OT: Is randomization for targeted cancer therapies ethical?

[Bert Gunter]

Hi Folks:

**Off Topic**

Those interested in clinical trials may find the following of interest:

http://www.nytimes.com/2010/09/19/health/research/19trial.html

It concerns the ethicality of randomizing those with life-threatening
disease to relatively ineffective SOC when new "biologically targeted"
therapies "appear" to be more effective. While the context may be new,
the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
can't remember for sure) about this about 30 years ago. I'm sure many
other also have done so.

Cheers,

Bert
--
Bert Gunter
Genentech Nonclinical Biostatistics

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[cberry]

On Mon, 20 Sep 2010, Bert Gunter wrote:

> Hi Folks:
>
> **Off Topic**
>
> Those interested in clinical trials may find the following of interest:
>
> http://www.nytimes.com/2010/09/19/health/research/19trial.html
>
> It concerns the ethicality of randomizing those with life-threatening
> disease to relatively ineffective SOC when new "biologically targeted"
> therapies "appear" to be more effective. While the context may be new,
> the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
> can't remember for sure) about this about 30 years ago. I'm sure many
> other also have done so.

Anscombe's remarkable (and influential) review of Armitage's 'Sequential
Medical Trials' back in 1963

  http://www.jstor.org/stable/2283272

is worth a look by any statistician who is interested in this topic.

It makes explicit several factors that weigh in the ethical assessment of
a particular trial design.

He discusses in formal terms the weighing of outcomes for patients
in the trial at hand aginst those of future patients and the impact that
this might have on design decisions.

HTH,

Chuck

>
> Cheers,
>
> Bert
> --
> Bert Gunter
> Genentech Nonclinical Biostatistics
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Charles C. Berry                            [phone]
                                             Dept of Family/Preventive Medicine
E mailto:[hidden email]            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[David Winsemius]

On Sep 20, 2010, at 2:28 PM, Charles C. Berry wrote:

> On Mon, 20 Sep 2010, Bert Gunter wrote:
>
>> Hi Folks:
>>
>> **Off Topic**
>>
>> Those interested in clinical trials may find the following of  
>> interest:
>>
>> http://www.nytimes.com/2010/09/19/health/research/19trial.html
>>
>> It concerns the ethicality of randomizing those with life-threatening
>> disease to relatively ineffective SOC when new "biologically  
>> targeted"
>> therapies "appear" to be more effective. While the context may be  
>> new,
>> the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
>> can't remember for sure) about this about 30 years ago. I'm sure many
>> other also have done so.
>
> Anscombe's remarkable (and influential) review of Armitage's  
> 'Sequential Medical Trials' back in 1963
>
> http://www.jstor.org/stable/2283272
>

Not all university libraries have access via that link and efforts at  
identifying the citation in Pubmed failed, so could I request a more  
complete citation, please? Oh never mind I got it with Google. If  
anyone else needs the ISSN of JASA in which both the cited article and  
Armitage's reply appeared for the purposes of JSTOR access, it's  
01621459.

Sequential Medical Trials
F. J. Anscombe
Journal of the American Statistical Association
Vol. 58, No. 302 (Jun., 1963), pp. 365-383
Published by: American Statistical Association
http://www.jstor.org/stable/2283272

  Sequential Medical Trials: Some Comments on F. J. Anscombe's Paper
  P. Armitage
Journal of the American Statistical Association
Vol. 58, No. 302 (Jun., 1963), pp. 384-387
(article consists of 4 pages)
URL: http://www.jstor.org/stable/2283273

--
David.

> is worth a look by any statistician who is interested in this topic.
>
> It makes explicit several factors that weigh in the ethical  
> assessment of a particular trial design.
>
> He discusses in formal terms the weighing of outcomes for patients  
> in the trial at hand aginst those of future patients and the impact  
> that this might have on design decisions.
>
> HTH,
>
> Chuck
>
>
>>
>> Cheers,
>>
>> Bert
>> --
>> Bert Gunter
>> Genentech Nonclinical Biostatistics
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
> Charles C. Berry                            [phone]
>                                            Dept of Family/Preventive  
> Medicine
> E mailto:[hidden email]            UC San Diego
> http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego  
> 92093-0901
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
West Hartford, CT

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[JLucke]

Clearly inferior treatments are unethical.

Donald Berry at MD Anderson in Houston TX  and Jay Kadane at Carnegie
Mellon have been working on more ethical designs within the Bayesian
framework.  In particular, response adaptive designs reduce the assignment
to and continuation of patients on inferior treatments.






Bert Gunter <[hidden email]>
Sent by: [hidden email]
09/20/2010 01:31 PM

To
[hidden email]
cc

Subject
[R] OT: Is randomization for targeted cancer therapies ethical?






Hi Folks:

**Off Topic**

Those interested in clinical trials may find the following of interest:

http://www.nytimes.com/2010/09/19/health/research/19trial.html

It concerns the ethicality of randomizing those with life-threatening
disease to relatively ineffective SOC when new "biologically targeted"
therapies "appear" to be more effective. While the context may be new,
the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
can't remember for sure) about this about 30 years ago. I'm sure many
other also have done so.

