We propose a change point detection method for large scale multiple testing problems with clustered signals. Unlike the classic change point detection setup, the signals can vary in size and distribution within a cluster. The spatial structure on the signals enables us to accurately delineate the boundaries between null and alternative hypotheses. New test statistics are proposed for observations from one sequence and multiple sequences. Their asymptotic distributions are established with consistent estimators for unknown parameters. We allow the variance to be heteroscedastic in the multiple sequence case. Simulation studies demonstrate that the proposed method yields favorable performance. Dataset from aCGH are used to demonstrate the utility of the proposed methods.