Ci a ion: Cap ioli, S.; Casaleggio, A.;
Taglia ico, A.S.; Con o i, C.; Bo da, F.;
Fiannacca, M.; Filau o, M.; Iandelli,
A.; Ma chi, F.; Pa inello, G.; e al.
High-F equency In ao al Ul asound
o P eope a i e Assessmen o
Dep h o In asion o Ea ly Tongue
Squamous Cell Ca cinoma:
Radiological–Pa hological
Co ela ions. In . J. En i on. Res.
Public Heal h 2022,19, 14900. h ps://
doi.o g/10.3390/ije ph192214900
Academic Edi o : Ca melo Sa ani i
Recei ed: 10 Oc obe 2022
Accep ed: 10 No embe 2022
Published: 12 No embe 2022
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2022 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
In e na ional Jou nal o
En i onmen al Resea ch
and Public Heal h
A icle
High-F equency In ao al Ul asound o P eope a i e
Assessmen o Dep h o In asion o Ea ly Tongue Squamous
Cell Ca cinoma: Radiological–Pa hological Co ela ions
Simone Cap ioli 1,* , Alessand o Casaleggio 2, Albe o S e ano Taglia ico 2,3 , C is ina Con o i 2, Fabio Bo da 3,
Ma ina Fiannacca 3, Ma a Filau o 4,5,6, And ea Iandelli 4, Filippo Ma chi 4,5 , Giampie o Pa inello 4,
Gio gio Pe e i 4,5 and Giuseppe Ci adini 2,*
1Depa men o In e nal Medicine (DIMI), Uni e si y o Geno a, Viale Benede o XV 6, 16132 Genoa, I aly
2Depa men o Radiology, IRCCS Ospedale Policlinico San Ma ino, La go Rosanna Benzi 10,
16132 Genoa, I aly
3Depa men o Heal h Sciences (DISSAL), Uni e si y o Geno a, Via Pas o e 1, 16132 Genoa, I aly
4Depa men o O o hinola yngology, Head and Neck Su ge y, IRCCS Ospedale Policlinico San Ma ino,
La go Rosanna Benzi 10, 16121 Genoa, I aly
5
Depa men o Su gical Sciences and In eg a ed Diagnos ics (DISC), Uni e si y o Genoa, 16132 Genoa, I aly
6Depa men o Expe imen al Medicine (DIMES), Uni e si y o Genoa, 16132 Genoa, I aly
*Co espondence: [email p o ec ed] (S.C.); [email p o ec ed] (G.C.)
Abs ac :
The eigh h edi ion o he TNM classi ica ion o icially in oduced “dep h o in asion”
(DOI) as a c i e ion o de e mining he T s age in ongue squamous cell ca cinoma. The DOI is a
well-known independen isk ac o o nodal me as ases. In ac , se e al expe s s ongly sugges
elec i e neck dissec ion o ongue cance wi h a DOI > 4 mm due o he high isk o ea ly and occul
nodal me as ases. Imaging plays a pi o al ole in p eope a i e assessmen s o he DOI and, hence,
in planning he su gical app oach. In ao al ul asound (IOUS) has been p oposed o ea ly-s age
SCC o he o al ongue as an al e na i e o magne ic esonance imaging (MRI) o local s aging. The
aim o his wo k is o in es iga e he accu acy o IOUS in he assessmen o he DOI in ea ly o al
SCC (CIS, pT1, and pT2). A o al o 41 pa ien s wi h ongue SCCs (CIS-T2) unde wen a p eope a i e
high- equency IOUS. An IOUS was pe o med using a small-size, high- equency hockey-s ick linea
p obe. The ul asonog aphic DOI (usDOI) was e ospec i ely compa ed o he pa hological DOI
(pDOI) as he s anda d e e ence. In pa ien s who unde wen a p eope a i e MRI, hei usDOI,
magne ic esonance DOI (m iDOI), and pDOI we e compa ed. Speci ici y and sensi i i y o he
IOUS o p edic a pDOI > 4 mm and o di e en ia e in asi e and nonin asi e umo s we e also
e alua ed. A high co ela ion was ound be ween he pDOI and usDOI, pDOI and m iDOI, and
usDOI and m iDOI (Spea man’s
ρ
= 0.84, p< 0.0001, Spea man’s
ρ
= 0.79, p< 0.0001, and Spea man’s
ρ
= 0.91,
p< 0.0001
, espec i ely). A Bland–Al man plo showed a high ag eemen be ween he
usDOI and pDOI, e en hough a mean sys ema ic e o was ound be ween he usDOI and pDOI
(0.7 mm), m iDOI and pDOI (1.6 mm), and usDOI and m iDOI (
−
0.7 mm). The IOUS was accu a e a
de e mining he T s age (p< 0.0001). The sensi i i y and speci ici y o he IOUS o p edic a pDOI
≥
4 mm we e 92.31% and 82.14%, espec i ely, wi h an AUC o 0.87 (p< 0.0001). The speci ici y,
sensi i i y, nega i e p edic i e alue (NPV), and posi i e p edic i e alue (PPV) o he IOUS o
p edic an in asi e cance we e 100%, 94.7%, 60%, and 100%, espec i ely. The AUC was 0.8 (95% CI
0.646–0.908, p< 0.0001). The IOUS was accu a e in a p eope a i e assessmen o a pDOI and T s age,
and can be p oposed as an al e na i e o MRI in he p eope a i e s aging o ongue SCC.
