Environmental and Molecular Mutagenesis 48:354^358 (2007)

Perspective

Implementing the U.S. FDA Guidance on

Pharmacogenomic Data Submissions
Federico Goodsaid* and Felix W. Frueh
Genomics Group, Office of Clinical Pharmacology, Office of Translational
Science, Center for Drug Evaluation and Research, U.S. Food and Drug
Administration, Silver Spring, Maryland 20903-0002

The FDA Guidance for Industry: Pharmacogenom- the past 2 years by a number of initiatives. These
ics Data Submissions was issued in 2005. This initiatives have included the development of the
guidance document covers a broad area associ- Interdisciplinary Pharmacogenomics Review Group
ated with how and when to submit genomic data to for review of pharmacogenomic data submissions,
the FDA. Additional tasks associated with genomic the pilot process for qualification of biomarkers,
data submissions include the implementation of and the concept paper on recommendations for the
genomic data submissions; the process for qualifi- generation and submission of genomic data. These
cation of exploratory biomarkers into valid bio- initiatives have contributed to the effective imple-
markers; and technical recommendations for the mentation of the Pharmacogenomics Guidance at
generation and submission of genomic data to the the FDA. Environ. Mol. Mutagen. 48:354–358,
FDA. These tasks have been addressed throughout 2007. Published 2007 Wiley-Liss, Inc.{

Key words: genomic; biomarker; regulatory

INTRODUCTION  Concept paper on recommendations for the generation

and submission of genomic data [U.S. FDA, 2006a]—
In anticipation of a significant increase of the use of Preliminary document summarizing consensus and dis-
genomic data in drug development and to foster the use cussions on recommendations for the generation, analy-
of such information, the Food and Drug Administration sis, submission, and review of genomic data to the
(FDA) issued a document: Guidance for Industry: Phar- FDA.
macogenomic Data Submissions [U.S. FDA, 2005a]
(‘‘Pharmacogenomics Guidance’’) in final form in March VOLUNTARY GENOMIC DATA SUBMISSIONS: A MODEL
2005. Catalyzed by this guidance, applications of genomic FOR INNOVATION THROUGH REGULATION
data in drug development have been integrated at the
FDA by several activities, for example: Genomic information has been integrated in drug devel-
opment for a number of years, but the pharmaceutical
 Voluntary genomic data submissions [U.S. FDA, industry requires a clear view of how this information
2005b]—Novel way to share information with the FDA. will be reviewed by the FDA. The pharmacogenomic
Exploratory or research data for which the Guidance guidance defined what type of genomic data needs to be
does not require that these data be submitted to an inves- submitted to the agency and when, and introduced the
tigational new drug (IND), or that complete reports be
submitted to an new drug application (NDA) or biologics
license application (BLA). Voluntary submissions benefit Views expressed in this manuscript are those of the authors and not nec-
both the industry and the FDA by ensuring that regula- essarily those of the U.S. Food and Drug Administration.
tory scientists are familiar with and prepared to appropri- *Correspondence to: Federico Goodsaid, Genomics Group, Office of
ately evaluate future genomic submissions. Clinical Pharmacology, Office of Translational Science, Center for Drug
 Biomarker qualification pilot process [Goodsaid and Evaluation and Research, U.S. Food and Drug Administration, 10903
Frueh, 2006]—Process developed to qualify biomarkers New Hampshire Avenue, Building 21, Room 3663, Silver Spring, MD
20903-0002, USA. E-mail: [email protected]
for use in submissions to the FDA. This process covers
the need identified in the Guidance for qualification of Received 15 January 2007; in final form 9 February 2007
‘‘exploratory’’ into ‘‘probable valid’’ or ‘‘known valid’’ DOI 10.1002/em.20294
biomarkers. Published online in Wiley InterScience (www.interscience.wiley.com).

