The association between Parkinson's disease and melanoma: a systematic review and meta-analysis

Abstract

Objective: To assess the association between Parkinson's disease (PD) and melanoma via systematic review and meta-analysis.

Methods: Comprehensive search in PubMed, Web of Science, Embase and four China databases (SinoMed, WanFang data, CNKI and VIP database) of epidemiologic evidences on PD and melanoma published before April 30, 2015. Studies which reported risk estimates of melanoma among PD patients or risk estimates of PD in patients with melanoma were included. Pooled odds ratios (ORs) with 95 % confidence intervals (CIs) were calculated by random-effects models. Heterogeneity across studies was assessed using Cochran Q and I(2) statistics. Subgroup analyses and sensitivity analyses were conducted to evaluate sources of heterogeneity. Subgroup analyses were done according to temporal relationship, geographic region and gender respectively. We assessed publication bias using the Begg and Egger test. In addition, study appraisal was done using a scale for observational studies to ensure the quality of evidence.

Results: We identified 24 eligible studies on PD and melanoma with a total number of 292,275 PD patients: the pooled OR was 1.83 (95 % CI 1.46-2.30) overall, subgroup analyses by temporal relationship showed that risk of melanoma after PD diagnosis was significantly higher (OR 2.43, 95 % CI 1.77-3.32), but not before the diagnosis of PD (OR 1.09, 95 % CI 0.78-1.54). Subgroup analysis by geographic region showed that increased risk of melanoma in PD was found both in Europe (OR 1.44, 95 % CI 1.22-1.70) and in North America (OR 2.64, 95 % CI 1.63-4.28). Gender-specific subgroup analyses did not show difference between men (OR 1.64, 95 % CI 1.27-2.13) and women (OR 1.38, 95 % CI 1.04-1.82) in the risk of melanoma. In addition, we found the risk of non-melanoma skin cancers in PD was slightly higher (OR 1.20, 95 % CI 1.11-1.29) than general population. It was impossible to evaluate the association between PD and melanoma according to use of levodopa or gene polymorphism via meta-analysis since few observational or cohort studies have focused on it.

Conclusions: An association between PD and melanoma was confirmed. Most of the evidences were of high quality, and the conclusion was robust. Further research is needed to explore the mechanisms underlying this relationship.

Keywords: Melanoma; Meta-analysis; Odds ratios; Parkinson’s disease.

Fig. 1

Flow diagram: Risk estimates on the association between PD and melanoma

Fig. 2

Association between melanoma and PD according to temporal relationship. Subtotal = pooled odds ratios (ORs) within each subcategory. Overall = pooled OR for all studies. Squares indicate study specific ORs; error bars indicate 95 % confidence intervals (CIs); diamonds indicate ORs and 95 % CIs from pooled analyses

Fig. 3

Association between melanoma and PD according to geographic region. Subtotal = pooled odds ratios (ORs) within each subcategory. Overall = pooled OR for all studies. Squares indicate study specific ORs; error bars indicate 95 % confidence intervals (CIs); diamonds indicate ORs and 95 % CIs from pooled analyses

Fig. 4

Association between melanoma and PD according to gender. Subtotal = pooled odds ratios (ORs) within each subcategory. Overall = pooled OR for all studies. Squares indicate study specific ORs; error bars indicate 95 % confidence intervals (CIs); diamonds indicate ORs and 95 % CIs from pooled analyses

Fig. 5

Association between non-melanoma skin cancers and PD. Subtotal = pooled odds ratios (ORs) within each subcategory. Overall = pooled OR for all studies. Squares indicate study specific ORs; error bars indicate 95 % confidence intervals (CIs); diamonds indicate ORs and 95 % CIs from pooled analyses