Keywo ds: ongue squamous cell ca cinoma; dep h o in asion; in ao al ul asound
1. In oduc ion
O al squamous cell ca cinoma (SCC) is he mos equen head and neck neoplasm,
and he o al ongue is he mos common si e o p esen a ion [
1
]. Despi e inno a ions in
In . J. En i on. Res. Public Heal h 2022,19, 14900. h ps://doi.o g/10.3390/ije ph192214900 h ps://www.mdpi.com/jou nal/ije ph
In . J. En i on. Res. Public Heal h 2022,19, 14900 2 o 11
ea men , he p ognosis o ongue SCC is s ill di icul o p edic : e en hough o e all
su i al and disease-speci ic su i al a e sa is ac o y a ea ly s ages, he isk o lympha ic
dissemina ion is high and ep esen s he mos c i ical p ognos ic ac o [
2
]. Mo eo e , he
p e alence o occul nodal me as asis in a clinically nega i e neck anges om 8.2% o
46.3%, and mo ali y is inc eased i e- old i occul nodal me as ases occu [
2
–
5
]. In 2017,
he eigh h edi ion o he umo –node–me as asis (TNM) s aging sys em o he Ame ican
Join Cance Commi ee (AJCC)/Union o In e na ional Cance Con ol (UICC) o icially
in oduced he dep h o in asion (DOI) as a s aging c i e ion o he T s age along wi h he
su ace dimension in o al SCC [
6
]. The DOI is de ined as he dep h o he umo in asion
measu ed om he le el o he basemen memb ane o he closes no mal mucosa ollowing
an ideal “plumb line” [7] (Figu e 1).
In . J. En i on. Res. Public Heal h 2022, 19, x 2 o 13
1. In oduc ion
O al squamous cell ca cinoma (SCC) is he mos equen head and neck neoplasm,
and he o al ongue is he mos common si e o p esen a ion [1]. Despi e inno a ions in
ea men , he p ognosis o ongue SCC is s ill di icul o p edic : e en hough o e all
su i al and disease-speci ic su i al a e sa is ac o y a ea ly s ages, he isk o lympha ic
dissemina ion is high and ep esen s he mos c i ical p ognos ic ac o [2]. Mo eo e , he
p e alence o occul nodal me as asis in a clinically nega i e neck anges om 8.2% o
46.3%, and mo ali y is inc eased i e- old i occul nodal me as ases occu [2–5]. In 2017,
he eigh h edi ion o he umo –node–me as asis (TNM) s aging sys em o he Ame ican
Join Cance Commi ee (AJCC)/Union o In e na ional Cance Con ol (UICC) o icially
in oduced he dep h o in asion (DOI) as a s aging c i e ion o he T s age along wi h
he su ace dimension in o al SCC [6]. The DOI is de ined as he dep h o he umo
in asion measu ed om he le el o he basemen memb ane o he closes no mal mucosa
ollowing an ideal “plumb line” [7] (Figu e 1).
Figu e 1. Ca oon demons a ing how o measu e DOI in la (a), exophy ic (b), o ulce a ed lesions
(c). The e e ence line is he basal memb ane. DOI is measu ed pe pendicula ly o he basal
memb ane, including ulce a ed pa s o he lesion, bu excluding he exophy ic po ions. Tumo
hickness is g ea e han DOI in exophy ic lesions and smalle in ulce a ed neoplasms. Solid line =
DOI; dashed line = umo hickness; a ow = basal memb ane.
The DOI is a well-known p ognos ic ac o . In addi ion, se e al au ho s ha e
demons a ed a di ec co ela ion be ween he DOI and he incidence o nodal me as asis
[2,3,8,9]. Fo hese easons, elec i e neck dissec ion has been p oposed o ea ly-s age o al
ongue SCC. Fu he mo e, he DOI can be used o dic a e p ophylac ic neck dissec ion in
clinically N0 pa ien s. No wi hs anding, he e is no global consensus o he h eshold,
wi h a ange a ying be ween 3 mm and 10 mm [3,10,11]. Howe e , a DOI cu -o o 4 mm
was ecen ly p oposed by se e al expe s [12–14].
Imaging plays a pi o al ole in he local s aging o ongue cance by es ima ing he
DOI and guiding he su geon o plan he su gical in e en ion p ope ly and de ine he
need o an elec i e neck dissec ion. In ac , cT1 and selec ed cT2 can be sa ely emo ed
anso ally. In con as , mo e ad anced umo s equi e a pull- h ough app oach o ob ain
be e con ol o deep ma gins [15]. Compu ed omog aphy, magne ic esonance imaging
(MRI), and an in ao al ul asound (IOUS) can be used o assess he local and egional
ex en o o al cance (Table 1). MRI is now conside ed he i s choice o he p eope a i e
s aging o o al cance [16], bu he in e es in IOUS has been p og essi ely g owing o e
ime. Se e al s udies ha e epo ed he u ili y o an IOUS in he p eope a i e s aging o
o al ongue SCC [17]. In an ea ly s udy, Shin ani and cowo ke s compa ed IOUS wi h CT
and MRI using his ology as he gold s anda d. They epo ed ha ul asonog aphy was
supe io o CT and MRI in an assessmen o he p ima y lesion o o al ca cinoma, mos ly
because CT and MRI could no de ec he p ima y umo i he hickness was less han 5
mm [18]. IOUS is also used o in aope a i e umo hickness assessmen s and o imp o e
loco egional con ol du ing su ge y [19,20]. Howe e , mos s udies in he li e a u e
e alua ed IOUS in he assessmen o umo hickness [21–30].
