Dr. E. David Crawford spoke at the 17th annual Future Directions in Urology Symposium on Sunday, September 11, 2016  on “An Approach to Early Detection of Significant Cancers.”



Keywords: prostate cancer, biomarkers, biopsy, PSA, SelectMDx

How to cite: Crawford, E. David. “An Approach to Early Detection of Significant Cancers” Grand Rounds in Urology. September 11, 2016. Accessed May 2021. https://dev.grandroundsinurology.com/prostate-cancer-e-david-crawford-approach-to-early-detection-of-significant-cancers/.


An Approach to Early Detection of Significant Cancers


One of the real exciting parts of this meeting that’s evolved over the years has been this whole thing about genomic markers and other markers in prostate cancer, and we’re going to talk about probably one of the most important tumor markers in oncology and that’s PSA. A lot of people don’t agree with that, but it does–what other tumor do you have a marker that you can measure that gives you sort of the outcome of a treatment, particularly radical prostatectomy. We don’t have any early warning signs or recurrence of breast cancer, colorectal and lots of other ones. And it’s unfortunate that as you all well know there’s been a lot of controversy that I think it’s actually good, about the over diagnosis and overtreatment of prostate cancer.

So this is a really hot topic, I was asked at the AUA to give a state of the art talk on markers, so part of this is that. At that same time, in May, at the AUA, right after the AUA and the Wall St Journal, there was a discussion about prostate cancer test, and all these markers and what role they play and I thought that that was pretty well done. There’s a little hyping and things like that that went on. But my talk is going to focus on what I call PCM or prostate cancer markers.

In a couple of days, we’re going to hear about another important arena and that’s all these new drugs that have been approved for advanced prostate cancer. And I think these two arenas, the prostate cancer markers and these new drugs of prostate cancer are probably two of the most important things that have happened in this disease in the last decade. And so I’m going to focus on the first one.

As a way forward we have a lot of challenges in prostate cancer, and one is to improve the interpretation of PSA, to allow, and I’ll show you why in a minute, primary care physicians to enhance referral of the right patients to a urologist for evaluation of a prostate cancer that is at significant risk for the patient. So that’s changed a lot in the last couple of years, and we’ll go into that. We also need a test which will decrease unnecessary negative re-biopsies and enhance detection of important cancers. And third, we need something to risk stratify. And I would say if we had some of these things that present a few years ago when the U.S. Services Task Force, the hammer came down on prostate cancer we wouldn’t be in the mess we’re in right now.

So what are biomarkers? Everybody is familiar with those, they can come from tissue, blood, urine. Obviously blood and urine are the easy ones. There are markers that give you a sign of a normal or an abnormal process.

So Neal Shore and I put this talk together, parts of this, and we’ve done it at AUA in sectional meeting and other things but what we try to do is educate people on these biomarkers and where they fit, because there is a lot of confusion. The average urologist doesn’t understand how all these biomarkers work, and so basically we put an end to three buckets, three and a half buckets. And that’s who to biopsy, next one is who to re-biopsy, and the third one is who to treat. And then there’s a little one about who to give asumen radiation or hormone therapies to following a radical prostatectomy. So the big one right here is obviously PSA. We have phi, 4K. SelectMDx. We have who to re-biopsy, ConfirmMDx, PCA3, 4K Score in that. And then who to offer interventional therapy to, and basically that is Oncotype Dx and Prolaris right now. There’s some other players here. And then who to give post-op treatment to in high risk radical prostatectomy. So there are three buckets.

