Automated Bone Scan Index as an Imaging Biomarker to Predict Overall Survival in the Zometa European Study/SPCG11

Mariana Reza, Manfred Wirth, Teuvo Tammela, Virgilio Cicalese, Francisco Gomez Veiga, Peter Mulders, Kurt Miller, Andrea Tubaro, Frans Debruyne, Anup Patel, Christien Caris, Wim Witjes, Ola Thorsson, Per Wollmer, Lars Edenbrandt, Mattias Ohlsson, Elin Trägårdh, Anders Bjartell

Research output: Contribution to journalArticlepeer-review

4 Citations (SciVal)


BACKGROUND: Owing to the large variation in treatment response among patients with high-risk prostate cancer, it would be of value to use objective tools to monitor the status of bone metastases during clinical trials. Automated Bone Scan Index (aBSI) based on artificial intelligence has been proposed as an imaging biomarker for the quantification of skeletal metastases from bone scintigraphy.

OBJECTIVE: To investigate how an increase in aBSI during treatment may predict clinical outcome in a randomised controlled clinical trial including patients with high-risk prostate cancer.

DESIGN, SETTING, AND PARTICIPANTS: We retrospectively selected all patients from the Zometa European Study (ZEUS)/SPCG11 study with image data of sufficient quality to allow for aBSI assessment at baseline and at 48-mo follow-up. Data on aBSI were obtained using EXINIboneBSI software, blinded for clinical data and randomisation of zoledronic acid treatment. Data on age, overall survival (OS), and prostate-specific antigen (PSA) at baseline and upon follow-up were available from the study database.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Association between clinical parameters and aBSI increase during treatment was evaluated using Cox proportional-hazards regression models, Kaplan-Meier estimates, and log-rank test. Discrimination between prognostic variables was assessed using the concordance index (C-index).

RESULTS AND LIMITATIONS: In this cohort, 176 patients with bone metastases and a change in aBSI from baseline to follow-up of ≤0.3 had a significantly longer median survival time than patients with an aBSI change of >0.3 (p<0.0001). The increase in aBSI was significantly associated with OS (p<0.01 and C-index=0.65), while age and PSA change were not.

CONCLUSIONS: The aBSI used as an objective imaging biomarker predicted outcome in prostate cancer patients in the ZEUS/SPCG11 study. An analysis of the change in aBSI from baseline to 48-mo follow-up represents a valuable tool for prognostication and monitoring of prostate cancer patients with bone metastases.

PATIENT SUMMARY: The increase in the burden of skeletal metastases, as measured by the automated Bone Scan Index (aBSI), during treatment was associated with overall survival in patients from the Zometa European Study/SPCG11 study. The aBSI may be a useful tool also in monitoring prostate cancer patients with newly developed bone metastases.

Original languageEnglish
Pages (from-to)49-55
JournalEuropean Urology Oncology
Issue number1
Early online date2019 Jun 8
Publication statusPublished - 2021

Bibliographical note

Copyright © 2019. Published by Elsevier B.V.

Subject classification (UKÄ)

  • Cancer and Oncology


Dive into the research topics of 'Automated Bone Scan Index as an Imaging Biomarker to Predict Overall Survival in the Zometa European Study/SPCG11'. Together they form a unique fingerprint.

Cite this