Figu e 1.
Ca oon demons a ing how o measu e DOI in la (
a
), exophy ic (
b
), o ulce a ed lesions (
c
).
The e e ence line is he basal memb ane. DOI is measu ed pe pendicula ly o he basal memb ane,
including ulce a ed pa s o he lesion, bu excluding he exophy ic po ions. Tumo hickness is
g ea e han DOI in exophy ic lesions and smalle in ulce a ed neoplasms. Solid line = DOI; dashed
line = umo hickness; a ow = basal memb ane.
The DOI is a well-known p ognos ic ac o . In addi ion, se e al au ho s ha e demon-
s a ed a di ec co ela ion be ween he DOI and he incidence o nodal me as asis [
2
,
3
,
8
,
9
].
Fo hese easons, elec i e neck dissec ion has been p oposed o ea ly-s age o al ongue
SCC. Fu he mo e, he DOI can be used o dic a e p ophylac ic neck dissec ion in clinically
N0 pa ien s. No wi hs anding, he e is no global consensus o he h eshold, wi h a ange
a ying be ween 3 mm and 10 mm [
3
,
10
,
11
]. Howe e , a DOI cu -o o 4 mm was ecen ly
p oposed by se e al expe s [12–14].
Imaging plays a pi o al ole in he local s aging o ongue cance by es ima ing he DOI
and guiding he su geon o plan he su gical in e en ion p ope ly and de ine he need o
an elec i e neck dissec ion. In ac , cT1 and selec ed cT2 can be sa ely emo ed anso ally.
In con as , mo e ad anced umo s equi e a pull- h ough app oach o ob ain be e con ol
o deep ma gins [
15
]. Compu ed omog aphy, magne ic esonance imaging (MRI), and
an in ao al ul asound (IOUS) can be used o assess he local and egional ex en o o al
cance (Table 1). MRI is now conside ed he i s choice o he p eope a i e s aging o o al
cance [
16
], bu he in e es in IOUS has been p og essi ely g owing o e ime. Se e al
s udies ha e epo ed he u ili y o an IOUS in he p eope a i e s aging o o al ongue
SCC [
17
]. In an ea ly s udy, Shin ani and cowo ke s compa ed IOUS wi h CT and MRI
using his ology as he gold s anda d. They epo ed ha ul asonog aphy was supe io o
CT and MRI in an assessmen o he p ima y lesion o o al ca cinoma, mos ly because CT
and MRI could no de ec he p ima y umo i he hickness was less han 5 mm [
18
]. IOUS
is also used o in aope a i e umo hickness assessmen s and o imp o e loco egional
con ol du ing su ge y [
19
,
20
]. Howe e , mos s udies in he li e a u e e alua ed IOUS in
he assessmen o umo hickness [21–30].
In . J. En i on. Res. Public Heal h 2022,19, 14900 3 o 11
Table 1.
Ad an ages and disad an ages o compu ed omog aphy (CT), magne ic esonance imaging
(MRI), and in ao al ul asound (IOUS) in o al cance local and egional s aging.
Ad an ages Disad an ages
CT
Spa ial esolu ion
Visualiza ion o bone e osion
Pano amic
Low con as esolu ion
In luenced by me al a i ac s
Supe icial lesions no always isible
MRI
Con as esolu ion
Mul ipa ame ic
Pano amic
In luenced by me al a i ac s
Supe icial lesions no always isible
IOUS Ve y high spa ial esolu ion
Di ec isualiza ion o lesions
Ope a o dependen : o a ional
o e es ima ion, comp ession dis o ion,
need o expe hands
No possible i ismus is p esen
Expensi e and no di use echnology
Al hough concep ually di e en , umo hickness (TT) and he DOI demons a ed
a good p ognos ic pe o mance due o he high co ela ion wi h he isk o lymph node
me as ases. Howe e , he use o TT ins ead o he DOI can cause he isk o ups aging
in a small pe cen age o pa ien s; mo eo e , he DOI is now conside ed o be he mos
eliable pa ame e o p edic he isk o lymph node me as ases and p ognosis [
31
,
32
]. Only
ou s udies ha e in es iga ed he ole o an IOUS in he assessmen o he DOI o da e.
Iida and colleagues demons a ed a good co ela ion be ween an ul asonog aphic DOI
(usDOI) and pa hological DOI (pDOI), e en in ea ly ongue SCCs (<5 mm) [
33
]. Filau o
and cowo ke s compa ed he usDOI and MRI-measu ed DOI (m iDOI) and demons a ed
a be e co ela ion be ween he m iDOI and pDOI han be ween he usDOI and pDOI [
34
].
Rocche i and colleagues ound a s ong co ela ion be ween he usDOI and pDOI, bu a
mode a e co ela ion be ween he usDOI and US-measu ed diame e ; mo eo e , hey ound
a high sensi i i y, speci ici y, and PPV in he assessmen o he in il a ion o he umo
beyond he lamina p op ia in o he submucosa (93.1%, 100%, and 100%, espec i ely) [
35
].
Takamu a and cowo ke s ound a high adiological–pa hological ag eemen and showed
ha an IOUS was mo e accu a e han CT and MRI a de ec ing T1 and T2 in squamous cell
ca cinomas [
36
]. To da e, ew s udies ha e analyzed he ole o IOUSs in he p eope a i e
s aging o ea ly ongue SCC [
22
–
24
,
26
,
28
,
35
,
37
], bu only he s udy by Takamu a e al.
e alua ed he adiological–pa hological ag eemen be ween he pDOI and usDOI [
36
].
