Cancer Screening in Transition: From Imaging to Innovation

In this episode of Breaking Down Health Care, John Hennessy and Dr Michael Kolodziej explore the evolving landscape of cancer screening—from the advances in imaging and artificial intelligence to the promise and pitfalls of new DNA- and blood-based detection tests—and discuss how technology, access, and policy are reshaping early cancer detection and patient care.

John Hennessy, MBA: Hi, welcome to Breaking Down Health Care. I'm John Hennessy. I'm an associate principal at Payer Sciences. I'm joined by my good friend and colleague, Mike Kolodziej. Mike, today we're going to talk about cancer screening. It's been a few years, but I remember when mammography was an early part of cancer screening and we were using what we would consider now primitive devices to do that imaging.

Now we've got some really amazing technology that allows us to see things in ways we couldn't see them in the past. So, maybe talk a little bit about how the improvement in our ability to see things when we're doing cancer screenings has changed the calculus for patients, health plans, and practitioners.

Michael Kolodziej, MD: It's an interesting way to discuss the current landscape in cancer screening, it sets the table for the big changes that are likely to happen soon. For those of us of a certain vintage, you and I especially, we think that cancer screening has been around forever, but it really hasn't. It is really in our lifetimes.

Mammography, of course, was the first and biggest, starting in the 1960s and then rolling up to where we are today. We have 3 other cancers for which screening is well established and endorsed: cervical cancer, colorectal cancer, and lung cancer. For those 4 cancers we have shown—using what would today be considered relatively simplistic technology—how to reduce cancer specific mortality.

To say that screening saves lives is a little bit of an exaggeration. Mammography reduces your risk of dying of breast cancer but doesn't save your life from anything else. The one exception is lung cancer, where, for reasons that are not at all clear, low-dose CT screening reduces the risk of all-cause mortality, at least in one of the biggest studies.

It is true that the breast tomosynthesis or some of the other more sophisticated imaging studies probably do pick up more cancers. We will never repeat the screening studies that led to the current recommendation, it's just unfeasible to study tens of thousands of patients. That's what the original studies did. It's probably fair to say that, because they do have increased sensitivity, whatever the mortality-specific benefits that were observed will certainly be maintained.

What we need to be more concerned about, perhaps, is that those increased sensitivity technological advances probably reduce the specificity. Even for the 4 cancers that we we've embraced screening for, for at least 2 of them—and arguably 3—false positives have a serious consequence. For mammography, not only do you drive a lot of women nuts, but you also have unnecessary biopsies, follow-up imaging, etc. In lung cancer, of course, the indeterminate pulmonary nodule creates a whole series of problems.

Now, I think that as we use artificial intelligence (AI) more intelligently with these higher technology imaging modalities, we're going to get a lot better. We're going to know, to a much greater degree, how worried we have to be about a particular finding on an imaging study. That's going to be really good.

I think the other thing is that screening modalities are population-level tests. They're screening for a large population with the recognition that you're going to test a lot of people to save only a couple of people heartache. There is a thought that you can use a more intelligent risk-adapted approach that will improve your test performance. There's been a lot of work done in this in breast cancer. In prostate cancer, for example, where there is not an endorsed screening modality, a lot of people have—almost by design but without good evidence—adapted the screening strategies for prostate-specific antigen testing, digital rectal examination, and imaging, so that more screening is done in African American men and those who have a family history of prostate cancer, which are enriched populations.

We will continue to have advancements in imaging modalities, but I believe they're going to largely be eclipsed by some of the other DNA-based technologies, which we're going to talk about here in a second.

Hennessy: It's a great point. When you talk about prostate cancer, for instance, I can tell you that my annual physicals have become much more pleasant in the last couple years as guidelines have changed.

Dr Kolodziej: Prostate cancer is such an interesting subject because when former president Biden was diagnosed with advanced prostate cancer recently, there were some really stupid things being said. Zeke Emanuel said some incredibly stupid things about why he thinks Joe Biden should have undergone screening.

We can argue about whether men in their 40s, 50s or even 60s should be screened, there aren't a lot of guidelines that promote prostate cancer screening in men in their 70s. When Joe Biden was president, he was in his 70s. His primary care physician should not be held up for scrutiny or criticism. The guidelines don't support screening. Should you screen a president differently? I don't know. On the other hand, with prostate cancer, if you screen and you treat and the president winds up being incontinent, that's not a good look for the United States and the international state.

All I'll say is this: I think there's so much emotion tied up in this subject. As health care providers, we should take a step back. Prostate cancer is a great example of where shared decision-making should be the methodology that's used to determine whether or not screening is even a good strategy.

Hennessy: That's a great point. We have different options now. We talked about DNA, and last year at one point I was shipped a box to my house and that box sat up on a shelf for quite some time, but it was eventually used. We have these trade-offs of convenience and, in our lifetime—certainly as professionals—we went from mammography in the hospital to freestanding breast centers that were almost boutiques.

Now the box comes to your house, it's easy to do. You don't have to worry about prep for a colonoscopy. Although the propofol, I must say, was one of the best experiences of my life once I got to that point. But talk about the idea that we're trading access for specificity and things like that.

