Management of Facial Palsy following Head and Neck Cancer Therapy

June 26, 2024

ID11836

To CiteDCA Citation Guide

00:00Hi, good morning, everyone.
00:02Thank you and welcome to Yale
00:05Cancer Center Grand Rounds.
00:07My name is Sarah Omar.
00:08I'm the Chief of Head
00:09Neck Surgery at at Yale.
00:10And it's my distinct pleasure to
00:13introduce our speaker for today,
00:16one of my colleagues in ENT head
00:18neck surgery, Doctor Suresh Mohan.
00:21So Doctor Mohan is assistant professor in
00:24surgery in otolaryngology here at Yale.
00:27He earned his medical degree
00:29from Brown University.
00:30He did a one year research fellowship
00:32at the NIH and then he actually did
00:35A7 year research track residency
00:37at the Massachusetts Eye and Ear
00:39Infirmary and Harvard Medical School.
00:41Normally ENT is a five year residency
00:44program and so he did an extra 2 years
00:47dedicated to research while in training.
00:51Then not only that,
00:52he went on to do a fellowship in
00:54facial plastics and micro vascular
00:57reconstructive surgery at UCSF,
01:00a very renowned program in
01:02this field here at Yale.
01:04Dr.
01:05Mohan's clinical expertise is
01:06focused on a lot and everything
01:09to do with facial reanimation.
01:12For patients with facial paralysis,
01:14that includes dynamic and static reanimation.
01:17He also is a force in microvascular
01:20reconstruction for patients with head
01:23and neck cancer as part of our team.
01:25And he's also expert in MO's reconstruction
01:28for patients with skin cancers.
01:30He also another part of his practice has to
01:33do with functional aesthetic rhinoplasty,
01:35facial trauma,
01:36gender affirmation surgery,
01:38and the cosmetic treatment of the aging face.
01:41In terms of research,
01:42which she's going to talk to
01:44us a lot about today,
01:45it's all related to clinical
01:47outcomes in facial policy and nerve
01:50repair through the understanding
01:52of molecular foundations of
01:54peripheral nerve regenerations.
01:56He has won numerous awards for his
01:58research and teaching and has over 30
02:01peer reviewed publications to date.
02:03And he just joined the faculty
02:04less than a year ago.
02:07It's rare that we bring in a
02:09new recruit to the Yale Cancer
02:10Center and Department of Surgery,
02:12where we're actually bringing something new.
02:15And that's what Doctor Mohan's
02:18recruitment was all about.
02:19He is bringing advanced facial nerve,
02:23facial paralysis reconstruction,
02:25functional rehabilitation
02:27to patients here at Yale.
02:29And it's, it's he's
02:31just getting started, but he's
02:32already very busy and we're very,
02:34very happy to have him here.
02:35And he's going to tell
02:36us all about the management of facial policy
02:38and head neck cancer therapy. Doctor Mohan,
02:45good morning. Thanks so much,
02:47Cyril, for the kind words.
02:48It's truly an honor and privilege
02:50to be invited to speak here.
02:52So thank you all for being here.
02:54I wanted to start into with
02:56talking a little bit how I got
02:59interested in facial palsy.
03:00I had the privilege of training
03:02with Tessa Hadlock at Mass Sioneer,
03:05who's one of The Pioneers and
03:08I would say gurus of facial
03:11nerve reanimation in the world.
03:13And I fell in love with this patient
03:16population because not only is it a
03:18condition that affects form and function,
03:20but truly psychosocial well-being overall.
03:24And especially in the context of
03:26dealing with a cancer diagnosis,
03:28having this life changing sequelae
03:31of treatment,
03:32I think really affected patients
03:33in a deep way.
03:34And being able to help with that seemed
03:37like a really special opportunity.
03:41So I wanted to start with what I hope to
03:46leave you with at the end of this hour.
03:48Facial palsy is not just a
03:51physical visible appearance change.
03:53It's a functional problem.
03:54It's a psychological problem
03:56for these patients.
03:58For the longest time,
03:59patients have been told there's
04:01nothing to do, I'm sorry,
04:02go to Boston or go to New York and no
04:05longer is that true in Connecticut?
04:06And I've heard many patients
04:07say that and over and over.
04:09And it's, it's a privilege to be able to,
04:11to bring this care here.
04:14And finally,
04:16facial reanimation surgery
04:18re establishes smile blink,
04:20symmetry of the face and self-confidence
04:24psychologically for patients.
04:26So all the sequela have head and
04:28neck cancer and its treatment.
04:30I would say facial palsy is one
04:31of the most devastating because
04:32of its broad reaching effects.
04:36I could describe to you in words or I
04:39could show you what facial palsy does
04:41to the cancer patient. This is Carol.
04:45She was diagnosed with a aggressive
04:48DLBCL of the neck that invaded the skull
04:51base and left her with facial palsy.
04:53You can see she can't raise her eyebrows,
04:55she can't close her eye,
04:56she can't breathe through her left nostril,
04:58she can't smile, she can't speak
05:00and she can't eat without drooling.
05:04The facial nerves has such broad
05:06functions in terms of expression,
05:08corneal protection,
05:09hair cell protection in the ear,
05:11breathing through the nose,
05:13chewing and speaking,
05:14and it's injury leaves patients with
05:17functional psychosocial impacts.
05:23This picture is special to me
05:24because I think in this expression
05:27Carol really depicts what this
05:29patient population goes through.
