular since the early 2000s and its publications have been rapidly increasing ever since. Theranostic is a portmanteau of the Greek words therapo and gnosis, translating to therapy and knowing, i.e., diagnostics [1,2]. There is a constant debate whether 'theranostics' or 'theragnostics' should be used, however both spellings are acknowledged. The principle of theranostic is to identify the right molecular probe (diagnostic/therapeutic) for the right patient in order to maximize subsequent treatment outcome while minimizing toxicity. It stratifies upfront future responders from non-responders, hence preventing unnecessary treatments, sparing patients from the usual trial and error approach, and saving unnecessary drug costs for the healthcare system. The theranostic concept applies to many different areas of science depending on the platform used to convey the main principle. In that sense, theranostics could be further categorized as: radiotheranostic, nanotheranostic, magnetotheranostic and optotheranostic, by using radionuclides, nanoparticles, magnetic particles Ivyspring International Publisher and optical probes, respectively [3,4]. Out of all the possible categories, radiotheranostic has been so far the only one widely adopted and integrated clinically, and therefore commonly referred to as theranostics. Radiotheranostics uses low penetration radiation emitted from radionuclides to deposit high levels of energy in the nucleus of the targeted cells to induce DNA strand breaks and activate programmed cell death. The concept of radiotheranostics has been clinically adopted for over 80 years now. Among the earliest examples of theranostic is the use of Phosphorus-32 ( 32 P) for leukemia in 1941 [5], and Iodine-131 ( 131 I) for Graves' disease published by Saul Hertz in 1942 [6], and subsequently thyroid cancer in 1946 [7]. Radioiodine therapies became one of the cornerstone treatments in nuclear medicine and is markedly used every day worldwide. Radiotheranostic complies with the ultimate concept of personalized medicine by using paired diagnostic/therapeutic radionuclide probes for the selective and targeted diagnosis and treatment of specific (usually cancer) cells, tailored to the patient's specific underlying disease. These theranostic probes allow for molecular characterization of cells and cancer cells in vivo, i.e., within the whole body, early detection of disease, disease staging, assessment of tumoral molecular heterogeneity by imaging, therapy selection, treatment planning, and subsequent targeted and tailored treatment based on the diagnostic molecular imaging results. The diagnostic part further allows for early assessment of treatment response and detection of recurrence, long before there are measurable anatomical changes, and plays a critical role in the differentiation between progression and pseudoprogression. The use of dosimetry to predict and determine the specific radiation absorbed doses by the tumor, organs at risk, and healthy tissues of every individual patient ensures the desired goals of efficacy and safety of the treatment. Advances in dosimetric software like voxel-based dosimetry will make dosimetry-guided targeted radionuclide therapies (TRT) more feasible. In pursuing a more personalized medicine, radiotheranostic is slowly moving from a fixed schedule of treatment cycles and dose activity to a more individualized approach based on the patient's specific tumor burden, biology, and dosimetry, which will allow immediate adjustment for toxicity and individualized decision making.

Radiotheranostic Pairs
A radiopharmaceutical consists of three components: the radionuclide (with diagnostic and/or therapeutic properties), a chelator (which links the radionuclide to the ligand/probe), and the ligand/probe (which targets a cancer-specific molecular marker on the tumor cell with high affinity). Sometimes, a radionuclide by itself can serve as a radiopharmaceutical without the need for a chelator or radiolabeling. The purest concept of a 'theranostic pair' consists of a chemically and structurally identical (or nearly identical) probe labeled with either a diagnostic or therapeutic radionuclide. This ensures targeting of the same molecular marker for diagnostic imaging and molecular targeted treatment. Radiolabeling is a critical step in the synthesis of a radiopharmaceutical since the receptor binding affinities may be negatively affected by a decreasing degree of similarity between the diagnostic and therapeutic molecule. In that sense, the 'perfect' theranostic pair would be two isotopes of the same element. The prime example is radioiodine, where for instance 123 I (single photon emitter) or 124 I (positron emitter) can be used for diagnostics and 131 I (beta emitter) for treatment of thyroid diseases. These isotopes are chemically 'identical', and only differ in their emissions and physical half-lives, which is favorable for their respective purposes.
The diagnostic counterpart can be performed by employing either single photon emission computed tomography/computed tomography (SPECT/CT) or more commonly used positron emission tomography either with computed tomography (PET/CT) or magnetic resonance imaging (PET/MRI) to obtain molecular diagnostic images. The chosen radiopharmaceutical is either a gamma emitter for SPECT or a positron emitter for PET. Both gamma and positron emitting radiopharmaceuticals have high tissue absorption, a low energy transfer and a long radiation range, resulting in low-level radiation exposure for the patient with optimal imaging condition. In contrast to anatomical imaging like CT or MRI, molecular imaging visualizes tumor molecules and characterizes tumor tissue, function, and biology. This allows not only for disease localization, staging and restaging, but also, and a unique feature of theranostic, the ability to effectively select patients for subsequent TRT based on their chances to have a positive response to therapy. Molecular imaging determines whether there is sufficient expression of the molecular target based on tumor uptake compared to normal tissue and background uptake, and therefore whether the patient will benefit from TRT. This principle indicates that treatment with the same compound will be delivering a tumoricidal radiation dose to the cancer cells.
