In response: Letter on update to the Vitamin C, Thiamine and Steroids in Sepsis (VICTAS) protocol.

TRIAL REGISTRATION: ClinicalTrials.gov: NCT03509350. Registered on 26 April 2018.

To the Editors,

We thank Drs. Frommelt, Kory, and Long for their interest in the Vitamin C, Thiamine and Steroids in Sepsis (VICTAS) trial and their recommendation to carefully consider time to treatment in our analysis plan. As they have pointed out, our inclusion criteria require study drug administration within 28 h of the onset of first qualifying organ dysfunction [1]. While it is unknown what differential effect time to hydrocortisone, ascorbic acid, and thiamine (commonly referred to as HAT therapy) may have on patients with sepsis, we agree early treatment with other therapeutics has improved outcomes in sepsis [2]. However, in planning the VICTAS trial, we chose to allow a moderately wide enrollment window reasoning that HAT therapy, if effective, would be of interest to providers managing patients along a wide spectrum of illness severity.

Since completing enrollment in the VICTAS trial in late 2019, discussions of trials exploring the benefits of therapeutic regimens like the one we tested have broadened. While much focus at the time VICTAS was designed was on the question of the efficacy of HAT, negative trials have stimulated a discussion as to when HAT therapy could be effective [3, 4].

In response, we have calculated the time between first qualifying organ dysfunction and first dose of study intervention in the VICTAS trial. This will be reported along with other participant characteristics and will be included in our adjusted analyses following the same approach as described for other covariates. Specifically, we will use restricted cubic splines to address potential non-linearities, and we will assess for a differential treatment effect. This will be done by testing the interaction between time to treatment and treatment assignment. If the interaction achieves a P value ≤ 0.2, we will consider the possibility of subgroup analyses with grouping informed by the relationship between time to treatment and outcomes. Since the approach was specified a priori, this helps to mitigate the post hoc addition of this covariate. We expect that including this important variable will strengthen our findings given the suggestion that time to treatment may modify the response to treatment.