05:31Not only is she suffering from a cancer
05:33diagnosis, but the stigma of having
05:36facial palsy and trying to face the
05:39world with with this altered condition.
05:42So let's take a step back and talk
05:44about the facial nerve anatomy.
05:46Cortical projections from the motor cortex
05:48synapse on the facial nucleus in the ponds,
05:51and the facial nerve loops around
05:53the abducense nucleus and exits
05:55at the pontomedullary junction,
05:57enters the internal acoustic mediatus,
05:59traverses the temporal bone and exits
06:01the skull base at the Stalin masted
06:04foramen where it divides into several
06:07branches of muscles that supply
06:09different branches and nerves that
06:11supply different muscles of the face.
06:13Not only does it provide motor
06:16movement to these muscles,
06:18there are also branches to the greater.
06:21The greater superficial petrosal nerve
06:24provides preganglionic parasympathetic
06:26innervation to the lacrimal glands,
06:29and also then stupedius nerve provides
06:31protection of hair cells in the middle ear.
06:37When we look deeper at the
06:38facial nucleus in the brain stem,
06:40it's divided into several subnuclei.
06:42Each of these nuclei are correlated
06:45with a specific function,
06:48so the medial submuculus, for example,
06:50correlates to the auricular muscles,
06:51so on and so forth.
06:53And even along the course of the nerve,
06:55there's variable topography,
06:56and so different axons are found
06:59at different places in the nerve
07:01as it travels into the face.
07:03This has significant implications
07:06for what happens after injury to
07:08the nerve and how important the
07:11location of the nerve injury is.
07:13What causes facial palsy?
07:15There are many causes.
07:17Probably the most common that
07:19you've heard of is Bell's palsy.
07:20I will say,
07:21I'll take this moment to emphasize
07:23the importance of distinguishing
07:25between Bell's palsy and facial palsy.
07:29Facial palsy is the final diagnosis.
07:31Bell's palsy is a type of facial palsy.
07:33I think very commonly we hear
07:35people use them interchangeably.
07:36In fact,
07:37they're very different because the
07:39etiology determines the treatment.
07:42Of course,
07:42today we will focus on facial
07:44palsy associated with benign or
07:46malignant tumors and their treatment.
07:50What happens after the
07:51facial nerve is injured?
07:53The face is in normal condition,
07:55it suffers an insult which results
07:57in valerian degeneration of the
07:59nerve and the face goes flaccid.
08:03If the nerve remains intact,
08:05eventually regenerating axons traverse
08:07the nerve and the face recovers.
08:1130% of patients will go on to
08:13develop aberrant regeneration or
08:15overzealous regeneration that
08:16results in post paralytic facial
08:19palsy or what we call synkinesis.
08:21If the nerve was cut or if there's
08:23an ongoing insult such as a tumor,
08:25then the state of the face
08:27remains in the flaccid state.
08:31On the left you see one of our
08:34patients with flaccid facial palsy
08:36for several years nearly 20 and
08:38a patient on the right with post
08:40paralytic or synkinetic facial palsy.
08:41You can see these are very different
08:43conditions, both categorized as facial palsy.
08:47In the cancer patient,
08:48the flaccid state is much more common cause.
08:50Typically the tumor or the treatment
08:53for the tumor results in transection
08:56or complete destruction of the nerve.
08:58We define synkinesis here as
09:00abnormal involuntary movements
09:01associated with a voluntary movement.
09:04And this begins usually six
09:05months after an insult.
09:09Muscle has a lifespan, denervated
09:11muscle has a much shorter lifespan.
09:15And so when you injure the nerve to a muscle,
09:18you get 18 to 24 months to
09:20reinnervate that muscle before the
09:22muscle is no longer receptive,
09:24the neuromuscular junctions fade away
09:26and there's no ability to reinnervate.
09:29So when we have a a facial nerve injury,
09:32the sooner reconstruction happens
09:33the better the long term outcome.
09:39The way we think about the face is
09:42by dividing it into sub sub areas
09:44and each zone of the face has a
09:46specific treatment associated with
09:47it for both the flaccid state here,
09:50I know this is a busy slide,
09:51but we'll walk through it in detail
09:53for the flaccid state as well As for
09:56the paralytic post paralytic state.
10:02What are the most common head and neck
10:04neoplasms associated with facial palsy?
10:05From a benign perspective,
10:07we have vestibular swanomas, facial swanomas,
10:09geniculate hemangiomas and of course
10:11from a malignancy perspective,
10:13high grade parotid tumors
10:14such as mucadermoid, adenoid,
10:16cystic or aggressive cutaneous glamosyl
10:19carcinomas are the most common.
10:21And as we know,
10:22the most common treatment for
10:24parotid tumors is surgical resection,
10:26and therefore the most common 'cause a
10:30facial palsy of with treatment of parotid
10:34tumors is requires surgical repair.
10:37So what is our algorithm?
10:38How do we repair the facial nerve after
10:41resection of a parotid malignancy?
10:45If the parotid tumor can be removed,
10:47for example with benign tumors
10:48with no damage to the nerve,
10:49then of course no treatment is necessary.
10:52Sometimes there can be temporary neuropraxia
10:54of the nerve from from dissection and
10:57that will typically recover as long
10:59as there's not a a deep nerve injury.
11:03If the nerve is transected in the
11:05process of removing the tumor,
11:07then we have multiple options for treatment.
11:11Depending on which branches transected.
