Sepsis subtypes and differential treatment response to vitamin C: biological sub-study of the LOVIT trial.

Summary

In 457 LOVIT participants with biomarker data, three inflammatory sepsis subtypes were identified. Intravenous vitamin C did not show anti-inflammatory effects across subtypes; time since randomization and concomitant hydrocortisone, not vitamin C, related to anti-inflammatory changes. Treatment effects trended toward harm in all subtypes with significant heterogeneity in magnitude (heterogeneity p=0.002).

Key Findings

• Three inflammatory sepsis subtypes were identified using hierarchical clustering of biomarker profiles.
• Vitamin C showed no discernible anti-inflammatory effects; changes were associated with time and hydrocortisone, not vitamin C.
• Treatment effects favored harm across subtypes with significant heterogeneity in magnitude (heterogeneity p=0.002; ORs ~1.04, 1.33, 1.95).

Clinical Implications

Routine intravenous vitamin C for sepsis should be avoided; biomarker-defined subtypes did not identify a benefiting group. Focus should shift to targeted therapies and robust phenotyping.

Limitations

• Sub-study included 53% of trial participants with available plasma; potential selection bias.
• Not powered for definitive subgroup treatment-effect interactions on clinical endpoints.

Future Directions

Integrate multi-omics phenotypes with adaptive trial designs to match therapies to endotypes; prioritize therapies with biological plausibility in defined subtypes.