Bioequivalence and Bioavailability Forum

Hi Łukasz!

❝ As per reference product SmPC prasugrel may be administered with or without food. Wchich means that in EU one study under fasting conditions would be sufficent.

True. But I don’t know whether Sam wants to apply in the EU.

❝ Based on fast study, Sam's formulation would be approved, although clearly it is not bioequivalent under fed conditions.

OK, for the FDA Sam would have to show BE fasting+fed. Even for EMA we have to show the synopsis of all studies performed during product development. Too bad that the C_max ratio was that high in fed state. I would expect questions approaching.

❝ Taking this into consideration and this Helmut's statement:

❝ ❝ General remark: For the majority of formulations BE in fed state is more difficult to show than in fasted state.

❝ what rationale for this guideline wording:

❝ "In general, a bioequivalence study should be conducted under fasting conditions as this is considered to be the most sensitive condition to detect a potential difference between formulations." is?

❝ In this case it looks like it is totally other way round.

I should have been more precise. EMA’s (and FDA’s) point of view generally is correct for most IR formulations. For MR formulations (more of the delayed release type) and IR formulations of drugs with presystemic metabolism and/or a high first pass (especially pro drugs), this might not be the case, IMHO.

Hi
For all who want to know more about prasugrel bioavailability and hints for bioequivalence I suggest:

Seiler D, Doser K, Salem I.: Relative bioavailability of prasugrel free base in comparison to prasugrel hydrochloride in the presence and in the absence of a proton pump inhibitor. Arzneimittelforschung. 2011;61(4):247-51.

Please be advised that if lansoprazole is used to raise the pH level in the upper gastro-intestinal tract, AUC is decreased by 25% after administration of prasugrel hydrochloride and by 41% after prasugrel free base. In addition, the peak plasma levels are decreased by 52% and 72%, respectively (geom. means). The relative bioavailability of the prasugrel free base compared to prasugrel hydrochloride, both in the presence and in the absence of the proton pump inhibitor lansoprazole, differs so much that most probably a generic formulation containing prasugrel free base will not be equivalent in all aspects to the originator product.

Kind regards
Dan

❝ Dear Sir,

    ▲

Hi Sam,
I called you “Madam” in my last post as a joke. I don’t know your sex. When you call me Sir, that’s fine. But in a first post to all members of the forum, please avoid a sex-specific salutation in the future. The forum is not a “males only” club.

❝ You have clearly stated all the recommendations given by the USFDA and it is not very clear to me.

In which respect? FDA’s guidance is comprehensive. Two studies, fasting + fed.

❝ Sir I have one doubt that in case of fasting study the ratio is very good and it is around 100 for all the three parameters like AUCs and Cmax. But in case of the Fed study the ratio for AUCs are good and it is around 100 but the ratio of Cmax is very abrupt which is 130. Ideally if the ratio changes, it should change in all the parameters but in my case all the parameters are coming in range but the Cmax is only outside the range.

↑ This is a misconception. C_max is a surrogate for the rate of absorption (k_a). Did you see a sub­stantial difference in t_max between T and R? Imagine a situation where both the absorption rate and fraction absorption (f) of test and reference are different.
Example (one-compartment, no lag time, k_a 0.6 h^-1 (R) and 3.905 h^-1 (T), f 1 (R) and 0.85552 (T); linear trapezoidal rule for simplicity):
time      C(R)     C(T)
 0.00      BQL      BQL
 0.25     11.46    51.21
 0.50     21.04    69.24
 0.75     29.01    74.80
 1.0      35.60    75.69
 1.5      45.41    73.39
 2.0      51.75    70.01
 2.5      55.57    66.62
 3.0      57.55    63.38
 3.5      58.22    60.29
 4        57.96    57.35
 6        52.15    46.95
 9        40.21    34.78
12        30.04    25.77
16        20.18    17.27
24         9.07     7.76     T/R
——————————————————————————————————
AUCt     748.96   762.32   101.78%
AUC∞     839.85   839.93   100.01%
AUCext    10.82%    9.24%          ← ≤20% (sufficient sampling)
Kel        0.10     0.10           ← identical elimination (drug specific)
Cmax      58.22    75.69   130.00%
tmax       3.5       1.0           ← shift in t_max due to faster absorption
Lower extent of absorption together with a faster rate of absorption will result in 100% for AUC_∞, but 130% for C_max.

❝ So I am wondering that is it a really the formulation problem or there is problem with the food given as breakfast and all other parameters which can likely effect the formulation release.

Both is possible. We simply don’t know.

❝ If I will reformulate and I will go to control the Release of the drug from the formulation in the fed condition as per the current data then it may likely impact the fasting data also which has a ratio of around 100 and it may come down to 90 also.

Possible. It seems that test and reference are subjected to a different food effect. You have to find a compromise.

❝ So sir please suggests me the best way as you have stated in the previous reply.

You can try to find a dissolution method which is discriminatory. There’s a problem that in vivo you are measuring the metabolite and in vitro the parent. You can only cross fingers that the enzyme system does not get saturated by different initial concentrations. Start with biorelevant media (FaSSIF, FeSSIF). If you have (a lot) of money to spare, try an artificial digestive system (f.i. TNO’s TIM-1).

Hi Sam,

Did you look at the individual data (and profile) for the fed study. Any strange individuals?

John

Dear Krishna,

❝ We are going prasugrel for EU submission. Kindly advise which metabolite to be evaluate (i.e R 95913 or R 138727) and show bio equivalence.

See my post above. R 95913 is inactive and metabolized to the active R 138727. If you have strong evidence that it is not possible with state of the art bioanalytics to measure the parent you may go with either of them. Personally I think EMA would prefer R 95913 since R 138727 is further “downstream” and differences (especially in C_max) will be less pronounced. If in doubt go for a scientific advice. Suggestion: Select a ‘difficult’ agency, e.g., the Spanish one.