TRT is a systemic cytotoxic treatment which is applied either intravenously or orally. The ionizing radiation aims directly at the cancer-specific target and induces deoxyribonucleic acid (DNA) doublestrand breaks and subsequently organized cell death through apoptosis. Therefore, choosing the most appropriate radionuclide is key. The higher the linear energy transfer (LET) to the target cell, the higher the damage to the target cell and treatment efficacy. Also, the longer the emission range, which is the tissue penetration range, the larger the perimeter of the irradiated tissue area/treated area (measured in microns up to 2 mm). Preferably, a radionuclide with a relatively long half-life (days to 1-2 weeks) is chosen to prolong the therapeutic effect. The most commonly used radioemitters in the clinic are beta particles like 131 I, Lutetium-177 ( 177 Lu), Samarium-153 ( 153 Sm), and Yttrium-90 ( 90 Y). They are characterized by a high energy transfer to the tumor cell and a short radiation emission range, which is favorable to spare surrounding healthy tissue cells. Treatment with alpha particles like Radium-223 ( 223 Ra) has been approved by the US Food and Drug Administration (FDA), others like Actinium-225 ( 225 Ac) are being actively researched in human clinical trials. Compared to beta emitters, they are distinguished by a very high LET and an even shorter path length in the dimension of microns (<100 µm). Another group of emitters that have been used in the past for theranostics are Auger electron emitters. However, these radionuclides are generally less effective as they provide very low energy electrons that decay by electron capture. The energy is deposited over a very short distance, so they become most effective strictly intracellularly. Examples of Auger electron emitters are 123 I, Indium-111 ( 111 In), Gallium-67 ( 67 Ga), and Technetium-99m ( 99m Tc), which are currently used for SPECT/CT at very low diagnostic doses, but some of them like 123 I and 111 In have been introduced in clinical trials at high doses for treatment of thyroid diseases and neuroendocrine tumors (NET), respectively [8].
In addition, radionuclides usually have two or more types of emission with different energy peaks. This characteristic makes certain radioisotopes used for therapy to be suitable for non-diagnostic imaging. This non-diagnostic imaging can be of great utility to obtain post-treatment SPECT/CT imaging to confirm molecular targeting of the treatment, rule out pharmacologic interference and stunning. This is usually the case with beta emitters, which tend to have a certain abundance of gamma emission suitable for post-treatment imaging with SPECT/CT. These scans can also be used for dosimetry to determine the absorbed doses of the tumor(s) and healthy tissues.

Thyroid Diseases -Radioiodine and Beyond
Thyroid cancer is the most prevalent endocrine malignancy, and the incidence of papillary thyroid carcinoma (PTC) has increased over the past few decades due to improved diagnosis [9]. The American Cancer Society estimates 44,280 new cases of thyroid cancer in 2021 in the United States [10]. Differentiated thyroid cancer (DTC) accounts for the vast majority of thyroid cancers. They arise from follicular epithelial cells and are divided in PTC (85%), follicular thyroid cancer including conventional (5%) and oncocytic (Hürthle cell) carcinomas (5%), poorly differentiated (<3%) and anaplastic thyroid cancer (<3%) [11].
Radioiodine treatment utilizes the underlying thyroid physiology of using iodide to synthesize thyroid hormones. Iodide is taken up from the blood stream by follicular thyroid cells through the sodium-iodide symporter (NIS) localized in the basolateral membrane. The NIS cotransports two sodium ions along with one iodide ion into the cytosol whereas the sodium gradient serves as driving force [12]. The efflux of iodide across the apical membrane into the follicular lumen is mediated by pendrin channels [13]. Thyroid peroxidase organifies iodide by oxidation and attachment to thyroglobulin. Iodinated thyroglobulin re-enters the follicular cell via endocytosis, undergoes hydrolysis, and the thyroid hormones T3 and T4 are subsequently secreted into the blood stream at the basolateral membrane [14]. Radioiodine is taken up and trapped in the thyroid cell in the same manner as any other iodine molecule. The thyroid cell cannot differentiate between the structurally identical radioactive and non-radioactive molecule. Several members of the radioiodine family are used for diagnostic or therapeutic purposes for hyperthyroidism and differentiated thyroid carcinoma (Table 1).
