UPMC Physician Resources
Sorry CME is not currently available for this activity.
Contemporary Management of Head and Neck Cancer: Role of Robotic Surgery
Dr. Umamaheswar Duvvuri discusses the the role of robotic surgery in the contemporary mangement of head and neck cancer.
Upon completion of this activity, participants should be able to:
- Review the role of HPV in the context of oropharyngeal cancer
- Describe the standard of care for patients with oropharyngeal cancer
- Discuss surgical management and treatment for head and neck cancers
- Byrd JK, Duvvuri U. Current trends in robotic surgery for otolaryngology. Curr Otorhinolaryngol Rep. 2013 Sep 1;1(3):153-157
- McDermott M, Hughes M, Rath T, Johnson JT, Heron DE, Kubicek GJ, Kim SW, Ferris RL, Duvvuri U, Ohr JP, Branstetter BF. Negative predictive value of surveillance PET/CT in head and neck squamous cell cancer. AJNR Am J Neuroradiol. 2013 Aug;34(8):1632-6
- Mehta V, Johnson P, Tassler A, Kim S, Ferris RL, Nance M, Johnson JT, Duvvuri U. A new paradigm for the diagnosis and management of unknown primary tumors of the head and neck: a role for transoral robotic surgery. Laryngoscope. 2013 Jan;123(1):146-51
Dr. Duvvuri has reported no relevant relationships with proprietary entities producing health care goods or services.
The University of Pittsburgh School of Medicine is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.
The University of Pittsburgh School of Medicine designates this enduring material for a maximum of 0.75 AMA PRA Category 1 Credits™ and self-assessment credits toward Part 2 of the ABS MOC Program. Each physician should only claim credit commensurate with the extent of their participation in the activity. Other health care professionals are awarded (0.075) continuing education units (CEU) which are equivalent to 0.75 contact hour.
For your credit transcript, please access our website 4 weeks post-completion at http://ccehs.upmc.edu and follow the link to the Credit Transcript page. If you do not provide the last 5 digits of your SSN on the next page you will not be able to access a CME credit transcript. Providing your SSN is voluntary.
Release Date: 12/27/2013 | Last Modified On: 5/21/2014 | Expires: 5/23/2016
Good afternoon ladies and gentlemen, it’s my pleasure to present a talk today entitled the Contemporary Management of Head and Neck Cancer: the Role of Robotic Surgery. I’m Umameshwar Duvvuri and I’m an Assistant Professor of Otolaryngology at the University of Pittsburgh School of Medicine. As a disclosure I have served as a proctor for intuitive surgical in the past year.
Squamous cell carcinoma of the head and neck is the 8th most common cancer in developed regions of the world comprising approximately 38 to 40,000 new cases per year in the U.S.; however it is one of the most common cancers worldwide, causing up to 500,000 cases throughout the world and up to 300,000 deaths worldwide. Indeed we know that the incidence of head and neck cancer is strongly related to tobacco and alcohol use and that the decreasing incidence of tobacco use in the U.S. has potentially contributed to a declining trend in the incidence of head and neck cancers that are typically associated with such risk factors and these include cancers that arise from the oral cavity, larynx and pharynx including the hypopharynx. However we have noted recently that tumors arising from the oral pharynx or the region of the posterior tongue, that is the base of tongue or the palatine tonsils is in fact increasing in trend and this increase is likely related to the incidence of human papilloma virus induced tumors.
To refresh our memories, the oropharynx is separated into four parts, the brace of tongue, the soft palate, the tonsillar pillars which encompass the palatine tonsils and the posterior pharyngeal wall. Of note Waldeyer’s Ring which is a ring of adenoidal tissue which comprises the typical adenoidal pad at the nasopharynx, the palatine tonsils laterally and the base of tongue inferiorly is an important lymphoid tissue that is associated with the mucosa of the base of tongue and is thought to harbor many patients with occult primary tumors also known as unknown primary carcinomas. However this oropharynx is also particularly sensitive to infection by the human papilloma virus and Chaturvedi, et al initially published in the Journal of Clinical Oncology data that showed that human papilloma virus and associated tumors seem to be increasing at a rather fast rate, up to 5-fold increase in risk over the 1975 to 2005 time span; however human papilloma virus unrelated tumors such as tumors arising from the oral cavity, the larynx and the pharynx seem to be declining. Of note human papilloma virus positive tumors tend to occur in middle aged men, typically the demographic is a Caucasian nonsmoking, nondrinking male.