11:15Periocular work may be done,
11:17may be necessary for corneal protection.
11:19A primary repair of the nerve
11:20can be performed,
11:21or an interposition graft can be can be
11:25inserted if more distal branches need
11:27to be resected but the proximal stump
11:30of the facial nerve is still preserved.
11:33In that case,
11:34a static suspension is typically
11:37performed based on the age of the patient.
11:40Older patients need more support in the face.
11:44A nerve transfer can be performed
11:46if there's insufficient donor
11:49axons from the native nerve.
11:51And finally interposition grafts again,
11:54if if, if there is a distal stump to sew too.
11:57In the most radical of resections of of
12:01parotid tumors where the lateral temporal
12:04bone must be resected into the middle ear,
12:07the proximal stump may be absent.
12:09And in those cases we depend completely
12:11on nerve transfer procedures to
12:13reanimate the face and sometimes
12:15requiring free muscle.
12:16So
12:19I'm going to walk through
12:20each of these strategies and,
12:21and some of the surgical details
12:24and then I will go into some of my
12:26research at the at the end of the talk.
12:30So we'll begin with eye care,
12:31protecting the eye,
12:33the cornea is the single most biggest
12:36priority in in treating facial palsy.
12:38And so when we do these
12:41radical parotid resections,
12:42staging the periocular work is
12:45helpful because we don't necessarily
12:47know the amount of of tightening
12:50of the upper eyelid or the lower
12:51eyelid that needs to be done.
12:52And so I'd like to stage this
12:54work until one to two weeks after
12:56the initial ablation so that the
12:58correct lid weight can be placed or
13:00the correct amount of tightening
13:01of the lower lid can be done.
13:03From a static perspective perspective,
13:05we can we teach patients to do
13:09lid stretching and eye patches.
13:10So this is the the lowest option
13:14on the totem pole.
13:15When patients develop paralytic lag of
13:18thalamus from injury to the facial nerve,
13:20they can't close their eye usually
13:22because the upper eyelid is unable to
13:24come down low enough to cover the cornea.
13:26By stretching the upper eyelid
13:28on a consistent basis,
13:311 to 2mm of improvement can be
13:35obtained and moisture chambers are
13:36used to keep the cornea moist.
13:41Next step up from this is performing
13:43a lateral tarsoraphy and so this
13:45is basically a procedure to close
13:48the lateral aspect of the eye and
13:50this helps with corneal protection.
13:52The downside to this is it's very temporary,
13:55doesn't last very long and and
13:58patients can have irritation
13:59from the suture being there.
14:01There's also decreased palpable
14:03fissure with and so vision can be
14:06occluded and also the cosmetic
14:08appearance of half your eye being
14:10closed is is problematic for patients.
14:15The next option we have to close the
14:17eye is by adding weight to the upper
14:19lid so that it descends further.
14:21The way we do this is by using a
14:23test weight before the surgery to
14:26determine the exact amount of weight.
14:28It's usually around 1.2 to 1.4g and
14:31we used to do this with gold weights.
14:34You may have heard of that,
14:36but the problem with gold is
14:37that it's very reactive,
14:39can trigger inflammatory reactions.
14:42Also, it has a high density.
14:44And so nowadays we use platinum
14:46because it's lower density.
14:48So you can produce a thinner profile
14:50weight that's less likely to extrude
14:52and achieves the same effect.
14:54And this comes in both a chain or
14:57format or just a regular profile.
15:00And so here you create a pretarsal
15:02or supertarsal pocket,
15:03insert the weight and secure it.
15:07What about for the lower eyelid?
15:09The other way we can tighten the
15:10lower eyelid of something called
15:12a lateral tarsal strip.
15:13So here there's a lateral canthotomy being
15:16performed with an inferior cantholysis.
15:18And so the inferior tarsus is
15:21released away from the canthus
15:23and the edges are stripped,
15:25as the procedure name indicates.
15:29And then this is sutured to the Whitnalls
15:33tubercle or the prominence on the medial
15:36aspect of the lateral orbital wall.
15:38And by doing this,
15:39we're able to tighten the lid.
15:40And you can see here a patient
15:42who's had the procedure,
15:43he had Atropian before and
15:45able to achieve much better eye
15:48closure with with the tightening.
15:50And in some cases,
15:52the lateral tarsal strip is insufficient
15:54because they have medial lag athalmos
15:55in addition to the lag lateral portion.
15:58So in those cases,
15:59we'll take,
16:00we'll perform what's called
16:01a lower lid tarsal sling.
16:02And so you can take fascia
16:04from the leg and thread it,
16:05think of it as a belt for the lower lid.
16:08And so you secure it to the
16:11lacrimal bone medially.
16:12Here you can see it's being tacked
16:14to the bone medially and brought
16:17through between the anterior posterior
16:19Lomella across to the orbital
16:21rim laterally and secured there.
16:22And this improves the medial lag of thalamus.
16:28Everything we've talked
16:29about so far is static.
16:30Is there a way for the upper eyelid
16:33to blink again? Indeed there is.
16:35So dynamic blink reanimation has been
16:38growing in popularity in the last few
16:40years due to some novel techniques.
16:43So here what's done is a
16:44cross faced nerve graft.
16:46So a nerve graft is harvested from the leg,
16:47it's A and you find a suitable
16:50eye closure branch on the
16:52other side that has redundancy.
