Archived Content

The content on this page is provided for reference purposes only. This content has not been altered or updated since it was archived.

Vaccines, Blood & Biologics

Threshold Pertussis Antibody Levels Memo - May 27, 2009 - Hiberix

From:                     Farizo, Karen

Sent:                      Wednesday, May 27, 2009 11:28 AM

To:                         Humbert, Jason (CBER)

Subject:                 FW: Hiberix Study Hib-097


Please include these Emails in the Hiberix BLA.

Subsequent to the Emails below, Drusilla and I spoke by phone about the threshold pertussis antibody levels proposed by GSK for evaluation of concomitant Pediarix in Study Hib-097. As per Drusilla's recommendations conveyed in the Emails, we will accept GSK's proposed definitions for PT, FHA, and pertactin seroresponse post-dose 3 Pediarix-- i.e., the percentage of subjects with a titer above a threshold level that leads to 95% seroresponse in the control group.  




From:                     Burns, Drusilla L. 

Sent:                      Thursday, May 21, 2009 12:13 PM

To:                          Farizo, Karen

Subject:                RE: Hiberix Study Hib-097


I too was tempted to ask for the 90% level, but I became hesitant when I looked at some RCD curves.  If we use a threshold level that is on the very steep part of the curve, then relatively small differences in response titers can give fairly large differences in the proportion of responders.  A 10% difference may then become too stringent of a requirement; it might require large sample sizes to rule out a 10% difference.

Am I thinking about this correctly?  Your thoughts are greatly appreciated since use of threshold titers for pertussis is new and I don't have a lot of experience with it.



From:             Farizo, Karen 

Sent:              Thursday, May 21, 2009 12:02 PM

To:                  Burns, Drusilla L.

Subject:        RE: Hiberix Study Hib-097


My only concern was whether a threshold level corresponding to a titer achieved by 95% of control Ss might be too high-- as you stated, they are "just barely" on the linear portion of the RCD curves. I suppose using a titer achieved by 90% of control Ss may be more sensitive, but asking for this instead of the 95% may be "splitting hairs". I trust/defer to your judgement on this, and if you are ok with what GSK has proposed, I am fine with that. It sure beats just looking at seropositivity (>5 EU/ml).

I will ask GSK to include the RCD curves in the study report.

Thanks again,



From:             Burns, Drusilla L. 

Sent:              Thursday, May 21, 2009 11:47 AM

To:                  Farizo, Karen; Ghebregiorgis, Ghideon

Subject:        RE: Hiberix Study Hib-097


Regarding the proposed threshold antibody levels, I think that a threshold level defined as "the titer achieved by 95% of individuals in the control group after vaccination with a 3 dose primary series" is a definition that is reasonably likely to have adequate sensitivity to detect meaningful differences between vaccine groups. In order to have sufficient sensitivity to detect differences, the threshold level must be significantly greater than pre-vaccination levels and should be on the "linear portion of the dose response curve".  GSK will not measure pre-vaccination titers in study HIB-097 (which is the whole point of using threshold levels) and therefore we will not know for certain the relationship between the threshold titers and pre-vaccination titers for that study.  However, we can glean some information from HibMenCY-TT-005.  In that study, pre-vaccination GMCs were 3.4, 8.7, and 5.1 EU/ml for PT, FHA, and PRN, respectively whereas the threshold levels (defined as that achieved by 95% of the individuals after vaccination) were significantly higher (25, 80, and 15 EU/ml, for PT, FHA, and PRN, respectively).  Also, if you look at the RCD curves, the 95% point is on the "linear portion of the dose response curve" (although just barely) and not on the upper asymptote of the curve.  Therefore, if the concomitant vaccine group responded less well to the vaccine, using this threshold level, we should detect differences.

Of note, the threshold definition proposed by GSK is essentially identical to the one that we settled on with Wyeth for a concomitant vaccination study with 13-valent Prevnar.

So I think that this definition for threshold level is a reasonable one.  If we have not already done so, we should state that we would also like to see the RCD curves from the study in the final study report.

Does all this sound reasonable to you? 



From:     Farizo, Karen 

Sent:      Wednesday, May 20, 2009 12:36 PM

To:          Burns, Drusilla L.; Ghebregiorgis, Ghideon

Cc:          Slater, Jay

Subject:                Hiberix Study Hib-097

Drusilla and Ghideon,

Could you please provide feedback on Item 7 in the attached document. Item 7 is GSK's response to our IR letter comment regarding the evaluation of post-dose 3 threshold levels of anti-PT, FHA, and PRN, in lieu of pre- to post-vx fold-rises, in Hiberix study Hib-097.

I do not really understand the statistical methods proposed for the comparison of the Hiberix arm to the ActHIB arm. 

Regarding the proposed threshold antibody levels-- GSK indicates that they are based on data from Study HibMenCY-TT-005. The report for this study is available in the EDR: IND ---b(4)---, module hib-mency-tt-005. The pertussis RCD curves are in Supplements 65, 66, 67, 115, 116, and 117 of the report.



 << File: Response to CBER 23Apr09 Hib-097 Review.pdf >>

Contact FDA

(800) 835-4709
(240) 402-8010
Consumer Affairs Branch (CBER)

Division of Communication and Consumer Affairs

Office of Communication, Outreach and Development

Food and Drug Administration

10903 New Hampshire Avenue

Building 71 Room 3103

Silver Spring, MD 20993-0002

Page Last Updated: 03/23/2015
Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.
Language Assistance Available: Español | 繁體中文 | Tiếng Việt | 한국어 | Tagalog | Русский | العربية | Kreyòl Ayisyen | Français | Polski | Português | Italiano | Deutsch | 日本語 | فارسی | English