Challenges Remain With Pancreatic Cancer Staging

Afonso Ribeiro, MD, discusses imaging modalities as well as challenges and future goals for the pancreatic cancer paradigm.

Afonso Ribeiro, MD

Although there are 4 imaging modalities available to detect pancreatic cancer, there are still challenges regarding early identification of the disease as well as making screening more cost effective, explained Afonso Ribeiro, MD.

“Primary care physicians have to be aware of the symptoms [of pancreatic cancer], such as pancreatitis that doesn't have an explanation,” said Ribeiro. “A patient with new-onset diabetes who has gained a significant amount of weight could usually be [experiencing symptoms of] underlying pancreatic cancer. Many times, I see patients who missed their diagnosis 6 months ago when they had a pancreatitis attack, and they were in the hospital for 2 days, and nobody thought that it could have been cancer.”

Additionally, the definition of “borderline resectable” in pancreatic cancer shifts depending on the institution, creating a lack of standard of care in staging for this patient population.

In an interview during the 2019 OncLive® State of the Science Summit™ on Gastrointestinal Malignancies, Ribeiro, medical director of Advanced Endoscopy and Gastroenterology at Memorial Healthcare System, discussed imaging modalities as well as challenges and future goals for pancreatic cancer treatment.

OncLive: What imaging modalities are currently being used to detect pancreatic cancer?

Ribeiro: There are 4 imaging modalities that are mainly available. The most commonly used one is CT scan, which is usually the first one used, and then endoscopic ultrasound. This can be complemented by a PET scan in some cases, or an MRI, if needed. Not all patients will need all 4 imaging [procedures]. Usually, just a CT scan and endoscopic ultrasound will be enough for most patients.

What are the criteria for determining whether a patient is borderline resectable?

It's an interesting question, because [“borderline resectable”] is a somewhat subjective definition and it changes group by group or institution by institution. In general, this involves a large group of patients; a majority will not have an R0 resection if they’re taken to surgery. Those patients will have a higher risk of residual disease after surgery. [The "borderline-resectable" label] usually involves patients who have a lot of tumor contact with the vein or some contact with the artery, so we don't see a clear tissue plane on those blood vessels. Instead of taking a patient to surgery and running the risk of heavy residual disease, we now group them as borderline resectable and give them treatment before surgery.

What advice do you have to make the “borderline-resectable” definition more standardized?

We are having meetings with surgeons and medical oncologists together. A lot of it depends on the skill of your surgeon, how much they are willing to do for that patient, and the patient characteristics. Can [the patient] really withstand an operation that will require arterial reconstruction? For big meetings, we have consensus guidelines that were developed in 2016 that describe the different groups and criteria for borderline-resectable disease.

What factors do you use when considering a patient for neoadjuvant chemotherapy?

Not all patients are the same. There are patients who we may want to receive chemotherapy but can't due to other conditions, such as renal failure or advanced age. On the other hand, there are patients who may seem resectable but, for example, they have multiple lymph nodes or have a very high C19-9 [level], meaning this could be systemic disease or metastatic disease. Those patients usually benefit from additional chemotherapy before surgery. Therefore, yes, there are patients who we will decide on treatment based on their characteristics—not just the anatomical stages.

What challenges remain with imaging modalities for pancreatic cancer?

We still struggle with seeing peritoneal disease in any of the imaging modalities, particular after chemotherapy. Additionally, when you restage a patient, it is very often difficult to make sure that we're seeing the tumor and not necrotic fibrotic tissue.

In the future, with more tumor-targeting images that show whether you're dealing with a viable tumor, [tumor-targeting images] may be [more effective] than just an anatomical imaging that we use currently. There are [also] challenges with [detecting] residual disease after chemotherapy.

Have there been any surgical advances in pancreatic cancer?

Pancreatic cancer is a systemic disease and should be treated as such. Therefore, surgery has a very limited role. We have been doing surgery for this disease for decades, and we still haven't changed the survival [outcomes]. That is probably not the way to go. You can do the surgery better and quicker with laparoscopy, but you still need to get rid of residual disease.

Pancreatic cancer is a disease that likes to stay on nerve cells, which is a sanctuary for pancreatic cancer. We need to get agents that can actually treat the residual disease in nerve cells, which is an area of concern because we can't see that on the imaging test. We need better systemic chemotherapy.

A significant issue in pancreatic cancer is not often being able to detect the disease until it is already in the advanced stage. How can early detection rates be improved?

This is the biggest challenge that we have. How do we detect those tumors? Awareness is the first thing. We don't want to scare the population, but I see a great increase of pancreatic cancer. Almost every week, I see a new case. We have looked at, for example, patients with diabetes who have new-onset diabetes after a certain age. Showing that it's cost effective to screen [for pancreatic cancer] is difficult, because you will screen a lot of patients and find very few cancer cases. We don't have a blood test or a great imaging test to diagnose [pancreatic] cancers early.

With genetic testing becoming more prominent across oncology, how have you seen it affect practice for pancreatic cancer?


At our institution, we see patients who have mutations related to pancreatic cancer or [have] a family history [of the disease]. Examining patients who have the mutation that is associated [with pancreatic cancer] may be more cost effective, because you are looking at a very small group of patients with a likely higher yield.

In initial data that were published, [researchers didn’t identify many patients with these mutations]. They found 4 [pancreatic] cancers out of 120 to 130 patients in their cohort. Looking at patients with a family member who harbors the pancreatic cancer mutation might be more effective. You will probably have a better yield and it will be more cost effective rather than screening the entire population of patients.