Recent Advances and Impending Challenges for the Radiopharmaceutical Sciences in Oncology

Introduction

The use of radionuclides for the imaging and treatment of cancer has undergone a dramatic transformation in the 80 years since Saul Hertz first attempted to treat thyroid cancer patients with radioiodine. While the field has undergone ebbs and flows in the intervening decades, it is hard to deny that the present represents a pivotal moment for the use of radiopharmaceuticals in oncology. Indeed, the increased clinical use of radiopharmaceuticals for both diagnosis and therapy, a surge in regulatory approvals, and an upwelling of preclinical research have combined to bring targeted radiopharmaceuticals — and especially radiopharmaceutical therapies (RPTs) — into the oncologic mainstream. In light of this advent, we offer this timely review of the most exciting recent developments in the field as well as the obstacles that must be overcome for it to continue on its remarkable trajectory. For the ease of the reader, we have divided the work into two sections: the first addresses the basic, preclinical, and clinical science underpinning the field (Scientific Developments), while the second addresses how the scientific and clinical work is performed (Operational Developments).

Before we begin, however, it is important to cover the fundamentals so that readers — from medical students to veterans in the field — begin on the same page. A radiopharmaceutical combines a radionuclide that emits imageable and/or therapeutic radiation with a carrier molecule that delivers the radionuclide to the target (though in some cases, radioiodine for example, the nuclide itself is responsible for both targeting and irradiation) (Figure 1). Many radiopharmaceuticals are labelled with metallic radionuclides and thus need a third component: a chelator that stably binds the radiometal to the radiopharmaceutical during its transit throughout the body. For imaging, the radionuclide should emit either positrons for positron emission tomography (PET) or photons for single-photon emission computed tomography (SPECT). For radiopharmaceutical therapy, radionuclides that emit ß⁻ particles (electrons) have proven highly effective in clinical practice (1-3). In addition, isotopes that produce α particles (4-6) — emissions capable of depositing extremely high amounts of energy across their short path-lengths — have been used clinically [e.g. ²²³RaCl₂ (Xofigo^®)] and have shown promise in advanced clinical trials (e.g. [²²⁵Ac]Ac-DOTA-TATE; [²²⁵Ac]Ac-PSMA-617). Finally, radiopharmaceuticals containing Auger-electron-emitting nuclides remain predominantly in the preclinical arena, though a handful of centres have translated them to the clinic (7)

FIGURE 1:

Illustration of how radiotheranostics are used: imaging scans acquired using a tumour-targeting vector (green hexagon) labelled with a diagnostic radionuclide (yellow) facilitate patient selection, individualized dosimetry, and treatment monitoring for radiopharmaceutical therapy using the same vector labelled with a therapeutic radionuclide (orange).

The efficacy of a radiopharmaceutical ultimately depends on its component parts. The pharmacokinetic profile of a radiopharmaceutical is generally determined by the carrier molecule. Ideally, radiopharmaceuticals should accumulate rapidly in tumour tissues, eschew healthy tissues, and be excreted as quickly as possible from non-target tissues. Internalising radiopharmaceuticals are often desirable(8) insofar as the sequestration of radioactivity inside tumour cells increases the cytotoxicity of many radionuclides (for RPT) and minimizes the radiopharmaceutical’s redistribution throughout the body (for both RPT and imaging). Likewise, the rapid excretion of any unbound radiopharmaceutical decreases background signal in healthy organs and reduces radiation damage to non-target tissues. Given these requirements, vectors including small molecules, peptides, and low-molecular-weight proteins such as antibody fragments are particularly well-suited to the creation of radiopharmaceuticals. Monoclonal antibodies have also shown promise (9) as radiopharmaceutical vectors due to their exquisite affinity and specificity, but their long biological residence times (≥ 3 d) require care to avoid high radiation doses to healthy tissues, especially in the context of RPT.

Shifting gears to the other major component, radionuclides that emit positrons and photons can be employed for PET and SPECT, while isotopes that emit α particles, β⁻ particles, or Auger electrons are suited for RPT. Pairing the targeting molecule with the right radionuclide (and vice versa) is critical.(10) In nuclear imaging, it is generally held that the physical half-life of the radionuclide should align with the biological residence time of the targeting molecule. For example, rapidly clearing peptides and small molecules should be labelled with short-lived nuclides (e.g. ⁶⁸Ga), while proteins that are excreted more slowly should be labelled with long-lived nuclides (e.g. ⁸⁹Zr).(11) For therapy, selecting the appropriate radionuclide for a targeting vector requires the careful evaluation of the radiation doses delivered to both the target and healthy tissues through systemic dosimetry. Indeed, each type of therapeutic emission has a different linear energy transfer (LET) profile, and thus the ultimate cellular fate of the targeting molecule (i.e., extracellular, intracellular, or intranuclear) is an important consideration when selecting a radionuclide for RPT. Finally, as we will see, some radionuclides produce both therapeutic and diagnostic emissions, facilitating the creation of radiotheranostics that allow for simultaneous imaging and treatment with the same radiopharmaceutical.

Scientific Developments

The Supply of Radionuclides

After more than seven decades of clinical use, ¹³¹I-iodide remains the most widely used therapeutic radionuclide(12) in the clinic due to its exceptional biological targeting properties for thyroid diseases like hyperthyroidism and differentiated thyroid carcinoma. Advantageous emission and dosimetry properties have also led to the use of ¹²³I and ¹²⁴I for imaging patients prior to treatment with ¹³¹I. In addition, there are at least 60 ongoing clinical trials of¹³¹I-labelled radiopharmaceuticals for imaging and therapy across the globe(13). Most of the worldwide demand for ¹³¹I focuses on its [¹³¹I]NaI form, a trend that is likely to continue. Over the last few years, disruptions in reactor performance(14) and supply chain issues (especially during the COVID-19 pandemic) have emerged as challenges in the global market for ¹³¹I(15, 16); nevertheless, theranostic clinics will continue to rely on clinical procedures with ¹³¹I-labelled radiopharmaceuticals, generating an urgent need for solutions to the ongoing reactor issues that threaten the supply stability of ¹³¹I (and other radionuclides). In addition, clinicians continue to use [^99mTc]Tc-MDP to identify patients with osteoblastic bone metastases suitable for treatment with approved treatments such as [⁸⁹Sr]Sr Cl₂ and [¹⁵³Sm]Sm-EDTMP.

