Functional imaging: what evidence is there for its utility in clinical trials of targeted therapies?

Abstract

Key issues in early clinical trials of targeted agents include the determination of target inhibition, rational patient selection based on pre-treatment tumour characteristics, and assessment of tumour response in the absence of actual shrinkage. There is accumulating evidence that functional imaging using advanced techniques such as dynamic contrast enhanced (DCE)-magnetic resonance imaging (MRI), DCE-computerised tomography (CT) and DCE-ultrasound, diffusion weighted-MRI, magnetic resonance spectroscopy and positron emission tomography-CT using various labelled radioactive tracers has the potential to address all three. This article reviews this evidence with examples from trials using targeted agents with established clinical efficacy and summarises the clinical utility of the various techniques. We therefore recommend that input from specialist radiologists is sought at the early stages of trial design, in order to ensure that functional imaging is incorporated appropriately for the agent under study. There is an urgent need to strengthen the evidence base for these techniques as they evolve, and to ensure standardisation of the methodology.

Figure 1

Schematic showing the currently most used functional techniques in the clinic and illustrating their mechanism of action. The output measures from each of these are shown in pink. Dynamic contrast enhanced MRI with output measured as rate constants (K^trans and K_ep) of gadolinium transfer between intravascular and extravascular compartments measured in ml min^–1. Diffusion weighted MRI with measured apparent diffusion coefficient (ADC) measured in mm² s^–1. Intrinsic susceptibility weighted MRI with output the relaxation rate constant R2^* measured in s⁻¹. Dynamic contrast enhanced CT, with output of relative blood flow (rBF) and relative blood volume (rBV) measured in ml min^–1 and ml ,respectively. Parameters in PET are measured as maximum standardised uptake values (SUV_max) of the radioligand. In US, quantified parameters are change in US backscatter before and after injection of microbubbles and represented as integrated area under the curve (IAUC) and rBF.

Figure 2

Dynamic contrast enhanced-MRI parametric maps before (left column) and 28 days after (right column) VEGF inhibitor therapy in a patient with metastatic colorectal carcinoma illustrating a down-stream effect with significant reduction in tumour vascularity (Courtesy to Dr C Messiou & M Orton, Royal Marsden Hospital).

Figure 3

Differential response in ¹⁸F-FDG PET uptake after 1 cycle of targeted therapy with a bRaf and Mek inhibitor in a patient with metastatic melanoma. The mediastinal nodal mass has increased in size and demonstrates a heterogeneous increase in SUV (short arrow) while lung nodules are smaller in size and show significant reduction in SUV (long arrow). The tibial bony lesion seen on the pre-treatment scan is smaller, but a new tibial lesion is visible. The new cervical nodes (arrowhead) were felt to be inflammatory in aetiology.

Figure 4

Multiparametric MRI protocol in the coronal plane in a patient with liver metastases from colon cancer showing a combination of anatomical information (T2-weighted, A) with functional imaging parametric maps, which quantify tumour vascularity (K^trans, B), areas of hypoxia (R2^* maps, C) and tumour cellularity (ADC and DWI, D and E, respectively) together with single voxel ¹H-MRS (F). The metastasis shows a vascular (B), hypercellular (D and E) rim with a hypoxic (C), necrotic (D) centre with increased proliferative activity as evidenced by choline signal in (F).