Common single-nucleotide variation in the host accounts for 25% of the variability in the plasma levels of HIV during the clinical latency stage (viral load set point). However, the role of rare variants and copy number variants remains relatively unexplored. Previous work has suggested copy number variation of a cluster of beta-defensin genes affects HIV load in treatment-na " ive sub-Saharan Africans and rate of response to antiretroviral treatment. Here we analyse a total of 1827 individuals from two cohorts of HIV-infected individuals from Europe and sub-Saharan Africa to investigate the role of beta-defensin copy number variation on HIV load at set point. We find no evidence for association of copy number with viral load. We also compare distribution of beta-defensin copy number between European cases and controls and find no differences, arguing against a role of beta-defensin copy number in HIV acquisition. Taken together, our data argue against an effect of copy number variation of the beta-defensin region in the spontaneous control of HIV infection.