On the Use of ²⁰³Pb Imaging to Inform ²¹²Pb Dosimetry for ^203/212Pb Image-Guided Alpha-Particle Therapy for Cancer

Abstract

Alpha-emitting radiopharmaceutical therapy shows promise for improving the therapeutic efficacy of existing and future targeting ligands by limiting off-target irradiation and by preempting many cell survival mechanisms. Dosimetry-guided therapies are emerging as potentially safer and more effective than approaches based on a fixed-activity-administration paradigm. Among the candidates of alpha-emitting radionuclides, ²¹²Pb shows promise for use under an image-guided dosimetry-informed theranostic paradigm, whereby ²⁰³Pb can be used for dosimetry and treatment planning. In this chapter, we model an approach to accurately estimate the dosimetry of ²¹²Pb-based radiopharmaceuticals using ²⁰³Pb as a surrogate. However, uncertainties arise in dosimetric predictions for ²¹²Pb based on ²⁰³Pb imaging due to the potential for migration of ²¹²Pb radionuclide progeny (i.e., ²¹²Bi, ²¹²Po, ²⁰⁸Tl) from the site of ²¹²Pb decay. On the other hand, based on distinct gamma-ray energies of the ²¹²Pb progeny, the design of in vivo experiments is described that have the potential to define these uncertainties more precisely, so as to gain insights into the potential toxicity of bioconjugated and potentially decoupled ²¹²Bi in tissues. The promise of alpha-particle radionuclide therapy is evidenced by a tenfold increase in publications over the last 30 years, and it is anticipated that the elementally matched ²⁰³Pb/²¹²Pb radionuclide pair will play a key role in our progress toward personalized receptor-targeted alpha-particle therapy for cancer.

28.1 Introduction

Emerging evidence suggests that receptor-targeted radionuclide therapy for cancer has the potential to be transformative for cancer patient care [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]. Alpha-particle radionuclide therapy (α-RT), in particular, is receiving considerable attention given the potential advantages of α-RT relative to (beta) β-RT. [1, 3, 5, 6, 16] Of these advantages (relative to β-emitters), higher linear-energy transfer (LET) (100 keV/μm) and resulting increases in primary and secondary ionizations along a relatively short path length in tissue is considered a primary advantage [3, 5, 11, 12, 15, 17]. The major underlying reason for this is that high LET radiation deposition over this short path length results in an increase in double-strand DNA breaks, which is thought to improve cytotoxicity via an improved relative biological effectiveness (RBE) when compared to β-RT. [18,19,20,21, 5, 16, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36] Of the radionuclides under investigation for α-RT, ²²⁵Ac, ²¹¹At, ²¹²Pb, ²¹²Bi, and ²¹³Bi have generated considerable enthusiasm [5, 16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. Of these, the only available elementally identical radionuclide pair for image-guided radionuclide therapy for cancer is ²⁰³Pb/²¹²Pb, where gamma-emitting radionuclide ²⁰³Pb can be used for single photon emission computed tomography (SPECT) and ²¹²Pb represents a potentially ideal radionuclide for specific classes of radiopharmaceuticals for delivering alpha particles to cancer cells.

Generator-produced ²¹²Pb (t_1/2 = 10.6 h; 100% β decay to alpha emitters ²¹²Bi and ²¹²Po) is recognized as a promising radionuclide for receptor-targeted α-RT (Fig. 28.1) [10, 33, 37,38,39]. However, α-decay cannot be used directly for molecular imaging. Therefore, a surrogate imaging radionuclide is required to perform complementary diagnostic imaging. The primary rationale for use of an elementally matched pair of radionuclides for this application is highlighted by recent comparisons of tumor and normal organ uptake of ⁶⁸Ga- and ⁹⁰Y-labeled small peptides in which measurable differences in pharmacokinetics were observed in in vivo biodistribution studies in mice [40]. Thus, for ²¹²Pb α-RT, the cyclotron-produced gamma(γ)-emitting radionuclide ²⁰³Pb can be used as an elementally identical imaging surrogate [10, 33, 37, 38]. To wit, it can be expected that isotopes of the same element will have identical chemical and biochemical behaviors, adding confidence to predictions of ²¹²Pb α-RT outcomes using ²⁰³Pb SPECT and SPECT/CT. However, in this context, uncertainties arise in these assumptions given the relationship between α-RT and imaging that must be considered in evaluation of ²⁰³Pb SPECT. Stability of the ²¹²Pb-ligand complex and the potential for biological redistribution of daughter progeny gives rise to a potentially significant uncertainty in the use of ²⁰³Pb SPECT for ²¹²Pb α-RT dosimetry. Here, we discuss factors relating to the introduction of these isotopes for image-guided radionuclide therapy for cancer.