From a diagnostic point of view, SPECT imaging with 123 I can be used for diagnosis and planning of radioiodine therapy in cases of DTC, hyperthyroidism like Graves' disease and toxic uni-or multinodular adenomas. PET imaging with 124 I is also a great diagnostic agent, although not as easily accessible as 123 I. 124 I is used for staging and restaging of DTC, although its long half-life of 4 days makes it an excellent theranostic pair for dosimetry. Currently, the doses of radioiodine prescribed for radioiodine treatment of DTC are based on disease risk, disease burden and location of metastatic disease. This can lead to under-or overtreatment with subsequent adverse effects. Lesional dosimetry aids in moving towards an individualized treatment approach in order to maximize outcome and minimize toxicity. The feasibility of dosimetry with 124 I in predicting and estimating absorbed doses of individual tumor lesions is currently evaluated in various institutions in the United States and Germany (ClinicalTrials.gov identifiers NCT03647358, NCT03841617 and NCT01704586). From a therapeutic point of view, treatment with 131 I is given in various settings: In benign hyperthyroidism to ablate uni-or multinodular toxic goiter or whole organ ablation in Graves' disease; in DTC, as adjuvant therapy to irradiate local or distant metastases and thereby reducing the risk of recurrence, or for residual or recurrent disease, or to ablate any normal thyroid remnant after total thyroidectomy to ensure an undetectable thyroglobulin level which subsequently can be used as tumor marker in follow-up [15,16]. The American Thyroid Association (ATA) guidelines from 2015 recommended radioiodine treatment for high-risk patients, selected cases with intermediate-risk, but not in patients with low-risk disease [15]. However, two randomized clinical trials from Europe validated the treatment of low-risk patients with a low activity of 1.1 GBq 131 I [17,18] and showed that the rate of relapse was not higher in patients who received 1.1 GBq vs 3.7 GBq 131 I [18]. Further large, randomized studies are underway to evaluate the usefulness of radioiodine treatment in low-risk DTC patients (ClinicalTrials.gov identifiers NCT01837745, NCT01398085). In a joint statement of the ATA, the European Association of Nuclear Medicine, the Society of Nuclear Medicine and Molecular Imaging, and the European Thyroid Association, they acknowledged the lack of substantial data pro or contra radioiodine treatment in low-risk DTC patients and concluded that the decision to treat should be made on an individual basis depending on disease risk factors, patient-related factors (concern, co-morbidities) and healthcare setting [16].
Theranostic with radioiodine is an integral part of (re-)staging, therapy, and surveillance of DTC after initial thyroidectomy. Although the prognosis of DTC is usually favorable, recurrence is seen in 5-30% of the cases [19]. Treatment of recurrent or metastatic disease is problematic, especially when these tumors have become refractory to radioiodine [20]. MEK inhibitors such as selumetinib have shown ability to induce redifferentiation of radioiodine refractory cancer cells and hence increased radioiodine uptake in a small cohort [21]. The SEL-I-METRY trial, a multicenter phase II study (European Union Clinical Trials Register identifier EudraCT 2015-002269-47), is underway to shed more light on the redifferentiation potential of selumetinib.
Medullary thyroid cancer (MTC) derives from C-cells which arise from the neural crest and is therefore considered NET. Primary and curative treatment is surgery. In case of progressive, symptomatic disease, systemic therapy with cabozantinib and vandetanib is recommended [22,23]. However, these multikinase inhibitors bear serious adverse reactions and have not shown a significant survival benefit. As MTC are NET, peptide receptor radionuclide therapy (PRRT) with 90 Y-DOTA TOC, which targets somatostatin receptors (SSR), has been evaluated. A phase II trial evaluating response, survival, and long-term safety of 90 Y-DOTATOC in patients with metastatic MTC and increasing tumor marker calcitonin showed that 29% of patients demonstrated a decrease in tumor marker which was associated with a significantly higher survival [24]. Other studies exploring 177 Lu-DOTATATE [25] and 177 Lu-octreotate [26] concluded that PRRT can be considered as alternative. However, PRRT has not yet become clinical routine. Lodewijk et al showed expression of prostate-specific membrane antigen (PSMA) in 90% of the 104 included MTC patients [27]. This might warrant future TRT with 177 Lu-PSMA which is currently used to treat prostate cancer (PC) in clinical trials. Interestingly, a recently published case report showed uptake of 18 F-FDG, 68 Ga-DOTATATE and 68 Ga-PSMA in a patient with radioiodinerefractory, poorly differentiated thyroid cancer [28]. Rottenburger et al reported the first in man administration of the cholecystokinin-2 receptor agonist 177 Lu-PP-F11N in patients with MTC. 177 Lu-PP-F11N showed promising biodistribution with accumulation in MTC lesions [29]. Clearly there is an unmet need for treatment of metastatic MTC. Further randomized clinical trials are needed to compare different TRT with current state-of-the-art multikinase inhibitors in patients with MTC.

