Is a TNF Alpha Polymorphism Responsible for Mutations in the TP53 Gene?


Ulf Thunberg, PhD, Departments of Oncology, Radiology and Clinical Immunology, Rudbeck Laboratory, Uppsala University, SE-751 85 Uppsala, Sweden

To the Editor:

In a recent paper by Skibola et al., it is speculated that ‘…TNF-α may function as a bridge between inflammation and cancer…’ [1]. The aforementioned study reported a significantly increased frequency (P = 0.004) of the tumour necrosis factor-αΑΑ genotype, (TNF-308G>A, rs 1800629) in diffuse large B-cell lymphoma (DLBCL) (4%) compared with normal controls (2%) [1].

Subsequent investigation into the same polymorphism in our own material, consisting of 109 DLBCL cases, revealed a very interesting ‘connection’ (P = 0.04) between mutations in TP53 and TNF-308AA. Could the TNF-308AA genotype together with TP53 mutations be responsible for cancer development, particularly in DLBCL?

The TP53 gene has a dominant role in tumour biology and is probably the most investigated gene in the world. Mutations in the TP53 gene have been linked to several different diseases, and particularly to cancer development. The majority of TP53 mutations are missense substitutions (75%) with the remaining mutations consisting of frame shift insertions and deletions, non-sense mutations and silent mutations [2].

However, the cause of TP53 mutation is debated, some of the mutations being inherited germ-line mutations, while other mutations arise spontaneously owing to, for example, UV damage, ionizing radiation, activated oncogenes and inflammation [3–5]. In fact, inflammation and chronic infections are believed to be responsible for 25% of all cancers according to epidemiological studies [6].

TNF-α, has previously been showed to be involved in inflammatory diseases such as rheumatoid arthritis, and several studies have showed a higher expression of TNF-α in individuals carrying the TNF-308GA or AA genotype [7–10]. These reports, together with Skibola et al.’s finding of an overrepresentation of the TNF-308AA genotype in DLBCL, strongly imply an association between inflammation and TNF-308AA and cancer development.

Our material consists of 165 Swedish patients with de novo DLBCL : Within this cohort, we observed a 9% (16/165) carried the TNF-308AA genotype compared with only 5% (11/238) normal controls (P = 0.1). Thus, we saw the same trend as that reported by Skibola et al. with DLBCL patients carrying the TNF-308AA genotype at higher frequency than normal controls (P = 0.004) [1].

Moreover, of 109 de novo DLBCL patients with both TP53 data and TNF-308 genotype data, as many as 33.3% (3/9) of the TNF-308AA patients displayed TP53 mutations, while only 8% (8/82) of the TNF-308GG genotypes had TP53 mutations (P = 0.04). This finding indicates that de novo DLBCL with the TNF-308AA genotype more frequently carry TP53 mutations, although the low number of cases in this group confers a low statistical power in this analysis (Power = 0.53).

To our knowledge, it has not so far been shown that a polymorphism can be directly responsible for an increased frequency of mutations in any gene, yet in our material we propose that it is possible that TNF-308AA may be in some way responsible for mutation of the TP53 gene.

This finding may indicate that those patients with the TNF-308AA genotype are prone to more severe inflammation compared with patients with other genotypes. A longer or more severe inflammation could essentially be the early event of cancer formation, creating an environment that promotes mutation formation in, for example, the TP53 gene in a ‘dependent mechanism’ (Fig 1a). A second explanation could be that these genotypes confer a higher risk of cancer, but a second mutation hit targeting a tumour-suppressor gene such as TP53 is necessary for the development of cancer, as an ‘independent mechanism’ (Fig 1b).

Figure 1.

 Hypothesis of how polymorphisms can be involved in inflammation and cancer development. (A) More severe inflammation creating mutation formation in TP53 gene in a ‘dependent mechanism’. (B) Genotypes confer a higher risk of cancer, but a second mutation hit targeting a tumour-suppressor gene as TP53 is necessary for the development of cancer, as an ‘independent mechanism’.

To elucidate whether this, as yet undefined, mutation mechanism is generic and can result in mutation of genes other than the TP53 gene, whether it can play a role in other types of cancers or whether indeed such a mechanism exists at all, further research must be undertaken. While this finding is remarkable, it may merely be attributed to chance, and additional studies must be performed to determine whether there is a genuine biological link between TNF-genotype and TP53 mutation.


The authors are grateful to Prof Richard Rosenquist and Prof Göran Roos for help with the collection of patient material. Also thanks to Dr Thomas Axelsson for the single nucleotide polymorphism (SNP) genotyping performed by the SNP technology platform at Department of Medical Sciences, Uppsala University.