Published 2007 Wiley-Liss, Inc. {This article is a US Government work and,

as such, is in the public domain in the United States of America.
 Environmental and Molecular Mutagenesis. DOI 10.1002/em
Implementation of Pharmacogenomic Guidance 355

concept of ‘‘Voluntary Genomic Data Submissions’’ integration in submissions to the FDA. A qualification
(VGDSs). VGDS submissions teach both sponsors as well process at the FDA must first identify the context in
as reviewers through multiple examples how genomic which exploratory biomarkers are proposed, and then gen-
data is generated and analyzed by sponsors and how it erate a scientific consensus within the agency and, if
should be reviewed by the FDA. needed, within advisory committees convened for this pur-
VGDS has reached a milestone in 2006 of 2 years since pose, that data submitted for qualification supports the
implementation. Over 30 submissions have been received context proposed. This process would not preclude inde-
in this program, which is supported by the Interdiscipli- pendent efforts for qualification in contexts different from
nary Pharmacogenomic Review Group (IPRG) at the those proposed to the agency.
FDA. The IPRG is a FDA-wide, interdisciplinary group VGDS data have opened a productive discussion about
created to ensure high-quality review of these voluntary biomarker qualification at the FDA. Preclinical and clini-
submissions, while also having the responsibility to ensure cal biomarkers have been traditionally accepted for regu-
proper partitioning of voluntary from nonvoluntary data. latory review over the course of time, seldom with a
The VGDS program was launched to discuss exploratory focused debate and rarely has an active consensus build-
pharmacogenomic data sets, addressing a need for infor- ing process been used. While a passive process may be
mal interaction between sponsors and regulators to evalu- intellectually painless, it has often required many years to
ate such exploratory data with no immediate regulatory reach a consensus on qualification. The urgent need for
impact. an accelerated application of biomarkers in drug develop-
VGDS submissions have not only helped communicate ment requires an enabling process for accurate, compre-
genomic data between pharmaceutical companies and the hensive, and aggressive qualification of biomarkers from
FDA, but have also had a positive impact on the content the perspective of the FDA. This is not a trivial extension
of genomic data in regulatory submissions. As we have of the review of drug submissions: it is a new and com-
approached the 30th VGDS submission, we have also had plex structural problem that requires a collaborative effort
increasing numbers of consults for regulatory submissions throughout the agency.
by the Genomics Group in the Office of Clinical Pharma- The Pharmacogenomics Guidance classifies biomarkers
cology at CDER, indicating that industry is getting more as exploratory, probable valid, or known valid. However,
comfortable sharing this type of information with the it does not describe a process by which an exploratory
FDA. biomarker can be qualified as a valid biomarker. A known
Approximately two-third of these VGDS submissions valid biomarker is defined as ‘‘A biomarker that is meas-
have focused on clinical study design issues. The remain- ured in an analytical test system with well-established
ing submissions include both toxicogenomic as well as performance characteristics and for which there is wide-
other genomic data, including data from prototype phar- spread agreement in the medical or scientific community
macogenetic devices. Clinical submissions have had data about the physiologic, toxicologic, pharmacologic, or clin-
associated with multiple types of oncology therapies as ical significance of the results.’’ A qualification process to
well as with Alzheimer’s disease, hypertension, hypogly- bridge the gap between exploratory and known valid bio-
cemia, depression, obesity, and rheumatoid arthritis. Joint markers must be closely aligned with the review of drug
VGDS meetings with the European Medicines Agency submissions at the FDA. It is imperative that FDA and its
[U.S. FDA, 2005b] have helped generate consensus on sponsors collaborate in the design of an efficient qualifica-
opportunities and limitations of genomic data in drug de- tion process and that FDA establishes a review structure
velopment and regulatory review. to reflect this context.
VGDS submissions have emphasized the need to pro- The FDA has worked over the past 2 years on the
vide additional information in areas of the Pharmacoge- design of a qualification process map [Goodsaid and
nomics Guidance that have not been described in full Frueh, 2006], reflecting qualification needs for preclinical
detail. For example, these areas include both the path for biomarkers. This process map reflects the expectation for
qualification of exploratory biomarkers into known valid a continued interaction between sponsors and the FDA in
biomarkers as well as technical recommendations on the the critical steps in this process, including initial evalua-
generation and submission of genomic data. These are tion, qualification protocol draft, and data review. One of
currently being addressed by IPRG and the Genomics the early proofs-of-concept for this process will be the
Group. review of the work of the Predictive Safety Testing Con-
sortium [U.S. FDA, 2006b] coordinated by the C-Path
BIOMARKER QUALIFICATION PILOT PROCESS Institute. This work is expected to lead to multiple pre-
clinical qualification submissions within the next year. It
Biomarker qualification at the FDA is needed both as a is anticipated that next to a thorough scientific review of
regulatory tool as well as to encourage accelerated identi- the data, the FDA will be able to confirm the assumptions
fication of new biomarkers in drug development and their of the proposed process map.
Environmental and Molecular Mutagenesis. DOI 10.1002/em
356 Goodsaid and Frueh

Fig. 1. Biomarker qualification pilot process at the FDA.