Fig. 28.1

Decay series for the production and use of ²¹²Pb for image-guided radionuclide therapy for cancer

28.2 ²⁰³Pb SPECT/CT Imaging in Advance of ²¹²Pb α-RT

One of the potential advantages of ²¹²Pb-based alpha-emitting radiopharmaceuticals is the potential for using quantitative imaging by single photon emission computed tomography (SPECT) and computed tomography (CT) with ²⁰³Pb-based surrogates to perform patient-specific dosimetry prior to therapy. It is anticipated that this could be most beneficial early in development process of new radiopharmaceuticals, such as in the preclinical setting and in early clinical trials (e.g., where organ doses can be monitored to develop understanding of the potential for other organ toxicities). SPECT imaging utilizes parallel hole collimation or pinhole collimation to obtain two-dimensional projections of the activity distribution within a patient. From these projected images and density information from CT imaging, a quantitative three-dimensional distribution of the radioactivity can be generated. Each quantitative image describes the activity distribution at a particular point in time, so multiple imaging time points are typically required to characterize the spatiotemporal characteristics of the radiopharmaceutical following administration.

Patient-specific pharmacokinetic data obtained through quantitation can be used to develop understanding by performing retrospective calculations regarding the absorbed dose to tumors and normal tissues. Dosimetry can enable patient-specific treatment optimization by delivering the maximum possible radiation dose to tumors without exceeding normal organ dose limits, which has the potential to improve the overall safety and efficacy of radiopharmaceutical therapies [41, 42]. Without dosimetric guidance, dose to normal tissues has been reported to vary by up to a factor of 5 per administered activity [43]. Based on these findings and observations, a dosimetrically informed therapy planning paradigm may be advantageous vs a “fixed activity” treatment strategy.

In some cases, dosimetry can be performed by administering a relatively low amount of the therapeutic radiopharmaceutical prior to treatment. This has been extensively demonstrated in the setting of radioiodine treatment for thyroid cancer—and this approach is now also the standard practice prior to treatment with iodine-131 labeled metaiodoguanidine (¹³¹I-MIBG) for pheochromocytoma or paraganglioma [44, 45]. In other cases, a surrogate radiopharmaceutical is used to predict the biodistribution and pharmacokinetics of the therapeutic radiopharmaceutical. Use of ^99mTc-labeled macroaggregated albumin (^99mTc-MAA) to predict the distribution of ⁹⁰Y-microspheres is a prominent example, although it is known that the biodistribution of these two radiopharmaceuticals is somewhat dissimilar [46]. In the case of fractionated delivery of radiopharmaceutical therapy (routinely performed with ¹⁷⁷Lu-DOTATATE; 7.4 GBq per fraction for a total of 29.6 GBq), dosimetry can be performed directly following each treatment via SPECT CT imaging and medical physics analysis. This retrospective dosimetry approach allows for modification of subsequent administrations to target a specific cumulative dose. In all cases, it is important to note that the dose per administered activity can vary within a given subject based on differing metabolic states at the time of administration, or due to radiation-induced changes in patient physiology over the course of treatment. These observations suggest that a pretreatment dosimetric assessment repeated prior to each therapeutic fraction may be optimal.

In the case of image-guided radionuclide therapy using ²¹²Pb, an elementally identical gamma-emitting radionuclide (i.e., ²⁰³Pb) can be employed for patient dosimetry. Practical considerations for quantitative SPECT imaging with ²⁰³Pb are likely to parallel methods that have been developed for SPECT/CT following therapy with ¹⁷⁷Lu-based agents [47, 48]. Specific considerations include the collimator selection, the number of SPECT projections, the camera orbit trajectory, scatter window selection, dead time correction, attenuation correction, collimator detector response modeling, number of iterative updates during the reconstruction, and partial volume correction. One notable difference compared with methods developed for ¹⁷⁷Lu is that a high-energy collimator will likely be needed for use with ²⁰³Pb due to the higher energy (279 keV vs 208 keV). In addition, it is possible that the size and makeup of scintillator crystals of the SPECT system may impact the efficiency of detection due the higher energy of the ²⁰³Pb emission. A representative small animal image of ²⁰³Pb-DOTATOC demonstrates the potential for ²⁰³Pb-based SPECT CT (Fig. 28.2).