Neuroendocrine Tumors -PRRT
NETs are a heterogenous group of rare tumors. The majority are gastroenteropancreatic NETs (GEP-NET) (60-70%) and bronchial NETs (30%). Although considered rare, their incidence is rising, which might be due to more awareness and incidental findings, but also related to rapidly developing diagnostic methods [30]. The classification and nomenclature of NET is complex and not uniform as studies have mostly been focusing on organ-specific NETs. The World Health Organization proposed a 'common framework' to classify NETs: Depending on the site of origin, neuroendocrine neoplasia are divided into NETs and neuroendocrine carcinomas (NEC). NETs are further characterized by grading (low, intermediate, and high grade) based on mitotic rate and ki-67 index, which is a marker for cell proliferation [31]. The distinctive feature of well-differentiated NETs is the overexpression of SSR on their cell surface [32,33]. Somatostatin is in control of hormone secretion such as glucagon, insulin and growth hormone, and cell proliferation [34]. Five subtypes of human SSRs have been identified: 1, 2A, 2B, 3, 4 and 5. The majority of NETs show overexpression of subtype 2, especially 90% of GEP-NETs [35][36][37]. Poorly differentiated NETs and NECs have less or no expression of SSR [37]. Somatostatin analogs (SSA) were developed to target SSR and were initially administered to relieve symptomatic burden of functioning NETs. As they also showed an antiproliferative effect, their use was expanded to include non-hormone-secreting NETs [38,39]. SSA can be radiolabeled to image SSR expression for theranostic purposes. The diagnostic counterpart for SPECT imaging is 111 In-octreotide or 99m Tc-Hynic-toc. However, SPECT imaging has been largely replaced by PET due to better image quality and its potential for quantification [40]. Currently, there are three well established SSA available that are conjugated to DOTA and labeled with 68 Ga: 68 Ga-DOTA-1-Nal3octreotide (DOTANOC), 68 Ga-DOTA-D-Phe1-Tyr3octreotide (DOTATOC) and 68 Ga-DOTA-D-Phe1-Tyr3-Thr8-octreotide (DOTATATE). DOTATATE has a 10-fold higher and selected affinity to SSR2 receptor followed by DOTATOC which also binds to SSR5, whereas DOTANOC is selective towards SSR2, SSR3 and SSR5 [41]. Despite their different affinity to the various SSR, they all show high affinity to SSR2 and are equally effective in diagnostic accuracy with a pooled sensitivity of 93-96% and specificity of 85-100% [42][43][44][45][46][47][48][49][50] making PET imaging with 68 Ga-labeled SSA a powerful tool. In a systematic review and meta-analysis on the impact of 68 Ga-SSA PET on patient management including 1,561 patients, a management change due to PET/CT findings was seen in 44%. Most interestingly, 77% were intermodality changes, meaning a change in type of treatment (e.g., surgery to chemotherapy). Recently, the FDA approved Copper-64 ( 64 Cu)-DOTATATE for PET imaging of SSR positive NETs. 64 Cu has the advantage of a longer half-life (12.7 hours vs 68 minutes of 68 Ga), which makes a central production with long haul distribution possible. In a phase III clinical trial evaluating 64 Cu-DOTATATE PET/CT imaging for NET, a 100% sensitivity and 96.8% specificity was found with no adverse events [51]. 68 Ga-DOTATOC has also been FDA approved based on two clinical trials [52,53], however, it has no commercial partner yet.
The therapeutic counterpart uses the same SSA radiolabeled with beta emitters like 177 Lu and 90 Y or more recently with alpha emitters such as 225 Ac. Treatment with PRRT has been employed for a quarter of a century now. A magnitude of clinical trials and studies evidence the efficacy of PRRT in terms of decrease in tumor size and tumor marker, symptomatic relief, overall improvement of quality of life, and increase in overall survival (OS). In general, PRRT is known to have a safe profile and possesses little side effects. However, different radioisotopes have different safety profiles. Renal failure due to radiation induced inflammation and fibrosis of the kidneys has been reported, especially when using 90 Y, but less since the introduction of nephroprotection with amino acids. Hematological toxicity is usually mild and in the majority of cases reversible [54][55][56][57][58]. The NETTER-1 clinical trial was the first prospective randomized phase III trial in midgut NETs comparing treatment with high dose SSA vs a combination of normal dose SSA and PRRT with 177 Lu-DOTATATE. At the data-cutoff date, the median progression free survival (PFS) had not been reached in the PRRT group vs 8.4 months in the SSA group, meaning a 79% lower risk of disease progression or death in the PRRT arm. Partial response of 17% was seen in the 177 Lu-DOTATATE arm vs 3% in the SSA group. The patients reported a significant improvement in quality of life compared with high-dose SSA. Adverse events were higher in the PRRT group, however only mild in nature including nausea and vomiting due to the concomitant infusion of nephroprotective amino acids, fatigue or abdominal pain, and diarrhea. Hematotoxicity grade 3-4 was low and included neutropenia, thrombocytopenia, and lymphopenia in 1%, 2%, and 9% of patients, respectively, vs none in the SSA arm [59]. This game changing study led to the FDA and European Medicines Agency approval of 177 Lu-DOTATATE for unresectable or metastatic, progressive, well-differentiated, SSR-positive NETs in 2018. As median PFS and OS had not been reached at the data-cutoff date, final analyses are forecasted either after 158 deaths or 5 years after randomization of the last patient and are anticipated soon.
Currently, research is shifting towards PRRT with alpha emitters. Beta emitting radionuclides have a longer range and hence penetration, which impacts adverse events. Alpha emitters not only have a short range but also deliver a high linear dose to the tumor cells, potentially increasing efficacy. Patients with GEP-NET frequently have liver-dominant disease.
