At the heart of this story is an enzyme with a deceptively simple job. CD73, also known as ecto-5′-nucleotidase, sits on the surface of cells and converts adenosine monophosphate, or AMP, into adenosine. It acts as the final and rate-limiting step in a two-enzyme cascade: CD39 first strips phosphate groups from extracellular ATP, a molecule released in abundance by dying and stressed cells, and CD73 then finishes the conversion. In healthy tissue this pathway helps resolve inflammation and prevent excessive immune damage. In tumors, however, hypoxic and necrotic conditions flood the microenvironment with extracellular ATP, effectively handing CD73 the raw material it needs to saturate the tumor surroundings with immunosuppressive adenosine. The result is a biochemical fog that blunts the activity of T cells, natural killer cells, and dendritic cells precisely where the immune attack on cancer needs to be sharpest.

One of the review’s central contributions is its detailed mapping of where CD73 appears within the NSCLC tumor microenvironment. The enzyme is not confined to a single cell type. Malignant cells themselves frequently display high levels of CD73 on their surfaces, and this expression often correlates with more aggressive disease, greater metastatic potential, and poorer survival. But the story extends well beyond the tumor cells. Immune cell populations within the tumor, including subsets of T cells and myeloid-derived suppressor cells, can also express CD73, effectively turning the body’s own defensive forces into adenosine-generating factories. Stromal cells, the connective and supporting tissue that scaffolds the tumor, contribute to the axis as well. This multicompartmental distribution matters clinically, because it suggests that therapies targeting CD73 must contend with adenosine production from several cellular sources simultaneously, and that measuring CD73 in only one compartment may seriously underestimate the pathway’s activity in a given patient.

The regulatory network controlling CD73 expression is equally intricate, and the review devotes considerable attention to untangling it. Hypoxia stands out as a dominant driver: low oxygen conditions within tumors stabilize hypoxia-inducible factors, particularly HIF-1, which binds to the CD73 promoter and ramps up enzyme production. This creates a vicious feedback loop, because the very oxygen deprivation that characterizes rapidly growing tumors directly instructs them to build their immunosuppressive shield. Beyond hypoxia, inflammatory and oncogenic signaling pathways converge on CD73 regulation. The transcription factor NF-kappaB, a master regulator of inflammation, along with pathways such as TGF-beta, Wnt, and various oncogenic signaling cascades, can modulate CD73 expression in response to cues from the microenvironment. Epigenetic mechanisms, including DNA methylation patterns at the CD73 gene locus, add another layer of control, and the interplay of these pathways helps explain why CD73 levels vary so dramatically between patients and even between regions of the same tumor.

Perhaps the most conceptually important section of the review addresses the fact that CD73 promotes tumor progression through both adenosine-dependent and adenosine-independent mechanisms. The adenosine-dependent arm is the classical story: once generated, adenosine engages a family of G-protein-coupled receptors on immune and stromal cells, chiefly the A2A and A2B receptors. Signaling through these receptors raises intracellular cyclic AMP in T cells, dampening their activation, proliferation, and cytotoxic function. Adenosine simultaneously skews the tumor microenvironment toward immunosuppression by promoting regulatory T cells and M2-like macrophages, stimulating angiogenesis, and encouraging tumor cell migration and invasion. In this way, a single enzymatic reaction cascades into a coordinated suppression of nearly every arm of the anti-tumor immune response.

The adenosine-independent actions of CD73, by contrast, reveal the molecule as more than a metabolic enzyme. CD73 can participate directly in cell adhesion and signaling, influencing epithelial-mesenchymal transition, the process by which cancer cells acquire migratory and invasive properties. It has been implicated in supporting cancer stem-like cell populations, which are thought to seed relapse and resist conventional therapies. These functions mean that even if adenosine signaling were fully blocked downstream, CD73 itself might continue to drive malignancy through physical and signaling interactions at the cell membrane. For drug developers, this dual identity argues strongly for targeting the enzyme itself rather than only its product, and it helps explain why complete CD73 inhibition may deliver benefits beyond what adenosine receptor antagonists alone can achieve.