As we all know, the USPSTF has been following prostate cancer for over a decade. And not too long ago they said don’t screen men over the age of 72. Three years ago they came out at the AUA when it was in San Diego, and they said, we give this a D and we recommend that you don’t screen but if you do, you use shared decision making. This was supported by a number of organizations. And right away what happened? The AUA went after them. We’re going to get these guys impeached. We’re going to go to our Congress person. We’re going to our Senator. We’re going to get this USPSTF put out of business. We want to change this, this isn’t right, we made all this progress. And then when I give talks, I say, okay, Mr. Urologist out in the audience, do you know who the USPSTF is? And they go, no. And you go, well, maybe you should know who you’re going after before you go after them. Why did they come around? They came around because they were by an act of Congress and that may be bad by itself, but nonetheless in the 60s to advice family practice, internal medicine, guys on guidelines. What blood pressure do you treat? What’s cholesterol? Do you screen for this? The hot button was prostate cancer, and several other things, and also breast cancer screening, but what they did was they watched this for a while and then PLCO trial came out. We were one of the centers in PLCO trial. Andriole was the guy that was in charge of it for prostate so you can blame him for everything that went wrong.

And then we had several other things occur. But they said after looking at this, we think screening does well, and I said, how many of you… let me ask this group, how many of you have actually read Virginia Moyer’s paper? Okay, so that’s about a third of the people maybe. But if you go to a urological audience they don’t even know who Virginia Moyer is or they’ve never read the paper. But if you read it, they said, gee, I believe screening does some good, but it also does a lot of harm. And on balance, we don’t think you should do it. All right, so that’s where it sort of set the field.

I think the way forward is, we don’t need to dump PSA, we need to refine it. We need to increase the likelihood of the initial positive biopsy of a significant cancer it should say here.

MALE VOICE: Dave, I’m sorry, over here. I think it’s very important to mention that the primary care physicians took this to heart and they stopped doing PSAs.

CRAWFORD: No question. And they wanted to, there’s a whole bunch of reasons why, and I’ll tell you why in a minute. One of them is lawsuits. We need to stratify high versus low risk, and so where we are right now is not one size fits all. We use this thing precision medicine, a personalized medicine and all this other stuff we throw around, what does that all mean? But what it does mean is we should try to tailor something for people, one size does not fit all.

I asked this question, you probably all know the answer to this. What percent of all PSAs ordered in the United States are ordered by urologists? Let me, let’s see here. Miner, what percent are ordered by urologists?

MINER: Twenty-five.

CRAWFORD: Twenty-five, wrong. Six percent are ordered by urologists. And the urologists don’t even know that. They only order 6% of the PSAs in the United States. As a matter of fact, this is where most of the PSAs, 90% come from internal medicine, family practice, urologists are 6%, hem/onc are 1.3%.

So if you’re going to start out with the issue about PSA ordering and educate, who are you going to start with? The people that are ordering it. The family practice doctors. You need to help them. Here’s the problem, and Alan alluded to this, we’ve got to talk to them about PSA cutoff of 1.5, 2.5, 4. We have percent free PSA, we have complex PSA, we have PSA velocity, we have PSA density, we have PCA3, we have 4K, we have 5, we have select, and they go, oh my God. They need a simple message. They need a simple message. And otherwise these are the people they need some sort of blood pressure glucose, this is the cutoff we use. So we need to refine that a little bit and we need to identify who needs to be referred.

So from the PLCO trial, Andriole and I worked in this and I got to present it at the plenary session at ASCO 10 years ago from the PLCO trial. That tens and thousands of men, if your PSA was less than 1 when you entered the PLCO trial your PSA did not get above 4, or even 2.5 in five years in less than 1% of people. So that was–if you did that in the United States, you would save a billion dollars a year, just for screening. So I said, okay, we’ve got to go contemporary.

VESTAL: [Interposing] Say that again, that’s an important statistic. Say it again.

CRAWFORD: It’s nothing great. I mean if your PSA is less than 1 when you went to PLCO trial, your chance of having an abnormal PSA in the next five years was less than 1%.

VESTAL: That’s amazing statistic.

CRAWFORD: I know, and we reported that and it never went anywhere.

MALE VOICE: Hans Lioture and a lot of these other studies to look at PSA, early in life, if it’s very, very low, you don’t need to be screened very intensively because you’re 10 year, 15 year risk of death is very, very low.

CRAWFORD: PSA is like 0.4 of – – .

MALE VOICE: That’s based on the NCCN guidelines, right? I mean that’s—

CRAWFORD: [Interposing] Right.

