Moffitt Expert Emphasizes Need to Improve Sarcoma Diagnoses

In Partnership With:

Partner | Cancer Centers | <b>Moffitt Cancer Center</b>

Marilyn M. Bui, MD, PhD, highlights the importance of pathology in sarcoma and how it is the cornerstone of precision medicine.

Pathology is the cornerstone of precision medicine, explained Marilyn M. Bui, MD, PhD, who added that, as a whole, the health care specialty needs to step it up in the diagnosis of sarcomas and the differentiation between the various histological subtypes.

“Pathologists are diagnostic physicians; we could be the bottleneck for the entire operation, or we could be the catalyst for the change the transformation in precision medicine,” said Bui, a senior member in the Department of Anatomic Pathology and director of the Sarcoma Cytopathology Fellowship at Moffitt Cancer Center. “This is a very important profession.”

As sarcomas in general are rare, Bui emphasized the importance of getting the right histological subtype diagnosed in an individual patient. For example, in a paper published in Cancer Control, results of a meta-analysis showed that common pitfalls in the pathology of retroperitoneal sarcoma exist and informative diagnoses are necessary to accurately and effectively make treatment decisions.

In an interview with OncLive, Bui, who is also the scientific director of Moffitt’s Analytic Microscopy Core, highlighted the importance of pathology and its digital transformation, especially as it relates to sarcoma diagnoses, and some of the diagnostic pitfalls that remain in this landscape.

OncLive: How would you define the importance of pathology in modern cancer care?

Bui: If a patient feels like there is a lump in her breast, she's worried about breast cancer. She will go to a cancer hospital. The surgeon will look at it and say "Yes, I'm worried about it too." They're going to send the patient for a radiology review, and we'll take a picture. Yes, I can grade [the suspicious tumor], but nobody knows exactly whether this is a cancer or not until a biopsy is done and the tissue is taken from the patient. [The results] then come to my desk, and I will make that diagnosis.

From that point on, the patient care journey starts. [From my research], not only was I able to provide a diagnosis with the tumor type and tumor grade, I'm also able to provide the biomarker status for patients [which will inform the] prognosis and prediction of their disease. That's why I say, "Pathology is the cornerstone of all medicine."

There is a report coming out showing that there is a decreased number of practicing pathologists. On the other hand, there is a shortage of [pathologists in] our pipeline. There is a 27.5% decrease of US medical students who apply to pathology. However, because of precision medicine [efforts], there is a 40% increase in diagnostic needs. For example, lung cancer used to be either non–small cell lung cancer or small cell lung cancer; that is what was required from pathologists [when making a diagnosis]. Now, we need to know if it's nonsquamous non–small cell adenocarcinoma or squamous cell carcinoma. We also need to know what the PD-L1 status is. What's the molecular profile for that patient?

There will be a lot more tests that need to be done by a pathologist; that demand has increased. We need to emphasize that pathologists are such an important, integrated part of the clinical team. Any successful hospital or cancer center needs to have a very strong pathology department. That's point number 1.

Point number 2 is that digital pathology is important for digital health. We [evolved to] immunohistochemistry. The second transformation was molecular pathology. Now, we're going through this digital transformation. 

There are a lot of retrospective studies looking at biomarkers and factors that could be linked to poor prognosis. What other studies can be done to better identify these patients?

As you know, sarcoma is very rare, it represents 1% of the cancers. Because of the rarity, people in the community don't usually don't see them. They all come to places like Moffitt Cancer Center, since we have the expertise to be able to look at [these types of tumors] collectively. One challenging diagnosis is for malignant peripheral nerve sheath tumors [MPNST]. There are 2 types, and one of the types is genomically programmed. There are patients who, down the road, eventually develop the tumor, but there are those random ones [that develop], and we don't know what the reason [is for its development].

That tumor, morphologically, has a very different diagnosis. However, if you know the molecular profile, [you can find] immunohistochemical diagnostic markers that will really help to make the diagnosis. If you know what the driver is, that really helps with choosing targeted therapy. There is the Genomics of MPNSTs Consortium, which [is a group that focuses on advancing preclinical research in this disease.]

Moffitt is one of the contributors to that [program]. We recently published in Genes introducing this program, because we're going to map out the molecular landscape of MPNST. The second paper is going to come out with the details of [our findings]. Then, in the following paper, we'll come up with potential diagnostic markers, so I'm very excited about that project.

This is a good opportunity to give a shout-out on projects like that, because [the disease is] so rare. We have collected close to 100 cases, and the collaborators are from the United States, Europe, and Canada. It's very exciting.

How would you define the diagnostic challenges across the sarcoma subtypes?

As far as diagnosis goes, we have to standardize the language and classification of these tumors. Something new in the diagnostic world, is the updated World Health Organization Classification of Tumors of Soft Tissue and Bone. There is a lot of information on that. One part focuses on the molecular and immunohistochemical biomarkers that are used for diagnosis, prognosis and prediction. I will go the extra miles trying to hit every part of the checklist. What's the diagnostic type? What's the subtype? What are the molecular markers for diagnosis and processes? That works very well for me diagnostically and also [for education] when you see that type of tumor.

Secondly, my research interest is to identify diagnostic pitfalls. We did a study [in retroperitoneal sarcoma] that number one, tumors should be differentiated from liposarcoma [when subtyping retroperitoneal sarcoma]. There are ancillary testings you can do to confirm this. I published a paper on this, and I'm following up, to see if the cases coming back to me have changed.

From the community, I can see people started picking up those tips [to improve diagnoses], but there are still some people in the academic setting that for some reason, [subtyping is] still a blind spot for them. This is a still a challenge. I still have to do more to really reach out to our pathology community, and talk about what the tumors [to better diagnose so] we can really overcome those diagnostic pitfalls.

Reference:

Pham V, Henderson-Jackson E, Doepker MP, et al. Practical issues for retroperitoneal sarcoma. Cancer Control. 2016;23(3):249-264. doi: 10.1177/107327481602300308