The fact that viruses of the human herpesvirus family can promote the development of malignancies has been demonstrated for EBV. EBV-patients have an increased risk of developing Hodgkin’s lymphoma in the follow-up of the infection (Shannon-Lowe et al. 2017; Roderburg et al. 2022). For gastrointestinal tumors in general a strong association with gastric cancer has been reported in several studies, including a large meta-analysis by Tavakoli et al. (Mirzaei et al. 2018; Tavakoli et al. 2020). Concerning Germany, however, only two small case–control studies were included in the aforementioned meta-analysis. A large cohort study based on data from the Disease Analyzer database (IQVIA) and involving 24,190 outpatients in Germany did not find a significant association between EBV and gastrointestinal tumors (Roderburg et al. 2022). However, the authors didn’t distinguish between the different GI tumor entities, so a possible significant association with gastric cancer cannot be ruled out. For colorectal cancer, the data are inconclusive regarding a possible association with EBV (Bedri et al. 2019). In contrast, in an Asian collective, EBV-associated intrahepatic cholangiocarcinoma has been described as a separate subgroup (Huang et al. 2021). However, it remains unclear whether other members of the herpesvirus family, in particular VZV, influence tumor development as it was first suggested in 1955 (Wyburn-Mason 1955).

By using a large database of more than 200,000 outpatients in Germany, we conducted a retrospective analysis to investigate the association between the occurrence of GI cancer after a diagnosis of HZ. Strikingly, in contrast to data suggesting increased cancer rates after HZ (Buntinx et al. 2005; Chiu et al. 2013; Cotton et al. 2013), our results did not identify a positive association between HZ and subsequent GI cancer after HZ during a 10-year follow-up period in a very large cohort of well-matched patients.

Both HZ and GI cancers are closely associated with immune dysfunction. Therefore, it has often been hypothesized that HZ may be associated with an occult tumor that is not yet clinically apparent but is already affecting the immune system (Patil et al. 2022). Although recent studies investigating this relationship have reported a significant association, the data remain largely inconclusive, particularly with regard to GI tumors (Buntinx et al. 2005; Schmidt et al. 2017; Sim et al. 2021). In a retrospective study published in 2005, Buntinx et al. described an increased risk of overall tumor development in women older than 65 years after an initial diagnosis of HZ. In an exploratory subgroup analysis, the authors further demonstrated an HR for the development of colorectal cancer of 4.001 (95% CI 1.123–14.257). However, only 1211 patients with HZ were included in this study, of which only 25 patients developed tumor disease during follow-up, limiting the statistical power of the results (Buntinx et al. 2005). Chiu et al. performed a retrospective study including a total of 38,743 patients with HZ and 116,229 patients without HZ. The risk of developing any tumor disease was increased in the first 2 years after diagnosis of HZ. Notably, after a total follow-up period of 5 years, there was no significant difference compared to the control group (Chiu et al. 2013). Cotton et al. (2013) also assessed the risk of developing tumor disease after HZ and showed that the risk was increased, especially in the first year after HZ diagnosis. Finally, Iglar et al. (2013) also showed that the association between HZ and tumor development is particularly strong within the first 6 months after diagnosis of HZ and tends to decrease over time. Therefore, one might speculate that the diagnosis of tumor disease within a short period after HZ might be indicative of a pre-existing but occult and undiagnosed tumor at the timepoint of HZ. In fact, HZ may be a symptom of tumor-associated immunodeficiency that favors VZV reactivation in these cases (Kim et al. 2021). In our study, we also assessed the incidence of GI cancer at different timepoints after HZ diagnosis. In contrast to the findings reported by Cotton et al. and Chiu et al. we did not find increased rates of GI cancer in the first years after HZ diagnosis. Therefore, regarding our data we cannot support the theory of HZ being indicative of occult GI cancer disease that might be more likely diagnosed after HZ-diagnosis.

Interestingly, the clinical course of HZ also seems to be relevant for the association with tumor development. Post-herpetic neuralgia (PHN) is an important complication of HZ that is particularly challenging to treat and occurs preferentially in immunocompromised patients. A study in an Asian cohort found out, that the occurrence of PNH was associated with subsequent tumor diagnosis, including the GI system, whereas patients with HZ but without PNH had a lower risk of tumor development (Sim et al. 2021).

To the best of our knowledge, we report for the first-time on the specific association of HZ and the subsequent GI cancer. In contrast to the aforementioned studies, in the present retrospective study, we focused on different GI cancer entities and examined also the long-term follow-up of the patients and did not identify a significant association between the two disease entities.

The main strength of our study is the large number of patients analyzed with a long follow-up time of up to 10 years. We performed sufficient matching of control patients without HZ, which further increased the quality of our data. However, the main limitations of our study are the retrospective study design and analyses based on ICD code data, lack of hospital data, mortality information, data on smoking status, alcohol use, physical activity, and detailed information on laboratory parameters, such as inflammation. Further, we did not review individual medical records in detail. It should be noted that we focused our analysis on the most common tumor entities of the GI tract and thus cannot exclude an association with less common tumor subtypes.

In conclusion, there was no association between HZ and subsequent GI-tumors. Our data therefore do not support the hypothesis, that VZV has pro-oncogenic effects during reactivation.