So why is this important? This is particularly important for the medical community at large because the incidence of oropharyngeal carcinoma seems to be increasing and at least at the University of Pittsburgh Medical Center we have gone back and looked retrospectively at our cases from about 1980 until present to 2005 era which gives us time for follow-up on these patients and we seem to find an increase of approximately 4 to 5% per year of patients presenting with human papilloma virus related tumors. And these tumors typically tend to occur in patients who are relatively young, at the age of about 50. So the classic presentation of these patients is most often a painless neck mass and this can many times by cystic in nature, it does not have to appear as a solid mass. And so these cystic neck masses can often be confused or misdiagnosed as branchial cleft cysts; however it is important to remember that in adult neck mass, that is a neck mass that presents in a patient over the age of 40 is cancer until proven otherwise. That should be the leading thing on the differential diagnosis list is malignancy and as such an index of suspicion, a high index of suspicion is really required to allow us to make that diagnosis. These patients can often present with a sore throat, that can present with dysphagia and sometimes dysarthria or odynophagia if the tumor if is in fact large enough. However many of these tumors are quite small in the base of tongue and may not present with any of these common symptoms until the tumor presents in the more locally advanced stage.
So to go back to this concept of human papilloma virus and why it’s important in the oropharynx we know that HPV is a known contributor to cervical dysplasia and carcinomas and this is typically associated with HPV types 16 and 18, these same high risk subtypes are important for patients with oropharyngeal carcinomas. Dr. No-Hee Park at UCLA performed some pioneering studies which demonstrated a similarity between the cervical/vaginal mucosa and oropharyngeal mucosa and its ability to be infected by the virus and in fact applying tobacco smoke extracts to epithelial cells that are grown in culture when infected with the human papilloma virus tobacco smoke seems to increase the incidence of HPV infection, which may explain why some of these patients tend to also be smokers.
The most common histology is one of a basaloid squamous pattern, these tumors tend to be dedifferentiated or nonkeratinizing and all of these words can often seen in pathology reports and these all give us a clue that this is a human papilloma virus related tumor, and HPVomas must be thought of in the context of HPV prone anatomical regions such as the oropharynx including the tonsil and the base of tongue.
So what is HPV? Human papilloma virus is a small unencapsulated DNA virus, there are many, many subtypes as I said, 16 and 18 are the high risk types. These tumors contain viral oncogenes known as the E6 and E7 oncogenes. These oncogenes now once the virus has infected the cell usually a terminally differentiated cell interacts with P53, a retinoblastoma protein RB and ultimately affects the P16 status and can actually convert a terminally differentiated keratinocyte into a non-terminally differentiated keratinocyte and cause uncontrolled growth. And this uncontrolled growth really is the hallmark of cancer biology and tumor development. It is still unclear as to why the oropharynx is preferentially targeted in the head and neck region as opposed to let’s say the oral cavity, the oral tongue or the larynx; but we do know that the oropharynx is preferentially targeted.
So how do we check for human papilloma virus in HPV detection? We usually obtain samples, tissue samples from either the lymph node or the primary tumor site which includes – which could include FNA biopsy, finite aspiration biopsy or histologic biopsies and these tissues are then subjected to HPV in situ hybridization looking for the high risk subtypes and staining with P16, which is the protein that ultimately controls tumor regulation. P16 levels are high in HPV positive tumors. Due to the technical ease HPV in situ hybridization and P16 are recommended, PCR tests can also be performed but they tend to be far more expensive and more tedious to conduct in tissue samples and this can be done in nodal tissue from these samples. And in fact the latest data suggests that P16 positivity is sufficient to predict HPV infection in these cases. So now that we understand that oropharyngeal cancer is increasing in incidence, we understand that oropharyngeal cancer is caused by human papilloma virus in many cases, up to 80% of cases, let’s move on to discussing the treatment of these patients.
There are really two main forms of treatment for patients with advanced – locally advanced or any head and neck cancer really, and that’s surgery with adjuvant therapy which may include chemotherapy and/or radiation therapy and nonsurgical therapy that some might term organ preservation which really revolves on the use of chemotherapy and radiation therapy.
The surgical approaches are the same as what has been proposed by Halstead and others for many, many years. The general principle is to take a wide surgical resection with negative surgical safety margins followed by you know adjuvant therapy as indicated. However the oropharynx is prime real estate, there are many, many critical structures that live in this area, it includes the carotid artery, the vagus nerve, the otopharyngeal nerves, the jugular veins and so while many of these tumors might present initially as small lesions in order to be able to access them surgical through conventional techniques we had to perform a rather large operation. And this slide depicts a patient who has undergone a lip slit mandibulotomy so the nose is shown here, the maxillary teeth, the dentition is demonstrated towards the top of the slide, this is the tongue and the jaw has been split in the midline and opened in order to allow us access to a very small tumor which is demarcated by the black arrow. So as you can see this is a relatively long run for a short slide and these patients need to often undergo tracheotomies, gastrostomy feeding tubes and a long operation which could last as long as 12 or 14 hours if there is need for reconstructive surgery.
