The study under discussion today represents a valuable resource due to its thorough and educational approach to a complex topic such as the treatment of primary cardiac sarcomas. Authored by a multidisciplinary team that includes surgeons, cardiologists (Houston Methodist, Texas), and oncologists (Anderson Cancer Center, Houston, Texas), this work reflects extensive experience, far surpassing that available at any center in our country, and even the individual experience any of us could accumulate over our professional career.

Primary cardiac sarcomas are a diagnostic rarity, with an incidence of 1-3 cases per 100,000 people. Although they do not technically qualify as rare diseases, which are defined by a prevalence of fewer than 5 cases per 10,000 people, the low frequency of these tumors explains the lack of extensive knowledge in this field. This rarity is further compounded by the fact that the ratio of benign to malignant heart tumors is approximately two to one. However, primary cardiac sarcomas are particularly uncommon among malignant neoplasms, with heart involvement more commonly resulting from infiltration by neighboring organs (such as lung, breast, or lymphoma) or from metastatic disease (such as melanoma). Additionally, the metastatic spread to the heart follows a distinct pattern, with the pericardium and epicardium being the most commonly affected areas (adenocarcinomas and squamous cell carcinomas), followed by the myocardium (melanoma). Endocardial involvement is rare and is mainly observed in tumors with intravascular growth, such as renal cell carcinoma, hepatocellular carcinoma, and uterine neoplasia. Other aspects related to cardiac surgery in oncologic patients have been analyzed in previous blog entries.

The study, beyond the wealth of details derived from the extensive experience accumulated by this team, follows a simple design as a classic case series summarizing the institution’s retrospective experience. They compare their results from 1998 to 2021 with those from a U.S. database, dividing their cases into two evolutionary periods (1998-2010, 2011-2021) to determine if there was any impact on outcomes with the accumulation of experience, modernization, and implementation of postoperative care and therapies. All procedures were performed by a single senior surgeon (Michael J. Reardon), and the series includes 122 cases (excluding metastatic tumors, benign tumors, and malignancies originating from other locations with cardiac involvement). Survival rates from the U.S. database at 1, 3, and 5 years were 40%, 15%, and 9%, respectively, while the Houston group’s experience showed improvement with rates of 88%, 43%, and 27%, respectively. No significant differences in survival were observed between the analyzed periods, suggesting the prognosis is more heavily influenced by the pathology than by potential technical or therapeutic advancements. It is also worth noting that the U.S. statistics span 43 years, and given such a wide time frame, a temporal analysis similar to that used for the Houston group’s experience might also have been appropriate. Median survival in the U.S. database was 7 months, while in the Houston group, it was 467 days (15 months) from surgery and 871 days (28 months) from diagnosis.

The authors conclude that managing these complex patients requires centers with extensive experience and multidisciplinary teams. They observed no survival improvements over time, likely due to the systemic nature of the disease.

COMMENTARY:

The lessons gathered from this study are numerous. First, the authors characterize the profile of patients with primary malignant cardiac tumors. These patients have few relevant medical histories that might correlate with the origin of such neoplasms, are relatively young (average age 45 years), and have a slight male predominance (52%). Most are White (73%), although, as commonly observed in the American healthcare system, access to high-level and costly resources often correlates with coverage disparities. The most common location was the left atrium (40%), followed by the right atrium (32%). In fact, tumor location influences clinical presentation, with left-sided tumors typically manifesting symptoms earlier and thus being diagnosed sooner. Right-sided tumors, however, tend to exhibit greater intraluminal growth, leading to later diagnoses and higher degrees of extension to neighboring structures (superior vena cava 1-3%, right ventricle 6%) and presenting distant disease. Ventricular involvement is, fortunately, much less frequent (right 5%; left <2%), as is involvement of the great vessels (pulmonary artery <2%; no cases were reported in the aorta). However, as an extension from the atrial origin, pulmonary vein involvement is not uncommon, which has significant implications for surgical technique.

The authors highlight that many patients were referred from other institutions (54%) where the diagnosis of malignancy was made only after the excision of a tumor initially considered benign, with affected resection margins subsequently identified upon histopathological analysis. The presentation symptoms were varied, with dyspnea (59%), chest pain (29%), cough (12%), and syncope (10%) being the most common. Regarding disease extension, 31% of patients already had metastases at the time of intervention. Although this was not statistically significant, this percentage showed a notable decrease between the two analyzed periods of the Houston experience, from 39% between 1998-2010 to 25% between 2011-2021. Metastases primarily occurred through hematogenous spread, with lung (17%), bone (7%), liver (6%), and brain (3%) as the main sites. The predominant etiologies included angiosarcoma (39%), high-grade sarcoma (15%), anaplastic sarcoma (11%), leiomyosarcoma, and intimal sarcoma (7% each). Remarkably, rhabdomyosarcoma accounted for only 1.6% of the cases.

