Prediction of Prostate Cancer Gleason Score Upgrading from Biopsy to Radical Prostatectomy (RP) using a Validated 17-Gene Panel Assay

Date 22 October 2018
Event ESMO 2018 Congress
Session Poster display session: Breast cancer - early stage, locally advanced & metastatic, CNS tumours, Developmental therapeutics, Genitourinary tumours - prostate & non-prostate, Palliative care, Psycho-oncology, Public health policy, Sarcoma, Supportive care
Topics Prostate Cancer
Translational Research
Pathology/Molecular Biology
Presenter Jennifer Cullen
Citation Annals of Oncology (2018) 29 (suppl_8): viii271-viii302. 10.1093/annonc/mdy284
Authors J. Cullen1, H.J. Lawrence2, Y. Chen1, R. Lu2, S. Srivastava1, I. Rosner1, T. Brand1, I. Sesterhenn1
  • 1Surgery, Center for Prostate Disease Research, USUHS, 20852 - Rockville/US
  • 2Urology, Genomic Health, Redwood City/US



We have previously shown that a 17-gene, biopsy-based RT-PCR assay (Oncotype Dx® Genomic Prostate Score™, GPSTM) is a strong independent predictor of 1) adverse pathology (defined as surgical GS > 4 + 3 and/or non-organ confined disease) at radical prostatectomy (RP), and 2) biochemical recurrence after RP in men with biopsy GS 6 and 7 prostate cancer. We performed a more refined analysis of the association between GPS and upgrading from biopsy to surgery.


Patients in a prior clinical validation study of the GPS assay were assigned a biopsy Gleason sum (BxG) and pathologic ISUP 2014 surgical Gleason grade (pGS) by a single uropathologist (IS) in a blinded fashion. We performed logistic regression analyses to model the association of GPS with Gleason upgrading (GU) defined as (i) Any downgrade, (ii) Any increase in GS, (iii) BxG 6 to pGS 3 + 4, (iv) BxG 6 to pGS 4 + 3, and (v) BxG 7 to pGS 8-10, all versus no change in GS. GPS values, as determined from diagnostic biopsy tissue, are reported on a scale of 0-100, and predictions of GU were modeled in 20-unit GPS increments.


Among 395 patients, median patient age at RP and GPS score was 62 years and 30.3, respectively. There were 249 patients (63%) who had no change in GS from biopsy to surgery, 45 patients (11%) were down-graded (BxG 7 to pGS 6) and 101 patients (31%) were upgraded (85 BxG 6 to pGS3 + 4, 10 BxG 6 to pGS 4 + 3, and 6 BxG 7 to pGS 8-9). GPS was a significant predictor of any GS upgrade, BxG 6 to pGS 3 + 4, and BxG 7 to pGS 8-9 (see Table).Table: 837P

BxG vs pGSOR/ per 20 units in GPS95% CI of ORP value
Any GS Downgrade1.000.581.740.978
Any GS upgrade1.771.202.600.004
bGS6 to pGS3 + 41.611.062.410.027
bGS6 to pGS 4 + 32.460.906.730.079
bGS 7 to pGS 8-94.491.3215.5330.017

Referent in all models is no change in GS.


GPS is a significant predictor of the likelihood of Gleason score upgrading from biopsy to RP. Incorporation of GPS results into treatment decision making can improve risk assessment for newly diagnosed patients with clinically low-risk prostate cancer.

Clinical trial identification

Legal entity responsible for the study



Genomic Health, Inc.

Editorial Acknowledgement


H.J. Lawrence: Consultant, stock owner: GHI. All other authors have declared no conflicts of interest.