Immune-Scape in Myeloid Sarcoma and Acute Myeloid Leukemia

Background. Acute myeloid leukemia (AML) is an aggressive bone marrow (BM) malignancy with less than a third of patients surviving beyond 5 years (National Cancer Institute, 2024). Myeloid sarcomas (MS) are tumors of extramedullary myeloid blasts which may confer a worse prognosis in AML (Zhao et al., 2022). Although most MS harbor driver genetic abnormalities such as NPM1, MLL aberrancies, t(8;21), inv(16), TET2 deficiency, among others, there are no clear somatic mutation profile that distinguish MS from bone marrow (BM)-only disease (Shallis et al., 2021; Zorn et al., 2023). The microenvironment in AML plays a critical role in treatment resistance and disease recurrence, although data in this field is lacking (Tettamanti et al., 2022). In the post-HSCT setting, MS is linked to increased risk of AML recurrence (Zhao et al., 2022); in fact, up to 25% of patients with AML who relapse after HSCT are found to have MS (Solh et al., 2016). This could be due to evasion of the graft-versus-leukemia effect secondary to an immunosuppressive microenvironment within the MS (Harris et al., 2013; Shimizu et al., 2013). Among different pathways, TGF-β is strongly associated with immune evasion and migration in cancer cells (de Streel & Lucas, 2021; Qi et al., 2022; Tettamanti et al., 2022), as well as worse survival in patients with AML (Bottomly et al., 2022; de Bruijn et al., 2023). Understanding the molecular mechanisms surrounding MS and their immune-evasive pathways is essential to aid in the development of immuno-oncology and precision oncology approaches.

We performed bulk RNA sequencing (RNA-seq) from patient and mouse samples of MS and BM AML and identified that the top ranked gene set enriched in MS was Epithelial-Mesenchymal Transition (EMT); we also identified that the key EMT transcription factor TWIST1 showed increased expression in the MS vs. BM samples, for both species (Zorn KE et al., 2024). We hypothesized specific gene signatures present within MS are essential players in generating an immunosuppressive microenvironment which aids in MS development and disease recurrence.

Methods and Results. We mined our published RNA-seq data (Zorn KE et al., 2024) and performed a differential expression (DESeq2) and gene set enrichment analysis (Love et al., 2014; Subramanian et al., 2005) to identify potential pathways within EMT that could lead to immune-escape mechanisms within the MS microenvironment. The paired MS-BM data and samples were obtained from the Froedtert Hospital/ Medical College of Wisconsin database for patients with AML. The mouse samples were obtained from the Npm1^cA mouse model which spontaneously generates MS. We observed that TGF-β3 was significantly differentially expressed in patient MS vs. BM AML, with log2fold change of 4.69 (p adj <0.001). TGF-β1 was not differentially expressed in MS vs. BM in the patient samples (log2fold change 0.78, p adj 0.509). This expression pattern and statistical significances for TGF-β3 and TGF-β1 were also reflected in the mouse samples. Regarding the TGF-β receptor expression, in patient samples TGF-βR3 was significantly differentially expressed in MS vs. BM (log2fold change 3.85, p adj 0.002); however, there was no significant differential expression for the other TGF-β receptors in patient samples or for any of the receptors in the mouse samples. We performed immunohistochemistry (IHC) staining of TGF-β3 in patient MS and BM samples and identified an increased expression in the MS vs. BM tissues, consistent with the RNA-seq findings. We are currently planning further studies in the Npm^1cA mouse model in which we evaluate an immune-escape profile of MS vs. BM via IHC and flow cytometry assays for secreted and cell surface markers (e.g., TGF-β, PD1/ PDL1, CTLA4, B7, IL10) and perform an adoptive T-cell transfer model evaluating for antitumor activity within the MS vs. BM. We will evaluate its correlation with TWIST1 via RT-qPCR in the samples.

Conclusion. MS exhibit an EMT-like signature compared to BM AML. Within this signature the TGF-β-TWIST1 pathway appears to mediate immune-escape in the leukemic blasts, leading to resistance and recurrence after therapy. TGF-β signaling can be a key driver of EMT gene expression and this may help explain the ability of MS to evade immunological therapies. Future studies are ongoing and will likely shed light on this important knowledge gap, which could help open new avenues for the optimization of patient outcomes.