20 Staging

Introduction to Staging

We're gonna start the morning off with a discussion about staging.

Our talk is on staging, and it's between myself and Andy Rosecrans.

We divided it in half.

Current Clinical Practice

The current clinical practice for staging involves use of part tables.

These are predictive tools that combine the stage, clinical information to predict the stage of cancer and the predict the risk of extra prostatic extension, but do not provide information regarding localization or extent of extra prostatic extension.

And obviously those two things are important for treatment planning.

Past Performance in Staging

How have we done though, in the past?

Not so well because as you can see, this is a study, quite old study 2002.

And this was a meta-analysis that just showed all over the place.

In fact, the sensitivity and specificity was quite low.

Obviously we want a high specificity and sensitivity reading.

We wanna be up here, but we're not.

Recent Advancements

Things have changed since then with a lot of the functional imaging and on board.

And several recent studies have found very high sensitivity and specificity for preoperative MR in evaluation of extra prostatic extension, and high resolution MR has demonstrated high accuracy between T two versus T three disease.

However, as a group as a whole, even on three Tesla, we have not, this is yet to be proven that we're as good.

These have several single institution studies have shown very good results.

Protocol for Staging

In terms of protocol, we wanna do local, regional survey and to assess prostate cancer in that, is it confined to the gland or is it extending outside of the gland?

My preference is to use Endorectal coil in this.

And a lot of studies have actually shown that it's useful for staging.

And in terms of 1.5 Tesla or three Tesla, you can do either, but I prefer three T because as it is fairly challenging to see very small amount of extra prostatic extension as you go and lower and lower Tesla scans that the signal you want, as best signal as you can get.

So my preference is three Tesla, and you want an optimized diffusion weighted sequence and a dynamic contrast sequence with temporal resolution of less than 10 seconds would be optimal.

So again, to keep in mind for staging, you do want indirect coil.

And so I'd shown a few these images yesterday.

And this is an example of the med rat indirect coil that is all vendors have it.

It's a flexible, expandable, and disposable one.

Hologic has a rigid and rectal coil also.

And you do want, if you place the un rectal coils, it's very important to be a little bit more, you want it to be exactly straight horizontally and make sure you spend some time seeing that the placement is correct.

And then the coils are originally designed to be filled with air.

But however, if we use diffusion being a EPI type sequence, you're gonna get a lot of distortion.

And so it's best to fill that with some fluid.

Per fluorocarbon.

It's something I use, but barium works as well.

So anything that will reduce the susceptibility artifact.

And because, and when you do fill, you wanna fill it enough, so you want about 40 to 60 ccs, because if you don't, just having a coil can actually be more of a problem than not having a coil.

If it's not well distended, you're gonna get all this artifact and really we're evaluating this portion of the gland and that's being distorted here and when that translates to very poor a DC maps.

And so we certainly wanna avoid that.

Even when it's well distended.

This is a these coils are dual channel.

There's a small amount of air anyway, so you get some distortion, but certainly if you're meticulous and distend, well then this the posterior portion of the land can be better seen.

This is an example given to me by Raj Gupta from Duke.

And this is a coil that was filled with barium and the sal is just not superior enough to cover the seminal vesicles.

And then you can place it again where it shows nice horizontally and then it covers the seminal vesicles.

This is using barium, so you can see, you can fill it up quite well with that.

Staging Categories

So in terms of staging, we are primarily looking between T two and T three disease and t T two being confined within the gland, either on one side or both sides, and T three going out extra prostatic or going above into the seminal vesicles and beyond.

One thing to keep in mind is the T four is that urethral sphincter involvement, because often when we think of extracapsular extension, we're primarily looking posteriorly or posterior laterally, but this could be very central and especially important to look at the apex very carefully, where the lesion, if it's anterior and in volume, the u the urethra.

And if you go further down, this could really involve a urethral sphincter.

And certainly if they were to operate on this patient, then the urethral sphincter would be the urinary continence would be compromised.

So something to keep in mind that not all extra prostatic extension has to be outside.

That's something that I talk a lot with my fellows and residents because they're concerned more about this area.

Patient Preparation

What about patient prep?

