Aims: This study investigated whether common CYP3A4 and CYP3A5 variants are associated with clinical outcomes in hospitalized COVID-19 patients treated with MP in Manaus, Brazil. Methods: A pharmacogenetic analysis was conducted on 100 hospitalized COVID-19 patients enrolled in a randomized, placebo-controlled clinical trial receiving heavy-dose intravenous MP (n = 49) or placebo (n = 51), for five days. CYP3A4: c.-392G>A (rs2740574, CYP3A4*1.002) and CYP3A5 (rs776746 - *3 and rs10264272 - *6) variants were genotyped. Associations between genotypes, treatment outcomes, and time to discharge or death were analyzed using χ 2 tests, Kaplan-Meier curves, and Cox regression. Results: Allele frequencies of CYP3A4*1.002 (86%), CYP3A5*3 (70%), and CYP3A5*6 (2%) were within those reported in admixed Latin American populations . Genotype distributions did not differ significantly between treatment groups or outcome categories. Carriers of CYP3A5 wild-type (*1/*1) genotypes had shorter hospital stay (p = 0.012), and this genotype did not influence mortality. Adjusted Cox models showed only sex, comorbidities, and weight predicted discharge likelihood in both CYP3A4 and CYP3A5 models (p<0.05). Conclusion: CYP3A4 and CYP3A5 variants were associated with modest differences in hospital length of stay but did not independently influence survival among MP-treated COVID-19 patients. Host genetic variation in CYP3A-mediated metabolism may contribute to variability in recovery dynamics, although clinical characteristics remained dominant determinants of clinical outcome.