And here is a picture of a patient of such as this who has a tumor demonstrated in panel A which is on the left tonsil, we can see malignant lymph adenopathy of the neck which is how this patient presented. The patient looks quite normal at baseline, there is not a large amount of deformity at baseline but once she’s subjected to a mandibulotomy with resection of the mandible and mandibulectomy, a resection of the tumor at the oropharyngeal region and neck dissection as well as a tracheotomy performance we can see the patient is left with the stigma of these operations. And so while these patients can undergo the operation safely three is great morbidity of this surgery. There is a need for tracheostomy, there is a need for gastrostomy tubes, there is a need for an intensive care unit stay as well as prolonged convalescence and rehabilitation. And indeed many of these patients ultimately go on to need adjuvant therapy anyway.
As such the scientific community began to wonder is there a need to treat these patients surgically if we are going to give them chemotherapy and radiation anyway? And this shift towards nonsurgical therapy really was sparked by the desire to obviate some of these aggressive surgical procedures in these patients. And indeed definitive therapy has been shown to be useful in these patients. Definitive radiation therapy alone has been advocated for low volume disease by the data from MD Anderson Cancer Center, however patients with locally advanced disease are typically treated with chemotherapy as well as external beam radiation.
So what is the data on radiation therapy alone? Adam Garden presented a series of 175 patients who were all stage I and stage II who were treated with 50% concomitant boost radiation therapy but these patients all had irradiation alone and he found that there was excellent survival at 5 years, 85% ranging from 88% for stage I to 72% for stage II and about 29% or 30% of these patients developed a second primary tumor which then leads us to the question how do you address the second primary tumor? How do you salvage these patients should they fail their initial surgery, I’m sorry their initial treatment with radiation?
Calais et al purported randomized control trial from Europe which really shifted the paradigm of treatment for patients with oropharyngeal cancer. The Calais trial took 222 patients all of whom had advanced stage, either stage III or stage IV oropharyngeal cancers. These patients were randomized to undergo irradiation therapy verus irradiation therapy with chemotherapy. And in this study overall survival at 5 years was improved in the patients who underwent chemoradiotherapy, however these patients also had an increased rate of complications. And one measure of that is a need for long term feeding tubes, gastrostomy tubes which was almost 3 times as high. However the local control rates were significantly better in the CRT group and this study established that chemoradiotherapy, here abbreviated as CRT, is the standard of care for patients with advanced stage oropharyngeal carcinomas.
Now more recently we have been able to tailor the adjuvant radiation therapy or the definitive radiation therapy to use techniques such as IMRT, or intensity modulated radiation therapy, to treat patients with oropharyngeal cancer and these IMRT techniques have also shown to have really good outcome and be able to preserve normal structures and avoid some of the morbidity. But as I said the major morbidity of radiation therapy once again is the need for gastrostomy tube placement and the need for these patients to have feeding supplementation.
So Avi Eisbruch at Michigan reported in JCO, the Journal of Oncology, 73 patients who were treated in a prospective fashion with follow-up of 36 months, all treated with IMRT and what he found that every patient had worsening dysphagia scores, meaning they were given a survey to look and see whether or not they had worse dysphagia before versus after treatment. Everybody had worse dysphagia, most people recovered, and they felt that they recovered, their subjective scores recovered over time but their objective scores really didn’t change. In other words, if an outside observer was looking at these patients and trying to determine whether their dysphagia, their difficulty in swallowing was better it wasn’t. So what this study told us is that even though chemoradiotherapy can cure these patients or control these patients’ disease it still carries with it a stigmata of morbidity which is the difficulty in swallowing.
Now people have attempted, the investigators have attempted to tweak these paradigms, so what if we didn’t give the chemotherapy with the radiation therapy? What if we gave the chemotherapy first? And so the TAX 324 study which was published in the New England Journal of Medicine showed that giving Cisplatin and 5 Fluorouracil and Taxol as an induction regimen upfront was better than giving Platin and 5FU alone; however many of these patients, I shouldn’t say many, 10 to 15% of these patients with induction chemotherapy regimens do not go on to get consolidation therapy, that is complete their therapy with adjuvant chemoradiation. So these induction therapy protocols really ought to be studied and implored in the context of clinical trials and in my opinion they are not yet – they have not yet superseded concurrent chemoradiotherapy as a standard of care for patients with advanced or locally advanced oropharyngeal carcinomas. It is important to remember that all of these data that I have shown you were collected before routine testing for human papilloma virus was implemented so we don’t know what the impact of HPV positive tumors or what the outcome of HPV positive tumors are, or patients are in the context of being teated on these protocols with chemotherapy and radiation therapy.