Ideally, eight to 12 weeks after biopsy because for staging, a lot of these already have known cancer and they wanna know how extensive the tumor is.

Avoid sexual activity for up to three days prior, and this is to keep our seminal vesicles distended.

I don't really have any data on that, but that's what is said.

And I do follow that light meal and avoid caffeine.

I do find that caffeine causes a lot of peristalsis in the bowel and fleet mima prior exam.

I do use these this prep.

Overt Extracapsular Extension

So something to remind ourselves is this, there's obviously this overt extra capsular, extra prostatic extension, and then there's this very minimal.

So if you when you start working and have a multidisciplinary meetings, et cetera, it's important to stress to the urologist, and you'll be surprised when you discuss this, is that they are not aware of that that they think that we're not good at any staging.

So gross measurable tumor extending outside the prostate.

We could have seen that on any T two weighted image.

We don't need a lot of the stuff we're doing now for that, but they don't know that.

We have very high specificity of that on mr.

It's uncommon in era of PSA screening, although we have seen a comeback of that because of the midway where there are a lot of people not using PSA at all.

And it's independent predictor of clinically significantly worse outcome compared to match controls.

And these are some examples of overt extra capsular extension where really we don't need any functional sequences, but of course this could be hemorrhage.

You wanna see T one to make sure this is not hemorrhage, but this is gross tumor extension that's involving it's irregular capsule.

This is a PI on T two PY rats, five, there's extra capsular.

And here's another tumor that's extending.

This is involving the neurovascular bundles on the right side.

This is extending out, and this is prostatic and semial vesco.

This is just extending grossly out of the gland.

This is not where we have the trouble.

Subtle Extracapsular Extension

It's really the the more the subtle extra prosthetic extension.

And so then we look at these secondary findings of it.

And some of those are described at the bulge, the capsular irregularity, the broad capsular contact greater than a centimeter or a centimeter.

And these can be fairly challenging, especially in the less than one millimeter range.

And that's where we have the suboptimal sensitivity and specificity.

So I think an important factor is to have this discussion with your clin or referring physicians.

They understand that what we are talking about is that it's even the pathologists have debate over when they call very minimal extra capsular extension.

And here's example of a small dark, so markedly hyper intense nodule that is fally bulging right there as well as a little larger nodule.

And these would be even though the size is small, these would be this would be four on the new py rads capsular bulge.

This is fairly large and this is dark on a DC, right on the high B 2000.

And this was established extra prostatic extension.

This is extending to where the neurovascular bundles lie.

So these little dots are the neurovascular bundles posterior laterally.

And here's an example of another tumor that has a broad capsular contact.

So that's another sign that if you see a lesion that's about centimeter or so along the capsule that's suggestive of extra extension.

And here's an anterior extra ca prosthetic extension because this is where this is also important to talk to the your urologist.

They're gonna do surgery to say that there is extra cap extension because this is easily, they can do wide margin here and resect everything out because they're not worried about neurovascular bundles, et cetera.

Far anteriorly, and this is another example of nice example of where the T four disease happens.

And this was a large exophytic tumor that extended below the urethra the apex and there was large involving the sphincter.

So important to remember T four disease even centrally, because if it goes further inferiorly, so reporting of extra prosthetic extension is important.

For each lesion, it's important to say if it's involves the extra if it's extra prostatic or not as it will impact the surgical planning and the risk for positive margins.

Because traditionally what we have said, well you don't if unless you're a hundred percent sure it's extra prostatic, don't call, don't over call it because the patient may not get a definitive prostatectomy because of that, not necessarily so anymore.

And I think that's where good communication is important, because they can do nerve sparing surgeries very easily.

And so if one side is involved and not the other, it's okay to call the extra prostatic then as long as you have good communication, they understand that this patient is not undergo prostatectomy because of what she called.

So now I could suggest the sensitivity of extra prostatic extension.

Currently, most of this is microscopic and does not preclude surgery.

And when there's no suspicion, nerve sparing approach can is preferred.

And if you suspect extra prostatic extension, they can do a wider margin with to lower the positive margin risk.