16:53You connect it to that nerve
16:55and then you tunnel across the
16:58forehead and bring it into the
17:00upper eyelid on the other side.
17:02In the second stage of the surgery,
17:03six months later,
17:04you come in with a free muscle graft and
17:07allow the nerves to grow into the muscle.
17:09And as this happens,
17:10the eye is able to blink once again.
17:14So this is a patient that came to me
17:16with flaccid facial palsy after a
17:18cerebellar hemangioblastoma resection.
17:20And so I performed the first stage.
17:23Here you can see the nerve is brought
17:25through into the upper eyelid.
17:26And she also had a a graft for
17:28a future smile or animation and
17:30the fascicles are splayed here.
17:32And so though in the second,
17:33her second stage is coming up
17:34in in six months.
17:39Free muscle transfer for blank re
17:40animation has also been described.
17:42A cross faced nerve graft can be placed
17:45and a free platysma flap can be taken
17:47and muscle strips placed in the upper and
17:50lower eyelids to reestablish blank as well.
17:52The problem with this technique is it's,
17:54it's very laborious and and the outcomes are
17:58not nearly as good as the dynamic blink.
18:00We just, we just saw, OK,
18:03so we talked about the eye.
18:05What about the nerves that have been
18:07cut after the tumor was resected?
18:10So we can repair those nerves with
18:12either either primarily if they're
18:13close enough together or with a graft if
18:16there's too much space in between them.
18:19One of the fundamental tenets of nerve
18:22repair is attention free coaptation.
18:25This was a case from residency where
18:29a parotid cancer was resected and
18:33the superior division of the facial
18:35nerve was transected and the surgeon
18:37brought them together,
18:38brought the ends together and and sewed
18:40them as you can see in the top right.
18:43Unfortunately,
18:43it was under significant tension and
18:46it was eye opening to me because we
18:48actually took the patient,
18:49even though the nerve was repaired,
18:50we took the patient back to put in an
18:53interposition graft because of the tension.
18:55And so I looked more into why,
18:58why, you know,
18:59where did this tension concept come from?
19:02So it began very early canine
19:04studies in the 40s.
19:05This was a study where they used
19:09primary repair or interposition
19:10grafting and then they used
19:12inter arterial lead injection.
19:14So you can see on the right side here
19:18where they perform a tension high
19:21tension repair versus interposition
19:23graft and you just have much better
19:26vascularization of your distal
19:28segment inferiorly here with using
19:31the interposition graft.
19:32So this is kind of where the concept
19:35of avoiding tension in nerve repair
19:37began and and why it's so important
19:40in in in the face as well.
19:43There have since been other strategies
19:45I've been developed using nerve
19:47wraps or other ways to mechanically
19:50distribute the tension.
19:51As you can see,
19:52these wraps can be sutured at other
19:54parts of the nerve to minimize
19:56tension at the actual interface.
19:58Nerve lengthening procedures.
20:00While this sounds counterintuitive
20:02because we know neuropraxia is is
20:04damaging to nerves when stretching
20:06of the nerve is.
20:08Performed in a slow fashion over 20 minutes,
20:12you can actually take advantage
20:14of the elasticity of the nerve
20:17to to improve tension.
20:19And finally,
20:19this was a a product that came out
20:21recently where there are micro
20:23hooks in the in the wrap that can
20:25actually hold the nerve together
20:26to reduce tension as well.
20:30OK, so if we have nerves to repair,
20:33we've done that.
20:34What about for the facility of the face?
20:37So let's move on to the static suspension.
20:42So this was a gentleman who came
20:44in with a who who needed a radical
20:49temporal bone resection and protectomy
20:51with facial nerve sacrifice.
20:53And so for him we performed a,
20:56amongst other things, static suspension here.
20:57And so the way we do this is we take
20:59fascia from the thigh, the fascia lata,
21:01and then we cut it into strips and attach
21:04it to different points in the face.
21:06Here you can see to the nasal base,
21:08to the nasolabial fold and
21:10to the oral commissure here.
21:12And while it's not a dynamic movement,
21:15it does help with resting position.
21:18And here he is three weeks after.
21:22And functionally this improves drooling,
21:24that's probably one of the main things,
21:26and also breathing.
21:28So the nasal,
21:29the nose here collapse and collapses.
21:31And so being able to pull that valve
21:33laterally improves air flow through the nose,
21:35improves drooling and appearance.
21:39This is another gentleman who you can
21:42see on the left side here had a static
21:45suspension along with a direct brow lift.
21:48And you can see also how the the brow
21:50can really descend and cause dramatic
21:53Oasis and and vision occlusion as well.
21:55So both of these are improved
21:58important techniques.
22:01OK, so we've addressed the mid face,
22:02what about dynamic movement
22:05of the mouth and upper lip?
22:07So this is what takes us
22:09into our nerve transfers.
22:11And so when you don't have the native facial
22:13nerve driving the muscles of the face,
22:15you need alternate nerve sources.
22:17And so that's where nerve transfers come in.
22:20So we can use the contralateral facial nerve,
22:22which is great because
22:25it provides spontaneity.
22:26And then we can also use the trigeminal
22:28nerve on the same side or the opposite side,
22:3012th nerve,
22:31the ANSA or even the the
22:34spinal accessory nerve.
22:35So the crossface nerve graft,
22:37the way this works is so on.
22:38In this scenario,
22:39the patient has weakness on
22:41the left side of her face.
22:42And So what we do is we harvest
22:44a nerve graft again and open the
22:46face and find a branch that's
22:49redundant and connect it to here.
