Screening for Nasopharyngeal Carcinoma

Podcast Transcript

Dale Shepard, MD, PhD: Cancer Advances, a Cleveland Clinic podcast for medical professionals, exploring the latest innovative research and clinical advances in the field of oncology. Thank you for joining us for another episode of Cancer Advances. I'm your host, Dr. Dale Shepard, a medical oncologist here at Cleveland Clinic, directing the Taussig Early Cancer Therapeutics program and co-directing the Cleveland Clinic Sarcoma program. Today, I'm very happy to be joined by Dr. Jacob Miller, a radiation oncologist and Assistant Professor of Radiation Oncology here at Cleveland Clinic. He's here today to talk to us about screening for nasopharyngeal carcinoma. So, welcome.

Jacob Miller, MD: Thanks for having me, Dr. Shepard. Pleasure to be here.

Dale Shepard, MD, PhD: Absolutely. So, to start off, let us know a little bit about what you do here at Cleveland Clinic.

Jacob Miller, MD: As you mentioned, I'm a radiation oncologist, so I spent about half of my time in the clinic treating patients. My clinical subspecialization is in head and neck cancers, cancers of the throat, larynx, the skin, as well as lower GI cancers. And the other half of my time, I primarily do translational research, and my research interests really focus on virally associated cancers. So, cancers like cancers that are caused by HPV, EBV, and other viruses, which are very common here in the United States and in other parts of the world.

Dale Shepard, MD, PhD: Excellent. Well, we're going to focus on some work you've done looking at screening for nasopharyngeal carcinoma. So, let's take a big step back, because there are a lot of different people that might be listening with different backgrounds. What is nasopharyngeal carcinoma?

Jacob Miller, MD: So, nasopharyngeal carcinoma, in Ohio, it's actually a very rare tumor. This is a carcinoma that arises in the nasopharynx, as it sounds like. And worldwide, believe it or not, it's actually the second most common head and neck cancer after oral cavity tumors. In the United States, the most common head and neck cancer is probably oropharynx cancer followed by oral cavity tumors, but nasopharynx cancer is really a rare disease in the United States.

I did my residency training in radiation oncology in the Bay Area, and we actually saw nasopharynx cancer relatively commonly, because of differences in the patient population. There are a relatively large population of patients from Southern China and Southeast Asia. And worldwide, the burden of nasopharyngeal carcinoma is disproportionately shared in those regions of the world. And it really is a fascinating tumor, for a number of different reasons. Worldwide, probably about 95 percent are caused by the Epstein-Barr virus, or EBV, and just like oropharyngeal carcinoma, it's a virally associated tumor. It's caused by a lifelong latent infection with this virus. But what's really fascinating about it is that this is a virus, the mononucleosis virus, that we all get. So, if you test all of us for antibodies, about 95 percent of us have antibodies against EBV, indicating that we've been infected at some point in our life.

But for some reason, and we don't really understand why, this is a cancer caused by this virus that, by an incidence ratio of probably a thousand to one or maybe even 10,000 to one, is primarily found in Southern China and Southeast Asia, and is a rare disease otherwise. And that's really interesting. In other parts of the world, there's a strong relationship between exposure, like smoking and maybe lung cancer, but this exposure has a heterogeneous effect upon the incidence of cancer. And it's geographically restricted, it's restricted to certain populations, and we really don't know why.

Dale Shepard, MD, PhD: Interesting. Tell us a little bit about the screening efforts that maybe were in place in China at this point and how you're trying to look into better ways to think about screening.

Jacob Miller, MD: Yeah, that's a great question. When I was a resident learning about nasopharynx cancer, it's a disease that's primarily treated with radiation, and most of the advances, in terms of therapies, have actually been an intensification of chemotherapy over the years. It's a radio-sensitive tumor. It's a chemo-sensitive tumor. Most of the therapeutic advances have been added more chemotherapy, more recently, adding immune therapy to chemotherapy, and there have been incremental and real clinically significant improvements in outcome. But when you see these patients in clinic, it's a location of the body that typically doesn't become symptomatic until somebody has very large lymph nodes of the neck or has a very large symptomatic tumor, that's often-causing skull base erosion or bleeding from the nose. So, most patients are present at an advanced stage, and that's actually reflected why the therapeutic advances have been an intensification of chemotherapy, long courses of radiation, long courses of chemotherapy. And we've been able to cure incrementally more and more people.

