Flu and COVID-19 can 'wake up' dormant breast cancer cells in the lungs

A paper published in Nature connects infectious diseases and oncology with a direct thread: common respiratory viruses - influenza and SARS-CoV-2 - are able to "wake up" disseminated breast cancer cells that have been dormant in the lungs for years after successful treatment. Using mouse models, the authors showed that just a few days after infection, such cells lose their "dormant" phenotype, begin to divide, and in two weeks develop metastatic foci. The key to the switch is the inflammatory mediator interleukin-6 (IL-6). Analysis of the UK Biobank and the Flatiron Health database added a human context: cancer "survivors" who had COVID-19 had an almost twice as high risk of dying from cancer, and patients with breast cancer had a higher risk of subsequent detection of metastases in the lungs.

What exactly did they do?

• We modeled “dormant” disseminated cells (DCC) of breast cancer in the lungs on the MMTV-Her2 line: single HER2⁺ cells maintain a “quiet” mesenchymal phenotype for years and almost do not divide. Then we infected mice with influenza A virus or mouse-adapted SARS-CoV-2 MA10 and tracked the fate of these cells over time.
• “Awakening” was measured by an increase in the number of HER2⁺ cells, the appearance of the division marker Ki-67, and a shift from mesenchymal features (vimentin) to more epithelial ones (EpCAM).
• We repeated the experiment in Il6-knockout mice to test the causal role of IL-6, and analyzed the immune “background” in the lungs – what CD4⁺ and CD8⁺ T cells do after infection.
• In the “human part”, two databases were studied: UK Biobank (survivors of various cancers) and Flatiron Health (36,845 patients with breast cancer) to understand how the history of COVID-19 correlates with the risk of death and pulmonary metastases.

Key results and figures

• In mice: "awakening" in days. After influenza and after SARS-CoV-2, the number of HER2⁺ cells in the lungs increases stepwise by days 3 and 9 and strongly by day 28; the proportion of Ki-67⁺ (dividing) cells increases; the phenotype shifts from "quiet" mesenchymal to proliferative. All these transitions depend on IL-6: in Il6-KO mice, there is almost no "rise", although the virus itself replicates in the lungs comparable.
• The immune “architecture” is against us. In the post-viral period, CD4⁺ T cells paradoxically support the metastatic burden by suppressing the activation and cytotoxicity of CD8⁺ cells; DCCs themselves also interfere with full T-cell activation in the pulmonary microenvironment.
• In humans: risk signal after COVID-19. In the UK Biobank, among cancer patients diagnosed in the distant past (≥5 years before the pandemic), a positive SARS-CoV-2 PCR was associated with increased mortality:
  - from all causes: OR 4.50 (95% CI 3.49-5.81);
  - non-COVID mortality: OR 2.56 (1.86-3.51);
  - cancer mortality: OR 1.85 (1.14-3.02).
  The effect was maximal in the first months after infection (in the short observation window, the OR for cancer mortality jumped to 8.24), then significantly weakened. In Flatiron Health, among women with breast cancer, a history of COVID-19 was associated with an increased risk of a subsequent diagnosis of lung metastases: HR 1.44 (1.01-2.05).

Why is this important?

• A new mechanism for relapse. The work shows that “normal” lung inflammation from viruses may be the very trigger that turns off the dormancy program in single tumor cells and unties their hands for growth. This partly explains the excess cancer mortality of the first years of the pandemic, which is not limited to delays in screening and treatment.
• Precise target and time window. The IL-6/STAT3 signaling axis appears to be critical precisely in the early phase after infection, suggesting that potential preventive interventions should be time-sensitive and targeted.

What this might mean in practice

• For cancer survivors
  • Prevention of respiratory infections (vaccination against influenza and COVID-19 according to recommendations, seasonal caution, timely treatment) takes on additional meaning - this is not only protection against severe course, but also a potential reduction in cancer risk in the coming months after the disease.
  • In the case of a previous infection, it makes sense to increase oncovigilance in the short “post-infection” window (for example, do not postpone follow-up visits/examinations if they are already indicated according to the plan).
• For physicians and health systems:
  • There is reason to consider risk stratification in cancer survivors who have recently had a viral infection and to test targeted anti-inflammatory prophylaxis in clinical trials (including with IL-6 blockade), taking into account the risks and contraindications.
  • It is important not to generalize the findings to everyone and everything: we are talking about risk groups and a clear time interval, and not about chronic suppression of inflammation.

How does this compare to previous data?

It has been argued before that inflammation is a “push” for metastasis; the pandemic has provided a unique “natural” test of the hypothesis. The new paper connects the causal mouse experiment with real cohorts and points to IL-6 as the central node. The popular retelling by Nature itself and the specialized media emphasize the same connection between mechanism and epidemiology.

Restrictions

• Mouse models are not equivalent to humans: the doses of the virus, the timing and the scale of the effect cannot be directly transferred.
• UK Biobank and Flatiron are observational: there are possible residual confounding factors (unaccounted for infections in “negatives”, differences in access to care, testing, vaccination).
• Focus is on breast cancer and lung metastases; other tumors/organs require separate testing. However, the consistency of signals increases confidence in the overall model.

What's next?

• Clinical trials of time-sensitive strategies in cancer survivors of respiratory infections: from IL-6 blockers to “enhanced surveillance” protocols in the first months.
• Refinement of biomarkers of awakening (IL-6, DCC transcriptional signatures, lung immune profiles) and mapping of risk windows by time after infection.
• Testing whether the mechanism extends to other tumors and other triggers of lung inflammation.

Source: Chia, SB, Johnson, BJ, Hu, J. et al. Respiratory viral infections awaken metastatic breast cancer cells in lungs. Nature (2025). (Online 30 July 2025). Key mechanistic and epidemiological findings, including the role of IL-6, UK Biobank and Flatiron Health risk assessments, are reported in the original article and further discussed in the Nature editorial.https://doi.org/10.1038/s41586-025-09332-0