In recent years, the advent of RPT in the clinic and the increasing use of diagnostic and therapeutic radiopharmaceuticals labelled with isotopic pairs have led to surges in the demand for both established radionuclides and newer isotopes. With respect to RPT, the past decade has witnessed the regulatory approval of a range of radiopharmaceuticals labelled with several different therapeutic radionuclides (Table 1). Notably, radiopharmaceuticals labelled with ⁹⁰Y and ¹⁷⁷_Lu — including ⁹⁰Y-ibritumomab tiuxetan in the treatment of non-Hodgkin lymphoma, ⁹⁰Y-glass microspheres (TheraSphere^™) for the treatment of hepatocellular carcinoma, [¹⁷⁷Lu]Lu-DOTA-TATE (Lutathera^®) for the treatment of somatostatin receptor-expressing neuroendocrine tumors, and [¹⁷⁷Lu]Lu-PSMA-617 (Pluvicto^®) for prostate cancer therapy — have been approved in multiple countries and used widely for in the clinic. The production and availability of these therapeutic isotopes and radiopharmaceuticals vary between countries, but significant efforts are underway to secure reliable commercial supply chains.(12)

Table 1.

Radiopharmaceuticals Approved for Radiopharmaceutical Therapy

Agent	Approval (FDA, EMA)	Companion diagnostic(s)	Indication
[131I]Nal	1971	[131I]Nal	Differentiated thyroid carcinoma
[131I]I-tositumomabb	2003, 2003	[131I]I-tositumomabb	CD20+ R/R low-grade, follicular, or transformed non-Hodgkin lymphoma (NHL) following disease progression during or after rituximab.
131I-chTNT	2007d	-	Advanced lung cancer
[131I]iobenguane (or MIBG)	2018	[123I]iobenguane	NEN+ pheochromocytomas or paragangliomas2
[177Lu]-Lu-DOTA-TATE	2018, 2017	[68Ga]Ga-DOTA-TATE (USA); [64Cu]Cu-DOTA-TATE (USA); [68Ga]Ga-DOTA-TOC (EU and USA)	SSTR-positive gastroenteropancreatic neuroendocrine tumours
[177Lu]Lu-PSMA-617	2022	[68Ga]Ga-PSMA-11 (FDA approved in 2021 and 2022)	Metastatic CRPC following disease progression on AR inhibitors and taxane-based chemotherapy
[223Ra]RaCl2	2013, 2013	99mTc-bone scan	CRPC with symptomatic bone metastases and no known visceral metastases.
188Re-resin	2020e		Non-melanoma skin cancer
[153Sm]Sm-EDTMP	1997, 1998	99mTc-bone scan	Palliation of bone pain in patients with multiple painful skeletal metastases
[89Sr]SrCl2	1993, 1986	99mTc-bone scan	Palliation of bone pain in patients with painful skeletal metastases
[90Y]Y-ibritumomab tiuxetana	2002, 2004	[111In]In-ibritumomab	Relapsed and/or refractory (R/R), low-grade or follicular B-cell NHL; Previously untreated follicular NHL who achieve a partial or complete response to first-line chemotherapy.
90Y-microspheresc	2002, 2002	99mTc-hepatic artery shunt scan	Unresectable metastatic liver tumours from primary colorectal cancer with adjuvant intra-hepatic artery chemotherapy of FUDR (Floxuridine).
90Y-Glass microspheresc	2021	99mTc-hepatic artery shunt scan	Unresectable hepatocellular carcinoma

^aDiscontinued in the USA in 2021

^bApproval withdrawn and discontinued in 2014.

^cAdministered via hepatic artery.

^dApproved by the Chinese FDA.

^eApproved by TGA. CRPC, castration-resistant prostate cancer; NEn, norepinephrine transporter; DOTATATE, DOTA0,Tyr3-octreotate; EDTMP, ethylenediamine tetra(methylene phosphonic acid; MIBG, meta-iodobenzylguanidine; PSMA, prostate-specific membrane antigen; AR, androgen receptor.

One source of increased demand for new radionuclides has been the rise of RPT with α-emitting radionuclides.(5, 6, 17). The higher LET of α particles (relative to β particles) leads to more ionization events along the particle track. As a result, α particles cause a greater fraction of double strand breaks per track length than β particles, resulting in greater biological effectiveness.(18) Indeed, molecules labelled with α-emitting radionuclides have shown significant promise as radiotherapeutics in patients who have not responded to previous therapy with β-particle-based RPT.(19) In this context, ²²⁵Ac (t_1/2 ~10 d) has emerged as a particularly important radionuclide. Its decay chain to stable ²⁰⁹Bi results in the release of 4 different α particles, a phenomenon that increases its therapeutic efficacy as long as these particles can be constrained to the tumour site (but complicates both dosimetry calculations and quality control measures). Several ²²⁵Ac-labelled radiotherapeutics are currently in clinical trials, as are several other agents labelled with shorter-lived α-emitting radionuclides like ²¹³Bi (t_1/2 ~46 min) and ²¹¹At (t_1/2 ~7.2 h). In light of this surge, the field urgently needs to secure a reliable supply chain of α-emitting radionuclides to realize the full clinical potential of these tools.(20) Traditionally, ²²⁵Ac is derived from the decay of ^229Th, which itself is obtained from the decay of ²³³U. Alternatively, researchers are also investigating accelerator-based methods predicated on the proton irradiation of ²³²Th as an avenue for increasing the global production capacity of ²²⁵Ac. One concern with this strategy, however, is the co-production of the undesirable byproduct ²²⁷Ac (with a half-life of 21.8 years). Furthermore, while challenges exist for the use of electron accelerators or medium energy cyclotrons for the production of ²²⁵Ac (e.g. the availability and handling of ²²⁶Ra target material and the relative scarcity of such equipment), we are nonetheless encouraged by the work being done in this area and recommend the continued funding of these efforts until a stable supply of ²²⁵Ac is guaranteed.(21)