But like for other cancers, early detection facilitates diagnosis at an earlier stage, where the prognosis is better, and the intensity of therapy is a lot less. So really, the standard of care for patients with stage one NPC is radiation by itself, and so, we get to avoid the acute and late toxicities of chemotherapy and shorten their overall treatment duration. So, if we can identify patients at an early stage, before they become symptomatic, they get less intense treatments, and they have less toxicity from the treatment and better long-term outcomes.

So, what's interesting about EBV-associated NPC, EBV has sort of been one of the original cancer biomarkers. So even in the 1970s and the 1980s, they recognized the association between the Epstein-Barr virus and Burkitt lymphoma, and the same observations were made between the Epstein-Barr virus and nasopharyngeal carcinoma. And that was actually initially done using serologic studies. So, there are these very old clinical trials published in the '80s and '90s, where they did serologic mass screening, so screening tens of thousands of individuals for antibodies against the Epstein-Barr virus. And they showed that that is a real, but relatively crude, biomarker for this disease.

People that have high titers to specific EBV antigens in their blood have a higher incidence of this tumor, and so, they were subsequently referred for getting mostly nasal scopes and examinations. And you could indeed detect cases in an early stage. So that has actually been the state of the art in terms of NPC screening, until probably the 2010s, and led to the motivation for this trial.

Dale Shepard, MD, PhD: So, tell us a little bit about the trial. What did you do?

Jacob Miller, MD: This trial was led by a group at Sun Yat-sen Cancer Center, which is in Guangzhou in Guangdong Province in Southern China. And this is a center that treats thousands of cases of nasopharyngeal carcinoma annually. So, it's an enormous cancer center. Most of the research, in terms of therapeutic advances, has come out of this center in terms of what the standard of care is, and this center has also led efforts in modernizing screening trials. The other major center that has really emphasized advances in molecular diagnostics has been in Hong Kong.

But this was a trial that started in approximately 2008, and it's a very large cluster randomized controlled trial. There were 16 towns within Guangzhou that were randomly assigned to a screening intervention or no intervention, a control arm. So individual participants weren't randomized, but they randomized entire towns. And there were about 350,000 residents that were randomized. The target population were adults, both men and women, between the ages of 30 and 69, and in the control population, the incidence, stage, and mortality of NPC was recorded via linkage to cancer registries.

And in the screening arm, individuals in the screening towns were invited to participate in the screening intervention, which was an invitation to complete serology against the Epstein-Barr virus. And via previous trials, the serology is against two antigens. One is called viral capsid antigen, the other one is called EBNA1, which is a nuclear antigen. It's a transcription factor. And the combination of the titers to these two antibodies dictates a relatively high sensitivity and specificity for presymptomatic occult nasopharyngeal carcinoma. So as part of this trial, participants who had high serologic risk were referred for nasal endoscopy, so fiber optic nasal endoscopy to look in their nasopharynx and look for these cancers.

And there were several additional sub-cohorts that looked at the utility of MRI, as well as nasopharyngeal swabs, which are the same types of swabs that we get for COVID testing. But instead of looking for COVID via PCR, they're looking for EBV via PCR, EBV DNA. So, the trial has about 10 years of follow-up. Now it's been reported on multiple different occasions, but what the trial has observed is that individuals randomized to the screening town have a lower risk of NPC mortality. Among about 1200 subjects that were diagnosed with NPC in the trial, the risk of death at about eight years is 10 percent lower in the screening arm. And that's because the stage distribution of these cases has flipped. So normally in clinics, when I see patients in the Bay Area or when I see patients here with EBV associated NPC, about 80 percent have stage three or stage four disease. And with screening, it's flipped, about 80 percent have stage one or stage two NPC, and the therapeutic implications are different. So, it's really with that earlier stage of diagnosis, we can offer them less intense therapy. And the cure rate is also higher.

Dale Shepard, MD, PhD: When we think about the stage in itself, there were a number of different screening strategies. Is there anything that stood out as being sort of the optimal screening strategy?

