To detect HuD-specific T cells in patients with Hu-antibody associated paraneoplastic neurological syndromes (Hu-PNS), we used short-term stimulation assays with HuD protein spanning peptide pools (PSPP) with purities of at least 70% and found reproducible false-positive CD8+ T-cell responses in three of 127 individuals (two healthy controls and one Hu-PNS patient), which all shared HLA-A*2402 and HLA-B*1801. After testing the 15-mer peptides of the HuD antigen separately, we discovered that the same three 15-mers yielded the CD8+ T cell response in those three individuals. This highly unusual result could not be reproduced when using new batches of peptides with a higher level of purity (>82% and >95%). Therefore, we assumed this response was not directed against the HuD peptides and analyzed the HuD 15-mers by Fourier transform ion cyclotron resonance (FT-ICR) tandem mass spectrometry (MS/MS), which showed the presence of a cytomegalovirus (CMV)-encoded peptide (AIAEESDEEEAIVAY) as a contaminant. The three responding individuals all were CMV-seropositive and the contaminating peptide appeared to fit in the binding groove of HLA-B*18. Our data reveal that synthetic PSPP may contain immunogenic contaminations which may cause false positive results in T-cell stimulation assays.