Dual Radionuclide Theranostic Pretargeting

Abstract

Recent years have played witness to the advent of nuclear theranostics: the synergistic use of ‘matched pair’ radiopharmaceuticals for diagnostic imaging and targeted radiotherapy. In this investigation, we report the extension of this concept to in vivo pretargeting based on the rapid and bioorthogonal inverse electron demand Diels-Alder reaction between tetrazine (Tz) and trans-cyclooctene (TCO). We demonstrate that a single injection of a TCO-modified immunoconjugate can be used as a platform for pretargeted PET imaging and radiotherapy via the sequential administration of a pair of Tz-bearing radioligands labeled with the positron-emitting radiometal copper-64 (t_1/2 ~ 12.7 h) and the beta-emitting radiometal lutetium-177 (t_1/2 ~ 6.7 d). More specifically, a mouse model of human colorectal carcinoma received a dose of the A33 antigen-targeting immunoconjugate huA33-TCO, followed 24 and 48 h later by injections of [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz, respectively. This approach produces high activity concentrations of both radioligands in tumor tissue (16.4 ± 2.7 %ID/g for [⁶⁴Cu]Cu-SarAr-Tz at 48 h post-injection and 18.1 ± 2.1 %ID/g for [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz at 120 h post-injection) as well as promising tumor-to-healthy organ activity concentration ratios. Ultimately, we believe that this work could not only have important implications in nuclear theranostics — most excitingly with isotopologue-based radioligand pairs such as [⁶⁴Cu]Cu-SarAr-Tz and [⁶⁷Cu]Cu-SarAr-Tz — but also in the delivery of fractionated doses during pretargeted radioimmunotherapy.

Graphical Abstract

Introduction

Over the past decade, the concept of ‘nuclear theranostics’ has emerged as a critical clinical tool.^1,2 Broadly defined, nuclear theranostics describes the synergistic use of ‘matched pair’ radiopharmaceuticals for diagnostic imaging and targeted endoradiotherapy. Indeed, the use of companion nuclear imaging agents based on the same platform as targeted radiotherapeutics facilitates the selection of patients most likely to respond to treatment, enables the optimization of the dose of the radiotherapeutic, and allows for the longitudinal monitoring of patients during treatment. As a result, the advent of theranostics clearly offers a path to more personalized and precise care. This paradigm has been leveraged with great success in the clinic. For example, [⁶⁸Ga]Ga-DOTA-TATE has been effectively employed for the diagnostic PET imaging of patients with neuroendocrine tumors receiving peptide receptor radiotherapy with [¹⁷⁷Lu]Lu-DOTA-TATE.^{3, 4} Likewise, [⁶⁸Ga]Ga-PSMA-617 has shown promise for the PET imaging of prostate cancer patients undergoing treatment with [¹⁷⁷Lu]Lu-PSMA-617 or [²²⁵Ac]Ac-PSMA-617.^5–8 This approach is not limited to ⁶⁸Ga-labeled PET imaging agents, of course: several ⁸⁹Zr-labeled antibodies have also been used as companion imaging agents for radioimmunotherapeutics labeled with nuclides including yttrium-90, lutetium-177, and actinium-225.

In the investigation at hand, our goal was to develop a theranostic approach to in vivo pretargeting. While radioimmunotherapy has long shown promise as a treatment strategy, the prolonged circulation times of radioimmunoconjugates — and, consequently, their high radiation doses to healthy tissues — has hampered their clinical utility. One way to circumvent this limitation is in vivo pretargeting. Pretargeting, in essence, relies upon performing radiosynthesis within the body. The antibody is injected prior to the radionuclide, and the two components are designed to combine at the tumor itself. In this way, this approach effectively delivers radioactivity to target tissues while shortening the circulation time of the radioactivity and facilitating the use of short-lived nuclides (e.g. copper-64, fluorine-18, and gallium-68) that are normally incompatible with vectors with long pharmacokinetic half-lives.

