In this study, we investigated a large cohort of more than 24,000 outpatients in Germany for a possible association between infectious mononucleosis, representing the clinical manifestation of EBV infection, and cancer risk using data from a representative real-world database. According to our findings, infectious mononucleosis is significantly associated with an increased incidence of tumors of hematopoietic and lymphoid tissues. In contrast, no significant association was found for the overall risk of cancer in patients with infectious mononucleosis compared to the matched non-mononucleosis cohort, but there was still a tendency for a higher incidence of cancers in infectious mononucleosis patients in the age group > 50 years and among men.

Infectious mononucleosis has been associated with the development of Hodgkin’s lymphoma for over 30 years, suggesting a causal role of EBV infection [10,11,12,13,14,15,16,17,18]. Interestingly, studies identified the presence of EBV in the characteristic Sternberg-Reed cells of Hodgkin’s disease in about 40–50% of cases [19,20,21,22]. A UK case–control study demonstrated an Odds ratio of 9.16 (95% CI = 1.07–78.31) for EBV-positive Hodgkin lymphoma after infectious mononucleosis [12]. Similarly, Hjalgrim and colleagues, who examined cancer incidence in a population-based cohort from Denmark and Sweden of more than 38,000 patients with infectious mononucleosis, found that patients with infectious mononucleosis were at increased risk for Hodgkin’s disease. It was concluded that the relative risk of Hodgkin’s disease will remain elevated more than twofold for up to 20 years after the diagnosis of infectious mononucleosis [11]. However, apart from a moderately increased risk of skin cancer, the authors found no increased cancer risk associated with infectious mononucleosis for cancers other than Hodgkin’s disease. This observation is consistent with the results of our study, which also revealed no significantly increased overall cancer risk associated with infectious mononucleosis. To our knowledge, our study was further the first to indicate a strong trend toward an increased risk of prostate cancer associated with infectious mononucleosis, although the corrected level of statistical significance was not reached. The role of EBV in the development and progression of prostate cancer has remained controversial in the literature; however, EBV is reported to be detectable in 40–50% of prostate cancer specimens [23]. It is suggested that EBV infection with subsequent expression of EBV latency genes could enhance survival of premalignant cells and provide anti-apoptotic properties to epithelial cells in the premalignant stage [24]. At the same time, changes such as inflammation in stromal parts of the tissue could lead to modulation of EBV latency gene expression, altering the growth characteristics of the tissue and triggering carcinogenesis [25].In contrast, for respiratory organ cancer, we found a non-significant trend toward a negative association with infectious mononucleosis. Indeed, there are no reports in the literature to date of a possible association between infectious mononucleosis and respiratory organ cancer. A relationship between respiratory organ cancer and EBV, however, has already been investigated in several studies [26,27]. Epidemiologic evidence suggests that EBV may play a role in a very limited subgroup of lung cancer [26,27]. In this context, pulmonary lymphoepithelioma-like carcinoma (PLELC), which is a rare form of non-small cell lung cancer closely associated with EBV infection, has to be mentioned [27]. One possible explanatory approach for a potentially negative association between infectious mononucleosis and respiratory organ cancer could be the life cycle of EBV, which occurs mainly in oral epithelial cells and in B cells [27]. In contrast to the well-established pathway by which EBV enters B cells, it is still largely unclear how EBV is able to invade other epithelia such as those of the lung [26,27]. Possibly, demographic aspects could play a role here, supported by the fact that EBV-positive lung carcinomas are predominantly found in Asian populations [27].Another finding of our study was a significantly higher prevalence of thyroid and liver disease in patients with infectious mononucleosis. There are few case reports in the literature that indicated an association between infectious mononucleosis and the occurrence of thyroid disease [28]. However, a possible etiologic role of EBV itself has been discussed, particularly for autoimmune thyroid diseases. In this context, nuclear expression of EBV non-coding RNA EBER (EBV-encoded RNA) was detected in 80.7% of Hashimoto’s thyroiditis and in 62.5% of Graves’ disease cases [29]. Interestingly, it is already known that infectious mononucleosis is associated with other immunological diseases, with evidence for multiple sclerosis in particular [9,30]. With respect to liver involvement, transient elevation of serum aminotransferase levels was noted in up to 90% of young adults with infectious mononucleosis, with older adults having higher liver values than adolescents [31,32]. However, chronic liver damage following infectious mononucleosis has not been described [33]. Infectious mononucleosis occurs in 25–70% of cases in the Western population during initial infection with EBV [34] and has a peak between the ages of 15 and 25 years [35]. Once