And here's an example of a subtle left neurovascular bundle involvement where see the tumor that's extending outside the capsule, and here it's irregular, the capsule.

And on DCE any functional sequence would've shown this, but this was operated on and the they did a right nerve sparing approach on this case.

Limitations: Hemorrhage

Now, there are some limitations, and one of the ones that I really would want to stress on is hemorrhage.

In this case, you can see this was a 62-year-old man who had Gleason seven that was diagnosed on the Trus biopsy, and this was six weeks ago.

So even though I have this rule about not doing scans eight to 12 weeks after the biopsy, patients always sneak in and the call the office calls and say they really wanted, they're having surgery tomorrow, et cetera.

So we always end up having some of these.

And often I'll just go and say we need to repeat it when I do the T one, but in this case they were being operated on and so we just completed the scan.

So here you see the hemorrhage.

And then on the on the T two weighted image, it's hard to tell.

I mean, you see irregular capsule on this site as well on this site, but there's so much hemorrhage, how do you know the capsules involved or not?

And so it's really hard.

So we go on to do the functional sequences, and on diffusion you say, see this large mass and it's extending out, but on the right side, can't really tell.

You see something here, but not quite sure.

And I spoke with the urologist and I said I really can't give you a state.

I mean, I'm a little worried, but I can't really tell because there's so much hemorrhage distorting it.

I said, ideally we would wanna wait and repeat this.

And he said, well, no, we're already operating.

And so they went on and this guy actually had nerve sparing on the right.

He did.

And they were negative margins.

So this was lucky.

And this would go along with the hemorrhage exclusion sign where there's a lot in the literature where the tumor will hemorrhage doesn't show up on the side of the tumors, et cetera.

But I would have to say that I have found opposite much more common.

And so looking at this case, and I specifically left the the the dates from the examination because I didn't want you to think I changed the dates or anything.

So this is right from the scanner where this guy who at least in seven tumor and in December, 2012, he this is gross hemorrhage, and this guy, again snuck in four weeks after biopsy.

'cause he just had to know where how much tumor he had.

And T two is just you can see this wet shaped low T two signal.

And if you were just to look at the PY rads on this, you would call it really two because, or three you could say.

But really nothing more than that.

So if we did diffusion, nothing, this is just very distorted.

So I spoke, actually, I spoke with the patient.

I said, really, if you really wanted to get it staged, I'm unable to stage it.

There's so much hemorrhage.

And he was fine with that.

He said, I'll come back.

He said, I really need to get to know.

And so he was willing to come back and he came.

And for one of the problems that we have with this post biopsy hemorrhages is it decreases the a DC value in benign tissue and can increase the image distortion due to susceptibility artifact.

And so that's why ideally we wanna wait longer.

So he waited and he came back in February of 2013.

Waited a long time, but it's kind of funny, when he came back, he you still see some high density and I was still a little concerned that there was, but at this point he's this long enough time and it really shouldn't affect the functional sequences.

Although in the T two, it's really mindly high point intense.

So this is not the typical markedly hypo intense lesion that we would like to see for something that's Gleason seven.

But diffusion now you can see the large mass.

And so hemorrhage, just to remember it can help, sometimes there is a hemorrhage exclusion shine, but I don't know why it is, it may be the stage of hemorrhage, et cetera.

But certainly I've seen both sides where hemorrhage may or may not involve it or oftentimes a diffusion is so distorted you can't read it.

False Positives for EPE

What about some false positives for EPE?

This was a lesion that was considered high suspicion bulging on markedly hyperintensity T two and this had no extra prostatic extension.

So those are they're those two.

Another thing to remember is that the central zone, and Kasha showed some nice images of that yesterday where it's low T two posteriorly and it's high above the peripheral zone.

And sometimes when you get these scans from outside, the red is gross, this is RADS five lesion.

But really that's the central zone.

And just keep that in mind that that's rare.

First of all, it's rare to have tumors in the central zone, and when it do occur, it's usually extension from the peripheral zone.

And if you go up and down the you know, it's fine, the peripheral zone.

So to keep in mind, that could be a potential pitfall.

Another false positive for EPE, this was there was no extra prostatic extension.

