Diagnosis, Staging, and Prognosis of Head and Neck Cancer

Transcript:Robert Ferris, MD, PhD, FACS: The question is often what is the appropriate workup for a new head and neck cancer patient? The otolaryngologist is usually the first physician or specialist on the treatment team to meet this patient. The patient usually goes to the PCP, or sometimes a dentist, with a lump in the neck, or sore throat, or an ulcer in the mouth. They’re then referred on to usually an ear, nose, and throat otolaryngologist. Some of these tumors present with a lump in the neck, the man may notice while shaving. It turns out this disease is more prevalent, about 3:1, in males, so that’s why I bring up this. It’s very common to say, oh, I just noticed this while shaving, and that’s the first sign. So a neck mass in an adult is cancer until proven otherwise. We kind of pound that into our trainees and to the community. So, when the patient arrives in the office of the otolaryngologist, a complete head and neck exam is conducted. This includes a mirror exam looking at all mucosal surfaces, usually a digital finger manual examination to feel for any masses or lumps, and then, often, a flexible nasopharyngoscopy, where we slide a fiber optic telescope through the nostril or other strategy to look at the voice box, and the base of the tongue, and the hypopharynx. We can’t see below the level of the vocal cords or into the esophagus.

For HPV-negative cancers, we are concerned about potential synchronous second primary tumors, and often we’ll perform imaging. We’ll look in the esophagus and perform what’s called, a panendoscopy, in the operating room. For HPV-positive patients, never-smokers, there’s a less frequent chance of second primary tumors. But an intravenous-contrasted CT scan of the neck usually ensues, a needle biopsy to document that it’s in fact a squamous cell carcinoma, since one could have a thyroid cancer with a lump in the neck, a salivary gland cancer. And, occasionally, cancers in the lung or breast can metastasize to the lower neck and the supraclavicular fossa. So, one really needs to document that it’s a squamous cell carcinoma in order to narrow it down to the mucosal surfaces to find the primary site.

From a needle biopsy, one can usually test for HPV. Often times, these days, we use a surrogate marker, immunohistochemistry for the p16 protein, which is dramatically upregulated when the tumor is HPV-positive. We use that surrogate and you can do that from a needle biopsy. That would help the clinician to narrow the search for the primary site to the base of the tongue and the tonsil, and help avoid biopsies unnecessarily from regions that really don’t have HPV.

In addition to CT scans one may get for advanced stages, a PET-CT, Positron Emission Tomography overload with a CT scan, so you have a co-registered scan with both PET and CT. You get functional, as well as anatomic imaging. Often that’s useful to identify the primary site, the extent of disease, and other lymph nodes maybe on the other side of the neck. And then often with a PET CT, you get the benefit of ruling out and excluding distant metastasis in the lungs or the liver. It’s not been proven whether a PET CT is really clearly better than a contrasted neck CT scan and a chest CT scan. That’s probably two standards of care in the workup. Often times, the otolaryngologist will take the patient to the operating room to perform endoscopy under anesthesia, not the sort of exam we can get in the office. And so, this is really the workup to accurately perform a clinical staging. Pathologic staging can be different. Sometimes upstaging and down staging usually requires a surgical therapeutic approach, including removal of the lymph nodes since, even in a clinically negative neck based on a CT scan, there will be metastatic disease, micrometastasis 20% to30% of the time. We can’t rely on PET scans or CT scans to do the accurate nodal staging. We have to remove lymph nodes to get the accurate pathologic staging.

Jared Weiss, MD: There are multiple factors that go into prognosis for the head and neck cancer patient. Let’s start with staging. Obviously, the worst stage is worse, but head and neck cancer has a really funny staging system. In almost every other cancer, stage IV means metastatic or incurable disease. This is not the case for most stage IV head and neck cancer. Most stage IV head and neck cancer is actually locally advanced, and we cure stage IV all the time. This is a little bit more useful in the TNM staging system, T being relevant to the size of this tumor. This is actually different for the different subtypes of head and neck cancer. The N system is somewhat useful, describing the nodal burden. A node under 3 cm is N1, N2 is a node between 3 and under 6 cm. If you get into multiple lymph nodes on the same side of the neck, that’s N2b. N2c reflects bilateral disease, and N3 is any node greater than 6 cm. Just about all locally advanced diseases will be stage III and stage IV. Obviously, worst stage is worse. There’s also a difference in prognosis by site of origin. For example, carcinomas of the lip tend to convey a superior prognosis, but the hypopharynx and larynx are a little bit worse. Finally, the biology is relevant here. In particular, we’re talking about HPV. Some cancers are driven by this virus as opposed to smoking, and the virally driven cancers have a better prognosis than these smoking driven cancers.

Robert Ferris, MD, PhD, FACS: The prognostic impact in head and neck cancer is important because patients want to know, and it helps us to design appropriately intensified therapy based on the potential for a good or worse outcome. So, it turns out that the strongest prognostic factor in head and neck cancer is the presence of HPV infection. That gives a dramatically better prognosis if the tumor is HPV-driven, something like an 80%, 2-year disease-free survival versus 40%, for instance. It’s really quite dramatic, 30% or 40% difference in overall survival. There are some modifiers such as the smoking status of the patient. And heavy smokers who are HPV-positive have a less good prognosis. A never-smoker may have a 90% to 92% survival. A smoker HPV-positive patient may have a 65% to70% survival. And then you have the HPV-negative patient with a 40% to 45% survival. So, smoking impacts. The extent of disease, the bulk of disease contributes. So T4 advanced primary tumor, extensive nodal disease on both sides of the neck, an N2c, or a big lymph node greater than 6 cm, which we call an N3—those patients, even when HPV-positive, are called intermediate-risk, and have a worse outcome than the patients with lower disease burden that are never-smokers. And that’s where you get the over 90% survival.

The tumor heterogeneity, likely it plays an important role. The more heterogeneous a tumor, the more difficult it will be to treat, and the more likely escape subsets, subclones of the tumor can arise. And you might cure 90% to 95% of the tumor, but because of heterogeneity, then you have an outgrowth of resistant cells and have recurrence. So, tumor heterogeneity through genomic analysis is just emerging as an important feature. And there are some scores out there that quantify the heterogeneity genomically. That’s important. P53 alteration has been demonstrated to be a poor prognostic factor. In particular, a subset of HPV-positive mutations that are called disruptive mutations really interfere with the effective p53 of the guardian of the genome, preventing further alterations and DNA damage, and preventing the cell from dividing and replicating if there is a lot of DNA damage. These disruptive mutations occur about 20% or 25% of the time. They appear to do much worse than the non-disruptive mutations, or where p53 is wild-type. We can look at genetic means for characterizing prognosis through single alterations like p53. We can aggregate them and say the more heterogeneous a tumor is, the worse a patient will do. But, overall, the most powerful prognostic factor is HPV status.

One question is, if you believe that there’s tumor heterogeneity, should we biopsy different areas of the tumor? I wouldn’t say that’s considered standard practice. I don’t think we really transmit that to our trainees, at present. These data are emerging. And we can identify tumor heterogeneity even with the small biopsies we traditionally have taken. This information comes from small biopsies from tissue that was in our tissue banks, so we don’t need to biopsy multiple areas of the tumor just to find the heterogeneity. We can actually identify it in the small pieces that we take routinely.

Transcript Edited for Clarity