Sentinel Node Detection in Breast Malignancy - HD

Sentinel Node Detection in Breast Malignancy

My brief was to talk about sentinel node detection in breast malignancy. My interest is that I'm a screening radiologist and I'm the director of breast screening here in South London.

We know that as breast radiologists, we are key to making the diagnosis of early breast cancer by using traditional imaging techniques such as mammography and ultrasound, and using gray scale ultrasound particularly to guide us in our biopsy technique.

Once we either know that there's a biopsy proven cancer, or that our suspicions are so high, we would naturally do an ultrasound of the axilla so that we can, with imaging locally, stage that tumor. And if we find an abnormal node, and that's the current guidelines here in the United Kingdom, we would proceed to a biopsy at that time, either with a fine needle aspirate or with a wide board needle depending on access and safety.

Now, if that biopsy comes back as a negative but adequate sample IE we know we've got lymphoid cells in the pathology report, but there are no malignant cells that patient will go on to have a sentinel node biopsy. If however, we have picked up positive metastatic disease in the node, they would be automatically triaged to an auxiliary lymph node de dissection or an auxiliary clearance as it's often called.

The problem for us as images is that if we have sent a patient down the arm of sentinel node biopsy, if one of those one to four nodes that are usually taken out by the surgeons at the time of surgery comes back positive, that lady needs to go back to theater, have a second general anesthetic and a second operation and proceed to a completion auxiliary node clearance. And that's where all of the discussions around imaging and finding this sentinel node is centered around.

And why do we care about the sentinel node versus auxiliary clearances? Because lymphedema is one of the major complications for patients who have had an auxiliary node clearance. And as you can see, the incidence of having lymphedema after a sentinel node is a quarter to a half that of having an auxiliary node clearance. And that's why it's important.

And since the almanac trial in Europe, in the United Kingdom, we've been doing sentinel node biopsy as a standard of care in imaging normal axilla since about the mid 2000. And this is what we're trying to avoid. The ladies who come back to us and as hard as it is it is this, that they struggle with it is that lymphedema. It is the mobility, the pain, the stiffness that is what keeps them awake at night rather than the psychology as you wouldn't as you would you may think from the surgery.

Ultrasound Staging of the Axilla

So we've been doing ultrasound staging for quite some time and when we look at the gray scale imaging and use the morphological criteria of lack of fatty hilar cortical thickness, and the jury's out to where we draw the line of where is abnormal and where is normal as far as cortical thickness is concerned.

If we find that it's abnormal, then we would stick a needle in it and get some kind of biopsy. And if we have an ultrasound and a biopsy, we have a 50% chance of picking up auxiliary nodal metastasis. I would say nothing to write home about, but we are when we put a needle in that auxiliary node very specific. And we have very a rare false positives about when we do find that there isn't anything there.

So using ultrasound and biopsy as a triage for auxiliary nodes status is important and that avoids a second operation and an unnecessary sentinel node biopsy for all breast radiologists who have to do lectures who saw me is a goddess because she does fantastic metal meta-analysis of very good papers. And she has written two meta-analysis about auxiliary staging, one in 2011 and even better, one in 2014. And that is what she came out with. We are highly specific but not so sensitive when we're looking.

The other thing to bear in mind is even with a lot of experience of using ultrasound as we are in Europe, probably more so than in America, to be fair, we don't see that lymph node on gray scale in imaging in up to 40%. And that comes from Ali Sever's work down in Maidstone and Peter Britton's work in Cambridge.

When you look at all of the stu a lot of the studies and these are very early studies and admittedly, albeit there you can see it reflects that same mantra. Our sensitivity is nothing but mediocre I would say. And our specificity is good. And I would draw you to the fact that it isn't all just FNAs that is doing that. We have got some studies with core biopsy and that's some of the discussion that we will draw out a little bit later on.

Contrast Enhanced Ultrasound for Sentinel Node Detection

So contrast enhanced ultrasound finding the sentel node uses this anatomical knowledge. So we have the epidermis up here and in that very superficial dermal layer, we have these very small lymph capillaries that are sort of pre collectors and it is very superficial under the skin. And they then drain into larger superficial lymph collecting vessels in the subcutaneous layer.