At the core of the process is a pilot structure (Fig. 1) to This IPRG Biomarker Qualification Review Team will
start a qualification process for biomarkers in drug devel- evaluate study protocols and review study results for the
opment. This pilot structure leverages the IPRG to allow purpose of qualifying novel biomarkers of drug safety.
comprehensive participation of expertise available at the This includes the evaluation of the appropriate preclinical,
FDA across different centers in the FDA, including clinical, and statistical considerations for the context in
CDER, CBER, CDRH, and NCTR, as well as across clin- which the biomarker is proposed. It is expected that the
ical divisions and pharm/tox reviewers in CDER. New expertise resulting from this function will also lead to the
responsibilities of the IPRG in this pilot structure include development of recommendations and future guidance for
the creation of a specific review function for the assess- the submission of biomarker data. The IPRG Biomarker
ment of biomarker qualification data sets: the IPRG Bio- Qualification Review Team will assess the original bio-
marker Qualification Review Team. marker context proposal through a voluntary data submis-
 Environmental and Molecular Mutagenesis. DOI 10.1002/em
Implementation of Pharmacogenomic Guidance 357

Fig. 2. Areas covered in concept paper on recommendations for the generation and submission of genomic data.

sion (VXDS, where the X underlines a wide range of data a consensus can be identified. Other areas in this applica-
sources) and then evaluate the qualification study protocol tion would benefit from an intensive discussion and scien-
together with the sponsor to reach a consensus protocol, tific exploration needed for a consensus on their use.
which if successful, will result in a regulatory decision On the basis of its experience with VGDSs as well as
about the use of the newly qualified biomarker for its with its review of numerous protocols and data submitted
determined, specific context. under IND applications, NDAs, and BLAs during the last
2 years, FDA believes that guidance [U.S. FDA, 2006a]
will benefit sponsors considering the submission of either
CONCEPT PAPER ON RECOMMENDATIONS FOR THE voluntary genomic data or marketing applications contain-
GENERATION AND SUBMISSION OF GENOMIC DATA ing genomics data. FDA plans to develop such guidance
based on the concepts in this paper and input from a pub-
A universal consensus on how to generate, analyze, and lic meeting. As technology changes and more experience
report genomic data is not yet available. There are some is gained, these concepts also may change. Figure 2 out-
areas in this application of genomic data, however, where lines areas covered in this concept paper. This is a timely
Environmental and Molecular Mutagenesis. DOI 10.1002/em
358 Goodsaid and Frueh

effort, both for its content as well as for the discussions valid biomarkers, and a concept paper has been published
and debates it is likely to generate. It summarizes experi- on recommendations for the Submission and review of
ence gained from the analysis of genomic data submis- genomic data.
sions and provides a reference document for future sub-
missions and analyses as well as for discussions at the
FDA regarding best practices for submission of genomic
data.
REFERENCES
Goodsaid F, Frueh F. 2006. Process map proposal for the validation of
SUMMARY genomic biomarkers. Pharmacogenomics 7:773–782.
U.S. FDA. 2005a. Guidance for industry—Pharmacogenomic data sub-
The Guidance for Industry: Pharmacogenomic Data missions. Available at http://www.fda.gov/cder/guidance/6400fnl.
Submissions set up a framework for the use and regula- pdf. Accessed on December 16, 2006.
tory review of genomic data at the FDA. The guidance U.S. FDA. 2005b. Guiding principles processing joint FDA EMEA vol-
introduced the process of voluntary submissions of untary genomic data submissions (VGDSs) within the framework
of the confidentiality arrangement. Available at http://www.
genomic data, which has been useful to exchange infor- emea.europa.eu/pdfs/general/direct/pr/FDAEMEA.pdf. Accessed on
mation and ideas about the application of this exploratory December 16, 2006.
science in drug development and the regulatory review of U.S. FDA. 2006a. Concept paper on recommendations for the generation
such information. The original guidance text did not spec- and submission of genomic data. Available at http://www.fda.
ify a process for biomarker qualification or technical rec- gov/cder/genomics/conceptpaper_20061107.pdf. Accessed on De-
cember 16, 2006.
ommendations on the submission of genomic data. To U.S. FDA. 2006b. Predictive safety testing consortium. Available at
address this need, a pilot process has been developed for http://www.fda.gov/oc/initiatives/criticalpath/projectsummary/con-
the qualification of exploratory biomarkers into known sortium.html. Accessed on December 16, 2006.