chatGPT:
Paper
Baker, Ansere, Sanqui & Benayoun, “The problem with one-size-fits-all medicine: Biological sex and the aging immune system”, PLOS Biology, 2026. This is an essay/review, not a primary experimental paper. Its central claim is that immune aging is shaped not only by age and biological sex separately, but by nonlinear sex-by-age interactions that are often ignored in research, trials, dosing, and clinical guidance.
Summary
The paper argues that aging produces shared immune decline in both sexes—immunosenescence and inflammaging—but also produces divergent trajectories in males and females. Females generally show stronger adaptive immune activity and cytokine responses, partly because of X-linked immune genes, X-inactivation escape, and estrogen effects. Males tend to rely more on innate immune responses with age, and androgen decline may contribute to altered inflammatory and vaccine responses.
A major theme is that these differences are not linear across life. In women, menopause is presented as a key biological inflection point, with abrupt estrogen decline and changes in T-cell dynamics, including post-menopausal TH1 bias and nonlinear regulatory T-cell changes. The authors stress that menopause timing varies widely, so using chronological age without menopause status can obscure real immune-aging biology. In men, testosterone decline is described as more gradual, but still biologically relevant for immune aging.
The paper then applies this framework to several clinical areas:
Infectious disease: older people are generally more vulnerable to infection and severe outcomes, while males often have worse infection outcomes than females. The paper highlights examples where sex differences emerge only after certain ages, suggesting interaction rather than simple additive effects.
Autoimmunity: females have higher overall autoimmune disease risk, but the age profile differs by disease and sex. The authors connect this to stronger female immune activation, menopause, and inflammaging.
Cancer: cancer risk rises with age, and males have higher incidence and mortality for many non-reproductive cancers. The authors argue that sex-by-age patterns in cancer incidence may partly reflect hormonal and immune shifts.
Vaccines: younger people and females generally mount stronger vaccine responses but also more adverse events. In older adults, sex differences in antibody response can widen with age; booster dosing may reduce such disparities. The paper discusses SARS-CoV-2, influenza, HPV, and pneumococcal vaccination examples.
Cancer immunotherapy: age can reduce benefit and increase immune-related adverse events; sex can also influence efficacy and toxicity. The authors note that interaction between age and sex is under-studied in immunotherapy.
Drug response and adverse events: the paper argues that age and sex affect pharmacokinetics and pharmacodynamics. Females often have higher blood concentrations and longer elimination for many drugs, which may contribute to higher adverse drug reactions. The paper gives examples of drugs withdrawn or relabelled due to female-biased toxicity, including zolpidem, cisapride, terfenadine, troglitazone, and others.
The authors conclude with recommendations: include both sexes in preclinical work, report sex and age clearly, record sex chromosome karyotype in cell studies, document hormonal and reproductive histories in clinical trials, use factorial or stratified designs, and apply nonlinear statistical models such as mixed-effects models, generalized additive models, and piecewise regression. They also call for post-marketing surveillance to test sex-by-age interaction effects.
Claimed novelty
The novelty is not a new dataset or experiment. The contribution is conceptual synthesis.
The main novel element is the insistence that sex and age should be treated as interacting, nonlinear biological variables, rather than as simple covariates to adjust away. The paper’s Figure 1 makes this explicit: female immune aging is depicted with a menopause-associated inflection, while male decline is shown as more gradual; linear models are presented as misleading simplifications.
A second novelty is the paper’s broad integration of immune aging with real-world clinical consequences: infection severity, autoimmunity, cancer risk, vaccine immunogenicity, vaccine adverse events, immunotherapy outcomes, general pharmacology, NSAIDs, glucocorticoids, and TNF inhibitors. This makes the argument more clinically grounded than a standard immunology review.
A third useful contribution is the paper’s practical methodological agenda. It does not merely say “include sex as a variable”; it recommends specific design and analysis approaches: sex-by-age powered trials, menopause and hormonal-history recording, sex chromosome karyotyping in cell studies, and nonlinear models to capture breakpoints.
Critique
The paper is persuasive as a position essay, but its evidentiary strength varies by section. Some claims are well supported by large epidemiological or pharmacovigilance patterns, but others rely on examples from different disease areas, vaccines, and drug classes that may not share the same mechanisms. The review sometimes moves from “sex and age are associated with different outcomes” to “immune aging explains these outcomes” more strongly than the evidence always allows.
A key limitation is that biological sex, hormones, reproductive history, body composition, kidney function, drug exposure, reporting behavior, and gendered healthcare behaviour are difficult to disentangle. The authors acknowledge that gender may influence immunity and adverse-event reporting, but they deliberately restrict the essay to biological sex. That keeps the argument focused, but it may understate how much real-world clinical outcome data are shaped by non-biological factors.
The menopause argument is biologically plausible and important, but the paper risks making menopause the dominant explanatory breakpoint for female immune aging. Menopause timing, hormone therapy, parity, contraceptive history, surgical menopause, autoimmune status, obesity, infection history, and socioeconomic factors can all modify immune trajectories. The recommendation to collect these data is strong, but the paper does not fully solve how trials should remain feasible once all these covariates are included.
The clinical pharmacology section is highly relevant, especially for older adults and polypharmacy, but “female-biased adverse events” can reflect multiple causes: pharmacokinetics, pharmacodynamics, prescribing patterns, dose selection, body composition, renal function, and reporting bias. For example, using sex-specific dosing may help in some cases, but the paper could better distinguish when sex itself is the right dosing variable versus when measured parameters such as creatinine clearance, lean mass, fat mass, hormone status, or drug level monitoring are better.
The paper’s recommendations are sensible but demanding. Properly powered sex-by-age interaction studies require much larger sample sizes. In aging biology, especially animal studies and early clinical trials, this may be expensive and slow. The authors are right that oversimplified studies can mislead, but a stronger framework for prioritising when sex-by-age modelling is essential would make the recommendations more operational.
Overall assessment
This is a useful and timely essay. Its strongest message is that immune aging is not a single curve shifted up or down by sex. It is a set of changing trajectories shaped by chromosomes, hormones, reproductive transitions, age, disease context, and treatment exposure. The paper is most valuable as a methodological warning: studies that “adjust for age and sex” may miss the biology that matters most.
The main weakness is that it is more of a broad manifesto than a mechanistic or quantitative synthesis. It would be strengthened by formal meta-analysis, clearer grading of evidence, and more precise decision rules for when clinicians should use sex-specific, age-specific, hormone-status-specific, or biomarker-guided interventions.