RSABE - 95%UCB estimation for Highly variable Drugs_FDA

Phoenix Certara Forums Phoenix WNL basics
1

Dear Team,

Is there any provision to determine Reference scaled average bioequivalence for Highly variable drugs?

In which, We are interested to calculate 95% Upper confidence bound. The parameter in which we are interested to estimate is critbound. How we can use WinNonlin (version 8.3.1.5014) to estimate it? Suggest if any additional module require for this.

We went through the AAPS Poster presentation,2013

https://www.certara.com/app/uploads/Resources/Posters/AAPS2013_RSABE_in_PHX_WNL.pdf

Note: I have an active license access towards Basic WinNonlin. (v 8.3.1.5014)

Kindly support by providing any relevant examples, Templates, Workflow etc.

Awaiting for your revert.

Regards,

d_nil

2

Dear D_nil,

You can find the updated templates for RSABE for HVD at the link below:

https://www.certarauniversity.com/store/2383230

These templates calculate the critical bound parameter you mention. Please make sure to download the templates for version 8.3.3. of Phoenix and read the instructions.

Sincerely,

Ana Henry

3

Hi Ana,

Thank you for giving valuable time.

Its meeting my criteria.

Thanks..! :slight_smile:

Regards,

d_nil

4

Hi d_nil,

did you initiate the poll? If yes, the second part of the question is wrong.

Widening the acceptance range for the 90% is applicable for Average Bioequivalence with Expanding Limits (ABEL) if CVwR > 30%. That’s the approach used in all jurisdictions except the U.S. FDA and China’s CDE.

For these two agencies you have to apply Reference-scaled Average Bioequivalence (RSABE). That’s a different pot of tea. If swR < 0.294 (CVwR < 0.3004689) you have to apply conventional ABE. If swR ≥ 0.294 you may apply RSABE.

See the FDA’s Draft Guidance – Appendix B and a comparison.

As shown in the article, you may face problems if you have to evaluate a partial replicate study for ABE (happens quite often if you may scale Cmax but not AUC due to its lower variability). Avoid the lousy partial replicate design whenever possible.

5

Hi Helmut,

The poll was not initiated,

I understand the differences among criteria for Both agencies with respect to ABE & RSABE.

Also, i appreciate your add-on comment on study design.

Regards,

d_nil

6

Note that the Assessing Reference Scaled Average Bioequivalence (RSABE) for Narrow Therapeutic Index Drugs (NTID) template has been updated. Please ensure you replace any earlier versions you may have saved with the updated file (provided August 2025). Two versions are provided: Phoenix 8.3.5 and Phoenix 8.6 (utilizing the FINV function).

Remember we are in the process of retiring the Sentinel server software used for licensing 8.4 and earlier; from March 2026 we will no longer be able to generate these older auth codes, so we highly recommend you to upgrade to 8.6.1, as this new licensing process via CAD (my.certara.net) makes acquiring licenses much easier

Also ALL official information can be found by logging into CAD (Certara Admin) and going to the Help centre; and then Software Documentation

I would recommend both users and any of your supporting team e.g. IT create accounts in CAD so you can manage acquire licenses for 8.5.x and later, plus of course all the other great resources here including a free e-book of Gabrielsson and Weiner/

Then **Phoenix 8.6 Documentation (**the official comparison is at https://help.certara.net/en/articles/11089683-phoenix-version-comparison-what-s-new-and-improved )