Statistical considerations

Data were analyzed using R v.3.0.3. Data are presented as medians along with a Q1–Q3 interquartile range. Data collected at baseline and at week 12 were compared using a predefined paired non-parametric statistical test due to the study design and the small sample size. A P value <0.05 was considered statistically significant. The number of participants needed to achieve sufficient statistical power was calculated based on the primary efficacy marker using a slightly modified Ahern’s study design (14). The minimum response was determined based on TAX327-controls (15). p₀ was estimated as 5%, p₁ =30%, α =0.05 and power =80%. Further investigations were warranted if ≥3 of the 20 patients met the primary endpoint for efficacy.