22:50So when the graft is connected and
22:52brought across the other side of the face,
22:55whatever this is connected to
22:56will have the same function.
22:58So it'll mirror the function
23:00of the healthy side.
23:01And so this is a very powerful
23:03technique that we, we,
23:05we will revisit here shortly.
23:07This can be done for the smile or,
23:09or the eye as we talked about earlier.
23:12The next is the mesoteric nerve.
23:14The mesoteric nerve is the branch that
23:17that control of V3 that controls the
23:20masculatory function of the master muscle.
23:22And you can see here that the
23:25nerve is basically because there's
23:27redundancy of the muscle.
23:28The patients don't have any
23:30donor site morbidity in terms
23:32of not being able to chew.
23:34The nerve is then connected to a
23:37distal branch of the facial nerve.
23:38And so when they bite down,
23:40instead of chewing, they can smile.
23:43What does that look like?
23:50And relax. And now give me a
23:53gentle smile, biting down.
23:56And relax, give me a big smile biting down
24:02and relax. So you can see here
24:06when the patient, one second
24:14the patient bites down,
24:15he's able to smile.
24:17And so the the reason is because the
24:21mesoteric nerve has excellent exonal load,
24:25nearly 1500 axons in the mesoteric nerve
24:28compared to the facial buccal branch,
24:30which only has about 800.
24:32And by the time we connect that
24:33branch to a cross faced nerve graft,
24:34there's only about 200 axons or 100
24:36axons that make it across the face.
24:43So this is him. He had a salivary
24:46ductal carcinoma and had a radical
24:47pyridectomy that left him with this
24:49appearance in the left side of the face.
24:51And again with the five seven transfer,
24:53he's able to re establish smile.
24:57This was a young child
24:59with an AV malformation.
25:02You can see she has weakness on the right
25:04side. After the five seven transfer,
25:11she bites down with teeth
25:17and she bites down close smile.
25:24So very powerful technique.
25:26Downside is it's voluntary,
25:28it's not spontaneous.
25:29You have to remember to bite
25:31down to be able to smile.
25:32And so an important part of it
25:34is the physical therapy that
25:36goes along with this in terms
25:38of making sure patients remember
25:39to use it and to bite down when
25:42they when they want to smile.
25:44Another example of a patient,
25:48OK, what are other nerves?
25:49So the hypoglossal nerve,
25:50the 12th nerve, is also a donor source.
25:53It can be used when the main trunk of
25:57the facial nerve has been transected
25:59and it's mostly useful for tone,
26:02but sometimes for movement.
26:04Of course, the risks are tongue weakness,
26:06dysarthria, dysphasia.
26:09Here was a patient who had a facial
26:11spinoma that was resected and outside
26:14hospital the facial nerve of course,
26:16was injured and there was no proximal stump.
26:18I need it to be identified and therefore
26:21the 12th nerve was transferred.
26:22And so you can see here for her,
26:25she has to move her tongue against the
26:27side of her mouth to be able to smile.
26:33This is her at rest
26:37and that's her pushing her
26:38tongue against her cheek.
26:44Again, voluntary is the downside.
26:48So this was a group from Spain who published.
26:52So the technique here is actually
26:53a side to end. So we don't take
26:55the entire hypoglossal nerve.
26:56So you maintain function of the nerve by
26:59only taking 1/3 of the axons and you're
27:02borrowing basically a portion of it.
27:04And with the endoside technique,
27:07So like we talked about for
27:09these voluntary nerve transfers,
27:10it's really important that,
27:11that we have facial rehabilitation
27:13that goes along with it.
27:14If they are not taught to use it,
27:16it's like anything else, they,
27:18it's, it's not functional and
27:19not useful to the patient if
27:21they're not trained to use it.
27:22So we have, we take facial
27:24rehabilitation very seriously.
27:25We have a strong partnership with
27:27here with our physical therapy
27:29department and we recently had
27:31a visiting lectureship as well.
27:32And, and so focusing on,
27:34on neuromuscular retraining of,
27:36of the muscles is, is a,
27:38is a important part of this.
27:41You can see here
27:44smile out by relaxing the tension in your eye
27:48and in your cheek and in your neck muscles.
27:51So form a small smile and at the same time
27:55you smile, you're gonna relax the tension
27:59around your eye and into your cheek,
28:03focusing on the cheek muscles
28:05forming the smile motion.
28:07I think of it like your cheek
28:09balls trying to form.
28:11There you go, that motion.
28:13Excellent.
28:16So you can imagine this takes a
28:18very cooperative patient and a
28:19very understanding therapist to
28:20make this that make this work.
28:24OK, so we've talked about
28:26static options so far.
28:27How about dynamic smile
28:29reanimation when there's no,
28:31when we're outside of the
28:32window for nerve transfers.
28:33So nerve transfers only work if the native
28:35musculature in the face is still alive.
28:37Like we talked about earlier,
28:38the native musculature only survives
28:40for 18 to 24 months after the injury.
28:42So if you're five years out,
28:45nerve transfers are not an option for you.
28:46So in that case, we need to bring in
28:49fresh muscle that can be innervated.
28:51And so that's what we do in this scenario.
28:53And so here there's multiple
28:54options of muscles that can be
28:56transferred for dynamic reanimation.
28:58The temporalis tendon can be brought
29:01down to connect to the oral commissure.