Cheers,

Bert
--
Bert Gunter
Genentech Nonclinical Biostatistics

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[Jim Lemon]

On 09/21/2010 05:02 AM, David Winsemius wrote:
>...
> Not all university libraries have access via that link and efforts at
> identifying the citation in Pubmed failed, so could I request a more
> complete citation, please? Oh never mind I got it with Google. If anyone
> else needs the ISSN of JASA in which both the cited article and
> Armitage's reply appeared for the purposes of JSTOR access, it's 01621459.
>

For those of us not fortunate enough to have access to JStor directly
(and how I miss it), try this link:

http://ml.stat.purdue.edu/bayes/writings/anscombe.sequentialmedical.1963.pdf

Jim

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[Liaw, Andy]

> From: [hidden email]
>
> Clearly inferior treatments are unethical.

The Big Question is:  What constitute "clearly"?  Who or How to decide
what is "clearly"?  I'm sure there are plenty of people who don't
understand much Statistics and are perfectly willing to say the results
on the two cousins show the conventional treatment is "clearly
inferior".  Sure, on these two cousins we can say so, but what about
others?
 
> Donald Berry at MD Anderson in Houston TX  and Jay Kadane at Carnegie
> Mellon have been working on more ethical designs within the Bayesian
> framework.  In particular, response adaptive designs reduce
> the assignment
> to and continuation of patients on inferior treatments.
 
I've heard LJ Wei talked about this kinds of designs (don't remember if
they are Bayesian) more than a dozen years ago.  Don't know how common
they are in use.

Andy

>
>
> Bert Gunter <[hidden email]>
> Sent by: [hidden email]
> 09/20/2010 01:31 PM
>
> To
> [hidden email]
> cc
>
> Subject
> [R] OT: Is randomization for targeted cancer therapies ethical?
>
>
>
>
>
>
> Hi Folks:
>
> **Off Topic**
>
> Those interested in clinical trials may find the following of
> interest:
>
> http://www.nytimes.com/2010/09/19/health/research/19trial.html
>
> It concerns the ethicality of randomizing those with life-threatening
> disease to relatively ineffective SOC when new "biologically targeted"
> therapies "appear" to be more effective. While the context may be new,
> the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
> can't remember for sure) about this about 30 years ago. I'm sure many
> other also have done so.
>
> Cheers,
>
> Bert
> --
> Bert Gunter
> Genentech Nonclinical Biostatistics
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Notice:  This e-mail message, together with any attachme...{{dropped:11}}

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[Marc Schwartz-3]

On Sep 21, 2010, at 6:52 AM, Liaw, Andy wrote:

>> From: [hidden email]
>>
>> Clearly inferior treatments are unethical.
>
> The Big Question is:  What constitute "clearly"?  Who or How to decide
> what is "clearly"?  I'm sure there are plenty of people who don't
> understand much Statistics and are perfectly willing to say the results
> on the two cousins show the conventional treatment is "clearly
> inferior".  Sure, on these two cousins we can say so, but what about
> others?
>
>> Donald Berry at MD Anderson in Houston TX  and Jay Kadane at Carnegie
>> Mellon have been working on more ethical designs within the Bayesian
>> framework.  In particular, response adaptive designs reduce
>> the assignment
>> to and continuation of patients on inferior treatments.
>
> I've heard LJ Wei talked about this kinds of designs (don't remember if
> they are Bayesian) more than a dozen years ago.  Don't know how common
> they are in use.
>
> Andy
>

They are becoming more popular as the FDA itself becomes more comfortable with both Bayesian approaches and adaptive designs.

The FDA released draft guidance earlier this year on adaptive designs for drugs and biologics:

  http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM201790.pdf

and final guidance for Bayesian approaches in device trials:

  http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm071121.pdf


There are also several presentations at this week's 2010 FDA Workshop in DC:

  http://www.amstat.org/meetings/fdaworkshop/index.cfm?fuseaction=onlineprogram


Beyond the ethical issues raised, I believe that industry sees these techniques as opportunities to streamline phased trial design, resulting in a compression of the overall timeline to make decisions on the viability of products in the development pipeline.

HTH,

Marc Schwartz

>>
>>
>> Bert Gunter <[hidden email]>
>> Sent by: [hidden email]
>> 09/20/2010 01:31 PM
>>
>> To
>> [hidden email]
>> cc
>>
>> Subject
>> [R] OT: Is randomization for targeted cancer therapies ethical?
>>
>>
>>
>>
>>
>>
>> Hi Folks:
>>
>> **Off Topic**
>>
>> Those interested in clinical trials may find the following of
>> interest:
>>
>> http://www.nytimes.com/2010/09/19/health/research/19trial.html
>>
>> It concerns the ethicality of randomizing those with life-threatening
>> disease to relatively ineffective SOC when new "biologically targeted"
>> therapies "appear" to be more effective. While the context may be new,
>> the debate, itself, is not: Tukey wrote (or maybe it was talked -- I
>> can't remember for sure) about this about 30 years ago. I'm sure many
>> other also have done so.
>>
>> Cheers,
>>
>> Bert

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.