ALL is a challenging target for immunotherapy due to its poor antigen presentation, lack of costimulatory molecule expression, and resistance to cytotoxic killing by immune effector cells. We previously described a model in which we transfected B cell precursor ALL cell lines derived from leukemic Eμ-ret transgenic BALB/c mice with a lentiviral construct encoding green fluorescent protein (GFP) and firefly luciferase (luc) in tandem. The luciferase protein serves a dual purpose as a potent neoantigen capable of stimulating leukemia-specific immune responses and as a bioluminescent reporter for highly sensitive detection of minimal residual disease (MRD).

Intravenous injection of wild type Eμ-ret leukemia is consistently lethal in wild type BALB/c mice within 3-5 weeks. However, we previously showed that adoptive transfer of GFP/luc-expressing leukemia cells into BALB/c mice leads to detectable disease in bone marrow by bioluminescent imaging but not by peripheral blood flow cytometry that increases until day 7-10, followed by a spontaneous reduction of leukemia burden to levels undetectable by imaging. Mice lacking T cells, either constitutively or following antibody depletion, are unable to control the leukemia, as are mice transgenic for GFP/luc, suggesting immune tolerance may undermine host responses against ALL.

While most mice go into long-term remissions both by imaging and peripheral blood sampling, a subset have evidence of MRD detectable by bioluminescent imaging that waxes and wanes in luminosity but may not break through to a fulminant relapse. Of the 14% of mice who developed lethal leukemia before day 60 (range: day 27-51), 8/8 had GFP/luc-expressing leukemia. In the 31% of mice who develop lethal leukemia beyond 60 days (range: day 65-562), only 6% (1/18) had leukemia expressing GFP/luc (Fisher's exact test, p<0.0001), showing antigen loss is a mechanism for immune evasion in this model.

Mice who undergo NK and T cell depletion beginning at day 28 from adoptive transfer of leukemia are more likely to develop overt disease than immunocompetent mice or mice with single lineage depletions (logrank for survival, p=0.0005), suggesting both cell types cooperate to provide durable immune control of disease.

Finally, mice who clear their GFP/luc leukemia upfront are able to reject a day 28-rechallenge with wild type leukemia (median survival of leukemia-naïve mice 27 days versus 70% survival at day 324 in mice with previous GFP/luc-ALL exposure and clearance; p<0.0001). Mice receiving concomitant administration of GFP/luc and wild type leukemia are similarly able to clear their leukemia and reject engraftment upon rechallenge with wild type ALL. These studies suggest that protection is enhanced by epitope spreading after the initial immune response.

Overall, these results reveal that both the innate and adaptive arms of the immune system are required for prolonged control of overt leukemic cells in the setting of persistent MRD. The finding that mice who clear the neoantigen-bearing leukemia have immune memory for wild type leukemia suggests that, with adequate immunostimulatory signaling, additional tumor antigens may be recognized and serve as targets for immune control of ALL. We are currently evaluating the cytokine milieu induced by this neoantigen, as well as additional immunostimulatory approaches such as adding Toll-like receptor agonists to break immune tolerance, thereby driving down MRD and prolonging relapse-free survival.

Citation Format: Alix E. Seif, David M. Barrett, Stephan A. Grupp, Gregor SD Reid. Innate and adaptive immune control of neoantigen-bearing acute lymphoblastic leukemia (ALL) enhanced by epitope spread. [abstract]. In: Proceedings of the AACR Special Conference on Pediatric Cancer at the Crossroads: Translating Discovery into Improved Outcomes; Nov 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2013;74(20 Suppl):Abstract nr B20.