Liver-directed treatment can deliver a high tumoricidal dose to the hepatic metastases while sparing surrounding healthy liver tissue. A first in human study showed promising results with liver targeted treatment with intraarterial application of Bismuth-213 ( 213 Bi)-DOTATOC [60]. Other strategies include the use of somatostatin antagonists as antagonists seem to bind to more sites on receptors, even though they are not internalized into the cell. SSR antagonists have a favorable pharmacokinetic and better tumor visualization than agonists [61,62]. The SSA antagonistic theranostic pair 68 Ga-NODAGA-JR11 and 177 Lu-DOTA-JR11 ( 177 Lu-OPS201) are currently evaluated in phase I/II multicenter studies (ClinicalTrials.gov identifiers NCT02592707, NCT02609737).

Prostate Cancer -PSMA and GRPR
PSMA is a transmembrane glycoprotein which is highly overexpressed in prostate cancer (PC) [63]. Despite the suggestive name, PSMA is also expressed on various other cancers including breast [64] and renal cell cancer [65]. High expression is correlated with advanced, high-grade, androgen-independent disease [66]. PSMA-targeted compounds are now the most widely used PET radiotracers for staging in intermediate-and high-risk disease and at biochemical recurrence, especially with low prostate specific antigen (PSA) level [67][68][69]. Once conjugated with the chelator DOTA, they can be labeled with positron emitters like 18 F and 68 Ga for diagnostics and with beta ( 90 Y, 177 Lu) or alpha emitters ( 213 Bi, 225 Ac) for therapy. 68 Ga-labeled PSMA inhibitors such as PSMA-11, PSMA-I&T (imaging and treatment) and PSMA-617 are the most commonly used. However, 18 F-labeled tracers have the advantage of a slightly longer half-life (110 minutes) which allows a wider distribution along with a high yield production in a cyclotron facility. As highly anticipated, the FDA just recently approved the 18 F-labeled PSMA PET agent DCFPyL [70]. PSMA-617 is the most studied ligand for therapy as it showed favorable pharmacokinetics with high internalization, tumor retention and rapid renal clearance [71]. Treatment with 177 Lu-PSMA for metastatic castration resistant PC (mCRPC) has been introduced nearly a decade ago. Results from prospective phase II clinical trials were pivotal with a decline of PSA levels of greater than 50% in 45% of patients whereas any decrease was observed in 60% of patients. Treatment response was registered in up to 82% according to Response Evaluation Criteria in Solid Tumors (RECIST) and Positron Emission Tomography Response Criteria in Solid Tumors (PERCIST) criteria. PFS and OS were improved, and patients reported pain relief and increased quality of life. Treatment side effects were mild, including fatigue, nausea, and pain. Hematological toxicity was also reported mild and reversible in 90%. However, due to the high uptake in salivary glands, xerostomia was common, however, transient [72][73][74][75][76]. The TheraP study, a randomized phase II trial, compared treatment with 177 Lu-PSMA-617 to chemotherapy with cabazitaxel. A higher PSA response rate (66% vs 37%) and lower adverse events (33% vs 53%) were seen in the 177 Lu-PSMA-617 arm compared to cabazitaxel [77]. In contrast, the VISION trial compared treatment with 177 Lu-PSMA-617 and standard of care (SOC) to SOC alone, i.e., no active chemotherapy regimen in the control group, in PSMA-positive mCRPC patients who had previously received treatment with next-generation androgen receptor signaling inhibition (abiraterone, enzalutamide, etc.) and one or two prior lines of taxane chemotherapy. The first results of this international, randomized, open-label phase III study were just presented at the annual conference of the American Society of Clinical Oncology (ASCO) in the beginning of June 2021: The treatment with 177 Lu-PSMA-617 plus SOC showed a 60% reduced risk of progression (median PFS 177 Lu-PSMA-617 and SOC vs SOC: 8.7 vs 3.4 months), and a 38% reduced risk of death (median OS 177 Lu-PSMA-617 and SOC vs SOC: 15.3 vs 11.3 months). The objective response rate was significantly higher in the 177 Lu-PSMA-617 plus SOC arm than in the SOC arm (29.8% vs 1.7%), as was the disease control rate (89.0% vs 66.7%) and time to first symptomatic skeletal event (11.5 vs 6.8 months). Despite a higher rate of high-grade (grade 3 -5) 177 Lu-PSMA-617 related adverse events, the treatment was well tolerated. Adverse events included bone marrow suppression, xerostomia, nausea and vomiting [78]. The sponsoring company announced that they will file for regulatory approval for 177 Lu-PSMA-617 in the US and Europe.
Alpha emitting PSMA agents for treatment have been a field of active research in the last years. In a first in man study, 225 Ac-PSMA-617 achieved biochemical and radiological complete remission with low hematotoxicity but severe xerostomia, which remains the dose-limiting toxicity [79]. A retrospective analysis of 225 Ac-PSMA-617 TRT in 73 patients showed a decline in PSA greater than or equal to 50% in 70% of patients, which was associated with a longer OS. The median PFS was 15.2 months and OS was 18 months. Xerostomia was seen in 85% of patients but did not lead to discontinuation of treatment [80]. However, the patient population was less pre-treated than in comparable studies [81]. More prospective studies comparing alpha emitting agents to chemotherapy are needed.
