Oncotype DX: Algorithm behind the recurrence score

Options
jenrio
jenrio Member Posts: 558
edited June 2014 in Clinical Trials, Research News, Podcasts, and Study Results

http://www.cap.org/apps/docs/membership/molecular_markers_in_breast_cancer.pdf 

RS= + 0.47 x HER2 Group Score
- 0.34 x ER Group Score
+ 1.04 x Proliferation Group Score
+ 0.10 x Invasion Group Score
+ 0.05 x CD68
- 0.08 x GSTM1
- 0.07 x BAG1

Their ER+ score ranges from 7 up to the teens.  the the maximum contribution to RS from ER is approximate 4.

Their Her2+ score also has a range, so maximum contribution from HER2 score is also around 6.

The main contributor as you could tell from the formula would be Proliferation.   I plugged in my independent Ki-67 score and ER score (Her2-), voila, my RS score. 

So the main contributor in this algorithm is Ki-67 and Her2 and ER/PR (costs <1000  for independent pathologist).   Oncotype DX charges $3000, is it covered by medicare?   

· Share on FacebookShare on Twitter

Comments

  • doxie
    doxie Member Posts: 1,455
    edited June 2012

    Using my pathology report and after stumbling onto this algorithm, I pretty much figured out I would need chemo.  Went ahead and requested the oncotype because I didn't want to do chemo and was hoping I was wrong.  It would always have nagged me not to know.  Also is a motivator to stay the course with AIs and other treatments. 

    I am curious if the staining of the ER and PR is considered also.  Weakly vs highly stained.  I suspect so from anaylizing my results, but have never come across this information.   

    · Share on FacebookShare on Twitter
  • jenrio
    jenrio Member Posts: 558
    edited June 2012

    Oncotype uses a different technique to find ER/PR than normal pathologist.   There are several ways they could come to different conclusions.   The biggest issue is that tumor is heterogeneous.    So maybe one part of it has lots of ER, no HER2, another part has little ER and lots of HER2.   Add to it Ki-67 differences in different part of tumor.

    If you have a few thousand of your own money, you may want to spend it

    a.   give multiple part of tumor to different pathologists. and collect all the information together.   A lot of trouble.

    b.   ask your pathologist to send the most aggressive part of tumor to "theraprint" or foundationmedicine, which gave you more detailed information on the tumor gene (than oncotype DX).   Both are covered by insurance.

    c.   send it to an university genomics lab:  ask for deep sequencing or next generation sequencing.    May still be covered by insurance.   If not, well you are contributing to a research facility. 

    · Share on FacebookShare on Twitter
  • bevin
    bevin Member Posts: 1,902
    edited July 2012

    ok, so I'm not a math head, and did have the Onco score done, but I dont understand this alogorythm.

    · Share on FacebookShare on Twitter
  • jenrio
    jenrio Member Posts: 558
    edited July 2012

    Try putting the equation into a spreadsheet, and plug in a few numbers.   It might help you understand a little bit more about your risks. 

    · Share on FacebookShare on Twitter

Categories