B7-H3, also known by its scientific name CD276, belongs to the B7 family of immune checkpoint proteins, the same extended family that includes the famous PD-L1 target of modern immunotherapy. In many tumors, B7-H3 is displayed at high levels on the cell surface, where it is thought to help the cancer evade immune surveillance. Earlier clinical work had already shown that patients with triple-negative breast cancer whose tumors carry abundant B7-H3 tend to fare worse, but the question of what the molecule actually does inside these cells remained open. The new study set out to answer that question by combining large public datasets, tissue microarrays from patient samples, multiplex immunofluorescence imaging, RNA sequencing, and a battery of genetic perturbation experiments in the laboratory.

The evidence that B7-H3 matters in this disease is striking. Across the datasets and patient tissue analyses, the team found that B7-H3 was consistently elevated in triple-negative tumors compared with normal breast tissue, and that high expression correlated with poorer overall survival and earlier recurrence. When the researchers forced cancer cells to overproduce B7-H3 in culture, the cells became more aggressive on every measure they tested: they migrated farther, invaded through barriers more effectively, formed more colonies in clonogenic growth assays, and resisted apoptosis, the programmed cell death that would normally cull damaged or abnormal cells. Silencing B7-H3 with small interfering RNAs produced the mirror image of these effects, calming the cells down and restoring their vulnerability to death signals.

To find out how B7-H3 exerts these effects, the team turned to RNA sequencing, a technique that reads out the activity of thousands of genes at once. Comparing cells with high and low B7-H3, one gene stood out from the crowd: HSPA8, which encodes a member of the HSP70 family of heat shock proteins. These chaperones are the cell’s folding assistants, stabilizing newly made proteins and protecting clients from stress-induced damage, and cancer cells often lean on them heavily to survive the hostile conditions inside a tumor. When the researchers depleted HSPA8, the malignant phenotypes that B7-H3 overexpression had created largely collapsed, indicating that this chaperone is a critical downstream effector of B7-H3’s tumor-promoting activity.

The next challenge was to trace the path from a cell-surface protein to a chaperone gene inside the nucleus. The team’s experiments pointed to a two-step relay. First, B7-H3 knockdown reduced the phosphorylation of p38 MAPK, a stress-activated kinase that relays signals from the membrane into the cell. Pharmacological inhibition of p38 MAPK reproduced the same molecular changes, lowering both SP1 and HSPA8. Second, the levels of SP1, a well-known transcription factor that binds gene promoters and switches them on, dropped in parallel. Promoter-reporter assays showed that SP1 drives the transcription of the HSPA8 gene, and when the researchers re-introduced SP1 into cells where B7-H3 had been silenced, HSPA8 expression partially recovered, along with the cells’ ability to form colonies and their resistance to apoptosis.

The authors are careful about how far they push this interpretation. Their data, they note, are consistent with p38 MAPK being involved in the pathway, but the phosphorylation and pharmacological experiments do not establish p38 as an obligatory intermediate in a strictly linear signaling cascade from B7-H3 to HSPA8. This kind of caution matters, because signaling networks in cancer cells are notoriously tangled, and a drug that blocks one kinase often triggers compensatory routes. Still, the chain of evidence, from B7-H3 to p38 phosphorylation to SP1 abundance to HSPA8 transcription, offers a coherent mechanistic account of how a surface immune molecule can shape the internal survival machinery of a tumor cell.

Perhaps the most clinically provocative result came from an experiment that stacked the odds against the cancer. When the researchers added HSPA8 knockdown on top of B7-H3 silencing, the malignant phenotypes were suppressed even further than with B7-H3 knockdown alone, suggesting that the two targets act on partially independent vulnerabilities. Translating that idea into a living animal model, the team tested a combination therapy in syngeneic 4T1 tumors, a mouse breast cancer line that recapitulates key features of the human disease. Mice received an antibody against B7-H3, the HSP70-family inhibitor VER155008, both, or neither. Among the four treatment groups, the combination produced the greatest observed reduction in tumor burden, a result that will surely fuel interest in pairing B7-H3-directed immunotherapy with chaperone blockade.

The implications reach beyond triple-negative breast cancer. B7-H3 is overexpressed in a wide range of malignancies, and several B7-H3-targeting agents, including antibodies, antibody-drug conjugates, and CAR-T cell constructs, are already in clinical development for other cancers. If the intracellular program described here, the p38-SP1-HSPA8 axis, operates in other B7-H3-high tumors as well, it could explain why simply blocking the surface molecule has produced mixed results in some trials, and it offers a rationale for rational combinations that attack both the immune-facing and the survival-facing roles of the protein. HSP70 inhibitors, for their part, have struggled to reach the clinic on their own, partly because chaperones are needed by many normal cells; pairing them with a tumor-selective target like B7-H3 could widen the therapeutic window.

There are, of course, substantial gaps between a mouse experiment and an approved regimen. The combination study used a single syngeneic model, and the authors themselves frame combined targeting of B7-H3 and the HSP70 family as a strategy that warrants further investigation rather than a ready-made therapy. Questions remain about toxicity, dosing, biomarkers for selecting patients, and whether the SP1-dependent regulation of HSPA8 holds in human tumors in real time. The study also relied on RNA interference for much of its genetic work, and validation with additional tools such as CRISPR editing or clinical-grade antibodies will be important. Nevertheless, the work exemplifies a growing trend in cancer biology: treating immune checkpoint molecules not merely as shields against T cells but as signaling hubs whose intracellular functions can be drugged.

For patients with triple-negative breast cancer, the study offers something the field badly needs, a new map of dependencies in a disease defined by the absence of targets. By showing that B7-H3 sustains SP1-dependent HSPA8 expression, and that dismantling this axis weakens the cancer’s migratory, invasive, and survival advantages, the researchers have converted a prognostic marker into a mechanistic hypothesis and, potentially, a combination strategy. The next steps, testing chaperone inhibitors alongside B7-H3 antibodies in larger preclinical models and eventually in carefully designed trials, will determine whether this molecular relay can be turned into a clinical weapon. For now, the message is clear: one of immunology’s favorite surface molecules has a secret life inside the tumor cell, and that secret life may be its greatest weakness.

Subject of Research: The role of B7-H3 in sustaining SP1-dependent HSPA8 expression and driving progression in triple-negative breast cancer

Article Title: B7-H3 promotes triple-negative breast cancer progression by sustaining SP1-dependent HSPA8 expression

Article References: Wang, Y., Zhang, F., You, H., Wang, C., Qian, Z., Yang, X., Ning, X., Jiang, Y., Wu, D., Liu, L., Mei, J., Zhao, X., Chen, D., & Zhang, Y. (2026). B7-H3 promotes triple-negative breast cancer progression by sustaining SP1-dependent HSPA8 expression. Cancer Immunology, Immunotherapy. https://doi.org/10.1007/s00262-026-04601-x

Image Credits: AI Generated

DOI: 10.1007/s00262-026-04601-x

Keywords: triple-negative breast cancer, B7-H3, CD276, HSPA8, HSP70, SP1, p38 MAPK, VER155008, immunotherapy, chaperone proteins, cancer signaling, combination therapy

The study, led by Haibin Xia of Shaanxi Normal University in Xi’an and published in Cellular and Molecular Life Sciences, centers on a protein called FAM76B, short for family with sequence similarity 76 member B. Compared with famous immune molecules, FAM76B has been an obscure figure, its functions largely unmapped. The researchers suspected it might matter for macrophage behavior, and their experiments revealed something striking: when FAM76B was absent, macrophages could still engulf apoptotic cells normally, but they could not finish the job. The swallowed corpses piled up inside the cells, undigested. In other words, the defect was not in the eating but in the digesting, a selective failure in the post-phagocytic processing of apoptotic material.

That distinction matters enormously. Engulfment is only the first half of efferocytosis; the second half requires the macrophage’s internal degradation systems to break down the ingested cargo. The cell’s primary self-cleaning and cargo-disposal pathway is autophagy, a process in which cellular material is sequestered into membrane-bound vesicles and delivered to destructive enzymes. The team found that loss of FAM76B crippled autophagic activity in macrophages, leaving them unable to properly degrade the apoptotic cells they had taken in. Because undegraded cellular debris is a well-known trigger of anti-nuclear antibodies and other hallmarks of lupus, the finding immediately suggested a mechanistic bridge between a single protein and systemic autoimmunity.

To understand how FAM76B controls autophagy, the researchers delved into the world of RNA modifications, one of the hottest areas in molecular biology. Messenger RNA molecules carry chemical tags that influence their stability and translation, the most common being N6-methyladenosine, or m6A. This tag is read and interpreted by specialized m6A-binding proteins. One such reader is HNRNPA2B1, which the team discovered was being physically restrained by FAM76B. In the presence of FAM76B, HNRNPA2B1 stays out of the cytoplasm, away from its mRNA targets. But when FAM76B is missing, HNRNPA2B1 translocates into the cytoplasm, where it can bind m6A-modified transcripts and stabilize them.

