Contrast Enhanced Ultrasound in Pediatric Oncology: Focal Liver Lesions and Beyond - HD

Introduction

Hello, my name is Beth McCarville.

I'm a pediatric radiologist at St. Jude Children's Research Hospital in Memphis, Tennessee.

Today I'm gonna be talking about our experience using contrast enhanced ultrasound in pediatric oncology.

I do receive product support from GE Healthcare.

So today I am gonna be talking about the benefits of contrast enhanced ultrasound in children, and then how we've been using it in our pediatric oncology population to measure tumor blood flow and as a problem solving tool in pediatric oncology.

Benefits of Contrast Ultrasound in Children

So the benefits of contrast ultrasound in children, as I'm sure you are aware, are that children are the ideal population for using ultrasound in general because of their small body habitus that allows us to put the transducer very close to the region of interest, thus reducing imaging artifact.

Our ultrasound in general is very well tolerated and we don't need to sedate our patients to perform ultrasound with or without contrast.

It also has the benefit of being portable.

It can be taken to the ICU or to the operating room.

It's less expensive than other cross-sectional imaging modalities.

And most importantly, it does not involve ionizing radiation, which is an issue of considerable concern, especially in children and particularly in pediatric oncology because these patients undergo so many imaging examinations.

Improving Visualization of Tumor Margins, Invasion, and Adenopathy

So we've previously shown that contrast ultrasound can improve visualization of tumor margins, visualization of invasion and adenopathy.

In this study that was published in 2012, we compared ultrasound to CT contrast enhanced ct, both before giving ultrasound contrast material and after giving contrast material.

And we used a grading system to rate our ability to identify the tumor conspicuity margins, invasion and adenopathy.

And you can see that the scores for the post contrast imaging were slightly better than the pre contrast ultrasound imaging when compared to contrast ct.

Examples

So just a couple of examples.

This was a 17-year-old boy with a renal roid tumor with this very large pelvic recurrence seen nicely here on this contrast enhanced CT scan on the pre contrast ultrasound.

We can see the tumor here, which is hypoechoic relative to surrounding tissue.

And on the post contrast images, we can see that this tumor is diffusely enhancing.

In this example, this is a 6-year-old boy with a wils tumor shown here on ct.

You can see the tumor and you can very nicely see the margin between the tumor and the underlying normal kidney.

On the pre contrast enhanced ultrasound, we can identify that there is a tumor in the right kidney, but we don't see tumor margins.

On the post contrast imaging, we are still unable to identify the margin between the tumor and normal kidney.

So contrast ultrasound, contrast in this case did not help to identify tumor margins.

And more work is needed to know which tumors are best suited for contrast ultrasound in children.

Assessing Tumor Angiogenesis

More recently and more importantly, we've been using contrast ultrasound to assess tumor angiogenesis.

Angiogenesis or the development of new blood vessels is critical for tumor growth, survival, and metastasis and targeted anti-angiogenic therapy is being increasingly used in clinical cancer trials.

There are a variety of different anti-angiogenic agents that have different mechanisms of action, but these therapies are cytostatic that is they prevent tumors from growing, but they may not cause tumors to shrink, unlike the conventional cytotoxic chemotherapies, which cause when they're effective cause tumors to shrink.

So the conventional methods of assessing tumor response to therapy, which rely on a change in tumor size are not suitable for assessing anti-angiogenic therapy.

So there's a really a crucial need for functional and quantitative modalities that will allow us to assess tumor response to these anti-angiogenic therapies.

And there are a number of imaging modalities that are currently under investigation, including dynamic contrast enhanced ct, oxygen labeled water, PET CT dynamic contrast enhanced MRI and dynamic contrast enhanced ultrasound.

And of these modalities, contrast enhanced ultrasound is the most appealing for the pediatric population for the reasons I've already described.

Ultrasound Contrast Agents for Measuring Tumor Blood Flow

So ultrasound contrast agents are the ideal contrast agent for measuring tumor blood flow for a number of reasons, they are micro bubble contrast agents.

