3.2 Overall analysis
Overall meta-analysis was performed on all studies that reported one or more cancer events associated with BD (pooled RR, 1.19; 95% CI: 1.09–1.30, Fig. 2). However, the studies exhibited significant heterogeneity (I2 = 81.3%; P = .000).
3.3 Subgroup analysis
We explored factors affecting heterogeneity through subgroup analysis based on gender, country, and types of malignancy (Table 2). We found an increased risk for hematological cancer (pooled RR, 2.58; 95% CI: 1.61–3.55), especially the leukemia and non-Hodgkin lymphoma, and thyroid cancer (pooled RR, 1.25; 95% CI: 1.04–1.47) but not for other organ cancers. Interestingly female BD patients from Korean population showed higher risk for developing malignancy (pooled RR, 1.73; 95% CI: 1.36–2.10; pooled RR, 1.18; 95% CI: 1.07–1.29, respectively). However, female patients and those from Korean population also reflected heterogeneity (I2 = 50.7%; 83.5%, respectively).
3.4 Publication bias and sensitivity analysis
We addressed publication bias through Funnel plot and Begg test. We did not observe any obvious asymmetry in funnel plot (supplementary Fig. 1, http://links.lww.com/MD/D316). The P value of Begg test was >.05, clearly showing no publication bias in our meta-analysis. Pooled RR values did not alter drastically upon sequential omission of individual studies. Thus, the sensitivity analysis showed our meta-analysis method to be reliable.
Our study is a novel work in the field of systematic review and meta-analysis highlighting the connection of BD with malignancy. We concluded association of BD with overall increased cancer susceptibility (pooled RR, 1.19; 95% CI: 1.09–1.30). We also found high degree of heterogeneity. Thereafter, the source of heterogeneity was assessed via subgroup analysis, which found that most of the heterogeneity was attributable to gender and country factors; especially male and Korean (Table 2).
During subgroup analysis of specific cancer types, we observed that BD reflects an increased risk of association with hematological cancer (pooled RR, 2.58; 95% CI: 1.61–3.55) and thyroid cancer (pooled RR, 1.25; 95% CI: 1.04–1.47) through subgroup analysis of different cancer types. Our findings were consistent with previous studies.[18,19] Besides, Turesson et al and Baecklund et al reported that broad categories of autoimmune disorders, including rheumatoid arthritis, Sjogren syndrome, are associated with an advanced risk for hematologic cancer.[4,20] As for the mechanism, some researchers considered that chronic activation, B or T cell stimulation, and different inflammatory cytokines including interleukin (IL)-6, IL-8, TNF are involved in BD just like the pathogenic mechanisms of hematologic cancers. Moreover,HLA-B27, as the major MHC class I, is also associated with BD.[13,20,21] Furthermore, thyroid cancer in solid tumors was the first time shown to have correlation to BD. In the past, BD was observed to be connected to incidence of solid tumors of breast, uterus, thyroid, and stomach in some case and case series, but it was difficult to determine whether BD had a greater risk of solid cancers. Our meta-analysis indicated it to be sufficient to monitor patients with BD for the development of hematological and thyroid cancer. However, our results do need to be further examined because of a low number of studies.
It was interesting to note that females reflected an increased risk of association with cancer (pooled RR, 1.73; 95% CI: 1.36–2.10) during the subgroup analysis. As we all know, inflammatory could accelerate epigenetic alterations and could cause inflammation-related carcinogenesis, which chronic inflammation is thought to be a causal factor in cancer briefly.[2,6,20] However, it is well appreciated BD runs a less severe disease course among the females. To the contrary, males have a distinctly more severe course, particularly associated with pulmonary vascular disease.[22,23] This counter-intuitive finding of more cancer among the females deserves attention. Some scholars suppose that sex hormones may contribute to the pathogenesis of cancer because of the relationship between breast cancer and BD. However, our analysis did not reflect the increased risk of association with breast cancer, and past studies did not prove significant changes in estrogen serum levels included the higher prolactin serum levels. Therefore, the result needs to be further investigation in the future.
The strength of our studies is as follows: Firstly, it is a novel meta-analysis with a large cohort to evaluate the association between BD and malignancy, wherein all of the included studies are population-based work. Moreover, subgroup and sensitivity analyses helped to determine the possible influential causes to affect the results and thereby increased the strength of our findings. Thirdly, the study involved exhaustive literature review to explore the connection between BD and the risk of malignancy. Finally, most of the studies included in our work were of high quality. These characteristics make our research findings more convincing.
Like other studies, our work was also not devoid of limitations. There are several limitations that must be considered. First, though our work focused on the risk for malignancy in patients with BD. The included studies involved different design and demographic characteristics, which can introduce bias in the readouts. Second, the malignancies reported in the studies were not characterized in detail, thereby limiting definitive conclusions through subgroup analyses. Third, all of the included studies were performed in Asian population and hence extrapolating the conclusions to worldwide population would be erroneous. Fourth, the meta-analysis could be questioned of publication bias as we included only 5 studies though the P value of Begg test was >.05. At last, there was bias in use of literature search portals which omitted out Google scholar (GS) due to institutional limitation in access.