One transcript in particular proved consequential: the mRNA encoding ISG15, an interferon-stimulated gene famous for its roles in antiviral defense and inflammation. With HNRNPA2B1 free to act, ISG15 mRNA was stabilized and ISG15 protein levels climbed. Here the story took another unexpected turn. ISG15 is best known for ISGylation, a process in which it is covalently attached to other proteins, modifying their behavior. Yet the researchers showed that the suppression of autophagy they observed was independent of ISGylation. Instead, ISG15 acted through a noncanonical mechanism: it physically interacted with HSPA8, a molecular chaperone, and disrupted HSPA8’s association with BECN1, also known as beclin 1, a core component of the autophagy-initiation machinery. Severing that partnership throttled the autophagic pathway at its source.

The team then moved from cells to living animals to test whether this molecular cascade mattered for disease. Using a pristane-induced model of lupus-like disease, they compared normal mice with mice lacking Fam76b. The deficient animals fared markedly worse, developing exacerbated disease characterized by defective clearance of apoptotic cells, elevated ISG15 expression, and reduced autophagy in their macrophages. The picture that emerged was coherent: without FAM76B, the m6A reader ran loose, ISG15 accumulated, autophagy stalled, dead-cell debris accumulated, and the immune system tipped toward autoimmunity.

Crucially, the researchers did not stop at demonstrating the damage. They tested whether intercepting the pathway could undo it. When they targeted ISG15 for inhibition in the disease setting, the intervention partially restored autophagic activity, improved the macrophages’ ability to process apoptotic cells, and alleviated disease manifestations in the mice. This therapeutic experiment elevates the finding from descriptive to actionable. It suggests that ISG15, or the molecular interactions it makes with HSPA8 and BECN1, could serve as a druggable node for treating lupus and potentially other conditions in which faulty efferocytosis fuels inflammation.

The study is notable for how many layers of cell biology it stitches together into a single regulatory axis. At the top sits FAM76B, a nuclear gatekeeper controlling where HNRNPA2B1 can operate. HNRNPA2B1, in turn, interprets m6A marks on RNA to determine the lifespan of specific transcripts, including ISG15. ISG15 then reaches into the autophagy machinery through a protein-protein interaction with the chaperone-mediated arm of the pathway. Each link in this chain, from RNA modification to chaperone sequestration to vesicle formation to debris clearance, represents an area of intense research interest in its own right, and the demonstration that they form a connected circuit in macrophages offers investigators a map of previously hidden connections.

For patients with systemic lupus erythematosus, the work offers a fresh explanation for a long-recognized puzzle. Defective clearance of apoptotic cells has been implicated in lupus for decades, and elevated type I interferon activity, the very signal that drives ISG15 expression, is a hallmark of the disease. What has been missing is a precise account of how these observations connect at the molecular level. By tracing a continuous pathway from an RNA-binding gatekeeper through a chaperone-disrupting interferon-stimulated protein to the autophagy engine, the Xi’an team has supplied a candidate answer, and one that comes with an obvious intervention point.

Caution is warranted, as it always is in translational science. The findings derive from cell culture experiments and a chemically induced mouse model, and lupus in humans is far more heterogeneous, involving multiple immune cell types, genetic risk factors, and environmental triggers. Whether FAM76B levels or ISG15 activity reliably track with disease in patients, and whether blocking ISG15’s interaction with HSPA8 is feasible and safe in people, remain open questions. ISG15 also performs beneficial antiviral functions, so any therapeutic strategy would need to navigate the balance between taming autoimmunity and preserving antiviral defense. Nevertheless, the identification of a FAM76B-HNRNPA2B1-ISG15-HSPA8 axis linking RNA modification to autophagy and efferocytosis gives lupus research a new set of molecular levers to pull, and it reminds us that sometimes the key to understanding a complex disease lies in a protein whose job, until now, nobody had bothered to read.

Subject of Research: The role of the FAM76B protein in regulating macrophage efferocytosis and systemic lupus erythematosus through an m6A-ISG15-autophagy axis

Article Title: FAM76B controls macrophage efferocytosis via m6A–ISG15–autophagy axis

Article References: Weng, G., Zhang, N., Wang, L., Liu, S., Zhao, W., Cheng, J., Yang, Y., Tang, Z., Zhao, J., Yang, P., Mao, Q., & Xia, H. (2026). FAM76B controls macrophage efferocytosis via m6A–ISG15–autophagy axis. Cellular and Molecular Life Sciences. https://doi.org/10.1007/s00018-026-06419-3

Image Credits: AI Generated

DOI: 10.1007/s00018-026-06419-3

Keywords: FAM76B, efferocytosis, macrophages, autophagy, m6A, ISG15, HNRNPA2B1, HSPA8, systemic lupus erythematosus, RNA modification, apoptotic cell clearance, autoimmune disease