Semi-quantitative metabolic values on FDG PET/CT including extracardiac sites of disease as a predictor of treatment course in patients with cardiac sarcoidosis

Sarcoidosis is an idiopathic multisystem inflammatory granulomatous disease, for which various environmental causes have been proposed in relation to its development, but not yet elucidated [1]. It frequently involves the lungs, eyes, lymph nodes, and skin; however, many of patients do not require treatment until they develop disability or impaired organ function. While cardiac involvement has historically thought to be quite rare, autopsy studies suggest that about 25% of sarcoidosis patients have cardiac involvement [1]. Although approximately 5% of sarcoidosis patients present with cardiac symptoms, anti-inflammatory treatment is generally recommended due to high risk ventricular arrhythmias in this population [1].

Oral corticosteroid is the mainstay among treatment options; however, adverse effects are a major concern with long-term use. Predicting treatment response or prognosis would be beneficial to healthcare providers managing sarcoidosis patients. However, assessing the inflammatory activity of sarcoidosis remains challenging despite current multimodality imaging and laboratory studies.

F-18 FDG PET/CT has emerged as a potentially useful modality for the evaluation of inflammatory diseases such as sarcoidosis [2‐17]. FDG PET/CT has been shown to be of superior diagnostic ability when compared to Gallium-67 scintigraphy, the conventional assessment tool for sarcoidosis inflammation. The main advantage of FDG PET/CT is its ability to demonstrate the metabolic activity as opposed to anatomic imaging such as radiography, CT and magnetic resonance imaging (MRI), which are unable to differentiate an active focus from inactive or fibrotic changes. Moreover, FDG PET/CT allows for the semi-quantification of metabolic activity with standard uptake value (SUV), metabolic volume, or total lesion glycolysis (TLG), which have been shown useful in predicting prognosis or treatment response in oncology. To date, there have been encouraging reports about the usefulness of FDG PET/CT in the evaluation of sarcoidosis regarding to the extent and activity of inflammation [4, 8, 9, 11, 15‐17], response assessment [4, 5, 7, 9], and prognostic assessment [6, 10, 12]. However, the relationship between semi-quantification of FDG uptake and treatment response and prognosis has not been fully investigated. Also, the prognostic value of extracardiac active inflammation noted on FDG-PET/CT in the setting of cardiac sarcoidosis is unclear.

The aim of this study was to investigate the prognostic significance of FDG uptake in both cardiac and extracardiac disease sites as measured by semi-quantitative values derived from F-18 FDG PET/CT (maximum SUV, metabolic volume, and TLG), in predicting treatment course in patients with cardiac sarcoidosis.

Of the 16 patients, 81.3% (13/16) of the patients showed hypermetabolic extracardiac involvement in lymph nodes, 37.5% (6/16) in lung parenchyma, 18.8% (3/16) in the liver, 25% (4/16) in the spleen, and 18.8% (3/16) in bones. The lesion with the highest SUV was identified in the heart in 11 patients (68.7%), in the liver in 1 patient (6.3%), and in lymph nodes in 4 patients (25%).

Twelve patients were classified into a group with a daily oral maintenance dose of prednisone of ≤ 10 mg, and 4 patients were classified into a group with a dose of prednisone of > 10 mg. Representative cases from each group are shown in the Figs. 2 and 3.

The differences in the semi-quantitative values of all involved organ systems and those of the heart between two patient groups are shown in the Table 2 and Fig. 4. There was a statistically significant difference in the maximum SUV among all involved organ systems between the groups, but none of the values in the heart alone showed any significant statistical difference.

As shown in Table 3, there were no statistically significant differences between the presence or absence of pulmonary, lymph node or extrathoracic organ system involvement (liver, spleen, and bone), and oral steroid dose at 6 months.

The relationships between the max SUV and metabolic volume of all involved organ systems, heart, lung, and lymph nodes were shown in Fig. 5. There were statistically significant relationships between the max SUV and metabolic volume of the heart, lung, and lymph nodes, but not in the values of all involved organ systems.

Maximum SUV across all involved organ systems was a significant predictor of maintenance dose of oral corticosteroid at 6 months following initiation of treatment in patients with cardiac sarcoidosis. Neither metabolic volume nor TLG of all involved organ systems showed a significant relationship with maintenance dose of corticosteroid at 6 months. On the other hand, none of the three semi-quantitative values in the heart predicted corticosteroid dose at 6 months. Maximum SUV of the heart did not reach statistical significance, though it showed weak correlation with daily steroid dose at 6 months. This was attributed to extracardiac organ system involvement with greater maximum SUV than the heart, which was noted in about one third of the patients. Similarly, none of the involved extracardiac organ systems (lung, lymph node, and extrathoracic disease sites) independently demonstrated significant relationship with daily steroid dose at 6 months.

The present study suggests that evaluation of extracardiac organ systems, in addition to the heart, is helpful to estimate the short- to intermediate-term treatment course in patients with active cardiac sarcoidosis. While corticosteroid therapy is the first line treatment option for active sarcoidosis, its adverse effects could limit the benefit of treatment, especially for those patients requiring long-term treatment. Assessment of FDG PET/CT of the thoracic and upper abdomen is beneficial for patients expected to require prolonged treatment by providing prognostic information so that treatment regimen could be tailored, such as early addition of or switch to methotrexate or other immunomodulating therapy, to maximize the patient benefit [1].

