Acute myeloid leukemia (AML) does more than infiltrate the lungs: it also transforms lung tissue to create an inflammatory environment that promotes the expansion of tumor cells and impairs respiratory function. This is demonstrated by a study published in Nature Immunology , co-led by Dr. Manel Esteller , an ICREA research professor, head of the Cancer Epigenetics Group at the Sant Pau Research Institute (IR Sant Pau) , and professor of Genetics at the University of Barcelona, and Dr. Iannis Aifantis of the New York University Grossman School of Medicine (NYU Grossman School of Medicine).
Using single-cell and spatial transcriptomics, experimental animal models, human samples, and clinical data, the researchers observed that leukemia cells accumulated in alveolar and vascular regions, disrupted capillary integrity, and altered both structural cells and the lung’s immune response . The study also identified galectin-9 and the IL-33/IL1RL1 axis as potential targets for reducing this serious complication of the disease.
AML is an aggressive leukemia characterized by the accumulation of immature myeloid cells in the bone marrow. Although advances in understanding its biology and the development of targeted treatments have improved survival, serious complications during the early stages of the disease remain common. These include respiratory failure caused by the infiltration of leukemia cells into the lungs , a phenomenon whose biology has remained poorly understood.
“We designed a broad study because this was a complex question that required us to combine basic research with potential clinical translation. We used next-generation technologies to study individual cells, analyze their location within lung tissue, and conduct functional experiments in both animal models and human samples,” explains Dr. Esteller.
In different animal models of AML, the researchers observed that leukemia cells reached the lungs through the bloodstream and accumulated particularly in alveolar and perivascular regions , close to the capillaries and epithelial cells responsible for gas exchange between the air and the blood. From these areas, the cells could cross the vascular wall and enter the lung stroma.
“The first thing we observed was that when leukemia cells reach the lungs, they alter the function of the organ’s different cell types. They do not simply invade the tissue; they create a niche—an environment that supports their persistence and expansion,” says Dr. Esteller.
This favorable environment was characterized by extensive remodeling of the lung architecture . The lungs of the affected animals showed increased vascular permeability, a reduction in endothelial cells, and a decline in populations essential for gas exchange, including capillary aerocytes and type 1 alveolar epithelial cells.
At the same time, the numbers of fibroblasts and myofibroblasts—cells involved in tissue repair and fibrosis—increased. The analyses also detected greater collagen deposition , indicating that leukemic infiltration can induce fibrotic remodeling of the lungs and progressively reduce their functional capacity.
Functional experiments in animals confirmed that these changes had respiratory consequences. Animals with lung infiltration showed an increase in the volume of air moved with each breath, consistent with a compensatory response to impaired lung function . The loss of vascular integrity also increased as the leukemic burden in the tissue grew.
Another major finding of the study was the identification of an extensive inflammatory response in infiltrated lungs . This response affected cells in the blood vessels, lung tissue, and immune system and activated mechanisms related to inflammation, oxygen deprivation, and cellular damage, including pathways mediated by IL-6, TNF, and interferons.
This activation was also reflected in increased levels of numerous inflammatory cytokines and chemokines in the animals’ plasma and bronchoalveolar fluid. The pattern was reproduced in different mouse models of AML with distinct genetic alterations, indicating that the phenomenon is not restricted to a specific subtype of the disease .
Spatial transcriptomics showed that the inflammation was not confined to areas surrounding the tumor cells but extended widely throughout the lung tissue. Some of the strongest inflammatory signals were observed in the endothelium, one of the cell populations located closest to the leukemia cells.
“Tumor cells establish a network of interactions with the cells surrounding them. This communication alters both the lung structure and the local immune response and helps amplify inflammation,” explains Dr. Esteller.
This disruption of the immune response resulted in a profound transformation of the cell populations present in the lungs. Several populations that participate in inflammation increased: activated neutrophils, which act as a first line of defense; interstitial macrophages, which monitor lung tissue and clear cellular debris; and nonclassical monocytes, immune cells that help regulate inflammation. At the same time, T and B lymphocytes, which are essential for recognizing and fighting tumor cells , decreased. The result was a more inflammatory lung environment that was less effective at controlling the leukemia.
This pattern more closely resembled that described in COVID-19 than that observed in lung cancer . In both AML and COVID-19, monocytes increase, while T and B lymphocytes and NK cells—which help identify and eliminate abnormal cells—decrease. In lung cancer, by contrast, inflammatory activation is more limited and is concentrated primarily in first-line immune defense cells, such as neutrophils, monocytes, macrophages, and dendritic cells, rather than extending broadly throughout the immune system. AML also displayed specific features in the models analyzed, including an increase in nonclassical monocytes and particularly extensive activation of the IL-6/JAK/STAT3 pathway , one of the mechanisms that helps amplify and sustain inflammation.