He ein, we assessed i he IOUS could be an al e na i e s aging ool, especially o ea ly
SCC. In pa icula , we in es iga ed he abili y o he IOUS o p edic he pDOI and T s age
in o al ongue SCC, as well as o p edic a pDOI > 4 mm, which is he h eshold alue o
pe o m elec i e neck dissec ion in ongue SCC.
2. Ma e ials and Me hods
2.1. Pa ien s
A o al o 72 pa ien s unde wen an IOUS o o al ongue mucosal lesions om 2017 o
2021 a ou ins i u ion. The inclusion c i e ia o he s udy we e: (1) pa hologically demon-
s a ed ongue o ongue pel is SCC; (2) a usDOI assessmen ; (3) comple e su gical excision
and his opa hological measu emen o he DOI. The exclusion c i e ia we e: (1) benign
lesion, dysplasia, nonsquamous cell ca cinomas; (2) diagnosis o dis an me as ases and/o
synch onous head and neck SCC; (3) ea men wi h neoadju an he apy; (4) T3 o T4 SCC.
Among hose who me he inclusion c i e ia o he s udy, 29 also unde wen p eope a-
i e MRI.
All pa ien s had been submi ed o su ge y a e a mul idisciplina y eam (MDT)
discussion and p eope a i e counseling be ween head and neck su geons, adiologis s,
adia ion, and medical oncologis s. All pa ien s we e p eope a i ely e alua ed by a dedi-
ca ed head and neck su geon by igid endoscopy unde whi e ligh (WL) and na ow-band
imaging o he assessmen o he supe icial bounda ies o he lesion.
In . J. En i on. Res. Public Heal h 2022,19, 14900 4 o 11
2.2. Measu emen o Radiological and Pa hological DOI
An IOUS was pe o med using a 22-8 MHz 8 mm oo p in hockey-s ick p obe; o
pa ien s who we e scanned be o e 2018, a hockey-s ick 15-7 MHz p obe was used. The p obe
was shielded wi h a la ex co e on which a small amoun o ul asound gel was in oduced.
The examina ion was pe o med wi h he pa ien ex ending he ongue, which was gen ly
held wi h gauze on he con ala e al side by he ope a o . The ul asound examina ion
was pe o med using ligh p essu e o a oid comp ession dis o ion. The en i e lesion
was examined o de e mine he deepes poin o in il a ion. The usDOI was measu ed
pe pendicula ly o he mucosal su ace using he closes no mal mucosa as he e e ence
line; he exophy ic pa s o he lesion we e excluded om he measu emen , while he
ulce a ed pa was included (Figu e 2).
In . J. En i on. Res. Public Heal h 2022, 19, x 4 o 13
All pa ien s had been submi ed o su ge y a e a mul idisciplina y eam (MDT) dis-
cussion and p eope a i e counseling be ween head and neck su geons, adiologis s, adi-
a ion, and medical oncologis s. All pa ien s we e p eope a i ely e alua ed by a dedica ed
head and neck su geon by igid endoscopy unde whi e ligh (WL) and na ow-band im-
aging o he assessmen o he supe icial bounda ies o he lesion.
2.2. Measu emen o Radiological and Pa hological DOI
An IOUS was pe o med using a 22-8 MHz 8 mm oo p in hockey-s ick p obe; o
pa ien s who we e scanned be o e 2018, a hockey-s ick 15-7 MHz p obe was used. The
p obe was shielded wi h a la ex co e on which a small amoun o ul asound gel was
in oduced. The examina ion was pe o med wi h he pa ien ex ending he ongue, which
was gen ly held wi h gauze on he con ala e al side by he ope a o . The ul asound ex-
amina ion was pe o med using ligh p essu e o a oid comp ession dis o ion. The en i e
lesion was examined o de e mine he deepes poin o in il a ion. The usDOI was meas-
u ed pe pendicula ly o he mucosal su ace using he closes no mal mucosa as he e -
e ence line; he exophy ic pa s o he lesion we e excluded om he measu emen , while
he ulce a ed pa was included (Figu e 2).
MRI was pe o med wi h a 1.5 T scanne and a 3.0 T scanne , wi h he manu ac u e ’s
phased-a ay head and neck coils. The exams we e conduc ed on axial and co onal planes
using u bo spin echo (TSE) T1 and T2 weigh ed sequences, and di usion-weigh ed im-
aging sequences (b- alues: 50, 800) wi h an appa en di usion coe icien map and a -
sa u a ed gadolinium-enhanced g adien echo T1 weigh ed sequences. I needed, T1 and
T2 weigh ed sequences o mo ion a i ac educ ion we e used (using, o ins ance, adial
sampling o he k-space sequences).
The pDOI was assessed using a mic ome e in o malin- ixed pa a in-embedded
specimens. The DOI was measu ed ollowing a e e ence line pe pendicula o he plane
o he basemen memb ane o he closes no mal mucosa.
Figu e 2. Two cases o squamous cell ca cinoma o he la e al su ace o he ongue. Fla (a) and
exophy ic (b) lesions a e shown. In each image, DOI (solid line) was measu ed pe pendicula ly om
he mucosal su ace o he deepes poin o in il a ion (cu ed a ow) using he closes no mal mu-
cosa as e e ence line (a ow). The exophy ic po ion o lesions (b) was no included and conse-
quen ly DOI (solid line) and umo hickness (dashed line) we e signi ican ly di e en . A owhead:
supe io longi udinal muscle.
