PET Assessment of Left Main Coronary Arterial Inflammation with Coronary CT Angiography Validation Before and After Statin Therapy: More Promise for FDG Vascular Uptake?

The field of cardiovascular imaging has been searching for the ideal imaging technique that will quantitatively and reliably assess vulnerable coronary plaques and also assess for cardiovascular risk noninvasively. Myocardial infarction (MI), a consequence of coronary atherosclerosis, remains the leading cause of death in the Western world despite advances in pathophysiology¹. The development of novel therapies to treat vascular disease beyond current treatment recommendations remains slow, in part due to the high cost of conducting cardiovascular outcome trials and the challenge of demonstrating that a novel therapy has effects on atherosclerotic plaques which translate into MI reduction if advanced into a large outcomes study².

Recently, many studies have demonstrated that FDG PET/CT accurately identifying regions of arterial inflammation in vivo which can be reliably assessed longitudinally following an intervention. Within this large body of work, FDG PET/CT arterial inflammation estimates current CV risk³, relates to severity of inflammatory diseases⁴ and is prognostic of future CV events^{5, 6}. Specifically, higher uptake in the carotid arteries was related to occurrence of stroke in the future⁶. FDG uptake in the aorta has been associated with increased risk for future myocardial infarction⁵. Higher areas of FDG uptake were associated with experiencing the MI sooner than those with lower uptake⁵. Finally, FDG uptake in the artery modulates with interventions known to be of benefit in cardiovascular outcomes trials. For example, use of statin therapy was associated with a reduction of arterial inflammation at 12 weeks^{7, 8}, and therapeutic lifestyle changes over the course of six months led to a decrease in arterial inflammation⁹. Most of these studies utilized arterial inflammation measured in the aorta or the carotid arteries, and very few studies have attempted to measure arterial inflammation directly in the coronary arteries using FDG PET/CT. Development of this measurement to assess left main coronary artery inflammation has likely been hampered by the inherent spatial resolution limitation of PET (4mm) as well as myocardial uptake of FDG. This latter issue though can be overcome with a diet-based intervention of high-fat low carbohydrate food content prior to the study.

In this issue of Circulation: Cardiovascular Imaging, Singh and colleagues utilize FDG PET CT to assess arterial inflammation in the left main coronary artery in addition to the carotid arteries¹⁰. The study was performed in a subset of patients who were part of a prospective study of statin effects on arterial inflammation. In order to enter the study, participants had to have a TBR >1.6 on FDG PET CT in the carotid arteries. Following a myocardial suppression protocol to reduce FDG uptake in the heart, the authors quantified arterial inflammation in the proximal portion of the left main coronary artery. Specifically, 4 regions of interest were placed: at the LMCA ostium, 5 and 10 mm distally, and that at the LM bifurcation. These four values of maximal FDG uptake were averaged to produce a single value. Then, participants underwent coronary CT angiography to perform luminography and assessment of plaque morphology in the coronary arteries for high risk features. Techniques to identify high risk plaque included presence of non-or partially calcified plaque as well as positive remodeling were used. These approaches are well-accepted in the field and have been shown to be predictive of future CV events¹¹. In an exploratory analysis, the authors also performed inflammatory biomarker phenotyping to probe mechanism associated with potential improvement of arterial inflammation following therapy. The overall study design permitted the authors to assess relationships between left main coronary artery arterial inflammation detected by FDG PET/CT, high-risk plaque assessed by coronary CT angiography and the effect of statin therapy in double-blinded, active comparator trial.

Seventy-one patients, previously on either no- or low-dose statins completed the twelve-week study period following randomization to atorvastatin 10mg vs 80 mg. Of these, sixty-eight had PET/CT images of high quality, and within this group, fifty-five had CTA images of high enough quality for analysis. First, the authors found that LMCA inflammation was approximately 15% higher in those patients who had coronary plaque features with high risk morphology with weak statistical significance (p=0.043). Next, the authors observed that statin therapy at 12 weeks reduced LMCA inflammation in those with high risk plaque morphology, a finding that persisted after adjustment for several important confounders including baseline LDL. In their exploratory analysis with serum biomarkers, the author report that higher hsCRP was associated with high risk plaque morphology, but not with LMCA inflammation. Furthermore, they report a weak correlation between LMCA inflammation and MMP-3 (r=0.42, p=0.04).

Perhaps one of the most interesting findings of the study was that there was a strong correlation between LMCA inflammation and TBR in extra coronary vessels, with the aorta being the strongest correlate (r=0.6, p<0.001) and carotids being the weakest (r=0.32, p=0.04). Strikingly, there was a strong association observed between extra-coronary arterial inflammation and coronary high-risk plaque. This finding particularly suggests the robust quality of risk data assessed by aortic vascular inflammation and support findings that aortic vascular inflammation increases risk for future MI⁵. Furthermore, they found that following statin therapy, the decrease in left main coronary artery arterial inflammation correlated with changes in extra-coronary arterial inflammation in the aorta (r=0.51, p=0.001).

There are several important findings from this study which are novel and of interest to understanding arterial inflammation. First, in patients who had high-risk plaque features in the coronary arteries by CCTA experienced the greatest reduction of arterial inflammation both within the LMCA and in extra-coronary arterial inflammation. Indeed, these findings suggest that in patients with high risk plaque features, there is a potentially greater anti-inflammatory effect of statins. This concept of imaging-guided statin therapy has been demonstrated using CAC scoring to guide statin therapy^{12, 13}, and adds to a body of evidence supporting a “personalized medicine” approach for allocation of statin therapy. Further, the observation of a relationship between high risk plaque morphology and arterial inflammation provides further evidence that arterial inflammation by FDG PET CT can serve as a reliable surrogate marker of coronary atherosclerosis at high risk for future rupture. These findings support the notion that FDG PET CT imaging provides a reliable and accurate surrogate marker for future CV events which may aid in evaluation of emerging interventions in CVD trials. In a recent review¹⁴, the author review five classes of drugs that have been studied with imaging arterial inflammation before and after therapy, as well as have executed large clinical outcome trials. Only in those therapies where arterial inflammation improved was there a benefit in clinical outcome studies. Therefore, this concordance between improvement in arterial inflammation and future CV events provides valuable preliminary data when planning large outcome studies.

There are important limitations which should be considered about the study. The patients were part of an ongoing clinical trial and there was attrition of patient data due to image quality which may bias the results if those patients had more severely diseased vessels. Furthermore, the authors excluded those with TBR<1.6 so these results should not be generalized to those with milder arterial inflammation. Finally, and the most significant limitation of the study relates to the actual imaging of the LMCA inflammation. Similar to a student being told to suspend disbelief in a creative writing class, one must first take the plunge and believe that FDG PET CT can reliably assess left main coronary artery inflammation given that it is a small structure difficult to resolve with current spatial resolution of PET CT. However, it should be noted that the study limited measures to the LM, a structure that does not take an epicardial course and thus is further from myocardial activity, and is subject to less motion during the cardiac cycle (relative to more distant coronary segments). None-the-less, the inability of FDG-PET to reliably measure activity beyond the most proximal segment of the coronary tree represents a major limitation of this approach. Notwithstanding these limitations, the study adds to the growing, strong body of work demonstrating extra coronary arterial inflammation by FDG PET CT provides a reliable surrogate marker to understand the effect of CV interventions on vascular outcomes, including coronary artery disease.