Transcript Role of Imaging in the Management of Oral Cancers

Control #: 1372
Retromolar Trigone Cancers: The overlooked
ability of 16 section MDCT in evaluating bone
erosion and assessing resectability.
eEdE-90-8913
Dr. Supreeta Arya1,
Dr. Peter Paul1,Dr. J. P. Agarwal2
1. Dept of Diagnostic Radiology
2.
Dept of Radiation Oncology
Tata Memorial Centre
Mumbai, INDIA
Disclosure of Commercial Interest
Neither we nor our immediate family members
have a financial relationship with a commercial
organization that may have a direct or indirect
interest in the content.
Purpose
1. To demonstrate the normal anatomy of the
retromolar trigone (RMT)
2. To depict the ability of multidetector CT
(MDCT) to accurately evaluate bone erosion and
soft tissue extent in RMT cancers.
using axial sections and multiplanar reformations
(coronal and oblique) with puffed cheek
technique on a 16 slice MDCT scanner
RETROMOLAR TRIGONE -Anatomy
“Retromolar trigone” is the triangular mucosal area behind the
last molars on each side 1
Fig 1. Retromolar trigone- Graphical representation of frontal view.
Courtesy- http://www.headandneckcancerguide.org
RMT -Anatomy
•
RMT overlies the ascending
ramus of the mandible 1
•
Behind the RMT is the
Pterygomandibular raphe
(superiorly attached to
pterygoid hamulus and
inferiorly to posterior end of
mylohyoid line) 2
Figure : courtesy Arya S, Rane P, Deshmukh A.
Clin Radiol. 2014 Sep;69(9):916-30 2
Fig 2. Sagittal line diagram of oral cavity
shows the RMT (shaded region shown by
arrowhead), Pterygomandibular raphe
(long arrow) that is attached to the
Pterygoid hamulus (*) and Mylohyoid line
(dotted line)
SUBSITES OF ORAL CAVITY
The 7 subsites are—
^
*
1. Lips
2. Buccal mucosa ( )
3. Upper alveolus () & lower
alveolus() with gingiva
4. Hard palate (^)
5. Tongue- anterior 2/3rds ( *)
6. Floor of mouth (│)
7. Retromolar trigone (not
shown—depicted in next slide)
Fig 3.Coronal reformation on 16 slice MDCT scanner using puffed
cheek technique (requires patient to blow uniformly through pursed
lips during quite breathing while separating tongue from hard palate)
Clinical examination versus Imaging
•Clinical evaluation is
difficult and incomplete
because of the posterior
location of RMT, dentition
and trismus if present 1
•Entire upper to lower limit
of RMT can be visualized on
the Oblique reformation on
a 16 or higher section MDCT
scanner using puffed cheek
technique 1
•Puffed cheek technique is
possible even in those
patients where trismus is
present 1
Fig 4. Visualization of RMT mucosa and the
contiguous raphe on an oblique reformation -requires use of puffed cheek technique that
can separate the buccal and gingival mucosa
with air that helps define the anterior margin
of the RMT .
Also seen is the ascending ramus of mandible.
Retromolar trigone Anatomy on
axial CT sections
A
B
Fig 5. Axial CT sections, A. shows the upper limit of RMT behind the
maxillary tuberosity (ellipse)
B. Shows lower limit behind the last mandibular molar ( arrow)
Relation of RMT with Pterygomandibular
raphe & Pterygopalatine fossa
*
Fig 6. 1 Sagittal bone algorithm reformat : shows
the pterygoid hamulus () that gives
attachment to the pterygomandibular raphe,
* shows the greater palatine foramen anterior to
it leading to greater palatine canal ( ) and
further to the pterygopalatine fossa () .
Perineural spread along V2 occurs along this
path.