The analysis of cell-to-cell interactions identified several molecules involved in communication between AML cells and lung tissue. The researchers focused particularly on galectin-9 and the axis formed by the cytokine IL-33 and its receptor IL1RL1 , also known as ST2. Both pathways appeared to help leukemia cells take advantage of signals generated in the inflamed lungs to persist and expand within the organ.
“Galectin-9 acts as one of the tools leukemia uses to interact with the lung environment and promote conditions that allow it to persist and expand ,” explains Dr. Esteller. The analyses showed that this protein was highly expressed in leukemia cells and participated in their communication with different components of the lung microenvironment. In addition, in patient databases, high expression of the LGALS9 gene was associated with greater inflammatory activity, extramedullary involvement, a higher percentage of blasts in peripheral blood, and poorer survival outcomes.
When the researchers blocked galectin-9 with a monoclonal antibody in animal models, they observed a significant reduction in leukemic infiltration of the lungs , as well as a decrease in inflammation and fibrosis . The treatment also increased the presence of B cells, NK cells, and T lymphocytes, reduced the proportion of exhausted T cells, and, in a second animal model with different genetic alterations, also reduced the leukemic burden in the bone marrow and prolonged survival.
The researchers also studied the IL-33/IL1RL1 axis. IL-33 is produced mainly by structural cells in the lungs, particularly type 2 alveolar epithelial cells, while its receptor is found on leukemia cells. This interaction could allow the leukemia to respond to damage signals generated in lung tissue. Blocking the IL-33 receptor likewise reduced the leukemic burden in the animals’ lungs, decreased fibrosis, and altered tumor cell activity: the expression of inflammatory genes declined, while programs associated with cellular stress and apoptosis increased.
“Our preclinical animal studies show that interfering with these pathways can disrupt the niche that supports leukemia and reduce its ability to infiltrate the lungs,” Dr. Esteller emphasizes. “The findings provide a scientific basis for continuing to explore these strategies in clinical studies.” However, these therapeutic results were obtained in animal models , and the safety and efficacy of these interventions will need to be evaluated in studies specifically designed for patients.
The study also evaluated prednisone, a glucocorticoid that reduces inflammation and is already used, based on clinical experience, in some patients with AML and respiratory failure attributed to lung infiltration. Until now, however, there has been little specific evidence regarding its usefulness in this complication. In animal models, the treatment significantly reduced the presence of leukemia cells in the lungs and decreased their accumulation both in the alveolar capillaries and in the tissue between lung structures.
The researchers also retrospectively analyzed the outcomes of eight patients with newly diagnosed AML who presented with respiratory failure and suspected lung infiltration, without symptoms or imaging findings consistent with an infection. After receiving prednisone, all patients experienced respiratory improvement and significantly reduced their oxygen requirements within 12 hours. At 24 hours, the median reduction in supplemental oxygen requirements was 75% compared with baseline.
These results support the hypothesis that inflammation directly contributes to respiratory deterioration and that controlling it can lead to rapid improvement in carefully selected patients. However, this was a small retrospective series without a control group , and prospective studies will therefore be needed to confirm the benefit and precisely define the circumstances in which this treatment should be used. It is also essential to distinguish leukemic infiltration from lung infections, which are common in people with AML, because glucocorticoids can be harmful when an uncontrolled infection is present.
Until now, most research on the AML microenvironment has focused on the bone marrow. This study demonstrates that leukemia is also capable of actively remodeling the tissues it infiltrates outside the bone marrow .
The findings show that the lungs do not act as passive recipients of tumor cells. Endothelial disruption, fibrosis, the loss of cells essential for gas exchange, widespread inflammation, and reorganization of the immune system collectively help create an environment favorable to the disease.
The study therefore opens up new therapeutic possibilities aimed not only at leukemia cells themselves but also at the interactions they establish with infiltrated organs. In addition to galectin-9 and IL-33, the molecular map generated by the researchers identified other potentially modifiable pathways, including IL-1, TGF-β, CXCL2, CD44, and various integrins.
“Lung infiltration is the result of a complex relationship between leukemia cells and tissue cells. Understanding this communication allows us to identify vulnerable points and raises the possibility of targeting the microenvironment to limit a particularly serious complication of the disease,” concludes Dr. Esteller.
Nature Immunology
Experimental study
People
Inflammatory immune modulators of AML lung infiltration and respiratory failure
29-Jun-2026