29:03Even the masseter muscle has been
29:05described to be transferred over.
29:06And from a functional muscle
29:09free flat perspective,
29:11many muscles have been used but the
29:14most common remains the gracilis muscle.
29:18So how does the temporalis
29:19transfer work here?
29:20Remember the temporalis muscle assists
29:22in in chewing and so when we flip
29:26the muscle down when it contracts
29:28and can pull on the oral commissure.
29:31Another way to do it was is with
29:33a orthodontic tendon transfer.
29:34In here the coronoid process is actually
29:40osteotomized and brought forth to
29:42connect to the oral commissure and this
29:45is accessed through a nasally beautiful
29:47incision or intraorally and this can
29:49cause smiling with biting down just
29:50like we would expect with the five 7.
29:54When we have to brew free muscle transfer,
29:56we go to the leg and we harvest a
29:59muscle flap with a nerve supply,
30:02the interruptory nerve as well as the
30:05blood vessels to the to the muscle.
30:08And previously they've already
30:10had a first stage crossface nerve
30:12graft placed for spontaneity.
30:14And so when you bring the muscle
30:16in and connect the blood vessels
30:18and connect the 5th nerve,
30:20you can get a nice smile.
30:22Again in a patient who's more than
30:24two years out from nerve injury.
30:31And in children,
30:32they have such remarkable nerve
30:33regeneration that you don't even
30:35need to go to the mesoteric nerve.
30:37Just across face nerve graft,
30:39despite the lower Axon load is
30:40sufficient to produce a smile.
30:42You can see here,
30:44this is her spontaneously smiling.
30:46And of course this is our gold
30:48standard outcome to be able to smile,
30:50not have to think about
30:51it and it just works.
30:55There are variations of the gracilis,
30:58not just using one nerve,
31:00but using both nerves using the mesoteric
31:02nerve and the crossface nerve graft,
31:04as you can see here, as well as
31:06the mesoteric nerve as an insurance
31:08policy if the crossface nerve graft.
31:10One of the downsides of this procedure
31:12is that while the spontaneous
31:14smile is our gold standard,
31:15the procedure fails nearly
31:1730 to 40% of the time.
31:19That is an unacceptably high
31:21rate in adults at least.
31:26There are also variations here where
31:28you can do intramuscular dissection
31:29and do different vectors of muscle.
31:32Of course the the facial,
31:33the smile mechanism takes nearly ten
31:36different muscles to engage a smile
31:39and so only one vector is is only.
31:42It's a small piece of the pie in
31:45terms of what allows natural smile.
31:48And so variations here,
31:49putting in a muscle graft to the lower
31:52lip can allow lower lip depression,
31:54which is normal when we smile.
31:59So of course with our cancer patients,
32:00what, what about after surgery?
32:03What about adjuvant radiation?
32:04How does that impact our, our repairs?
32:08There's not great data,
32:10mostly because facial reanimation still
32:13remains fairly uncommon at at the time of,
32:16of radical resection,
32:17though it's improving.
32:19This was a a small study that looked
32:22at 3039 patients and they found
32:25no significant difference in house
32:27Brackman outcomes between patients
32:29who had radiation and didn't.
32:32So it supports a theory that
32:36getting radiation should not
32:39preclude upfront nerve repair.
32:43Timing of facial nerve repair has
32:44also been brought up as an issue,
32:46but we know that time is muscle.
32:48So it's very clear now that the
32:51sooner you repair the nerve,
32:52the better outcome.
32:54But a recent study looked at almost
32:5811,000 cases of of parotidectomy
33:00and found that only 24 percent,
33:0325% are undergoing reanimation at
33:05the time of facial nerve sacrifice.
33:08So I think that is an indication of
33:11of how much room we have in terms
33:13of improving outcomes for patients.
33:19So as I mentioned earlier,
33:22the failure rate in this procedure
33:24is 30 to 40% and many theories
33:27have been proposed as to why.
33:29But one of the prevailing theories
33:31is that as these axons start from the
33:34contralarial side and are growing across,
33:37Schwann cells in the distal nerve graft
33:40lose their ability to support regeneration.
33:43And so by the time axons make it across,
33:46very few Schwann cells are alive to
33:49support and get the Axon into the muscle
33:52into the neuromuscular ejection junction.
33:54And because of that the failure
33:56and the the procedure fails.
33:58So this got me interested in thinking
34:01about what are these repair swan
34:02cells doing at a molecular level?
34:04What is the state that gets turned
34:07off and how do we keep that state
34:10for a longer period of time to
34:12maybe allow better regeneration?
34:14So when a nerve is cut,
34:15the distal nerve undergoes valerian
34:17degeneration and the Schwan cells
34:20that were myelinating Schwan cells
34:22take on a repair repair phenotype.
34:24And we know this is a transient state.
34:26So the repair swan cells come from either
34:29the myelinating differentiated cells
34:30or the non myelinating Schwan cells
34:33and they become repair Schwan cells.
34:35After injury,
34:35macrophages are brought in for
34:38breaking down the myelin debris.
34:40There's an inflammatory response.
34:45So this led to the question,
34:47can we exploit the repair Schwan
34:50cell response to injury to
34:52improve regeneration outcomes.
34:56So our objectives were first,
34:57what cells are in the facial nerve,
35:00Can we characterize this repair Schwan
35:02cell phenotype at a transcriptional
35:04level and can we potentially
35:07identify any therapeutic targets?
35:09So what do we do?
