Evolving Use of Molecular Imaging in Research and in Practice

Combined molecular and structural imaging using positron emission tomography (PET) has long been clinically employed in the evaluation of cardiovascular and oncologic disorders. More recently, PET imaging has been accepted as an assessment tool in rheumatology, where it has proven value in the evaluation of vasculitis. Emerging research continues to expand the uses for molecular imaging in rheumatic diseases. Such imaging holds the potential to improve the identification of individuals at the highest risk for disease complications and those most likely to benefit from more aggressive treatments.

Among imaging approaches that target inflammation in humans, ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET with computed tomography (CT) has been the most extensively studied. ¹⁸F-FDG accumulates within tissues in proportion to its glycolytic rate (1). Inflammatory cells have particularly high rates of glycolysis, especially after proinflammatory activation (1), and thus accumulate relatively larger amounts of ¹⁸F-FDG. The relationship between ¹⁸F-FDG uptake and tissue inflammation has been best evaluated in the context of atherosclerosis, which is itself a chronic inflammatory condition. In human studies, atherosclerotic ¹⁸F-FDG uptake measured using PET-CT imaging has repeatedly been shown to correlate closely with macrophage density on histopathologic evaluation (1). Thus, ¹⁸F-FDG–PET-CT provides a noninvasive index of tissue inflammation.

The ability of ¹⁸F-FDG–PET-CT to characterize tissue inflammation has been leveraged in a wide range of inflammatory cardiovascular conditions (Figure 1). In the evaluation of atherosclerotic disease, arterial ¹⁸F-FDG uptake has been shown to independently predict incident cardiovascular disease (CVD) events beyond traditional risk factors (1). Since therapies that improve clinical CVD outcomes also tend to attenuate the PET imaging signal, ¹⁸ F-FDG–PET-CT has been repeatedly used to test novel therapies that target atherosclerosis (1). In the evaluation of cardiac sarcoidosis (a condition characterized by macrophage infiltration of the myocardium), ¹⁸F-FDG–PET-CT can be used to detect cardiac involvement with high sensitivity (2). Further, heightened myocardial ¹⁸F-FDG uptake has been shown to predict a higher risk for cardiac events in patients with cardiac sarcoidosis and can be used to evaluate the efficacy of therapies targeting this disease process (3). Additionally, ¹⁸F-FDG–PET-CT has been proven to be clinically useful in detecting infections of intracardiac prosthetic materials, such as prosthetic valves, cardiac implantable electronic devices, and left ventricular assist devices (4).

Figure 1.

Clinical uses of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET) with computed tomography (CT) inflammation imaging. In the clinical entities depicted (suspected prosthetic valve endocarditis, known or suspected cardiac sarcoidosis, suspected device infection, and aortitis/arterial inflammation), ¹⁸F-FDG–PET-CT has demonstrated utility and is routinely employed. Arrows indicate increased FDG uptake consistent with higher inflammation. A PET-CT device is also depicted. Reproduced, with permission, from refs. 20–22.

¹⁸F-FDG–PET-CT has also been used with increasing frequency in rheumatic conditions and has been demonstrated to be clinically useful in the evaluation of large vessel vasculitis. ¹⁸F-FDG–PET-CT provides both high sensitivity and high specificity in the detection of arterial wall inflammation and, along with biopsy and other imaging modalities, has become more frequently used in the evaluation of giant cell arteritis (GCA) and Takayasu arteritis (5). Additionally, ¹⁸F-FDG–PET-CT may have a role in the longitudinal assessment of disease activity and may be useful in the prediction of long-term arterial complications (e.g., aortic aneurysm) (6). In recognition of the utility of PET-CT, a European League Against Rheumatism working group has recommended early ¹⁸F-FDG–PET-CT imaging in individuals with suspected GCA (7).

While ¹⁸F-FDG–PET-CT imaging has been less extensively studied in rheumatoid arthritis (RA), emerging research suggests that it may have a potential role in the evaluation of the disease. Inflamed synovium in RA is characterized by considerable immune cell infiltration, proliferation of resident synoviocytes, and neovascularization (8). These synovial immune cells have heightened glycolytic metabolism as a result of their immune activation (9). Perhaps unsurprisingly, relatively high synovial ¹⁸F-FDG uptake has been frequently observed among patients with RA who have undergone ¹⁸F-FDG–PET-CT for oncologic indications. Further, multiple studies have demonstrated that in individuals with RA, synovial ¹⁸F-FDG uptake is associated with severity of joint inflammation and predicts drug therapy response and subsequent chronic joint damage (10,11).

While there has been growing interest in advancing the use of molecular imaging in RA, significant limitations have curtailed its clinical application. Importantly, previous studies have used a broad range of ¹⁸F-FDG end points to characterize joint activity and there are limited data to compare the various ¹⁸F-FDG–PET-CT synovial measures with clinical measures of disease activity, thus limiting standardization of the technique. Accordingly, key questions remain regarding the use of ¹⁸F-FDG–PET-CT synovial imaging, including how best to measure synovial activity with PET-CT, and how much those measures add to currently utilized clinical parameters.

