In patients with relapsed or refractory acute myeloid leukemia (AML), blocking immature leukemia cells from migrating to the bone marrow increased the cancer-killing effect of chemotherapy in a recent phase Ib/IIa trial. “Homing of leukemia cells to the bone marrow depends on CXCR4, a chemokine receptor, which the drug BL-8040 blocked,” said Gautam Borthakur, M.D., associate professor in the department of leukemia at the University of Texas MD Anderson Cancer Center in Houston. Blocking CXCR4, which is involved in cell migration and adhesion, prevented immature leukemia cells from hiding in the bone marrow, where they can cause recurrence, said Borthakur, lead researcher of the study that he presented in Orlando, Fla.at the 58 th annual meeting of the American Society of Hematology (ASH) in December 2015 (abstr. 2546; https;/ash.confex.com/ash/2015/webprogram/Paper82437.html). The composite complete remission rate in 22 patients was 38% at the three highest doses.
Gautam Borthakur, M.D.
“The idea is that if we can disrupt cancer cell adhesion to the bone marrow, we can flush out cancer cells hiding there,” Borthakur said. BL-8040 also increased cancer cell death by mobilizing immature AML cells to move into peripheral blood, where cytarabine, given later, killed them. Two days after taking BL-8040, patients showed a 40.2-fold mobilization of immature AML cells, as well as substantial apoptosis (programmed cell death) of immature progenitors remaining in the bone marrow. Granulocytes, a type of white blood cell, increased 3.1-fold on day 3, and blood cells began differentiating into mature cells. A maintenance trial with BL-8040 and cytarabine is being planned, Borthakur said.
In AML the bone marrow overproduces immature, dysfunctional blood cells, or blasts, that do not develop into normal, functional blood cells. Chemotherapy kills cancer cells in the peripheral blood. But chemotherapy does not kill immature cells sheltered by the bone marrow and its microenvironment, which is thought to be a reason for high rates of minimally residual disease after treatment and eventual disease recurrence. In recent years researchers have begun focusing on the bone marrow and the microenvironment, progenitor and stem cells, and adhesion molecules like CXCR4 to target the disease at its roots. “The microenvironment has long been considered a target in solid tumors, but only recently have researchers focused on it for blood cancers,” said Craig Jordan, Ph.D., Nancy Carroll Allen Professor and chief of the division of hematology at the University of Colorado in Aurora. Researchers debate whether the immature cells must leave the bone marrow to be killed or whether cells are more vulnerable while residing in it, Borthakur said.
Targeting the microenvironment and stem cells is somewhat controversial, since many researchers believe that it is more effective to target mutations, said Michael Andreeff, M.D., Ph.D., professor of medicine in the departments of leukemia and stem cell transplantation at MDAnderson. “Mutations are necessary for cancer development and progression, but they are not sufficient,” he said. “To target AML in particular, it is necessary to target the stroma and bone marrow microenvironment, where blood cells differentiate,” Andreeff said. If immature cells can be induced to differentiate, cure may be possible. “The microenvironment, which has a different set of signaling pathways, can mimic mutations and must be targeted to prevent relapse,” Andreeff said.
Another inhibitor of CXCR4 in trials for AML and other blood cancers is plerixafor. In a phase II trial with 69 patients, combining plerixafor with the hypomethylating agent decitabine in newly diagnosed older AML patients resulted in an overall response rate of 43%. Median overall survival for responders was 18 months, compared with 5 months for decitabine alone (ASH 2013abstr. 621; https://ash.confex.com/ash/2013/webprogram/Paper62535.html ). Prior treatment with hypomethylating drugs was the strongest independent predictor of overall survival: 52% of drug-naïve patients achieved complete responses, compared with only 14% of those previously treated with these drugs. Leukemia stem cells (LSCs) and progenitor cells were mobilized in a subset of patients as a result of CXCR4 blockade.
CXCL4 acts through several pathways to promote leukemia progression and activates the PI3K–Akt and mitogen-activated protein kinase signaling pathways, which directs leukemia cell survival and proliferation. Higher levels of CXCR4 are found in these AML cells, and interaction between CXCR4 and the microenvironment can occur in response to cancer therapies and is involved in drug resistance. “The highest levels of CXCR4 are found in in FLT3-mutated AML, one of the most difficult types of AML to treat, and a good reason to target CXCR4,” Andreeff said.
Eradicating Stem Cells, Targeting Stroma, and Microenvironment
“Treatments that kill bulk disease do not target leukemic stem cells,” Jordan said. Using drugs to treat AML and other stem cell–derived cancers without targeting stem cells is like mowing a lawn full of dandelions,” Jordan said. It kills visible weeds, but without reaching the roots, it’s only a temporary solution, he said. Progenitor and LSCs have different properties, such as quiescence, from most AML cells, which makes them difficult to eradicate with chemotherapy but easy to identify and target.
