Bioequivalence and Bioavailability Forum

Hi Laura,

❝ I know this topic was already discussed here but I am not clarified.

You are not alone…

❝ What do you think is the best statistical analysis to analyse Tmax? Hodges-Lehmann Estimate for example?

Some stuff to chew (not to say to digest):

• BioBridges (Prague, 22–23 September 2022)
  • Graham Pennick. Product specific guidelines – T_max criteria
    
  • Helmut Schütz. t_max Evaluation. Where are we and where will we go…
    
  • Vít Perlik. Partial AUCs: tool for development vs regulation
• BioBridges (Prague, 21–22 September 2023)
  • Jiří Hofmann. Bioequivalence of Early Exposure: t_max and pAUC
    
  • Paolo Biffignandi. Tadalafil T_max – case study
• Helmut Schütz. ‘Apparent’ Difference in t_max? June 2023

The last one is pretty technical, I must confess, and it was written before the EMA’s product-specific guidances were finalized.
Personally I’m happy with nonparametric methods. They were recommended for ages till 2010 in Europe and still are in some other jurisdictions. I used them in hundreds of my studies. IMHO, it would only make sense if you define a priori a clinically not relevant range and assess the ≈90% CI for inclusion. This range should be defined case by case and not based on this crappy 20% of the median as the EMA stated. As shown in the presentations of Jiří and myself it would require an extremely tight sampling schedule. Furthermore: You might have different ranges dependent on the indication. Think about paracetamol: The range might be narrow as an analgesic, but is completely irrelevant if administered in multiple doses as an antipyretic in children.
There will be an entire session dealing with early exposure at the 6^th GBHI workshop (Rockville, 16–17 April 2024).