Improving on the Kattan nomogram through use of genetic data

There is an interesting new paper forthcoming in the Journal of Urology in which the research team describes how the addition of data from genetic profiling is able to improve the accuracy of the Kattan pre-treatment nomogram specifically in men believed to have localized prostate cancer at the time of diagnosis.

What Talantov et al. have been able to do is to modify the Kattan pre-treatment nomogram (used to project outcomes to various types of treatment based on things like age, PSA level, clinical stage, Gleason score, etc.) by the inclusion of data based on a three-gene signature (DPT, MYH11, and SSBP1).

Technically, this is an extremely complex paper, and we aren’t going to try to explain all of the details. But here are the basics:

  • The research team built the model using data from 316 consecutive patients treated at St. Vincent’s Hospital in Sydney, Australia, all of whom had pathologically localized prostate cancer (pT2a through pT3a) after radical prostatectomy  and were followed for a minimum of 5 years.
  • The 316 patients were randomly assigned to a “training” set (for construction of the model) and a “validation” set for confirmation of the accuracy of the model.
  • The adjusted model was able to predict biochemical progression-free survival at 5 years with significantly greater accuracy than the current Kattan nomogram available on the Memorial Sloan-Kettering web site.
    • Specifically, the “concordance index” for the current Kattan nomogram is 0.67, and the concordance index for the revised model is 0.77 (a 15 percent improvement in predictive accuracy).

The authors draw the very reasonable conclusion that “the development of an improved prognostic model for localized prostate cancer has the potential
to facilitate better treatment decisions.”

We should note, however, that what this study really shows is the potential to be able to improve our prognostic tools rather than the development of a practical tool for the immediate future. We don’t yet know if these are the “best” genetic markers to use in such a nomogram. We also don’t have the systems in place to gather the genetic information on every patient and use it in such prognostic algorithms. But we are clearly getting closer, and the practical demonstration that we can develop these more accurate algorithms is clearly an important new step forward. It is also important to note that the data set used to create this specific algorithm was entirely focused on patients who are highly representative of most of the men being diagnosed with prostate cancer today using early detection protocols.

Leave a Reply

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

You are commenting using your 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: