Precision pre-HSCT conditioning by targeting cMPL Free

With the success of hematopoietic stem cell gene editing and transplantation, precision hematopoietic stem cell (HSC) replacement therapy, without collateral damage to mature hematopoietic cells and other tissues, is in high demand. In this issue of Blood, Araki et al¹ demonstrate that this might be possible by targeting cMPL with an immunotoxin.

About 30 years ago, both thrombopoietin (TPO) and its receptor (cMPL) were identified.² The primary function was understood to be a master regulator of megakaryocyte proliferation and maturation and platelet production. cMPL loss of function leads to amegakaryocytic thrombocytopenia, and its constitutive activation leads to high platelet counts. cMPL is expressed on hematopoietic stem and progenitor cells, megakaryocytes, and platelets; ligand binding leads to heterodimerization, signaling, and receptor internalization. Although mainly produced by the liver, TPO is also made in the kidney and in bone marrow, with both homeostatic as well as regulated expression. Plasma TPO levels are usually inversely correlated with platelet counts, and platelets are postulated to be involved in regulating TPO levels by scavenging the cytokine. TPO receptor agonists (TPO-RAs) have been developed and shown to be effective, first receiving approval in immune thrombocytopenia and later also in specific settings of thrombopenia in liver disease.

However, there is another role of TPO and cMPL, now at center stage in the work published here by Araki and colleagues: TPO also acts as a stem cell factor. TPO and cMPL were shown to be essential for adult HSC maintenance.³ TPO-RA treatment leads to a quantitative increase of long-term repopulating HSCs, and, when combined with chemotherapy, allows for more efficient HSC niche clearance and the subsequent engraftment of transplanted HSC in mice.⁴ Importantly, clinically approved TPO-RA application in aplastic anemia enhances multilineage reconstitution, possibly via reexpansion of HSCs.⁵ 

Araki et al first explored cMPL expression in hematopoietic stem and progenitor cells (HSPCs). They show that (1) phenotypically and transcriptionally defined human and rhesus macaque long-term HSCs express relatively high levels of cMPL, (2) cMPL expression on human HSPCs correlates with primary and secondary repopulation ability in immunodeficient mice, and (3) cMPL expression in humans is limited to HSPCs and megakaryocytes with little expression in nonhematopoietic tissues. They then use this knowledge to generate a new pre-HSPC transplantation conditioning strategy by selectively targeting cMPL-expressing cells. To this end, they generated an anti-cMPL immunotoxin, consisting of a truncated diphtheria toxin (DT), fused to a bivalent anti-cMPL single-chain variable fragment (biscFV[cMPL]), derived from a minibody cMPL agonist, which is internalized upon receptor binding. The truncated DT lacks the regular surface-receptor binding moiety and thus should only be effective if internalized via cMPL in cells that actively synthesize protein. They demonstrate that the DT390-biscFV(cMPL) immunotoxin impairs proliferation of cMPL⁺ human cells in vitro, depletes human cMPL⁺ CD34⁺ cells in vivo in a xenograft mouse model, and safely eliminates cMPL⁺ CD34⁺ cells in rhesus macaques, allowing for low-level, multilineage, long-term engraftment of lentivirally transduced autologous HSPCs, without major disturbance of peripheral blood hematopoietic cell counts, causing only very minor, short-term thrombocytopenia.

Precision pre-HSPC conditioning (HSPC niche clearance), followed by subsequent HSC transplantation and engraftment, allowing seamless takeover of “new” hematopoiesis over “old” hematopoiesis without the morbidity consequences of temporary cytopenia, is a long-standing wish of physicians and patients. This is now becoming even more pressing with the availability of gene-corrected hematopoietic cells in settings of nonneoplastic diseases. Since the seminal study by Czechowicz et al,⁶ the receptor for stem cell factor (cKIT, CD117) has been in focus for selective immunotargeting to achieve this goal.^7-9 The current study by Araki et al introduces a second (possibly more selective) target, the receptor for TPO, cMPL, which now requires further investigation. Specifically, it will be important to clarify the following aspects in more depth: (1) is the truncated DT immunogenic? The presented work suggests that some neutralization by preexisting antibodies can happen (even against the modified DT); however, it might be overcome if antibody responses are minor. Also, the brief one-time clinical use for preconditioning purposes might have a low risk for generating or boosting preexisting antibodies sufficiently to result in neutralizing the response. (2) A related aspect could be a theoretical neutralization of DT, leading to prolonged half-life of the complex and extended agonist action on the cMPL receptor. This might induce temporary thrombocytosis, a phenomenon observed in this study in rhesus macaques, although it was likely due to a reactive increase in endogenous TPO in this setting, causing a rebound effect. (3) Will the on-target binding on platelets be a safety concern? The expected procoagulant, platelet-activating effect might be low, given the experience with clinically approved TPO-RAs. Also, the direct DT effect on platelet numbers might be minor, given the mode of DT action. (4) Are there, although unlikely, nonhematopoietic on-target and off-target effects, not evident in nonhuman primates in this and prior studies on the cMPL targeting minibody? (5) Can the approach be escalated via increasing or modifying dosing to achieve higher levels and, in the best case, full donor hematopoietic chimerism, which is desirable for many HSPC transplantation settings? Which combinatorial approaches might be necessary and feasible? (6) How does cMPL targeting compare to, for example, CD117 targeting regarding safety and efficacy? The authors reasonably suggest that the more restricted expression pattern in both hematopoiesis, as well as nonhematopoietic tissues, might favor cMPL. (7) Will cMPL targeting with other effectors, such as alternative antibody-drug conjugates, radionuclide antibody conjugates, bispecific effector cell-activating constructs, chimeric antigen receptor T cells, or even particles carrying, for example, genetic payloads, be possible? Here, platelets expressing cMPL might pose a greater challenge for translation. (8) Will DT390-biscFv(cMPL) be developable for myeloid malignancy treatment? cMPL expression and TPO use of some AML subtypes suggests that this could be a potential option.¹⁰ (9) And finally, will DT390-biscFv(cMPL) be developable for nonhematopoietic diseases, for example, for hematopoietic chimerism mediated tolerance in the context of solid organ transplantation?

In sum, the work by Araki et al is highly exciting news to the field of precision HSPC conditioning. It will likely boost further basic and translational studies for a broad spectrum of diseases with high unmet medical needs.

Conflict-of-interest disclosure: M.G.M. is an inventor on a patent application that describes an anti-CD117 × anti-CD3 bispecific T-cell engager. The commercialization rights and plans for this patent are overseen and regulated by the Technology Transfer Office of the University of Zürich and ETH Zürich, which granted an exclusive license to ATLyphe (www.atlyphe.com), of which M.G.M. is a cofounder.