Tumors and masses of the heart are extremely rare, but remain an important component and a relevant issue in the field of cardio-oncology. Cardiac masses are a heterogeneous group of disorders and include primary cardiac tumors, both benign and malignant, secondary cardiac tumors, and tumor-like conditions (e.g., thrombus, pericardial cyst).1,2 The incidence of clinically diagnosed primary cardiac tumors is approximately 1380/100 million individuals, and approximately 75% representing benign lesions. Malignant tumors are extremely rare, represent only 5% to 6% of primary cardiac tumors and have typically rapid expansion, invading several cardiac structures such as myocardium, cardiac chambers, and pericardium.3,4,5 Sarcomas are most common (64.8%), followed by lymphomas (27%) and mesotheliomas (8%).1 Cardiac lesions could be asymptomatic and incidentally discovered, or could present with various clinical manifestations such as embolization, obstruction, and arrhythmias. Symptoms depend on tumor size and location, on systemic involvement/response, and on possible thromboembolic phenomena.1,3,6 Early diagnosis and identification of malignant cardiac tumors and masses as well a prompt multidisciplinary treatment may have an important impact on the prognosis. As stated above, primary cardiac sarcoma (PCS) represents the most common primary cardiac malignancy, with a wide range of histological subtypes represented. Symptoms depend on the tumor size and localization (left or right heart, vessels) and could include arrhythmias, pericardial effusion, cerebral or coronary embolization, pulmonary hypertension or pulmonary venous obstruction. Despite aggressive treatments, the overall prognosis for PCS remains unfavorable with a median survival of approximately 1 year.4 The second most frequent primary malignancy involving the heart is the primary cardiac lymphoma (PCL).5 Commonly arise in the right atrium and ventricle and the majority of cases of PCL are diffuse B-cell lymphoma.7,8 Like other cardiac tumors, the symptoms depend on the tumor localization. The most frequent cardiac clinical manifestations associated with PCL are pericardial effusion, heart failure, and atrioventricular block.7 Without treatment, the survival of patients with PCS or PCL may be limited to just a few months. An early diagnosis and promptly therapies may prolong the five-year survival.5 The best treatment for malignant cardiac tumors, when possible, is the surgical excision in combination with systemic chemotherapy. Nevertheless, the tumor could involve extensively cardiac structures, the pulmonary arteries and/or the pulmonary veins, and that impedes an adequate resection.3 Due to the technical risks and difficulties of biopsy, it may be possible to get close to a diagnosis using a structured imaging approach with surgery serving both a diagnostic and a curative role.1,9 Hence, only prompt and robust diagnostic modalities accelerate the initiation of the therapy programs and improve patient prognosis significantly.5

The clinical setting provides critical diagnostic leads in helping establish the etiology of cardiac lesions. Two-dimensional transthoracic echocardiography (TTE) is often the first diagnostic approach in terms of imaging modality, because of its wide availability.1 Previous studies demonstrated that cardiac computed tomography (CT) and magnetic resonance imaging (MRI), both with or without a contrast agent, could help to define the exact localization and the extent of the tumors, to get a differentiation between various cardiac masses such as cardiac thrombi as well to conduct a safe biopsy.5 Positron emissions tomography (PET) is a key molecular imaging modality to noninvasively evaluate the metabolic activity of tumors using 2-[18F]fluorodeoxyglucose (FDG). Moreover, assessment of maximum standardized uptake value (SUVmax) is useful for differentiation between benign and malignant lesions. The combination of FDG PET with other imaging modalities such as CT (PET/CT) or MRI (PET/MRI) allows, using whole body scans, also the identification of possible metastasis and distant embolization foci, improving its sensitivity and specificity. In addition, assessment of FDG PET with a thoracic contrast-enhanced CT (CECT) helps in assessing primary cardiac tumors in adult patients, compared with CECT or PET/CT alone.9 FDG PET/CT imaging provides a cancer disease staging before treatment, a restaging after treatment, and an optimization of a safe biopsy location.1

In the current issue of the journal, Yuan et al.10 investigate the role of PET-morphology and clinical characteristics to distinguish PCL from PCS. Pretreatment PET/CT with FDG and CECT was performed in 14 patients with PCL and in 15 patients with PCS. Patient demographics, clinical symptoms, cardiac conduction disorders, and survival data were collected and imaging morphological and metabolic features of both malignancies were assessed. In particular, metabolic parameters such as SUVmax, SUVpeak, SUVmean, metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were measured. An interesting PET-morphology parameter reflecting the expansion of tumor within the heart, R_Kurtosis, was additionally derived. Regarding demographic data, the patients with PCL were older, presented a higher incidence of cardiac arrhythmias and dysfunctions, and a lower left ventricular ejection fraction compared to patients with PCS. CT images showed a similar incidence of cardiac structures involvement in both diseases, while PET images proved higher SUV values (SUVmax, SUVpeak, SUVmean) and larger metabolic tumor volumes (MTV and TLG) in PLC compared to PCS, with a significant linear correlations between both, PET metabolic and morphologic parameters, and patient demographics. Interestingly, the two parameters independently predictive of PCL were PET-derived tumor expansion pattern (R_Kurtosis) and cardiac conduction disorders.

