Recognition of Myeloma by MAGE-A3 Specific Cytotoxic T Lymphocytes Induced by Treatment with Azacitidine and MGCD0103

Demethylating agents and histone deacetylase inhibitors (HDACi) are epigenetic modulators that can induce re-expression of tumor suppressor and cell cycle proteins that have been silenced through aberrant hypermethylation associated with tumoral transformation. Azacitidine (Aza) is a cytosine analogue that primarily affects RNA during transcription. However, DNA is also a target for demethylation during replication thereby increasing gene re-expression. Treatment with HDACi, such as MGCD0103 (MGC), can synergize with demethylating agents to boost the epigenetic effects of either drug used alone. Our gene expression profiling data shows that the cancer-testis antigen MAGE-A3 is expressed in 31% of myeloma patients at diagnosis and the frequency of expression is increased at relapse to 49% (n=51 paired samples, p<0.001, unpublished data). However, expression of MAGE-A3 is often heterogeneous. We hypothesized that the combination of Aza and MGC could induce MAGE-A3 expression, thus facilitating killing of myeloma cells by MAGE-A3 specific CTLs isolated from a HLA-A68 positive patient post MAGE-A3 protein vaccination (J Immunother2007; 30:847). The MAGE-A3 negative myeloma cell line LP1 was first transfected with HLA-A68. MAGE-A3 protein production was optimized by dose finding and time course experiments using Aza alone or Aza and MGC sequentially. Induction of MAGE-A3 RNA expression was assessed by real time PCR and protein expression by Western blotting. 51Cr-release assays were used to measure killing of Aza/MGC treated cell lines by MAGE-A3 specific CTLs. MAGE-A3 RNA expression was detected in LP1-A68 treated with 500nM Aza for 3 days and expression was enhanced by sequential treatment with 1mM MGC for 1 day when compared to Aza treatment alone. However, protein expression was low. In an effort to optimize protein production, we increased the time of treatment with 500nM Aza to 5 days and with 500nM MGC to 2 days. After this sequential treatment, protein was clearly expressed (Figure 1) and LP1-A68 cells were killed by MAGE-A3 specific CTLs (specific lysis: 70% ± 9% at E:T ratio of 5:1), whilst untreated controls only showed background killing (specific lysis: 12% ± 5%) (Figure 2). Repeat experiments are in progress to verify these results. 500nM Aza in vitro is comparable to a clinically achievable in vivo dose of 12.5mg/m2 (Leukemia2008; 22:965). 500nM MGC is comparable to a 280mg/m2in vivo dose (Blood2006; 108: 1954). Additional titration experiments with MGC will be tested to achieve clinically relevant concentrations in vivo that can induce MAGE-A3 expression. In conclusion, epigenetic modulation by Aza and MGC can enhance MAGE-A3 expression and result in increased killing by MAGE-A3 specific CTLs. Hypomethylating agents and HDACi may be useful to sensitize tumor cells to immune effectors. Figure 1. Treatment with 50nM Aza and/or sequential MGC at 500 nM induces de nono expression of MAGE-A3 protein in the myeloma a cell line transfectant LP1 A68. Figure 1. Treatment with 50nM Aza and/or sequential MGC at 500 nM induces de nono expression of MAGE-A3 protein in the myeloma a cell line transfectant LP1 A68. Figure 2. Lysis of LP-1 A68 Aza/MGC treated targets by MAGE-A3 specific CTL effective. Figure 2. Lysis of LP-1 A68 Aza/MGC treated targets by MAGE-A3 specific CTL effective.