The rise in the clinical use of PET and SPECT agents for theranostic imaging has also created pressure to increase the production of radionuclides (Table 2). Both ⁶⁸Ga and ¹⁸F are increasingly paired with ¹⁷⁷Lu to create theranostic radiopharmaceutical pairs for imaging and therapy. Perhaps most famously, [⁶⁸Ga]Ga-DOTA-TATE (NETSPOT^™)/[⁶⁸Ga]Ga-DOTA-TOC (SOMAKIT^™) and [¹⁷⁷Lu]Lu-DOTA-TATE (LUTATHERA^®) have been effectively deployed for the theranostic imaging and RPT of patients with neuroendocrine tumours that overexpress somatostatin receptors (SSTRs). The use of imaging with the ⁶⁸Ga-labelled tracer to confirm the presence of the target and thus identify patients likely to respond the corresponding ¹⁷⁷Lu-labelled agent has succeeded in several settings, paving the way for the incorporation of radiotheranostics into mainstream patient care. However, the mismatched half-lives and physicochemical properties of ⁶⁸Ga and ¹⁷⁷Lu can lead to radiopharmaceuticals with different binding affinities and biodistributions, making accurate dosimetry calculations difficult at best.(22) Differences in the pharmacokinetic and pharmacodynamic behaviour of vectors when they are labelled with radionuclides of different elements has spurred research into the use of radiotheranostics bearing true elementally matched radionuclides (i.e. isotopologues) or, at the very least, radionuclides with closer chemical properties. The accurate prediction of radiation doses to tumour lesions and healthy tissues becomes especially important in the context of high-dose therapy or RPT with α-emitting radiotherapeutics. Examples of radioisotope pairs that can be used for imaging and therapy include ⁶⁴Cu (for PET) and ⁶⁷Cu (for β-RPT), ⁴³Sc (for PET) and ⁴⁷Sc (for β-RPT), and ²⁰³Pb (for PET) and ²¹²Pb (for α-RPT), as well as a variety of radioisotopes of terbium (¹⁵²Tb, ¹⁵⁵Tb, ¹⁶¹Tb and ¹⁴⁹Tb).(23) ⁸⁹Zr has also been used to identify targets in tumors suitable for treatment with ¹⁷⁷Lu-antibodies. [24]

Table 2.

Selected radionuclides in (or close to) clinical use for nuclear imaging and radiopharmaceutical therapy.

Nuclide	Application	Half-life (t½)	Notes
Imaging
61Cu	PET	3.3 h	Paired with 67Cu
18F	PET	109.8 m	Routinely available
68Ga	PET	67.7 m	Paired with 177Lu
124I	PET	4.2 d	Paired with 131I
123I	SPECT	8.0 d	Paired with 131I
43Sc	PET	3.89 h	Paired with 47Sc
44gSc	PET	3.97 h	Paired with 47Sc
203Pb	SPECT	51.9 h	Paired with 212Pb
152Tb	PET	17.5 h	Can be paired with 161Tb
99mTc	SPECT	6.01 h	Paired with 153Sm, 186Re and 188Re, 89Sr
86Y	PET	14.7 h	Paired with 90Y
89Zr	PET	78.4 h	Readily available
Imaging and Radiopharmaceutical Therapy
64Cu	PET, β therapy	12.7 h	Readily available
67Cu	β therapy, SPECT	61.8 h	Availability limited
67Ga	SPECT, Auger therapy	3.3 d	Readily available
123I	SPECT, Auger therapy	13.2 h	Readily available
131I	β therapy, SPECT	8.0 d	Readily available
111In	SPECT, Auger therapy	2.8 d	Readily available
177Lu	β therapy, SPECT	6.7 d	Paired with 68Ga; Readily available
186Re	β therapy, SPECT	3.7 d	Availability limited
188Re	β therapy, SPECT	17.0 h	Availability limited
47Sc	β therapy, SPECT	3.4 d	Availability limited
153Sm	β therapy, SPECT	46.3 h	Readily available
117mSn	SPECT, Auger therapy	13.6 d	Limited availability
161Tb	β therapy, SPECT	6.9 d	Paired with 152Tb and 68Ga
Radiopharmaceutical Therapy
225Ac	α therapy	9.9 d	Availability limited
211At	α therapy	7.2 h	Availability limited
213Bi	α therapy	45.6 min	Availability limited
212Pb/212Bi	α/β therapy	10.6 h/60 min	Availability limited
223Ra	α therapy	11.4 d	Readily available
227Th	α therapy	18.7 d	Availability limited
90Y	β -therapy	64.1 h	Paired with 86Y

PET, Positron Emission Tomography; SPECT, Single-photon emission computed tomography; d, days; h, hours

Methods of Radioisotope Production

As we have discussed, the increased demand for emerging radionuclides has inspired and generated a great deal of research into improving the efficiency of production methods (see, for example, our discussion of ²²⁵Ac above). However, demand is not the only impetus for the development of new approaches to radionuclide production. Indeed, several recent challenges — like the 2008 disruption to the ⁹⁹Mo supply chain caused by the shutdown of major reactors as well as interruptions in the supply of enriched materials(24) — have spurred investigations into alternative routes for the production of both new and established radionuclides. For instance, the use of electron beams for the production of radionuclides (e.g., ⁹⁹Mo, ⁴⁷Sc, ⁶⁷Cu, and ²²⁵Ac) via the photonuclear route has shown promise by successfully generating clinical-grade doses in some cases, but the technology needs further improvement before its widespread adoption.(25) In another example, nuclear reactors (used in many countries to produce long-lived radionuclides (e.g. ⁶⁰Co)) are currently being investigated for the production of short-lived radionuclides, providing the field with a potential new radionuclide source. Additionally, advances in accelerators and targetry for charged-particle reactions with energetic protons, deuterons, and α particles continue to expand the availability of both diagnostic and therapeutic radionuclides. For example, advances in the production of ²¹¹At using a cyclotron (26, 27) and the creation of production networks through cross-site collaborations may expand access to this short-lived alpha-emitter. Along the same lines, while the cyclotron-based production of ²²⁵Ac is under investigation, the handling of long-lived radioactive target material remains a significant challenge.(28)

Many isotope production routes require the use of stable, isotopically enriched material. The availability of existing separated isotopes and the relative lack of suitable isotope separation facilities worldwide may limit the development of some of these otherwise feasible isotope production routes.

Chelators

The rise of new radiometals for nuclear imaging and therapy has necessitated the creation of new chelators that can provide stable coordination architectures for these isotopes. This is an active area of research that has been the topic of a number of recent reviews.(29-31) In essence, each radiometal has an optimal chelator (or chelators) that, when attached to a vector, will provide a kinetically and thermodynamically inert environment for the nuclide during the radiopharmaceutical’s transit throughout the body. The design and synthesis of new radiometal chelators, however, requires intensive efforts in design, synthesis, and testing. As a result, groups will often turn to well-known and frequently deployed chelators such as DOTA or NOTA, even if they are not ideal. DOTA, for example, was designed to sequester Gd in MRI contrast agents and has been shown to complex both Ga and Lu efficiently and stably. Yet while DOTA has shown suitability in some agents labelled with radiocopper (e.g. [⁶⁴Cu]Cu-DOTA-TATE), several other studies have shown that it provides a suboptimal coordination environment for Cu²⁺, which may lead to the inadvertent release of the radiometal in vivo. The need for better chelators for radiocopper has led to the exploration of other coordination scaffolds, including cross-bridged tetra-azamacrocycles and sarcophagines. Emerging radionuclides like ²²⁵Ac and ²¹²Pb do form reasonably stable complexes with DOTA; however, the development of more specific and stable chelators for each radiometal remains an area of important research. In the end, regardless of the specific radiometal or chelator, the underlying tenet is clear: when designing a radiopharmaceutical, careful preclinical and clinical research is needed to ensure that the chelator is an appropriate match for the radiometal.

