Early detection pancreatic cancer test 85% effective

A new blood test, dubbed PAC-MANN, to help detect pancreatic cancer earlier has been engineered by researchers in America. Pancreatic cancer is notoriously hard to spot early; it usually presents late and with a grim prognosis. It can also be hard to distinguish in the initial stages from some other pancreatic problems, muddying the diagnostic water. The new test looks for pancreatic protease enzymes that have spilled over into the bloodstream, where they should never normally be. It can correctly distinguish patients with pancreatic cancer from non-cancerous pancreatic problems 98% of the time, and used alongside existing cancer diagnostic markers, it can spot early-stage cancer with 85% accuracy. Jose Montoya and, first, Jared Fischer from Oregon Health and Science University…

Jared - So there's currently no good screening method for pancreatic cancer. It's really typically only detected accidentally from an unrelated medical condition or, and also the early warning signs are very vague, such as just, you have pain in your abdomen, you have yellowing of the skin, fatigue, recent onset diabetes, weight loss. I mean, these kinds of conditions are very common for many other types of diseases outside of pancreatic cancer.

Chris - So what did you do, Jose, to try to come up with a better way to find this and find this early?

Jose - It was clear to us that priority was early detection, meaning that we needed to be able to detect that transition from healthy tissue to malignant or cancer disease. And that meant that we needed to find something that was present in early stages and later stages as well. We decided on a type of protein that is called a protease, which if you just allow me this analogy, they sort of work like a really well regulated pair of scissors that the body uses. The body uses proteases to chop proteins when they are not needed. And one of the most common ones is in your stomach when you ingest food, these proteases basically digest the food and cut proteins so then the gut can absorb those nutrients.

Chris - And the pancreas makes one of these proteases and you're saying, well, let's use that because if it's in the blood, it shouldn't be. So therefore it might be a marker of something being amiss with the pancreas.

Jose - That is exactly what it is. So we actually found that even cancer uses these proteases in order to grow. So then we can utilise these markers, these proteins that are present in and look at their activity in order to know how much the cancer is progressing or what stage it is at.

Chris - So Jared, when you do this, how much better is it than what we have at the moment, which is not very good by the sound of it. When can you pick up cancer and how early?

Jared - So how we address that initially was what's called retrospective, where these are samples that are already collected and are frozen and that are found at different stages, which are different levels of the disease, the pancreatic cancer. So stage one is considered a disease that's very local and confined just to the pancreas and can go all the way up to what we call stage four, where it has spread throughout the entire body. And we found that even when the disease is just localised to the pancreas, we get very, very strong detection.

And then the other aspect is how many of the patient's samples that have cancer do you call correctly? So that's called sensitivity. 73 to 85% of the time we can correctly call samples that have cancer. We say that we are highly probable, highly likely that they have cancer.

Chris - The key thing is, though, it's got to be very early. So therefore, how sensitive is it and how likely is it to make a difference to those people who could use this as a screening test and find they have very early pancreatic cancer that would then be amenable to curative or potentially curative treatment?

Jared - In the general population, your chance of getting pancreatic cancer is well less than 1%. But in a subset of patients that have pancreatic damage or even patients with diabetes, their chance of getting pancreatic cancer is elevated to 2%, upwards of 20%. And what we showed is that we never call those patients with these pancreatic inflammation, pancreatic disease that's non-cancer, we never call them with cancer.

Chris - Basically, you've got a really good rule out test here. So if you see something going on, and you apply your test, and it is not positive on your test, it's almost certainly not cancer. And therefore, you can reassure the patient, but it also means that you don't then go hunting for something they haven't got.
That's the critical thing, isn't it?

Jared - Yeah, that's where, especially for early tests, I mean, there's two ways of thinking about it. Do you want to make sure you pick up all the cancers, which is one way of thinking about it. But you also then you never want to call a healthy person to have say that they have cancer. So we're very good at that aspect.

Chris - Jose, how can you take this forward? You've got a good rule out test, it looks like you can call a spade a spade and say this is not cancer. But you're slightly less good at picking up cancer is there, but it's still very, very good if it does pick it up early. But how can you now apply this is the next step to do a really big proper clinical trial rather than a retrospective analysis.

Jose - We're exploring four different avenues right now to try to grow this study. One of them is to make a bigger study similar to the one that we're doing. But we're actually going to use other hospitals to patients from other hospitals, and not just other hospitals, but also other countries, since part of what we want to do is be able to use these not just here where we are, but everywhere in the world. A second one is reiterating what Jared was mentioning is using this high risk patient cohort. We're hoping to start a clinical trial, a proper clinical trial, like you were mentioning, where these high risk patients that have inflammation in the pancreas or other diseases, and hopefully we're able to pick up that some of these patients are transitioning from a non pancreatic cancer to a pancreatic cancer state. We also have the idea of expanding to other cancers. Foundationally and fundamentally, if you look at the biology of these proteins that we call proteases, they are also present in many other cancers. The other cancers use them for the same purpose or slightly different purposes. And I think we can exploit that difference to be able to, again, create early detection tests for other cancers. We are also going to keep improving the technology. The PAC-MANN test that we developed, that needs to be, we need to keep working on it so we can have the best foundation to offer all of these new findings that are coming.