Study links specific inflammatory markers to increased cancer risk

By Neha MathurFeb 5 2024Reviewed by Lily Ramsey, LLM

In a recent study published in eBioMedicine, researchers evaluate the causal relevance of circulating inflammatory markers in cancer risk using a two-stage Mendelian randomization (MR) design and a quasi-experimental approach.

Study: Association between circulating inflammatory markers and adult cancer risk: a Mendelian randomization analysis. Image Credit: Gorodenkoff/Shutterstock.com

Background

Preclinical studies have shown that pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukins (ILNs) 1/6, are associated with site-specific cancers.

Accordingly, their transcription factors (TFs), such as nuclear factor kappa B (NF-kB) and signal transducer and activator of transcription 3 (STAT3), up-regulate during cancer cell proliferation, invasion, and metastasis.

Furthermore, clinical trials suggest that targeting these inflammatory markers by drugs may, thus, be effective in cancer prevention.

However, the relevance of these data is unclear, given the susceptibility of observational studies to confounding and reverse causation and the limited translational utility of preclinical studies to humans.

So far, MR studies have not comprehensively evaluated different classes of circulating inflammatory markers across adult cancers, focusing only on one marker or a specific class in one type of cancer.

Therefore, the causal nature of the relationships between circulating inflammatory markers across adult cancers and the suitability of these markers as chemotherapy targets is unclear.

About the study

In the present systematic MR analysis, researchers meta-analyzed data from six genome-wide association studies (GWASs) comprising ~60,000 participants of European ancestry.

They then generated estimates of their effects (causal) on the risk of 30 types of adult cancers using inverse-variance weighted random-effects models. 

Of 204 inflammatory markers for which genetic association data was available, 88 were meta-analyzed, but only 45 showed significant associations.

Across studies, the median Pearson correlation coefficient was 0.66 when using a stricter P-value threshold, and 93.4% of single nucleotide polymorphisms (SNPs) showed consistent effects, which helped construct genetic instruments proxying 66 circulating inflammatory markers.

These cis-acting SNPs were within ±250 kb (proximity) from the gene encoding the relevant marker, as observed in linkage disequilibrium (LD).

Previously reported associations between inflammatory markers and cancer with an overall association score of ≥0.05 were included in the validation MR analyses.

The authors additionally performed a "hypothesis-free" pan-cancer assessment to identify potential new associations between inflammatory markers and cancer.

The researchers used F-statistics to evaluate the relevance of the instruments. Further, they performed colocalization analysis to identify potential horizontal pleiotropy and determine shared genetic variants across inflammatory markers and cancer outcomes, where a posterior probability (PPH4) >70% indicated support.

Likewise, a threshold of <0.05 and between ≥0.05 and <0.20 in Benjamini-Hochberg false discovery rate (FDR) correction indicated strong and suggestive evidence, respectively. Finally, the team replicated and pooled these findings in the FinnGen study.

Results

The authors found consistent evidence for the potential association of four genetically proxied inflammatory markers (pro-adrenomedullin, interleukin-23 receptor, prothrombin, and interleukin-1 receptor-like 1) in the risk of four site-specific cancers. 

There were positive associations between pro-adrenomedullin concentrations and breast cancer risk and interleukin-23 receptor concentrations with pancreatic cancer risk, with respective ORs and PPH4s of 1.19, 84.3%, and 1.42, 73.9%, respectively.

On the contrary, there were inverse associations between interleukin-1 receptor-like 1 concentration with triple-negative breast cancer risk and prothrombin concentrations with basal cell carcinoma risk, with respective ORs and PPH4s of 0.92, 85.6%, and 0.66, 81.8%, respectively. 

Further, the authors found suggestive evidence for an association of genetically proxied macrophage migration inhibitory factor (MIF) concentrations with bladder cancer risk, with OR and PPH4s of 2.46 and 76.1%.

Accordingly, the researchers could not replicate or pool these findings with the FinnGen study. 

Intriguingly, there was minimal evidence for the association of circulating inflammatory markers with cancer risk for 22/30 cancer outcomes examined in this study, likely because 13 of 30 outcomes were restricted to <10,000 cases.

Conclusions

The authors emphasized the need to replicate and validate these findings, particularly for novel associations identified between inflammatory markers and cancer risk.

Subsequently, these markers may be prioritized for further evaluation as potential pharmacological targets for cancer prevention.

Nonetheless, this comprehensive and systematic MR evaluation of the inflammation-related markers in cancer risk offers insight into tissue-level regulatory mechanisms influencing cancer development. 

Future studies should focus on identifying inflammatory markers mediating cancer risk in participant sub-groups with elevated risk of cancer (e.g., smokers and individuals with autoimmune diseases).