Abstract
Background
Primary cardiac leiomyosarcoma is an extremely rare entity. There is no accumulated knowledge about the clinical characteristics and therapeutic strategies for this tumor. This study aimed to systematically review the available literature to investigate the clinicoradiologic and clinicopathologic characteristics, treatment, and outcomes in patients with primary cardiac leiomyosarcoma.
Methods
We identified 79 instances derived from 72 unique reports and carefully analyzed all clinical presentations, imaging, pathology, treatments, and outcomes.
Results
The mean age at onset was 48 years. A symptom of obstruction was the most frequent complaint at diagnosis. Routine imaging methods such as echocardiography, computed tomography, and magnetic resonance imaging had high sensitivity in diagnosis. The tumor has a poorer life expectancy than its counterparts in other sites. The 5-year overall survival and recurrence-free survival rates were 25.4% and 14.7%, respectively. Overall survival was affected by age, surgery, and adjunctive chemotherapy or radiotherapy.
Conclusions
Our findings suggest that primary cardiac leiomyosarcomas tend to be biologically more aggressive compared to their counterparts in other sites. However, complete resection and adjunctive chemotherapy or radiotherapy may help to increase the life expectancy.
Introduction
Primary cardiac tumors are uncommon with an incidence of 0.02% per person-year, and only a small minority are malignant. 1 Most primary cardiac malignancies are soft tissue sarcomas, of which there are many types, and all have a common mesenchymal origin. The most frequent cardiac sarcoma is angiosarcoma, followed by undifferentiated high-grade pleomorphic sarcoma and rhabdomyosarcoma. 2 Primary cardiac leiomyosarcoma (PCLMS) is an extremely rare entity. In most studies, it has been defined as a leiomyosarcoma arising only from the heart chambers. According to this definition, there have been no more than 200 cases reported in the PubMed database to date, including publications in non-English languages and those without full-text sources. Because of its rarity, there has been no large cohort study to date, although series with as many as 4 cases have been published. 3 According to our previous studies, the outcome of cardiac malignancies is dependent on tumor type, clinical features, and treatment strategies.4–8 For example, the overall survival (OS) pattern of primary cardiac synovial sarcomas was affected by age and chemo-/radiotherapy. 6 The prognosis of primary cardiac undifferentiated pleomorphic sarcomas was very poor even though complete resection and aggressive chemotherapy were performed. 5 It is well known that leiomyosarcoma is a common form of soft tissue sarcoma that is composed of cells showing distinct smooth muscle features. However, there is a paucity of literature that defines the outcomes for primary non-uterine leiomyosarcoma only. Although previous reports have noted that tumor grade, depth, and size are potential prognostic factors for tumor recurrence, metastasis, and OS,9,10 the prognostic factors in PCLMS are still very unclear. In addition, there is limited knowledge about the clinico-radiologic and clinicopathologic features of PCLMS. In this study, we carefully reviewed 79 cases of PCLMS and investigated the relationship between clinical characteristics and outcomes.
Methods
We performed PubMed searches using the terms “cardiac AND leiomyosarcoma”, “heart AND leiomyosarcoma” to identify potential eligible reports. Cases meeting the following criteria were included in the study: the tumor arose from heart chambers and/or pericardium, but not the great vessels; the cardiac tumor was the primary lesion; the source article was written in English and the full-text, no matter electronic or printed, could be obtained; cases without detailed descriptions were excluded; the final diagnosis was definitely confirmed by histopathology. All references cited in these reports were also evaluated to include any potential missing cases. In addition, autopsy instances (diagnosed postmortem) were included to expand the sample size, if the history was clearly presented. Approximately 200 PCLMAS cases could be retrieved from the database, however, a considerable number of them were described in a non-English language, and some indexed items did not have an available literature source. Finally, a total of 79 instances derived from 72 reports until 30 June 2014 were included in this study3,11–47,52,54–86 (see Supplementary Table 1).
