2 Prostate Cancer- Why MRI?

Introduction to Speaker and Topic

Peter Choki is the head of the imaging section and molecular imaging program and program director at the Center for Cancer Research at the NCI.

Peter is really gonna introduce the whole topic of prostate cancer, and his talk is called Prostate Cancer: Why MRI?

Today I'll be talking about the case for using MRI in prostate cancer, and my talk is divided into two sections, basically, first, talking about why we should even bother with prostate cancer, its importance, and then discussing evolving concepts that are occurring in the field of prostate cancer that make the introduction of MRI so important, including things that are going on with screening, the rise of active surveillance, the concept of multifocal disease versus dominant lesion, and the use of MRI from staging to diagnosis to treatment.

Importance of Prostate Cancer

First, a few numbers.

Prostate cancer is a most frequent diagnosed cancer in the United States, 233,000 new diagnoses, almost 30,000 deaths per year from prostate cancer.

One in seven men diagnosed with PCA during their life, and one in 36 will die.

The average age of diagnosis 66.

But perhaps the most interesting of these figures are the last two, that at its peak in the 1990s, the death rate was 40,000, 40 per hundred thousand.

Whereas today it's at 25 per hundred thousand, a substantial decrease, hearkening back to the days of the early 1930s, when prostate cancer was very poorly understood.

We're definitely making significant progress in this disease in terms of mortality, and that is very good news.

As you look at the various death rate curves from our major cancers, you can see lung and bronchial cancers decreasing markedly prostate cancer, also decreasing markedly along with colon cancer, although some of the less common diseases and cancers are more steady.

There's progress that has been made, and of course, various factors and various factions are claiming credit for this decline.

We'll look at this in a little bit more detail.

What is Prostate Cancer?

Let's go back to the beginning. What is prostate cancer?

As you know, the prostate is a glandular organ.

It has glandular epithelium and something a genetic abnormality or mutation causes proliferation of glands.

We have an initiating event, and we now know, at least in many men, that an initial event is called the emus gene, and that initiates proliferation of cancer.

Most of what we're gonna talk today about is this localized cancer, but we should keep in mind that even while we think of prostate cancer as localized, it is shedding cells into the blood, so-called circulating tumor cells.

Most of these die the vast majority, but some go on to form metastases.

And even with margin negative cancers at the time of surgery, the patient can recur.

They usually recur in the castrate sensitive that is sensitive to androgen deprivation therapy, but eventually, almost all cancers will become androgen independent, and they're called castrate resistant prostate cancer.

And that is a lethal form of cancer for which there is currently no cure, although there are many new drugs.

Gleason Scoring System

So how do we judge a prostate cancer?

Today we use a Gleason scoring system that has was first described by Dr. Gleason from a VA hospital in the mid 1960s.

He proposed a five point scoring system, one to five well organized cancers to very undifferentiated cancers.

Now, I will say that we have truncated this Gleason scoring system by eliminating for all practical purposes, ones and twos from use pathologists simply don't use this very often.

And so we're left with this three, four, and five to decide how a malignant a tumor is.

Most people believe that the Gleason three type is a very indolent type of tumor, but as that there's a transition to four, the tumor becomes more aggressive, and certainly by five, it is very aggressive.

Now, the genetic mutations underlying this are being worked out today, and there's a lot of evidence that we can now differentiate between threes and fours that is benign or indolent tumors and more aggressive tumors using panels of mutation analysis.

And these are now commercially available in the marketplace and are being increasingly used.

So, Gleason score for all its flaws and its antiquity are none, is nonetheless pretty predictive of outcome.

So if the patient has a Gleason score of six, which is basically a three plus three, that is no elements of Gleason four, the recurrence free probability is very, very low, and the patients do very well.

If the Gleason score is seven, which obviously combines some elements of three and four, it's an intermediate, and if it's greater than if it's all four or greater, then there's a significant recurrence rate and subsequent death rate.

So you can really, for all it's faults, Gleason is a pretty successful way of looking at prostate cancers.

This is what pathologists are looking at.

The Gleason three basically are these well-formed glands that are fairly uniform, and you can see that they contain a lot of fluid, which will account for their relatively higher signal on T two weighted scans, for instance, and maybe they're higher diffusion on diffusion weighted scans.

And as we look at a Gleason four, we sort of lose those cystic spaces because there's collapse of the glands and it's more difficult to differentiate clear glands.

And then of course, there are mixtures of both three and four.

So this would be an area of three, this would be an area of four.

And so this will highly influence our imaging as we move forward.

Clinically Significant Prostate Cancer

A concept that has evolved that's extremely important is the idea of clinically significant prostate cancer.