Expanded Uses for Radiopharmaceuticals

Earlier-Line and Earlier-Stage Disease

While RPT has gained a great deal of clinical momentum in recent years, it has generally been deployed as a late-line palliative treatment. This status quo is now changing, with ongoing investigations into somatostatin receptor 2 (SSTR2)- and prostate-specific membrane antigen (PSMA)-targeted RPT in the setting of earlier-line and earlier-stage disease. For example, [¹⁷⁷Lu]Lu-DOTA-TATE (LUTATHERA^®) recently demonstrated statistically significant and clinically meaningful progression-free survival in first-line advanced gastroenteropancreatic neuroendocrine tumors (NCT03972488). Similarly, ¹⁷⁷Lu-labeled PSMA-targeted small molecules are now being tested in men with chemo-naïve metastatic castrate-resistant prostate cancer (NCT04647526; NCT04689828; NCT04419402), metastatic hormone-sensitive prostate cancer (NCT04720157), oligometastatic or biochemically recurrent prostate cancer (NCT04443062; NCT05496959; NCT05079698), and locoregionally advanced or high-risk prostate cancer (NCT04430192; NCT05162573).

More Disease Indications

Clinical trials are also exploring the use of established RPT agents for new disease indications (see Table 3). For example, [¹⁷⁷Lu]Lu-DOTATATE is being investigated in patients with many other SSTR-positive diseases, including breast cancer (NCT04529044), meningioma (NCT03971461), pheochromocytoma (NCT04711135), and paraganglioma (NCT03206060). Xofigo (Ra-223 dichloride) is under investigation for other cancers with bone metastases, including non-small cell lung cancer (NCT02283749), renal cell carcinoma (NCT04071223), and breast cancer (NCT02366130, NCT02258451). In addition, [¹⁷⁷Lu]Lu-PSMA-l&T and [¹⁷⁷Lu]Lu-EB-PSMA-617 are under investigation for PSMA-positive adenoid cystic salivary carcinoma (NCT04801264; NCT04291300). Given the discovery that PSMA is expressed in many different cancers beyond prostate, it is expected that new applications of PSMA-targeting theranostics will continue to emerge (32).

Table 3.

Examples of radiopharmaceuticals currently in clinical trials in cancer patients

Nuclide	Therapeutic (Trade Name/Scientific Name)	Target	Tumour Type	Industrial Sponsor
Neuroendocrine tumours
177Lu	Satoreotide (OPS201, JR11)	STTR2	NENs	Ipsen/Satosea
177Lu	Edotreotide (DOTATOC, ITM-11)	SSTR2	NENs	ITM
177Lu	Lutathera (DOTA-TATE)	SSTR2	GEP-NETs, SCLC, GBM	Novartis (Advanced Accelerator Applications)
177Lu	DOTA-LM3	SSTR2	NENs	(42)
212Pb	AlphaMedix (DOTAM-TATE)	SSTR2	NENs	Radiomedix
225Ac	DOTATOC	SSTR2	NENs	(93)
225Ac	DOTATATE	SSTR2	NENs	(94)
225Ac	DOTA-LM3	SSTR2	NENs	(43)
Prostate Cancer
225Ac	RYZ101	SSTR2	GEP-NETs	Rayzebio
177Lu	TLX591 (Rosopatamab)	PSMA	mCRPC	Telix Pharmaceuticals
177Lu	PNT2002 (PSMA I&T)	PSMA	mCRPC	POINT Biopharma Lantheus
177Lu	PSMA I&T	PSMA	mCRPC	Curium US LLC
177Lu	PSMA-617	PSMA	mCRPC, mHSPC	Novartis
Other Cancers
225Ac	Actimab-A (Lintuzumab)	CD33	AML	Actinium Pharmaceuticals
225Ac	FPI-1434	IGF-1R	Solid tumours	Fusion Pharmaceuticals
225Ac	FPI-1966	FGFR3	Solid tumours	Fusion Pharmaceuticals
131I	TLX101 (Iodophenylalanine)	LAT-1	GBM	Telix Pharmaceuticals
131I	HER2-VHH	HER2	breast	VU Brussels, Belgium
131I	Iomab-B (Apamistamab)	CD45	AML	Actinium Pharmaceuticals
131I, 177Lu	Omburtamab	B7-H3	NB, MB	Y-mabs Therapeutics
177Lu	TLX250 (Girentuximab)	CA9	ccRCC	Telix Pharmaceuticals
177Lu	Betalutin (Lilotomab)	CD37	NHL	Nordic Nanovector
177Lu	FAP-2286	FAP	Solid tumours	Novartis
177Lu	PNT6555	FAP	Solid tumours	POINT Biopharma
177Lu	Hurlutin	αvβ3	Breast	Advanced Imaging Projects
177Lu	NeoB	GRPR	Solid tumours	Advanced Accelerator Applications
177Lu	DOTA-FAPI	FAP	various	Xiamen University, China
177Lu	EB-FAPI	FAP	various	Peking Union Medical College Hospital and Xiamen University, China
177Lu	PP-F11N	CCK-2R	thyroid	University Hospitals Basel, Freiburg, Zurich, Bern and PSI Villigen, Switzerland, Germany

STTR2, somatostatin receptor 2; PSMA, prostate specific membrane antigen; FAP, fibroblast activation protein; NENs, neuroendocrine tumours; GEP-NETs, gastroenteropancreatic neuroendocrine tumours; FGFR3, fibroblast growth factor receptor 3; GRPR, gastrin-releasing peptide receptor; SCLC, small cell lung carcinoma; GBM, glioblastoma; AML, acute myeloid leukaemia; mCRPC, metastatic castrate resistant prostate cancer; mHSPC, metastatic hormone sensitive prostate cancer; NHL, non-Hodgkins lymphoma; NB, neuroblastoma; MB, medulloblastoma; ccRCC, clear cell renal cell carcinoma. HER2 human epidermal growth factor receptor 2; CCK-2R, Cholecystokinin-2 receptor

Combining RPT with Immunotherapy and Other Therapies

Strategies predicated on the combination of RPT with immunotherapies are being widely explored in clinical and preclinical research.(23) Mechanistic evaluations of the effects of radiation on immunotherapy suggest numerous benefits, including the improved release of neoantigens, heightened pro-inflammatory effects, and the suppression of immunosuppressive cell populations within the tumour microenvironment. Clinically, external beam radiotherapy can sometimes lead to the induction of systemic anti-tumour immune responses and the generation of anti-tumour T cells.(33, 34) It is of great interest whether systemic β⁻ or α-emitters can have similar effects. Thus, multiple active clinical trials are exploring immune checkpoint therapy in combination with RPT, including [¹⁷⁷Lu]Lu-PSMA-617 (NCT03805594), [¹⁷⁷Lu]Lu-DOTA-TATE (NCT04525638), and ²²³Ra (NCT02814669; NCT03996473).

