Background
Parkinson’s disease (PD) is one of the most common neurodegenerative diseases characterized by motor dysfunction, such as resting tremor, rigidity, hypokinesia and postural instability [
1], as well as non-motor symptoms including constipation and depression [
2]. The etiology of PD remains unclear, both genetic and environmental factors contribute to it [
3]. While the motor symptoms likely resulted from the loss of dopaminergic neurons in the substantia nigra [
4], the non-motor symptoms of PD are less well understood. The recent theory on the “gut-brain axis” postulates that the enteric microbiota may influence the cognitive behavior of the brain [
5]. On the other hand, the characteristic protein aggregate in the PD brain, α-synuclein, was also found to present in the enteric system [
6,
7].
Colorectal cancer (CRC) is the third most common cancer and the second leading cause of cancer death worldwide [
8]. CRC ranks the top five cancers in new diagnostic cancers and cancer-related death in China [
9], making CRC one of the most serious health problems. CRC develops as the result of the accumulation of genetic and epigenetic alterations [
10]. More recently, studies have also suggested that the alteration in the microbiota could generate local and systemic changes to influence oncogenesis [
11].
These evidences from both the PD and CRC research fields have suggested a potential interaction between the pathogenic mechanisms of the brain and the gut. Previous epidemiology studies conducted in China showed inconsistent results, indicating the necessity of our research. The goal of the present study is to use the public database to explore the disease risk association between PD and CRC. In the recent 20 years, accumulating epidemiological studies have revealed that patients with PD may be associated with a lower risk of certain cancers [
12‐
14], however, the association between PD and CRC remains controversial [
15]. Therefore, we conducted this meta-analysis to provide a quantitative assessment of current epidemiological evidence on CRC in relation to PD and to explore the potential factors affecting the association between the two.
Discussion
PD is an age-related neurodegenerative disorder commonly diagnosed at the age of 60 and above. CRC is increasingly common in people over the age of 60. PD and CRC are totally different illnesses and are thought to have different pathogenic mechanisms. For example, PD is characterized by the death of dopaminergic neurons in the substantia nigra, while CRC derives from the inappropriate cell proliferation with a selective growth advantage. Growing evidence have suggested that patients with PD may have a substantially lower incidence of cancer, but the risk association with CRC is less clear.
Prior to our study, there was only one meta-analysis examining the association between PD and CRC. Unfortunately, there were several caveats and weaknesses associated with that study: (1) It was unclear if the results represented the risk of cancer influenced by PD, or vice versa; (2) The data collected in that study was incomplete; (3) It is methodologically wrong to mix the results from the cohort studies and the case-control studies due to the different design of the two studies.
Our study has overcome the aforementioned weaknesses in the previous work and examined specifically the risk of CRC in patients with diagnosis of PD. We analyzed 17 studies involving 375,964 PD patients and 879,307 cancer patients. The pooled results for all populations indicated that PD patients have a decreased risk of CRC in Western population. Subgroup analysis showed that the significant inverse relationship between PD and risk of CRC is not affected by differences in types of study design, tumor location, or different regions of the Western population.
The inverse association between PD and CRC in the Western population is statistically significant in our study. However, the association in the Asian population remains obscure. We found contradictory conclusions in published studies. In our meta-analysis, we identified a modest lower risk of CRC in patients with PD (RR = 0.83) with a high level of heterogeneity (I
2 = 97%) and the significance is not as clear as data from the American and European groups(
P = 0.442). We think the reason could be due to the wide variability of the Asian data. Some large-scale studies on the Asian population are unfortunately not completed like the similar studies for the Western population. For the two studies that are completed in Taiwan, China, the results are opposite. The RR of the study completed by Sun [
14] in 2011 is 0.72 (95% CI 0.53–0.99). The study consisted of 4957 newly diagnosed PD cases in the cohort and 19,828 non-PD controls during the period of 2000–2005 from the Taiwan National Health Insurance (NHI) Research Database (NHIRD), a nationwide population-based database containing more than 24 million subjects covering 99% of the entire population in Taiwan, China. The other study that completed by Lin [
15] in Taiwan, China in 2015 is the main source of heterogeneity, the pooled OR of this study is 1.47 (95% CI, 1.31–1.65), and the cohort study included 133,322 individuals with PD newly diagnosed between 2004 and 2010, also from the NHI database. Freedman [
32] et al. adjusted for a surrogate for surveillance (number of physician visits) and found that the odds of cancer in the total population after PD was reduced, suggesting that the medical surveillance contributed to the risks.
The key question that our study begs is the potential mechanism that may account for the negative association between PD and CRC. The level of melatonin [
36], dopamine [
37], smoking [
38] and diabetes [
39] have all been proposed to account for such mechanisms. More importantly, patients with PD often have microflora alterations in their feces and colonic mucosa, which may lead to non-motor symptoms such as constipation [
40]. Interestingly, recent studies have also pointed to the role of microbiome and their secretion in inducing local and systemic effects on cancer onset and progression [
11]. Thus, a detailed analysis on the types of microbiota may provide clues to a negative association between PD and CRC. At the molecular level, dysfunction of the ubiquitin-proteasome system (UPS) leads to an accumulation of intracellular proteins and formation of Lewy bodies containing α-synuclein, which is the characteristic pathological feature of PD [
41,
42]. In contrast, the function of UPS is usually up-regulated in CRC [
43]. Whether the UPS stands at the crossroads of dysregulation for PD and CRC awaits further studies. Moreover, studies have shown that the PI3K /AKT/mTOR pathways are hyperactive in patients with CRC [
44] while the activation of the PI3K/AKT/mTOR pathway may promote the survival of dopaminergic neurons by inhibiting apoptosis, thus preventing PD [
45].
Our meta-analysis has provided a most up-to-date picture for the CRC risk in PD patients. As CRC patients often exhibit a much more aggressive disease course than the PD patients, the risk of PD in CRC patients is hard to calculate. Our study, thus, made a strong conclusion for the inverse CRC risk from the Western PD population, while that for the Asian population remains obscure due to large heterogeneity and a small number (4) of available datasets. It is worth noting that publication bias and other forms of bias may still exist, and a more detailed subgroup analysis is incomplete due to the insufficient data from the primary articles. Nonetheless, our study has several strengths, including its comprehensive literature search for the latest data, large number of cases, careful assessment of the quality of evidence, which altogether made the results more reliable compared to earlier studies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.