Our study concludes higher risk of BD patients to be afflicted with malignancy. The finding should be validated in bigger cohorts for firm conclusions. Moreover, we wish our analysis will identify future direction for research in BD.
Conceptualization: Xin Wang, Yu Peng, Jun Gao, Yongning Li.
Data curation: Xin Wang.
Formal analysis: Yu Peng.
Investigation: Xin Wang, Yongning Li.
Methodology: Yu Peng, Shiyuan Han.
Visualization: Jun Gao.
Writing – original draft: Xin Wang.
Writing – review & editing: Xin Wang, Yongning Li.
. Melikoglu M, Melikoglu MA. What affects the quality of life in patients with Behcet's disease? Acta Reumatol Port 2014;39:46–53.
. Direskeneli H. Autoimmunity vs autoinflammation in Behcet's disease: do we oversimplify a complex disorder? Rheumatology (Oxford, Engl) 2006;45:1461–5.
. Sakane T, Takeno M, Suzuki N, et al. Behcet's disease. N Engl J Med 1999;341:1284–91.
. Turesson C, Matteson EL. Malignancy as a comorbidity in rheumatic diseases. Rheumatology (Oxford, Engl) 2013;52:5–14.
. Liang JA, Sun LM, Yeh JJ, et al. Malignancies associated with systemic lupus erythematosus in Taiwan: a nationwide population-based cohort study. Rheumatol Int 2012;32:773–8.
. Abu-Shakra M, Guillemin F, Lee P. Cancer in systemic sclerosis. Arthritis Rheum 1993;36:460–4.
. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol 2009;62:e1–34.
. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis
of observational studies in epidemiology: a proposal for reporting. Meta-analysis
Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12.
. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603–5.
. Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev 1987;9:1–30.
. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis
detected by a simple, graphical test. BMJ (Clin Res Ed) 1997;315:629–34.
. Stuck AE, Rubenstein LZ, Wieland D. Bias in meta-analysis
detected by a simple, graphical test. Asymmetry detected in funnel plot was probably due to true heterogeneity. BMJ (Clin Res Ed) 1998;316:469author reply 470−461.
. Han M, Jung YS, Kim WH, et al. Cancer risk in patients with intestinal Behcet's disease: a nationwide population-based study. Gut Liver 2018;12:433–9.
. Jung YS, Han M, Kim DY, et al. Cancer risk in Korean patients with Behcet's disease: a nationwide population-based study. PloS One 2017;12:e0190182.
. Yu KH, Kuo CF, Huang LH, et al. Cancer risk in patients with inflammatory systemic autoimmune rheumatic diseases: a nationwide population-based dynamic cohort study in Taiwan. Medicine 2016;95:e3540.
. Na SJ, Kang MJ, Yu DS, et al. Cancer risk in patients with Behcet disease
: a nationwide population-based dynamic cohort study from Korea. J Am Acad Dermatol 2018;78:464–70.e2.
. Wang LH, Wang WM, Hsu SM, et al. Risk of overall and site-specific cancers in Behcet disease
: a nationwide population-based study in Taiwan. J Rheumatol 2015;42:879–84.
. Ahn JK, Oh JM, Lee J, et al. Behcet's disease associated with malignancy in Korea: a single center experience. Rheumatol Int 2010;30:831–5.
. Lin Y, Li G, Zheng W, et al. Behcet's disease associated with malignancy: a report of 41 Chinese cases. Int J Rheum Dis 2014;17:459–65.
. Baecklund E, Smedby KE, Sutton LA, et al. Lymphoma development in patients with autoimmune and inflammatory disorders − what are the driving forces? Semin Cancer Biol 2014;24:61–70.
. Bardak Y, Aridogan BC. The demonstration of serum interleukin 6-8, tumor necrosis factor-alpha, complement, and immunoglobulin levels in Behcet's disease with ocular involvement. Ocul Immunol Inflamm 2004;12:53–8.
. Kural-Seyahi E, Fresko I, Seyahi N, et al. The long-term mortality and morbidity of Behcet syndrome: a 2-decade outcome survey of 387 patients followed at a dedicated center. Medicine 2003;82:60–76.
. Hatemi G, Yazici Y, Yazici H. Behcet's syndrome. Rheum Dis Clin North Am 2013;39:245–61.
. Song GG, Lee YH. Circulating prolactin levels and Behcet's disease: a meta-analysis
. Cell Mol Biol (Noisy-le-Grand) 2018;64:14–8.
Behcet disease; malignancy development; meta-analysis; risk factors
Supplemental Digital Content
Copyright © 2019 the Author(s). Published by Wolters Kluwer Health, Inc.