We've spoken about Gleason three plus three being fairly indolent disease, three plus four, four plus four, four plus five being clinically significant.

So prostate cancer that is likely to progress and affect a man's life is dependent on the amount of four. Gleason four in the specimen.

Larger volume disease is also likely to be more clinically significant, and a commonly used threshold is 0.5 milliliters or ccs.

Now, in all these discussions about whether something is clinically significant or not, one has to consider the life expectancy of the patient because the expected time course of prostate cancer is measured in tens of years.

And so if the patient has only a five year or three year prognosis from some other disease that they have, for instance, hypertension or diabetes, or even another cancer, something that is quote clinically significant, becomes less important in that it is unlikely to affect their life.

Trends in Prostate Cancer Management Over the Last 15 Years

I want to talk about some trends that have occurred in prostate cancer between in the last 15 years.

For sure, prostate cancer has not changed that much in the last 15 years, but the way we look at it has really changed.

And so I want to take you through some timelines of what's occurred.

Now, I'm sure you're all familiar with public service announcements that came out in the early two thousands that you should go out and get your PSA, it'll save your life and get your PSA at health fairs, at hospitals, whatever.

And of course, it was a huge moneymaker for those who did it, but you don't hear much about that anymore.

And there's a good reason because we quickly became aware that we were overtreating too many men by getting PSAs and then doing random biopsies.

We were detecting a lot of Gleason three disease.

And eventually this came to the attention of authorities such as the US Preventive task Force, and they assigned the PSA score of D, which has actual significance because in as the Affordable Care Act rolls out, this agency has the ability to determine reimbursement for tests that that is not in effect today, but it could be in the near future.

So as a result, what we're seeing is sort of a decrease in the amount of interest in PSA screening, although it's still widely done, and prostate cancer continues to march on.

So there's some kind of concern that maybe by just withdrawing all kinds of screening, we're now going to be confronted with patients presenting with spinal metastases and need for emergency radiation therapy, which is clearly not where we should be going.

Now, part of the problem with PSA is that it only applies to a small number of cancers.

Many of the cancers are very indolent, so they will never actually affect the patient's life.

So it's hard to show an impact from PSA screening on a disease that won't affect your life.

On the other hand, there are some cancers about 5% in the population that are so aggressive, they violate all the rules that we've just talked about, and they proceed to metastatic disease and death in a time period that makes annual PSA screening irrelevant.

So it's really only this middle course of the trajectory of disease, so-called tumor C here that potentially could benefit from early detection with A PSA screening technique.

Part of the problem with PSAI, I think in some ways we have to look at the PSA issue as not just a problem with the PSA number.

PSA is just absolutely a number, but it another big part of the problem is what it caused, which is a random truss biopsy.

And so there are two major problems with the random truss biopsy.

One is that significant lesions could be outside the normal template.

So these biopsies are tend to be done in a very routine way.

The urologist puts the needle in in a template, and if the tumor happens to be off the template, then it's not going to be seen.

And significant cancers probably 15 to 20 or 20% of cancers can be missed in that way.

Another, but probably even more significant problem is that because it is random, these small inconsequential Gleason three plus three cancers can be detected in 5% or 10% of individual cores.

So we have the problem of both under-diagnosis with trust biopsy and over-diagnosis.

So the US PSTF considered a two major studies when they decided to give P-S-A-A-D screen D degrade.

One was the PLCO trial in the United States, and this showed actually no reduction in the amount of prostate cancer.

But it PLCO has been highly criticized because it was performed in the United States where PSA is very available, even though the patient may have said look I wasn't screened, chances are they got a PSA somewhere along the line, and that contaminated the two groups.

So you couldn't see a big difference between the two groups in Europe, particularly in Sweden, where PSA it's a government controlled healthcare system.

PSA was off the screen.

You couldn't order PSA, you could see a profound a significant decrease in the death rate from PSA screening and different other countries where the rate of PSA screening varied, had a modest but definite effects on mortality.

So anything to the left of the line represents an improvement in mortality.

Anything to the right represents a decrement in mortality.

But so there were many studies, as you can see from these two trials that showed some benefit immortality from PSA screening.

But at the end of the day, the task force believed that this was a very modest effect, and that the quality of life issues far outweighed the these small benefits.

So in their opinion the the quality of life was too decreased.

This is important, and you may see it again on a question.

So we've covered that slide.

Multifocal Disease

I want to talk about a multifocal disease.

Multifocal disease is a characteristic of prostate cancer.

These cancer these prostates don't harbor one cancer.

They harbor many. And it turns out that there's this huge reservoir of cancer.

