key: cord-0980916-4p7vly9e
authors: Barbui, Tiziano; Carobbio, Alessandra
title: Response to: Ruxolitinib withdrawal due to the COVID-19
date: 2021-02-25
journal: Leukemia
DOI: 10.1038/s41375-021-01187-4
sha: 7f5eadd87d9afe073f920fa03a6d6f18218aae83
doc_id: 980916
cord_uid: 4p7vly9e

nan

The comment raised by Dr. Lucijanic and Kusec that the effect on high mortality in patients with MPN and COVID-19 could not so much be due to the suspension of ruxolitinib but to the greater severity of the disease was obviously considered in our analysis [1] , both in the multivariate model and in the propensity score (PS) matching.

First of all, we clarify that the logistic model presented was not only performed on 45 patients exposed to ruxolitinib. The model considered the entire cohort and adjusted the exposition to ruxolitinib for its discontinuation through the technique of mediation analysis [2] .

We were aware that, given the low number of events, the logistic model did not allow to correct for many confounders and for this reason we also performed a sensitivity analysis applying the PS matching. As Supplementary  Table 1S shows, there were no significant differences between those who discontinued and those who did not discontinue after PS matching.

This finding should alert the haematologist who cares for patients with MPN treated with ruxolitinib, who contract COVID-19.

Conflict of interest The authors declare no competing interests.

Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

High mortality rate in COVID-19 patients with myeloproliferative neoplasms after abrupt withdrawal of ruxolitinib

Mediation analysis in epidemiology: methods, interpretation and bias