The authors10 should be congratulated for their incredible effort in recruiting 29 patients with these rare oncological diseases and for improving our understanding of the diagnostic process and the subsequent clinical and therapeutic management of patients with PCL or PCS. A further strength of the study is the use of PET/CT with FDG, as main diagnostic imaging modality. TTE is usually the first-line diagnostic tool in patients with cardiac masses, and when malignancy features have been shown, additional imaging modalities like CMR, CT, or PET are required. As described in the recent multicenter study of Shenoy et al., CMR imaging is a key diagnostic tool to evaluate patients with suspected cardiac tumors. It is useful for tissue characterization of cardiac masses, to detect the tumor intra- and extra-cardiac extension, and for a precise risk stratification of patients with suspected cardiac tumors.11 Nevertheless, as mentioned by Yuan et al.10, many patients because of arrhythmias or cardiac dysfunctions require implantable intra-cardiac devices, which are often not MR-compatible. Furthermore, CMR provides images only of a limited region of the body and does not allow a complete staging of disease, hence other image modalities such as whole body PET is anyway required.

Although PET/CT with FDG is not yet included in the diagnostic routine of cardiac masses, it shows the intra- and extra-cardiac expansion of tumors and allows to assess different tumor metabolic parameters. Indeed, PET/CT with FDG is a validated, noninvasive diagnostic tool in oncology that provides tomographic images and semi-quantitative parameters concerning metabolic activity of target tissues and degree of cell proliferation in the lesions through the intensity of uptake of FDG, which is directly proportional to the degree of malignancy of the lesion.12 PCL shows a higher FDG-uptake compared to PCS. Indeed, PCL progressed faster than PCS, according to the findings of Yuan et al.10 Meng et al. and Qin et al. demonstrated that PET/CT with FDG, and in particular the SUV quantification, has an important diagnostic and prognostic value to distinguish between benign and malignant masses, with improved specificity from 79 to 93%, positive predictive value from 73 to 89%, and accuracy of diagnosis from 85 to 93%, if combined with CECT.9,13,14 Furthermore, the combination of FDG PET/CT and CECT improved the diagnostic accuracy in differentiating between PCL and PCS, as described by the authors10 and by Liu et al.15 Beyond cancer disease staging, PET/CT with FDG could help in performing a targeted safe biopsy of primary cardiac tumor or of metastasis, to get a restaging after treatment and to perform the patients follow up over time.1,12

R_Kurtosis, the PET-derived tumor expansion pattern identified by the author10 as helpful parameter in distinguishing PCL from PCS, deserves a special mention. Yuan et al.10 found higher conventional FDG PET metabolic parameters, such as SUV values and metabolic tumor volumes, in PCL than those of PCA, in line with previous reports.15 Nevertheless, these could be overlapping and new indicators such as R_Kurtosis could improve the diagnostic and prognostic values of FDG PET/CT imaging. Indeed, machine learning-based approaches and artificial intelligence are increasingly used in various fields of nuclear medicine imaging like oncology, neuroimaging, and also in the setting of cardiovascular imaging.16 These new techniques might help to better evaluate also the heart metabolism and to improve the diagnostic accuracy of FDG PET/CT. Indeed, in cardiac studies, high physiological myocardial FDG-uptake may interfere with both analysis and interpretation of the PET images. For this reason, for the study of cardiac diseases, many strategies to reduce physiological FDG-uptake of myocardium, how the high-fat low-carbon diet regime for at least 24 hour prior to the images acquisition performed by Yuan et al.,10 have been proposed and included in the patients preparation.

Another fundamental point of the research of Yuan et al.10 is the integration of FDG PET/CT with CECT. As mentioned before, cardiac CECT could provide anatomical information about tumor location and adjacent infiltration, tissue density, tumor morphology, and contrast enhancement patterns and could help in distinguishing between PCL and PCS, as explained previously by Liu et al.15 The lack of CECT image analysis and of description of tumors characteristics in the present study10 represents a limitation. Indeed, it could be interesting to observe the correlation between CECT characteristics and PET-derived tumor expansion patterns between both CECT features and other variables, such us cardiac conduction disorders or survival.

Despite these limitations, the study of Yuan et al.10 represents a very important step toward incrementing diagnostic and prognostic role of PET/CT with FDG imaging using a PET-derived tumor expansion pattern (R_Kurtosis) in differentiating between the two most common types of primary cardiac malignancies. Further studies, in particular regarding the possible application of artificial intelligence and machine learning in the field of primary cardiac tumors, are required to assist cardiac evaluations, to improve the diagnostic accuracy of PET/CT with FDG, and to perform proper risk stratification in such patients, in order to help a prompt treatment process and to improve the patients long-term survival.