35:11So I used a mirroring facial
35:13nerve transection model.
35:14So as you see here in Figure 1,
35:18the right side of the face
35:19of the mouse is shown here.
35:21Of course the mouse facial
35:23nerves are sub millimeter,
35:25but we harvested.
35:26So first we opened the face and
35:29performed an injury and closed
35:31up on one side of the face.
35:33Five days later we came back,
35:35harvested the nerve from this
35:37the the distal nerve from the
35:39site of transection to the
35:41to the distal buckle branch.
35:43And then we harvested contralateral facial
35:45nerves that are healthy for controls.
35:48We took those nerves,
35:49dissociated them down and
35:50ran them through the single
35:53cell sequencing platform 10X.
35:56And then what did we find?
35:59So here we're looking at
36:00something called a fate plot.
36:02Each dot on here represents
36:04a cell and as you can see,
36:07the difference in each color
36:08represents a different cell type.
36:09As shown on the right.
36:11When you compare between the
36:12wild type or healthy condition
36:14and the injured condition,
36:15there's obviously a new
36:16cluster that has shown up.
36:18And this correlates with the repair
36:19Schwann cells that we'd expect
36:24when you do an ontology analysis.
36:26They have expected gene functions,
36:28isonagenesis guidance,
36:29and restructuring of the
36:32extracellular matrix as Schwann
36:34cells undergo conformational
36:36changes to allow for reinnervation.
36:39We do Ade analysis and we found
36:41several genes that were correlated
36:44with the repair phenotype.
36:48Some of these upregulated genes were
36:52established and some of them were novel.
36:57So to understand this better,
36:58we used a technique called RNA velocity.
37:02I wanted to be able to ideally would
37:03look at several different time points,
37:05but for anyone who's done
37:06single cell or any sequencing,
37:08you know, it's very expensive.
37:09And so there are bioinformatic ways of
37:12trying to understand gene trajectories
37:14without doing every time point that you have.
37:17I explain it this way,
37:20you see a picture,
37:21but what was happening before and after?
37:25Can you use what's in the picture to
37:28give you a sense of the trajectory?
37:31So that's what we did.
37:32And the way RNA velocity works is
37:35it basically takes the ratio of
37:37unspliced to splice transcripts and
37:40can predict the number of genes that
37:43are moving at that point in time.
37:45It's it's kind of like taking the
37:47time derivative of gene expression
37:49to get a sense of of trajectory.
37:51So when we look at this map,
37:53we see that in in teal are the
37:56myelining Schwan cells in in all of
37:58are the non myelin Schwan cells.
38:00And we see vectors of differentiation towards
38:03the repair swan cell phenotype in blue.
38:06In pink,
38:07we see the macrophage cluster also
38:09showing a differentiation as we'd
38:11expect because they have similar
38:13functions in in breaking down myelin.
38:17And when we zoom in here again,
38:19we see that five days post injury,
38:21there's a significant increase in
38:23the repair swan cell phenotype.
38:25So what are the genes that are
38:28driving this this transformation?
38:30We performed a driver gene analysis
38:32and identified a few different genes
38:34that that had the greatest impact.
38:38But we so far had been looking at
38:41individual genes. But we know that
38:42genes don't work in isolation.
38:44So we did a regulon analysis
38:46and a regulon is basically a
38:48regulatory network of genes.
38:50And that way we can understand
38:52what is impacting what in terms of
38:54downstream to upstream effects.
38:57And so with the regulon,
38:58we found regulon analysis,
38:59we found that the ATF 3 regulon
39:02had the highest specificity.
39:03So when we look at the cluster,
39:04we see here in the bottom right that this
39:08correlates with with our Schwann cell,
39:10a repair Schwann cell phenotype.
39:12So we realized that the ATF 3 regulon
39:14was very specific for this cluster.
39:20What are the genes in the ATF 3 regulon?
39:22It came up we were able to narrow down
39:25from thousands of genes that were involved
39:28in nerve injury to a couple to 16.
39:31And of these, Gal three was
39:33was the most expressed.
39:34But how do we know that these in silico
39:36findings are truly reflected in biology?
39:38So we went back and did validation
39:41with in situ hybridization and
39:44found that indeed Gal three's
39:46upregulated after injury in in our
39:51in our cell population as well.
39:53And our positive control was C
39:54June which we is a well established
39:56marker of injury in nerves.
40:00So in conclusion, we've performed
40:02profiling of the facial nerve cellulum.
40:04We've identified driver genes and
40:06found that the ATF 3 regulon and
40:08GAL three may be a targetable
40:10factor of the program and.
40:12Next steps would be trying to
40:14assess this in a more realistic
40:17cross faced nerve graft model.
40:19We have models with fluorescent
40:21reporter mice where we're able
40:24to harvest the nerves and look at
40:27specific transcriptional changes
40:29and then potentially create a
40:31knockout of the gall 3 gene and
40:33see impacts on nerve regeneration.
40:39I'd like to thank my collaborators
40:42and mentors and specifically
40:44my Yale order oncology family,
40:47Ben Judson, Sarah Mara,
40:49Zephyr Syed and Sarah Pay.
40:51Thank you all for for being here.
40:52And those who are not here,
40:54thank you as well.
40:55It's been a wonderful year so far and
40:58I'm excited to help grow this program.
41:01We have over 75 patients already in a year.
41:04We're rapidly growing as word gets out.