Fig 7. Axial CT section at the level of base
skull shows the pterygopalatine fossa ()
between posterior wall of maxillary sinus
and base of pterygoid plates
RMT squamous cell cancers (SCC)
• Common in South East Asia , risk factors being betel quid
chewing, tobacco and alcohol 2
• Relatively rare site of oral SCC in the West ( where it forms
7% of oral cancers3,4)
• RMT cancers usually present late due to indolent growth3,4
• The importance of RMT SCC lies in the complex spread
patterns that need recognition by the radiologist to guide
the clinician to optimize management
How RMT cancers spread
 Invades structures in close
proximity 3-5
• Tonsil and anterior tonsillar
pillar (  )
• Buccal mucosa is in continuity
anteriorly ( )
• Closely abuts ascending ramus
of mandible (#)
• Maxillary alveolus and inferiorly
the body of mandible (  )
• Medially related to base tongue
and inferiorly to Floor of
mouth(  )
• Posteriorly related to Masticator
space ( black arrow shows
medial pteryoid muscle)
•
Perineural spread via V2 (shown
in slide 9 & V3 shown in slide
18)
Fig- Courtesy Arya et al. Imaging in oral
cancers. IJRI 20125
#
Fig 8. Pterygomandibular raphe
shown by * is not seen on
imaging -- superior spread along
this extends into the
pterygopalatine fossa ( shown
earlier in Fig 6 & 7)
RMT SCC-Imaging Methods
MDCT with puffed cheek technique &
reformations
• High accuracy for bone
invasion1,6-9
• Adequate for soft-tissue extent
• Perineural spread may be missed
when early
• Dental amalgam artifacts can be a
problem
Contrast enhanced MRI
• Accurate for T staging and
relations10
• Perineural spread-- more
accurate than CT
• Problem- Can overestimate
mandibular cortical invasion and
inferior alveolar nerve invasion 7
• Problem of motion and
swallowing artifacts
RMT SCCs- CT features
Contrast enhanced CT (CECT)-soft tissue algorithm :
1. Mildly enhancing infiltrative mass in the region of the RMT.
2. Loss of adjacent fat planes may be a diagnostic clue.
3. Very small lesions may not be picked up due to opposition of mucosal surfaces.
'Puffed-cheek' technique helps to detect small obscure lesions 1,5.
Bone algorithm CT with multiplanar reformations
Features s/o osseous involvement are--
• Cortical erosion --- seen as cortical discontinuity adjacent to the enhancing tumor
(remember to look at cortex on bone window/algorithm image + soft tissue window to
see the enhancing tumor)1
• Marrow invasion ---seen as trabecular destruction with abnormal attenuation of bone
marrow
1,7
Perineural spread (remember to see both soft tissue and bone windows)
•Loss of normal fat density /excessive enhancement in various foraminae (foramen
ovale, mandibular foramen, greater palatine foramen) in early cases 2
•Foraminal widening or erosion in advanced cases 2
RMT SCC- MRI findings
2,7,10
T1W MRI:
•Tumor is usually isointense to muscle. (Clue - Asymmetry of
adjacent fat planes)
•Replacement of normal hyperintense fat signal indicates marrow
involvement .
T2W & STIR MRI: Tumor is usually hyperintense to muscle
T1W MRI +Contrast: Shows moderately enhancing infiltrative
mass, adjacent soft tissue spread and can show perineural spread.
Perineural spread is seen as increased enhancement along the course of the
nerve from mandibular foramen reaching foramen ovale ( V3) or along
greater palatine canal reaching pterygopalatine fossa (V2) . Retrograde
perineural spread can occur through the mandibular foramen along the
inferior alveolar nerve .
Spread Patterns
Fig 9. Axial contrast enhanced CT sections show in A. a large RMT SCC ( t )
at maxillary level extending into masticator space posteriorly ( black arrow) ,
tonsil medially ( block arrow) & retro-antral buccal space anterolaterally
(white arrow) .
B. shows superior extension into pterygopalatine fossa ( V2 invasion)
resulting in soft tissue/ tumor density (shown by black arrow); compare with
fat density on normal opposite side( white arrow).