Here we have a contrast agent optisan shown in a Petri dish with blood, red blood cells in the background.

And you could see that these microspheres approximate the size of a red blood cell.

And because of this, they remain in the vascular space.

Unlike other imaging contrast agents that are very, very small and diffuse freely across the vascular membrane.

These microbubbles contain a gas that are highly reflective on ultrasound imaging, and therefore we can use very small doses, less than one ml.

And with newer contrast specific software, a now available from all vendors, we can quantitate the amount of contrast agent flow into a region of interest.

Pilot Study on Quantitative Dynamic Contrast Ultrasound

So we previously performed a pilot study to assess the value of quantitative dynamic contrast ultrasound in the assessment of tumor response in children and adolescents with recurrent or refractory solid tumors who are being treated at our institution on an institutional phase one trial of anti-angiogenic therapy.

So in this clinical trial, these patients were not expected to be cured, and the main endpoint was to see if this therapy could prolong survival or increase the time to progression of their tumor.

So our study was comprised of patients who were less than or equal to 21 years of age with a refractory or recurrent solid tumor who were enrolled on this phase one therapy.

And the contrast ultrasound component was optional and required a separate informed consent.

So to be eligible for contrast ultrasound, the patient needed to have a target lesion that was visible on non-contrast gray scale ultrasound.

They needed to undergo a complete history and physical and 12 lead EKG and echocardiography, and they needed to have no history of an ultrasound contrast allergy right to left cardiac shunt, pulmonary hypertension, and they had to have an oxygen saturation of greater than 92% on room air or no history of oxygen saturation, less than 92% on room air.

So we performed contrast ultrasound at the following time points at baseline before the therapy was initiated, and then three days and seven days after the initiation of the anti-angiogenic therapy, and then again at the end of course one which was 21 days, and at the end of course two, and then every other course until the patient progressed.

So this is an example of a target lesion.

This was a supraclavicular mass, a synovial sarcoma that was adjacent to the carotid artery here in the lower neck.

So in general, we chose the largest area of the visible target lesion to image.

We held the transducer in this single slice position throughout the examination.

And we recorded dynamic imaging for 60 seconds after the bolus injection of IV contrast material.

So this is what the tumor looked like using the contrast specific software.

The orange area is where contrast agent flowed and the blue areas are where there is no contrast agent or just background non enhancing tissue.

So using the vendors online software package, we were able to do region of interest analysis by drawing a region of interest slightly inside tumor margins to avoid including any normal adjacent tissue.

And then from this data, we created time intensity curves, and from the time intensity curves, we could evaluate six different parameters.

So this is the time intensity curve that was generated from that region of interest.

So on the X axis we have time up to 60 seconds, and on the Y axis we have signal intensity in decibels.

So the six parameters that we investigated were the peak enhancement, which was measured in decibels, the time to peak, which was measured in seconds, the rate of enhancement, which was measured in decibels per second, and then the area under the curve.

The area under the curve was divided into two sections, because most of our tumors reached the peak of enhancement by 10 seconds.

We measured the first 10 seconds after arrival of contrast material and called that the AUC one or the wash in.

And then we measured the second 10 seconds and called that the AUC two or the wash out.

And then we also looked at the total AUC, which was the AUC one plus the AUC two.

Analysis and Results

So we then looked at the association between each of these parameters and time to progression.

First of all, we just looked at the baseline values themselves to see if they predicted which patients would respond and which would not.

And then we looked at the change between the baseline value and the value at each of the imaging time points.

We also looked at the percent change in parameters at each time point.

And for this analysis, we separated our cohort into two groups, non-responders, which we defined as those patients who progressed at or less than 42 days, which would've been the end of course two or patients who progressed at greater than 42 days.

So we had 13 subjects who underwent a total of 74 ultrasound examinations.

They had a variety of primary tumor types as shown here.

Target lesion sites included the liver, the pleura, and then a variety of other soft tissue sites.