Clinical trials are also underway exploring other combination strategies that may enhance the efficacy of radiotherapeutics or offer synergistic effects. For example, several trials combine RPT with DNA repair pathway inhibitors such as PARP inhibitors (NCT03317392; NCT04375267). Other groups are exploring methods to upregulate target expression. Along these lines, redifferentiation therapy has been used in thyroid cancer patients refractory to radioiodine therapy.(23) Investigations combining RPT with chemotherapy agents have been limited, but this approach is certainly also feasible and is likely to have increasing importance as RPT is incorporated into the treatment of earlier stage disease.(35)(36)

Combination Therapies with α- and β-Labelled Radiotherapeutics

The FDA approval of [¹⁷⁷Lu]Lu-DOTA-TATE and [¹⁷⁷Lu]Lu-PSMA-617 — both ¹⁷⁷Lu-based β-RPTs — has been transformative for patients with neuroendocrine tumours and prostate cancer, respectively. Nonetheless, not all patients respond to RPT monotherapy. One key parameter in predicting response to ¹⁷⁷Lu-based RPT is the number of DNA strand breaks created by the radiotherapeutic. Because α particles have a higher LET than β particles, α-emitting radiopharmaceuticals can create a greater number of complex cytotoxic double-strand DNA breaks, potentially endowing them with greater therapeutic efficacy. Indeed, recent reports suggest that α-RPT has a more potent antitumour response than β-RPT and that its use should be expanded.(37) While still very preliminary, remarkable results have been achieved in advanced clinical trials by switching patients unresponsive to ¹⁷⁷Lu-based RPT to treatment with ²²⁵Ac-labelled radiopharmaceuticals.(38) Clinical trials should be encouraged to confirm and validate these results. However, it is important to note that α-RPT also carries a significant risk of toxicity, including dose-limiting xerostomia in the case of [²²⁵Ac]Ac-PSMA-617. Recent approaches to balancing efficacy and toxicity include the use of a tailored mixture of ¹⁷⁷Lu- and ²²⁵Ac-labelled radiopharmaceuticals. In fact, early preliminary reports investigating this approach in a small number of patients indicate that combination treatments may be feasible, safe, and effective and may also benefit from synergistic effects between the two radionuclides.(39)

A Wider Array of Targets for RPT

In recent years, the field of RPT has shifted toward radiotherapeutics labelled with ¹⁷⁷Lu, including agents developed for the treatment of lymphoma and hepatic cancer as well as SSTR-targeting agents for neuroendocrine tumours such as [¹⁷⁷Lu]Lu-DOTA-TATE. These tools are expertly discussed in recent reviews by Bodei, et al. and Sgouros, et al.(40, 41) Along these lines, novel SSTR antagonist-based peptide receptor radionuclide therapies (PRRTs) (e.g. [¹⁷⁷Lu]Lu-DOTA-LM3 and [²²⁵Ac]Ac-DOTA-LM3) are also gaining traction.(42) α-RPT using ²²⁵Ac-labelled SSTR agonists and antagonists(43) has shown favourable biodistribution and higher tumour radiation doses than SSTR agonists alone and may lead to improved outcomes in patients with neuroendocrine neoplasms (NENs), as may the addition of imaging (e.g. with [⁶⁸Ga]Ga-LM3 and [⁶⁸Ga]Ga-LM4(44) (Figure 2A; unpublished data). Another very important recent development in RPT has been the advent of PSMA-targeting radiopharmaceuticals that address the urgent need for better approaches to the treatment of prostate cancer. There are several entrants into this field, but [¹⁷⁷Lu]Lu-PSMA-617 (Pluvicto^™) is the frontrunner and was approved by the United States FDA in March 2022. Critically, the PSMA- and SSTR2-targeted radiotherapeutics serve as excellent examples of the clinical utility of combining imaging (e.g. with [⁶⁸Ga]Ga-PSMA-11 [FDA-approved] or [⁶⁸Ga]Ga-DOTA-TATE) and therapy (e.g. with [¹⁷⁷Lu]Lu-PSMA-617 or [¹⁷⁷Lu]Lu-DOTA-TATE) in a radiotheranostic approach to improve patient care via patient selection and treatment monitoring.

Figure 2A:

[¹⁸F]FDG-PET scans of a 60-year-old male diagnosed in April 2014 with pancreatic NEN (Ki67 52%) who underwent Whipple surgery, CAPTEM chemo, liver resection, and 4 cycles Lutathera in 2020. Upon recurrence in late 2021, further disease progression was observed despite numerous other treatments. Nearly complete hepatic remission was observed after PRRT with [¹⁷⁷Lu]Lu-DOTA-LM3 (timing of doses shown with blue arrows) and has persisted for 2 years^.

Figure 2B: [¹⁸F]FDG-PET scans of a 53-year-old patient diagnosed in 2010 with ovarian adenocarcinoma G3 who underwent surgery and chemotherapy. After recurrence in 2021, progressive disease was observed despite various chemo- and immunotherapies. Complete remission was achieved after tandem [¹⁷⁷Lu]Lu/[²²⁵Ac]Ac-3BP-3940 treatment (timing of doses shown with blue arrows) and is ongoing after 19 months.