Translating this biology into medicine has produced a growing portfolio of clinical candidates. The review surveys the latest developments in CD73-targeted therapies in NSCLC, including monoclonal antibodies such as oleclumab and other agents designed to block the enzyme’s active site or flag CD73-expressing cells for immune destruction. Clinical trials have explored these drugs in combination with the workhorses of modern lung cancer care: PD-1 and PD-L1 immune checkpoint inhibitors, chemotherapy, and radiation. The biological rationale for these combinations is compelling. Checkpoint inhibitors release the brakes on T cells, but in an adenosine-rich environment the unleashed cells remain metabolically paralyzed; pairing CD73 blockade with PD-1 or PD-L1 inhibition addresses both the ignition and the fuel supply of the anti-tumor response. Similarly, chemotherapy and radiation kill tumor cells, releasing ATP that CD73 would otherwise convert into immunosuppressive adenosine, so adding a CD73 inhibitor may convert treatment-induced cell death into productive immune priming rather than immune escape.

The clinical results to date show significant promise, though the review is careful to note the challenges that remain. Early-phase trials have demonstrated that CD73 inhibition is generally feasible and can produce meaningful activity in selected patients, particularly when layered onto existing immunotherapy. Yet responses have been heterogeneous, and not every combination has cleared the bar of randomized testing. This variability points to one of the field’s most pressing needs: better biomarkers. CD73 expression measured at a single time point on a single cell type may not capture the dynamic, spatially variable nature of the adenosine axis in a living tumor. The authors highlight the development of dynamic biomarkers, capable of tracking pathway activity over the course of treatment, as a key future research direction. Such tools could identify which patients are most likely to benefit from CD73 blockade and reveal when resistance emerges, enabling the kind of adaptive, precision-guided treatment decisions that have transformed other areas of oncology.

The review also looks ahead to novel combination strategies that could extend the reach of CD73 targeting. Beyond checkpoint inhibitors, chemotherapy, and radiotherapy, the authors point toward rational pairings with agents that modulate other metabolic pathways in the tumor microenvironment, with drugs targeting additional adenosine receptors, and with emerging approaches that reshape the immune landscape more broadly. Because the CD73-adenosine axis intersects with hypoxia, inflammation, and stromal biology, it offers numerous points of therapeutic leverage, and the optimal combinations will likely differ between patients whose tumors rely on different regulatory programs. The authors, supported by funding from the National Natural Science Foundation of China and Guangdong provincial research programs, frame these questions within the larger goal of precision immunotherapy: matching each patient’s tumor to the specific combination of agents most likely to dismantle its particular immune defenses.

For patients with NSCLC, the stakes of this research could hardly be higher. Immunotherapy has already extended survival for thousands, but resistance through metabolic immunosuppression remains one of the most stubborn barriers to durable cures. The CD73-adenosine axis sits at the intersection of tumor metabolism, immune regulation, and treatment resistance, and the systematic synthesis provided by Cai, Ren, Li, and colleagues clarifies both why the pathway matters and how best to attack it. As clinical trials mature and biomarker strategies evolve, blocking the final step of adenosine production may prove to be one of the pivotal advances that converts lung cancer from a frequently fatal disease into a manageable chronic condition for a far larger share of the people it touches.

Subject of Research: The role of the CD73-adenosine axis in immune suppression, tumor progression, and targeted combination therapy in non-small cell lung cancer

Article Title: The CD73-adenosine axis in NSCLC: expression regulation, pro-tumor mechanisms, and combination therapy

Article References: Cai, D.-X., Ren, Z.-R., Li, J.-T., Xu, C.-R., Chen, Z.-H., Deng, Y., & Zhou, Q. (2026). The CD73-adenosine axis in NSCLC: expression regulation, pro-tumor mechanisms, and combination therapy. Journal of Cancer Research and Clinical Oncology. https://doi.org/10.1007/s00432-026-06612-8

Image Credits: AI Generated

DOI: 10.1007/s00432-026-06612-8

Keywords: NSCLC, CD73, adenosine, immunotherapy, tumor microenvironment, immune checkpoint inhibitors, hypoxia, combination therapy, cancer metabolism, biomarkers, lung cancer, targeted therapy