Talking Tall Rounds®: Cardiac Sarcomas

Podcast Transcript

Announcer:

Welcome to the Talking Tall Rounds series, brought to you by the Sydell and Arnold Miller Family Heart, Vascular and Thoracic Institute at Cleveland Clinic.

Eric Roselli, MD:

Good morning everyone. Welcome to another Tall Rounds. Again, agenda is highlighting the multidisciplinary management of a lot of complex things that we see here at the Cleveland Clinic. One of the things I always say is that we see uncommon problems commonly. And certainly, cardiac sarcomas are a rare disease, but one that requires significant attention from a multidisciplinary team. It's something that we do, I think, quite well here. And again, part of the agenda for Tall Rounds is to focus it on patients. We're going to lead off this excellent Tall Rounds session on the multidisciplinary management of cardiac sarcomas with Holliann Wilkes, one of our fellows presenting a case of a patient. And then, we'll be discussing several other things, and we'll keep it moving along. Holliann, if you want to get us started.

Holliann Willekes, MD:

Good morning. This case was a 67-year-old female who presented to an outside hospital with acute mid-sternal chest pain. Her past medical history included obesity and uncontrolled hypertension. Labs were notable on arrival for elevated troponins. They obtained a left heart cath to rule out NSTEMI, but it showed no coronary artery disease. They also obtained an echocardiogram, which showed a preserved LVEF and no wall motion abnormalities. However, there was an incidental finding of a left atrial mass. So here you can see the mass up in the left atrium where it's originating, and then, it crosses the mitral valve during diastole. There's a couple different views. The outside hospital also obtained a cardiac MRI, which confirmed that there was a large mass originating in the left atrium and then, prolapsing across the mitral valve. Surgical options were discussed with the patient. The patient requested transfer to the Clinic.

When the patient arrived to the Clinic, we got a CT scan, and it shows here the mass extending from the roof of the left atrium. It involves the right superior pulmonary vein, and then, you can see it again across the mitral valve during diastole. So this case was discussed by a multidisciplinary team to determine the best course of action. Given the fact that there was no evidence of metastases and the tumor appeared amenable to resection on imaging, the decision was made to proceed with surgery. Biopsy was not necessary prior to surgery, as the imaging was sufficient to determine the need for surgery and then, would provide us a better specimen for a definitive histologic diagnosis. Operative planning included a 3D model that was constructed, given the complexity of the mass and the structures that it involved. This was set up prior to surgery, and then, the patient underwent surgery on the 27th, which Dr. Roselli is going to discuss in more later in this presentation.

Eric Roselli, MD:

Patrick Collier is the co-director of our Cardio-Oncology Center and co-director of our echocardiography lab and sees a lot of our patients with oncology-based cardiac disease, sometimes related to the effects of therapy, but also, those rare patients that have primary cardiac tumors, such as the one we presented today. And he's going to speak to us about imaging modalities to assess cardiac sarcomas. Dr. Collier?

Patrick Collier, MD, PhD:

Thank you very much, Dr. Roselli. It's an honor to talk today, and it's been a pleasure to look after patients with you and the others on this team. So just to begin, primary cardiac sarcomas, we've heard they're rare tumors, but really, it's fair to say they're rare forms of rare forms of rare tumors. Meaning that, when we think about malignant cardiac tumors, they're certainly a rare subset compared to benign cardiac tumors. And of cardiac malignant tumors, primary forms are much rarer compared to secondary tumors or cardiac metastases. And then, when we think of the body as a whole, in terms of the sarcomas we see in the body, cardiac sarcomas would be a rare subset of those.

And again, echoing what Dr. Roselli mentioned, the importance of the multidisciplinary team. When these cases arise, it's very clear, from the outset, they tend to be complex. And invariably, the first thing we do, when we see these pictures from an imaging perspective, was we ask for help or advice. And I discuss these cases, typically, with my imaging colleagues, many of whom are in this room. And oftentimes, a team is built, because multiple expertise is required and different perspectives are very useful. And this has been outlined in multiple papers about how this might best work in each institution.

So imaging of cardiac masses has evolved over time, and there have been many improvements. And I think, ultimately, it's fair to say that it has meant that we've had a better ability to plan and deliver therapy. When these imaging findings arise, it's important to put these pictures in context with the history. If there's any prior imaging, that can be very helpful. We then think about which other modality we may use, and there are pros and cons with each imaging modality, which I'll discuss a little further. Patient factors may influence, and then, this is a scenario where each modality may add incremental value.

So some patient factors to think about. Well, body mass index can affect image quality. The stability of the patient, the heart rate can affect temporal resolution, for example. The respiratory rate, can the patient tolerate breath hold, for example? Can the patient lie flat? In terms of renal function, are we able to deliver contrast? And then, other comorbidity can also influence things like sedation, breath hold, devices, and that can affect your imaging modality of choice.