So you'll have those cases where you call, but as long as your surgeon realizes that they're not gonna give definitive treatment to a patient, it's fine.

If you you will have those misses.

Changes in Interpretation for Extracapsular Extension

And at that time, I will switch with Andy.

So I would wanna emphasize the point.

I think the way we read or interpret for extra capsular extension has changed as radiologists.

I think given changes in the surgical management from a specificity to a sensitivity approach, I don't think is much the case now that our calling EC would determine whether or not a patient has or undergoes a prostatectomy.

I find all the time that we could raise suspicion and the patient will still undergo that the surgery rather, our interpretation is influencing the surgical mar the decisions regarding neuro sparing or not, and the margins at the time of a prostatectomy.

So it's important for us to raise the concern when we have it on based on imaging so the surgeon can be alerted to take a wider margin.

Impact of MRI on Surgical Planning

So to kind of to this end, this is I think a from a nice study from UCLA looking at the impact of MRI findings on surgical planning for prostatectomy.

And they recorded the surgeon's initial plan regarding nervous resection prior to Mr data.

And then the plans after reviewing the MR findings, they looked at cases in which it the MR changed their intentions about the resecting the neurovascular bundle and then compared with the final outcomes after prostatectomy in terms of presence of extra cap extension and positive margins.

And they report about 27% of the time was there a change in plan.

And in no case when they changed to the nurse hearing approach based on MRI, was there an ipsilateral positive margin.

So this was favorable for the patient.

So just some cases to this point, this was a patient with least in three plus four tumor.

PSA is almost 10.

And the point to make here is that that it's a side specific decision that the surgeon can on one side have a tight margin on the other side, go wider, and we can help direct that with MRI, which might not always be so clear cut based on say PSA and biopsy data otherwise.

So on MRI, there was a five outta five lesion on the right with extra pathetic extension on the left, no no clear lesion or EPE.

So the patient did a had had a radical prostatectomy now with a wide wide margin nerve sacrifice unilaterally on the right, but preserved on the left.

This patient had a high volume bilateral tumor on a biopsy on MRI.

We had a five out of five on the left with no lesion on the right, and the urologist did a nerve excision on the left, but a partial nerve sparing on the right in consideration of the bi of the biopsy, and then what the MRI showed and negative margins bilaterally.

This patient had a Gleason six with PSA of 6.7.

Patient was electing for radical prostatectomy, MRI, we called a three out of five on the left.

This sm small lesion on T two and a DC, the capsule was intact, no visualized DPE.

So this this supported a narrow margin with nerve sparing surgery bilaterally and negative margins.

Improving Staging Nomograms with MR

So there there's been so here we showed data from one study.

There's been a number of similar studies looking at the ability of MR findings to improve staging nomograms, essentially trying to combine data such as PPSA biopsy findings, and now MR findings for predicting extra relic extension on on pathologic findings and showing significant improvements when including the MR parameters.

So Saana had a made a point about this earlier, and I'll show some cases.

There's been a number in the literature showing improved staging using DCE and diffusion.

And to a large extent, I don't think this is that we're directly visualizing the ep itself on the functional sequences.

Rather, I think we can better localize the dominant lesion with diffusion DC and then knowing where the dominant lesion is, we can then better assess for ECE with higher sensitivity and specificity.

So for instance, this was one study where they had multiple readers experienced an inexperienced reader, and each reader had a significant improvement in staging compared with compared to T two alone when combining T two with DCE this study was looking at the role of diffusion for staging and different parameters for predicting the presence of TIC extension and compared with standard clinical or biopsy parameters a DC here he for 77%, so performed well in terms of some cases this patient we see in the left posterior peripheral zone area of low T two.

And then again dark and a DC and bright on the high B image.

So this the very low a DC along the capsule.

And then on prostate there's established EP at the left apex.

This patient on T two heterogeneous prostate, a lot of non-specific areas of decreased signal.

There is a region on the right, but not all that different from other areas.

But then that that pans out as that dominant lesion on a DC and the high B image with again the very low A DCA lot up along the capsule.

It just gonna direct our attention in that region, maybe raise our concern in that area.

And on prostatectomy established ep from that lesion.