Spread patterns
 Only 15-26 % RMT SCC
are confined to the RMT ;
rest spread to as many as
3 other subsites of oral
cavity or oropharynx 11
*
 80- 84% extend to
anterior tonsillar pillar
( ATP) - part of oropharynx,
resulting in labelling of these
as RMT-ATP SCC 11
Fig 10. Axial CT shows RMT SCC ( * )
spreading to the ATP and tonsil (arrow)
Differences between RMT and ATP SCC
11
It is important to distinguish between ATP and RMT SCCs as their
growth patterns & prognosis differ
RMT SCC
ATP SCC
RMT -Subsite of oral cavity
ATP-Subsite of oropharynx
Higher incidence of 2nd primary in
aerodigestive tract
Lower
Bone invasion is more frequent( 12-53%
for mandible & 3-22% in maxilla)
Less frequent
Spread to pterygopalatine fossa and
masticator space is more frequent in
RMT SCC
Less common
Incidence of neck node metastasis is
around 40%. RMT SCC preferentially
spread to level I B nodes.
ATP SCC preferentially spreads
to level II nodes.
Spread patterns- Perineural spread
Fig 11. A. Coronal CECT shows SCC ( * ) spreading to the foramen ovale
along V3 (arrow). Examine region from mandibular foramen to foramen
ovale B. Bone window shows foraminal widening (arrow)
Fig- Courtesy Arya et al. Imaging in oral cancers. IJRI 20125
Spread Patterns– tongue muscles and
mandible erosion
Fig 12. Axial CT sections in same patient shows the tumor extension at its
lower limit as in
C. Medial extension into tongue muscles ( white arrow) and across the
gingivobuccal sulcus into buccal mucosa ( black arrow)
D. Bone algorithm image shows erosion of the anterior border of mandible
(arrow)
Mandibular assessment with MDCT
 16 slice MDCT scanner & above can provide very thin sections (
0.625 mm to 0.75mm)
 Thin sections along with bone algorithm reconstructions make
possible high resolution reformations
 As the mandible is curved, sagittal reformation cannot display
the entire body and the inferior alveolar canal
 Oblique reformation best depicts the alveolar crest as well as
almost entire inferior alveolar canal in RMT SCC 1
 Coronal reformations in bone algorithm is best for measuring
depth of erosion 1
 Coronal, oblique and curved planar reformations with 16 or
higher section CT obviate the need for Denta scan software 1
Oblique reformation—
Bone window Vs Bone algorithm
*
*
A
B
Figure 13. Oblique reformations on 16 slice MDCT with puffed cheek
method. Shows eroded alveolar crest and invaded inferior alveolar canal
A. Bone window image on the soft tissue algorithm
B. Bone algorithm image shows a RMT SCC (*) well outlined by air in
the buccal vestibule, invading cortex and marrow reaching upto inferior
alveolar canal (arrows). Note the higher resolution of the cortex in B.
Figure : courtesy Arya et al , Clinical Radiology 2013 1
Bone erosion – Role of Oblique reformations
A
BB
B
c
D
A
C
Figure 14 .Oblique reformations on
16 slice MDCT
A. Subtle erosion of alveolar crest
(arrow)
B. Irregular pattern of erosion of
alveolar crest (arrow) extending
above to vertical ramus.
C. Lysis around root of third molar
(short arrow) with erosion of alveolar
crest (long arrow). This was false
positive, cause of erosion being
infective.
D. Smooth scalloping of cortex
(arrow) without discontinuity (due to
pressure of tumor). This was
negative on HP.
Hence use of oblique reformation
increases specificity for mandibular
invasion.
Figure : courtesy Arya et al , Clinical
Radiology 2013 1
Evidence for accuracy of MDCT In RMT SCC
Using ad hoc generated oblique reformations in addition to axial images +
coronal reformations,
The reported sensitivity, specificity and accuracy of 16 slice MDCT for
• Mandibular cortical erosion = 94%, 90% and 91.8% respectively.
• Marrow invasion = 83%, 92% and 89% respectively.
• Accuracy for inferior alveolar canal invasion = 100%.
Reference - Arya S, Rane P, Sable N, Juvekar S, Bal M, Chaukar D.
Retromolar trigone squamous cell cancers: A reappraisal of 16 section MDCT for
assessing mandibular invasion. Clin Radiol. 2013 Dec;68(12):e680-8.
How to generate oblique reformats ad hoc
A
Sagittal
Figure 15 A-D.