Gastrin releasing peptide receptor (GRPR) has emerged as a promising novel target. It is highly expressed in PC and shows only low expression in benign prostate tissue [82]. Initially, GRPR agonists were synthesized but due to their internalization into the cell, caused side effects. Inspired by research on SSR antagonists, research shifted towards GRPR antagonists. The most studied and promising compound is 68 Ga-RM2 which has been evaluated for initial diagnosis [83,84] and biochemical recurrent disease [85,86] and showed promising results with high tumor uptake and favorable biodistribution. However, one caveat is that GRPR imaging might be limited in patients with hormonal treatment as this may affect GRPR expression in the tumors [87]. More prospective trials with larger cohorts using GRPR antagonists are needed, especially in comparison to the currently most widely used 68 Ga-PSMA-11, to better understand their relationship and potential predictive value of tumor biology and aggressiveness in different types and stages of PC.
Several theranostic pairs were investigated preclinically, however the first compound explored in human is 177 Lu-RM2 [88]. Four patients with mCRPC without any further treatment options were treated with 177 Lu-RM2 after prior confirmation of GRPR expression through 68 Ga-RM2 PET/CT. The highest absorbed doses were seen in bone lesions, followed by lymph node and soft tissue metastases and were all therapeutically relevant. The highest physiologic uptake was measured in the pancreas, which became the dose limiting organ [88]. A phase I/IIa open-label, multicenter trial is currently underway evaluating biodistribution, dosimetry and safety, tolerability, and antineoplastic activity of the GRPR antagonist 177 Lu-NeoB in patients with GRPR expressing, metastatic solid tumors (ClinicalTrials.gov identifier NCT03872778).

Bone Seeking Pseudotheranostic Pairs
Many cancers metastasize to the skeleton causing pain, pathologic fractures, and spinal cord compression thus driving morbidity and mortality. Osteoblastic metastases are characterized by increased bone metabolism which can be targeted by bone seeking radiopharmaceuticals such as 99m Tc-biphosphonates for SPECT and sodium fluoride-18 (NaF 18 ) for PET imaging, and subsequent treatment labeled to an alpha or beta emitter. Bone seeking radiopharmaceuticals accumulate in the bone depending on the degree of osseous metabolism, meaning the higher the bone turnover, the higher the accumulation of the radiopharmaceutical, hence higher affinity to osteoblastic metastases than healthy bone [89]. Bone palliation has been first treated with 32 P and subsequently other beta emitters such as Strontium-89 ( 89 Sr), Rhenium-186-hydroxyethylidene diphosphonate ( 186 Re-HEDP), 153 Sm-ethylenediamine tetramethylene phosphonate ( 153 Sm-EDTMP), 177 Lu-EDTMP, and the alpha emitter 223 Ra. Bone seeking agents can be grouped into calcium analogs like 89 Sr and 223 Ra, and radionuclides labeled to a phosphate like 186 Re-HEDP and 153 Sm-EDTMP. For radiolabeled phosphates and 89 Sr, bone palliation has been shown to occur 2 -10 days after treatment with overall pain relief response rate of 59 -86%. However, survival is not significantly impacted [90][91][92][93][94][95][96][97]. 223 Ra-dichloride is the currently most widely and commonly used bone targeted treatment. It has been FDA approved in 2013 for symptomatic bone metastases in mCRPC without known visceral metastases. 223 Ra is a calcium analog, which forms complexes with the bone mineral hydroxyapatite in areas of high bone metabolism. It therefore does not target the tumor cell itself but rather accumulate in between them. The short range of the alpha emitter allows high LET to the tumor stroma, causing DNA double-strand breaks, but minimizes damage to the surrounding normal tissue, especially the bone marrow [98]. The landmark study which led to approval was the ALSYMPCA trial, a randomized double blinded multicenter phase III clinical trial. Patients, before or after chemotherapy with docetaxel, were stratified in a treatment group which received six cycles of 223 Ra with an interval of four weeks, and a placebo group receiving best supportive care. The key results were a significant delay in time to first symptomatic skeletal event in the 223 Ra group vs placebo group (15.6 vs 9.8 months) and better OS (14.9 vs 11.3 months). Adverse events were seen slightly more frequently in the placebo than in the 223 Ra group (96% vs 93%). 223 Ra related side effects included mainly hematotoxicity, nausea, and other gastrointestinal reactions. Quality of life has been significantly improved in the treatment group [99]. PSA showed to be unreliable for response assessment. First treatment effects were reduced bone pain (typically after two weeks) or decreased alkaline phosphatase. Furthermore, the ALSYMPCA trial showed that 223 Ra was safe and effective when used before or after chemotherapy with docetaxel. However, concurrent 223 Ra therapy and chemotherapy is not recommended. Adding 223 Ra therapy to abiraterone or enzalutamide has been reported safe, however the combination treatment did not improve bone palliation, on the contrary, was associated with increased incidence of pathologic fractures and mortality [100]. A small clinical trial showed that patients who had previously been treated with 223 Ra and progressed afterwards can be safely re-treated without serious drug-related adverse events in a two year follow up after re-treatment [101].