In the wake of the encouraging clinical results obtained with SSTR2- and PSMA-targeting radiopharmaceuticals, new targets continue to arise. For a summary of the targets currently being explored in the clinic as well as their corresponding radiopharmaceuticals, see Table 3. An even wider range of tumour cell- and microenvironment-specific targets are currently being interrogated in preclinical research. An exciting emerging target for RPT is fibroblast activation protein (FAP). FAP inhibitors (FAPIs) were first described in 2014 by Jansen and colleagues at the University of Antwerp.(45) Diagnostic (and possibly therapeutic) radiopharmaceuticals based on radiolabelled low-molecular-weight FAPIs were first presented in 2018 by Haberkorn et al.(46) Later research produced encouraging clinical results for these quinoline-based inhibitors labelled with theranostic radionuclides.(47) One of the agents in particular — [⁶⁸Ga]Ga-FAPI-04 — showed rapid tumour uptake and has subsequently been investigated in clinical trials for multiple tumour types.(48-50)

The pioneering work of Osterkamp, Zboralski and others(51) led to the development of several ¹⁷⁷Lu-labelled FAP-targeted peptides and peptidomimetics that are currently in clinical use for different types of cancer.(52) First-in-human results on the compassionate use of [¹⁷⁷Lu]Lu-FAP-2286 have been obtained by Baum et al.(52) Recently, promising results have been shown for a theranostic approach using [⁶⁸Ga]Ga-3BP-3940 for PET/CT-based patient selection for FAP-targeted RPT with ¹⁷⁷Lu, ⁹⁰Y, ²²⁵Ac, and combinations of these radionuclides (Figure 2B; unpublished), especially in combination with immunotherapeutics (e.g. anti-VEGF antibodies like bevacizumab) or radiosensitising chemotherapeutics (e.g. doxorubicin or irinotecan). These results have created excitement for FAP-targeted radiotherapeutics, which could potentially fill a gap in current clinical practice and can be applied in several indications.(42) Beyond FAP, a diverse set of other exciting tumour-specific antigens are currently being evaluated as targets for RPT, including neurotensin, B7-H3, CXCR4, and carbonic anhydrase 9 (CAIX).(53-57)

Different Administration Routes

Radiopharmaceuticals are typically administered intravenously, although some approved radionuclide therapies are given topically or via intra-arterial injection (Table 1). Once either a PET or SPECT scan — increasingly acquired using a companion radiotheranostic — has confirmed a patient’s suitability for RPT, a dosing plan for the radiotherapeutic must be implemented. Given the importance of delivering an appropriate therapeutic dose to the tumour while sparing normal tissue, such planning should consider the radiotherapeutic’s pharmacokinetic and pharmacodynamic properties.(58) Along these lines, both radiotherapeutics and their companion imaging agents may benefit from state-of-the-art drug delivery systems.(59) For example, ongoing studies are investigating the potential advantages of intra-arterial (hepatic artery) vs. intravenous administration of [¹⁷⁷Lu]Lu-DOTA-TATE in patients with NET liver metastases.(60) In these studies, intra-arterial [¹⁷⁷Lu]Lu-DOTA-TATE was shown to be safe but did not lead to a clinically significant increase in tumour uptake. That said, the results suggest that the intra-arterial administration of [¹⁷⁷Lu]Lu-DOTA-TATE results in higher activity concentrations and higher absorbed doses in the hepatic metastases of patients with gastro-entero-pancreatic neuroendocrine tumours compared to standard intravenous doses.(61) While further clinical testing with other agents is needed, these preliminary reports suggest that intra-arterial administration may improve the efficacy of radiotherapeutics, a phenomenon that has also been seen with nanomedicines, chemotherapeutics, and ⁹⁰Y-microspheres (Table 1) (62, 63). In humans, RPT via the local delivery of [²¹³Bi]Bi/[²²⁵Ac]Ac-DOTA-substance P for gliomas has produced promising, prolonged survival parameters.(64, 65)

The Advent of Computational Methods

The use of artificial intelligence (AI) algorithms is expected to impact many aspects of nuclear medicine, from the discovery, development, and distribution of radiopharmaceuticals to the acquisition, reconstruction, and analysis of images.(66) AI is also expected to improve nuclear medicine image analysis in multiple areas, including the detection of lesions and dosimetry.(67-69) Indeed, AI techniques are already being used to standardize the quantification of PSMA PET/CT images and have already garnered FDA approval.(70) PYLARIFY AI^™, for example, is an FDA-approved medical device that enables standardized quantitation and reporting of PSMA PET/CT images.

While AI will not replace nuclear medicine physicians or medical physicists, its use in clinical practice could reduce the burdens of image analysis (thereby freeing up time for other tasks and reducing misinterpretations) and improve accuracy (thereby reducing inter-observer variability). In the coming years, as the use of AI becomes more widespread, trustworthiness and reliability will be key. Reproducibility, generalizability, and ethics also represent valid concerns that must be addressed. As such, the use of AI in nuclear medicine should follow the best practices defined for the field (66) as well as concepts like current Good Machine Learning Practice (cGMLP), for which regulatory agencies like the FDA are already developing guiding principles.(71) Developers must ensure that algorithms earn the trust of healthcare adopters and patients alike.

Leveraging Imaging and Dosimetry Techniques for Theranostics

The emergence of more sensitive PET scanners — in particular, whole-body scanners (e.g. the UExplorer Total Body PET scannerTM and the Biograph Vision QuadraTM) — means that both short- and long-lived PET radionuclides can be imaged for longer periods of time, thereby enhancing biodistribution and dosimetry studies. In addition, advances in quantitative and digital SPECT(72) have enabled the use of true elemental matched pairs of radiopharmaceuticals for imaging and therapy in cases where no suitable positron-emitting nuclide exists. These two advanced imaging techniques may prove especially useful during the early phases of the development of imaging radiotherapeutics, when they can inform structural modifications that may improve pharmacokinetic behaviour or help determine dosing strategies by facilitating serial quantitative imaging at multiple time points to produce time-integrated activity curves.

Advances in SPECT imaging and dosimetry also provide the opportunity for personalized dosimetry-guided RPT. With these techniques, clinical trials may more safely and rigorously investigate dose escalation or re-treatment. In the landmark DOSISPHERE trial, personalized dosimetry with dose escalation of ⁹⁰Y-microspheres resulted in improved objective response rates for patients with locally advanced hepatocellular carcinoma.(73) Current trials are investigating dosimetry-guided [¹⁷⁷Lu]Lu-DOTATATE in adults (NCT02754297) and children (NCT03923257).

