research reports

University of Hull

2001/2

Centre for Magnetic Resonance Investigations
Scientific Director: Professor L W Turnbull

Breast cancer

Primary and translational research co-exist in the breast cancer research portfolio. Both rely on fast, user friendly, automated analysis of the signal intensity time changes generated during dynamic contrast-enhanced imaging. Work continues in the development of models which accurately reflect patho-physiological processes as determined by immuno-histochemistry based measurement of neovascularisation, measurement of apoptosis and conventional histopathology. At present this work is ongoing in both DCIS, early and in advanced primary breast malignancy. Determination of tumour presence, extent and response to treatment for appropriate and cost effective patient management is in progress. Classification of breast masses and determination of tumour extent are important for cost effective and improved quality of life patient management. These issues form the basis of a Hull based multi-centre study examining use of MR imaging in patients scheduled for wide local excision. In patients with indeterminate microcalcification secondary to a spectrum of pathologies ranging from benign fibrocystic disease to DCIS, comparison is made between MR findings, electrical impedance mammography, measurement of MVD, Tie-1 and Tie-2 and subsequent histopathological grading. Patients with advanced pathologically proven primary breast cancer on standard treatment with FEC, undergo early assessment by DCE-MRI, diffusion and proton spectroscopic imaging, serum tumour markers and repeat core biopsy. Analysis of MR-derived parameters of neoangiogenesis, serum markers and apoptosisis is carried out prior to mastectomy and after tumour bed resection.

Prostate cancer

The majority of newly diagnosed patients with adenocarcinoma of prostate, currently undergo radical radiotherapy with the aim of curative treatment. Prostate-related research at the CMRI reflects current patient management. Studies continued on the differentiation of benign and malignant disease, employing dynamic contrast-enhanced MRI and the use of tetrahedral, trace and anisotropic diffusion imaging examined the apparent diffusion coefficient of contained water. Factors which influence radical radiotherapy are also under consideration namely gross prostate movement due to bladder/rectal filling; and accurate determination of prostate contour with specific delineation of the location and extent of tumour to allow target-within-target boost dose escalation, by employing inversely planned intensity modulated radiotherapy planning. These are informed by fused CT/MR images which incorporate electron density, detailed anatomical and tumour vascularity information. Precise, reproducible, 3D quality assurance of radiotherapy planning prescriptions is necessary. As conventional information obtained by plain films, ionization chambers and portal imaging is insufficiently detailed for IMRT plans, novel gel-dose dosimetry using poly-acrylamide gels is being developed to aid 3D assessment. A stable gel whose relaxation time characteristics demonstrate a linear response to radiotherapy dose has been developed, and preliminary comparison with treatment plans is underway. Retrospective analysis of DCE-MRI has failed to detect correlations with currently recognized American College of Pathology category I predictive factors, although pharmacokinetic parameters correlate with tumour volume, and stage. However good correlations were obtained between MR parameters reflecting contrast leakage space and tumour size, and strong intra-subject correlations noted for the same parameter between tumour and contra lateral normal peripheral zone, suggesting a 'field of change' throughout the prostate. Work is ongoing to determine the significance of these results. Further perfusion studies await the installation of additional fast receivers.

Ovarian cancer

Clinical presentation of ovarian adenocarcinoma is traditionally late due to the insidious nature of this disease and consequently research has concentrated on the diagnostic accuracy of conventional MR imaging, combined with DCE-MRI in comparison with CT and ultrasound. An overall diagnostic accuracy rate of 86.5, 67.2 and 47.2% response has been recorded for MRI, US and CT respectively. Evaluation of uptake characteristics lead to highly significance differences between benign and malignant lesions, which could potentially reduce the number of falsely positive conventional MR scans by 5-10%. Work is currently ongoing to compare MR staging accuracy, with respect to surgical findings and potential alteration in patient management, as well as correlation of pharmacokinetic parameters with response to conventional treatment with first line cytotoxic chemotherapy. MR derived-pharmacokinetic parameters have also been compared with expression of native somatostatin, somatostatin receptors, VEGF and the receptors Flt, Flk and neuropilin-1. Native somatostatin and its receptors were strongly correlated with VEGF and its receptors suggesting a strong positive feedback loop. MR derived parameters were also strongly correlated with MVD, somatostatin and SSTR2, while vascular SSTR1 and 5, Flt, and Flk were negatively correlated suggesting inhibition of neoangiogenesis. MR derived parameters also correlated with the tumour marker CA125, and with FIGO stage. As somatostatin analogues are widely available and have few serious associated side effects, treating with these agents is therapeutically possible.