Ontogenetic variation in susceptibility of Spodoptera frugiperda (Lepidoptera: Noctuidae) to azadirachtin under high-altitude conditions: an integrated dose–response and survival analysis approach

Abstract Azadirachtin, the principal bioactive limonoid of Azadirachta indica A. Juss., is a widely used botanical insecticide in Integrated Pest Management programs targeting Spodoptera frugiperda (J.E. Smith), yet the ontogenetic modulation of its toxicity under high-altitude Andean thermal conditions remains uncharacterized. This study quantified the susceptibility of six larval instars of S. frugiperda to azadirachtin under high-mountain laboratory conditions (2,586 m a.s.l.; 17 ± 1°C) using an integrated framework combining Probit dose–response analysis, Kaplan–Meier survival estimation, and Cox proportional hazards regression. Survival analysis was performed on 1,488 time-to-event records (n = 248 per instar; 1,009 events, 479 right-censored observations). Early instars (L1–L3) exhibited high susceptibility, with 87.9–89.9% cumulative mortality within 96 h, whereas late instars (L4–L6) showed markedly lower mortality (42.7–51.6%), with more than half of the individuals surviving the observation period. Kaplan–Meier curves differed significantly among instars (log-rank combined trend: χ² = 119.35, df = 1, p < 0.001). The Cox proportional hazards model was highly significant (χ² = 95.77, df = 2, p < 0.001): each successive larval instar reduced instantaneous mortality hazard by approximately 15.4% (HR = 0.846; 95% CI: 0.816–0.878), while each 100 µL/L increment in azadirachtin concentration increased mortality risk by 2.6% (HR = 1.026; 95% CI: 1.013–1.039). The convergence between dose–response and survival modeling confirms a consistent ontogenetic gradient in susceptibility under cooler Andean thermal conditions and provides a quantitative framework for timing neem-based interventions in high-altitude maize agroecosystems.