Evaluation of the RMT on
multiplanar reformations
with triangulation on a
GE volume viewer (AW
suite). The spatial cursor
is seen in the left RMT in
all images.
Axial
Oblique
Obliq
ue
B
Axial
C
D
Coronal
Figure : courtesy Arya et al , Clinical Radiology 2013 1
The oblique reformat is
generated by placing the
spatial cursor in the diseased
RMT on the axial image (Fig
15 B) and rotating the
oblique image ( fig 15A) on
the work station/ volume
viewer.
Figure 15A shows the final
plane of oblique reformation
with the entire curved RMT
region seen in one section .
Table 1. Features to be studied on axial images & multiplanar reformations
(bone and soft tissue algorithms) on16 slice MDCT in RMT SCC for
evaluating bone invasion
Axial images
Coronal reformations
Oblique reformations
Sagittal
reformations(for
upper limit of RMT)
Anterior cortex of
vertical ramus
Superior alveolar
margin
Superior alveolar
margin
Posterior cortex of
maxillary alveolus
Buccal cortex
Inferior alveolar canal
Anterior cortex of
vertical ramus
Greater palatine
canal
Lingual cortex
Depth of erosion
Inferior alveolar canal
Pterygoid hamulus
Posterior margin of
maxillary alveolus
Buccal cortex
Length of erosion
Pterygopalatine fossa
Lingual cortex
Depth of erosion
Tooth sockets
Mandibular foramen
Mental foramen
Mandibular foramen
Pterygopalatine
fossa
Tooth sockets
Greater palatine
foramen
Greater palatine canal
Tooth sockets
Reference –Arya et al , Clinical Radiology 2013
Management of RMT SCC– Principles
 Goals of Management
a) Curative treatment of primary & nodes
b) Preservation of function & cosmesis  Reconstruction
 Curative treatment
RMT SCC can be managed by one of following -i) Only Surgery
ii) Only Radiotherapy ( RT)
iii)Surgery followed by RT +/- Chemotherapy
3-4,11-12
Management of RMT SCC
3-4,11-12
 Early lesions --uni-modal/ combined modality ( increasing
evidence that latter improves locoregional control)
 Advanced lesions treated by combination modality
 * When clinical/ imaging evidence of bone involvement seen,
surgery is mainstay (RT is high risk due to osteonecrosis)
RMT SCC –treatment of Neck
3-4,11-12
 If metastatic nodes present -- Modified radical neck dissection

If neck is clinically and imaging negative (N0) 
a) Selective neck dissection ( levels I-III) if primary treated with
surgery
a) Neck irradiation if primary treated with RT
Teaching Points–
1) Demonstration of metastatic nodes can influence extent of neck
dissection
2) Bulky nodes at multiple levels necessitate postoperative radiation ±
chemotherapy
Imaging in RMT cancers--- Why?
1,2,5
 Usually not for detection except in cases of trismus where
clinical examination is difficult
 For accurate staging
 T staging--- a ) Bone erosion & b) Soft tissue spread
 N staging
Teaching points
•Clinical evaluation of bone erosion unreliable & requires imaging
• Deep posterior soft tissue extent is revealed only by imaging
Oral Cavity SCC- AJCC 7th edition TNM staging12
T staging
Neck Nodes
Tx
Nx Regional LN cannot be assessed
Primary tumour cannot be assessed
T0
No evidence of primary tumour
N0 No regional LN metastasis
Tis
Carcinoma in situ
N1 Ipsilateral Single node < 3cm
T1
Tumour 2cm or < in greatest diameter
N2a Ipsilateral Single node 3-6cm
T2
Tumour > 2cm , < 4 cm in greatest
diameter
Tumour > 4cm in greatest diameter
N2b Ipsilateral multiple nodes <6cm
T3
T4a (oral cavity) - Tumor invades adjacent
structures (eg, cortical bone, deep extrinsic
muscle of tongue, maxillary sinus, skin of
face).
T4b - Tumor invades masticator space,
pterygoid plates, or skull base and/or encases
internal carotid artery.