Dr Kolodziej: It's kind of the old saying that if a doctor prescribes a medicine, you don't take it, and it sits in the bottle, there's no chance of it helping you. In the same fashion, adherence to a guideline recommended cancer screening is extremely variable, depending on the cancer.

Cervical cancer is the gold-medal example. Cervical cancer screening, in this country, is close to 90%. That's great, but we probably don't need that as much anymore, partially because of the HPV vaccine, partially because HPV serotype testing is a routine part of a GYN exam these days. We know when trouble is coming, and we have headed it off in the past with the vaccine.

The next best is breast cancer screening mammography (75%-80%) Next is colorectal cancer screening with colonoscopy (65%-70%). Clearly, the loser here is lung cancer screening, which is only about 20%.

Why is it so low? Part of it certainly is convenience. Part of it is patient preference. I cannot tell you how many times a friend, a family member, or a patient's family said, "Oh, I don't want to have a mammogram. I hear they hurt." The answer is yes; the compression can be uncomfortable. That's true. When I tell people I've had 3 colonoscopies, they look at me like I'm insane. There's a reason for that, let's leave that aside. But the fact of the matter is, colonoscopy preps are horrible. They are horrible.

Lung cancer is a special thing because I'm not sure why we're so low. Part of it is because primary care doctors are not ordering the test. They're just not ordering the test. We have to solve for the adherence problem.

You've mentioned Cologuard. It's DNA-based testing of stool to identify patients who are likely to harbor colorectal cancer and, to some extent, a polyp. It sounds like a great solution, but a lot of people find it icky. In addition, some payers don't pay for it. Although, because it's United States Preventive Services Taskforce (USPSTF) approved, most do.

Sometimes the payer pays for colorectal cancer screening but has a preferred methodology that they want you to use first. I was shocked, actually, that Cologuard didn't move the needle so much because it's icky, but we don't have a lung cancer test like that and we don't have a breast cancer test like that. We don't have a prostate cancer test like that. We need something that people find acceptable and convenient, and that's where we get into the blood-based testing.

Hennessy: Let's open that up: multi-cancer early detection (MCED) tests. It's something that, from a patient perspective, sounds very promising. Rather than the logistics, toxicities, pain, or the inconvenience involved in current screening, I just drop some blood and a miracle occurs. But one of the things we're challenged by is also adding other traditional screening tools to it or dealing with this challenge of the false positives.

To your point, those are not insignificant. They are not only challenging from a patient perspective, in terms of going through procedures that maybe you wouldn't have to, but it also costs money. If someone says, "Listen, you have a positive MCED, you need to get a PET scan and you have to pay for that with cash," that's going to be a problem.

Let's talk beyond that, once you've identified someone at risk like this, who's going to take care of these folks? Primary care? Oncology? What's the plan?

Dr Kolodziej: Let's talk about what you do with a positive test. At present, for the most part, a positive MCED test falls upon the physician who ordered the test and usually results in a referral.

Let's say you have a positive test for a gastrointestinal malignancy and you're referred to a gastroenterologist for endoscopy and colonoscopy. Because of the Affordable Care Act, cancer screening that's endorsed by the USPSTF is provided to both commercial health plan patients as well as government payer patients without any beneficiary contribution.

In other words, you don't have to pay anything out-of-pocket to go and get your colonoscopy if you're a Medicare beneficiary. However, if you had a positive MCED, it's not technically a screening test anymore. What we have seen is that there's a significant out-of-pocket consequence if you go and have your upper endoscopy and your colonoscopy. For Medicare, of course, it's 20%.

There's a significant out-of-pocket cost, and so we need to find a way to deal with that scenario. None of the manufacturers of the MCED test suggest that you skip routine, recommended cancer screening; they recommend that it be done in addition.

We try to jump right to these blood tests without doing the kind of studies that we required of every other cancer screening test. Thus far, they haven't done that great. People are skeptical. The United Kingdom was going to be a leader in this. They were going to do a big national study, and they signed a contract to do the Galleri test on a lot of people. After 1 year, the results were so disappointing that they were talking about terminating it.

So, we think we're going to do it in the US. The National Cancer Institute (NCI) has talked about doing a big study; who knows what's going to happen now with the current leadership, but there was talk about doing a big US study. The problem is that the test manufacturers are already selling the test, and they're trying to get a consumer and even a congressional groundswell to just push that test out there.

I think that's a terrible idea. We just don't know how good the test is as a cancer screening test. We do have pretty good screening tests for 4 malignancies, 3 of them actually being very common malignancies and common causes of cancer death.

What we need the blood test to do is identify ovarian cancer or pancreas cancer because those tests, we believe, if they work, will lead to earlier diagnosis and improved outcomes. It would be great if they worked in lung cancer too because nobody wants to go get a CAT scan. Let's say we tackle the really bad guys, if we could prove the test works in those guys, we should do it.

Hennessy: That's our discussion today about cancer screening. I hope you've enjoyed it. Please come back and watch our other videos and we'll have more to come. Mike, have a great afternoon, and for Breaking Down Health Care, we'll see you again soon.