AML was one of the first cancers in which the existence of cancer stem cells was proven, and consequently, they are among the best-characterized cancer stem cell population, said Daniel Pollyea, M.D., clinical director of leukemia services at the University of Colorado at Aurora. “We know properties in these cells, from lab studies, that represent a true Achilles’ heel,” Pollyea said. Although treating chronic myeloid leukemia without targeting stem cells is possible because one determinative mutation is responsible for that cancer, that is not the case in AML, he said.
Another advantage in targeting LSCs is that disease regression can occur even after clonal evolution has happened, Pollyea said. The self-renewing properties of LSCs depend on activating signaling pathways, including WNT–β-catenin, NOTCH, and Hedgehog, all of which are good LSC targets. In addition, therapies that inhibit antigens more highly expressed on LSCs than on hematopoietic stem cells, such as CD123 and CD47, and kinases such as c-Kit and Src kinases are also good strategies for eradicating cancerous stem cells. Clinical trials are under way for AML with all these targets.
A related issue is the bone marrow microenvironment and stroma’s relationship to LSCs and leukemia cells. The microenvironment was a prominent topic of discussion among scientists this year at ASH. Scientists from the German Cancer Research Center in Munich sought to disrupt the protective interaction of stroma in a line of AML cells that is particularly resistant to treatment (abstr. 676; https://ash.confex.com/ash/2015/webprogram/Paper84991.html ). Combining azacytidine with a new kinase inhibitor, crenolanib, of the resistant FLT-3 mutation, interfered with stromal protection of CD34 leukemia-initiating cells. They hypothesized that when the cells lost their quiescence disease cells became more susceptible to therapy with tyrosine kinase inhibitors.
Another strategy for understanding microenvironment-mediated signaling is proteomic profiling, said Marina Konopleva, M.D., Ph.D., associate professor in the department of leukemia of the division of cancer medicine at MD Anderson. Konopleva and colleagues studied stroma-mediated paths of resistance to targeted drugs that come with relapse in order to identify effective combinations to kill residual disease (abstr. 1398; https://ash.confex.com/ash/2015/webprogram/Paper85339.html ). They profiled 53 key molecules in 11 signaling pathways in 20 samples treated with several drugs. Changes in the stroma-regulated network responsible for cell survival are specific to each AML therapy. They found, for example, that activation of the PI3K–AKT–mTOR pathway was a common survival mechanism mediated by stroma in response to inhibition of two cancer-related molecules. By protein profiling, the team found mechanism-based drug combinations to address drug resistance in each case.
Focusing on Apoptosis
The Bcl-2 antiapoptosis protein is found in high levels in AML cells and plays a large role in disease pathogenesis, stem and cancer cell survival, and chemotherapy resistance. Having higher levels of Bcl-2 in CD34 + hematopoietic stem cells confers a survival advantage to diseased cells and resistance to chemotherapy and helps maintain a favorable antiapoptotic microenvironment for the survival of AML blasts.
An anti-Bcl2 drug discussed extensively at ASH is venetoclax, or ABT-199. In a phase Ib study of with decitabine or azacytidine in 22 treatment-naïve patients aged 65 years or older and ineligible for standard induction therapy, the overall response rate was 75% (abstr. 327; https://ash.confex.com/ash/2015/webprogram/Paper84265.html ). Three patients who did not experience objective responses still had beneficial decreases in bone marrow blast counts. “While still early, the responses were rapid and very encouraging, much better than with hypomethylating agents alone,” said lead investigator Courtney DiNardo, M.D., assistant professor in the department of leukemia at MD Anderson. Many patients had stable disease at 6 months and were no longer transfusion dependent. A phase II is ongoing in 125 patients.
Resistance to venetoclax is an important issue. Researchers from the University of Helsinki in Finland reported that combining venetoclax with ruxolitinib, a JAK1/2 inhibitor, was synergistic in combating stromal resistance to venetoclax (abstr. 867; https://ash.confex.com/ash/2015/webprogram/Paper84426.html ). Stroma-derived cytokines induce JAK–STAT signaling in AML cells, which increases Bcl-2 expression and creates venetoclax resistance. Blocking JAK1/2 with ruxolitinib restored sensitivity of AML cells to venetoclax.
Additional studies to tackle venetoclax resistance by using inhibitors of other signaling pathways were discussed. Lina Han, Ph.D., a postdoctoral student at MD Anderson combined it with cobimetinib, a mitogen-activated protein kinase inhibitor, in mice, and found synergistic activity in blocking both pathways, which overcame venetoclax resistance (abstr. 2544; https://ash.confex.com/ash/2015/webprogram/Paper81612.html ). Combining venetoclax with idasanutlin, which activates p53 and produced synergistic apoptosis, also overcame venetoclax resistance (abstr. 673; https://ash.confex.com/ash/2015/webprogram/Paper85569.html ). The hope is that by targeting AML at its roots it will be possible to raise what remains a dismal 5-year survival rate.