Fig. 28.2

²⁰³Pb-DOTATOC SPECT/CT in healthy ICR mice (HE pinhole collimator; Siemens Inveon)

28.3 Prediction of ²¹²Pb Dosimetry Based on ²⁰³Pb Imaging

An elementally identical imaging surrogate (i.e., ²⁰³Pb as a surrogate for ²¹²Pb) is potentially advantageous because the chemistry (and biochemistry) of nuclides of the same element are the same. However, potential uncertainties arise in the approach due to the subsequent nuclear transformations that generate ²¹²Pb radionuclide progeny in the ²¹²Pb decay series (Fig. 28.1). This scenario is common to other radiometals currently employed and under investigation for receptor-targeted alpha-particle therapy, including ²²⁵Ac and ²²⁷Th. A key distinguishing physical characteristic of these two radionuclides (i.e., difference from ²¹²Pb) is that their primary decay is directly by alpha-particle emission, while the ²¹²Pb nuclear transformation to radionuclide progeny ²¹²Bi occurs by beta-particle emission. This is important because the alpha-particle energy of the ²²⁵Ac and ²²⁷Th decay is undoubtedly sufficient to break the chemical bonds of the chelator-radiometal coupling of the daughter nuclei (i.e., ²²⁵Ac–²²¹Fr and ²²⁷Th–²²³Ra). This phenomenon creates an immediate separation of the entire decay-series progeny from the site of the parent radionuclide (i.e., the chelator-ligand complex) that cannot be overcome. This phenomenon is potentially lessened in the case of ²¹²Pb, because the recoil energy imparted to the transforming nucleus is relatively small compared to alpha-particle-induced recoil energy [49]. Nonetheless, a critical parameter in assessing the uncertainty of modeling ²¹²Pb-based radionuclide therapy using ²⁰³Pb imaging surrogates is an understanding of the potential for migration of ²¹²Pb decay series radionuclides from the site of ²¹²Pb decay. Within this context, the half-lives of ²¹²Po (t_1/2 300 psec.) and ²⁰⁸Tl (t_1/2 3 min.) are sufficiently short such that understanding of the potential for ²¹²Bi to migrate from the site of ²¹²Pb decay is considered sufficient information to inform the uncertainty in using ²⁰³Pb as a model for predicting ²¹²Pb alpha-particle dosimetry.

Within this context, one key parameter that remains in question is the kinetic stability of ²¹²Bi generated by the decay of ²¹²Pb within the chelator moieties employed for binding the radionuclides to the receptor-targeted ligands employed for delivering radiation to the tumor microenvironment. Extensive studies of the potential for differences in biodistribution of ²¹²Pb and ²¹²Bi (using the various chelator-ligand combinations for ²¹²Pb chelation and tumor delivery) have not been reported. The limited studies that have been reported explored the kinetic stability of ²¹²Bi generated by ²¹²Pb in the in vitro or in vivo setting. One study examined ²⁰³Pb(II) and ²⁰⁶Bi(III) chelate stability using a tetracarboxy chemical form of the chelator DOTA with no peptide ligand attached. This investigation showed that the chemical exchange of both Pb and Bi complexes with DOTA occur rather slowly in aqueous solution at physiologically relevant pH (pH 4–10). However, this study revealed that in the case of this tetra-carboxy DOTA derivative (free chelator without a peptide attached), approximately 30% of ²¹²Pb beta decays can result in the release of daughter ²¹²Bi from the chelator coupling, representing a potentially significant uncertainty that could be introduced with respect to pretreatment dosimetry that employs a ²⁰³Pb-labeled agent as a surrogate for ²¹²Pb [50,51,52]. Further studies are required to develop a more detailed understanding of the chelator coupling with ²¹²Bi created by ²¹²Pb decay and it is anticipated that the stability of the ²¹²Pb-chelator complexes will be chelator-specific. An examination of this type for specific chelator-modified peptide conjugates will be needed to develop a more empirical understanding for individual radiopharmaceuticals.