Clinical data including age at presentation, imaging examination, tumor size, surgical margin status (complete resection or not), adjunctive therapy, and outcome were collected and entered into Excel software (Microsoft Corp, Redmond, WA, USA). Prism 5.0 software (GraphPad Software, La Jolla, CA, USA) was used for statistical analysis. Survival analysis was performed using the Kaplan-Meier method. A log-rank test was performed for intergroup comparisons. A p value of less than 0.05 was considered statistically significant.
Results
The study included 79 patients with a median age of 48 years (range 6 months to 86 years). The total patient cohort comprised 43 women and 34 men (the sex information was not provided in 2 cases; Figure 1). Clinical presentations were recorded in 64 cases, in which a symptoms of obstruction (dyspnea, cough, chest discomfort, and edema) was the most frequent complaint at diagnosis (50 cases, 78.1%). Arrhythmia-related symptoms (palpitation, irregular heartbeat, tachycardia, and atrial flutter or fibrillation) were reported in 8 cases (12.5%).11–18 Central nervous system disorders (headache, syncope, dizziness, loss of consciousness, and paralysis) were also reported in this cohort, accounting for 10.9% (7 cases).11,19–24 Nonspecific systemic symptoms such as fatigue, malaise, fever, sweating, asthenia, and weight loss were reported in 19 (29.7%) cases.3,13,17,22–36 Only 3 patients were asymptomatic.37–39 PCLMS predominately arose from the left atrium (59.5%, 47/79). Other documented locations included the right ventricle in 16 (20.2%), right atrium in 12 (15.2%), and left ventricle in 4 (5.1%). Sometimes, the tumor may display an infiltrative growth pattern and involve the cardiac valve or occupy two or more heart chambers (17 cases, 22.1%). Unlike cardiac myxoma, PCLMS rarely had a stalk,35,40 and rarely originated from the interatrial septum.18,20,41
Age and sex distribution is shown of patients with reported primary cardiac leiomyosarcoma.
Imaging findings
Usually, the routine imaging examination methods, for example, echocardiography, computed tomography (CT) and magnetic resonance imaging (MRI), could well outline the tumor. Of the 49 patients with an echocardiographic diagnosis (transthoracic and/or transesophageal), only 2 were missed;34,42 these were located in the right ventricle outflow tract and the patients had not undergone transesophageal echocardiography. The sensitivity of CT and MRI detection was 100% (22 CT records and 19 MRI records). Sometimes, the tumor may be misdiagnosed as cardiac myxoma by the imaging examinations.20,43,44 Positron-emission tomography was documented in 3 case, of which 1 was missed. 23
Pathology
Grossly, PCLMS often appeared multilobular, nodular, or polypoid. Thus the tumor may often be taken as myxoma on imaging detection, especially when it is located in the left atrium. Tumor dimensions were reported in 48 instances, and of these, 32 (66.6%) showed disease greater than 5 cm and 4 (8.3%) were greater than 10 cm. The median maximal diameter was 6.0 cm (range 1.5 to 13 cm). The tumor cell may appear spindle, epithelioid, or pleomorphic. Interestingly, epithelioid tumor cells may show distinct positive staining for epithelial markers such as CAM5.2, cytokeratin and AE1/3. 18 Sometimes, an extensive, loose myxoid stroma was present, by which a diagnosis of myxoid leiomyosarcoma could be made.45,46 Necrotic foci could be easily identified in most cases. Mitotic figure values were given in 29 instances with a frequency of 1–100 per 10 high-power fields. Values for Ki-67 were reported only in 3 instances, with values of 20%, 30%, and 15%, respectively.13,37,47
Therapy
A total of 72 patients were treated by any modality. The most common modality was surgery, which was administered in 71 cases. Complete resection, either grossly or microscopically, was recorded in 16 instances, and incomplete resection was clearly reported in 26. A total of 32 patients accepted adjunctive chemotherapy, and 11 accepted radiotherapy. The reported agents included Adriamycin, cisplatin, dacarbazine, dexamethasone, doxorubicin, ifosfamide, gemcitabine, and vincristine.
Survival analysis
Overall survival data and clinical factors in 66 patients with primary cardiac leiomyosarcoma.