Jacob Miller, MD: Great question. So, there have really been about 10 prospective NPC screening trials run either in Taiwan or in Hong Kong or Southern China. And they've used a variety of different strategies. And all of these strategies are combinations of EBV biomarkers. So that can be EBV serology, EBV DNA in the blood detected with PCR, next generation sequencing, bisulfide sequencing in plasma. There was actually just last month, in the New England Journal, a new serologic mass screening study that was published with a new antibody to EBV. So, it's a combination of EBV antibodies and then, some sort of gold standard test or some sort of clinical workup test. And that is typically either an MRI or a nasal endoscopy. So, if you're biomarker positive, maybe you have a triage test, but then ultimately, you test for presence or absence of the disease with a scope in the nose or an MRI.

And as you can imagine with, let's say, five to 10 different biomarkers, a triage test, and then, a gold standard test, there's a variety of different combinations of doing these. And they all have different risks and benefits and advantages and disadvantages. And the way I look at them, obviously, the first thing is the performance of screening. So, the sensitivity and specificity, the number of people who are referred for unpleasant workup evaluations, like having a scope or undergoing an MRI. And then, the resource utilization, which is really what my main responsibility and my main participation in the study was. If we have these different competing screening strategies and they have different sensitivities and specificities and indirect costs to healthy individuals, how do we optimize the screening program within the constraints of Southern China? And those are economic constraints, but also resource constraints. How many MRI units do we have available? Just the province of Guangdong is over a hundred million people. And if you're proposing screening most adult men and women for this disease, you're going to need a lot of MRI units.

Dale Shepard, MD, PhD: And so, what did you find in terms of strictly from a resource standpoint and a cost standpoint? Was there sort of a sweet spot?

Jacob Miller, MD: Absolutely. So, in Southern China, dual antibody testing is really what the standard of care is. So, I anticipate they will likely not transition away from serologic screening, whereas in Hong Kong, molecular diagnostics are the primary screening methodology upfront. So, after an individual undergoes serology, there's really four main competing screening strategies that appear to be the sweet spot. And this depends upon what screening costs you want to expend and then, what reduction in NPC mortality is desirable. But those four screening strategies are just plain serology followed by a scope. The second would be serology followed by an MRI, and then, the third and fourth do serology with a nasopharyngeal swab.

And if you're positive for both of them, then you're referred for either an endoscopy or an MRI. And in terms of the cost effectiveness literature, these are all in what's called the cost effectiveness plane. So, they're all reasonable screening strategies, but likely, if we wanted to reduce NPC mortality by the maximum amount on a population basis, it seems that the integration of MRI really is key. It does boost sensitivity from about 70 percent up to about 85 percent. You do detect additional cases, although there are costs associated with doing the MRIs, but the integration of MRI seems to be key, if you want to reduce population-based mortality.

Dale Shepard, MD, PhD: I guess, simply from a very, very practical standpoint, it's hard to get MRIs here.

Jacob Miller, MD: Yes.

Dale Shepard, MD, PhD: If you have a really, really large population, a lot of people at risk and need screen, is that something logistically possible? Is there the infrastructure to do that?

Jacob Miller, MD: That's a great point, and I think that's why multiple different screening strategies that are reasonable within different populations, and if you have an urban population or a rural population, maybe a patient that has a contraindication to an MRI and then, thinking about simplicity. If you have multi-step cancer screening protocols, you're going to have attrition with every step of the way. So, if you refer somebody for a nasopharyngeal swab and then, a scope and then, an MRI, you're going to have attrition and you're going to lose the performance, despite spending all that money upfront for the serologic test. So, we looked at both upfront MRIs, if somebody is serologically positive, but recognizing that MRIs are difficult to get in Cleveland and they're going to be difficult to get all around the world. And some patients can't tolerate them. There are a few different approaches to that.

