EndoPredict breast cancer testing

 
 

With endoPredict, more patients can safely avoid chemotherapy

Direct Comparison in Node Negative Patients

In 591 node negative patients, all signatures identified patients with low risk of distant recurrence (<10%) in years 0-10.

Compared to all other signatures, EndoPredict identified most patients with node negative breast cancer as low risk (72.6%) with a 6.6% rate of recurrence. With EndoPredict, more patients might safely avoid chemotherapy.

Accuracy to identify node negative low risk patients in years 0-10 in TransATAC

§ Prosigna cut-off values were trained in TransATAC and therefore were optimized for these patients' cohort.

Direct Comparison in Node Positive Patients

In node positive patients, EndoPredict was one of only two signatures that identified low risk patients with a less than 10% risk of distant recurrence. The 10 year recurrence rate for patients identified as low risk by Oncotype DX was 19.4% and the recurrence rate for patients identified as low risk by BCI was 15.5%. In comparison, patients identified as low risk by EndoPredict had a true low rate of recurrence of 5.6%.

Among the other two signatures, EndoPredict identified almost three times more patients as low risk (23.5% of patients compared to 8.2% for Prosigna) who might safely avoid chemotherapy.

Accuracy to identify node positive low risk patients in years 0-10 in TransATAC

§ Prosigna cut-off values were trained in TransATAC and therefore were optimized for these patients' cohort.

Direct Comparison Regarding Late Distant Recurrence

Compared to all other signatures, EndoPredict identified more node negative patients as low risk for late recurrence in years 5-10 (73.5%). In node positive patients, EndoPredict was one of only three signatures that identified patients with a less than 10% risk of distant recurrence. The 10 year recurrence rate for patients identified as low risk by Oncotype DX was 17.9%.

Compared to Prosigna (ROR), EndoPredict identified more node positive patients as true low risk (26% vs. 9.7%).

EndoPredict identifies more patients for whom an extended endocrine therapy might not be justified.

Risk of distant recurrence in years 5-10 in TransATAC and size of low risk group

§ Prosigna cut-off values were trained in TransATAC and therefore were optimized for these patients' cohort.

EndoPredict has the Strongest Prognostic Accuracy

The prognostic accuracy of all tests was compared with the standard statistic C-index.
The EPclin Risk Score shows stronger predictive power than all other tests in years 0-10 and years 5-10.

Prognostic ability to detect distant recurrence in years 0-10 and 5-10 in TransATAC

The C-index is a standard statistic for prognostic power that is used to compare prognostic accuracy of different tests. The greater the C-Index, the better is the prognostic power of a test. The C-Index reflects the prognostic power of the continuous score independent from cutoff values.

More Patients Might Safely Forgo Aggressive Treatment

Compared to other prognostic signatures:

  • EndoPredict was the best signature in predicting distant recurrence in years 0-10 and in years 5-10 in all patients, including node negative and node positive disease.
  • EndoPredict identified the largest group of women with breast cancer, both in node negative and node positive disease, at low risk of distant recurrence who might safely avoid chemotherapy.
  • EndoPredict identified the largest group of women with breast cancer, both in node negative and node positive disease, at low risk of late distant recurrence for whom an extended endocrine therapy might not be justified.

Sestak I. et al.: Comparison of the Performance of 6 Prognostic Signatures for Estrogen Receptor-Positive Breast Cancer. A Secondary Analysis of a Randomized Clinical Trial. JAMA Oncol. 2018;4(4):545-553