A majority of energy released during the ²¹²Pb decay chain arises via alpha emissions of ²¹²Bi (t_1/2 61 min) and that of the short-lived ²¹²Bi daughter, ²¹²Po (t_1/2: 0.3 μs). The radioactive half-life of ²¹²Bi may be long enough to redistribute within the body according to its own pharmacokinetics depending on the tissue type—and depending on whether the bismuth atom is released within a cell or in the extracellular environment. Reported pharmacokinetic data provides some information regarding the biological fate of radioactive bismuth ions in humans. As is the case for numerous other elements, the International Commission on Radiological Protection (ICRP) has created a pharmacokinetic model for bismuth based on human data from accidental or intentional exposures to radiobismuth [53]. This model is reproduced in Fig. 28.3 with transfer rate constants specified in Table 28.1. Most forms of bismuth (those prone to ionic dissociation) clear rapidly from the blood with significant accumulation in the kidneys and liver. With consideration given toward bismuth released within tumors, activity that is released into extracellular fluid (represented by the rapid turnover compartment in Fig. 28.3) will clear quickly (k = 66 d⁻¹) into the plasma, and subsequently into the liver (k = 30 d⁻¹) and renal structures (k = 36 d⁻¹). For this reason, dose to tumors may be overestimated by ²⁰³Pb imaging, and dose to other normal tissues may be underestimated. The potential for new chelator technologies to be introduced that protect the integrity of the daughter-chelator coupling post-decay of ²¹²Pb has the potential to mitigate this uncertainty and more research in this area is needed.

Fig. 28.3

Pharmacokinetic model of bismuth in the body, reproduced with permission from ICRP 137 [53]

Table 28.1 Transfer coefficients for bismuth pharmacokinetic model. Reproduced with permission from ICRP 137 [53]

In order to estimate the degree of systematic error from ²⁰³Pb-based predictions of ²¹²Pb biodistribution, the pharmacokinetic model described by Fig. 28.3 and Table 28.1 was implemented in a MatLab script. Bismuth ions were assumed to be generated in various sub-compartments (soft tissue, kidneys, liver, blood), and the fate of ions at the time of radioactive decay were tallied. Organ-specific time activity curves for the case where ions are released within the tumor extracellular compartment is shown in Fig. 28.4. Transfer of activity from the extracellular soft tissue space to the plasma occurs rapidly, followed by localization in the kidneys and liver within approximately 1 h. Integration of these time activity curves reveals that 50% of released bismuth ions would decay prior to leaving the soft tissue compartment, while the blood, liver, and kidneys receive 8%, 13%, and 15% of decays, respectively. In the case of bismuth released while the radiopharmaceutical is in the blood, only 10% of ²¹²Bi decays occur prior to clearance from the blood, while the liver, kidneys, and other soft tissues receive 17%, 19% and 37% of decays, respectively. Other combinations of “source” and “target” organs are listed in Table 28.2. Data generated from this pharmacokinetic modeling, combined with information regarding the fraction of ²¹²Bi daughters that are released from a particular chelator, allows for improved accuracy when extrapolating from ²⁰³Pb imaging.

Fig. 28.4

Time-activity curves following a release of free ²¹²Bi within the extracellular soft tissue space, such as what would be observed with non-internalized tumor uptake of a radiopharmaceutical

Table 28.2 Fraction of energy deposited in normal structures as a function of where ²¹²Bi is released. When ²¹²Bi is released in liver, kidneys, and intracellular soft tissue compartments (ST1, ST2), the biological redistribution of ²¹²Bi is negligible prior to decay. Consideration should be given to ²¹²Bi that is released in the extracellular soft-tissue compartment (ST0) as well as ²¹²Bi that is released in the blood plasma

If the fraction of bismuth ions released (f) is known, a generalized formalism can be devised to account for this redistribution effect by correcting ²⁰³Pb-derived time-integrated activities:

$$ \overset{ }{\mathrm{A}}\left({r}_i\right)=\left(1-f\right){\overset{ }{\mathrm{A}}}_{\mathrm{raw}}\left({r}_i\right)+f\sum \limits_{r_S}{\overset{ }{\mathrm{A}}}_{\mathrm{raw}}\left({r}_s\right)\;\psi \left({r}_i\leftarrow {r}_s\right) $$

