javier Junior Spain, 20170316 17:49 Posting: # 17158 Views: 936 

Dear All Reading about nivolumab, in the EPAR, there is a confusing information regarding the half life, im wondering if you dont mind to use your expertise to answer this question, in the single dose pk study (with full pk sampling ) the half life was reported in 17 days, while in the multiple dose study (with sparse sampling) the half life was reported in 26 days, How could it be possible? How would you calculate the steady state in this situation? I quote the exact word of the EPAR "the mean terminal elimination halflife of nivolumab ranged between 17 and 27.5 days following single dose (study MDX110601) and Q2W administration (MDX110603) across the range of 0.1 to 10 mg/kg dose." Thanks in advance for your time and dedication answering it Best Regards Javier Edit: EPAR linked and category changed; see also this post #1. [Helmut] 
Helmut Hero Vienna, Austria, 20170317 13:09 @ javier Posting: # 17160 Views: 762 

Hi javier, » in the single dose pk study (with full pk sampling ) the half life was reported in 17 days, while in the multiple dose study (with sparse sampling) the half life was reported in 26 days, How could it be possible? First of all I don’t think that half lives are normal distributed. Hence, I don’t like the reported arithmetic means. There were different number of patients enrolled in the cohorts. Therefore, you should use weighted means which are 22.0 days (single dose) and 26.1 days (multiple dose). Such a difference does not worry me. Furthermore, half lives in the SD study were obtained by NCA and in the MD study by a twocompartment PopPK model. Since there is substantial accumulation it might be possible that in the SD study the terminal phase was not completely “visible” – especially after low doses and NCA underestimated the true (i.e., higher) value. » How would you calculate the steady state in this situation? What do you mean by that? — All the best, Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. ☼ Science Quotes 
nobody Senior 20170317 16:35 @ Helmut Posting: # 17163 Views: 734 

btw nice example why I don't like discussions based on t_{1/2}. If you ever determined it yourself you know what kind of mess this parameter is, not to speak of the pain trying to compare results from different trials (SD/MD, sampling times/duration, noncomp./comp. models, etc. pp...). — Kindest regards, nobody 
javier Junior Spain, 20170320 12:56 @ Helmut Posting: # 17168 Views: 641 

Hi Helmut » » How would you calculate the steady state in this situation? » » What do you mean by that? first, thank you for your swift and ilustrated response, regarding the steady state question, I was wondering about how to calculate the steady state of a drug when you have two different values of half life (If I understand correctly the steady state of a drug takes 35 half lives) thanks again and best regards 
Helmut Hero Vienna, Austria, 20170320 14:40 @ javier Posting: # 17170 Views: 634 

¡Hola Javier! » I was wondering about how to calculate the steady state of a drug when you have two different values of half life See this post (especially the last sentence) and the warfarin example (slides 21–22). If you extrapolate from SD to MD you have to rely on assumptions (linear PK, no saturation/induction/inhibition/capacitylimited excretion). » (If I understand correctly the steady state of a drug takes 35 half lives) It depends on how one defines sufficiently ‘close’ to true steady state (at t = ∞!). For a onecompartment model, IV administration
Coming back to your OP: I would trust the results of the MD study’s PopPKmodel more than the SD study’s results derived by NCA. — All the best, Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. ☼ Science Quotes 