Although 223 Ra was only approved for PC, 223 Ra could target osteoblastic metastases of any malignant origin. A recently published phase II study showed high disease control rate of 49% and tumor response rate of 54% of 223 Ra combined with hormonal therapy in hormone receptor-positive, bone-dominant metastatic breast cancer [102]. Osteosarcoma cells are also known to have a high avidity for bone seeking agents, not only at the primary site but also in soft tissue metastases, therefore making 223 Ra an ideal systemic therapy for patients with metastatic osteosarcoma. A single phase I dose escalation study has been published hitherto showing feasibility and safety of up to six cycles 223 Ra with minimal hematologic toxicity in a small cohort of 18 patients with high-risk osteosarcoma [103]. Further studies are expected.

Neuroblastoma, Pheochromocytoma, Paraganglioma -mIBG
Neuroblastoma is the most frequent extracranial solid tumor in infancy which derives from primitive neural crest cells from the sympathetic nervous system. These tumors are heterogenous in nature, varying in location, and clinical behavior ranging from spontaneous regression to aggressive, metastatic disease [104,105]. Pheochromocytoma and paraganglioma (PPGL) are also rare tumors arising from the neural crest. Pheochromocytomas originate from the adrenal medulla and paragangliomas from extra-adrenal, the sympathetic or parasympathetic ganglia. As these tumors are very heterogenous, their management is challenging, especially in the advanced, metastatic setting [106][107][108]. The distinct feature of tumors deriving from the neural crest is the overexpression of norepinephrine transporter on their cell membrane. This is found in 90% of neuroblastomas [109] and 50-60% of PPGL, whereas expression might be low in head and neck paragangliomas [110]. High expression of norepinephrine transporter is also found in other NETs like GEP-NET, MTC, Merkel cell carcinoma, and ganglioneuroma. Meta-iodobenzylguanidine (mIBG) is a norepinephrine analog which is taken up by norepinephrine transporters and subsequently stored in neurosecretory granules via vesicular monoamine transporters 1 and 2, thus irradiating the cancer cell when radiolabeled to 131 I [107]. 131 I-mIBG has been approved by the FDA in late 2018 for the treatment of mIBG positive, unresectable, locally advanced, or metastatic pheochromocytoma or paraganglioma based on the results of an open-label, single-arm, multicenter clinical trial involving 68 patients. Treatment response was achieved in 22% of patients receiving a single dose and increased to 32% in patients who received two doses. Ninety-two percent of patients who received at least one dose, showed partial response or stable disease. The median OS was 36.7 months. Clinical benefit was reflected in a 50% or greater reduction of all antihypertensive medication for at least six months in 25% of patients. Patients without hypertension were not included in this study, however, this might not have an impact on overall oncologic course of disease. The most common side effect was of hematologic nature in 90% of patients, whereas 72% were severe (grade [3][4]. Despite the high percentage of severe adverse events, they were transient and resolved without intervention in all but 25% who needed supportive care such as red blood cell and platelet transfusion, granulocyte colony-stimulating factors or erythropoietin for a limited time. In 16% of patients, 131 I-miBG treatment had to be discontinued due to persistent severe myelosuppression or other non-hematologic adverse reactions such as nausea. Myelodysplastic syndrome was seen in 4% and acute leukemias in 3% [111]. Of note is that patients in this study were not prospectively stratified by genetic mutations. Especially succinate dehydrogenase complex iron sulfur subunit B (SDHB) mutations are associated with an unfavorable prognosis, and false negative 123 I-mIBG scans have been reported [112]. Patient preparation for 131 I-mIBG diagnostic and treatment includes discontinuation of medication that interfere with norepinephrine transporters for at least five half-lives before and seven days after treatment (i.e., blood pressure medication such as combined alpha/beta blocker labetalol and calcium channel blockers, antidepressants, tramadol and pseudoephedrine) to avoid any false negative scans or transporter saturation a priori. As mIBG is radiolabeled with 131 I, thyroid blockade should be given at least 24 hours prior to treatment and ten days afterward.
For neuroblastoma, therapy with 131 I-mIBG showed promising results with a 30% -40% response rate, especially in high-risk neuroblastoma with refractory, relapsed or resistant conditions [113,114]. Hematological side effects are commonly seen, often amplified by combined chemotherapy.
As PPGL are also NETs, they express SSR on their cell surface which makes theranostic with 68 Ga-DOTATATE and 177 Lu-DOTATATE possible. Especially in patients with metastatic, extra-adrenal primaries or familial PPGL, the sensitivity of mIBG may drop to 53 -61%. SDHB mutation is the most common germline mutation seen in PPGL. It is characterized by a high malignant transformation (30%), shorter survival (5-year survival rate 36% vs 67% in patients without SDHB mutation) and 20% do not produce norepinephrine [115,116]. These patients might benefit from PRRT. In hereditary forms and germline mutations, the various radiopharmaceuticals for molecular imaging allow for stratification and selection of the right treatment for the patient.