41:06We're getting a lot of referrals
41:07from people who were going to Boston,
41:09NY and want to bring their
41:11care back to Connecticut.
41:12We have several reanimation procedures going.
41:14We got multiple clinics,
41:16established workflows.
41:17We had AI mentioned the the physical
41:21therapy visiting lectureship.
41:23So it's a very exciting time at
41:25the facial nerve center and I'm
41:27very grateful for all the support.
41:29And I leave you again that facial palsy
41:33results in not just visual appearance,
41:35but functional and psychological
41:37quality for these patients.
41:39They have nearly every patient
41:41has an option for for improvement
41:43in their condition.
41:45And we have a bit ways to re establish,
41:47smile and blink.
41:50And I think that's all I have.
41:51I can take questions.
41:51Thank you so much,
42:00Doctor Sayed. Sorry. So, yeah,
42:02I want to really applaud you for all
42:04the amazing work you're doing and the
42:05fact that you're in The Apprentice here
42:07in Yale and to Connecticut at large.
42:10With regards to the gene expression,
42:12the GAL 3, for example,
42:14are you able to tell if there's differential
42:16expression across the nerve that you took,
42:19like proximal versus distal?
42:21Is that something that needs to be looked at?
42:23Yeah, absolutely. Yeah.
42:25That's so the question was,
42:28is the repair SWAN cell gene expression
42:31variable across the length of the
42:34nerve from distal to proximal segment
42:36and that's a very good question.
42:39We know that there are, you know,
42:41by definition the proximal segment
42:43is not injured.
42:44And so although there are changes
42:46at the level of the facial nucleus
42:48as soon as exotomy occurs.
42:49So short answer is no, I haven't done it yet.
42:53But yes, there are expected changes and
42:57I think it would be very easy to do.
42:59It's just multiple time points.
43:01And at the time I didn't have
43:02the resources to do that.
43:04But great question, Sarah.
43:05Great talk.
43:05Thanks so much.
43:06Yeah.
43:07I was wondering if you guys
43:09were going to look at the
43:12potential interacting ligand.
43:20Absolutely, Yeah, I think the next
43:22step would be trying to yes, sorry.
43:24So the question is have we done a lagging
43:28analysis for Gal three and looked at
43:31what's interacting And absolutely
43:33we've found several chemokines and
43:37I think there are some GPC Rs that
43:39are also involved with with Gal 3.
43:42And so it, it, I think our sample size
43:44was not high enough to be able to to get
43:47a significant data on the lag analysis.
43:49So we did, we kind of ditched it,
43:50but that is definitely on the table by Ray.
43:55Yeah, so amazing work for the patient
44:00who let's say has product tumor,
44:02you know that you're going to
44:04be doing chemo radiation are
44:08are there any particular considerations in
44:11our, let's say antivant treatment planning,
44:14are there particular nerve branches
44:15is the person who's contralateral neck
44:18that we should be paying attention
44:20to so you have the optimal nerve to.
44:24Yeah, that's a great question.
44:26Question was in the patient with a
44:29parotid tumor who needs is expected
44:31to need chemo radiation afterwards.
44:33Are there any considerations about what
44:35nerves to preserve or protect? You know,
44:38if we're doing upfront facial reanimation,
44:41then it doesn't really change. I think.
44:44I think the prevailing opinion has been like,
44:48do your rename management early and then
44:50whatever treatment has to they need to get,
44:52they need to get anyway.
44:54And so I don't think anyone has thought
44:58of a way to alter their management.
45:01The only thing that might be different is,
45:03sorry, you're saying do the
45:05reanimation and then radiate that.
45:08Yeah.
45:09So those transplanted nerves and yeah.
45:11And so it turns out,
45:12turns out that the the data on radiation of,
45:16of the nerves is not,
45:17is not as bad as initially thought.
45:20You know that for a long time,
45:21people are like, oh,
45:21if you're going to get radiation,
45:22don't bother with facial reanimation.
45:25And the outcomes are not that different
45:27between.
45:27And so that's why now there's a
45:29push to just do it up front because
45:31then patients end up not getting
45:33it or for whatever reason.
45:35So I think that's that's
45:39my strategy currently.
45:46OK, thanks so much. All right,
45:48thank you. Doctor Mohan,
45:48I have one more question for you.
45:50And as I as I moderate and end this.
45:54So it sounds like you have some great
45:58opportunities here at Yale for research,
45:59not not only basic science,
46:02but also in clinical outcomes
46:03like the question that was
46:04just asked by Doctor Burtness,
46:05you know, what are the outcomes?
46:07You know, we, I think there's
46:08great opportunity here with the high
46:10volume head neck cancer we have.
46:11Have you found any collaborators at
46:14Yale Cancer Center or at Yale in nerve
46:16regeneration or you looking for some?
46:19And how can people contact you if there
46:20are some people listening who might
46:22be interested in working with you?
46:25Thank you. Yeah, I,
46:26I promise I didn't plant him for that.
46:28But absolutely, I am very
46:30eager to continue this work.
46:32And I've been trying to find the
46:35right collaborators and just haven't
46:37exactly found found them yet.
46:39So please do reach out e-mail.
46:41I would be would be great.
46:42And I think there's an opportunity
46:44to be able to bring tissue from from
46:46the operating room and and really do
46:48this because so far all the work has
46:50been in my sense of being able to
46:51bring human tissue into the end and play.
46:53It would be great. Thank you.
46:56All right. Thank you, everybody.