In this issue of Arthritis & Rheumatology, Lee et al (12) provide important insights that help address these questions. Lee and colleagues evaluated the relationship between ¹⁸F-FDG–PET-CT–derived measures of joint activity and clinical assessment of joint disease activity in 91 individuals with RA. The study consisted of 2 groups: a development cohort (n = 69) and a validation cohort (n = 22). The number of PET-positive joints (of 28- or 68-joint counts) significantly correlated with swollen and tender joint counts as well as with the Disease Activity Score in 28 joints using the erythrocyte sedimentation rate (DAS28-ESR) (13). Using multivariable analyses that included the ESR, the patient’s global assessment of disease activity (PtGA), and PET-CT–derived parameters, Lee et al developed a PET-based disease activity score (PET/DAS). When applied to the validation cohort using Pearson’s correlation coefficient (0.843), the PET/DAS highly correlated with DAS28-ESR scores (P < 0.001). Based on these observations, the authors proposed that PET-CT could serve as a sensitive and reliable method to evaluate joint inflammation. This study adds to the growing number of studies that have evaluated ¹⁸F-FDG–PET-CT imaging of joints. Importantly, it provides data that 1) identify the quantitative ¹⁸F-FDG–PET-CT measures that best correlate with concurrent clinically determined disease activity and 2) demonstrate that the PET-CT joint measures are highly reproducible.

The limitations of the standard clinical joint counts used to assess disease activity are well recognized. Joint counts are limited by an inherent lack of objectivity related to both operator factors (e.g., training, experience, perception of the patient’s pain level) and patient factors (e.g., body habitus, subjectivity of pain). Exclusion of the least reliable joints from the 28-joint count (i.e., the metatarsophalangeal [MTP] joints of the feet) has been demonstrated as a simplification that does not sacrifice utility across groups of RA patients (14). Yet, such simplifications may not apply equally across patients. For example, consider the dilemma illustrated when comparing a patient with 10 tender and swollen MTP joints (who receives a score of 0 using the traditional 28 tender and swollen joint count) with a patient with scant nontender synovitis in the small joints of the hands (who receives a score of 20). In this example, the higher joint count does not necessarily translate to greater disease activity. To complicate things further, there remains a concern that current joint counts are not sufficiently sensitive, as subclinical joint destruction continues despite apparent clinical remission (as assessed with typical clinical parameters) (15,16). Hence, more sensitive and reproducible methods are needed to quantify disease activity in RA. Indeed, the ¹⁸F-FDG–PET-CT–derived method described in the report by Lee and colleagues provides such a measure.

In the future, the assessment of synovial inflammation may be further enhanced by combining PET functional measurements with measurements derived from structural imaging modalities (e.g., volumetric indices derived using simultaneously acquired CT or magnetic resonance imaging data). Such multimodality imaging may further enhance measurement of disease activity by providing data on both the burden and the intensity of synovitis. However, additional study, including assessments of cost-efficacy, is needed before these imaging techniques can be used in the routine clinical care of patients with RA.

On the other hand, ¹⁸F-FDG–PET-CT imaging of joint inflammation can currently be leveraged in clinical trials on RA, where surrogate end points of disease activity have the potential to improve the identification of effective therapies. Akin to a model that has been highly successful in oncology, results of phase II PET-CT imaging end point trials can be used to inform the selection of treatments that should advance to phase III clinical end point trials. However, before relying on such a strategy, more data are needed to evaluate how well changes in the surrogate PET-CT imaging measures predict clinical efficacy, as has been done in oncologic and atherosclerotic ¹⁸F-FDG–PET-CT studies (1).

Additionally, ¹⁸F-FDG–PET-CT can be used to gain important pathobiologic insights into RA, including assessment of the biologic cross-talk that exists between organ systems impacted by RA. For example, previous ¹⁸F-FDG–PET-CT imaging studies have shown an association between synovial inflammation and arterial inflammation (17). This may represent an important observation since patients with RA have a substantially increased risk of death attributed to underlying CVD (18,19). A particularly vexing question remains as to whether reducing clinically evident synovial disease activity (by treating to alleviate joint symptoms) is sufficient to attenuate atherosclerotic inflammation and its accompanying risks of CVD in RA. To this end, the Treatments Against RA and Effects on FDG-PET/CT (TARGET) trial (ClinicalTrials.gov identifier ) is an ongoing National Institues of Health–sponsored, randomized, controlled, multicenter trial using ¹⁸F-FDG–PET-CT to compare the effects of 2 treatment regimens on arterial inflammation among RA patients with inadequate methotrexate (MTX) treatment response (the addition of a tumor necrosis factor inhibitor to background MTX versus the addition of sulfasalazine and hydroxychloroquine to background MTX [triple therapy]). Additionally, the TARGET trial provides a unique opportunity to prospectively evaluate several synovial PET-CT parameters (both at baseline as well as change over time) to assess which best predict clinical response; the trial should also provide needed data to further assess the value of ¹⁸F-FDG–PET-CT imaging in RA.

Multimodal molecular imaging of inflammation has become an important tool in the assessment of several inflammatory conditions. In RA, inflammation imaging with ¹⁸F-FDG–PET-CT is being increasingly used to study novel treatment approaches and to develop pathobiologic insights. More studies, such as the one reported by Lee et al, are needed in order to better understand the potential role of this imaging modality in the assessment of individuals with RA.