Molecular Imaging the Response to Immunotherapy

Theranostic imaging can be combined with immunotherapy to predict both response and early resistance. For example, the visualisation of PD-L1 expression via PET using ⁸⁹Zr-labelled atezolizumab correlated better with clinical responses to treatment with atezolizumab than immunohistochemistry or RNA-sequencing-based predictive biomarkers.(69)

Therapeutic effects can also be driven by changes in and modifications of the tumour microenvironment, providing an additional diagnostic and therapeutic opportunity. While methods exist for delineating specific T cell populations (CD8+, CD4+) in ex vivo tumour tissue, novel imaging approaches to image T cell populations in vivo have also been developed.(74) Furthermore, opportunities to use nuclear imaging to determine the impact of cancer-associated fibroblasts (CAFs) on the immune suppressive microenvironment using radiolabelled fibroblast activation protein inhibitors (FAPI) have been reported.(47) Tumor-associated macrophages — which are known to suppress the immune response — represent another target for the imaging of immunotherapy; to wit, imaging CD206-expressing TAMs was shown in a preclinical study to predict tumour relapse in breast cancer.(75)

Operational Considerations

The Production and Availability of Radionuclides

Radionuclides are considered the active pharmaceutical ingredients (API) of approved radiopharmaceutical products. Radionuclides are primarily produced using research reactors (fission and neutron activation), cyclotrons, and electron accelerators, as well as through the decay of other (parent) radionuclides. Worldwide, around 60 research reactors are currently active in the production of radionuclides according to the International Atomic Energy Association (IAEA) research reactor database.(76) Of these, 8–10 constitute the major worldwide producers of medical radionuclides. (Table 4). On top of this, the increasing demand for positron-emitting radionuclides has led to a global surge in the installation and operation of cyclotrons. According to the IAEA database on “cyclotrons used for radioisotope production”,(77) around 1200 cyclotrons have been installed around the globe, and at least 800 of these routinely produce medical radionuclides for PET and SPECT. The majority of these cyclotrons are used for the routine production of ¹⁸F for [¹⁸F]FDG and (increasingly) for ¹⁸F-labelled PSMA-targeting probes. In response to the growth of nuclear medicine and molecular imaging, the number of cyclotrons installed around the world is increasing, thanks in large part to versatile designs that make it easier to construct cyclotrons with different sizes, energies, and targetry systems.

Table 4.

Global nuclear reactors supplying medical radionuclides (95)

^a h/d/w = operating schedule expressed as hours per day / days per week / weeks per year

Reactor	Location	Power (MW)	Schedule(h/d/wa)	Age (years)
BR2	Belgium	10	24/7/21	61
FRM-II	Germany	20	24/7/34	18
HFIR	USA	85	24/7/24	57
HFR	Netherlands	45	24/7/44	45
LVR-15	Czechia	10	24/7/30	65
MARIA	Poland	30	24/5/40	48
MURR	USA	10	24/6.5/52	56
OPAL	Australia	20	24/7/52	16
SAFARI	South Africa	20	24/7/44	57

Recent years have witnessed explosive growth in nuclear medicine due to the regulatory approval of several new diagnostic agents and the emergence of theranostics for the imaging and RPT of neuroendocrine tumours and prostate cancer.(78) For many PET centres, this growth has been a double-edged sword, burdening legacy cyclotron, radiochemical, and scanner infrastructure that had primarily been built and optimised for [¹⁸F]FDG patient volumes. The broad spectrum of newly approved PET radiopharmaceuticals has left some facilities unable to manufacture each radiopharmaceutical every day, complicating patient care.(79) PET/CT and PET/MRI scanners at many sites are also nearing capacity. While demand has been managed since the approval of new PET agents began in earnest following the introduction of regulatory mechanisms,(80) this has mostly been due to the limited utilisation of tools like amyloid- and tau- targeted tracers and a lack of therapeutic options that could be informed by PET scans (i.e. no need for theranostics). However, the recent approval of aducanumab and lecanemab as well as the number of other Alzheimer’s therapeutics in advanced clinical trials (e.g. donanemab) could also lead to an increase in the demand for amyloid- and tau-targeted PET(81) at the same time as demand surges for PSMA-targeted PET(82). This could stress the existing PET infrastructure to potentially critical levels. Relief will be difficult, given the high cost and long lead times associated with expanding and/or building new PET centres. Because funding, planning, and building such centres is complex and can take a number of years,(83, 84) it is urgent that planning for such expansions begin immediately for academic medical centres, commercial nuclear pharmacy networks, and standalone imaging centres.

At the same time, demand for therapeutic radionuclides is higher than ever. The regulatory approval of Lutathera^™ ([¹⁷⁷Lu]Lu-DOTA-TATE) and Pluvicto^™ ([¹⁷⁷Lu]Lu-PSMA-617) has created significant demand for ¹⁷⁷Lu, which is severely testing the limits of current supply chain and reactor infrastructure.(85) The plethora of novel ¹⁷⁷Lu-labelled radiopharmaceuticals in advanced clinical trials suggests that the demand for ¹⁷⁷Lu could soon surge again, further straining supply. The production of ¹⁷⁷Lu is predicated on the irradiation of isotopically enriched ¹⁷⁶Lu (or ¹⁷⁶Yb) targets with reactor neutrons and the subsequent processing of the irradiated target material. Relying on ageing reactor infrastructure to meet and even increase production is expected to be problematic (Table 4). As the construction of new reactors is difficult, expensive, time-consuming, and politically delicate, we recommend the field engage scientific and political leadership around the world now to prepare the critical radionuclide supplies needed for nuclear medicine’s future. A recent report from the European Union overviews the supply chain for therapeutic radionuclides.(86)

The Production of Radiopharmaceuticals

Both local and centralised methods are used for the production of diagnostic and therapeutic radiopharmaceuticals.(87, 88) Generally speaking, the choice between the two depends on a number of factors, including the source of the radionuclide, the physical half-life of the radionuclide, the availability of synthetic kits, and the local regulations governing pharmacy and manufacturing.

In the context of local production, the widespread availability of ⁹⁹Mo/^99mTc generators has fuelled the creation of a wide variety of kits for the synthesis of ^99mTc-labelled radiopharmaceuticals.(89) As a result, local compounding — the combination of two or more drugs, in this case ^99mTc and the kit — for SPECT remains common. Similarly, in light of the short half-life of many positron-emitting radionuclides, the local production of PET imaging agents like [¹⁸F]FDG as well as ⁶⁸Ga- and ¹⁸F-labelled SSTR- and PSMA-targeting radiopharmaceuticals remains the norm at academic medical centres and commercial nuclear pharmacy networks. In the future, local production and compounding may also become the norm for the next generation of radiotherapeutics labelled with short-lived radionuclides like ²¹²Pb.

As we have discussed, the use of radiopharmaceuticals bearing longer-lived radionuclides — e.g. ⁶⁴Cu and ⁸⁹Zr for PET, ¹⁷⁷Lu for β-RPT, and ²²⁵Ac for α-RPT — has surged in recent years. In these cases, centralised production is more sensible, as limited infrastructure exists for the production of these radionuclides, and their half-lives are amenable to transport. For example, [⁶⁴Cu]Cu-DOTA-TATE is manufactured and distributed from a centralised location, as are ¹⁷⁷Lu-labelled radiotherapeutics.

