Sukalpa Biswas
☆    

India,
2020-05-26 14:47
(1402 d 23:31 ago)

Posting: # 21468
Views: 1,825
 

 ANVISA Regulation for NTI [Regulatives / Guidelines]

For NTI product for USFDA, we follow product specific guidance where 4WC study is mandatory with statistical criteria discussed below,

Un scaled Average Bioequivalence approach
• Primary pharmacokinetic parameter(s) for the 90% confidence interval of the geometric least square means must fall within 80.00% to 125.00% (both inclusive).
Reference Scaled Average Bioequivalence
• SWR will be determined for Primary pharmacokinetic parameter(s) of Cmax, AUC0-t and AUC0-∞.
The following criteria will be estimated for Primary pharmacokinetic parameter(s):
• The 95% upper confidence bound for (μT-μR)2-θ*S2WR must be ≤ 0
Where μT and μRare mean of test and reference formulations on ln-transformed scale, respectively, and θ = (In (1.11111)/ (σW0)2 (Scaled average BE limit); where σW0 = 0.10 (regulatory limit).
• The point estimate of the Test/Reference geometric mean ratio must fall within [0.80, 1.25] 90% CI for the ratio of the within subject SD of Test product to Reference product σWT/σWR
• The within-subject standard deviation of test and reference products will be compared, and the upper limit of the 90% confidence interval for the test-to-reference ratio of the within-subject variability should be ≤ 2.5.


For NTI product of ANVISA studies,
1. if 95% CI need to be ensured within 80-125, then can we prefer doing 2WCstudies(ISCV<30%)?
2. If ISCV>30%, can we go for full replicate/partial replicate design? If yes then what will be the statistical approach?
3. Is it necessary to get the statistical approach approved from ANVISA regulatory prior initiation of the study.


Edit: Please follow the Forum’s Policy[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-05-26 15:39
(1402 d 22:39 ago)

@ Sukalpa Biswas
Posting: # 21470
Views: 1,431
 

 ANVISA Regulation for NTI

Hi Sukalpa,

❝ For NTI product of ANVISA studies,

❝ 1. if 95% CI need to be ensured within 80-125, then can we prefer doing 2WCstudies(ISCV<30%)?


Calculating a 95% CI (instead of the common 90% CI) was dropped a while ago and for good reasons. It does not decrease the patient’s risk. That was a misconception by the ANVISA. Only a narrower acceptance range (EMA and most other agencies) or reference scaling acc. to the method of the FDA does.

See the draft of 27 Dec 2019 (in Portuguese). Inofficial Translation:

Chapter VI, Section V, Narrow therapeutic index drugs (NTID)
Article 82. In specific cases of narrow therapeutic index drugs, the acceptance range for AUC should be adjusted to 90.00 –  111.11%; the same can happen with Cmax where safety, efficacy or drug level monitoring is of particular importance.


❝ 2. If ISCV>30%, can we go for full replicate/partial replicate design? If yes then what will be the statistical approach?


It is extremely unlikely that an NTID shows high variability. The only ones I know are dagibatran and rivaroxaban. Whether you opt for a 2×2×2 or a replicate design, up to you. The evaluation has to be done for ABE. If you expect AEs in healthy volunteers and a high dropout rate, I would avoid a replicate design.

❝ 3. Is it necessary to get the statistical approach approved from ANVISA regulatory prior initiation of the study.


The protocol has to be approved by the ANVISA anyway. It is up to you to give a clinical justification for the limits (Cmax, AUC, both). Examples from the EMA’s product-specific guidances:
  • Colchicine
    AUC 90.00–111.11%, Cmax 80.00–125.00%
  • Everolimus
    Both 80.00–125.00% (oncologic-only indication, 2-10 mg)
    AUC 90.00–111.11%, Cmax 80.00–125.00% (transplant-only indication, 0.1–1 mg)
  • Sirolimus
    AUC 90.00–111.11%, Cmax 80.00–125.00%
  • Tacrolimus
    AUC 90.00–111.11%, Cmax 80.00–125.00%

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
106 visitors (0 registered, 106 guests [including 8 identified bots]).
Forum time: 13:18 CET (Europe/Vienna)

With four parameters I can fit an elephant,
and with five I can make him wiggle his trunk.    John von Neumann

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5