Prognostic Signatures in Breast Cancer


Prognostic signatures are also known as Gene Expression Profiles (GEP) are now commonly used in the clinic to stratify hormone positive breast cancer patients risk of distant recurrence at 10 years and aid treatment strategy decision making. In this blog post, I will review the commonly used GEP and hopefully illustrate their proper clinical utilization as well as limitations. In a previous post, I discussed the adjuvant endocrine therapy for hormone positive breast cancer, for details please click here.

The use of GEP in Early stage breast cancer signficantly reduced the use of Chemotherapy in this patient population.

In general, Gene Expression Profile tests are used to aid clinical decision making on hormone positive, node negative (there are some exceptions), Her2 negative breast cancer, GEP should not be used in Her2 positive patients or triple negative patients. The genetic profiling of tumors is an emerging field with intense research and  ongoing refinement, our understanding of such tests can be summarized as follow,

A. The commonly used genetic profile tests are

  • Oncotype Dx 21-gene recurrence score (RS).
  • EndoPredict (EP).
  • Breast Cancer Index (BCI).
  • PAM50 risk of recurrence score(ROR).
  • Amsterdam 70-gene profile(Mammaprint).

B. All above tests were validated in predicting prognosis and distant recurrences (Prognostic value). There is some performance variation between the above tests, as well as variation in the specific patient clinical presentation, where a given test is particularly useful.

C. Not all the tests were actually clinically validated in showing the benefit of chemotherapy(Predictive value) i.e. while all the tests can predict poor prognosis some of the tests were clinically validated to demonstrate the use of chemotherapy actually improved outcome.

D. At the time of this blog post, the NCCN guidelines recommend “Consideration of Recurrence Score (RS)” in patients who are hormone positive, node negative.

E. ASCO supports the use of all of them, however, not more than one test to be used in a given patient presentation. I will try to demonstrate below that some of the tests are particularly useful in a specific clinical presentation.

F. in the following presentation of different GEP available, I will attempt to illustrate,

  • Prognostic value of GEP.
  • The predictive value in determining the benefit of chemotherapy.
  • The predictive value in determining the benefit of extended hormonal therapy.
  • Comparison of different GEP tests.

The Oncotype Dx 21-gene recurrence score (RS):

  • The RS is a molecular only test, the RS is calculated based on RT-PCR expression of 21 genes. The test is performed on tumor paraffin blocks.
  • The RS is commonly used on Estrogen receptor positive, Her2 negative, lymph node-negative disease (see below for further discussion)
  • The current RS is reported in the following ranges, a. Low-risk distant recurrence  RS <18 b. intermediate risk of distant recurrence RS = 18-30 c. high risk of distant recurrence RS>31. The RS is then plotted on a curve with a corresponding 10-year risk% of distant recurrence for each RS. It is generally accepted to pursue endocrine therapy only and without chemotherapy in patients with low RS (<18), however, in a patient with high RS (>30) it is also acceptable to pursue chemotherapy. Patients with the intermediate score there is no current validated recommendation with regard to adding or withholding chemotherapy, and it is entirely clinical judgment decision. The result of the completed TAILORx hopefully will answer the question pertaining to some intermediate RS.
  • Retrospective validation, RS values as it stands today were retrospectively validated as a prognostic tool through the NSABP-14 patient population specimens, and as a predictive tool through the NSABP-20 patient population specimens (CMF chemotherapy).
  • Prospective validation of RS as also prospectively validated as a prognostic tool in TAILORx for RS of ≤10 and published here, also in PlanB trial for RS ≤11 and published here. Please note that such trial used different ranges of RS, and the prospective validation of RS as a predictive tool is yet to be published.
  • Lymph node-positive disease, the data to support the use of RS in women with lymph node-positive disease is limited, and can be summarized as follow,
  1. Prognostic value of RS in lymph node-positive women was shown in a retrospective way in NSBP-B28 can be found here
  2. Predictive value of RS in lymph node-positive women was shown in postmenopausal women with high RS in a retrospective way in the SWOG trial which can be found here.

Endopredict (EP):

  • Endopredict test is a prognostic score that combined molecular component (11 genes) and clinicopathologic component.
  • The EP test can be used in hormone positive, lymph node-negative or lymph node-positive, Her2 negative women.
  • The EP score is reported as either Low Risk = <3.5 or High Risk = ≥3.5.
  • The test was validated as a prognostic tool by using data from the ABCSG-6 trial and the ABCSG-8 trial.
  • Predictive value is not yet available.
  • EP score was shown to predict late distant recurrence, this may be of value in using extended endocrine therapy details here.
  • EP score was also shown to identify a subgroup of postmenopausal lymph node-positive patients who can safely avoid chemotherapy.

Breast Cancer Index (BCI):

  • Like the RS the BCI is a molecular only test.
  • It showed a particular advantage over other GEP tests in identifying late recurrences and can be of great value in determining extended endocrine therapy, for details click here.

PAM50 risk of recurrence score (ROR):

  • Like the EP score, this is a combined molecular and clinicopathologic score. PAM50 molecular component test for expression of 50-gene, and was originally developed to test for breast cancer intrinsic subtypes. The result of the test is often referred to as risk of recurrence score ROR score.
  • Like all the other tests discussed here the ROR prognostic value was validated and actually compared to RS, where it was shown to be superior to RS by having fewer intermediate scores for details click here.

Amsterdam 70-gene profile (Mammaprint):

  • This is another GEP test that also combines molecular and clinicopathologic finding resulting on a score that can aid therapeutic decision making, the test utilizes 70-gene.
  • The prognostic value of the test was validated in multiple studies. The predictive value of the test was also validated and published in the MINDACT trial.
  • ASCO published clinical practice guideline focused update, for details click here. My personal interpretation of the recommendation can be summarized as follow,
  1. Mammaprint assay may be used in high risk, hormone positive, node-negative, Her2 negative patients, to identify patients who may not need chemotherapy.
  2. Mammaprint assay may also be used in high hormone positive, node-positive, her2 negative with 1-3 lymph node positive, to identify the need for chemotherapy.
  3. of note, the Mammprint assay is not yet recommended by NCCN guidelines.

Comparison between all above GEP tests was published here, and it concluded that for women with the node-negative disease the ROR, BCI, and EP were significantly more prognostic for overall and late distant recurrence, for women with 1-3 node-positive disease data was unlimited to draw a conclusion.

As always, please refer to terms of use here