Astatine-211 for PSMA-targeted α-radiation therapy of micrometastatic prostate cancer: a sustainable approach towards precision oncology

Prostate cancer (PCa) is the second most prevalent cancer among men and is a leading cause of morbidity and mortality worldwide [1]. Despite being a clinically heterogeneous disease, radical prostatectomy and external beam radiotherapy are the widely established treatment regimens for the early-stage PCa [1]. After treatment, the prostate specific antigen (PSA) rapidly falls to negligibly low levels and can be used thereafter as a tumor marker for PCa monitoring [1]. Majority of the patients treated with the above-mentioned treatment protocol eventually have disease recurrence within a few years. For these patients, androgen deprivation therapy (ADT) is an approved therapeutic option because androgen is required for the growth of the PCa cells [1]. Though ADT restricts disease progression briefly, most patients become castration-resistant within a span of few years. Advanced therapeutic options including taxane-based chemotherapeutic drugs, next-generation anti-androgen agents, and radiopharmaceuticals have been used extensively in the management of metastatic castration resistant prostate cancer (mCRPCa) [1–4]. However, mCRPCa still remains a highly fatal disease and demands for concerted efforts in developing novel therapies to mitigate the same.

Prostate-specific membrane antigen (PSMA) is a transmembrane glycoprotein expressed in all forms of prostate tissues and also in some other kinds of cells such as central and peripheral nerve tissues, kidneys, parotid and lachrymal glands [5]. PSMA is expressed in high levels on prostate cancer cells and also neovasculature of several other solid tumors. This overexpression of PSMA forms the basis of using PSMA as a target for delivering cytotoxic drugs and radionuclides for diagnosis and therapy. More importantly, PSMA overexpression levels are accentuated in mCRPCa [5]. This has led to rapid evolution of PCa targeted molecular imaging and therapy strategies in nuclear medicine over the last decade, which have focused on the successful development of radiolabeled small molecules that act as inhibitors to the binding of the N-acetyl-L-aspartyl-L-glutamate (NAAG) substrate to the PSMA molecule [5, 6]. Several peptidomimetic inhibitors of PSMA (such as PSMA-11, PSMA-617, PSMA I&T, MIP-1095, DCFPyL, etc.) typically having a Glu-ureido component that serves as the PSMA binding motif have been designed and translated for routine clinical use [4, 7–9].

In nuclear medicine, PSMA-targeted radioligand therapy (PRLT) of mCRPCa has most commonly been performed using ¹⁷⁷Lu (t_½ = 6.7 d), a β⁻ emitting radionuclide coupled to PSMA ligand (e.g. PSMA-617, PSMA I&T, etc.) [7, 10, 11]. Other β⁻-emitters in clinical trials include ¹³¹I, ⁹⁰Y and ⁶⁷Cu [7]. Among them, the safety and efficacy of [¹⁷⁷Lu]Lu-PSMA-617 have been clinically established for treatment of mCRPCa. This radiopharmaceutical has recently been approved by the US Food and Drug Administration (US FDA) for patients with progressive disease [12]. Despite remarkable efficacy of ¹⁷⁷Lu-based PRLT in prolonging the overall survival of patients with mCRPCa compared with standard treatment alone, a significant percentage of the patients actually do not respond to this treatment regimen. It is also worth mentioning that many patients who initially demonstrate response to ¹⁷⁷Lu-based PRLT may also experience disease progression at a later stage. This has therefore led to interest in evaluation of safety and efficacy of PSMA-targeted α-radiation therapy (TAT) in mCRPCa patients who are either unsuitable for or resistant to ¹⁷⁷Lu-based therapy [2, 5].

Due to theoretical advantages in the physics and radiation-biology of α versus β⁻ particle emitters, TAT can deliver effective localized radiation dose selectively to the PCa cells as well as the metastatic, micrometastatic and overt metastatic sites in lymph nodes, the skeleton, and visceral organs and thereby control mCRPCa while minimizing toxicity [2]. This can also potentially overcome the resistance to PRLT using β⁻-emitting radionuclides and/or treatment with standard chemotherapeutic drugs. Despite a plethora of available α-emitting radionuclides, only a few of them have desirable nuclear decay characteristics and favorable chemistry to make them suitable for clinical applications in TAT. In the recent past, α-emitting radionuclides, such as ²²⁵Ac, ²¹³Bi, ¹⁴⁹ Tb, ²¹²Pb/²¹²Bi, ²¹¹At, ²²³Ra, and ²²⁷Th, have been labeled to PSMA ligands and evaluated in preclinical and clinical settings [2]. Of these, ²²⁵Ac (t_½ = 9.9 d) and its short-lived daughter radionuclide, ²¹³Bi (t_½ = 46 min), have been most extensively studied [2, 10, 13, 14]. Availability of ²²⁵Ac is the major challenge faced worldwide as it is produced in very few centers from legacy nuclear fuel materials (²²⁹Th or ²³²Th) [15]. This sourcing limitation constrains its use to less than 1000 patients per year worldwide, making it an exorbitantly costly treatment option. Pursuit of alternate production routes met with limited success because of the radionuclidic purity issues (presence of ²²⁷Ac) and challenging targetry involving use of a scarcely available radioisotope ²²⁶Ra as the target. Additionally, ²²⁵Ac generate multiple α-particles per decay and for each daughter product the energy imparted by the recoil effects is several orders of magnitude greater than the chemical bonds in the radiopharmaceutical [16]. This most likely leads to release of daughter radionuclides from the targeting ligand which might cause undesirable effects [15, 16].

In a recent issue of the European Journal of Nuclear Medicine and Molecular Imaging, Watabe and colleagues have reported the development of novel PSMA targeted radiopharmaceuticals using an emerging α-emitting radio-halogen, ²¹¹At (t_½ = 7.2 h) [17]. ²¹¹At has the potential to circumvent most of the limitations of the first-generation α-emitters and is arguably the best option for TAT of mCRPCa. The 7.2 h half-life of ²¹¹At is sufficiently long enough to allow multi-step synthetic procedures and is also acquiescent with the pharmacokinetics of most peptidomimetic PSMA targeting ligands [18]. An important nuclear decay characteristic of ²¹¹At that makes it different from other commonly used α-emitters for TAT is that it yields only one α-particle per decay which not only simplifies the dosimetry calculations but also evades the undesirable off-targeting of daughter products. The double-branched decay scheme of ²¹¹At is associated with either direct α-decay to ²⁰⁷Bi (42%) followed by electron capture (EC) to stable ²⁰⁷Pb or by EC decay to ²¹¹Po (58%) followed by α-emission to stable ²⁰⁷Pb. This second decay branch of ²¹¹At is an issue as it can lead to escape of radioactivity from the site of ²¹¹At uptake. However, this issue is much less severe as compared to more widely used ²²⁵Ac as the half-life of ²¹¹Po intermediate is only 520 ms which limits its diffusion distance prior to α-emission. Also, since the progeny ²¹¹Po is not produced by α-decay, the potential destabilizing effect of α-particle recoil is not a concern in this case [19]. On the other hand, this EC decay branch actually leads to emission of 77–92 keV x-rays which offer the scope of SPECT imaging. Besides, ²¹¹At can be produced in reasonably high yield via ²⁰⁹Bi(α,2n)²¹¹At reaction utilizing inexpensive natural Bi target [19]. The methodology for isolation of ²¹¹At from the irradiated target by dry-distillation or wet chemistry is well established and commercially sustainable [20].

Fulfilling the emerging demands of radiopharmaceutical science, ²¹¹At offers a flexible chemistry to synthesize a wide spectrum of radiolabeled agents [18, 19]. Unlike other α-emitting radionuclides, astatine is a halogen and has similar chemical characteristics with iodine, notwithstanding with more metallic properties [21]. Owing to its existence in several oxidation states, astatine provides multiple synthetic options but also contributing to its sometimes-unpredictable behavior [21]. Currently, the most widely implemented radiolabeling approach is the electrophilic astadometallation of organometallic derivatives [21]. However, most of these methods require use of hazardous chemicals or are unable to yield the final product with high specific activity. To circumvent these limitations, a robust method involving substitution reaction of ²¹¹At with dihydroxyboryl groups has been adopted for radiolabeling in the present study [17, 22]. In this pursuit, new PSMA inhibitors (PSMA1, PSMA5 and PSMA6) were synthesized using the structure [¹⁸F]PSMA-1007 as scaffold and aryl boronic acid moiety was introduced for the ²¹¹At labeling [23]. These PSMA analogs have three different amino acid residues in their side chains: Gly-Lys (in PSMA1), ©-G©(R)-Glu (in PSMA5), and (S)-Glu-(S)-Glu (in PSMA6). After purification procedure involving solid-phase extraction (SPE), the radiochemical purity of all these ²¹¹At-labeled agents were > 96%. The differences in amino acid residues in [²¹¹At]PSMA1, [²¹¹At] PSMA5 and [²¹¹At]PSMA6 were evaluated by in vitro and in vivo evaluations in preclinical settings.

In vitro cellular uptake (in LNCaP and PC-3 cell lines) studies revealed PSMA-mediated uptake of both [²¹¹At] PSMA1 and [²¹¹At]PSMA5, though higher uptake was observed with the latter. In vivo SPECT imaging and biodistribution studies in LNCaP xenograft mice demonstrated higher tumor retention of [²¹¹At]PSMA5 compared to [²¹¹At]PSMA1 and [²¹¹At]PSMA6, whereas renal excretion was superior in [²¹¹At]PSMA1 compared to the other two agents. As expected, the remarkably high uptake of these agents in the kidneys was due to high level of PSMA expression in the kidneys and primarily renal excretion. Additionally, mild uptake of radioactivity was observed in the thyroid, spleen and stomach which indicated deastatination in vivo. In fact, [²¹¹At]PSMA5 showed slow deastatination in mice resulting in < 1% of the injected dose to be present in urine at 3 h post-injection. In the treatment studies, best tumor growth suppression without significant change in body weight of mice was observed with [²¹¹At]PSMA5 and therefore this agent was chosen as the main candidate for future translation.

Histopathological evaluation after treatment did not show any significant changes in the parenchyma, salivary gland, stomach, thyroid, spleen, and small intestine. Nevertheless, in vivo deastatination is a serious issue as evident from the biodistribution study and thyroid can be a risk organ for PRLT using ²¹¹At-based radiopharmaceuticals. This aspect needs to be evaluated carefully and if required blocking dose of iodine may be given to protect the thyroid by inhibiting the uptake of ²¹¹At [24]. Also, histopathological studies showed regenerated tubules in the kidneys at 3 and 6 weeks after the administration of [²¹¹At]PSMA5, which might be a result of radiation-induced toxicity due to PSMA expression in the proximal tubule of the kidney. Apart from this observation, proteinuria or increased BUN and creatinine levels were not seen after the administration of [²¹¹At]PSMA5. Still, chronic long-term kidney toxicity study must be performed from the perspective of future clinical translation.

Xerostomia is the most common side effect of PRLT [25]. Careful toxicity study of the salivary glands must also be performed, though there were no histological anomalies in salivary glands in the present study. Additionally, a more rigorous evaluation of [²¹¹At]PSMA5 is warranted from perspective of its translation for clinical use. The treatment study was performed in the LNCaP xenograft model using a single-dose administration. In the future, repeated administration or progressive dose escalation study should be carried out to mimic clinical circumstances and thus define a minimum or maximum effective dose for treatment in humans. Further, a detailed examination of biodistribution pattern at different time points is essential for an accurate dosimetry approach. Further, pharmacokinetic studies are required to perform precise estimation of dose absorbed while comparing it with histological abnormalities. Last but not the least, biochemical and hematological toxicity study, including myelosuppression, should be performed in greater detail to arrive at an optimal dose for PRLT using [²¹¹At]PSMA5.

Overall, this preclinical study amply demonstrated that [²¹¹At]PSMA5 or close analogs holds promise for translation if and when TAT is to be considered in the treatment protocol of mCRPCa patients, which for sure is a giant leap towards cost-effective precision oncology. Despite abundant availability of target material and robust radiochemical separation technology, the feasibility of sourcing ²¹¹At and its implementation for widespread clinical use is still a challenge as most accelerators in the world do not have beam characteristics required for producing meaningful levels of ²¹¹At [19, 20]. Providing worldwide supply chain of ²¹¹At mandates efficient utilization of existing facilities, installations of new cyclotrons and employment of well-organized chemistry infrastructure. In this direction, the US Department of Energy has funded augmentations and upgradation of 5 existing cyclotrons to improve abilities for ²¹¹At production [19]. The European Cooperation in Science and Technology (COST) has also funded a multi-institutional program focused on ²¹¹At production in the European Union [19]. With an overall same goal, an accelerator network has been established in Japan [19]. The recent interest of the commercial nuclear medicine sector in ²¹¹At supply is also an encouraging progress. Expectantly, with these initiatives there will be substantial near-term improvements in ²¹¹At availability for use in treatment of mCRPCa and beyond.