By Cox regression analysis, there were three parameters that were significantly predictive of the time to progression.

And these were the peak enhancement, the rate of enhancement and the wash in all at the end of course one were associated with time to progression such that patients who had greater decreases between baseline and the end of course one of any one of these parameters also had a longer time to progression.

Additionally, there were a number of parameters that trended towards significance.

Most importantly, the peak enhancement, again, was trending towards significance as early as day three.

And again at day seven.

When we separated our patients into responders and non-responders and looked at the percent changes.

We did not have enough patients in each group to do a statistical analysis, but you can see just by looking that the responders had much greater decreases in general in all of the parameters at all of the time points when compared to the non-responders.

And again, this was most substantial for the peak enhancement, which was four to six times greater in the responders than in the non-responders.

The decrease in the peak enhancement was four to six times greater in responders compared to non-responders.

Examples

So just a couple of examples.

This is a 21 month old girl who had a recurrent roid tumor with this large peritoneal implant, you can see nicely on ct.

So this is this patient's contrast ultrasound at baseline, and again at day seven at baseline, this is the tumor this round mass adjacent to the tip of the liver.

You can see it enhances very intensely.

It enhanced 33.5 decibels.

And then at day seven, the patient jumps around a little but then settles down and here's the tumor, here's the tip of the liver.

And you can see it's still enhancing quite intensely.

So the peak enhancement on day seven was 30.6 decibel, and this was only an 8.7% reduction in the peak enhancement between baseline and day seven.

And this patient had a very short time to progression of only 22 days.

In contrast, this is a 16-year-old girl who I showed you earlier with a large synovial sarcoma in the base of the neck surrounding the carotid artery.

This was an inoperable tumor at baseline.

Here is the tumor, you're gonna see the carotid artery enhanced first followed by the tumor, which is going to enhance very intensely.

And then on day seven you'll see a dramatic difference.

Here's the tumor, here's the carotid artery.

Almost no enhancement at all at day seven.

So this went from 28.9 decibels to two decibels, and that was a 93% reduction.

And this patient had a very long time to progression of 242 days.

Limitations

So our study did have some limitations.

We had a small sample size and a variety of tumor types.

Also, we only looked at one target lesion, and these patients had numerous metastatic lesions and it might be more accurate to look at more than one target lesion to get a better assessment of tumor burden and tumor response.

And also due to limitations in current software, we were able to only image one slice of a tumor and it might be better and more accurate to look at the entire tumor volume.

Conclusion on Tumor Angiogenesis Assessment

So in conclusion, we determine that quantitative contrast ultrasound does detect changes in tumor blood flow and can detect them very early in therapy.

The peak enhancement, the rate of enhancement in wash in at the end of course one all predicted time to progression in our study.

The peak enhancement appeared to be the most robust parameter and our approach avoids radiation or the need for sedation, but our findings do need to be validated in larger clinical trials.

Contrast Enhanced Ultrasound as a Problem Solving Tool

I'll move next to how we've been using contrast enhanced ultrasound as a problem solving tool in our patient population.

I'll be discussing assessing tumor residual tumor versus postoperative change assessment of malignant fusions, assessing thrombus versus stasis of blood flow, and then finite focal liver lesions.

Assessing Residual Tumor vs Postoperative Change

So this is an example of a boy who had a right adrenal neuroblastoma shown here preoperatively this large well defined mass with internal calcification.

And on a post-op follow-up CT had, again, a very large rounded, well-defined mass in that area of surgery.

And there was concern that this was a residual tumor.

We performed an ultrasound.

You can see that on the gray scale ultrasound.

This has sort of a solid periphery and has some septations internally with color Doppler evaluation.

It didn't appear to have internal blood flow, but we used contrast enhanced ultrasound to confirm that there was no internal blood flow and therefore we felt it was consistent with the hematoma.

And on follow-up imaging, you can see that this thing did shrink down and was probably just a post-op hematoma.