So notice that in the deep layer we're too deep. So if we start imaging down here or injecting in the wrong layer and going too deep, which is not, it's not that difficult to do to be honest, we're going to miss the actual anatomy we were trying to image. And we will have poor success with the technique.

We know that the anatomy and the lymphatic drainage of the breast is well documented from the time we were all first year medical students. And it's this sub subcutaneous superficial plexus that we're really using in this technique.

So Ali Sever's work is clearly the seminal work that started all of our interests off and started all of this work off. And just to summarize what he used in the first instance was local anesthetic to the subcutaneous layer. That area in the perio, that skin in the peri areola region is very sensitive. So that's a really important first thing.

I've highlighted two mills because we're using it at slightly stronger strengths than if we were using it. Vascular, we only use two mills to reconstitute the powder and then we take a 0.2 to 0.5 mil aliquot in a diabetics one mil syringe with a fine 26 gauge needle and pop it into the dermal layer. You have to see a skin bleb and some of the early failures and every group that does it found that that's what we did wrong the first time we went into the subcutaneous layer too quickly. We have to raise the skin blow. We can use up to a maximum of three injections and the lymphatics are pretty quickly visualized.

To be fair, if you don't see a lymphatic within this sort of 15 to 45 second window, then you can go back to that skin bleb and massage it and you often then can pick up the lymphatic and start scanning from the peri alar skin in the upper out quad up to the axilla of the breast with the breast cancer in it.

Having dual screen is clearly very nice because you can follow the contrast images as well as looking at the gray scale, which is your comfort zone 'cause that's what you know that you look at normally. And then once you found the node, you know what you are going to target for your biopsy.

So Ali's paper at first paper was a hundred and 36 patients and he had a technical success of 93%. That is probably the highest technical success rate of all the papers and of all the anecdotal evidence that we have from the from my colleagues in other screening programs in the country, and Professor Cross Grove, professor Lim and I had the pleasure of going down to maiden and meeting Ally and he's showing us his unit and his technique. And he did say to us that when he first started this study, he was so fastidious about trying to get po you know, positive results, that sometimes he would spend 45 minutes doing this and massaging and massaging, massaging. So that and that is a credit to him to be fair. And you can see again very few false negatives in his series.

Early Work and Examples

So I'm just gonna show you some pictures of our early work and then I'll go through what we found with our study that we did. And I'm crossing toes and fingers to know that this works.

So here we have the skin. So we are looking, when we are here in this very superficial layer, that's where we'll see the first parts of the lymphatics. If we're looking down here, we're too low and here we are, we're seeing something. And then about two thirds of the way out towards the axilla, we'll suddenly see it dip down. And that's a really reassuring sign 'cause then you know you're gonna find your lymph node. So you're following it and you're following it and then you are gonna suddenly see it dip down. And then you know that you're gonna find your node and there it goes. And the loop's not quite long enough. So we're gonna go to the next slide and press again.

So we're coming down and here we have the lymphatic making the node look very nice and positive. And because we've got that, we now know where to put our needle on the gray scale and which is really helpful.

So one more, more, I'm just gonna go to the next one. We've another case this time. It just shows actually how tortuous and long-winded these lymphatics can be and how patient you need to be. So again, we are looking in the superficial layers here we've got two, we're diving down and we're getting quite excited 'cause we're getting near the node and then we're going off to the right and then no, we're not, we're gonna come back 'cause we're gonna still keep following it. And we're coming and we're coming and there we are and we've got the blush and we now know where our lymph node is on gray scale.

And I hope you can see how difficult it can be to see lymph nodes. It is easy to see a replaced metastatic lymph node, but we don't need contrast for that. We need it for the sentinel node triaging so that we get the right patients to the right operation quickly.

So that's what we did. So once we've done that, we then try and target the correct node so we can see that, we can see a lymph node here and it would be easy to just head for that. But if you then look the darker, I'm finding it really difficult on here, we can see a darker lymph node and that's the one that's been identified with the contrast enhance. And that's the one that we target. That's the one we wanna know what the cells are like and we use whatever we think is safe either FNA or core biopsy.