Typically, echo is the baseline modality where these tumors may first come to light. Sometimes CT also can be that baseline first test where these tumors arise. We find that CT is probably best for location of masses, and this is quite simplified. MRI for mass content or tissue characterization, PET for mass activity, CAT to identify mass vasculature, and transesophageal echo for peri-procedural guidance. In terms of MRI tissue characterization, here's a patient with a right atrial mass with extension into extracardiac structures on the SSFP image on the left. In a T-1 weighted sequence, isointense at the myocardium. T-2, it's hyper intense and somewhat heterogeneous. With first pass perfusion, this mass uptakes contrast significantly, and that's also evidence on delayed hyper enhancement. And this sort of tissue pictures would be consistent for a sarcoma. In this case, it was a high-grade spindle cell carcinoma.

So there are many questions that we try to answer with imaging. Is it benign or malignant? What is this cardiac mass? How might we get tissue? Very important, is it the only mass we see in the body? It's important to look for metastatic disease and consider whole body scanning, image of the brain. An important question for Dr. Roselli, can it be removed safely? And then, in terms of overall, will the patient likely benefit? In terms of rules of cardiac imaging, malignant tumors are more likely if it's a large size. More than five centimeters is more likely to be malignant. Irregular, ill-defined borders, if there's direct invasion through tissue planes or encasement, right heart involvement, or pericardial or pleural involvement, particularly with the effusions or nodular masses, if there are multiple sites or lesions, if there are lymph node involvement or distant spread, this would all be suggestive of malignant tumors.

Highly perfused tumors, as we've mentioned, tissue heterogeneity, which can represent hemorrhage and the central necrosis within the mass, and if there's rapid growth over time. But masses don't always play by the rules. And ultimately, imaging can suggest a differential, but it's not diagnostic. And ultimately, we require a pathological diagnosis. We need tissue.

So what are the main utilities of imaging? Well, we can provide a cardiac mass differential diagnosis. We can help with staging and assessing tumor effects. As we saw very nicely in the case example, we can help with 3D modeling in specific cases to help guide procedural planning. During procedures, TEE can help with peri-procedural guidance. And then, in terms of treatment follow-up and restaging. Here's some cases. On the right, this patient had multiple imaging modalities, had echo on the top left, CT on the bottom left, MRI top right, and PET bottom right. And in this case, imaging again was strongly consistent with a malignant tumor. I think involvement of the anterior left atrial wall in particular does tend to be very suspicious for sarcoma as you see here. This sort of circumferential mass, it's unlikely to be anything else but a malignant tumor. In this particular case, it was a right atrial sarcoma with extracardiac expansion, again, strongly consistent with a malignant tumor.

And I think, as I mentioned, that imaging has helped over time, really where it's highly unusual for us ever to be in a situation where these are first approached as myxomas and incompletely resected. Usually, we have a good idea before, as we proceed, what we're dealing with. Pulmonary artery angiosarcoma is, I think, historically, quite a rare scenario. But as we're doing more CT-PE scans, and again, our understanding of what this differential may be, we're detecting more of these lesions. My CT colleagues will say, "Beware enhancement of these lesions, beware extravascular spread, and beware central PA expansion." Because these are not features typically of pulmonary emboli, and it would raise at least that differential, rarely. Can we remove it safely? What's the likelihood of achieving negative margins? This is really the opportunity for the patient to achieve a cure. Will affected heart chambers remain functional?

Is there a need for valve intervention, coronary bypass? Is there a need for reconstruction? And these are the key questions that we would interact with Dr. Roselli and the surgical teams with. And finally, when we're considering patient treatments, we have to think about the net benefit of therapy. These are typically poor prognosis cancer types with few existing treatment possibilities, and oftentimes, we're forced to accept toxicities or complications of therapy. The treatment intent is important, if there's a curative intent versus a palliative intent, because really, we're comparing often with limited alternatives or sometimes the risk of doing nothing. Thank you.

Eric Roselli, MD:

Thank you. That's great. Thank you, Dr. Collier. When these tumors come out, again, we have an idea of what they might be, but we count on our pathologists to give us the details. Marc Halushka has been working with our cardiovascular pathology team along with Drs. Tan and Rodriguez. And we send these, sometimes, funny looking specimens to them and ask them to help us with these decisions. He's going to speak to us about the pathology of cardiac sarcomas and give us some of the basic science of what we're dealing with here. Dr. Halushka?

Marc Halushka, MD, PhD:

Thank you very much for inviting me to speak at the Tall Rounds today. It's quite an honor. From a pathology perspective, here's really the rub. Cardiac sarcomas are rare, with an incredibly low incidence. They are ugly, frequently high grade lesions with a lot of pleomorphic looking cells, and they're bad actors, that they have very poor survival characteristics, particularly when they're not resectable. The most common type of cardiac sarcoma is the undifferentiated pleomorphic sarcoma. And this may be a new term for you, but it's the term we're using now in pathology. And it's encompassed a lot of older terms, including malignant fibrous histiocytoma, undifferentiated sarcomas, undifferentiated sarcomas with pleomorphic spindle cells, and sarcoma NOS.