Assessment of Malignant Effusions

This is a 16-year-old boy who had metastatic osteosarcoma.

This is the compressed right lung with all these calcified metastatic deposits within it.

And this very large malignant pleural effusion around it, which you can see is ated, but you can't really appreciate septations within it.

The patient did have a pigtail pleural catheter in place, shown here in the lateral lower right pleural space, which drained initially and then stopped draining.

So they wanted to know where this pleural catheter was in relationship to their pleural effusion.

So we did just gray scale ultrasound of the pleural effusion.

And you can see that the collapse lung here on this longitudinal image is surrounded by a very large effusion, but you can't really appreciate septations within this fluid collection or the relationship of the tip of the catheter to the fluid collection.

So we installed a small amount of contrast material through the patient's pigtail catheter, and you can see that the contrast agent pooled in this very small pocket ated pocket within this very large pleural effusion.

So they treated the patient with TPA, they drained off 750 ml of fluid.

And we've repeated the ultrasound the next day.

And here you can see that the lung has re expanded.

And when we instilled the contrast through the pleural tube, it layered out in the anterior pleural space.

Assessing Thrombus vs Stasis of Blood Flow

This was a 9-year-old girl who had a pad beo occlusive disease as a complication of bone marrow transplant.

In these patients, the flow in the main portal vein becomes slower and slower and decreases and diminishes, while flow through the hepatic artery increases in response to the venal occlusive disease.

So this was this patient's main portal vein on August 4th.

It was widely patent, had good flow, it was in the right direction.

And then two days later when we repeated the examination, there was no flow within the portal vein.

We were not able to identify a signal from the vessel and we were not able to identify any flow within it with color doppler evaluation.

So the question was, is this a portal vein thrombus or is this just stasis of blood flow within the portal vein?

And this is a very important question for this patient because if this is thrombus, this patient was going to require intravascular TPA administration directly into the portal vein.

And there were problems with performing intravascular TPA because this patient was already a bleeding risk due to thrombocytopenia caused by her bone marrow transplant.

And she also had a underlying coagulopathy.

It was imperative to confirm whether this was a thrombus, but the patient also had renal insufficiency and therefore doing computed tomography angiography was contraindicated and the patient was in the ICU and also not a good candidate for doing MR Angiography.

But she was an ideal candidate for contrasted enhanced ultrasound because we could take the machine to her if needed.

And there is no risk to the kidneys from ultrasound contrast materials.

So whether or not the patient has renal insufficiency is not a contraindication to using these contrast agents.

So this was her contrast ultrasound that we performed at the bedside.

And you'll see this is the gray scale image and the contrast image on the right shows good flow in the hepatic artery initially.

And then later on you can see reversal of flow in the main portal vein, which eventually completely fills in.

So we were able to confirm that this patient did not have thrombus.

She had reversal of flow in her main portal vein, and there was no need to do intravascular TPA.

Focal Liver Lesions

So probably the most common indication for using contrast ultrasound in our patient population is to evaluate liver lesions, focal liver lesions that are identified on surveillance imaging.

This is from a very nice publication from Seattle Children's Hospital where they looked at follow-up imaging in 273 solid tumor patients to determine the incidents of focal liver lesions in this population.

And they found that about 17% of patients developed focal liver lesions.

About a third of those were due to focal nodular hyperplasia and about 15% due to metastatic disease, while the others were a wide variety of other lesions.

Our main concern in these patients is trying to determine whether these new liver lesions are due to metastases from their solid tumor or if they're benign in nature.

You'll notice that of the focal nodular hyperplasia lesions, 86% of patients have multiple lesions, and about 60% of patients with metastatic disease also have multiple lesions.

So whether or not the lesion is multiple or solitary does not help you distinguish whether it's benign or malignant in this patient population.

So contrast and enhance ultrasound is extremely useful for discriminating benign from malignant focal liver lesions.

And the main feature that helps is on the delayed imaging where malignant lesions demonstrate washout on delayed imaging while benign lesions do not.