I think we're all becoming much more comfortable using core biopsy. But frankly when that auxiliary neurovascular bundle is looking at me and pulsating and I've got two centimeter throat on a on a 16 or a 14 gauge even I start getting a few ations. So that's where some of our kind of, I suppose our learning and our bravery come needs to come from. 'cause we need to push this forward.

Feasibility Study Results

So I'm just gonna go through some of our early experience of doing it. And this was just as a feasibility. So this was just after Ali's paper had been published and we were all thinking, gosh, is this the way forward and is this possible or is this just something that the Great Ali Saver could do? And so it was a feasibility study.

We asked patients or we recruited patients from the breast multidisciplinary team meeting who had a biopsy proven breast cancer who then had a normal ultrasound. So they were always gonna go to a sentinel node. They were then referred and in the department we had a Toshiba. We did the same technique as Ali, but the only thing that we did do is we used slightly more. 'cause we diluted it slightly more than we we would today.

And when we first started it was exciting. We did find the lymphatics, but we had an awful lot of breakthrough and it was really difficult with the breast, with the number of interfaces in that subcutaneous layer to know where that superficial initial lymphatic was. And so there was quite a lot of work with the app specialist to be able to use it. And actually the best thing that we did was to change probe from those high frequency linear probes that we use in breast imaging to a linear f uh intermediate frequency vascular probe. And I think that that is still the case.

So we had an 80% success rate and when we had a little straw poll of screening units in England, that was the figure that everyone else had. So we take our hats off to Ali to be fair. We had 48 patients recruited over 20 months and we had 10 failures for a variety of reasons as you can see there. But I think it is worth pointing out that four of those 10, almost half was in that first period where we were still learning what to do and what was normal and where to look and how to refine our technique.

We had a nor we had a pretty standard distribution of cancers within that group from tumor size from five to 50 millimeters and 36 cases, we were able to correlate with the final surgical histology. And this was what we got. And I think that this was the standout figure that we had. We had six patients that were that we called negative and that were false. And so this gave us a sensitivity of 45% and a specificity of 96%. And it's those false negatives that has always been the criteria that holds us back as far as the MDM and buy-in from the surgeons and the oncologists.

Micro-Metastases and Guideline Changes

Then we had Juliana's paper, the ZZ 11 trial that started to talk about micro metastases and and I-T-S-I-T-C cells, which are micro metastases are metastases within a lymph node that is basically less than two millimeters. That is the gauge of your needle to be fair. And isolated tumor cells are less than that. And all of our six false negative cases were micro metastases or ITCs.

So we then had quite a few years where actually it was controversial to use the contrast enhanced define the sentinel node because we've had evidence from New York saying actually AL clearance is really not needed. So if you find your micro metastasis, we're not gonna do anything about it. And then we had oncologists, radiotherapy oncologists to be fair, saying actually you are gonna miss the micro metastasis 'cause you're only doing FNA and then we might be undertreating the axilla. And that argument has been going on within the breast fraternity for quite some time.

So last year there was a multidisciplinary meeting in London and we tried to get a consensus statement between the surgeons, the radiologists and the oncologists. And as you can imagine, it was a very long day. And to be fair the statement was not the cla the clear concise erudite thing that we wanted it to be 'cause we can't always agree on anything. But what we did take away, and it is now published and it will be included in our guidelines certainly is that ITC and micro metastases need no further auxiliary treatment in addition to their breast conserving surgery and mastectomy.

So now we've got the literature starting to tell us what to do with the axi. So we've got Juliana saying actually in America, patients who have T one or T two tumors, which is fine 'cause we know that's small, but even if a sentinel node, if they have one or two positive sentinel nodes with no extra capsular spread, they are gonna do nothing further. They're not gonna go back and do a completion auxiliary dissection, which would be the standard of care here.

However the patient has to accept and complete whole breast radio radiotherapy. And where there was lots of controversy, and I will be really interested to hear what happens in Europe is in America the field for radiotherapy is slightly angulated. So it actually catches the auxiliary tail and the lower levels of the axilla in Britain that did not happen. And so we were going to miss that part of the of that important critical part of the breast. And so there was a concern about us undertreating the axilla, whereas in America those small metastases were being mopped at by the adjuvant therapy that the patient was gonna have anyway.

