Paranasal sinus cancer: Radiology and management

Paranasal sinus cancer: radiology and management
Presenter: Simon Gane

I’m going to talk about sinus malignancy, the radiology, mainly the management action, not the radiology. I’m Simon Gane, I’m a rhinologist, I work at the Royal National Throat, Nose and Ear Hospital. I’m also now employed as the ENT surgeon to Moorfields Eye Hospital, which is how I happen to be here.

I’m going to talk about sinonasal malignancy, the radiology and management. When I think about how what I do relates to what most of you do here, I think I’m of kind of like the person who lives at 667, the sort of neighbour of the beast: sinonasal tumours are their own thing and we handle them, but when they become troublesome, is when they stretch into your areas of expertise: the brain and and the eye. You’ll see that theme coming up again in this talk, is that we can manage the nose but when they go into your domains, is when they start to cause trouble. Spreading up, you can see this is a histological slide through my main area of expertise, and when I think about that triangle, I think apply that triangle to the way we think about sinonasal malignancy as a whole.

When I’m thinking about how to approach a cancer, I’d really think of it as this triangle between the characteristics of the cancer, the characteristics of the surgeon, and the characteristics of the patient.

So for the cancer, I want to know its location, its extent, obviously the histology, and the cell type differentiated and undifferentiated -tumours obviously have very different natural histories-, want to know the staging ‘Has it progressed? Is it going to progress?’ As well as the structures that are involved around it, what we have to deal with. We were talking earlier about the optic nerve for instance, that’s key that position in this relation to the tumour is key in our management from the sinonasal point of view.

Then the patient, ‘what do we not want to know about the patient?’ we want to know previous treatment; ‘Have they had radiotherapy?’ My earlier speaking colleagues mentioned radiotherapy completely changes what we expect in the operative field and the post-operative phase. ‘Have they had chemotherapy, sclerosants, etc.?’ Also patient comorbidities obviously. Then, and maybe this should have been at the top, patient preferences; maybe we should be taking that first and their preferences for skin incisions versus endoscopic approaches. All of these have to be put together when we’re thinking about how to manage the sinonasal tumour.

From my side of the view, what we have to think about surgeon selection should be ‘Who’s going to do this? What is their experience? What is their ability? Where are they sited? What equipment do they have available? What other facilities?’ What I didn’t say, is that our new hospital in Huntley Street, they forgot to build the theatres for us, so we’re still operating in the old hospital for another year. The new theatres are going to be sitting on top of the country’s second proton beam accelerator, which we think is going to really change the ongoing management of these kind of tumours, and well, I’ll talk about that in a little bit.

‘What is the patient journey here? What are we talking about? What’s the overview of the way we manage these things?’ Well, all patients are going to present signs and symptoms and then they come in, clinical assessment including endoscopy, they get imaging dual modality usually sometimes shows results as well; MRI with GAD, CT with contrast, definitely need to get some cells, and then we talk about it for far too long in the multidisciplinary team. This is unsurprisingly a theme of this whole conference, is the pooling of expertise being vital to the excellent treatment of these patients. The multidisciplinary team comes up with options and our options at the moment: surgery, radiotherapy, chemotherapy, all of you know all this. I like to think of this as a circle actually, we’re not just one treatment and then they’re done in these patients; they need a lifelong follow-up, lifelong examination looking at signs and symptoms, repeat endoscopy, etc. This is something that should go round to keep an eye on them, and to keep them as healthy as possible.

‘What are the kinds of surgery?’ I said that cell type is really important in our management of these, and some cell types are better for surgery and some for radiotherapy. We know from their natural history for instance, adenocarcinoma, melanoma, adenoid cystic chondrosarcoma, these are all classically surgical tumours. There may be a role for chemoradiotherapy, but not really. Olfactory neuroblastoma does better with surgery and chemo-radiotherapy; the same show for ‘SCC’, the ‘SNUC’ sinonasal undifferentiated carcinoma and rhabdomyosarcomas. And then, there those tumours, that I always feel a little bit of relief when the cell histology comes back and I know that it’s going off to the oncologists: the lymphoma and the plasmacytomas, no role for surgery in those apart from diagnostically. The surgeons are really on the first list and the radiotherapists and oncologists on the second. But recently, they’ve been a sort of spread, a metastasis of the oncologists into our side of the treatment modality. For instance, the role of new immunotherapy and BRAF inhibitors- in controlling the size and the outcomes from these, is really shown use in mucosa malignant melanoma. And in chondrosarcoma as I said, the proton accelerator is probably going to change some of that too.

