[R] R and FDA trials
gregory.warnes at mac.com
Mon Oct 8 21:24:43 CEST 2007
I'm excited to see that R-Core has released the document "R:
Regulatory Compliance and Validation Issues
A Guidance Document for the Use of R in Regulated Clinical Trial
I know it represents a great deal of effort on the part of R core,
and it has taken many months to achieve.
Random Technologies, LLC (http://random-technologies-llc.com) has
been extended this work to cover additional R packages, as well as
offering additional documentation, products, and tools that
facilitate use of R for mission-critical and regulated contexts.
(Lots of this work falls into the "Make your *boss* happy" context,
<grin>). These efforts are supported by the sale of support
contracts and related services.
We are very interested in working with R users and developers to
identify additional 1) packages, 2) features, and 3) documents which
will (with appropriate documentation and testing, of course) make R
easier to use (both functionally and politically) for FDA sumbissions.
Please send us your suggestions and comments at <comments at random-
Gregory R. Warnes, Ph.D.
Random Technologies, LLC.
On Oct 8, 2007, at 8:54AM , Marc Schwartz wrote:
> On Sun, 2007-10-07 at 21:25 -0500, Frank E Harrell Jr wrote:
>> Ricardo Pietrobon wrote:
>>> Yesterday I just noticed the new document on R and regulatory
>>> for biomedical research posted at
>>> Coming from an institution that performs a large number of clinical
>>> trials for FDA and being an advocate of R myself, I have found that
>>> the following issues usually come up when discussing the use of R
>>> FDA trials:
>>> 1. Most FDA submissions come down to a series of r x k tables,
>>> and it
>>> is hard to claim that one system is better than another for that.
>>> 2. Data is to be submitted to the FDA in SAS (considered by many as
>>> the industry standard) or CDISC XML formats (http://www.cdisc.org/);
>>> there are pretty good SAS tools for that; does R have comparable?
>>> 3. Some packages in R provide acknowledgedly better functionality
>>> their SAS-equivalent, but an entire FDA validation would have to
>>> each time an enhancement is made to the R package because often an
>>> enhancement breaks something else or the syntax would change from
>>> release to another.
>> Your item 3. is up to the company's policy. FDA does not require
>> it and
>> the word "validation" is not well defined in this context. No
>> does a complete validation any time any piece of the package is
> Just to augment Frank's response:
> With respect to point number 2, I believe the only R based package
> offers the ability to write SAS XPORT format files at this point is
> SASxport CRAN package by Greg Warnes et al. There are several,
> Frank's Hmisc package and the standard foreign package that will read
> XPORT files.
> Keep in mind that it is the SAS XPORT files, not the proprietary SAS
> files, that the FDA wants, since the XPORT format is openly
> There are other third party packages, such as DBMS/Copy and
> Stat/Transfer, that will also write XPORT format files.
> At some point, CDISC based exports will likely need to be looked at by
> one or more folks motivated to do so. In the mean time, you would need
> to consider third party tools for that. My recollection of Mat
> comments from our panel at useR back in August, is that the FDA itself
> is not yet at the point where CDISC based file exchanges are without
> Finally, bear in mind that the document to which you refer covers
> specifically listed packages released via the R Foundation. See
> 2 of the document. Other CRAN/non-CRAN R packages are not covered. In
> either case, you would need to engage in appropriate risk mitigation
> processes when using those and as Frank notes, these would be internal
> procedures and policies as deemed appropriate within the GxP
> Marc Schwartz
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help