Loky do
☆    

Egypt,
2020-07-15 12:57
(140 d 07:25 ago)

Posting: # 21698
Views: 1,078
 

 Subject non compliance [Study As­sess­ment]

Dears

if a subject didn't have any evaluable concentrations for only test product all intervals are BLQ and according to our SOPs and SAP, the BLQ is considered "zeros", could be excluded from PK and statistical analysis?

thanks


Edit: Category changed; see also this post #1[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-16 14:30
(139 d 05:52 ago)

@ Loky do
Posting: # 21714
Views: 934
 

 Subject non compliance (or product failure?)

Hi Loky do,

» if a subject didn't have any evaluable concentrations for only test product all intervals are BLQ and according to our SOPs and SAP, the BLQ is considered "zeros", could be excluded from PK …

What does you protocol say? See also this recent thread.
Even if the outcome of investigations is negative, likely it would ring the alarm bells of regulators, esp. since you found no concentrations after the test product. What if it was a product failure (example: bad coating of a gastro-resistant PPI)? Regulators are more relaxed about the reference product since it “works” in clinical practice despite occasional failures. The EMA accepts exclusion in exceptional cases:
  • MR-GL Section 6.2.3
    […] non-existing or aberrant concentration profiles. If the incidence of this outlier behaviour is observed with a comparable frequency (e.g. the number of cases is not numerically higher in the test product) in both, test and reference product, data of a period with non-existing or aberrant profile can be excluded from statistical analysis provided that it has been pre-specified in the study protocol. In a 2-period trial this will result in the subject being removed from the analysis.
  • Dasatinib
    In order to avoid the bias introduced by the randomly occurring low-lier values under fasting
    con­ditions, it is considered acceptable that low-lier profiles can be excluded from statistical analysis […] if they occur with the same or lower frequency in the test product compared to the reference product.
(my emphases)

In your case (test only) cards are stacked against you.

» … and statistical analysis?

$$\lim_{x \to 0} \log x=-\infty.$$For simplicity we can say that \(\small{\log 0}\) is undefined. Hence, the common analysis based on log transformed PK metrics without excluding the subject is not possible at all.

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Loky do
☆    

Egypt,
2020-07-16 22:29
(138 d 21:52 ago)

@ Helmut
Posting: # 21724
Views: 907
 

 Subject non compliance (or product failure?)

Many Thanks, Helmut for your reply

» ... it is considered acceptable that low-lier profiles can be excluded from statistical analysis […] if they occur with the same or lower frequency in the test product compared to the reference product.
» (my emphases)

Is this case-specific for this product, and its formula? or it could be followed for other products with the same cases?

thanks in advance:-)
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-17 01:01
(138 d 19:21 ago)

@ Loky do
Posting: # 21727
Views: 899
 

 Bad science (statistics is taboo)

Hi Loky do,

» » ... it is considered acceptable that low-lier profiles can be excluded from statistical analysis […] if they occur with the same or lower frequency in the test product compared to the reference product.
»
» Is this case-specific for this product, and its formula?

Yes. It’s based on numerous replicate design studies reviewed by the PKWP. AFAIK, in all more outliers were seen after the reference. It’s a terrible product showing also funky batch-to-batch variability.

» or it could be followed for other products with the same cases?

If it’s a delayed release product, see my previous post. However, note the subtle different wording “number” vs. “frequency”.

In the draft MR-GL “comparable frequency” was used as well. From a statistical point of view that calls for a two-sided test. When I showed an example at the GL-meeting in Bonn (2013) I stirred up a hornets’ nest. Imagine: 64 subjects, 4 outliers after T and 3 after R. Is the frequency comparable? Yes, the p-value is 0.7139!1 The members of the PKWP did not like that at all. Oops, one (!) more. Changed in the final GL. Not acceptable. You can have it even more extreme. A 4-period full replicate study, one outlier (0.78%) after T and none after R → p-value 0.5.2 Not acceptable because the bloody number is higher!
Lesson learned: Bad science (statistics is taboo). If you have one more outlier after T than after R – independent of the sample size – you’re dead.