N2c
Bilateral/Contralateral
nodes<6cm
N3 Lymph node > 6cm
Role of Imaging in staging and management
Bone erosion
•
Presence or absence
influences decision
about mandibular
resection ---Yes/ No
Extent of erosion
influences extent of
resection
(marginal, segmental or
hemi-mandibulectomy)
•
Soft tissue spread
•
Influences resectability and
extent of resection.
1- 2
Bone erosion—planning resection
1 ,3-4,13
MARGINAL MANDIBULECTOMY
(MM)
SEGMENTAL MANDIBULECTOMY
(SM) / HEMI MANDIBULECTOMY (
HM)
When disease abuts mandible , but no
erosion seen– for oncologically safe
resection margins
When cortical erosion is extensive- SM
When cortical erosion is subtle and
does not invade marrow or inferior
alveolar canal
When inferior alveolar canal &/or
mandibular foramen are involved-HM
Contraindicated in irradiated or
edentulous mandibles
When soft tissue component is large
with significant posterior spread to
masticator space( even without
erosion)– debated indication.
Soft tissue spread-- When to Resect? 14
Fig 16. A. Axial CT
at level of
mandibular notch
between coronoid
process (thick
arrow) & condyloid
process ( thin
arrow)
 T1-T3 & T4 a disease are
resectable
 A subset of T4 b with
extension to masticator space
below the mandibular sigmoid
notch# is resectable (in South
Asia where disease is
prevalent)
 T4 b disease above
mandibular notch, and
disease encasing carotid
artery –unresectable
*
Fig 16 B. Oblique
reformat on 16
slice MDCT
scanner.
Horizontal black
line through
mandibular notch
divides masticator
space into a high
(*) & low
compartment
# - Notch between coronoid and condyloid process ( black line in Figure 16 A & B) .
High and low masticator space as related to oral cancer
spread 14
Figure 17 .shows high and low
masticator space (MS).
T
Lp
m
m
p
Dashed yellow lines depict the entire
MS. Horizontal yellow line is at
approximate level of mandibular
notch
The part of MS above this line (high
MS) consists of the lateral pterygoid
(Lp) and temporalis (T) muscles .
The part of MS below the notch ( low
MS) has bulk of the masseter (m) and
medial pterygoid muscles (mp) .
T4 b disease extent
Revealed only on imaging!
Unresectable
2
Resectable
A
*
Fig 18 A. High masticator space disease
extension (arrow) reaching skull
base, invading lateral pterygoid
muscle. Horizontal line shows the
approximate level of mandibular
notch. Bulk of medial pterygoid (*)
muscle is below this level.
*
Fig 18 B. Low masticator space disease.
Arrow shows destruction of anterior border of
vertical ramus of mandible with invasion only
up to anterior margin of medial pterygoid
muscle (* ).
Conclusion
 Contrast enhanced MDCT is a good first line choice for bone
evaluation and soft tissue extent in RMT SCC— “a one-stop
shop” imaging method.
 The speed of MDCT scanning is added advantage ( < 25
seconds).
 MRI reserved for problem solving such as—
a) Those presenting with pain and sensory symptoms that
suggest perineural spread
b) Extensive trismus with inability to perform adequate puffed
cheek
c) Whenever CT is equivocal for precise soft issue extent
Imaging Check list for RMT
SCC
Primary– Soft tissue
•
Tumor dimensions and thickness (
transverse extent)
Adjacent structures ( more
important than T stage)
 Buccal mucosa, gingiva
 Base tongue, Floor of mouth
 ATP & Tonsil
•
 *Masticator space including low or
high space extension
 *Pterygopalatine fossa (PPF)
 Perineural spread via V2 (greater
palatine canal to PPF) and V3
(mandibular foramen to foramen
ovale )
*--High masticator space and invasion of
pterygopalatine fossa are
unresectable
Primary– Bone
• Bone erosion ( maxilla &
mandible) ---present or
absent
• Mandible *– extent of
erosion ( length & depth),
invasion of marrow ,
inferior alveolar canal &
mandibular foramen.
*Use– thin sections, soft
tissue and bone algorithm,
axial images, ad hoc coronal
& oblique reformats with
spatial cursor & puffed cheek
technique!