So just some examples from our patient population.

The phases of enhancement are shown here.

The arterial phase lasts about 10 to 35 seconds, followed by the portal venous phase, which lasts anywhere from 30 seconds to 120 seconds.

And then the late phase imaging needs to be carried out for two to up to five minutes with intermittent scanning during that time period.

So just some examples, this was a 13-year-old boy who was previously treated for stage four neuroblastoma, who presented to his local er with a small bowel obstruction and a contrast enhanced CT was performed, which revealed this fairly large and very enhancing lesion in the liver.

So the question was is it metastatic or is it benign?

So we performed contrast ultrasound and this was a focal nodular hyperplasia.

And this ultrasound nicely demonstrates the contrast ultrasound features of focal nodular hyperplasia and the arterial phase.

You have these central feeding vessels in sort of a spoke wheel pattern that with centrifugal filling in of the lesion and then during the portal venous phase, these lesions can have this light bulb appearance.

Here's the enhancing portal vein and the lesion on delayed phase imaging.

The lesion is iso enhancing.

So there was no washout of this lesion, and therefore this is not a malignant lesion.

This is a 7-year-old girl who was also treated for neuroblastoma at the age of 13 months on surveillance imaging was found to have this enhancing nodule in her liver.

Again, we need to distinguish between a benign lesion or a metastatic lesion.

Gray scale ultrasound.

We could see the lesion here very nicely.

It's hypoechoic, so it's not a hema angio.

It had a little bit of internal blood flow.

On contrast ultrasound, we found that this was a regenerative nodule.

Regenerative nodules will be iso enhancing with surrounding normal liver on all phases.

So during the arterial phase, it's iso enhancing with normal liver portal venous phase, again, ISO enhancing and on the delayed phase ISO enhancing.

This was a 25-year-old girl who was treated for a ML at the age of 16 years and had a chest CT performed for some indication.

The chest CT was performed without IV contrast, but included the upper liver, and they noted all of these numer innumerable hypodense lesions scattered throughout the liver, some of which were poorly defined.

So there was concern for malignancy and we perform contrast ultrasound.

First of all, you need to make sure that the lesion is visible on gray scale.

We can see two lesions here that had pretty prominent internal blood flow by color doppler evaluation.

And these were adenomas on the early arterial phase.

Adenomas have more diffuse enhancement than do focal nodular hyperplasias.

They do not have that central feeding vessel or the spoke wheel appearance on the late arterial phases usually light up like a light bulb and are extremely hypervascular lesions.

Then on the more delayed phases, they become iso enhancing with surrounding normal liver.

Shown here in the portal venous phase and also in the delayed phase ISO enhancing no washout.

So this is not a malignant lesion.

This was a 20-year-old woman who was diagnosed with a gastric carcinoid at 18, at 14 years of age, who was currently symptomatic with flushing with cor carcinoid syndrome and concern for new liver metastases.

On her MRI, we saw several and small enhancing lesions in the liver.

We performed contrast ultrasound, first stone gray scale.

You can see a very sharply defined small round hypo coic lesion with a little bit of internal blood flow on the arterial phase, hyper enhancing on the portal, venous phase iso enhancing, but most importantly on the delayed phase, there was out of this lesion.

For a number of reasons, it was decided not to biopsy these lesions and we're following this patient, but we're concerned that these are metastatic deposits.

I wanted to show an example of an confirmed metastasis so you can see what these look like.

This was in a patient with a melanoma who had two numerous to count nodular hyper enhancing lesions scattered throughout her liver, which on the delayed phase imaging washed out.

Overall Conclusion

So in conclusion, contrast enhanced ultrasound is uniquely suited for use in pediatric imaging in general within pediatric oncology.

It's very useful for measurement of tumor blood flow for distinguishing benign processes from tumor for assessing fluid collections for distinguishing blood clot versus stasis, and for characterizing focal liver lesions.

I thank you for your attention.