Imaging dopamine neurotransmission in Parkinson's disease: biomarker versus surrogate end point

Imaging of dopamine (DA) neurotransmission in Parkinson's disease (PD) began with positron emission tomography (PET) measurements of dopamine synthesis using [18F]fluorodopa (FDOPA). This precursor is converted within DA neurons to the ionically charged [18F]fluorodopamine, and this radioactive metabolite is trapped within the cell. The rate of trapping is proportional to the amount of converting enzyme (DOPA decarboxylase), which itself is correlated with the number of DA terminals in the striatum.

Two other targets were subsequently imaged as biomarkers for DA neurotransmission: dopamine transmitter (DAT) and vesicular monoamine transporter, type 2 (VMAT2). DAT is located on the terminals of DA neurons in the striatum and functions to remove DA from the synapse to the intracellular space for recycling or metabolism. VMAT2 is located on the vesicle membranes of DA and noradrenergic neurons, and transports intracellular DA (or norepinephrine) into the vesicle, which is subsequently released by exocytosis on electrical stimulation. DA synthesis and VMAT2 are measured with PET, whereas DAT levels have been measured with both PET and single photon emission computed tomography (SPECT). All three targets (DOPA decarboxylase, DAT, and VMAT2) are clearly biomarkers for DA neurotransmission (Table I). Representative images in PD patients and healthy subjects are shown in Figure 1. Because they are biomarkers of DA neurotransmission, the imaging of these targets has clear utility in the study of the pathophysiology of PD. For example, imaging has demonstrated the following:

• The known loss of DA innervation in PD.

• A negative correlation between the brain imaging measurement, and symptom severity in groups of patients.

• The increasing progression of symptoms over time within individual subjects.

Figure 1. Representative radiotracer images of Parkinson's disease (PD) patients and healthy subjects. A. Single photon emission computed tomography (SPECT) images of dopamine transmitter (DAT) using [¹²³I]β-CIT(2β-carbomethoxy-3β-(4-[¹²³l]iodophenyl)tropane). B. Positron emission tomography (PET) images of DA synthesis with ¹⁸F]fluorodopa (FDOPA). C. PET images of vesicular monoamine transporter, type 2 (VMAT2) with [¹¹C]dihydrotetrabenazine. These images were acquired in PD patients with relatively early onset of disease and compared to age-matched healthy subjects. The color scale is slightly different for the three radioligands. Nevertheless, all three imaging agents show a dramatic and similar loss (~50%) of striatal uptake in PD patients compared to the healthy subjects.

A is from John Seibyl, MD (Institute for Neurodegenerative Disorders, New Haven, CT).

B is from David Eidelberg, MD (North Shore-Long Island Jewish Research Institute, Manhasset, NY).

C is from Kirk Frey, MD, PhD (University of Michigan at Ann Arbor, Ann Arbor, Mich).

Table I Three targets for imaging dopamine (DA) neurotransmission in Parkinson's disease. DOPA, dihydroxyphenylalanine.

At. least two of these targets (DA synthesis and DAT) have been shown to have modest diagnostic specificity. That, is, imaging of these two targets can clearly distinguish PD from benign senile tremor, but has marginal, if any, utility to distinguish idiopathic PD from other “parkinsonisms,” such as multisystem atrophy and striatonigral degeneration. All three targets have demonstrated significant “reserve function” in brain, such that >50% loss of the target is required for the onset of clinical symptoms. Serial studies of DA synthesis and DAT levels in individual patients have shown about 10% loss per annum in the early stages of the disease. As a rough back-extrapolation, these results suggest that the preclinical phase of the disease is ~5 years before the presence of symptoms adequate to make the diagnosis of PD (ie, 50%/10% per annum = 5 years). Taken as a whole, these results suggest imaging of DA neurotransmission in PD may have at least, two important clinical applications:

• Diagnosis in both clinical and preclinical phases.

• A biomarker for the efficacy of agents designed to slow progression of the disease, ie, neuroprotective agents.

Significant controversy surrounds the utility of imaging to provide a “biomarker for efficacy” (ie, a surrogate end point.) for a potential neuroprotective agent. On first consideration, it. seems obvious that DA imaging is a useful end point in a clinical study and provides a useful surrogate for clinical efficacy. However, this may not be the case. For example, a false-positive imaging end point could result, from an agent that slows the loss of the target, but has no clinically significant effect for the patient's symptomatology. Such a result may be the case for a study of SPECT DAT imaging in patients with pramipexole, where DAT loss was slower in the experimental group, but. with no clinically significant differences between groups.¹ One possible interpretation of these results is that pramipexole has odd effects on DAT levels, but. is not a good measure of overall DA neurotransmission. Even more dramatic examples of a falsepositive surrogate end point, exist. For example, the antiarrhythmic agents flecainide and encainide were evaluated with electrocardiogram (F.CG) as their “surrogate end point,” since agents that decrease ECG arrhythmias must certainly help the patient. In fact, these medications decreased arrhythmias, but, after approval by the Federal Drug Administration (FDA), were found to be associated with elevated cardiac mortality.² This example surely shows the need to validate surrogate end points, lest the treatment is found to cure the disease and kill the patient!

In the summer of 2003, the National Institutes of Neurological Disorders and Stroke organized a panel of experts with various backgrounds (including PD, imaging, and regulatory affairs) to assess the utility of DA imaging in PD.The consensus of this group was that DA imaging of these three targets provide useful biomarkers to study pathophysiology, but that additional studies are needed for them to be accepted by the field and the FDA as validated surrogate end points. Enthusiasm was strong for additional prospective studies to be performed by academic researchers, often in collaboration with industry and with input from regulatory authorities such as the FDA.