2.3. S a is ical Analyses
S a is ical analyses we e pe o med using MedCalc so wa e. A Shapi o–Wilk es
was used o s udy he dis ibu ion o he a iables. A co ela ion analysis was pe o med
wi h Spea man’s ank co ela ion. The ag eemen be ween he pDOI, usDOI, and m iDOI
was assessed wi h he Bland–Al man plo . An χ² es was used o e alua e he abili y o
he IOUS o co ec ly assign he umo in he co esponding pa hological T s age (pT
s age), es ing he null hypo hesis ha he e is no co ela ion be ween he ul asonog aph-
ically assessed T s age (usT s age) and pT s age. The speci ici y, sensibili y, and a ea unde
he ROC cu e we e calcula ed o de e mine he abili y o he IOUS o p edic a pDOI ≥ 4
Figu e 2.
Two cases o squamous cell ca cinoma o he la e al su ace o he ongue. Fla (
a
) and
exophy ic (
b)
lesions a e shown. In each image, DOI (solid line) was measu ed pe pendicula ly om
he mucosal su ace o he deepes poin o in il a ion (cu ed a ow) using he closes no mal mucosa
as e e ence line (a ow). The exophy ic po ion o lesions (
b
) was no included and consequen ly
DOI (solid line) and umo hickness (dashed line) we e signi ican ly di e en . A owhead: supe io
longi udinal muscle.
MRI was pe o med wi h a 1.5 T scanne and a 3.0 T scanne , wi h he manu ac u e ’s
phased-a ay head and neck coils. The exams we e conduc ed on axial and co onal planes
using u bo spin echo (TSE) T1 and T2 weigh ed sequences, and di usion-weigh ed imaging
sequences (b- alues: 50, 800) wi h an appa en di usion coe icien map and a -sa u a ed
gadolinium-enhanced g adien echo T1 weigh ed sequences. I needed, T1 and T2 weigh ed
sequences o mo ion a i ac educ ion we e used (using, o ins ance, adial sampling o
he k-space sequences).
The pDOI was assessed using a mic ome e in o malin- ixed pa a in-embedded
specimens. The DOI was measu ed ollowing a e e ence line pe pendicula o he plane o
he basemen memb ane o he closes no mal mucosa.
2.3. S a is ical Analyses
S a is ical analyses we e pe o med using MedCalc so wa e. A Shapi o–Wilk es was
used o s udy he dis ibu ion o he a iables. A co ela ion analysis was pe o med wi h
Spea man’s ank co ela ion. The ag eemen be ween he pDOI, usDOI, and m iDOI was
assessed wi h he Bland–Al man plo . An
χ2
es was used o e alua e he abili y o he
IOUS o co ec ly assign he umo in he co esponding pa hological T s age (pT s age),
es ing he null hypo hesis ha he e is no co ela ion be ween he ul asonog aphically
assessed T s age (usT s age) and pT s age. The speci ici y, sensibili y, and a ea unde he
ROC cu e we e calcula ed o de e mine he abili y o he IOUS o p edic a pDOI
≥
4 mm.
Speci ici y, sensibili y, and a ea unde he ROC cu e we e also assessed o he IOUS o
p edic he deep in asion beyond he epi helial laye (pT1 and pT2 s. CIS).
3. Resul s
Al oge he , 41 pa ien s me he c i e ia o he s udy and we e included. Clinicode-
mog aphic da a a e shown in Table 2. The p ima y si e was he la e al su ace in 33 cases,
In . J. En i on. Res. Public Heal h 2022,19, 14900 5 o 11
en al su ace in 3 cases, do sal su ace in 2 cases, and ongue pel is in 3 cases. On he US,
he ongue SCC appea s as a sligh ly hypoechoic lesion ha eplaces he no mal epi helial
laye , which is e y hypoechoic. I in asi e, i in il a es he deepe hype echoic laye ,
which may ep esen he subepi helial connec i e issue and possibly he in insic muscle
o he ongue laye .
Table 2. Clinicodemog aphic da a o pa ien s included in he s udy.
N(%)
Pa ien s 41 (100)
Females 16 (39)
Males 25 (61)
Mean age (s . de ia ion) 64.07 (17.67)
pTis 5 (12.20)
pT1 21 (51.22)
pT2 15 (36.58)
Mean pDOI (s . de ia ion) 3.07 (2.65)
Mean usDOI (s . de ia ion) 3.79 (2.74)
Mean m iDOI (s . de ia ion) 5.39 (2.81)
The mean usDOI, m iDOI, and pDOI we e 3.79 mm (95% CI 2.93–4.65), 5.30 mm
(95% CI 4.23–6.37), and 3.07 mm (95% CI 2.24–3.91 mm), espec i ely. The mean di e ence
be ween he adiological and pa hological DOI was 1.06 mm and 1.64 mm o he usDOI
and m iDOI, espec i ely. Howe e , he di e ence was no s a is ically signi ican (
p= 0.26
).
The spea man ank co ela ion be ween he pDOI and usDOI, pDOI and m iDOI,
and usDOI and m iDOI was 0.84 (95% CI 0.72–0.91, p< 0.0001), 0.79 (95% CI 0.60–0.90,
p< 0.0001
), and 0.91 (95% CI 0.80–0.95, p< 0.0001), espec i ely (Figu e 3). No s a is ically
signi ican di e ence was ound.
In . J. En i on. Res. Public Heal h 2022, 19, x 6 o 13
Figu e 3. Sca e plo s showing he co ela ion be ween usDOI and pDOI (a) and be ween m iDOI
and pDOI (b).
