In orthogonal frequency division multiplexing (OFDM) with massive multi-input multiple-output (mMIMO), the reduction of the high peak-to-average power ratio (PAPR) is a challenging problem. Recently, an adaptive peak cancellation is proposed to reduce the transmitted signalâ™s PAPR, while keeping the out-of-band leakage power (ACLR) as well as an in-band distortion power (EVM) below the predetermined and permissible value. In this paper, we propose an analytical method to evaluate achievable BER performance of downlink OFDM with the peak cancellation in massive multi-input-multioutput (mMIMO) systems using arbitrary numbers of transmit antennas and served users. In this method, bit error rate (BER) is derived based on the assumption that in-band distortion due to peak cancellation is approximated as random variable following Gaussian distribution, provided that variance of the Gaussian distribution in two user case is known. The results clarify that theoretical BER expressions for arbitrary numbers of transmit antennas and served users show good agreements with its simulation results. In addition, we clarified the impact of the increase of the number of transmit antennas on achievable BER and PAPR reduction capability in MRC precoded mMIMOOFDM system with the peak cancellation.