Imaging Check list for RMT
SCC
Nodes




*Level of abnormal nodes
Size & number
Presence of necrosis
Extracapsular spread
 Invasion of IJV
 Relation to common carotid artery and internal carotid artery
(circumferential contact , if < 180° –easily resectable,
if > 270° , unresectable)
Teaching points
 Understanding anatomy of RMT and spread patterns of
malignancies in RMT assumes significant relevance as the
site has limited clinical accessibility.
 Bone erosion, soft tissue extent and nodal status are
important as prognostic markers as well as for
management in RMT SCC
Teaching points
 Contrast enhanced MDCT with puffed cheek technique is a
good first line investigation for all the above. The thin
sections with MDCT afford high quality reformations ,
together with bone algorithms and puffed cheek need to be
exploited to a maximum.
 MRI, even though useful in better delineation of soft tissue
invasion and perineural spread, can be set aside as
problem solving tool rather than primary imaging modality;
with CT being a faster and cost-effective imaging modality.
References
1. Arya S, Rane P, Sable N, Juvekar S, Bal M, Chaukar . D.Retromolar trigone
squamous cell cancers: A reappraisal of 16 section MDCT for assessing
mandibular invasion. Clin Radiol. 2013 Dec;68(12):e680-8.
2. Arya S, Rane P, Deshmukh A. Oral cavity squamous cell carcinoma: Role of
pretreatment imaging and its influence on management. Clin Radiol.
2014;69 (9):916-30.
3. Genden EM, Ferlito A, Shaha AR, Rinaldo A. Management of cancer of the
retromolar trigone. Oral Oncol. 2003 Oct;39(7):633-7
4. Ayad T, Guertin L, Soulières D, et al. Controversies in the management of
retromolar trigone carcinoma. Head Neck. 2009 Mar;31(3):398-405.
5. Arya S, Chaukar D, Pai P. Imaging in Oral cancers. Indian J Radiol Imaging.
2012 Jul;22(3):195-208.
6. Mukherji SK, Isaacs DL, et al. CT detection of mandibular invasion by
squamous cell carcinoma of the oral cavity. AJR Am J Roentgenol , 2001
;177:237–43.
7. Imaizumi, N. Yoshino, I. Yamada, et al.. A Potential Pitfall of MR Imaging for
Assessing Mandibular Invasion of Squamous Cell Carcinoma in the Oral
Cavity .AJNR Am J Neuroradiol 2006 ;27: 114-122
References
8. Vidiri A, Guerrisi A, Pellini R et al. Multi-detector row computed tomography
(MDCT) and magnetic resonance imaging (MRI) in the evaluation of the
mandibular invasion by squamous cell carcinomas (SCC) of the oral cavity.
Correlation with pathological data. J Exp Clin Cancer Res. 2010 :29:7
9. Handschel J, Naujoks C, Depprich RA ,et al. CT- scan is a valuable tool to
detect mandibular involvement in oral cancer patients. Oral Oncol. 2012:
48(4):361-6.
10.Crecco M, Vidiri A, Angelone ML .Retromolar trigone tumors: evaluation by
magnetic resonance imaging and correlation with pathological data. Eur J
Radiol. 1999 Dec;32(3):182-8.
11.Antoniades K, Lazaridis N, Vahtsevanos K, et al. Treatment of squamous cell
carcinoma of the anterior faucial pillar-retromolar trigone. Oral Oncol. 2003
Oct;39(7):680-6.
12.NCCN Guidelines Version 1.2012, Cancer of the Oral Cavity, National
Comprehensive Cancer Care Clinical Practice Guidelines in Oncology (NCCN
Guidelines).NCCN.org.
13.Brown JS, Kalavrezos N, D'Souza et al. Factors that influence the method of
mandibular resection in the management of oral squamous cell carcinoma.
Br J Oral Maxillofac Surg. 2002 Aug;40(4):275-84
14.Liao CT, Ng SH, Chang JT, et a., T4b oral cavity cancer below the mandibular
notch is resectable with a favorable outcome. Oral Oncol. 2007 43(6):570-9.