Over the last half decade — in conjunction with our collaborators and in parallel with a handful of other laboratories — we have developed an approach to in vivo pretargeting based on the rapid and bioorthogonal inverse electron demand Diels-Alder (IEDDA) ligation between a trans-cyclooctene (TCO)-modified antibody and a tetrazine (Tz)-bearing radioligand (Figure 1A).^9–14 This methodology has five elemental steps: (i) the administration of the TCO-bearing immunoconjugate; (ii) an interval period during which the mAb-TCO slowly accumulates in the target tissue and clears from the blood; (iii) the administration of the small molecule Tz radioligand; (iv) the in vivo click ligation between the two components; and (v) the rapid clearance of any excess radioligand. In murine models of pancreatic, colorectal, and breast cancer, we have demonstrated that this approach can be effectively harnessed for PET imaging with ⁶⁴Cu (t_1/2 ~ 12.7 h), ¹⁸F (t_1/2 ~ 110 min), and ⁶⁸Ga (t_1/2 ~ 68 min).^15–20 Just as — and perhaps even more — importantly, we have also used mouse models of pancreatic and colorectal cancer to illustrate the efficacy of IEDDA-based pretargeted radioimmunotherapy (PRIT) with both ¹⁷⁷Lu (t_1/2 ~ 6.7 d) and ²²⁵Ac (t_1/2 ~ 10.0 d).^21–23 In both applications, this in vivo pretargeting strategy produces high activity concentrations in tumor tissue. Furthermore, the radiation dose rates to healthy tissues created by this methodology — particularly in the context of pretargeted PET with ⁶⁸Ga or ¹⁸F — are far lower than those created by traditional radioimmunoconjugates labeled with long-lived radionuclides such as zirconium-89 (t_1/2 ~ 3.3 d) or iodine-124 (t_1/2 ~ 4.2 d). As IEDDA-based PRIT continues its steady progression toward the clinic, it is likely that it will ultimately benefit from a theranostic approach, just as endoradiotherapy with [¹⁷⁷Lu]Lu-DOTA-TATE, [¹⁷⁷Lu]Lu-PSMA-617, and radioimmunoconjugates have

Figure 1.

(A) The inverse electron demand Diels-Alder (IEDDA) reaction between trans-cyclooctene and tetrazine; (B) a schematic of huA33-TCO and the structures of [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz.

Results

Our theranostic pretargeting strategy is predicated on following the injection of the mAb-TCO with the sequential administration of not one but two radioligands: one for PET and one for endoradiotherapy (Figure 2A). Critically, the PET imaging of the former could provide information on the biodistribution, and thus dosimetry, of the latter. The feasibility of this approach, however, requires that the trans-cyclooctene moieties of the TCO-bearing immunoconjugate are not fully saturated after the initial injection of a Tz-based radioligand and thus can subsequently react with a second radioligand. Fortunately, this seems to be the case. Based on previous biodistribution experiments and several inevitable (yet, we argue, reasonable) assumptions, we estimate that the injection of the initial radioligand occupies about ~15% of the antibody-bound TCO moieties and that 0.32 nmol of the TCO groups have isomerized to cis-cyclooctene (CCO) 48 h after the administration of huA33-TCO. Taken together, these two processes leave ~64% of the injected TCOs free to react with a second radioligand (see Supporting Information for calculation). The two radioligands we have used in this investigation — [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz — have already proven highly effective for pretargeted PET imaging and radioimmunotherapy, respectively (Figure 1B).^{13, 24} Admittedly, the use of two structurally different radioligands falls somewhat short of the theranostic matched pair ideal represented by [⁶⁸Ga]Ga-PSMA-617 and [¹⁷⁷Lu]Lu-PSMA-617 or, even more stringently, isotopologue-based radiopharmaceuticals. However, [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz boast very similar pharmacokinetic profiles, serum half-lives, and reaction kinetics with trans-cyclooctene, making the pair a suitable choice for this proof-of-concept investigation (Supporting Information Tables S1 and S2).^{13, 21, 22, 24} Finally, our model system is rounded out by the huA33 antibody 3/4 an IgG that targets a transmembrane glycoprotein (the A33 antigen) expressed by >95% of colorectal carcinomas 3/4 and the A33 antigen-expressing SW1222 human colorectal cancer cell line.^25–27

Figure 2.

(A) A schematic of dual radionuclide theranostic pretargeting based on the IEDDA reaction; (B) a timeline of the sequence of events.

