913P - Computed tomography (CT) changes precede malignant spinal cord compression (SCC) in patients with advanced prostate cancer (APC)

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Diagnostics
Prostate Cancer
Imaging, Diagnosis and Staging
Presenter Aurelius Omlin
Authors A.G. Omlin1, C. Pezaro2, D. Mukherji3, D. Bianchini1, G. Attard2, A. Mulick Cassidy1, C. Parker4, D. Dearnaley5, J.S. De Bono6, N. Tunariu7
  • 1Prostate Targeted Therapy Group, The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research ICR, SM25PT - Sutton/UK
  • 2Prostate Targeted Therapy Group, The Royal Marsden NHS Foundation Trust and Institute of Cancer Research ICR, SM25PT - Sutton/UK
  • 3Department Of Hematology Oncology Po Box: 11-0236, The Royal Marsden NHS Foundation Trust and Institute of Cancer Research ICR, SM25PT - Sutton/UK
  • 4Academic Urology, Royal Marsden Hospital NHS Foundation Trust, SM2 5PT - Sutton/UK
  • 5Academic Radiotherapy, The Royal Marsden NHS Foundation Trust and Institute of Cancer Research ICR, SM25PT - Sutton/UK
  • 6The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research ICR, SM25PT - Sutton/UK
  • 7Prostate Targeted Treatment Group, The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research ICR, SM25PT - Sutton/UK

Abstract

Background

SCC occurs in 5-7% of APC patients and is mostly due to epidural soft tissue disease associated with bony metastases. Early detection and treatment is critical to avoid permanent disability. Magnetic resonance imaging (MRI) can detect SCC in asymptomatic patients, but current guidelines do not recommend screening MRI outside of clinical trials. CT scans are performed regularly for restaging. We hypothesise that CT changes precede symptomatic SCC.

Methods

We identified patients with APC treated on clinical trials who developed MRI proven SCC. Evaluable patients had at least four CT scans prior to SCC. Each CT was jointly reviewed by a radiologist and an oncologist for three CT features: paravertebral fat stranding/infiltration (0: none; 1: small volume, suspicious of infiltration; 2: definite, large volume); presence of soft tissue epidural disease in spinal canal (0: none; 1: small volume; 2: definite >2mm antero-posterior diameter); periosteal reaction (0: none; 1: spiculated periosteal reaction in the absence of osteophytes from degenerative disease).

Results

A total of 43 patients were evaluable. At first CT 35 patients (81%) were asymptomatic and 8 patients (19%) had mild to moderate back pain without neurological symptoms. Prior to SCC, 84% of patients had all three CT signs at the corresponding level. Paravertebral fat infiltration was present in all patients but four had only small volume infiltration. Epidural disease was present in 42 patients (98%), with small volume disease in three. SCC was diagnosed on MRI in 81% of patients due to symptoms and in 19% due to suspicion raised on CT. Of 42 patients with definite epidural/paravertebral disease, 22 (52%) developed SCC within 3 months (m). In 31 patients with normal baseline CT, interval from development of CT sign to SCC was (mean, 95% CI): paravertebral fat infiltration 7.9 ± 1.9m; epidural soft tissue disease 7.1 ± 1.8m; periosteal reaction 5.5 ± 1.9m.

Conclusions

CT features of paravertebral fat infiltration, epidural soft tissue disease and periosteal reaction precede SCC in most patients. Presence of significant paravertebral fat infiltration or epidural disease on CT should prompt MRI evaluation of the spine.

Disclosure

All authors have declared no conflicts of interest.