Conditioning with α-emitter based radioimmunotherapy in canine allogeneic hematopoietic cell transplantation

Abstract

With the introduction of nonmyeloablative conditioning, hematopoietic cell transplantation (HCT) has become a viable treatment option for patients who due to age or comorbidities are ineligible for high dose conditioning. However, relapse and toxicities are still major problems in HCT. Radioimmunotherapy (RIT)-based conditioning is a promising approach that has the ability to specifically target radiation to hematopoietic cells. The most widely investigated isotopes are the β-emitters, but because of long path lengths and low linear energy transfer, α-emitters which have more favorable physical characteristics, might prove to be a better alternative. In the current study we have investigated the efficacy and safety of α-emitter based RIT as the only form of conditioning in a preclinical model of canine allogeneic HCT.

Keywords: :

• allogeneic transplant
• chimerism
• nonmyeloablative
• radioimmunotherapy
• α-emitter

This article refers to:

A major dilemma in allogeneic hematopoietic cell transplantation (HCT) is that the benefit gained by intensifying the conditioning regimen to reduce relapse rates, is negated by increased regimen related toxicity and mortality. The introduction of nonmyeloablative conditioning, where external total body (TBI) irradiation is decreased from 12 to 2 Gy, has reduced regimen related toxicity significantly.^1,2 However, as toxicity and relapse are still major problems in HCT, there is a desire to further decrease the toxicity associated with external γ-radiation and possibly increase the radiation dose deposited in target tissues without increasing exposure of surrounding tissues. Radioimmunotherapy (RIT) is a treatment strategy that targets radiation precisely to the desired tissues. The efficacy and safety of RIT in delivering targeted radiation therapy has been demonstrated in several clinical settings. The most widely used isotopes for RIT are the β-emitters, and they have previously been used to augment conventional external γ-radiation in a variety of conditioning regimens in allogeneic HCT.^3-6 β-emitters, such as yttrium-90 (⁹⁰Y), rhenium-188 and iodine-131 (¹³¹I), are characterized by low linear energy transfer and long path lengths (800–5000 µm). Characteristics like these make β-emitters ideal for the treatment of bulky, poorly perfused tumors. However, when targeting hematopoietic cells or malignancies, where tissues are well perfused, β-emitters induce off-target toxicity as significant amounts of radiation are deposited in the surrounding healthy tissues.⁷

Alternative sources of radiation are the less explored α-emitters. They have short path lengths (40–90 µm) which only span a few cell diameters in vivo, hereby limiting radiation exposure of surrounding tissues. Furthermore, α-emitters also have up to 400 times higher linear energy transfer, which combined with a limited ability of cells to repair α-radiation induced damage, results in a superior relative biological effectiveness.^8-10

In an effort to limit radiation-induced toxicity and replace external γ-radiation, we have investigated the safety and efficacy of α-emitter-based RIT, as the sole form of conditioning prior to allogeneic HCT in a preclinical canine model. In our initial experiments we used the α-emitter bismuth-213 (²¹³Bi) conjugated to monoclonal antibodies (mAb) against CD45. CD45 is expressed in high numbers on virtually all non-malignant and malignant hematopoietic cells through all stages of maturation,^11,12 and was chosen as antigen because of its ability to effectively target T-cells and natural killer (NK) cells responsible for rejection^13-15 and possibly eradicate minimal residual disease if present. Although dog leukocyte antigen (DLA) HCT using 1.0–5.9 mCi/kg ²¹³Bi labeled anti-CD45 mAb was successful, with low rejection rates, minimal toxicity and durable long-term donor chimerism in the mononuclear cell (MNC) compartment,^16,17 several factors precluded further investigation and translation into clinical trials. Logistics with ²¹³Bi were cumbersome due to its short half-life (t_½) (46 min). To deliver the planned doses, three to eight injections over two successive days were necessary, to avoid significant decay of the isotope before treatment. Furthermore, the availability of the ²¹³Bi parent compound actinium-225 (²²⁵Ac) was limited and costs excessively high. As an alternative, we investigated the α-emitter astatine-211 (²¹¹At). ²¹¹At has a t_½ (7.2 h) that allows treatments to be administered as one injection and is locally available in sufficient amounts and at a cost that permits upscaling to clinical trials. Furthermore, as part of the transition from ²¹³Bi to ²¹¹At, murine studies demonstrated that the radiation dose from 50 µCi ²¹¹At deposited in the spleen was almost 2-fold higher than from 500 µCi ²¹³Bi without a concomitant increase in toxicity, suggesting a superior nonhematologic toxicity profile of ²¹¹At-antiCD45 mAb.¹⁸ Canine biodistribution was favorable,¹⁹ and in dose-finding studies, treatment with 0.1–0.62 mCi/kg without HCT rescue demonstrated dose-dependent myelosuppression followed by autologous recovery.²⁰ Encouraged by these results DLA-identical HCT studies with a single dose of 1.6–6.3 mCi/kg ²¹¹At loaded on to 0.5 mg/kg anti-CD45 mAb were conducted.²⁰ Depth of granulocyte, lymphocyte and platelet nadirs were dose-dependent, and more profound than with 2 Gy external γ-radiation²¹ or ²¹³Bi-anti-CD45 mAb.^16,17 All dogs had long-term engraftment. Apart from the dog treated with the lowest dose (1.6 mCi/kg), which was euthanized with stable 5% donor chimerism in MNC after 18 weeks, all dogs had durable donor chimerism in the MSC compartment (19–58%) at 40–53 weeks post-transplant. Toxicity was mild, and consisted mainly of transient elevations in liver function enzymes and subclinical hypothyroidism in two dogs, which could not be directly attributed to radiation exposure.