To address this question Kian Ang and Maura Gillison published a very elegant review of a prospective trial that was conducted by the Radiation Therapy Oncology Group, RTOG, this was a retrospective analysis of a prospective trial and in this particular trial 323 patients were treated, however 206 of them were oropharyngeal cancers or HPV positive so 64% and when they looked at patients with HPV positive disease they found a much better overall survival at 3 years, 82% versus 57% for those patients who were HPV negative. Most importantly, smoking was also a poor predictor of survival in these patients. So what we would like to – the take home point from this slide is that HPV positive patients do better, HPV positive patients who smoke tend to do worse than nonsmokers and HPV negative smokers do the worst of them all, and there is a very nice delineation of these curves which can be seen in the article published in the New England Journal of Medicine. However this potentially good effect of having a human papilloma virus positive tumor is not limited to patients treated on chemotherapy and radiation therapy protocols alone.
Lisa Licitra published a very nice article from Italy once again Journal of Clinical Oncology which just looked at patients who were treated with surgery. So these patients were all treated with surgery prospectively, the tumors were then analyzed for human papilloma virus and risk factors for radiation therapy were also analyzed including multiple nodal disease, multiple nodal disease, positive surgical margins, lymphovascular invasion, etc. And what she found was that human papilloma virus positive tumors that did not require adjuvant radiation therapy had 100% survival over 8 years, 100% survival over 8 years; whereas HPV positive tumors that did require radiation therapy, that is they had multiple nodes, they had positive margins had a less favorable survival. However both of these groups did significantly better than HPV negative tumors whether they needed radiation or not suggesting that HPV positivity in and of itself is a predictor of good outcome whether we treat the patients with surgery or with chemotherapy and radiation therapy.
So why does this matter and why is this important to our discussion at hand? And what is the contemporary management of oropharyngeal cancer and head/neck cancer? I’ve told you that all treatments carry with them some risk of morbidity so when we are dealing with patients who tend to be high functioning patients, who tend to have active careers and then develop oropharyngeal carcinomas I told you the demographic is middle aged men, and here is an example of George Karl, who is the basketball coach of the Denver Nuggets who popularized his battle with oropharyngeal carcinoma as has more recently Michael Douglas for example with his HPV positive oropharyngeal carcinomas, there is a greater and greater emphasis for these patients to not only be cured of their disease but to have a functional quality of life after the disease. We don’t want the treatment to be worse than the disease itself. We don’t want to subject these patients to a lifelong dysphagia, lifelong need for gastrostomy tubes if we cure their cancer. So the challenge at hand is how do we reduce the morbidity of treatment for these patients while maintaining their oncologic outcomes? And this is not a simple question, I don’t have a simple answer to this question but this is where the field is moving and this is where we are moving in the near future.
And one of the questions at hand is can minimally invasive surgery help these patients? Minimally invasive surgery evolved from videoscopic surgery much like endoscopic sinus surgery for head and neck, for you know patients with sinus disease or laparoscopic surgery for patients with GI disease, many LAP CHOLIs that are done today, cholecystectomy, open cholecystectomy is a relatively underutilized procedure at this juncture, it is – cholecystectomy has really moved towards videoscopic surgery and the concepts of videoscopic surgery remain the same whether it’s in the head and neck space or in other parts of the body, we want to have good visualization and we want to have good access to the surgical site. And the surgical robot we show here a picture of the Da Vinci surgical console allows the surgeon both of these key elements, it allows the surgeon to have 3-D visualization as well as wristed technology.
And in the head and neck space what the Da Vinci Robot allows us to do is to place the surgical robot arms into the mouth of the patient thereby giving us direct access to the oropharynx, there is now no longer the need to make a transcervical incision, there is now no longer the need to perform a mandibulotomy for these patients and so we can get direct access to the tumor as opposed to going through all of the cervical structures to get access to the tumor. So we’ve converted the long run into a much shorter run for the same target, the same slide. The robotic arms are introduced into the patient’s mouth, the surgeon sits at a remote console which is in the same room and allows for surgical manipulation dissection and so this is really a transoral procedure. So the section is done through the open mouth thereby limiting the morbidity but it also limits the exposure and this is why we need the surgical robot because the line of sight is not good enough to allow us to use typical cautery or a laser or other technologies so we have to employ the surgical robot in this area. In my opinion I think the surgical robot improves visualization of this area. And so this is a typical setup for a patient who is undergoing TORS or a transoral robotic surgery procedure. Here the mouth is opened with a retractor system and access is gained to the oropharynx which is shown in panel B and then the surgical robot is docked and brought into play.