Stratifying EPE: Focal vs. Established

So I I I think I've been kinda using these terms a sta so the pathologist, they'll they'll stratify EP and the language they'll use will be focal or established.

And I think that's an important point that there's EP and then there's EPEI mean, it's not all the same.

So if so some there's been some more recent studies kind of stratifying our sensitivity by the extent of it.

And when it's established EPR sensitivity is very high.

There.

There's gonna be certain cases we're just gonna continually miss where it's a fraction of a millimeter.

I mean, there there's and you just kind of have to get used to that.

I mean, we've had sometimes our urologist a while back would say, well, why did you miss this case?

Or What happened?

Why did you say no eep?

And then there was, and I mean, you could you could really go cra crazy with this and we'd go back the slides and and and sit there with the and see what we missed.

And there'd be like a few malignant glands out in the fat right outside the prostate, and just some very microscopic focus.

And some of this you're just not going to see.

So so this was a slide that was shown earlier, but I mean, there's a report of of focal EPE.

And again, when we we reviewed the case and this was, it was about half a millimeter, and it's just gonna be tough to reliably call that call some of these cases based on MRI.

Seminal Vesicle Invasion

So moving along to higher stages of some of the vesical invasion, this typically it'll appears a mass like area of decreased T two signal with continuous tumor in the prostate base.

There's gonna be causes of false negatives here typically from microscopic extension of tumor along the e*******e duct without any without forming a discreet mass, as well as false positives from scar or fibrosis within the seminal vesicles.

This some will often manifest as a on T two A images with the diffusion, I think we can improve upon that and raise our specificity.

This was actually from a recent study radiology from the group at Chicago that had nicely shown for two independent readers, improved accuracy for some ation combining T two with diffusion weighted imaging.

So so in this case here we see some focal low T two signal on the base of the sum vesicle, again, with the with the decreased A DC the sum vesical invasion.

It will also in addition to the low A DC, we can see here the increased signal on the high B image, which can be helpful as well.

Multiplanar images can be sometimes be helpful for tricky or equivocal cases.

Here by the right base, we see some decreased T two, and then this was the the dominant lesion in conjunction with the diffusion in DCE.

And with the sagittal and coronal images were helpful for better appreciation.

It def that it did in fact extend above the base and into the seminal vesicle.

And this was confirmed surgically.

I think one thing to be aware of is that there there there actually is a small intra prostatic segment of the seminal vesicle and the urologist I mean the the pathologist won't classify it as T three B.

They they won't really deem the vascular invasion if it's involvement of just the intra prostatic segment of it.

So we we've had some cases where we were really concerned on Mr but then on prostatectomy, it did not it was not interpreted as vascular invasion by the pathologist, and we've compared it, it was it it was, this was kind that's kind of what was going on.

So in this case, here we see a dominant lesion of the left base, a rads five.

It's over the 1.5 centimeters.

And if you look at the higher up of the axial image than the coronal is, it's intimately budding the base of the seminal vesicle, but still confined to the prostate.

And prostatectomy there there was gross CPE, but no seminal vesicle invasion.

Again, this is kind of a tricky case.

This was a false negative, the extent I was discussing describing previously on pathology.

They described diffuse bilateral seminal vesicle invasion.

On MRI, we see the preserved kind of ovulate some vesical architecture with high a DC throughout bilaterally early.

This point has come up a number of times so far in this course.

Again, there isn't really data to support this, but many practices do it that we we give instruction to the patient to avoid e*********n for two to three days prior to the exam to avoid having a collapsed somal vesicle that would be difficult to assess for tumor.

So again, kind of a comparison here, you know what you're trying to avoid is these decompressed on the vesicles, and we want these to be nice and distended to better assess for tumor.

Lymph Node Assessment

So moving along, so lymph nodes, so I mean we've traditionally relied on size for identification of suspicious lymph nodes, but as we know, is in really anywhere in any organ, this will have false positives and false negatives.

So we have to combine it with other features.

So looking for, is there a round or ovary shape, lack of a fatty hilum heterogeneity, poorly circum margins of the node.

So so these would these would be example of metastatic nodes these kind of bulky rounded nodes.