A case with mitosis data of 4–6/10HPF was included. †Reports stating: “the tumor was resected as completely as possible” were included. HPF: high-power fields.
Overall survival curves of 7 variables. Age, surgery, and adjunctive therapy were found to significantly impact survival.
Discussion
In this study, we reviewed 79 unique PCLMS cases derived from 72 isolated reports. Our data show that this tumor had a poorer life expectancy than its counterparts in other sites. In addition, the present study has demonstrated for the first time that OS is affected by age, surgery, and adjunctive chemo-/radiotherapy. PCLMS did not show a distinct difference between sexes. The median age of PCLMS patients was younger than those with other non-uterine leiomyosarcomas.48,49 The clinical features of most primary cardiac tumors often appeared nonspecific. Just as with cardiac myxomas and other primary cardiac sarcomas, the most common presenting symptom of PCLMS was obstruction.4–7,50 The rate of occurrence of central nervous system disorders in PCLMS was almost equal to that in cardiac myxomas. We consider that these disorders are due to decreased cardiac output. Systemic disturbances may be caused by immune responses against the tumor.
The sensitivity of imaging (echocardiography, CT, and MRI) in detecting PCLMS was very high. According the current data, only 2 cases originating from the right ventricular outflow tract were misdiagnosed. In fact, to make a diagnosis of PCLMS in this site is very difficult because the clinical symptoms are usually similar to those of pulmonary thromboembolism, which makes not only early diagnosis difficult but also leads to misdiagnosis. 51 The leiomyosarcoma rarely originated from the interatrial septum, and rarely had an obvious pedicle, in contrast to myxoma. Some researchers considered that immobility of the mass, the presence of pericardial effusion, a broad base of the tumor, and neovascularity are suggestive of malignancy. 52 Interestingly, PCLMS is similar to myxoma in having a relatively well-defined border. 19 The definite diagnosis must be confirmed by both histopathological and immunohistochemical examination, where a strong positive reaction in the tumor cells to α-smooth-muscle actin and desmin is diagnostic.
The 5-year OS rate for PCLMS was far lower than that in other sites. In this series, the 5-year OS rate was only 25.4%. In contrast, earlier studies reported a 5-year OS rate for non-selected deep-seated leiomyosarcomas of 50%–66%.9,49,53 In addition, we found that age > 50 years and an inadequate surgical margin were the two most important risk factors for shorter OS, and adjunctive chemotherapy and radiotherapy were of benefit with a better prognosis. These findings are not in accordance with recent studies that showed that tumor size and malignancy grade were significant independent predictors of OS, whereas the margin status was only responsible for recurrence-free survival.9,49 However, an earlier study reported that age and mitoses correlated with OS, and tumor size, malignancy grade, and type of surgical margin did not influence OS. 53 Our results partly corroborate this view. In fact, we used 5 and 10 cm instead of 6 cm (median size) as the cutoff values of tumor size to make the log-rank analysis, but the differences were still not significant. The reasons for the different observations in these reports are still unclear. We consider the outcome patterns for leiomyosarcoma may be verified by body sites, and constituent ratio discrepancies existed in different studies; thus the results did not agree with each other. Nonetheless, the current data suggest PCLMS patients have a very poor life expectancy.
We believe our study may provide some useful information regarding the diagnosis and management of PCLMS. However, we must acknowledge that the present data have the limitations of a retrospective study along with the lack of availability of all of the pertinent clinical and demographic information. When using case reports as a source of data, there is a bias in favor of reporting those cases with unique features, and a lack of standardization as to what features should or should not be reported. While comparing the deceased and surviving patients, due to the lack of availability, we did not take into account certain factors such as the quality of medical service in different times and different medical centers. However, we concluded from these findings that PCLMS tend to be biologically more aggressive than their counterparts in other sites, and complete resection and adjunctive chemo-/radiotherapy may help to increase life expectancy.
Footnotes
Conflict of interest statement
The authors declare no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commerical, or not-for-profit sectors.