Document Type


Publication Date

Fall 10-26-2023


Acute myeloid leukemia (AML) encompasses a diverse group of cancers that originate in the blood-forming tissues of the bone marrow. Aside from the M3 subtype (PML-RARA+), AML carries a 5-year survival rate of 28% for patients 20+ years of age. AML is the most common cancer of the hematopoietic system and is slightly more common in biological males; the average age at diagnosis is 68 years. Standard frontline treatment for AML is a 2-phase regimen of intensive chemotherapy (CTx) employing daunorubicin and cytarabine. Despite 60-70% of patients achieving complete remission (CR), at least half of CR-achieving patients experience relapse within 3 years from their diagnosis. Additionally, 30-40% of patients present with refractory AML, experiencing little to no benefit from frontline treatment.

AML relapses when a pool of undetectable, CTx-resistant leukemia stem cells (LSCs) survives & proliferates after frontline CTx [1]. Notably, the poor performance status of many AML patients precludes use of the standard CTx regimen; while reduced-intensity CTx still offers therapeutic benefit, it is less effective at killing LSCs and, as a result, relapse is more likely. Goardon, et al. determined that AML patients harbor two types of LSCs: granulocyte-macrophage progenitor (GMP)-like LSCs and FLT3+ lymphoid-primed multipotential progenitor (LMPP)-like LSCs [2]. Eradication of both types of LSCs is necessary to maintain CR in AML.

Our group and others have established that ~40% of AML patients express upregulated Toll-like receptor (TLR) signaling (TLR+). TLR+ disease is associated with specific genetic abnormalities, such as MLL rearrangements (MLL-r+), and is inversely associated with prognosis (Figure 1) [3,4]. TLR+ AML represents a challenging, treatment-sparse subset of an already difficult-to-treat disease. To study TLR+ AML, we utilize an MLL-r+ model using the MLL-AF9 oncogene.

We have also demonstrated that both GMP- and LMPP-like LSCs require TLR-associated Ser/Thr protein kinases for their survival [5-7]. Specifically, GMP-like LSCs require TAK1 and LMPP-like LSCs require TBK1. The loss of either Tak1 or Tbk1 ablates the corresponding LSC pool and enriches for the opposite LSC pool in vitro and in vivo. Recently, our group determined that the genetic loss of Tak1 sensitizes mouse AML cells to TBK1 blockade in vitro. Strikingly, the loss of Tbk1 also seems to extend overall survival (OS) despite causing extramedullary AML.

While mice given Tbk1NULL AML cells develop a subcutaneous tumor of AML cells (chloroma) near the pelvis, they survive longer than mice given control (Tbk1WT) AML cells. The clinical significance is unknown, but these data support our impression that the loss of Tbk1 forces AML cells to differentiate; this should be therapeutically favorable, as inducing the differentiation of AML cells is an effective treatment strategy. Theoretically, chloromas may form in Tbk1NULL AML due to the enrichment of GMP-like LSCs, which express higher levels of chemokine receptors.

We hypothesize that the differentiation & eradication of LSCs can be induced by blocking TAK1/TBK1 in combination with standard CTx (and possibly targeted agents like Mylotarg®, Venclexta®, and/or Xospata®). We propose TAK1/TBK1 parallel blockade as augmentation to standard CTx, ideally allowing for a dose-reduction of CTx & promoting improved patient outcomes.


Author Posting © The Author(s), 2023. This was a presentation given on Fall 2023.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.