Even if you look at this classic study of 20 year olds, 30 year olds, 40 year olds who are who I think this was from Detroit, who unfortunately met untimely ends, they were autopsied and a very high percentage demonstrated low grade cancers, even in their thirties and forties.

So there is this huge reservoir of disease, which leads me to the second point, which is that Gleason six we know has a good prognosis, but how do you know it's really a six?

And that was the way we started the decade of two thousands.

And but people sort of pushed back from that.

They said okay look we know that this is a low grade disease.

Could we try active surveillance?

Active Surveillance

And the experience with active surveillance was this, that men went on, they got second PSAs, which went up, and their biopsies were higher grade.

And people said look I can't live with this.

I have to have treatment.

And the vast majority of people in the early days of active surveillance went on to definitive treatment.

So it really wasn't working.

And there were too many conversions to active treatment.

So that's a very important trend that's been happening, because now with the advent of MRI, we can see a role for active surveillance in patients with low grade disease and monitoring them non-invasively in that way.

So I'd like to also talk about the problem of upgrading.

And this is really what concerned so many physicians when they were counseling patients about active surveillance.

It turns out if you compare biopsy and surgery, that there's a very significant upgrading, probably 30 to 40% upgrading for to seven from six.

So there isn't a significant upgrading at for at between the the sevens less than sevens, but but at seven you can see an upgrading from three plus four to four plus three, very commonly 30 to 40%.

And at higher grades there's really not that much upgrading.

So this led to doubt in people's mind is your six really a six?

And it led to more active treatment.

Dominant Lesion Concept

So the idea of multifocal disease is also very important in in the beginning the of the decade.

The urology community was more on the page that it's mo multifocal.

So the whole gland really needs to be treated.

But I remember distinctly a change in philosophy that occurred in which people began to realize yeah may be multifocal, but many of those other tumors are very low grade.

And really we should be concerned about the so-called dominant lesion that is the largest lesion with the highest grade.

And that should really guide what we're how we're going to treat patients.

And it even led to the I idea that maybe even focal therapies were a valid way of treating dominant lesions.

And that of course requires good imaging.

So that leads us again to the imaging door.

Now multifocality is a real issue in a patient like this.

These are serial sections of the same patient that you can see tumors bilaterally, significant tumors in terms of size.

They're often not clonal.

That is this tumor has nothing this tumor has nothing to do with this tumor in terms of its genetic mutations.

They're independent actors, and they can behave independently.

Of course when they're large like this probably the whole gland does need to be treated.

But what about a patient like this with some very tiny Gleason sixes scattered about and a dominant lesion in the periphery of the gland?

Isn't this at least in theory a good candidate for a focal therapy?

Well take out this whole gland incur the consequences of incontinence and potentially impotence for this lesion that could be ablated by cryotherapy or laser therapy.

Evolution of MRI in Prostate Cancer

So MRI is certainly a game changer.

It plays into all the changes in philosophy and mythology that have grown up around prostate cancer.

So we have we ourselves have undergone a great change in the last 15 years.

I think in the beginning of the decade we were focused on staging prostate cancer.

That is Endorectal Coil MRI high resolution scans to tell whether there was extra capsular extension or not.

And it was maybe useful in surgical planning.

Many people didn't believe in it, but that's where it lay.

But as the concept of really needing to discover where the cancers are how where they where they were how large they were, and whether you could direct needles into them created the idea that multiplanar MRI including diffusion weighted I imaging could show these dominant lesions.

And I will say in the mid two thousands the advent of good diffusion weighted imaging was a huge game changer for MRI.

By the end of that decade we were wondering whether in recal coils were really needed, whether DCE was really needed and whether we could use a much simpler even screening examination.

And then in the last few years the development of multiple vendors who are offering MRI ultrasound fusion biopsies.

So not only can you see it but you can do something about it by directing needles into it.

So this led to the this idea of the multi-parametric MRI as potentially even a screening tech technique.

Now Mr. Spectroscopy which was part of the package in the beginning is no longer done very much.

It's technically difficult, and it's pretty much off the chart.

D-C-E-M-R-I is decreasing in importance.

And as you'll see in the PY rads classification that will be discussed in more detail it's it has a decreasing role as we as we roll forward.

So even MRI has been evolving, but it nonetheless is capable of showing a tumor that was missed on four prior negative TRUS biopsies and can can easily be biopsied under guidance.

So how critical is the Endorectal coil and D-C-E-M-R-I?

Well increasingly you can see the Endorectal coil patient here and the same patient on a follow-up scan without the Endorectal coil.

Certainly the anterior lesion is very well visualized on both of these and correlates very well with the pathology.