MALE VOICE: So I’d say—

CRAWFORD: [Interposing] Anyway.

MALE VOICE: This preceded all of that by three or four years.

CRAWFORD: It did. And that Ian Thompson picked up on this and reported for 10 years. Our study did not, it only looked at five years at that point.


CRAWFORD: All right, so—

MALE VOICE: [Interposing] But David, wait. Here’s where I’m stuck. You sort of almost just like run through it. It’s an amazing statistic, and where we’ve gotten stuck partly is by saying, using only PSA for predicting the presence of cancer.

CRAWFORD: Well, if you listen to my talk in a minute, you’ll get it, okay?

MALE VOICE: And what that was is about the negative predictive value which is phenomenal.

CRAWFORD: Okay. So let’s look at this. This is the–so I’ve worked with the Henry Ford database, they have a great database. So what we did is I knew what a cutoff should be, and I said, let’s look at in a contemporary database, with a lot of African-Americans I’m sorry, 30%, what’s going to happen if your PSA is between 1 and 5 over five years and you’re not on a 5ARI. Which’s your relative risk of having prostate cancer? The idea was to have a cutoff to get family practices guys a way to say, yes or no. And so what we did, we had 21,000 men that fit all of our criteria, and if your PSA was less than 1.5, you had a 0.5% risk of being diagnosed with prostate cancer in that thing in five years. Whereas, if it’s 1.5 or 4, your relative risk went from 7 almost 8% up to 10%, if you were African-American. That’s the cutoff, okay. Nothing perfect, and we can’t argue with all… what about age. The average age in here was like 52. So that’s what it is. And if you looked at the area under the curve, 0.7, it’s pretty good for a cheap test.

Right, but PSA, levels of 1.5 and above doesn’t necessarily mean prostate cancer. It’s more commonly associated with BPH and that’s why people will get biopsied. And so this was an article that we did from the MTOPS trial, when I looked at the data when we were in that, and if your PSA was 1.5, you had, and you look at 1.5, you had a 35 gram prostate and if you looked at men in the MTOPS study, PSA went up and we had very good volumes with MRIs and other things here. There was a linear relationship between PSA and prostate volume.

And so if you have a larger prostate, you will see and this is Clause Roehrbom’s work, is that your prostate volume will double over 15 years from age 55 to 70, and in general a larger prostate gives you more symptoms. Not always but it does.

And so here we got another thing. Let’s embrace a big–put our arms around men’s health here with PSA, not just prostate cancer, but BPH and as a matter of fact, in the PLUS trial, in the MTOPS trial and everything else like that, where that was looked at, and it was also looked at in the PLCO trial, looking at PSAs and urinary issues and things like that.

Here is where I think this all fits in in a minute. Anyway this is a practical algorithm that Steve Cochran came out with. It came out in MTOPS if that if you have an enlarged prostate, you use a 5ARI plus an alpha-blocker.

This came out last year in Renal and Urology News. Prostate size predicts less risk. No kidding. We showed that a long time ago.

All right, okay here’s how, now this is where all the controversies are going to begin. So anyway, here’s how I think we ought to go forward. I think PSA should be treated like every other thing a family practice doctor gives. When it goes up and he says, Dr. Lugg, I’m going to check your blood pressure. Dr. Lugg, I’m going to check your cholesterol. Dr. Lugg, I’m going to check your blood sugar. He doesn’t get informed decision to do that, he doesn’t. He gets your weight and everything else. And then if something’s abnormal, they talk to you. So they get a blood sugar, they do that routinely, and the next thing they do if they don’t really treat you for diabetes, they do an H1A1c hemoglobin and other things like that.

I think PSA should be treated like other tests that they draw. Informed decision occurs after the test is abnormal. That’s not what the USPSTF says. Matt Rosenberg and I have worked on this and others, informed decision is not going to happen. Now you don’t have time, it’s been looked at, it’s been studied, it’s not going to happen. And if you want the data, I can give it to you later.