One is shorter protocol MRIs. So instead of a traditional 45-minute MRI or 30-minute MRI, this group has actually developed non-contrast, just T1 weighted MRIs, that actually have really great performance and have more throughput of patients. That's one strategy. A second strategy is to forego MRIs altogether. Endoscopy has very reasonable sensitivity in about the 70 percent range and the lead time between essentially preclinical detection and symptomatic presentation for this cancer, it's not that fast. So, it's not as if, if you miss it in one year and you rescreen an individual one year later, they typically don't go from stage one to a T4N3 tumor within a year. We've done some modeling work, and the approximate sojourn time for this tumor is on the order of two to three years. And sometimes you actually see this with oropharynx cancers, in that, if a patient has a previous MRI or maybe they don't start treatment, they don't grow that quickly.

So doing an endoscopy over the course of subsequent years is a very reasonable and viable strategy. And then, what's also interesting is the utilization of these nasopharyngeal swabs. We think of that like a triage test. So, if somebody is serologically positive, rather than referring them directly for an MRI, we want to increase positive predictive value by increasing specificity. So, to accomplish that, if somebody is serologically positive, obtaining a nasopharyngeal swab, doing a PCR for EBV, really boosts the positive predictive value from about 4 percent to around 15 percent. So, you get to obviate and avoid a lot of MRIs and unnecessary scopes for healthy individuals and really select individuals that are at the highest risk.

Dale Shepard, MD, PhD: So, it looks like you've come up with some really interesting data, nasopharyngeal cancers in China. How can we use what you've learned and the methods you've used to apply to things like oropharynx cancer here in Cleveland and the states?

Jacob Miller, MD: I think there's two main lessons. One is that this is a disease that is indeed top five cancers in Southern China and that part of the world, but it is a cancer that we do see in our practice here in Cleveland and in North America. And it's a disease that arises in minority populations, that are identifiable via their self-reported ethnicity or ancestry. And there's no reason why these screening programs can't be developed and implemented in North America and offered to these individuals. We know that these screening methodologies work. They're successful at identifying individuals in an early stage. And right now, it seems that the barrier to implementing this is not logistical. We have the technology to do serology. We have MRIs and scopes and PCR and things like that. We actually even use these tests as part of our routine clinical care in post-transplant monitoring.

We do EBV DNA, we do EBV serology, when we diagnose people with mononucleosis. There's no real barrier why we can't extend these screening programs in North America and actually serve this patient population. So, I think that's one lesson from my research is that there really needs to be cancer screening programs that are tailored to individual populations, that may be rare diseases. And then, the second takeaway from this is that, in my clinic, I primarily see oropharynx cancers, larynx cancers, and oral cavity cancers. And HPV oropharynx cancer really is an epidemic in North America. The incidence is rising. About 90 percent of oropharynx cancers are now HPV related. We do have vaccines for them. So, I think that's the first point is all of us know we want to promote vaccination, so that we no longer see these cancers later in life. These are effective for prevention of cervical, anal, or pharyngeal cancers.

But for the population of healthy individuals, who are roughly in their thirties to age 90, they were blatantly infected with HPV. The prevalence of HPV in the saliva is on the order of about 10 to 15 percent in healthy individuals. So, it's measurable. And I think there are a lot of lessons that we can use from this group in Southern China from what they've observed over the past 30 years in nasopharynx cancer, using EBV as a biomarker, not just for screening, but also for risk adaptive therapy, for design of clinical trials, for post-treatment surveillance. And we're only beginning to see the signs of this in very early-stage trials here in North America. So currently, in our practice, we've just started to use plasma HPV DNA for post-treatment surveillance of oropharyngeal cancer. So that's a very useful biomarker and we can use that. We really need to move this upfront and use it for pre-treatment prognostication.

Do individuals that have high HPV titers in the blood have a worse prognosis? Stage for stage, does that warrant treatment intensification? Can we use mid treatment HPV DNA to adapt our therapies? If somebody is responding molecularly or not responding, can we offer them more intensive or less intensive therapies? So, I think that's a very valuable lesson that we can take from very large clinical trials that have already been conducted in nasopharynx cancer and apply them to oropharynx cancer.

And then, similarly, in the screening setting, we're all familiar with the success of cervical cancer screening. Cervical cancer is now rarer than oropharynx cancer in the United States, and we have a very large population of individuals that are latently infected, they're going to develop oropharynx cancer later in life, and it would be great if, with all this lead time of years that we know between latent infection and development of this cancer, there's some method that we can use to screen them, either in early detection or at least identifying individuals who are higher risk, and intermittently screening them or offering some sort of risk reduction, so that they're not presenting with very advanced disease. Because as we all know, the morbidity of our treatments for this disease is quite high.