Bland–Al man plo s be ween he usDOI and pDOI, m iDOI and pDOI, and usDOI
and m iDOI a e shown in Figu e 4a, 4b, and 4c, espec i ely. The mean bias be ween he
usDOI and pDOI was 0.7 mm (95% CI 0.26–1.16), whe eas he bias be ween he m iDOI
and pDOI was 1.6 mm (95% CI 1.56–1.90). The mean bias be ween he usDOI and m iDOI
was −0.7 mm (95% CI −1.24–0.24).
The null hypo hesis ha he e is no co ela ion be ween he usT s age and pT s age
was ejec ed, and he al e na i e hypo hesis ha he e is a ela ion be ween he wo clas-
si ica ions was accep ed (p < 0.0001) (Figu e 5). The sensi i i y and speci ici y o he IOUS
o p edic a pDOI ≥4 mm we e 92.31% and 82.14%, espec i ely. The a ea unde he ROC
cu e (AUC) was 0.87 (95% CI 0.73–0.95, p < 0.0001) (Figu e 6).
Using he IOUS, we ound ha 39 umo s wen beyond he epi helial laye and in il-
a ed he subepi helial connec i e issue, while 3 did no . Upon a his ological examina-
ion, we ound ha 36 umo s we e in asi e while 5 we e in si u. The speci ici y, sensi i -
i y, nega i e p edic i e alue (NPV), and posi i e p edic i e alue (PPV) o he IOUS o
p edic an in asi e cance we e 100%, 94.7%, 60%, and 100%, espec i ely. The AUC was
0.8 (95% CI 0.646–0.908, p < 0.0001) (Figu e 6).
Figu e 3.
Sca e plo s showing he co ela ion be ween usDOI and pDOI (
a
) and be ween m iDOI
and pDOI (b).
In . J. En i on. Res. Public Heal h 2022,19, 14900 6 o 11
Bland–Al man plo s be ween he usDOI and pDOI, m iDOI and pDOI, and usDOI
and m iDOI a e shown in Figu e 4a–c, espec i ely. The mean bias be ween he usDOI and
pDOI was 0.7 mm (95% CI 0.26–1.16), whe eas he bias be ween he m iDOI and pDOI was
1.6 mm (95% CI 1.56–1.90). The mean bias be ween he usDOI and m iDOI was
−
0.7 mm
(95% CI −1.24–0.24).
In . J. En i on. Res. Public Heal h 2022, 19, x 7 o 13
Figu e 4. Bland-Al man plo s compa ing usDOI and pDOI (a), m iDOI and pDOI (b), and usDOI
and m iDOI (c). Mean bias be ween usDOI and pDOI is 0.7 mm, be ween m iDOI and pDOI is 1.6
mm, and be ween usDOI and m iDOI is −0.7 mm. In e al ag eemen is smalle in he usDOI-pDOI
plo han in he m iDOI-pDOI plo ( ed lines: in e al ag eemen ; blue line mean di e ence: o ange
line: ze o line).
Figu e 4.
Bland-Al man plo s compa ing usDOI and pDOI (
a
), m iDOI and pDOI (
b
), and usDOI
and m iDOI (
c
). Mean bias be ween usDOI and pDOI is 0.7 mm, be ween m iDOI and pDOI is 1.6
mm, and be ween usDOI and m iDOI is
−
0.7 mm. In e al ag eemen is smalle in he usDOI-pDOI
plo han in he m iDOI-pDOI plo ( ed lines: in e al ag eemen ; blue line mean di e ence: o ange
line: ze o line).
In . J. En i on. Res. Public Heal h 2022,19, 14900 7 o 11
The null hypo hesis ha he e is no co ela ion be ween he usT s age and pT s age
was ejec ed, and he al e na i e hypo hesis ha he e is a ela ion be ween he wo classi i-
ca ions was accep ed (p< 0.0001) (Figu e 5). The sensi i i y and speci ici y o he IOUS
o p edic a pDOI
≥
4 mm we e 92.31% and 82.14%, espec i ely. The a ea unde he ROC
cu e (AUC) was 0.87 (95% CI 0.73–0.95, p< 0.0001) (Figu e 6).
In . J. En i on. Res. Public Heal h 2022, 19, x 8 o 13
Figu e 5. F equency cha s o clinically and pa hologically de i ed T s age. (a) = us-de i ed and
pa hologically de i ed T s age. (b) = m i-de i ed and pa hologically de i ed T s age.
Figu e 6. Recei e ope a ing cha ac e is ic (ROC) cu e o he IOUS assessmen o a pDOI > 4 mm
(a) and o di e en ia ing in asi e and nonin asi e (ca cinoma in si u) umo s (b) (blue line: ROC
cu e; ed line: andom classi ie line).
4. Discussion
The ongue is he mos common subsi e o o al SCC [1]. The eigh h edi ion o he
TNM classi ica ion o icially in oduced he DOI as a c i e ion o de e mining he T s age
in o al cance . The DOI is a well-known p ognos ic ac o . Indeed, se e al au ho s ha e
demons a ed a di ec co ela ion be ween he DOI and he incidence o nodal me as asis
[2,3,8,9]. A p ecise p eope a i e measu emen o he DOI is c ucial o planning he su -
gical app oach: cT1 and selec ed cT2 umo s wi h a DOI < 10 mm ha do no in il a e
ex insic ongue muscles can be emo ed anso ally [15]. Mo eo e , i a eliable
Figu e 5.