So what does that mean? Everybody goes, you can’t do that because the urologists are going to be overwhelmed with so many patients. Well, how many are they going to be overwhelmed with? That’s shooting from the hip. What it was is we looked at initially 150,000 men from Prostate Cancer Awareness week and now 500,000 men we’ve looked at that walk in the door with their first PSA, 70% are less than 1.5. There’s 30% would require discussion, and that’s where you go from there. It’s like 30% have hypertension, or things like that. And PSA elevation doesn’t necessarily mean prostate cancer, it’s a surrogate for prostate health.

This is sort of an algorithm we put together on this whole thing, on where the PCP is, urologists, where PCMs are, biopsies, and this just came out yesterday and we’re still working on it, so I’m going to go to my old one. But here’s what happens. A guy comes in, gets a PSA less than 1.5, routine, come back in five years. Greater than 1.5, then that’s 30% of people either family practice guy or urologist evaluate and they evaluate for BPH. They don’t necessarily do a biopsy. They might do some of these tests I’m going to talk about and decide whether to do a biopsy. All right, so this is the bucket we’re in right now. And we have four tests here that can help us in men that we think are going to do a biopsy. Might want to do a biopsy on.

All right, so the new paradigm is, we have this false positive rate leading to unnecessary biopsies. We find with Gleason 6 and we over treat non-lethal disease.

So the goal in 2016 is we, in my opinion, we want to eliminate most Gleason 6s and we want to eliminate active surveillance. What does that mean? Active surveillance costs money. It causes anxiety. Well you don’t want to find these, we want to get rid of it.

All right, so here are the four tests, and I don’t have time to go through all of them, and some of them are circling the drain unfortunately, and our PCA3 was just give up by Hologics, I guess that’s the company in San Diego, Promark, there’s other people that work on it. The phi test is I think is a good test, Beckman Coulter has not done a good job of marketing this and providing availability. We can’t get it periodically, but it is, it’s the Hopkins group, they’re significant-everybody’s done, we’ve done a lot of work with it, it’s a good test.

PCA3 is a urine test, it’s a good test. The problem with PCA3 and again we’ve done a lot of work with it too, is that it doesn’t discriminate low from high grade cancers. This was a study David Boswick and I did and published in 2012 with almost 2,000 men with PCA3 in their first biopsy. There’s a linear relationship between PCA3 and a positive biopsy, and it’s independent of volume of the prostate. And then you get a report Metamark is doing this now.

4K test is Hans Lilly’s work for years, Andrew Vikers and that group from Memorial put it together. It is four parts of PSA, the secret sauce is probably HK2, and then they threw in age, digital rectal exam and prior biopsy studies. So because we’re doing all these screening, I tell you, all of these tests I’ve done in myself. And last year when we did 4K, the person put in my information so that I had an abnormal digital rectal exam, and it came back I had a 40% chance having prostate cancer and I go oh my God, what’s going on, even my PSA was low. And I realized that when we sent back the correction, then my risk dropped down to like 4%. So what does that tell you about some of the things that go into it? At any rate, you get a result back like your relative risk of finding the cancer and also a significant cancer.

And Dr. Jetta’s group looked at this and this came out just recently. It’s sort of and we’ll talk about that with her, is that due to significant issues with assay validation and absence of clinic utility 4K testing is not reasonable, unnecessary and is not covered by Medicare. And I think these things are all important because other companies have taken things out of perspective and you’ve got to take them step by step by step.

All right, so another one is we want to talk about in risk stratification is a new one. It’s just a few months old that is called SelectMDx and what this is a test developed by Jack Schalken who developed that PCA3, and his goal was to identify a test that would predict a significant cancer, minus Gleason 7 and above. And so it helps risk stratify so it’s a personalized medicine of high risk patients versus low risk patients.