Dale Shepard, MD, PhD: Screening studies are hard to do and take a long time. What are going to be the biggest barriers to doing some of these screening tests for cancers in the oropharynx?

Jacob Miller, MD: Absolutely. I think, based upon the trial that I was fortunate to be involved in, that required 350,000 randomized subjects to demonstrate about a 10 percent improvement in overall survival. I think strategies that we can use are intelligent design of clinical trials. So, I think using cluster randomized trials is thoughtful, rather than individually consenting and enrolling participants for screening. We saw that with, for example, COVID masking and large public health interventions. So, design of efficient clinical trials. And then, some of the barriers that I do think they're going to be faced with oropharynx cancer is that, unlike cervical cancer, where you can do a colposcopy, you can visualize a risk, a pre-malignant lesion, and you can excise it for secondary prevention, just like with colonoscopies, that same paradigm is not true in the oropharynx. We've never identified a pre-invasive lesion. There's no oropharyngeal HPV associated dysplasia, like there is in the cervix.

So, you either have a benign infection or you have an invasive malignancy. So that's going to be a very big challenge of, if you identify somebody who's at high risk for this cancer, how do you prove they have a cancer? How do you survey them? What modalities are you going to use to monitor them? You can't do base of tongue biopsies serially over time, because it's just too morbid and then, the number needed to screen is too high. So, I think what's been observed, again, taking lessons from nasopharyngeal carcinoma, really leveraging next generation molecular diagnostics, advanced serologic tests, so that we can non-invasively identify the highest risk population and then, watch them very, very carefully.

And we might not be able to diagnose T1N0 or pharynx cancer and then, surgically resect it. We may have to wait until somebody has a T1N0 tumor or a limited nodal disease, but at least when we can detect them in an early stage, we know that treatment outcomes are excellent. By and large, the cure rates for this tumor are very high, but we really want to identify patients before they have very large primary tumors or advanced nodal burden.

Dale Shepard, MD, PhD: You mentioned you've talked about EBV and HPV as sort of biomarkers and specific antigens. Are there any additional biomarkers that you find interesting right now? Or is it too early to say?

Jacob Miller, MD: I think, speaking from in my practice, seeing patients with primarily head and neck cancers, I think, very practically, in terms of availability of biospecimens and accessibility, if you're ever going to screen a population, you have to subject them to something that is easily accessible. So, in the head and neck, thinking about mucosal immunity and mucosal biomarkers, so for oral cavity cancer or larynx cancer, can we use saliva as a screening methodology? Because it has already been proven to be feasible in oropharynx cancer, actually. So, there have been studies of dental clinics just doing salivary washes. NHANES, which is a very large cross-sectional public health intervention or a public health study, has shown that you can detect HPV in a very large cross-section of the population. Can we detect epigenetic changes in salivary DNA, that could predict who is going to develop an oral cavity cancer? I think that's the potential biomarker. And in our population, we're actually collecting saliva and selecting patients to do these types of studies.

Dale Shepard, MD, PhD: Well, certainly screening and early detection is tremendously important. You've done some fascinating work with nasopharyngeal cancer. Looks like exciting things are coming maybe for oropharynx cancers as well. Appreciate your insights today.

Jacob Miller, MD: Thank you so much, Dr. Shepard. Appreciate being on the podcast. Thank you.

Dale Shepard, MD, PhD: To make a direct online referral to our Taussig Cancer Institute, complete our online cancer patient referral form by visiting clevelandclinic.org/cancerpatientreferrals. You'll receive a confirmation once the appointment is scheduled.

This concludes this episode of Cancer Advances. For more podcast episodes, visit our website, clevelandclinic.org/canceradvancespodcast. Subscribe on Apple Podcasts, Spotify, or wherever you listen to podcasts. Don't forget, you can access real time updates from Cleveland Clinic's Cancer Center experts on our Consult QD website at consultqd.clevelandclinic.org/cancer.

Thank you for listening. Please join us again soon.