So what does the Da Vinci Robot allow us to do? The Da Vinci Robot allows for 3-D visualization and it allows for the surgeon to sit at a remote console and manipulate joy sticks which then allow the surgical robot to move in a tremor free system. So there is no tremor, the physician’s hands are effectively miniaturized to 5 mm instruments so imagine being able to do everything that your hands can do but at a fraction of the size. So we can really get into tight spaces, we can get into narrow areas, we can control vessels very, very precisely and we can make very precise dissection moves and eliminate the need for wide exposure. The wristed technology also allows a surgeon to really move the instruments in 540 degrees of freedom, so it allows me to make maneuvers within the body which I typically could not do with my bare hands. And this is another advantage to the system.
So why perform TORS? Well TORS really started at the University of Pennsylvania, Greg Weinstein was a pioneer in this area and used the surgical robot to treat 27 patients with small tonsil cancers, mostly T1 and T2. There were a few T3 tumors but 75% were T1, T2. He was able to show that all patients had a negative surgical resection margin and there was a very acceptable complication rate of 11%. Most importantly there was very good oncologic control of these patients.
So there are other studies that have been published on this topic, for the interests of time I’m not going to go into all of those studies but I do want to highlight a novel design that has been bred here at the University of Pittsburgh that has now moved forward in a cooperative group trial through the ECOG, the Eastern Oncology Cooperative Group which is ECOG 3311. This trial design seeks to evaluate the role of transoral robotic surgery and transoral surgery in general for patients with HPV positive, P16 positive oropharyngeal carcinomas that are of early stage, T1 T2 N0 to N2. These patients are basically subjected to surgery with the transoral robotic system and then are randomized to either low dose, that is 50 Gy radiation or standard dose, 60 Gy radiation. And the primary objective is to evaluate the 2 year progression free survival of these patients who are treated with low dose radiation therapy. That is assuming that if we get all the tumor out that these HPV positive patients are going to do well anyway, then can we take the tumor out, give them less radiation, treat them less intensively, maintain their oncologic outcome but most importantly improve their functional status, improve their ability to swallow, improve their ability to eat. And so this is what the schema looks like. This is the randomized controlled trial, well this is their ECOG 3311 trial, this is not randomized controlled yet but the patients are randomized into low dose radiation versus conventional high dose radiation depending on the pathologic outcomes. And this accrual target is set to 377, this trial has now just opened at the University of Pittsburgh and around the country and we hope to accrue patients to this trial very, very quickly.
So the other common problem that we’ve noted now with the increasing prevalence and incidence of human papilloma virus related tumors is the presentation of a patient with an isolated neck mass. If you will recall I told you that one of the most common ways that these patients present is with an isolated lump in the neck, it’s not uncommon to patients, my male patients to say well doc, I noticed this when I was shaving. It’s not uncommon for my female patients to say doctor I noticed this when I was putting on my makeup, I have this lump in my neck, it hasn’t gone away, it’s been 3 weeks. And this presentation of an isolated neck mass is usually followed by a needle biopsy because once again remember that adult patients with a neck mass is cancer until proven otherwise. So when these patients get biopsied and the squamous cell carcinoma is identified but mo primary site is identified these patients are then termed to have an occult primary colloquially known as a cancer of unknown primary.
The incidence of these patients or patients presenting with carcinomas of unknown primary occult primary tumors I have mapped here at the University of Pittsburgh Medical Center over the last 3 decades. In the 1990s we used to see about 2 of 3 of these patients per year; however in the 2000s just when the human papilloma virus if you want to call it epidemic became more popular and became more studied we suddenly noted that there was a jump from 2 per year to about 8 per year of these patients. And over the last 3 years in the decade of the 2010s we’ve seen an even more alarming rise to about 14 per year. So this is clearly a problem that’s increasing in incidence and we are seeing more and more patients like this and there will be more patients who present with isolated neck masses in all practices in medicine.
