Methamphetamine (METH) is a drug with high abuse potential that directly affects the central nervous system. METH use leads not only to detrimental health for the user but also to potential risks and costs for society. Impaired cognition has been shown to exist in individuals with long-term METH dependence, and this impairment may be the main cause of impulsive drug taking and high relapse rate. However, the molecular mechanisms underlying METH addiction and METH-induced cognitive decline remain unknown. To investigate potential molecular mechanisms of METH addiction, we compared serum protein expression levels by using label-free quantitative proteomics in 12 long-term METH users and 12 healthy control participants. In total, 23 differentially expressed proteins were found between the two groups. Bioinformatics analyses were used to identify functional networks and protein-protein interactions. The differentially expressed proteins were related to cognitive dysfunction, inflammation of the nervous system, immune system decline, energy metabolism, and calcium binding and regulation. Thus, these 23 proteins may be closely associated with the damage caused by METH to the nervous, immune, and metabolic systems. The results of this study provide additional insights into the molecular mechanisms of METH addiction and potential prevention and treatment options for METH-dependent individuals.