So now we have this conundrum of really do, does the radiologist even need to move their probe from their breast cancer up to the axilla? Do we need to do any of that? Can we save time in our already busy overrun clinics in symptomatic?

And fortunately Hasani has come to RA again and her more recent paper looking at the studies in the axilla shows that actually that decision to treat or to further treat the axilla after a positive sentinel node biopsy is based on how much tumor burden we think that there is in the axilla. And that can only come from radiology. And so hence we've now got another place to be. And another and another voice to be heard in the breast MDM and in her in her analysis there were seven paper seven studies which looked at the utility of ultrasound and biopsy. And if you had a positive outcome from that biopsy, you had a pretty good way of saying there is quite a high tumor burden in that axi. So you can triage those patients straight into auxiliary clearance. And if you had a negative biopsy, then that had a good clinical utility to make sure that those patients went into the sentinel node arm and they would get a good treatment pathway.

So if we took out our six cases from our small study our small feasibility study of micro metastases, you can see now that our false negatives gone up and all of our sensitivity and our other factors have gone up. So that's where I think in the last three and a half years, the staging of the axilla and the place for contrast enhanced ultrasound has changed. And Sani it does say it in her paper. Actually grayscale as we know is not that great, but anything that allows us to identify the sentinel node accurately and be able to tell the surgeons what the tumor burden in the auxiliary is and will allow us to go down that imaging algorithm that I went through earlier.

Unresolved Issues and Future Directions

So we've still got a few unresolved issues here. We've still got this. Do we do an FNA? Do we do a core biopsy? If we do an FNA, are we going to miss micrometastases? Well that doesn't matter anymore 'cause we now know that a two millimeter focus wouldn't get any further sentinel node.

We've now got very good ex at uh vacuum excisions of breast parenchymal lesions. So there was a natural step to say, could we go to the auxiliary and take out the auxiliary nodes with a vacuum? And that would really tell us the answer 'cause we'd pretty much get the whole node and the pathologist would be able to tell us from our nine gauge. Cause what we're we're talking about Now, Peter Britain's probably led the way here in the United Kingdom in Cambridge and as been has been doing a study with vacuum incision in the lymph nodes.

Most of the first cases were done in theater. So the utility of being able to use that technique in the outpatient clinic in which we all work every day is still out for the jury. But they do get good results and there's no doubt about it. However, they also, when there is a complication, it's a huge complication. So they've had patients taken back to theater that night to have the auxiliary artery tied off. And then unfortunately if that node is positive, they've then got to go back and do an auxiliary clearance in an axi that's had a hematoma a primary surgery and it made the whole thing really quite awful.

And so the breast surgeons are still at, you know, even in Cambridge, I'm still waiting to see how good that is and I'm not sure how many people are champing at the bit to put a nine gauge cutting needle into an axilla that's quite close to the neurovascular bundle in that in that way.

Conclusion

So in conclusion, I think contrast enhanced central node is clearly a feasible at detecting and biopsying and guiding biopsy. We had quite a lot of problems getting the imaging technique right, but when we did everyone was much more confident and I think that's getting better.

We are not having to reinvent the wheel as we go to different units. I think a multicenter trial is absolutely needed to validate not only the technique but what the clinical practice is as far as our guidelines are concerned. And the Royal College of Radiologists breast group has set up a MICROBUBBLE users group for exactly that purpose and we kind of encourage anyone who wants to take part.

So the units who have already done studies, so Imperial and Mastone leads Gates Head, which is near Newcastle and Manchester have put in their initial results. So at least we can do a joint paper on feasibility. And then from there we're going to get a study with a protocol together that we all use and then take that forward. 'cause I think that's what we need next.

And I suppose the last thing I just wanted to say is that remember the utility of a test is not just about the sensitivity and specificity, it's about how it affects patient management. It's about whether it's actually really useful when you are sitting in your breast MDMs talking about how you're going to make it better for that patient for both survival as well as morbidity and mortality. And I think that's where central contrast enhanced sentinel node, which kind of went off the boil if I'm absolutely honest in the last few years, is now coming back because now we know we're looking at estimating tumor burden and that's going to be our role as radiologists and I think that that's where it will grow.

I do wanna take this opportunity to thank both David and Adrian for all of the helps their slides and everything through this. And I thank you for your attention.