Why is this important? Because patients who present with an isolated neck mass and who do not have a diagnosis of cancer need to be worked up and need to be evaluated for the existence of metastatic carcinoma. 2 to 4% of head and neck cancers typically presented with cervical adenopathy as the initial symptom, as an unknown primary according to our old data; however these data seem to be quickly changing in the context of human papilloma virus. It should be noted that many of these patients used to be older but now not any more, they could be quite young patients in their 40s or 50s. These patients are not typically smokers and drinkers and so the treating physician needs to have a high index of suspicion for the diagnosis of such patients. And if you have a high index of suspicion then a thorough physical examination is indicated for these patients. A thorough physical examination of the oropharyngeal structures including palpation as well as flexible laryngoscopy or indirect laryngoscopy is important. Oftentimes this is performed by an otolaryngologist and so a prompt to an otolaryngologist is usually warranted. As I tell my colleagues we often look for the bloody glove sign; so on manual palpation of the base of tongue if the glove has blood on it that’s most likely a friable area which indicates the existence of a malignancy. And of course a needle biopsy, a finite aspiration biopsy of the neck nodes is very helpful.
Imaging is a valuable adjunct to the treatment of these patients, adjuncts and it complements to physical examination. Typically a CT scan with contrast is our first line of evaluation; however a PET CT scan can also be helpful to evaluate for metastases or to better stage a tumor. An MRI scan with contrast has been advocated by some authors but at the University of Pittsburgh we prefer CT scan with contrast. These patients all require operative laryngoscopy and operative endoscopy with biopsies because this allows for definitive staging and sometimes treatment planning. And tissue diagnosis is essential and larger biopsies allow for potential molecular testing.
So what is the role of imaging in these patients? As I told you that CT, PET CT and MRIs are often employed. PET CT scan has become currently the standard of care for these patients and with a PET CT alone there is about a 30 or 40% chance of identifying the primary tumor site. The sensitivity and specificity are quite high at about 85%; however PET CT has a 40% chance of finding the tumor, that means there is about a 60% chance of not finding the tumor.
And so the question becomes where is the tumor? You know there is metastasis in the neck, but where is the primary tumor? And so we really try to find this primary tumor to help guide our decision making for these patients. And so the index of suspicion is really based on anatomy as well as the physical exam, as well as the other testing that we have performed. Tumors that typically occur in levels II, III and IV of the neck tend to occur from the oropharynx. HPV positive tumors tend to arise in the oropharynx and this is what we are seeing more and more of. So you need to think about Waldeyer’s Ring, that lymphoid tissue in the oropharynx, you need to think about difficult to visualize areas such as the hypopharynx which cannot be seen very easily on indirect laryngoscopy and that’s why we take people to the operating theater to perform operative endoscopy in these patients, and of course you should always remember infraclavicular sources such as lung cancer, esophageal cancer, breast cancer in women, these should all be evaluated and thought of.
And so once again the site that the primary tumor – I’m sorry the site that the metastatic node resides in gives us a clue as to where the tumor could be starting, the primary tumor could be starting. And so the primary echelon of drainage for oropharyngeal tumors is the internal jugular vein chain, levels II, III and IV and retropharyngeal nodes are the second echelon. Level V spread, that is nodes behind the sternocleidomastoid muscle is very rare and so base of tongue, the lateral pharyngeal walls tend to spread to levels II and III, this is their primary drainage pattern and this is what we think about patients having primary tumors in the base of tongue when they present with level II or level III disease. The posterior pharynx typically drains superiorly to the Rouviere’s nodes or the high nasopharyngeal, retropharyngeal nodes and that’s also something to be borne in mind.
So why do we care about finding the primary tumor? Because it may be associated with a survival advantage but most importantly locating the primary tumor allows us to narrow the radiation field. It allows us to target the radiation so that we are not radiating these patients from the nasopharynx to the larynx in a wide field of radiation which can cause significant morbidity as anybody who is taking care of these patients knows. Most importantly if the tumor can be cleanly identified and potentially even resected there is a possibility to deescalate therapy. So once again what do we gain by finding the tumor for these patients? We facilitate staging so the patients are no longer stage TX they can be given a stage based on the size of the tumor. This potentially improves enrollment to clinical trials as I showed you that ECOG 3311 trial for example that’s currently undergoing accrual. It allows us to plan for definitive therapy and almost all HPV positive tumors originate in the oropharynx so with that knowledge we can really target where we want to look for these tumors and try to find them more cleanly.