Another consideration in the production of radiopharmaceuticals is the use of synthesis modules and single-use kits. In the context of PET, the traditional production paradigm involves the use of synthesis modules and is considered drug manufacture. However, approved radiopharmaceuticals labelled with radiometals like ⁶⁸Ga and ⁶⁴Cu are often amenable to kit production (as is used for ^99mTc-labelling). In some regulatory environments, the use of kits with radiometals such as these no longer meets the definition of either ‘drug manufacture’ or ‘compounding’ (in the U.S., for example, ⁶⁸Ge/⁶⁸Ga generators, unlike ⁹⁹Mo/^99mTc generators, are not approved drugs). Instead, this sort of production qualifies as “drug preparation” according to an approved package insert. Clearly, in the years to come, the field will need to carefully keep abreast of the ways in which the ever-changing regulatory environment will impact the production of radiopharmaceuticals.

Generic Radiotherapeutics

While many novel imaging agents (e.g. Pylarify^™, Amyvid^™, Tauvid^™, Neuraceq^™, Vizamyl^™) are marketed as new drugs and protected by patents, established agents like [¹⁸F]FDG, [¹³N]ammonia, and [¹⁸F]sodium fluoride have no patent protection and are thus marketed as generic drugs. To date, only a small number of approved radiotherapeutics — i.e., Bexxar^™, Zevalin^™, Xofigo^™, Lutathera^™, PSMA-I&T, and Pluvicto^™ — have been marketed by pharmaceutical companies, while others, like low molar activity [¹³¹I]MIBG and ¹³¹I-sodium iodide, have no patent protection. However, imaging and therapeutic radiopharmaceuticals remain patent-protected only for a limited time. Notably, protection for [¹⁷⁷Lu]Lu-DOTA-TATE is set to expire in 2025.(90) As a result, opportunities will soon arise for companies and academic medical centres to create generic theranostics ([¹⁷⁷Lu]Lu-DOTA-TATE and others later). To capitalise on this, a number of start-up companies and academic medical centres are already building labs dedicated to the cGMP manufacture of radiotherapeutics. We expect this trend to continue.

Reimbursement

Reimbursement policies for nuclear imaging agents and radiotherapeutics are of critical importance as our field continues to grow.(91) While reimbursement rules vary from country to country and depend upon local healthcare policy, the field needs to be responsive to global challenges to ensure that our high-value products are available to the patients who will benefit from them.

While a survey of radiopharmaceutical access and reimbursement issues around the world is beyond the scope of this overview, the topic was recently reviewed in detail.(12) In many countries, the cost of a diagnostic radiopharmaceutical is “bundled” with the cost of the imaging procedure in the outpatient setting. In the United States, for example, a significant challenge with this approach is that Medicare’s ‘bundled’ rates are the same for an established agent such as [¹⁸F]FDG as they are for a novel, patent-protected agent such as [⁶⁸Ga]Ga-DOTA-TATE. This is the case even when the cost of the new diagnostic is higher than the entire bundled reimbursement for the imaging procedure! Unfortunately, issues like this can usually not be fixed at the local level but rather require unified efforts from the field to petition governments for changes in healthcare policy. The nuclear medicine community in the United States has been highly active in championing the Facilitating Innovative Nuclear Diagnostics (FIND) Act, which directs the Department of Health and Human Services to pay separately for all diagnostic radiopharmaceuticals with a cost threshold of $500 per day. It is hoped that this will gain congressional approval in the near future.(92)

Another important reimbursement issue is the cost of dosimetry calculations, which, while rarely standard today, are likely to become more frequently used in order to optimise the delivery of radiopharmaceuticals. In the context of external beam radiotherapy, dosimetry is routinely paid for, and planning systems are well-developed. However, this is not presently true for most dosimetry in service of RPT. That said, additional planning systems are in development, and several new companies are delving into this radiopharmaceutical dosimetry space. Ultimately, the development of appropriate reimbursement rates for dosimetry will prove important for the future of RPT.

Moving Forward

The unprecedented growth in nuclear medicine and the radiopharmaceutical sciences over the last decade has been fuelled by the emergence of several exciting new radiopharmaceutical therapies. The clinical approval of these drugs for the treatment of patients with NETs and prostate cancer has been pivotal in establishing radiotheranostics as major tools in oncology. However, ensuring access and availability on a global scale will be challenging and will require a scale-up of both production capacity and the number of sites suitable for the delivery of these treatments to patients. This expansion will require training the scientists and clinicians necessary to deploy novel theranostics in the clinic (for more on this, see our companion piece on the challenges faced by the nuclear medicine workforce in the age of radiotheranostics). In order to take full advantage of the potential of theranostics for improving patient care, a coordinated effort will be needed that links academia, industry, government agencies, and intergovernmental bodies. In this context, we propose the following approaches to enhance the development of radiopharmaceuticals for oncology:

On the scientific side, we believe the field could benefit from…

• support for the development of novel, efficient, and reliable production methodologies for both established and emerging radionuclides;

• the development of widely accessible online databases on the availability of medical radionuclides;

• an increase in international and interdisciplinary collaborations focused on both the production of radionuclides and the synthesis of radiopharmaceuticals;

• an emphasis on clinical trials focused on radiotherapeutics as first-line therapies or as part of combination therapies, particularly alongside immunotherapeutics;

• an expansion in the number of clinical trials focused on the use of companion theranostic imaging agents, particularly to facilitate individualised dosimetry and the selection of patients;

• the development of new chelation architectures suitable for emergent metallic radionuclides.

  On the operational side, we believe the field could benefit from…

• support for the dramatic expansion of radionuclide production facilities to meet the rising preclinical and clinical demand for both established and new radionuclides, particularly α-emitters, Auger-electron-emitters, and radionuclide pairs that can be used to create isotopologous theranostics;

• the simplification and modernization of reimbursement protocols to adequately cover novel radiopharmaceuticals;

• enhanced collaborations to establish interdisciplinary theranostics centres to address the growing number of patients that need RPT as well as follow-up assessment after treatment.

Ultimately, we have no doubt that the difficulties of overcoming the challenges associated with our field’s rapid expansion will be worth it. We are eager to see where our field goes in the next few years and are extraordinarily excited about what it could mean for our patients.

Summary

This paper is one of a series summarizing the current landscape of the radiopharmaceutical sciences as they pertain to oncology. In this article, we describe exciting developments in radiochemistry and the production of radionuclides, the development and translation of theranostics, and the application of artificial intelligence to our field. These developments are catalysing growth in the use of radiopharmaceuticals to the benefit of patients the world over. We also highlight some of the key issues to be addressed in the coming years in order to realize the full potential of radiopharmaceuticals in the fight against cancer.