For the investigation at hand, huA33-TCO, [⁶⁴Cu]Cu-SarAr-Tz, and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz were synthesized, purified, and characterized according to previously published procedures.^{13, 15, 17, 21} Female athymic nude mice (5–7 weeks old) were inoculated with 5×10⁶ SW1222 cells in the right shoulder, and the subcutaneous xenografts were allowed to grow for two weeks. Once the tumors had grown to an appropriate size (~100 mm³), huA33-TCO (100 μg, 0.67 nmol, 2.4 TCO/mAb) was administered intravenously (i.v.) via the tail vein. The first radioligand — [⁶⁴Cu]Cu-SarAr-Tz (10.4–11.3 MBq, 0.64–0.70 nmol) — was then administered 24 h later, followed 24 h thereafter by the second radioligand, [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz (5.9–6.4 MBq, 0.69–0.74 nmol) (Figure 2B). PET imaging was then carried out 6, 24, and 48 h after the administration of [⁶⁴Cu]Cu-SarAr-Tz. In a parallel biodistribution experiment, mice were sacrificed 4, 24, 48, and 120 h after the injection of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz, and their organs were collected, weighed, and measured for both ⁶⁴Cu and ¹⁷⁷Lu using a gamma counter. The i.v. injections of the two radioligands were performed using opposite tail veins. This 24 h interval between the administration of the two radiotracers was chosen for three reasons: (i) it provides more than enough time for any unreacted [⁶⁴Cu]Cu-SarAr-Tz to clear from the blood prior to the administration of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz (Supporting Information Table S1); (ii) it facilitates the collection of several high quality PET scans before the injection of the ¹⁷⁷Lu-labeled radioligand; and (iii) it allows for a significant amount of time during which both types of radioactivity are present in the body, enabling comparisons between the biodistribution profiles of both radioligands.

The PET images and biodistribution results produced by [⁶⁴Cu]Cu-SarAr-Tz (Figure 3B and Table 1) closely mirror our previous work with this system (Supporting Information Tables S3 and S4).^{13, 16, 20} Tumoral uptake of the ⁶⁴Cu-labeled tetrazine is evident at early time points and increases to a maximum of 18.3 ± 3.1 %ID/g at 72 h post-injection. Some activity is present in the blood at early time points (3.5 ± 0.9 %ID/g at 28 h p.i.), likely the result of click reactions between [⁶⁴Cu]Cu-SarAr-Tz and still circulating huA33-TCO. However, these activity concentrations decrease with time, ultimately yielding a tumor-to-blood activity concentration ratio of 8.2 ± 1.6 at 72 h post-injection (Supporting Information Table S5). Other healthy organs generally exhibit very low activity concentrations: to wit, the non-target tissues with the highest uptake — the liver and the kidneys — contain <2% ID/g at 72 h p.i. The more notable biodistribution results are provided by the second radioligand: [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz. Surprisingly, the prior administration of [⁶⁴Cu]Cu-SarAr-Tz does not appear to dramatically hinder the in vivo performance of the ¹⁷⁷Lu-labeled tetrazine (Table 1 and Supporting Information Table S4). The tumoral activity concentration is 4.9 ± 0.4 %ID/g at 4 h p.i., a value which increases to 13.1 ± 2.7 %ID/g and 18.1 ± 2.1 %ID/g at 48 and 120 h post-injection, respectively. As in the ⁶⁴Cu case, activity can be observed in the blood at early time points, but this decreases with time to ultimately produce a tumor-to-blood activity concentration ratio of 17.9 ± 5.9 at 120 h post-injection (Supporting Information Table S5). Furthermore, the activity concentrations in healthy organs are also all quite low, generally under 1–2 %ID/g. Analogous theranostic pretargeting experiments with a 6 h — rather than 24 h — interval between the administration of [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz produced strikingly similar results, illustrating that this phenomenon is not the product of a single set of reaction conditions (Supporting Information Figure S1 and Tables S6 and S7).

Figure 3.

