What IS an “N-of-1” clinical trial?


As we come, slowly, closer and closer to truly individualized forms of medical practice, in which individual patients get treated individually based not just on “what works” for most men with (say) prostate cancer that has just progressed to the metastatic stage, clinical researchers are increasingly interested in so-called “N-of-1″ clinical trials.

N-of-1 clinical trials are basically trials of specific therapies in a single patient with very highly defined characteristics — often including a complete genetic profile as well as a whole bunch of other data. It would be impossible here to get into all the details about how N-of-1 clinical trials should be used and when they provide data that may be helpful in the management of larger number of patients, so here are a series of resources for those who might be interested in learning more — because we are going to be hearing about N-of-1 clinical trials in prostate cancer soon:

  • An article by Lillie et al., published back in 2011, and available as a full text article via PubMed, still provides an excellent overview of what N-of-1 trials are and what they may be used to accomplish. The article is not specific to prostate cancer … but that’s not really important. At the end of the article, there is a very nice “executive summary” for those who don’t want to read all the detail in between.
  • The Agency for Healthcare Research and Quality, in 2014, produced a report entitled Design and Implementation of N-of-1 Trials: A User’s Guide. The report gets into a lot of detail, but it is a valuable update for those who are really interested in understanding the concept of N-of-1 trials and everything from their design to the ethical issues involved in the conduct of such trials.
  • On the Nature web site, there is a nice and relatively brief article by Nicholas Schork entitled “Personalized medicine: time for one-person trials“.

It is important to understand that, in some types of N-of-1 clinical trial, the trial can even be randomized and controlled — because the order in which experimental and control medications are given to an individual patient can be randomly allocated or they can be fixed in advance by the clinical researcher. Thus, just as an example, a series of N-of-1 clinical trials of abiraterone and enzalutamide and docetaxel chemotherapy in a series of similar but very specific patients could specify that:

  • In patient A the docetaxel was always given first but the order of use of abiraterone and enzalutamide was randomized.
  • In patient B the docetaxel was always given last, but the order of use of abiraterone and enzalutamide was randomized.
  • In patient C the order of use of all three drugs could be randomized.

and so on.

When it comes to prostate cancer, the place where N-of-1 clinical trials are most likely to start having an impact is, indeed, in helping us to understand the appropriate sequencing of a series of potential therapies in the management of men with micrometastatic and metastatic prostate cancer who exhibit very specific genetic characteristics of that cancer. We already have multiple forms of treatment available that can be used in such patients but we have very little information yet as to how those treatments work in patients with particular genetic characteristics. Indeed, the only specific characteristic for which we have well-established information as yet seems to be the androgen receptor subtype known as AR-V7. Men who express this androgen receptor subtype do not seem to respond well to either abiraterone or to enzalutamide, but they do seem to still respond to docetaxel-based chemotherapy (see here for more info on this topic).

As we learn more about the classification of subtypes of prostate cancer — and particularly subtypes of advanced prostate cancer — based on the expression of specific genes (see, for example, this recent commentary), we may be able to apply N-of-1 clinical trial processes … and meta-analyses of data from serial N-of-1 trials in men with exactly the same (or very, very closely similar) biological and clinical profiles to tease out the very best ways to treat other men with closely analogous profiles.

2 Responses

  1. I’ve read the material. N + 1 may have excellent benefits to individualizing targeted therapies. But the downside appears obvious ~ reduced peer review, a lot of anecdotal comparing, and lots of individualized bias among the researchers. Not that this doesn’t happen today. :-)

    But the strength in a cohort of a large trial will still be the standard of approving new drugs. And I do not see an N + 1 approach that would be any improvement on that front. But we are strapped today for funding to be able to wait out the process of determining drug sequencing and genomic targeting. We may need N + 1 for that reason.

    It does sound like dinner table stuff to talk about at the next SWOG gathering in Chicago.

  2. Tony:

    I think one of the great potential advantages of N + 1 trials is that they may allow for earlier hypothesis generation that would then lead to trials of new drugs (and older drugs) in better defined groups of patients, which may then be able to be smaller groups of patients in order to observe meaningful endpoints.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: