Objective：To explore whether exist sexual risk compensation in the HIV pre-exposure prophylaxis clinical trial among men who have sex with men（MSM）. Methods：Totally 108 MSM were recruited with non-probability sampling method，and randomly divided into treatment group（n=77） and blank control group（n=31）. Clinical follow-up and questionnaire were carried at 12th，24th，36th，48th week；the socio-demographic characteristics and the HIV/AIDS-related knowledge and behavior were collected according to the ques－tionnaire. Univariate and multivariate generalized estimating equations were used to analyze the sexual risk compensation in the PrEP and the associated factors. Results：During MSM participated in the PrEP clinical trial，the median number of sexual partners of the treatment group at the 12th，24th，36th，48th week was 1（0，6），1（0，6），1（0，10），1（0，3），1（0，3），respectively，and the median number of high-risk sexual behavior was 1（0，26），1.5（0，8），1（0，12），1（0，9），2（0，30），respectively. On the other hand，the median number of sexual partners of blank control group was 1（0，21），1（0，2），1（0，3），1（0，3），1（0，3），respectively，and the median number of high-risk sexual behavior was 1（0，9），1（0，6），0.5（0，15），0（0，10），1（0，10），respectively. Multiariable analysis found that the number of the sexual partners and the number of high-risk sexual behavior were not changed in the PrEP trial（Z=-0.24，P=0.811；Z=0.28，P=0.779）；“1” and “0.5” style sexual behavior MSM had more sex partners compared with that of “0” type MSM（Z=2.47，P=0.014；Z=2.24，P=0.025）. MSM who ever had a commercial sexual behavior had fewer sexual partner and fewer high-risk sexual behavior（Z=2.82，P=0.005；Z=2.28，P=0.023）. Conclusion：There is no evidence of the sexual risk compensation in the PrEP clinical trial among MSM. “1” style and “0.5” style MSM seem to have more sexual partners，who may be the important population during the late follow-up management. We need to strengthen the promotion of the HIV/AIDS related knowledge.