Future Developments in Theranostics
Active preclinical research has resulted in the development of new molecular targets with high potential in translating into the clinic and application through theranostics. Here, we present just a few of the many novel targets evolved.

Fibroblast Activation Protein Inhibitors -FAPI-04
A malignant tumor contains not only cancer cells but also non-malignant cells such as inflammatory and vascular cells, and fibroblasts that form the tumor microenvironment, the so-called tumor stroma. Cancer-associated fibroblasts make 90% of the tumor mass and highly express the membrane bound glycoprotein fibroblast activation protein (FAP) in epithelial carcinomas, especially in colorectal, breast, pancreatic and PC [117]. FAP is involved in tumor proliferation and escape from immunosurveillance [118,119]. Various small molecule FAP inhibitors (FAPI) have been developed for imaging and treatment with promising results. 68 Ga-FAPI-02 and 68 Ga-FAPI-04 showed high uptake in several cancers with the highest uptake in sarcoma, esophageal, breast, lung and cholangiocellular carcinoma [120]. Preclinical studies on human pancreatic cancer cells with 225 Ac-FAPI-04 showed significant tumor growth delay [121]. In a first in man study, 90 Y-FAPI-04 was administered to one patient with metastasized breast cancer and reduced pain medication significantly. Despite the high uptake in the tumors, a fast clearance was observed [122]. Improved ligands with a longer tumor retention time are needed to improve tracer pharmacokinetics. Clinical data on treatment with FAPI are yet scarce.

Neurotensin
Neurotensin is a neuropeptide exhibited by the gastrointestinal and central nervous system as well as the myocardium. Neurotensin receptor 1 has been found to be overexpressed in various cancers such as colorectal, small cell and non-small cell bronchial, breast and pancreatic cancer. A first in man study showed the feasibility of treatment with 177 Lu-3BP-227 in six patients with pancreatic adenocarcinoma. The therapy was well tolerated with reversible grade 2 anemia as adverse event. One patient achieved partial response and significant improvement of symptoms while the other survived 11 months after treatment with 177 Lu-3BP-227 [123]. Currently, a multicenter phase I/II clinical trial is underway to evaluate the safety, tolerability, biodistribution and antitumor activity of 177 Lu-3BP-227 in solid tumors expressing neurotensin receptor 1 (ClinicalTrials.gov identifier NCT03525392).

Immuno-agents
The CXC-chemokine receptor type 4 (CXCR-4) is highly overexpressed in various human cancers like leukemia, lymphoma, and multiple myeloma [124]. Receptor imaging with 68 Ga-labeled pentixafor has shown promising results for imaging of multiple myeloma and lymphoma. A first in human study explored 177 Lu-and 90 Y-labeled pentixather for the treatment of relapsed multiple myeloma and showed partial and complete response to CXCR4-directed treatment [125].
The anti-CD20 monoclonal antibody rituximab is a standard treatment for B-cell non-Hodgkin lymphoma. In the setting of relapsed or refractory disease, 131 I-tositumomab and 90 Y-ibritumomab tiuxetan have demonstrated efficacy [126,127]. However, anti-CD20 resistant disease calls for the necessity of other targets. The CD37 targeting agent 177 Lu-lilotomab satetraxan has been explored in a phase I/IIa study with an overall response rate of 70% whereas 32% showed complete response [128]. Adverse events included transient grade 3 and 4 neutropenia and thrombocytopenia. These very promising results warrant future clinical trials.
Future objectives of theranostics not only include the development of new molecular targets, but also improvement of currently available treatments. The implementation of TRT earlier in the course of disease might improve quality of life as reported side effects from chemotherapies and other targeted systemic treatments are higher (52,53). Dosimetry estimates cumulated radiation dose and in a personalized medicine fashion will allow for individualized dosing and number of treatment cycles as opposed to current fixed regimen of dose and treatment cycles [129]. Currently, TRT mostly aim at stabilizing disease and improving quality of life in the palliative setting. However, a combination of treatments, especially in the early stage of disease might have potential of achieving complete remission. This is currently explored in small-cell lung cancer treated with 177 Lu-DOTATATE combined with nivolumab (ClinicalTrials.gov identifier NCT03325816) or in mCRPC combining treatment with 177 Lu-PSMA-617 with pembrolizumab (ClinicalTrials.gov identifier NCT03805594). Other combinations include TRT with chemotherapy and radiation therapy (65).

Conclusion
In this era of precision medicine in oncology, theranostic is the ultimate example of targeted, personalized diagnostic and treatment as we treat what we see. Each tumor's molecular profile is unique and so should the therapeutic approach be. Theranostic allows for imaging these specific tumor markers and patient selection and stratification so that the most beneficial treatment can be chosen. Active research in the field of theranostic will evolve more novel targets and improve radiopharmacokinetics. In the future, dosimetry will pave the way to an even more personalized treatment approach by moving from empiric standardized doses to individualized treatment doses and cycles. Molecular targeted radioligand therapy will become a major part of daily clinical nuclear medicine.