And these undifferentiated pleomorphic sarcomas are somewhat confused with intimal sarcomas, which is more of a tumor of the vasculature. But one third of these UPS have MDM2 amplifications, which are common in intimal sarcomas. These tumors also occur primarily in the left atrium. A quick word on MDM2 amplifications. This gene is the mouse double minute 2 gene, which is a proto-oncogene. Its upregulation leads to the loss of p53 activity, and I'm sure many of us are familiar with p53's important role in malignancies. About a third of all sarcomas are affected by the MDM2 amplification, with over 90% of liposarcomas and about 75% of the intimal sarcomas. As I mentioned, about a third of the UPS have this same amplification, which is probably the source of confusion with intimal sarcomas. There's been one therapeutic attempted as an MDM2 antagonist, RG-7112. However, it had poor responses in a small clinical trial.

This is what a UPS looks like by gross and histology, on the left side, you see a polyploid mass in the left atrium, and on the right side, you see, at low power in the middle, a very cellular tumor, as seen here. And over here, we have the cells, which are very pleomorphic and non-specific in their shape. Here's another picture of a UPS histology from a couple of different case reports. On the left, again, we see some of these funny-looking cells. This one highlights, in these red areas, areas of inflammation, which is an unusual feature of a UPS. These cells of a UPS have been described as atypical pleomorphic discohesive cells that may be spindled, epithelioid, or a combination. So again, a very non-specific finding that captures a lot of these different ugly sarcomas.

Some additional features of an undifferentiated pleomorphic sarcoma, there's no specific or unique staining pattern by immunohistochemistry, smooth-muscle actin, desmin, CD31, ERG, and FLI1, which are markers of smooth-muscle cells and epithelial cells can all be seen in a UPS but are not specific. There's also variable genetic amplifications, MDM2 is already mentioned, including CDK4, PDGF-RA, KIT, and CDKN2A. From a histologic perspective, the differential diagnosis is an angiosarcoma, myxofibrosarcoma, and an intimal sarcoma. An angiosarcoma is the second most common type of cardiac sarcoma, although it is the most common differentiated sarcoma. And in older literature, it's often regarded as the most common type of cardiac sarcoma. It is generally secondary to ionizing radiation or environmental exposure, such as carcinogens. However, cardiac angiosarcomas are not as tightly associated with those entities. It's composed of vascular channels lined by malignant endothelial cells, and it's most frequently seen in the right atrium, whereas the UPS is most frequently seen in the left atrium.

The histology of angiosarcomas are malignant endothelial cells with irregular anastomosing vascular channels, and they have a wide range of histologies. And we frequently see red blood cells, these little red areas, within the tumor, because of those vascular channels. Angiosarcomas will typically stain for endothelial cell markers. So CD31, and ERG are both endothelial cell markers. CD31 stains the cytoplasm of the endothelial cells and ERG stains the nuclei of those endothelial cells. Certain types of angiosarcomas do not stain as well with CD31 as they do with ERG. Additionally, antibodies to CD34, FLI1, and Von Willebrand factor can be useful for delineating these as angiosarcomas. Some additional features of an angiosarcoma, there has been described mutations in the POT1 gene, which was seen in Li–Fraumeni-like families. A number of genes have been implicated in angiosarcomas. And the differential diagnosis is, again, the UPS, mesotheliomas, papillary endothelial hyperplasia.

There's also a litany of other cardiac sarcomas, which are even more rare from this otherwise very rare collection. This includes myxofibrosarcoma, osteosarcoma, leiomyosarcoma, synovial sarcoma, rhabdomyosarcoma, MPNST, and liposarcoma. Some of the histologic features of these different types of sarcomas, we see bone formation, for example, in osteosarcomas, and we see some primitive adipocytes in a liposarcoma.

Going to end by saying there are a number of challenges in the field of pathology for cardiac sarcomas. The first is that we have no single established grading system. There's two different grading systems, the NCI system and the FNCLCC grading systems. And these differ by mitotic rate, the extent of tumor necrosis for staging, and nomenclature as well. So some of these tumors are called different things by these different grading systems, and we really need to get organized on that point. There's also an ongoing reclassification of tumor names. I mentioned an undifferentiated pleomorphic sarcoma as a new name for things that we may have learned in medical school, like malignant fibrosis histiocytoma. And finally, there's generally too few of these tumors for any one center to study very effectively, particularly those rarer entities on that last list. Thank you.

Announcer:

Thank you for listening. We hope you enjoyed the podcast. Like what you heard? Visit Tall Rounds online at clevelandclinic.org/tallrounds, and subscribe for free access to more education on the go.