And so does finding the primary site really matter? We are not sure. Some authors have argued that identification of the site is not as important as the histology of the tumor but in our mind we think that it does matter and so we sought to evaluate retrospectively the University of Pittsburgh data to determine whether finding the primary tumor really mattered. So we set out to look retrospectively at our series of 136 patients who were initially diagnosed as presenting with cancer of unknown origin or occult primary, we then found about 67 patients in whom the tumor was identified at the initial trip to the operating room and an additional 69 in whom the primary tumor was never found. We subsequently went on to identify 22 matched pairs from these two cohorts so that we could really sort of tease out whether finding the primary tumor mattered. Interestingly this just shows the demographics, there was a slight difference in age between the tumors that were found versus not found, but this was our most shocking data that identifying the primary tumor which is shown in this green graph, green line in this Kaplan-Meier survival curve was associated with significantly improved survival when compared to those patients in whom the tumor was never found. We don’t yet have any explanation for exactly why this is the case but these – but this outcome actually stratified across nodal status, meaning that this wasn’t just that all the patients in whom the primary tumor was found had small lymph nodes or small tumors, and patients in whom it wasn’t found had big lymph nodes, they were matched across those criteria.
Similarly within the subset of matched patients we found the same finding to hold true that the disease free survival and cause specific survival and overall survival were all significantly better for patients in whom the primary tumor was found. However this – we think that this is – this is potentially independent of HPV status but we are evaluating this more carefully because it turns out that the patients in whom the tumors were found tended to be HPV positive much more frequently than those in whom the tumor was not found. And once again here are cause specific and disease free survival curves which demonstrate the same data and all of these data sets are statistically significant.
So if we then believe that finding the primary tumor does matter and it is important the question becomes where are we missing the primary tumor? Why were we not able to find it, why is it that PET scan can only find 40%? Where are the rest of these tumors lurking? And so this requires us to really look for the primary tumor and we can’t just tell ourselves that the primary tumor isn’t there. Actually on this little cartoon that I stole from the internet the identification of the primary site is often facilitated by panendoscopy, which is a diagnostic endoscopy under the operating room. However we remove the palatine tonsils because we can see them cleanly but we’ve never been able to remove the lingual tonsils, the inferior portion of the Waldeyer’s Ring. And we know that palatine tonsillectomy is important because bilateral tonsil malignancies have been reported, we know that taking out the entire tonsil is better than doing a tonsil biopsy or tonsillotomy, and Leon Barnes, our head of Head/Neck Pathology when he was here used to advocate for complete tonsillectomy so that he could then section through that entire tonsil and find the tumor, find these small primaries because many of these could be less, much less than ½ an inch, could be on the order of a centimeter. And several studies have shown that performing a panendoscopy with a tonsillectomy with or without imaging has a hit rate of about 40%, 17 to 40% depending on the studies. So we are still not perfect, we still miss some of these tumors.
So we started with the hypothesis that these tumors must therefore be hiding in the lingual tonsils, and we note that we used to take random biopsies, we thought there were a lot of tissue that was biopsied, but typically it was multiple biopsies of about 5 mm each. So when you look at the entire tongue base and the entire oropharyngeal region taking multiple small biopsies may not be adequate. So Leon Barnes tasked me with the challenge of can we submit the entire lingual tonsil, is this doable? And this is when we started thinking about getting access to the tongue base. And once again these procedures, the open procedures to get to the tongue base are often morbid. Here is a patient who – in whom we had to perform a lip split mandibulotomy, breaking the mandible in the midline to get access to the tongue and the base of tongue. Now this is a very, very big operation to subject a patient to just to try to find an unknown primary tumor. It may not work 100% of the time, so this is a rather morbid procedure.
However both of the use of Da Vinci Surgical Robot has allowed us to perform a complete lingual tonsillectomy with relatively limited morbidity. So this is a patient in whom we are looking at the back of the tongue, so this is the lingual tonsil tissue and the back of the tongue, here is the endotracheal tube. The surgical robot is brought into play, here are those little 5 mm arms that I mentioned earlier and we can use the surgical robot to now perform an anatomical resection of the entire lingual tonsil tissue in the back of the tongue, both the right and left sides from the circumvallate papillae which is the anterior most limit of the oropharynx and the lingual tonsil tissue all the way down to the vallecula. Here you can see the epiglottis and we can remove all the mucosa to the vallecula. This now allows us to submit a much more impressive pathologic specimen for evaluation. So we are no longer sending 5 mm pieces of tissue, we can submit a piece of tissue that looks like it’s about 3 to 4 cm by 3 to 4 cm by about a cm in depth. And this now allows us to perform a complete resection of the tongue base.
