T cells reloaded after allogeneic HCT

In this issue of Blood, Rimando et al suggest that T-cell–engaging immunotherapeutic approaches may be especially suited to confront or respond to or salvage relapse of acute myeloid leukemia (AML) after allogeneic hematopoietic cell transplantation (alloHCT).¹ Both the dual-affinity retargeting (DART) compound flotetuzumab (FLZ) and chimeric antigen-receptor T (CART) cells against AML-associated surface antigens such as CD123, CD33, and CD371 induced upregulation of major histocompatibility complex (MHC)-class II molecules by interferon-gamma (IFNγ) signaling. This mechanism was confirmed in cell lines and primary AML cells in vitro, in xenogeneic transplant models, and in samples of patients treated with FLZ for relapsed disease. By elucidating the mode of action, the authors provide further rational for use of T cell–engaging therapies and/or CART cells in patients with relapse after alloHCT.

AlloHCT is the most effective strategy for treating suitable patients with AML, owing to the potent graft-versus-leukemia (GVL) effects mediated by donor-derived, allo-reactive effector cells. In vitro cytotoxicity studies of matched donor cells with AML blasts and adoptive transfer studies in NSG mice support the hypothesis that MHC class II molecules are major targets of GVL.² Despite the strength of allogeneic GVL effects, a significant proportion of patients with AML relapse after alloHCT. Treatment of relapse after alloHCT is an immense challenge, and a cure is elusive in most cases.³ The exact reason for immune escape is difficult to dissect at an individual case level, but several patterns of immune evasion have been described. Peripheral tolerance-induction and expression of immune checkpoints on leukemic cells after alloHCT may contribute to the loss of antileukemic immunity resulting in frank relapse.⁴ Due to inherent genetic instability, AML progenitor cells, like many other cancer cells, can downregulate MHC molecules. The mode and pattern of reduced human leukocyte antigen expression are shaped mainly by the pressures induced by allo-reactive effector cells. Complete loss of mismatched human leukocyte antigen haplotypes on the genetic level has been described after haploidentical alloHCT.⁵ Conversely, posttranscriptional downregulation of MHC class II expression has been detected on AML blasts at relapse after allogeneic HCT. In that same study, the incubation of leukemic cells with IFNγ led to upregulation of MHC class II expression.⁶ 

This report opens up a new and exciting strategy for treatment of relapsed AML after alloHCT. The authors of this study clearly demonstrate that both the CD123-targeting DART FLZ and AML-targeting CART cells induce paracrine IFNγ and subsequent upregulation of MHC class II (see figure). This effect was observed not only in vitro but also in a xenotransplant model and in paired patient samples analyzed before and after treatment with FLZ.

This study provides a clear rationale to test T-cell–engaging therapies in patients with relapse after alloHCT. An interesting question to answer is whether the effects of these therapies are more durable in such patients, compared to their effects in patients who have not undergone a transplantation procedure.⁷ This benefit is suggested by prior reports that unspecific reactivation of donor T cells—for example, after adoptive transfer—has been shown to be of limited efficacy, but durable, in a small subset of patients.⁸ The limitation of adoptive transfer always has been the potential induction of graft-versus-host disease (GVHD). Such an on-target, off-tumor toxicity has to be kept in mind when T cells are activated or transferred after alloHCT and MHC expression is induced, but the risk of GVHD induction, it is hoped, will be lower when the awakening of alloreactivity is targeted to AML cells. With the importance of IFNγ signaling for the observed effects, one has to be cautious when patients have received or are still on JAK-2 inhibitors (eg, ruxolitinib). However, the additional systemic administration of IFNγ might even boost the effects of T-cell–targeting therapies. Fortunately, we will not have to wait too long to get first insights into the feasibility and efficacy of such treatment strategies, as the use of both IFNγ and FZL are currently being tested in patients with relapse after alloHCT (https://clinicaltrials.gov: NCT04628338 and NCT04582864). As the response to FLZ has been shown to be related to expression of IFNγ-related genes in the nontransplant setting, a point of interest is whether similar predictive biomarkers can be identified in the post-HCT setting.⁹ The current paper by Rimando et al provides a convincing rationale to pursue the clinical testing of T-cell–engaging therapies and CART approaches in patients with relapse of myeloid malignancies after alloHCT.

Conflict-of-interest disclosure: M.B. reports being a scientific advisory board member of and receiving speaker honoraria from Jazz Pharmaceuticals and MSD.