F equency cha s o clinically and pa hologically de i ed T s age. (
a
) = us-de i ed and
pa hologically de i ed T s age. (b) = m i-de i ed and pa hologically de i ed T s age.
Figu e 6.
Recei e ope a ing cha ac e is ic (ROC) cu e o he IOUS assessmen o a pDOI > 4 mm
(
a
) and o di e en ia ing in asi e and nonin asi e (ca cinoma in si u) umo s (
b
) (blue line: ROC
cu e; ed line: andom classi ie line).
Using he IOUS, we ound ha 39 umo s wen beyond he epi helial laye and in il-
a ed he subepi helial connec i e issue, while 3 did no . Upon a his ological examina ion,
we ound ha 36 umo s we e in asi e while 5 we e in si u. The speci ici y, sensi i i y,
nega i e p edic i e alue (NPV), and posi i e p edic i e alue (PPV) o he IOUS o
p edic an in asi e cance we e 100%, 94.7%, 60%, and 100%, espec i ely. The AUC was
0.8 (95% CI 0.646–0.908, p< 0.0001) (Figu e 6).
In . J. En i on. Res. Public Heal h 2022,19, 14900 8 o 11
4. Discussion
The ongue is he mos common subsi e o o al SCC [
1
]. The eigh h edi ion o he
TNM classi ica ion o icially in oduced he DOI as a c i e ion o de e mining he T s age
in o al cance . The DOI is a well-known p ognos ic ac o . Indeed, se e al au ho s ha e
demons a ed a di ec co ela ion be ween he DOI and he incidence o nodal me as a-
sis
[2,3,8,9]
. A p ecise p eope a i e measu emen o he DOI is c ucial o planning he
su gical app oach: cT1 and selec ed cT2 umo s wi h a DOI < 10 mm ha do no in il a e
ex insic ongue muscles can be emo ed anso ally [
15
]. Mo eo e , i a eliable adiolog-
ical ool o es ima ing DOI was a ailable, i would be possible o plan a p ima y umo
emo al and elec i e neck dissec ion simul aneously.
Cu en ly, he e is no s anda d adiological echnique o es ima e he DOI. MRI is
conside ed he i s -choice imaging modali y o p eope a i e s aging o ongue SCC. How-
e e , ul asonog aphy may be a sui able al e na i e, hanks o i s high spa ial esolu ion,
especially in ea ly umo s.
In ou s udy, an excellen adiological–pa hological ag eemen in es ima ing he DOI
was ound. Compa ed wi h his opa hology, bo h ul asonog aphy and MRI showed a good
co ela ion, wi h sligh ly be e pe o mance o ul asonog aphy han MRI (0.84 and 0.79,
espec i ely), e en hough he di e ence was no s a is ically signi ican .
The adiological–pa hological ag eemen was s udied using he Bland–Al man plo
(se ing his ology as he e e ence s anda d), which enabled us o e eal he sys ema ic
mean bias. Bo h ul asonog aphy and MRI had a good pe o mance since only h ee cases
and one case we e ou side he limi s o ag eemen , espec i ely, o he i s and second
me hodology. No su p isingly, a sys ema ic e o was ound o bo h imaging modali ies.
Howe e , ul asonog aphy showed a smalle bias and 95% limi s o ag eemen han MRI:
he IOUS sys ema ically o e es ima ed he DOI by 0.7 mm, while he mean bias o MRI
was 1.6 mm. The e o e, ul asonog aphy was mo e p ecise han MRI wi h ega d o CIS,
T1, and T2 umo s. Only he s udy by Takamu a e al. s udied adiological–pa hological
ag eemen and mean bias be ween usDOI and pDOI, showing a highe eliabili y o IOUS
han MRI in he p eope a i e p edic ion o DOI o T1 and T2 SCCs and a 0.2 mm mean
bias be ween he usDOI and pDOI [
36
]. The adiological o e es ima ion o he DOI may be
ela ed o wo phenomena. Fi s ly, issue sh inkage has been demons a ed on o malin-
ixed su gical specimens o o al SCC [
38
,
39
]; mo eo e , umo s a e usually su ounded by
pe i umo al in lamma ion and edema ha may in luence adiological measu emen s o
he DOI, as se e al au ho s ha e demons a ed ha MRI signi ican ly o e es ima es he
DOI [
40
]. Based on ou esul s, we can hypo hesize ha an IOUS may be less in luenced by
pe i umo al edema. Mo eo e , o ea ly ongue SSC, an m iDOi mean bias >1 mm has o
be conside ed clinically meaning ul.
Howe e , despi e he mean bias, ou s udy showed he good pe o mance o he
IOUS in es ima ing he T s age co ec ly: 70.7% o pa ien s we e assigned o he co ec
pa hological T s age, 17.1% o pa ien s we e assigned o a highe T s age, and 12.2% o
pa ien s we e assigned o a lowe T s age.
The DOI was ound o be a good p edic o o occul nodal me as ases. Con o e sies
emain abou he p ope pDOI h eshold o a clinically ele an isk o occul nodal
me as ases; howe e , da a om he li e a u e sugges ha an elec i e neck dissec ion should
be pe o med wi h a pDOI
≥
4 mm [
12
–
14
]. An accu a e ins umen o p eope a i ely measu e
he DOI may allow p oposing concomi an umo esec ion and elec i e neck dissec ion,
educing he a e o wo-s ep p ocedu es. In ou se ies o ea ly ongue SCCs, he IOUS
was e y accu a e in de e mining whe he a umo had a DOI
≥
4 mm o no , which was
he h eshold chosen o elec i e neck dissec ion a ou ins i u ion, ega dless o he T
s age, wi h a sensi i i y o 92.31%, speci ici y o 82.14%, and AUC o 0.87 (95% CI 0.73–0.95,
p< 0.0001
). Howe e , u he s udies wi h a la ge sample a e equi ed o unde s and i he
usDOI alone can be conside ed as a p edic o o neck nodal me as asis and local ecu ence
in he same way as he pDOI.