I am going to show you a very – in the interests of time I won’t make this very long but a brief video to demonstrate the surgical technique. The patient is once again in the operating room, is completely anesthetized and here is the endotracheal tube, incision is made along the circumvallate papillae from left to right sides, the specimen is then bisected in the midline and then the specimen is then removed from the left side and then the right side allowing us to take the entire specimen, the entire lingual tonsil tissue. The specimen is then surgically oriented with ink for the pathologist to be able to tell us where the tumor came from and even be able to assess operative margins on these patients. And we take this dissection to the level of the base of tongue musculature, we don’t actually resect the tongue musculature, the lingual tonsils sit just on top of the tongue musculature so we cannot really subject the patients to a lot of morbidity because the tongue is in fact still there, this is just removing the tonsils. This is not that different from a palatine tonsillectomy in an adult patient, so we have removed one side and now we are removing the other side in a similar fashion performing a lingual tonsillectomy.
Once again, this is what the specimen looks like. It’s a really hearty piece of tissue which the pathologist can now section through closely and find the primary tumor and even though we remove such a large piece of tissue from the tongue base or from the lingual tonsil area postoperatively the patient feels quite nicely with limited complications. These patients do however have significant pain as they would if they had a simple tonsillectomy. But we are able to manage their pain and these patients all leave the hospital within one day of surgery.
So thus far we have presented 22 patients who – we have seen 22 patients who presented with carcinoma of unknown primary, all have had PET CT or CT scans of the neck which failed to reveal the tumor. Many of these patients underwent prior endoscopy and had failed endoscopy. In these patients when we subjected them to robotic surgery we successfully identified the tumor in 19 out of 22 patients which is 86.4%. So we have now increased our hit rate from 30% as an average number in the literature to almost 90%. So we’ve tripled our ability to find these tumors. These tumors tend to be very small, about a cm in size, less than ½ an inch which can explain why it’s so hard to find them. And most importantly we were able to get negative margins of resections in 10 of these patients, almost half of them.
Now this robotic surgical procedure does not in and of itself provide the definitive treatment for patients with unknown primaries. Many of these patients present with advanced nodal disease and in fact it’s been known that patients who present with larger nodes tend to have worse survival. And so patients with larger nodes such as N3 nodes will probably be treated with adjuvant therapy including chemotherapy and radiation therapy, and I shouldn’t even say adjuvant it really will be definitive. So in those patients the robotic procedure just serves as a biopsy, it allows us to find the primary tumor. However if the patients have small disease, N1 disease which many more patients have – many more of our patients seem to present in this stage then complete resection of the primary tumor can be followed with a neck dissection which will eliminate the need for adjuvant therapy in these patients. And so the nodal dissection is useful to reduce tumor burden first of all, but if we can remove all active disease in the patient we may be able to deescalate therapy and potentially irradiate all disease in the patient.
And so we advocate nodal dissection in selected cases but really we are advocating surgical therapy for the tongue base to identify the primary tumor. It should be noted however that extracapsular extension is still an indication for adjuvant chemoradiotherapy. So patients who have – who likely are getting of having ECS, that is N3 nodes which we know have a 75% chance statistically speaking of having extracapsular extension may not really benefit from having aggressive nodal surgery – nodal dissection but this is an active area of investigation. In our series we found that 4 out of 22 patients underwent surgery alone, sorry underwent radiation therapy alone, 3 out of 22 patients presented again with small volume disease, N1 disease, we took the primary tumor out, we took the lymph nodes out and they had no additional therapy, surgery alone.
So in summary, we changed the management of these patients, patients who would have been treated with definitive chemoradiotherapy with wide field in 30% of cases, which in my opinion is a significant advance in the field and serves to deescalate therapy for these patients because they will have a better functional outcome. For future directions we are obviously working to refine the management of the patients who present with unknown primary carcinomas. I would submit to you that identification of the primary tumor site is important in these patients as evidenced by our data both becomes it seems to impact overall survival of the patients, but most importantly because it provides accurate staging information for these patients allowing us to randomize them into clinical trials if the patients are so amenable and most importantly it allows the radiation therapists and the oncologists to perform rational care pathways for these patients as opposed to treating them as unknown primary carcinomas with wide field radiation. The initial results with the robotic surgery for the unknown primary carcinomas are encouraging, however this technology is currently not available at all UPMC sites and so if there are patients who have unknown primary carcinoma of the head and neck we encourage a referral to a dedicated specialist center such as the Department of Otolaryngology here for consideration of possible robotic surgery. And of course prospective trials are currently underway to define the role of transoral robotic surgery in the management of patients with oropharyngeal carcinomas.
I’d like to acknowledge my colleagues, Dr. Jonas Johnson, Dr. Bob Ferris, Dr. Steve Kim, my partners in the Department of Otolaryngology, the Division of Head and Neck Surgery and also the hard work of our trainees Cara Davis, Ken Byrd and Vikas Mehta who were instrumental in performing the retrospective analyses. Thank you.