Our studies show that conditioning with ²¹¹At-anti-CD45-based RIT in HCT is safe and has the ability to induce durable high level donor chimerism. There was low toxicity and more myelo- and lymphosuppression than what is observed with 2 Gy TBI in the same model. The data suggest that conditioning with ²¹¹At-anti-CD45 based RIT could expand the nonmyeloablative approach to patients who cannot endure conventional high dose TBI, but where the nonmyeloablative approach with only 2 Gy TBI is not sufficient for disease control or engraftment. In haploidentical donor HCT, where significant conditioning is required to overcome the immunological barriers and establish durable chimerism,²² we have previously conducted a pilot study with ²¹³Bi-anti-CD45 based conditioning in the DLA-haploidentical HCT model.²³ Although initially successful, with four of six dogs acquiring sustained donor chimerism, survival was short due to toxicity. In this setting, an ²¹¹At-based RIT regimen may also prove superior due to a more profound myelosuppression and favorable nonhematologic toxicity profile.

Prior to the transition from ²¹³Bi to ²¹¹At, we also explored the possibility of further minimizing toxicity, by targeting radiation specifically toward TCRαβ⁺ T-cells, which have shown to be of particular importance for rejection.¹⁴ Of 11 dogs conditioned with 1.1–5.6 mCi/kg ²¹³Bi-anti-TCRαβ mAb prior to DLA-identical HCT, four conditioned with less than 1.5 mCi/kg rejected their graft and had autologous recovery. Compared with the studies using anti-CD45 mAb, a steeper decline in peripheral lymphocyte counts was observed, and in dogs that did not reject long-term donor chimerism was stable, but at a lower level (MNC, 3–40%). Although the low level chimerism observed following anti-TCRαβ-based RIT may not be sufficient to cure malignant disease, it may be suitable for the treatment of nonmalignant diseases such as hemoglobinopathies, immunodeficiencies, inborn errors of metabolism, as well as in gene transfer studies where full donor chimerism is not a requirement for cure but the long-term toxicities associated with HCT conditioning are a major concern.

α-emitter-anti-TCRαβ-RIT-based HCT may also play a role in solid organ transplantations, where toxicity due to lifelong immunosuppressive treatment is associated with significant morbidity. Canine studies have demonstrated that mixed chimerism achieved through a HCT from the organ donor induces tolerance toward the donor organ, enabling long-term survival of kidneys and vascularized composite skin and muscle grafts without pharmacological immunosuppressive treatment.^24,25

A novel avenue of research where ²¹¹At may prove useful is radiovirotherapy (RVT) (reviewed by Touchefeu).²⁶ The principle of RVT is to deliver radioisotopes into virally infected malignant cells. The most common approach is to use oncolytic viruses that have been genetically engineered to express the thyroidal sodium/iodide symporter (NIS). Although the main physiological function of NIS is to transport into iodine thyroid follicular cells, it has also been demonstrated to efficiently transport ²¹¹At, with an approximately 15-fold higher tumor absorbed dose compared with ¹³¹I in tumor cell line experiments.²⁷ Several in vitro and in vivo animal studies have already demonstrated efficient tumor killing by NIS ²¹¹At-based RVT which, when combined with the ²¹¹At short path length that limits damage to non-target tissues, may make ²¹¹At a highly attractive isotope for clinical trials.

In conclusion, our studies demonstrate that conditioning with α-emitter-based RIT in allogeneic HCT is feasible and safe. They provide a platform for future investigation of α-emitter-based conditioning, and a basis for the translation of α-emitter-based therapy in a wide variety of clinical trials.

Abbreviations:
225Ac	=	actinium-225
211At	=	astatine-211
213Bi	=	bismuth-213
DLA	=	dog leukocyte antigen
HCT	=	hematopoietic cell transplantation
131I	=	iodine-131
mAb	=	monoclonal antibodies
MNC	=	mononuclear cell
NK	=	natural killer
RIT	=	radioimmunotherapy
RVT	=	radiovirotherapy
NIS	=	sodium/iodide symporter
TBI	=	total body irradiation
90Y	=	yttrium-90

Acknowledgments

B.K. drafted and revised the manuscript, Y.C. revised the manuscript and B.M.S. conceived, drafted and revised the manuscript. The authors declare no conflicts of interest.