infected, EBV persists in the organism life-long [34]. While approximately 30% have seroconversion for EBV by the age of 14, up to 95% are infected with the virus by the age of 40 [35]. Although almost the entire world population is infected with EBV, the virus seems to cause very different diseases in different regions of the world [36,37,38]. The majority of EBV carriers do not present with clinical symptoms due to viral infection. However, EBV is etiologically associated with the development of different cancers, including several types of B-, T-, and NK-cell lymphomas, as well as carcinomas of the nasopharynx, stomach, parotid gland, and thymus [36,38,39]. Globally, about 200,000 annual cancer cases and approximately 2% of all cancer-related deaths are thought to be EBV-associated [27,40,41]. Interestingly, there is a very heterogeneous prevalence of EBV-associated malignancies worldwide. Nasopharyngeal carcinoma, for example, is more common in Southeast Asia, where it accounts for the most common tumor in young adults, as well as in North Africa, whereas it is rare in Europe. While Burkitt’s lymphoma is endemic in central Africa, T-cell lymphomas are almost exclusively described in Japan [36]. The existence of different viral strains or subtypes found only in limited geographic areas, as well as deletions or polymorphisms in the EBV genome, have been discussed in the literature as possible explanations for the differences in disease pattern [42,43]. Delecluse and coauthors identified that virus type M81, isolated from a nasopharyngeal carcinoma, infiltrates epithelial cells of the nasal mucosa in addition to immune system B cells [39]. In contrast, virus types associated with causing infectious mononucleosis almost exclusively infect B cells. Age of onset of EBV infection, environmental factors such as nitrosamines or smoking are hypothesized to have an impact on cancer risk. In addition, cofactors such as concurrent infection with Plasmodium falciparum in the case of Burkitt’s lymphoma or immunosuppression in EBV associated B-cell lymphoma or post-transplant lymphoma (PTLD) have been attributed an important role [44,45]. Additionally, genetic determinants are discussed as contributing factors [46]. EBV has been shown to affect various cell types, including B cells, epithelial cells, and T cells [47]. Several glycoproteins of the virus are involved in enabling the virus to invade the host cell. The envelope protein gp350/gp220 is thought to be of major importance in this process [48]. It has a high affinity for the complement receptor type 2 CD21 (CR2), which is mainly expressed on B lymphocytes [48,49]. The potential of EBV to immortalize B cells is thought to play an important role in linking the virus to the pathogenesis of lymphoproliferative disorders. In this context, the expression of the six EBV nuclear antigens (EBNA1–6) and the two latent membrane protein genes (LMP1 and LMP2) have been found to be of relevance. The cellular tumor necrosis factor (TNF)-receptor-1-associated death domain protein (TRADD), which mediates the induction of apoptosis as well as the activation of nuclear factor kappa B (NF-κB) by cellular tumor necrosis factor receptor 1 (TNFR1), is thought to be a crucial signaling mediator of LMP1 and critically involved in the immortalization process [50]. Furthermore, EBV lytic replication has been identified as a cancer risk factor [51,52]. After all, the property of EBV gene products to be involved in cell proliferation as well as apoptosis and immunological processes may provide an explanation for why the virus is associated with the development of so many different diseases. [53]. Our study’s strengths are its population-based setting, cohort size, long follow-up period, as well as the use of a database that has already been proven to be representative for Germany [7]. Nevertheless, there are also limitations to our study, in part because of the study design. For example, diagnoses are based on documentation of ICD-10 codes by general practitioners. Therefore, we cannot exclude the possibility that diagnoses were sometimes misclassified or that the coding was missing from the ICD-10 coding system. In this line of thinking, it is important to note that our study includes only patients with a documented diagnosis of infectious mononucleosis. Thus, patients with asymptomatic EBV infection, which did not lead to a medical consultation, are not included and the non-mononucleosis cohort might include a proportion of patients with clinically inapparent EBV infection vice versa. Nevertheless, the study results of Hjilgram and colleagues indicated that the accuracy of the diagnosis of infectious mononucleosis did not differ significantly when evidence of acute infectious mononucleosis infection was established either serologically or clinically [11]. Moreover, our study explicitly examined the association between cancer and infectious mononucleosis but not EBV. As such, it is not feasible to draw any conclusions about EBV strains or subtypes. Furthermore, tumors of hematopoietic and lymphoid tissues could not be subdivided in more detail because of the existing ICD coding. Finally, the German Disease Analyzer database does not contain information on patients’ lifestyle or socioeconomic status, risk factors such as smoking, immune status, immunological markers, or genetic factors that would have allowed further analyses.