Individual differences in serotonin 1A (5-HT1A) receptor may result in variable response to antipsychotics with 5-HT1A receptor partial agonism. We investigated the relationship between 5-HT1A receptor gene (HTR1A) single nucleotide polymorphisms (SNPs) and efficacy of antipsychotics with 5-HT1A receptor partial agonism in Japanese patients with schizophrenia. Perospirone or aripiprazole was administered to 100 patients with schizophrenia in a randomized controlled study. Candidate SNPs were rs6295 (which affects HTR1A expression and function), rs1364043, rs878567, and rs10042486. Efficacy at week 12 of treatment was evaluated using the Positive and Negative Syndrome Scale (PANSS) 5-factor subscales (excitement/hostility, depression/anxiety, cognition, positive, and negative). Rs1364043 T allele was correlated with the percent change in the PANSS 5-factor negative score (P < 0.01). Haplotype analysis showed that the rs10042486-rs6295-rs1364043 T-C-G haplotype was correlated with worse negative score improvement (haplotype frequency, 0.675; P = 0.014), and the relatively rare T-G-T haplotype correlated with better efficacy (haplotype frequency, 0.05; P = 0.031). This is the first study to show that rs10042486-rs6295-rs1364043 HTR1A variants may be correlated with the improvement of the PANSS 5-factor negative score during treatment with 5-HT1A partial agonist antipsychotics. Studies with larger sample sizes and in different ethnic groups are warranted.