(A) Biodistribution data for in vivo pretargeting (n = 5) using two different tetrazine radiotracers — [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz — in athymic nude mice bearing subcutaneous SW1222 xenografts. The mice were first administered huA33-TCO (100 μg, 0.67 nmol, 2.4 TCO/mAb) via tail vein, followed 24 hours later by the i.v. administration of [⁶⁴Cu]Cu-SarAr-Tz (10.4–11.3 MBq, 0.64–0.69 nmol) and 24 h thereafter by the i.v. injection of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz (5.9–6.4 MBq, 0.68–0.73 nmol). The time values represent the number of hours after the administration of each Tz radioligand; (B) PET images of the same mice at 6, 24, and 48 h after the injection of [⁶⁴Cu]Cu-SarAr-Tz (11.2 MBq, 0.68 nmol). Top row: Coronal planar images that intersect the center of the tumors. Bottom row: maximum intensity projections (MIP).

Table 1.

Biodistribution data for in vivo pretargeting (n = 5) using two different tetrazine radiotracers — [⁶⁴Cu]Cu-SarAr-Tz and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz — in athymic nude mice bearing subcutaneous SW1222 xenografts. The mice were first administered huA33-TCO (100 μg, 0.67 nmol, 2.4 TCO/mAb) via tail vein, followed 24 hours later by the i.v. administration of [⁶⁴Cu]Cu-SarAr-Tz (10.4–11.3 MBq, 0.64–0.69 nmol) and 24 h thereafter by the i.v. injection of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz (5.9–6.4 MBq, 0.69–0.74 nmol). The time values represent the number of hours following the administration of each Tz radiotracer (i.e. the 28 h column under [⁶⁴Cu]Cu-SarAr-Tz denotes 28 h after the injection of [⁶⁴Cu]Cu-SarAr-Tz, whereas the 4 h column under [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz denotes 4 h after the injection of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz). The values for the stomach and small intestines include contents.

	[64Cu]Cu-SarAr-Tz			[177Lu]Lu-DOTA-PEG7-Tz
	28 h	48 h	72 h	4 h	24 h	48 h	120 h
Blood	3.5 ± 0.9	3.0 ± 9	2.3 ± 0.3	4.7 ± 0.8	3.4 ± 1.0	2.4 ± 0.4	1.1 ± 0.4
Tumor	11.0 ± 1.6	16.4 ± 2.7	18.3 ± 3.2	4.9 ± 0.4	11.3 ± 1.6	13.1 ± 2.7	18.1 ± 2.1
Heart	0.9 ± 0.2	1.0 ± 0.1	0.8 ± 0.2	1.2 ± 0.2	0.9 ± 0.2	0.6 ± 0.2	0.2 ± 0.1
Lungs	1.9 ± 0.5	1.9 ± 0.3	1.5 ± 0.3	2.4 ± 0.4	1.9 ± 0.5	1.2 ± 0.4	0.7 ± 0.3
Liver	2.0 ± 0.3	2.3 ± 0.2	1.6 ± 0.2	1.45 ± 0.2	1.2 ± 0.1	0.8 ± 0.2	0.5 ± 0.1
Spleen	1.4 ± 0.3	1.8 ± 0.6	1.1 ± 0.3	1.0 ± 0.2	1.0 ± 0.3	0.5 ± 0.1	0.4 ± 0.1
Stomach	0.3 ± 0.1	0.3 ± 0.1	0.2 ± 0.1	0.3 ± 0.1	0.3 ± 0.1	0.2 ± 0.1	0.2 ± 0.1
S. Intestine	0.4 ± 0.1	0.5 ± 0.2	0.4 ± 0.1	0.4 ± 0.1	0.3 ± 0.1	0.3 ± 0.1	0.2 ± 0.03
L. Intestine	0.3 ± 0.04	0.5 ± 0.1	0.3 ± 0.1	0.7 ± 0.2	0.4 ± 0.1	0.2 ± 0.1	0.1 ± 0.03
Kidneys	1.9 ± 0.4	2.3 ± 0.3	1.8 ± 0.2	1.6 ± 0.3	1.5 ± 0.3	0.9 ± 0.2	0.6 ± 0.2
Muscle	0.4 ± 0.2	0.5 ± 0.1	0.4 ± 0.1	0.4 ± 0.1	0.4 ± 0.1	0.2 ± 0.1	0.1 ± 0.03
Bone	0.5 ± 0.1	0.5 ± 0.1	0.4 ± 0.1	0.5 ± 0.1	0.4 ± 0.1	0.2 ± 0.04	0.2 ± 0.1