I’d like to give a brief overview of what our toolbox is for the surgical approaches. Where we started in and then 1980s was pretty terrible for our management of this. The standard approach was the maxillectomy or lateral rhinotomy open approach, plus/minus orbital clearance, plus/minus radiotherapy,. Our results were dismal, 35% had five-year survival and is not good news that you want to be giving anyone in the clinic.

Over the next couple of years through the 80s, things began to improve and our options began to increase. So open approaches went from the old school Egyptian total rhinectomy to the lateral rhinotomy, which is still a workhorse access to the lateral nose. Just to orientate you, we’re looking straight on the patient eyes at the top, the whole of the lateral nose has been incised there and you can see we’re coming into, you can see the greyish tumour there. It’s really nice, you get very good subperitoneal plane early on, and you can get access to the ethmoids quite nicely. The cosmetic result is not that bad, there’s a little bit of a typical, what Professor Lund calls an alar sneer, the lifting of the alar there which is difficult to fix, but it’s not too bad. They heal pretty well and you’ve got a good access.

In mid-80s, we developed the mid-facial degloving, pioneered in this country by professor David Howard. This is a fantastic approach, it’s really clever. A bilateral sublabial incision and then bringing it up and over the nose as sort of joining a rhinoplasty type approach to expose the entire mid-face; teeth, rostral septum, caudal septum and maxilla on the side. You can lift this right up, you’re basically only reliant on preserving the orbital nerves there as your upper limit, you get excellent access to the maxilla, which you can then take down, and excellent access to the lateral nose all the way back to even the post nasal space. This has really revolutionised the management of juvenile angiofibromas for instance. It’s fantastic approach, and again the cosmesis is excellent because you’ve got no exterior incisions. Controlling bleeding and other other complications are also easy.

Then going further, my maxillofacial colleagues will see this more often, a total maxillectomy with a Weber Ferguson incision, which is excellent for the access for total maxillectomy. Obviously these patients have maybe less good outcomes, but still acceptable and they obviously return and require repair and reconstruction of the maxillary defect over the long term.

In 1979, the first craniofacial resection was performed here in London by Professors Cheeseman, Lund and Howard. This a facial incision which you can see here, but it gives excellent access to the whole of the nose and to the anterior skull base through an osteoplastic flap on the front, and you can see here we’re looking down onto the anterior skull base from the top, you can see onto the cribriform plate and you can get access to the whole area, ‘en bloc’. The whole thing which can be lifted out ‘en bloc’, thus following solid oncological principles, you’re taking out the whole thing with margins and then reconstructing it afterwards.

The craniofacial resection is fantastic approach; you had access to the whole of the skull base, there you can perform orbital clearance if necessary and controlling CSF leakage and bleeding. However the disadvantage is that incision is pretty big on the face. In addition, you’re in the cranial cavity, so it’s expected risks include meningitis, CSF leak, pneumocoele, orbital problems with diplopia, vision loss, and if the bone flap is not pedicled on anything, it can die off and give you rather unsightly defect. This is evidence from a very large multicentre study, looking at about 1200 patients, 17 institutions, the post-op mortality was just sub 5%, which is not great as you might expect, and mainly related to medical comorbidities, but the complication rate was also pretty high (36.3%), the wound complications about 20%, the CNS about 15% and orbital much less (1.7%). Not the greatest operation, it’s got a lot of morbidity, but sometimes you don’t have a choice.

