Martin Carroll

Chemotherapy remains a primary treatment for younger AML patients, though many relapse. Data from our group have shown that highly phosphorylated S6 in blasts may predict response to sirolimus given with chemotherapy. We report the results of a phase I study of this combination in newly diagnosed AML and the pharmacodynamic analysis of pS6 before and after treatment. Subjects received sirolimus (12 mg on day 1, 4 mg daily, days 2–10), then idarubicin and cytarabine (days 4–10). Response was assessed at hematologic recovery or by day 42 using a modified IWG criteria. Fifty-five patients received sirolimus. Toxicity was similar to published 7 + 3 data, and 53% had high-, 27% intermediate-, and 20% favorable-risk disease. Forty-four percent of the high-risk patients entered into CR/CRp. Seventy-nine percent of the intermediate-risk subjects had a CR/CRp. All favorable-risk patients had a CR by day 42; 9/11 remained alive and in remission with a median follow-up of 660 days. Additionally, 41/55 patients had adequate samples for pharmacodynamic analysis. All patients demonstrated activation of S6 prior to therapy, in contrast to 67% seen in previous studies of relapsed AML. mTORC1 inhibition was observed in 66% of patients without enrichment among patients who achieved remission. We conclude that sirolimus and 7 + 3 is a well-tolerated and safe regimen. mTORC1 appears to be activated in almost all patients at diagnosis of AML. Inhibition of mTORC1 did not differ based on response, suggesting that AML cells may have redundant signaling pathways that regulate chemosensitivity in the presence of mTORC1 inhibition.

Despite recent therapeutic advances, the 5-year survival rate for adults with acute myeloid leukemia (AML) is poor and standard-of-care chemotherapy is associated with significant toxicity, highlighting the need for new therapeutic approaches. Recent work from our group and others established that the G protein-coupled estrogen receptor (GPER) is tumor suppressive in melanoma and other solid tumors. We performed a preliminary screen of human cancer cell lines from multiple malignancies and found that LNS8801, a synthetic pharmacologic agonist of GPER currently in early phase clinical trials, promoted apoptosis in human AML cells. Using human AML cell lines and primary cells, we show that LNS8801 inhibits human AML in preclinical in vitro models, while not affecting normal mononuclear cells. Although GPER is broadly expressed in normal and malignant myeloid cells, this cancer-specific LNS8801-induced inhibition appeared to be independent of GPER signaling. LNS8801 induced AML cell death primarily through a caspase-dependent apoptosis pathway. This was independent of secreted classical death receptor ligands, and instead required induction of reactive oxygen species (ROS) and activation of endoplasmic reticulum (ER) stress response pathways including IRE1α. These studies demonstrate a novel activity of LNS8801 in AML cells and show that targeting ER stress with LNS8801 may be a useful therapeutic approach for AML. Significance: Previous work demonstrated that LNS8801 inhibits cancer via GPER activation, especially in solid tumors. Here we show that LNS8801 inhibits AML via GPER-independent mechanisms that include ROS induction and ER activation.