Discussion

Taken together, these in vivo data prompt two important observations. First, the biodistribution profile of the pretargeted [⁶⁴Cu]Cu-SarAr-Tz is very similar to that of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz, suggesting that the former could be used as a companion imaging agent for the latter. Indeed, at 48 h post-injection, for example, the correlation between the uptake of the two radioligands in different tissues is remarkably high. Second, as we have noted, the previous administration of [⁶⁴Cu]Cu-SarAr-Tz does not appear to have significant adverse effects on the biodistribution of [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz. In an earlier study by Membreno et al.²¹ in which [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz was injected 48 h after the administration of huA33 — but without the intervening injection of [⁶⁴Cu]Cu-SarAr-Tz — the tumoral activity concentrations at 48 and 120 h post-injection were 13.0 ± 2.8 and 11.0 ± 2.4 %ID/g, similar to the 13.1 ± 2.7 and 18.1 ± 2.1 %ID/g value obtained in this experiment (Supporting Information Table S3). Critically, the activity concentrations in other non-target organs were similar across the two investigations as well.

In order to further explore the potential of this approach for PRIT, dosimetry calculations were performed based on the [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz biodistribution data (Table 2). For most tissues, the time-activity curves could be best fit using one- or two-phase exponential decay models. Yet for the tumor — which displayed increased uptake over the experiment — a trapezoidal model was used up until the last time point, after which it was assumed that clearance was based entirely on radioactive decay. The dosimetry calculations revealed a total absorbed dose to the tumor of 310.2 cGy/MBq and a tumor-to-blood therapeutic index of 12.4. These results are consistent with (though slightly lower than) those obtained by Membreno, et al. in their proof-of-concept study of colorectal cancer PRIT with huA33-TCO and [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz (Supporting Information Table S8).²¹ This strongly suggests that this theranostic pretargeting approach could likewise be used for effective radiotherapy in mice bearing SW1222 xenografts.

Table 2.

Radiation dose rates and therapeutic indices for [¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz calculated based on the biodistribution data from the huA33-TCO/[⁶⁴Cu]Cu-SarAr-Tz/[¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz pretargeting experiment.

Organ	Dose (cGy/MBq)	Therapeutic Index
Tumor	310.2	_
Blood	25.0	12.4
Heart	5.7	54.9
Lungs	11.7	26.5
Liver	11.7	26.5
Spleen	7.2	43.0
Stomach	2.4	131.2
S. Intestine	3.5	87.7
L. Intestine	2.5	124.7
Kidneys	10.6	29.3
Muscle	2.3	134.1
Bone	4.8	65.3

Herein, we have described the development of a proof-of-concept system for theranostic pretargeting based on a single immunoconjugate (huA33-TCO) and a pair of radioligands labeled with diagnostic ([⁶⁴Cu]Cu-SarAr-Tz) and therapeutic ([¹⁷⁷Lu]Lu-DOTA-PEG₇-Tz) radionuclides. We believe that this system could be used to help select patients for PRIT and optimize the dose and timing of PRIT regimens. Yet our observations regarding the sequential administration of the tetrazine-based radioligands could be just as important as the theranostic implications of this work. Our data clearly indicate that the administration of a second Tz-bearing radioligand is feasible in the context of IEDDA-based pretargeting. And even more importantly, the in vivo performance of the second radioligand seems to be largely unaffected by the administration of the first. We believe that this phenomenon could have a significant impact not only on theranostic pretargeting but also on the administration of fractionated doses during PRIT. Moving forward, we plan to leverage this work in three ways: exploring theranostic pretargeting with longer injection intervals (e.g. 72 h between the administration of the two radioligands), developing new theranostic pretargeting systems based on chemically identical isotopologue-containing radioligands (e.g. ⁶⁷Cu/⁶⁴Cu), and examining the benefits of fractionated dosing in the context of PRIT with ⁶⁷Cu, ¹⁷⁷Lu-, and ²²⁵Ac-labeled radioligands.

Abbreviations

%ID/g

percent injected dose per gram

IEDDA

inverse electron demand Diels-Alder

iTLC

instant thin layer chromatography

PET

positron emission tomography

p.i.

post-injection

PRIT

pretargeted radioimmunotherapy

RIT

radioimmunotherapy

TCO

trans-cyclooctene

Tz

tetrazine