Here’s what I like about this test. And I’ve also had this done to myself. If you have a negative test, you have a negative predictive value for a Gleason 8, 9, or 10 in 99.6% and a 98% negative predictive value for a Gleason 7. So that’s actually pretty good I think. It’s powered on negative predictive value. And then it also gives you relative risk of having a significant cancer and a non-significant cancer. So it’s gone through. This is in the early stages, it’s gone through a couple of testing and it needs to be looked at but at any rate, if you look at individual things, area under the curve, for PCA3 in the prostate cancer calculator, it’s better. And if you look at Gleason’s discriminating no cancer from 6s and 7s, it really does discriminate the 6s from 7. But one of the, none of these tests are perfect. None of them are 99 plus percent chance that there were 100% that’s going to make the diagnosis, the diagnosis and things like that. But this is easy because it’s a urine test.

And what you get back is it identified high risk aggressive cancers and when you get a report back, it gives you your chance of having any cancer, and you chance of having a high risk cancer. We’ve been doing a lot of these lately inpatients on active surveillance and things like that. I think it’s going to help eliminate the need to do a lot of follow-up biopsies and active surveillance, I really do from what I’m seeing so far and a number of other things. So, this is new and exciting and it’s sort of a non-invasive office space samples collection thing and it’s easy to do.

Now there are two genes, and we’ll talk about it, I think there’s going to be some presentations from the industry about this, so I’m not going to say anything more about it. And people when I go out, when I do these molecular markers, well, we’re testing 985 genes. Well, sometimes too many is too many. And if you’ve the hot buttons to press, that it’s sort of like the test is positive or negative, the pregnancy test and things like this, then it is somewhat better. So these are two genes. I think this is going to help do this.

This is a table that we have, we just published. We have an article coming out very soon, and let me see if I–it’s probably the hardest article I’ve ever worked on to get published in my whole career. And Dr. Gomella is on it, right? We have Dr. Carol, Dr. Coperberg, Alan Parton, Dr. Klein, Eric Klein, we have some folks from MD Anderson involved, we have one of your colleagues, and a few other people, and also we had Andrew Vikers on it. And so what it basically is, is this PSA 1.5 and then the way forward and then we had this table in here that looks at the company, the specimen, the cost, phi, 4K, PCA3 and SelectMDx, so this is part of it. But this will be out and available very shortly.

Along with this algorithm. I’m finishing up here right now. This is what I think happens in my world, that I would like to see happen. Patient comes in, routine, PSA, lab, less than 1.5, green light comes back in five years. If the PSA is greater than 1.5 or 4, they’re fine. Somebody’s PSA comes back 10 or 8, most people will say, I think we need to evaluate this and probably do a biopsy and I don’t see anything wrong with that. And then if your PSA is less than 1.5, then we practice guys can order some of these tests or refer to a urologist, and then they do this thing like phi, SelectMDx, whatever they’re going to do. Low risk, back into routine. High risk, these are the people that should get biopsy. If they get biopsy, a biopsy could be positive or negative and Lenny’s going to talk about that. If it’s negative, ConfirmMDx, if that’s negative, back into routine. And then if it’s positive, this is where I think multiparametric MRI comes in. If it’s Gleason pattern 4 and above, treat it in the healthy person. If it’s Gleason 6, he’ll talk about genomic markers in a minute. And active surveillance and so forth.

I think the markers have, and this whole thing is going to help us. I have not drank the MRI Kool-Aid I’m sorry, I know we’re having talks about MRI here. I think there’s a long way to go with this. This was something that I took pictures of it at an EAU meeting, 400 men undergoing MRI, then radical prostatectomy. There were 101 that had totally negative MRIs, and look at the biology that existed here. If you look at the next one, PT3, nodes, Gleason pattern 4, 4 or 5 pattern and all these things, so MRI is okay but it’s not perfect.

This is an article everybody is familiar with, breast cancer screening, the incidence of metastases has been unfaltering in the last… since 1975 we saw this happen with prostate cancer. The incidence of metastatic disease went down. And I think that is an argument about the value of screening but the cost was a lot of people were over diagnosed and over treated.

And so everybody shows this, don’t throw the baby away with the bathwater.

I’m going to end. There’s a whole bunch of things that we better do, so screening, we need to be smarter.