Failure to detect human herpesvirus type 8‐DNA sequences in systemic plasmacytosis from HIV‐uninfected Japanese patients

and 13, which our patient reacted to, belonged to the chemical family of aminoazobenzene (Fig. 1). However, our patient did not react to Disperse red 1, 17 or PPD, which allowed us to narrow the epitope to aminoazobenzene, while excluding the nitrophenyl group and azo structure as T-cell epitopes. This interesting ®nding should be correlated with the results of azo dye patch tests in the future, in which a larger number of patients would make the results more clear, especially in the context of heterogeneous human leucocyte antigen and human metabolic pathways. By performing patch tests on a group of structurally de®ned haptens, we may be able to map the T-cell epitope that is responsible for a patient's allergic contact dermatitis. The results from this study create a new meaning for patch testing, namely epitope mapping, by which one could even illustrate hapten±TCR interaction at the molecular level in order to design safer, less allergenic compounds.