Allergies show a small but significant link to later cancer risk

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Jul 5 2026

A new meta-analysis suggests allergies are not broadly protective against cancer, with asthma and regional differences shaping a small but important risk signal.

Study: Association between allergies and cancer incidence: a systematic review and meta-analysis. Image Credit: Belhadef.nabil / Shutterstock

In a recent Article in Press study published in the journal Scientific Reports, a group of researchers evaluated the association between allergic diseases and subsequent cancer incidence using cohort-based evidence from multiple countries.

Could a condition as common as an allergy influence the likelihood of developing cancer? An estimated 10%-30% of the world's population experiences allergic disorders, while at the same time, cancer remains a leading cause of suffering and death in both sexes.

There has been much debate among scientists over whether allergies actually assist the body's immune system in recognizing and removing cancer cells, or whether the long-standing allergic inflammation increases the risk of developing cancer. Different allergic conditions trigger distinct immune responses, further complicating this relationship.

Understanding how allergies and cancer are connected could improve public health strategies and guide future research. 

About the study

The researchers conducted a systematic review following the International Prospective Register of Systematic Reviews (PROSPERO) protocol (registration number CRD42024566792). A search for cohorts published between 1999 and 2024 was carried out using PubMed to identify eligible studies. To ensure no relevant evidence was missed, researchers checked the reference lists of selected articles using Google Scholar and Scientific Electronic Library Online (SciELO).

The study included patients who were diagnosed with allergies before they developed cancer, ensuring that allergic diseases preceded cancer diagnosis. Acute allergic reactions, studies involving special patient populations, and studies without cohort designs were excluded. Both solid tumors and hematological cancers were considered as outcomes.

Researchers extracted study characteristics, including country, study design, participant demographics, sample size, allergy type, cancer type, effect estimates, and 95% confidence intervals (CIs). The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of the studies, and Rayyan was used to screen the literature.

The meta-analysis was performed using Stata 18.0 statistical software, with a random-effects model. Effect estimates, such as odds ratios (ORs), hazard ratios (HRs), relative risks (RRs), and standardized incidence ratios (SIRs), were transformed to the log scale, with HRs, RRs, and SIRs treated as approximations of ORs under the rare-outcome assumption before the statistical analysis. Subgroup analyses examined differences by geographic region, allergy type, and cancer type, while I² statistics and Egger's test were used to evaluate heterogeneity and publication bias.

Study results

The systematic review included 28 eligible studies conducted across the United States, Sweden, Taiwan, the United Kingdom, Denmark, Australia, Finland, and South Korea. Most were cohort studies, with two nested case-control studies included because they met the predefined eligibility criteria.

The studies evaluated a broad range of allergic conditions, including asthma, hay fever, atopic dermatitis, and atopic allergic conditions, alongside multiple cancer types affecting different organs and tissues.

All the included studies were rated as moderate or high quality using the NOS, although substantial heterogeneity, possible publication bias, and subgroup limitations meant the pooled evidence required cautious interpretation.

The primary meta-analysis showed an overall weak but statistically significant association between allergic disease and cancer incidence. The combined OR for individuals with allergic disease and incidence of any cancer was 1.07 (95% CI: 1.03-1.11), which indicates that having allergic disease is associated with a small increased overall cancer risk. Although the effect size was limited, the findings suggested that allergies were not consistently protective against cancer across the included populations.

Examination of the funnel plot revealed visual asymmetry, suggesting publication bias could not be completely excluded, although Egger's test did not identify a statistically significant small-study effect.

Allergies were statistically associated with cancer incidence in the Western Pacific region (OR 1.65, CI: 1.22-2.21), whereas no statistically significant association was found between allergies and cancer in the Americas or Europe. These regional differences suggested that the relationship between allergies and cancer may vary across populations and geographic settings, although the Western Pacific subgroup was based on a limited number of studies and may reflect population, environmental, or cancer-type differences.

An additional subgroup analysis of cancer type found a relationship between allergies and the incidence of the following cancers: lung cancer, hematological cancers, kidney cancer, testicular cancer, and thyroid cancer. However, colorectal cancer and uterine cancer showed an inverse relationship (negative association) with allergies.

Several other cancer types, including breast, ovarian, prostate, skin, cervical, liver, endocrine, and upper aerodigestive tract cancers, were also evaluated, although not all demonstrated statistically significant associations. Furthermore, many of the type-specific analyses were small; therefore, the findings should be interpreted cautiously.

When considering allergy type alone, only asthma showed a statistically significant association with cancer incidence (OR 1.18, 95% CI: 1.10-1.28), whereas atopic dermatitis, hay fever, and atopic diseases showed no statistically significant association.

However, the authors noted that the asthma signal could partly reflect allergy severity or treatment-related confounding, particularly corticosteroid use, rather than allergic pathophysiology alone. 

The analysis also exhibited heterogeneity across studies due to differences in populations studied and in the definitions of allergy and cancer outcomes. 

Therefore, the findings suggest that the relationship between allergy and cancer is complex and varies by type of allergy and geographic area, with no consistent pattern across regions, allergy subtypes, and cancer outcomes.

Conclusion

The findings indicated a statistically significant but modest association between allergic diseases and overall cancer incidence. Although the overall increase in risk was small, important differences emerged across geographic regions and allergy types, with the strongest association observed in the Western Pacific region and among individuals with asthma.

The variation in results across cancer-specific evaluations, as well as the limited number of studies in several subgroups, highlighted the need for further research to identify subgroups at increased risk and to better define the relationship between allergy and cancer, rather than encouraging an immediate change in current clinical practice.