MUC5B Risk Allele Carries an Unexpected Survival Signal
A promoter variant in MUC5B long tied to idiopathic pulmonary fibrosis now sits at the center of a more unusual question: why would a common allele that raises the risk of a progressive, often fatal lung disease remain frequent in the population? In a large analysis of the All of Us Research Program, the rs35705950 T allele tracked with a markedly higher hazard of IPF, yet it also aligned with lower all-cause mortality and modestly better survival in later life. The pattern points to a biological trade-off rather than a simple disease gene.
The variant is common. The disease is not.
That imbalance has fueled speculation for years. The T allele occurs in roughly 20% of the general population and is present in at least half of people with IPF. In this cohort, the allele carried an adjusted hazard ratio of 4.0 for IPF diagnosis, with a 95% confidence interval of 2.7 to 5.7. Those estimates held after adjustment for smoking, sex, and predicted ancestry, preserving the core association in a population that had not yet been selected on the basis of having established fibrotic lung disease.
Population-Level Data Reframe the Allele’s Clinical Meaning
Previous work concentrated on patients who already had IPF, a setting vulnerable to survival bias and competing-risk distortion. By starting with 232,401 participants in All of Us and examining who later developed disease, this analysis shifted the question upstream. Participants carrying the T allele were diagnosed at a younger age than those with GG alleles, a result that contrasts with earlier clinic-based observations in which carriers appeared older at diagnosis. The difference likely reflects the advantage of studying people before the disease itself has reshaped who enters the sample and when.
The statistical signal remained visible across approaches. A Fine-Gray competing risks analysis produced an IPF hazard ratio of 3.9 with a P value below 0.001. The consistency matters because IPF is uncommon, and small shifts in how competing events are handled can alter the apparent direction of an association. Here, the risk estimate did not disappear when the model absorbed the usual clinical confounders that often shadow fibrotic lung disease.
Mortality Trends Tilted in the Opposite Direction
All-cause mortality moved the other way. Participants with the T allele had a lower adjusted hazard of death, with an estimate of 0.91 and a 95% confidence interval of 0.83 to 0.999. The P value for the fully adjusted model was 0.048, a narrow margin but still statistically significant. Kaplan-Meier curves separated late in life, with the T allele group showing at least a 1% survival advantage between ages 55 and 78 and at least a 2% advantage between ages 79 and 95.
Late-life survival differences of this size are small at the individual level and meaningful at the population level. A 2% survival difference across tens of thousands of people changes the balance between a deleterious pulmonary phenotype and a broader protective effect elsewhere in the body. The analysis also cites a UK Biobank finding that serendipitously linked the same allele to improved male survival, adding external support to the direction of the association, even if the precise mechanism remains unproven.
Airway Biology Offers a Plausible Explanation
The most plausible explanation sits in the airways rather than the fibrotic lung. MUC5B contributes to mucociliary transport, which clears inhaled particles and pathogens from the respiratory tract. The T allele increases MUC5B expression, creating a mechanistic path by which thicker or more effective mucus handling could improve host defense. In a world where respiratory infection has historically driven mortality, even a modest advantage in airway clearance could be selected over time despite the cost of increased fibrotic risk later in life.
That hypothesis is biologically attractive because it links a common variant to two very different phenotypes: early protection against infection and later susceptibility to fibrosis. The gene does not need to be uniformly beneficial to persist in the population. A variant that improves survival through midlife and beyond can remain common even if it raises the risk of disease in a smaller subset of people.
The authors tested the respiratory-infection idea against available data. Prior studies in UK Biobank and the Million Veterans Program had suggested fewer COVID-19–related hospitalizations in carriers. In the All of Us analysis, the T allele was not associated with a lower risk of COVID-19 itself, but it did associate with reduced hospitalization risk. No difference emerged for the combined group of all lung infections or accompanying hospitalizations. The result is narrower than a broad infection-protection model, yet it still fits a version of the airway-defense hypothesis in which severity rather than susceptibility changes.
What the Trade-Off Means for IPF Genetics
For clinicians, the finding does not change how IPF is diagnosed or treated today. It does, however, sharpen the view of MUC5B as more than a risk locus for fibrosis. The allele may sit in a zone of evolutionary compromise, where the pulmonary penalty of fibrosis is outweighed by a longer survival curve in the general population. In practical terms, a gene can be a liability in one disease context and still leave a net survival imprint across a lifetime.
The paper also gives a concrete population example. In 10,000 people carrying GG alleles versus 10,000 carrying the T allele, the T allele group would be expected to include about 10 additional people with IPF, based on an incidence difference of roughly 9.8 per 10,000. The same group would also be expected to include about 200 additional people surviving to age 79, using the reported 2% survival difference at that age. The arithmetic makes the evolutionary argument visible without overstating certainty.
Limits That Shape How Far the Result Can Be Read
Several constraints keep the inference grounded. Death counts were small and lacked cause-of-death data. IPF diagnoses came from electronic health records rather than centralized adjudication. The analysis was restricted to participants of European or Admixed-American ancestry because of database reporting rules, leaving other ancestries untested in this dataset. Enrollment also began at age 18 or older, so any advantage confined to childhood or early adulthood would not appear here.
Those limits matter because the finding is about survival, not just disease risk. A lower hazard of death can reflect any number of pathways, and without cause-of-death detail, the analysis cannot separate respiratory from nonrespiratory mechanisms. Even so, the central observation stands on its own: MUC5B rs35705950 increases IPF risk while tracking with lower all-cause mortality in this large general-population cohort. The next question is not whether the allele matters, but which physiologic systems pay the price and which ones gain the advantage.