FDG PET/CT has been shown to be useful in determining treatment option in patients with sarcoidosis [7, 9, 16], largely due to the detection of metabolically active disease [4, 8, 9, 11, 15‐17] and evaluation of treatment response [4, 5, 7, 9]. To date, a few studies have been published evaluating the prognostic value of PET/CT [6, 10, 12]. Keijsers et al. [10] investigated the prognostic value of metabolic activity in the lung parenchyma with 43 newly diagnosed sarcoidosis patients. They found that if untreated, diffuse parenchymal disease on PET/CT, as opposed to parenchymal disease on CT without metabolic activity, predicted future deterioration of diffusion capacity measured by pulmonary function test. Blankstein et al. [6] focused on the cardiac FDG PET/CT findings concurrently performed with Rubidium-82 perfusion PET/CT. They concluded that abnormal FDG activity accompanied by focal perfusion defect on Rubidium-82 PET/CT was a significant indicator of future major cardiac event. These studies primarily investigated patients who were not undergoing corticosteroid or other anti-inflammatory therapy and therefore, their findings would influence the timing of medical intervention, but not the treatment regimen. In contrast, the present study highlights the prognostic value of FDG PET/CT regarding treatment course in patients who are on corticosteroid therapy, and thus helpful in tailoring the treatment regimen. These data synergistically further enhance the value of PET/CT in evaluating patients with clinically active sarcoidosis.

The present study shows that the intensity of sarcoidosis expressed by the maximum SUV is a more significant prognostic indicator than metabolic volume, which reflects the extent of inflammation. Physiologically, this finding appears reasonable as the anti-inflammatory effect of corticosteroid would be expected to be less in highly active inflammatory foci even if small, while a pronounced response would be expected in less active foci even if extensive. Alternatively, it is noteworthy that there was no significant relationship between the intensity and extent of inflammation when all involved organ systems are considered as a whole. This may imply that a patient with extensive disease does not always present with intense disease or vice versa. Moreover, it is even more significant in terms of prognosis when a patient presents with limited but intense disease. Therefore, caution should be exercised during interpretation of FDG PET/CT not to underestimate the amount of corticosteroids needed to successfully treat small volume disease of high intensity.

However, as one may presume, the intensity and extent of the disease are not independent when examining each involved organ. There was a significant correlation between the intensity and extent of inflammation in each involved organ systems including the heart, lung, and lymph node, which are frequent sites of disease. This implies that, in daily clinical practice, the more extensively involved organs should be scrutinized for intensity of inflammation during image interpretation.

The study has several limitations. First, the present study is retrospective and the sample size was relatively small, partly because the focus of this study is patients with active cardiac sarcoidosis. Therefore, the results may not be generalizable to a larger population or to sarcoidosis patients without cardiac involvement. However, these findings are beneficial given the high morbidity and mortality of cardiac sarcoidosis. Second, we chose maintenance steroid dose at 6 months as a surrogate for treatment response, but this might not exactly reflect treatment response. However, patients’ response eventually reflect in their treatment, and treatment goal is to control cardiac sarcoidosis with medication including steroid of as low dose as achievable. Therefore, we believe it is justified to use steroid dose as a measure of treatment response. We were not able to incorporate objective follow-up findings such as PET/CT, pulmonary function test, and cardiac ultrasound into treatment response analysis because patient pool is heterogenous due to retrospective nature and not all patients had these follow-up studies. However, daily steroid use at the time of follow-up PET/CT might affect FDG activity in inflammatory foci in the heart and elsewhere due to elevations in blood glucose and insulin levels and compromise the results. Third, there was no reference standard for the diagnosis of cardiac sarcoidosis. None of the enrolled patients had positive endomyocardial biopsy; however, 10 of 16 patients had pathological evidence of sarcoidosis elsewhere and met the widely accepted diagnostic criteria established by the Japanese Ministry of Health and Welfare [18] or Heart Rhythm Society [19]. Six patients did not meet the criteria because these patients lacked extracardiac pathological confirmation of sarcoidosis, though three of them showed positive late gadolinium enhancement of cardiac MRI which was consistent with sarcoidosis involvement. These patients might have suffered from other cardiac inflammatory pathology though the diagnosis was made by our experienced sarcoidosis experts at the referring center. Fourth, patients underwent FDG PET/CT of the thorax and upper abdomen, not whole body coverage (from base of the skull to thigh), which is the standard protocol in oncology. The field of view from the thorax to the upper abdomen covers the most frequent sites of involvement in sarcoidosis, that said, there remains a possibility that disease involvement was underestimated.

The maximum SUV across all involved organs of the chest and upper abdomen, not that of the heart alone, could be a predictor of treatment course of corticosteroid therapy in patients with presumed active cardiac sarcoidosis. This finding underscores the importance of evaluating extracardiac sites potentially, when evaluating patients with suspected sarcoidosis.