About the same time that craniofacial resection was becoming more widely used, endoscopic sinus surgery was really taking off and developing. We talked about orbital decompression, which is probably where this started, and then it was used for controlling haemorrhage, CSF leak repair, and resection of benign tumours. Gradually people started to say ‘well in in certain cases it doesn’t have a role in malignant disease’ and turns out, of course, yes it does. This is very important to me, I once got as a trainee into trouble for suggesting that we might be debulking for radiotherapy, and Professor Valerie Lund was extremely upset with me for even suggesting that. It is excision with curative intent, this is what we’re doing, we’re not following oncological principles in that, we’re not doing ‘en bloc resection’ but we talk about it, as sort of ‘piece-meal’ ‘disassembly’ of the tumour within the sinonasal cavity, but it is excision with curative intent not debulking.

‘What kind of things can we do with the endoscopes?’ Well, we can do the tumour and wide field resection with a subperiosteal plane, we can do septal resection, go up into the frontal sinuses complete fronto-ethmoidectomy (‘DRAF III’), median sphenoidotomy, we can take the bone adjacent to these tumours, we can take dura and periorbita and we can repair those defects all through the nose with no disfiguring facial incisions and the results are pretty good.

There are some things that we can’t do: orbital exenteration, maxillectomy, you need to excise skin since it can’t be done endoscopically, the anterior/lateral frontal sinuses are very difficult to access endoscopically through the nose, dura and brain lateral to the mid-line of the orbit, again difficult when you’re into brain parenchyma. But it’s good, it’s a shorter hospital stay, they get to the chemotherapy and radiotherapy more quickly and better outcomes, similar outcomes for conventional surgery, but much better weirdly in mucous melanoma. This is an audit of cases, they’re all national, they mean for all histological types, mean day stay was about just under five days, with very low risk – maybe epiphora and CSF leaks.

Quickly about the management of the orbit: really we care that the lamina is intact. If it is fine we can leave the orbit alone, if not ‘is the periosteum involved?’ The only way to reliably demonstrate this is with frozen section, unless is obviously through, so that’s what you need to do: to resect the periorbita, send it off and see is it involved with frozen section, take the eye if not. Here you can see some cases, so this is an adenocarcinoma. This is the MRI, you can see it’s kind of bulging into the orbit not doesn’t seem to be infiltrating and again, we removed it, the periosteum was clear on frozen section and the eye was preserved. Here an antral squamous cell carcinoma, this looks like it’s bulging through the periosteum a bit more and in the frozen section showed that it was involved, we had to clear the orbit.

I would like to say thank you for your attention. Thank you very much to Valerie Lund, David Howard, for the use of some of the images. In response, I’d like to tell you about their book, which is fantastic, more than you’ve ever wanted to know about nose, sinuses, and the nasopharynx.

Three things you could take away: remember endoscopic sinus surgery has a role, the MDT is vital, the brain and the eye are the limits.

Q & As section

Question 1
The literature demonstrates that orbital exenteration really may increase local control but not long-term survival, particularly with chemo-therapy. So what is the trend at this point in London?

Answer
I think at the moment if there is local control, I think that is still necessary. I think watching them suffer with increased symptoms in the sinuses makes a lot of sense, but again patient preferences have a role, we have to give the option. It’s difficult for patients to realise how unpleasant it is to have a fungating sinus cancer in their head and I find it difficult to express that to them, but I think it has a role still.

Question 2
You alluded to ‘en bloc’ and peaceful resection, any more thoughts on that to say where we’re going?

Answer
So for ‘en bloc’-ing ‘piece-meal’ resection, I think that when the endoscopy management of the sinonasal malignancy really started, there was a question about your going through the tumour ‘Are you seeding it anywhere else? Are you leaving bits behind?’ I think the evidence has really supported the use of this technique to sequentially move through the tumour to find its limits, rather than going around the outside. Obviously that any sort of classically trained cancer surgeon feel slightly uncomfortable about that. It’s something reassuring about going around the outside and getting it all but in the nose that’s just not possible; we don’t have margins in the nose if something is hanging into an air-filled cavity, you have to just take it up to its margin.