2017-12-01 13:01

Posting: # 18030
Views: 1,265

 Highly variable BE study for a BCS Class III drug [Design Issues]

Dear All,

We are currently working on a BCS class III drug. Drug has almost no bioavailability. Absolute bioavailability is around 1% based on literature survey and bioavailability is around 5-10% but molecule is metabolized drug first pass. We requested biowaiver for this product but it was rejected.

Molecule is far highly soluble in all physiological pH. Regarding drug product, dissolution tests were performed at 50 rpm in order to provide a discriminative method. First raw time, second raw dissolution%, third raw RSD%

In 0.1 HCl, very rapid dissolution is achieved.
 5'   10'   15'   20'   30'   45'
75.7  81.8  82.4  82.8  83.3  83.9
 8.4   6.3   5.7   5.7   5.4   5.2

In p.H 4.5, very rapid dissolution is achieved, too.
 5'   10'   15'   20'   30'   45'
72.3  82.6  83.7  85.1  85.6  87.4
 9.9   7.6   8.6   6.1   6.0   5.2

In p.H 6.8, dissolution is slower in a highly variable manner.
 5'   10'   15'   20'   30'   45'
33.3  59.7  66.3  75.6  78.7  80.1
85.9  41.5  38.0  11.6  10.2   8.8

In Water, dissolution is slower in a highly variable manner, too.
 5'   10'   15'   20'   30'   45'
37.3  59.8  66.7  66.7  70.6  70.1
49.7  24.5  22.2  19.1  15.2  18.3

We also tested dissolution at 75 rpm, nevertheless, it is impossible to see discriminative dissolution among different formulations under this condition.

According to literature (very limited), tmax is either 20 mins or 30 mins which means dissolution within 15 minutes is crucial. As per results drug is highly variable within 15 minutes. Even in disintegration test, some reference tablets disintegrate in 4 minutes whereas some in 15 minutes (medium: water). That is because of coating of reference drug product. Test product we developed has very similar dissolution profile with that of reference product (F2>60). We will conduct BE study for this product however we can not judge whether to design study as semi replicate or full replicate.

I highly appreciate if you can reply to my questions below:
  • Considering we choose semi replicate design (n=24), is it only possible to widen BE criteria (ISCV>30%)? No more advantages?
  • Considering we choose full replicate desing (n=24), is it only possible to widen BE criteria (ISCV>30%) and see formulation quality of test product as well? No more advantages?
  • Which one would be better (in terms of enlarging population), a semi replicate design (n=36), or full replicate design (n=24/28)?
  • I do not know if all values for R and T products are pooled in different periods if we design a full replicate design. In other words, does a four way crossover replicate study (n=24) have a power equal to two way crossover study (n=48)?
Last, our test product is not variable in dissolution. RSD% for all media is less than half of reference product. Therefore we believe our test product in terms of variability much more than reference product.

Thank you in advance.

Kind Regards,

P.s. I am sorry if presented data are not seen well.

Concerning the P.S.: Tabulators changed to spaces and BBcoded; see also this post #6. [Helmut]

 Thread view
Bioequivalence and Bioavailability Forum |  Admin contact
19,604 posts in 4,158 threads, 1,340 registered users;
online 9 (0 registered, 9 guests [including 8 identified bots]).
Forum time (Europe/Vienna): 17:08 CEST

Nothing in the world is more dangerous
than sincere ignorance
and conscientious stupidity.    Martin Luther King, Jr.

BEBAC Ing. Helmut Schütz