In this formalism, ${\overset{ }{\mathrm{A}}}_{\mathrm{raw}}\left({r}_i\right)$ is the time-integrated activity for a given organ (r_i) without correcting for the redistribution effect, and $\overset{}{\mathrm{A}}\left({r}_i\right)$ is the corrected time-integrated activity for that organ. The fractional transfer of time-integrated activity from a given source organ (r_s) to the organ of interest (r_i) is represented by the product of f and ψ(r_i ← r_s) (Table 28.2), where ψ(r_i ← r_s) describes the probability of a given bismuth ion decaying in r_i when it was released from r_s. More sophisticated microdosimetric correction factors could potentially be developed if sub-organ pharmacokinetic models were utilized in a similar fashion to what has been described here. Once $\overset{}{\mathrm{A}}\left({r}_i\right)$ is determined for each tissue type, patient-specific dosimetry can proceed according to MIRD methods [54]. It is worth noting that within this formalism the release of activity within tumors should be treated separately from other soft tissues in the body, and that the additive correction—$\sum \limits_{r_S}{\overset{ }{\mathrm{A}}}_{\mathrm{raw}}\left({r}_s\right)\;\psi \left({r}_i\leftarrow {r}_s\right)$—should distribute time-integrated activity uniformly over all soft tissues, including tumors. Also notable is that this methodology could potentially be extended to ²²⁵Ac-based radiopharmaceuticals to estimate dose due to redistribution of daughters (²¹¹Fr, ²⁰⁷At, ²¹³Bi). In this way, a more detailed understanding of the potential off-target dosimetry can be obtained that can be used for more precise image-guided radionuclide therapy treatment planning.

28.4 Summary and Future Directions

Alpha-emitting radiopharmaceutical therapy shows promise for improving the therapeutic efficacy of existing and future targeting ligands by limiting off-target irradiation and by preempting many cell survival mechanisms (e.g., DNA repair, hypoxia). In addition to the potency of alpha radiation, dosimetry-guided therapies have been shown to be potentially safer and more effective than RPT administration under a fixed-activity paradigm. Among the candidates of alpha-emitting radioisotopes, ²¹²Pb shows promise for use under a theranostic paradigm, whereby ²⁰³Pb can be used for dosimetry and treatment planning.

In this chapter, we have presented an approach for accurately estimating the dosimetry of ²¹²Pb-based radiopharmaceuticals using ²⁰³Pb as a surrogate. Moving forward, it will be necessary to establish more precisely the uncertainties arising under this paradigm and to experimentally validate the model used to predict the redistribution of ²¹²Bi following the decay of ²¹²Pb. One way to approach this problem would be to perform comparative biodistribution studies, whereby a ²¹²Pb-bearing compound is administered to a mouse that is sacrificed at specified time points postinjection. Tissue samples would be acquired promptly, and quantitative gamma spectrometry could be performed to differentiate between the location of ²¹²Pb and ²¹²Bi (unsupported vs supported) (and progeny) in the body. The gamma emission energies of ²¹²Pb (E_ꝩ: 239 keV, 43.6%), ²¹²Bi (E_ꝩ: 727 keV, 6.7%), ²¹²Po (E_ꝩ: 570 keV, 2%), and ²⁰⁸Tl (E_ꝩ: 583 keV, 85%) are sufficiently distinct and abundant as to enable quantitative energy-peak spectroscopic measurements by sodium iodide solid scintillation detectors and by high-purity germanium detector. The challenging aspects of this experiment are (1) minimizing the time between animal sacrifice and spectroscopic measurements, and (2) appropriately modeling the decay of unsupported ²¹²Bi, and ingrowth of ²¹²Bi in tissues where ²¹²Pb is present. Careful uncertainty analysis will be required when considering the alteration of redistribution dosimetric modeling parameters.

Experiments are also needed to elucidate the toxicity of bioconjugated and free ²¹²Bi in tissues of interest, normalized to tissue mean dose. It has been shown that microdosimetric factors, such as what organ sub-structure the radioactivity resides in, can substantially alter organ-level toxicity [55]. Therefore, on a per-radiopharmaceutical basis, it may be important to consider the differential distribution of free ²¹²Bi ions and radiopharmaceutical-bound activity.

Research productivity in the field of αRPT is growing rapidly, as evidenced by the ~ten-fold increase in publications per year over the last 30 years (Fig. 28.5). Human therapy studies have shown promising preliminary results, and the field of αRPT is benefiting from the development of new and innovative beta-emitting radiopharmaceutical therapies. We foresee continued growth in research productivity in this area, as well as improved patient care standards as technologies progress. Lead-212 is likely to play a key role in the progress toward personalized αRPT.

Fig. 28.5

PubMed results by year for the search terms ²¹²Pb, ²²⁵Ac, ²¹²Bi, ²¹³Bi, ²¹¹At, and ²²⁷Th