This study examined the effectiveness of a linguistically and culturally adapted Applied Behavior Analysis (ABA) intervention targeting lexical-semantic language development in preschool children with Autism Spectrum Disorder (ASD) in Kazakhstan. Sixty children aged 3-6 were allocated to an experimental group receiving the adapted ABA program or a control group receiving routine high-intensity VB-MAPP-informed ABA therapy. The intervention incorporated discrete trial teaching and naturalistic learning across 720 instructional hours over nine months. Lexical-semantic outcomes were assessed using a linguistically adapted assessment instrument, and data were analyzed through a 2 × 2 repeated-measures ANOVA with age and gender as covariates. Results revealed significant main effects of Time [F (1, 56) = 84.73, p < .001] and Group [F(1, 56) = 7.92, p = .007], as well as a strong Group×Time interaction [F(1, 56) = 18.56, p < .001, η² = .25], demonstrating greater gains in the experimental group. Qualitative findings further suggested improvements in word production, semantic categorization, and vocabulary generalization, while persistent difficulties remained in abstract and relational term use. Given the non-randomized design and the link between group assignment and institutional site, the findings should be interpreted as preliminary quasi-experimental evidence of intervention-associated lexical-semantic gains rather than as definitive causal evidence.