No no hilum with them.

That hilum is one comparison with more typical reactive nodes here, kind of symmetric and elliptical shape in short axis, not only measuring a sm several millimeters.

There's been some literature on using diffusion weighted image imaging to improve our accuracy for metastatic retinopathy in prostate cancer.

There was a lot of attention very recently to this study from the group in Switzerland.

It was in radiology.

They showed improved sensitivity for small metastatic lymph nodes.

They they looked in prostate and bladder cancer using diffusion.

It was actually very interesting.

They they took patients who were other otherwise N zero.

They were looking really at just at small nodes of patients who based on anatomic imaging alone, no no suspicion would've been raised.

And they showed they actually had a a pretty good, a very good sample size.

And showed showed higher detection looking primarily at high B images to pick up to pick up more nodes and correlating back with the T two weighted images.

And here another case of of a small lymph node on T two weighted image bright on the high B image.

And in this case the node is dark on a DC i I think we have to be careful, and this point was kind of made in their paper, I mean just about any node of some increased signal on diffusion weighted images.

So I mean, benign nodes will will see this commonly just kinda these symmetric external or these symmetric iliac chain or some of these public sidewall nodes.

Again, they kind of have the a a flattened shape to them.

Maybe you can make out kinda a notch contour.

So so we have to be careful kinda high sensitivity with the fusion in the paper I was just citing the nodes were correlated back to T two and only considered suspicious if that some atypical morphology on t on T two in terms of their shape or presence of a fatty hilum.

The the diffusion was largely used for initial detection or depreciate the nodes visually, but then it was wasn't necessarily called if it was then considered not suspicious on other sequences.

The they did make the point in that paper of looking for nodes that were markedly hyper intense or brighter on the high B images than other lymph nodes in that patient.

So an example here, we just see nodes that are very very bright relative to other areas, and that was one of the features they described.

And another point that was made in that paper was it was rely generally interpreting based off the high B images, and that they actually found substantial overlap in a DC values between the benign and malignant nodes.

And most sometimes the the this was a this was a histologically proven metastatic node from a patient with prostate cancer.

And sometimes the a DC will I show an earlier example where it had reduced a DC, but but this isn't as reliable and sometimes they will be will have higher a DC values.

And then it's common to include in the protocol when staging a larger field of use sequence to assess for proximal lymph nodes not included in the high resolution images through the prostate and seminal vesicles.

Higher Stages: Bladder and Rectal Invasion

So again, kind of moving along to higher stages, bladder invasion kinda a similar theme.

The diffusion weighted images in DC can be helpful for this as well.

Here we see this irregular thickening of the bladder wall along the right lateral aspect on T two, on the post contrast image marked confluent increased enhancement.

So this was bladder invasion, this patient on we see this thickening by the posterior base of the bladder.

Some asymmetric increased signal on the high B diffusion image.

This patient here on T two probably wouldn't raise concern for bladder invasion.

But then on the a DC and high B image, we see a focal area restricted diffusion extending into the left posterior bladder wall.

So again just kinda a similar theme of using the functional sequences and finishing up in terms of rectal invasion.

I think this is another area where we've come across some pitfalls or some some over over calls where it can also maybe raise concern for rectal agent based on just broad contact or very intimate abutment of a of tumor with the the rectum.

And we've had some case over these, then go to the OR and come out come out very easily.

And then there's no involvement and really wanna look for a more infiltrative and irregular nodular involvement and not just close contact to make this call.

And this was a patient who did have rectal invasion.

Bone Metastasis

And finally for bone metastasis this is again kind of a similar theme as previously with with functional sequences.

This was a prostate cancer patient with widespread bone bone mets that here we're better appreciating on the diffusion weighted image than on po on the post contrast sequence.

Conclusion

So, in conclusion just some a few takeaway points from this.

There's can be a diagnostic challenge in diagnosing focal Protic extension.

Although we are reliable for gross ECE diffusion weighted imaging can help in determine the presence of eec ECE some ation and nodal metastases.

And just have a structured approach to this so that you consistently and reliably communicate your level of suspicion for all of these findings to your urologists.

Thanks.