A significant volume of early reoperations was motivated by patients referred from other centers who had undergone incomplete surgery with histopathological diagnoses of malignancy and affected resection margins. This altered the management protocol considerably compared to patients without prior intervention. The authors describe the therapeutic course followed for a typical patient. First, imaging studies to determine lesion anatomy (cardiac MRI) and disease extension (CT/PET) are essential. The addition of a biopsy, particularly in right-sided tumors, helps exclude etiologies like lymphoma, which may benefit significantly from neoadjuvant treatment. For patients with distant disease, biopsy of the lesion and/or metastases enables the initiation of neoadjuvant treatments before surgery (4-6 cycles of doxorubicin/ifosfamide, taking into account the cardiotoxicity of the former chemotherapeutic agent). The authors emphasize assessing lung function, particularly if resectability involves lobectomy/pneumonectomy techniques when pulmonary veins are affected and/or metastases are present in that location. Coronary angiography was also a relevant study, especially in cases of right-sided involvement where the right coronary artery is often implicated. After multidisciplinary evaluation and shared decision-making with the patient, therapeutic indications for both medical and surgical treatment are made. Eighty-two percent of patients received a neoadjuvant chemotherapy cycle. After the surgical technique, which will be analyzed below, oncological care and treatment were continued.

Regarding the surgical technique, the group has published several studies and proposed various classifications. First, primary cardiac tumors are classified according to the 4th edition of the WHO classification as benign, malignant, and of uncertain behavior (e.g., benign tumors but potentially causing severe or fatal complications such as obstructive myxomas, highly mobile pedunculated tumors, embolizing tumors, etc.). Many classifications focus on histopathology, but the Houston group recommends classification according to the chamber of origin (right heart, left heart, pulmonary artery), due to the technical implications for resection. Briefly, right-sided tumors often involve the right atrioventricular groove and the right coronary artery, infiltrating the ventricle. Their resection may require reconstruction of the coronary artery with an interposition graft and repair/replacement of the tricuspid valve (<10% of the series). In left-sided cases, involvement is typically more restricted to the atrial level, with one of the main limitations for resectability being pulmonary vein involvement. Extension through some of these veins required associated lobectomy and pneumonectomy procedures (24%). The group has described a protocol in which they perform the cardiac intervention with cardiopulmonary bypass on the first day, followed by the thoracic intervention the next day, once organ damage has recovered, and coagulopathy has been reversed. The high complexity of some cases, with biatrial involvement, required autotransplantation techniques in 37% of cases. In-hospital mortality for patients undergoing autotransplantation was 19%, compared to 18% for the entire series. Mortality for cases requiring both autotransplantation and pneumonectomy reached 50%, with the authors noting a reduction in mortality with the two-stage strategy. Notably, likely due to the poor prognosis of these patients given the systemic nature of the disease, the Houston group does not consider alternatives like the total artificial heart (preferred over transplantation to avoid the need for immunosuppression) when resection of the disease is extremely complex or limited by irreparable cardiac defects. Likely, the institution’s policy is reflected in the fact that the series presented consists only of selected cases with resectability options. The volume of patients rejected is not indicated, but the importance of neoadjuvant treatment response is suggested for considering candidates for surgery.

Finally, the authors analyze results based on resective success. They follow a classification of resection margin involvement, common to oncologic surgery and more familiar to other surgical specialties than to ours. R0 indicates the absence of disease at the margins, with adequate microscopic and macroscopic resection margins. R1 denotes microscopic involvement of the resection margin, or that some of the margins are insufficient microscopically to be considered free of disease. R2 is usually an intraoperative diagnosis where macroscopic involvement of the resection margins is evident. Despite good pathological outcomes (R0 47%, R1 39%, R2 5%), the authors conclude that survival was not affected by achieving better resection grades. This finding highlights the unique nature of these neoplasms, which, given their contact with the circulatory system, could be considered systemically disseminated from diagnosis, with the associated prognostic implications. In the case of other solid tumors, findings such as vascular infiltration or affected resection margins have clear prognostic implications. However, conceptually, the initial presentation of these neoplasms in such adverse circumstances almost inherently suggests that applying oncologic surgical principles from other specialties is not entirely relevant; instead, other concepts and classifications should be considered.

In conclusion, efforts to treat patients with primary cardiac tumors continue to yield poor outcomes. Median survival between 7 months (U.S. database) and 15 months (Houston group’s experience) seems inadequate given the monumental diagnostic, surgical, and oncological efforts involved. However, beyond this discouraging paradigm, lie the expectations and decisions of the patient. Thus, without ever losing sight of them, the apparent futility compared to the excellent clinical outcomes achieved in other areas of our specialty can become a benefit, particularly when considering that this devastating disease affects very young patients (35 to 56 years in the Houston series). Hopefully, future advancements in oncology therapy will improve survival outcomes for a disease almost systemic by definition, where resective efforts do not translate into a reduction in recurrence, which seems almost inevitable in the short to medium term.

REFERENCE:

Chan EY, Ali A, Zubair MM, Nguyen DT, Ibarra-Cortez SH, Graviss EA, et al. Primary cardiac sarcomas: Treatment strategies. J Thorac Cardiovasc Surg. 2023 Sep;166(3):828-838.e2. doi: 10.1016/j.jtcvs.2021.10.070.