Biopsy Techniques

So let's talk just a few thing words about in gantry biopsy.

We're gonna go into into much more detail as we go on.

In the beginning the approach was this which was to put a probe into the rectum in the MRR gantry and then insert needles through that probe.

And of course there were disadvantages to this approach.

There was time consuming. It used up a lot of mr time.

And so the concept that you could fuse MRI to ultrasound, MRI providing the anatomic information, ultrasound providing the real time information and the fusion enabling you to lock these two images together created big opportunities.

And this will be discussed in more detail.

There are various approaches to link the two images once they're fused, but all of these techniques require fusion of the the two measures.

So you have the Euro nav that does this by a GPS type method.

You have Artemis that uses an articulated arm and cos or Euros station that uses imaging itself.

But regardless of the technique that you use, the value of targeted biopsy becomes very clear.

If you look at the 12 core biopsy which is depicted on the top, it showed a three plus three lesion, 5% of that core.

The targeted lesions in the same patient certainly picked up the apical lesion, but it showed that it was a much higher grade higher volume tumor than the the standard biopsy was able to.

So we commonly see this this issue.

Now Jeff mentioned that we still are not perfect with MRI, and so I just want to mention a few consequences of using targeted biopsy that some are favorable, but some point out challenges for the future.

So if you compare the 12 core for clinically significant tumor upgrading at surgery, so this is clinically significant, this is nonclinical significant upgrading.

There was about a 30% increase in our in our 12 core population in the targeted population that was greatly reduced.

So now it wasn't reduced to zero.

We still have some upgrading, so we're not perfect.

And to some extent you may view the upgrading of non clinically significant upgrading as significant as well.

But certainly even this small number which is about five to 7% now requires more work.

But nonetheless we have integrated MRI now into virtually every facet of the workup and management of prostate cancer patients from patients who arrive with a positive biopsy who are basically undergoing staging to those who have never undergone biopsy before.

So so-called biopsy naive patients to patients who have received negative prior routine biopsies for whom the PSA continues to rise.

We're using it increasingly in active surveillance patients and then to monitor patients after their treatment for recurrent disease.

And this will all be talked about in great detail.

Summary: Why MRI?

So to summarize, why is MRI needed the first harkins to the myth of PSA screening saves lives.

It's quite clear that the the problem is not necessarily PSA alone, but the PSA combined with a blind biopsy is not a good combination.

It resulted in the discovery of too many inconsequential lesions.

There was minimal gain in survival.

There was a lot of loss of quality of life as a consequence of this overtreatment.

But PSA combined with MRI for imaging can direct biopsies to MR targets with increasing detection of clinically significant disease, at the same time not detecting as many non clinically significant lesions.

And I believe that if we had chosen a different threshold for PSA and had integrated MRI earlier in earlier than we have, that the the the way PSA would look today would be quite different.

It would be it would be much more of a winner.

And so this is a just a sample anterior lesion mist on routine biopsy but clearly seen on MRI.

So the second myth why and and why MRI is needed, all prostate cancer must be treated.

It's clear now not all cancers need to be treated, but this 30% conversion from two active treatment from active surveillance due to the inaccuracy of blind biopsy led to what's known as active surveillance fatigue.

And people jumped on the active treatment bandwagon and then had to deal with the consequences of that.

But MRI allows the identification of significant abnormalities, targeted biopsies, and continued follow up of documented lesions.

So here's a good example.

A patient in 2010 with a PSA of 3.1, and a targeted biopsy of that showed a three plus four lesion is very clearly seen, is really on the borderline of what you would want to put on active surveillance.

But the patient strongly want it to be on active surveillance.

A year later the PSA is not very high much higher.

And you might say well you're doing fine, except when you do the biopsy there's an increase in the amount of Gleason three plus four and more significantly.

Look at the volume change of this lesion over time.

This is an active growing lesion.

And so this patient went on to active treatment.

So MRI is a game changer in this respect.

We recognize that now that prostate cancer is multifocal, but dominant lesions do count.

MRI doesn't have the sensitivity for low grade tumors.

So we're not even needing that.

And we're picking up dominant lesions enabling focal therapy to be performed.

So here's just an example of focal therapy.

Vision for the Future

And so finally I just want to give my vision for the future that while PSA is on the decline, there's nothing on the scene to replace it.

And the advent of MRI I think will rejuvenate Mr. PSA because I don't think PSA is entirely at fault in this problem.

But the goal is clearly the identification of significant cancers.

And there are a number of markers that are in the on the scene PCA three genetic testing involving various mutations.

And I believe Mr. Screening with Mr. Guided biopsy will be a reasonable approach for the future.

With that I'd like to thank you for your attention.