  1. Two-sided test (is T = R?)
    subj <- 64
    per  <- 2
    seq  <- 2
    OL.R <- 3
    OL.T <- 4
    n.T  <- n.R <- subj*per/seq
    outliers  <- matrix(c(OL.T, n.T-OL.T, OL.R, n.R), nrow = 2,
                        dimnames=list(Guess = c("T", "R"),
                                      Truth = c("T", "R")))
    fisher.test(outliers, alternative = "two.sided")

            Fisher's Exact Test for Count Data
    data:  outliers

    p-value = 0.7139
    alternative hypothesis: true odds ratio is not equal to 1
    95 percent confidence interval:
      0.2296229 10.0829801
    sample estimates:
    odds ratio
      1.418408


  2. One-sided test (is T > R?)
    subj <- 64
    per  <- 4
    seq  <- 2
    OL.R <- 0
    OL.T <- OL.R + 1
    n.T  <- n.R <- subj*per/seq
    outliers  <- matrix(c(OL.T, n.T-OL.T, OL.R, n.R), nrow = 2,
                        dimnames=list(Guess = c("T", "R"),
                                      Truth = c("T", "R")))
    fisher.test(outliers, alternative = "greater")

            Fisher's Exact Test for Count Data

    data:  outliers

    p-value = 0.5
    alternative hypothesis: true odds ratio is greater than 1
    95 percent confidence interval:
     0.05262625        Inf
    sample estimates:
    odds ratio
           Inf

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Loky do
☆    

Egypt,
2020-07-22 11:29
(133 d 08:53 ago)

@ Helmut
Posting: # 21774
Views: 756
 

 unexpected behavior for test product

Thanks for your reply :-):-)

Now another case for a non highly variable drug, a bioequivalence study performed on 27 subjects, where only 2 subjects had an odd behavior for test product (almost double the concentration of the reference product also double concentration of the rest of test product's subjects) and accordingly the study failed :-| and by excluding the probability of double dosing the 2 subjects by mistake (accountabilities are correct), and the results of incurred sample reanalysis for both are satisfactory :confused:
Is it applicable to consider these subjects as outliers, is there any way to conclude this? or could we re-dose them again to be confirmed? or simply the product fails :-|?

thanks :-)
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-22 12:03
(133 d 08:19 ago)

@ Loky do
Posting: # 21775
Views: 753
 

 Subject-by-Formulation interaction?

Hi Loky do,

» Now another case for a non highly variable drug, […]

What you confirmed already:
  • Subjects were not erroneously administered two doses of T.
  • Subjects are not poor metabolizers or “super-absorbers” (since you observed “normal” concentrations after R).
  • No problems in bioanalytics.
What is still possible:
  • A Subject-by-Formulation interaction, i.e., due to different properties of the products (excipients, manufacturing, …) T performs ‘better’ than R in these subjects – possibly belonging to a sub-population.
    However, an S×F-interaction is extremely rare (the ‘Two Lászlós’ doubt that it exists at all and is not just a statistical artifact). Without a replicate design cards are stacked against you (see the FDA’s guidance, Section C.2).

» Is it applicable to consider these subjects as outliers, is there any way to conclude this?

That’s similar to the cases we discussed before. Even if you would perform a statistical test (which is not acceptable acc. to current regulatory practice), you would get a p-value of 0.2453 which would speak against excluding them. Even for Health Canada (where an outlier assessment of studentized model residuals can be done if pre-specified in the protocol) you could exclude only one subject (≤5% if the sample size is ≥20).

» or could we re-dose them again to be confirmed?

Re-dosing is discouraged and likely not accepted.

» or simply the product fails :-|?

I’m afraid, yes.

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Loky do
☆    

Egypt,
2020-07-26 11:30
(129 d 08:51 ago)

@ Helmut
Posting: # 21791
Views: 560
 

 Subject-by-Formulation interaction?

Thanks a lot :-)
Activity
 Admin contact
21,230 posts in 4,427 threads, 1,481 registered users;
online 7 (1 registered, 6 guests [including 5 identified bots]).
Forum time: Wednesday 19:22 CET (Europe/Vienna)

A big data-analyst is an expert
in producing misleading conclusions from huge datasets.
It is much more efficient to use a statistician,
who can do the same with small ones.    Stephen Senn

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