A new study published in Cell suggests that severe viral pneumonia—including severe COVID-19—may leave behind lasting changes in the lung that increase the risk of future lung cancer or accelerate tumor growth if cancer emerges.
The research, led by investigators at Duke University studying how infections shape immune responses in the lung, integrates large-scale human data with mechanistic studies in mice. Together, the findings suggest that severe respiratory infections can leave a long-lasting “memory” in lung tissue that alters the immune environment in ways that favor tumor development.
“Our data suggest that severe COVID-19 can reprogram the lung environment in a way that supports tumor growth later on,” said senior author Jie Sun, PhD. “Even after the virus is gone, the immune and structural cells in the lung retain an epigenetic imprint that can promote cancer progression.”
Evidence from patients and animal models
To explore whether severe viral infections influence cancer risk, the researchers first analyzed large human health datasets. They found that patients who had been hospitalized with severe COVID-19 showed a significantly increased rate of subsequent lung cancer diagnoses in the years following infection.
“In our epidemiological analysis, we saw that patients with prior severe COVID had about a 24 percent higher incidence of lung cancer diagnosis over the next four years,” Sun said. “We were surprised by the magnitude of that signal.”
The study examined an exceptionally large dataset, the Epic Cosmos database, including nearly 900,000 individuals hospitalized with severe COVID-19 and tens of millions of control subjects. Because lung cancer remains a relatively rare event, such scale was necessary to detect the association.
However, the human analysis alone could not determine whether severe infection directly contributed to cancer development or whether other factors—such as preexisting disease—might explain the association.
To address this question, the investigators turned to mouse models designed to mimic the sequence of events seen in patients.
In these experiments, animals were first infected with respiratory viruses to induce severe pneumonia. After the infection cleared, the researchers introduced oncogenic drivers known to initiate lung cancer. Across multiple models, tumors grew significantly faster in animals that had previously experienced severe viral infection.
An epigenetic “scar” in the lung
The researchers traced this effect to persistent changes in lung cells following infection.
Severe viral pneumonia triggered long-lasting chromatin remodeling in both immune cells and structural cells in the lung. These epigenetic changes altered how genes involved in inflammatory signaling were regulated. Sun describes the process as an “epigenetic scar.”
“All of our cells have the same DNA, but different genes are turned on or off depending on the cell type,” he said. “What we found is that severe infection changes the epigenetic state of lung cells, creating a memory that drives production of pro-tumor factors once cancer begins.”
These changes reshaped the tumor microenvironment in ways that favored tumor growth.
In particular, the lung accumulated high numbers of tumor-associated neutrophils—immune cells that can suppress anti-tumor responses. At the same time, cytotoxic CD8 T cells, which normally attack cancer cells, were reduced in both number and activity.
“The infection primes the lung so that when tumors arise, there are more pro-tumor immune cells and fewer anti-tumor T cells,” Sun said. “That allows the tumor to grow faster and escape immune control.”
Importantly, infection alone did not initiate cancer. Instead, it acted as a cooperating factor that accelerates tumor progression when oncogenic mutations—such as KRAS mutations common in lung cancer—are already present.
“Infection itself does not cause the mutation,” Sun said. “But if a cell already carries a mutation, the altered immune environment can help that mutated cell grow into a tumor.”
Potential therapeutic strategies
The study also explored ways to counteract this infection-induced tumor promotion.
The investigators found that blocking neutrophil recruitment through inhibition of the chemokine receptor CXCR2 reduced tumor growth in virus-experienced lungs. When combined with immune checkpoint blockade targeting PD-L1, the treatment restored CD8 T cell activity and further suppressed tumor progression.
While these experiments were conducted in mice, the findings suggest potential strategies for treating lung cancers that arise in patients with prior severe viral infection.
“We identified a couple of therapeutic approaches in our animal models that can reverse this tumor-promoting environment,” Sun said. “The next step is to determine whether the same mechanisms occur in human lung cancer.”
Implications for prevention
Beyond therapy, the study also highlights the potential importance of preventing severe respiratory infections.
Notably, vaccination prevented infection-enhanced tumor growth in animal models. Because vaccines reduce the likelihood of severe disease, they may indirectly reduce the long-term cancer-promoting effects of viral pneumonia.
“The key message is that severe infections should not be considered harmless once the virus is cleared,” Sun said. “The infection may be gone, but its impact on the body can persist.”
Sun emphasized that acute infections can leave long-lasting biological consequences, a concept increasingly recognized in conditions such as long COVID and post-acute infection syndromes.
“Historically, many people assumed that once an infection resolves, the body simply returns to normal,” he said. “But we now know that severe infections can leave durable changes in tissues.”
Next steps
The research team now plans to investigate whether similar epigenetic changes occur in the lungs of patients who recovered from severe COVID-19.
They are also seeking clinical collaborations to compare lung tumors from patients with and without prior severe viral infection.
“We want to determine whether lung cancers that develop after severe infection have distinct biological features,” Sun said. “If they do, that could guide how we monitor patients and how we treat those tumors.”
For clinicians and researchers, the findings underscore the complex interplay between infectious disease and cancer biology. “We should take severe infections seriously,” Sun said. “Preventing them—through vaccination and effective treatment—may also help prevent long-term diseases, including cancer.”