In . J. En i on. Res. Public Heal h 2022,19, 14900 9 o 11
One o he main s eng hs o ou wo k is he use o a e y high equency (22-8 MHz)
linea p obe, in con as o mos s udies in he li e a u e [
19
–
28
,
30
,
33
,
35
–
37
]. Mo eo e , a
small p obe allowed a ela i ely easy in ao al app oach, e en hough i is mo e di icul o
scan pos e io ly loca ed umo s. Fu he mo e, in his s udy, he use o a e y high equency
p obe enabled us o ecognize he supe icial laye s o he mucosa o he ongue and, in
such a way, o measu e he usDOI om he subepi helial connec i e issue o he deepes
poin o in il a ion o he umo . Howe e , cau ion has o be used o he usDOI assessmen
o mo e ad anced SCCs: ul asonog aphy is less eliable i he pDOI is
>5 mm
[
33
,
41
]. Fo
small high- equency p obes, he isk may be e en highe , bo h o he weak pene a ion o
he ul asound and he small ield o iew.
The p esen wo k enligh ens he ole ul asonog aphy in ea ly ongue SCCs. The clin-
ical impo ance o p ecisely es ima e he DOI in ea ly s ages is due o he ac ha o al
ongue SCCs ha e an ea ly lympha ic sp ead and a high isk o occul me as ases [
2
–
4
],
especially i hey ha e a DOI > 4 mm. To ou knowledge, his is he second s udy o in es iga e
adiological–pa hological ag eemen in ea ly ongue SCCs as a a ge and o include CISs,
con i ming he esul s in he li e a u e [
35
,
36
]. The e o e, we belie e ha a high- equency
IOUS may be a adiological ool o nonin asi ely es ima e he isk o he in asi eness o a
clinically e ealed lesion. Thanks o i s highes spa ial esolu ion among adiological de-
ices, ul asonog aphy appea s o be indica ed explici ly o s udy small, supe icial lesions.
Recen wo k by Rocche i e al. showed ha he sensi i i y, speci ici y, PPV, and NPV o an
IOUS in he assessmen o he ex ension o he umo beyond he lamina p op ia in o he
submucosa we e 93.1%, 100%, 100%, and 60%, espec i ely [
35
]. We epo a sensi i i y,
speci ici y, PPV, and NPV o he IOUS o p edic an in asi e cance o 94.7%, 100%, 100%,
and 60%, espec i ely, in acco dance wi h he li e a u e. Howe e , simila o he s udy by
Rocche i e al., we only e alua ed i e ca cinomas in si u. Fu he s udies a e equi ed o
es ablish he ole o an IOUS in di e en ia ing in asi e and nonin asi e umo s.
The main limi a ions o his s udy a e he small sample size and e ospec i e me hod-
ology. Mo eo e , a high- equency IOUS is a new and ope a o -dependen echnique, and
adiologis s need ime o lea n how o measu e he DOI co ec ly.
The sample was oo small and he ollow-up was insu icien o iden i y he co ela-
ion be ween he DOI and nodal me as ases. Howe e , since we ound a high co ela ion
be ween he usDOI and pDOI, we can indi ec ly assume ha he usDOI may be a p eop-
e a i e p ognos ic ac o o nodal disease: u u e s udies mus be conduc ed o suppo
his hypo hesis.
5. Conclusions
In conclusion, ou s udy showed a good ag eemen be ween he usDOI and pDOI.
We belie e ha an IOUS migh be indica ed o he local s aging o ea ly clinical N0 o al
lesions: his is he scena io in which he p ecision in es ima ing he DOI would ha e a
g ea e clinical impac , a leas o es ima e he isk o occul lymph node me as ases. In
con as , MRI migh be p oposed o mo e ad anced lesions o i suspicious o enla ged
lymph nodes a e p esen , gi en i s pano amic iew and be e sensi i i y in de ec ing bone
in asions and loo -o - he-mou h in il a ion. Howe e , la ge and p ospec i e s udies
mus be conduc ed o con i m his hypo hesis.
Au ho Con ibu ions:
Concep ualiza ion, S.C., A.C. and G.C.; me hodology, S.C. and A.S.T.; o -
mal analysis, S.C.; in es iga ion, S.C., A.C., F.M., M.F. (Ma a Filau o), A.I. and G.P (Giampie o
Pa inello).; da a cu a ion, S.C. and F.M.; w i ing—o iginal d a p epa a ion, S.C.; w i ing— e iew
and edi ing, A.S.T., G.C., G.P. (Giampie o Pa ienello), C.C., M.F. (Ma ina Fiannacca). and F.B.;
supe ision, G.P. (Gio gio Pe e i) and G.C. All au ho s ha e ead and ag eed o he published e sion
o he manusc ip .
Funding: This esea ch ecei ed no ex e nal unding.
Ins i u ional Re iew Boa d S a emen :
The s udy was conduc ed in acco dance wi h he Decla a ion
o Helsinki and was app o ed by he Regional E hics Commi ee o Ligu ia (CER Ligu ia 133/2021).
