PART TWO
Diagnostic Cytology
Table 29.2 Differential Diagnosis of Malignant Round Cell Tumors in Children—
Cant'd
Neoplasm type
Age group
Location
Cytologic features
Immunopositivity
Electron
microscopy
Genetic
disorders
Small-cell hepa-
toblastoma
< 3 years
Liver
Dyscohesive small
rounded cells with small
amounts of glycogen,
lipids, and biliary pigment
Cytokeratin,
alpha-fetoprotein,
Hep-Par1
Pronounced
expression of
extracellular matrix
proteins, fibers,
collagen type
Ill-positive
Loss of hetero-
zygosity for the
telomeric portion
11p (11p 15.5),
overexpression
ofTP53
Pancreato-
blastoma
< 5 years
pancreas
Epithelioid cells with wide
granular cytoplasm, trian-
gular spindle cells, stromal
fragments, rosette-like
structures, and squamoid
corpuscles
Pancreatic en-
zymes, endocrine
markers, and
markers of duc-
tal differentiation:
carcinoembryonic
antigen, DUPAN-2,
B72.3
Granulates of zy-
mogen, acinar cell
differentiation
pancreatoblas-
toma is often
a component
of the Beck-
with-Wiedeman
syndrome
NSE, neuron-specific enolase.
cells. H ep atocellular carcinom a differs fro m hepatoblastom a,
show ing cells s im ila r to a d ult hepatocytes. H ep atocellular car-
cinom as also p refere ntially affect child ren o ld e r th an 10 years
and w ith previous chronic hepatic damage. A n o th e r im p o rta n t
m a lig n a n t e p ith elial tu m o r is pancreatoblastom a. Together w ith
hepatoblastom a, it shows a characteristic clinical and im ageno-
logic profile. Pancreatoblastom as often show fine-needle aspi-
rates w ith e p ith e lio id , m onocytic/m acrophagic system cells and
in fla m m a to ry cells. Som e sarcomas can also show e p ith e lio id
cells, such as alveolar soft tissue sarcom a and sin ovial sarcoma. It
can be confused w ith adenocarcinom as and granular cell tum ors,
because it shows large, atypical, p leo m o rp hic e p ith e lio id cells
w ith vesiculous nuclei and p ro m in e n t n u c le oli and lig h t cyto-
plasm , som etim es vacuolated. C lin ica l, im agenologic, and lab o-
ra to ry exam inations are essential fo r establishing the d iffe re n tia l
diagnosis, such as in tu m o rs o f extragonadal germ cells, w h ic h
are often located in the sacrococcygeal region, retro p e rito ne um ,
o r m ed iastinum . A n im p o rta n t group o f th em are partial or
to ta lly cystic.35-37
Diagnostic Accuracy of FNA Cytology
P ublished sensib ilities and specificities obtained b y FN A in large
series o f pediatric tu m o rs are between 90 and 100% and are
s im ila r to those obtained in adults.4-6,49 In a retrospective review
o f 996 FNAs o f pediatric tu m o rs in o u r hosp ital, 197 m alig n an t
and 799 b enign and o f d iffe re n t locations, perform ed between
1982 and 2005, we ob tained a se n sitivity o f 95.3% and a spe-
c ificity o f 98.0% and a p ositive predictive value o f 95.5% w ith -
o u t sig nificant differences depending o n the sam ple o rig in site
(Table 2 9 .3 ).
Causes of Error
Causes o f error are s im ila r to those o f FN A in adults. T he m ost
im p o rta n t causes are unusual sm all tu m o rs w ith a lo c a tio n d if-
fic u lt to reach, inap p rop riate technique, p o o r fix a tio n and stain,
inap p rop riate screening, an inexperienced pathologist, and lack
o f consideration o f clinical antecedents.4,5
False negatives
False negatives are u su ally due to sm all o r scarcely m acroscopi-
cally visib le m a lig n a n t neoplasm s, aspirate o f m a lig n a n t neo-
plasm s w ith in su ffic ie n t o r m in im a l cellular m aterial, necrosis,
hem orrhage, little atypia, ab und ant and fib rous strom a, o r
accentuated in fla m m a tio n .4,5
False positives
False positives are m a in ly due to the presence o f lesions th a t
could be diagnosed as m a lig n a n t o n ly b y histology, errors o f
m orp h olog ic in te rp re ta tio n (benign neoplasm s w ith sig nificant
atypical cells, abund ant m itosis, in fre q u e n t histo lo g ic type, pres-
ence o f a h ig h n um b er o f lym phocytes), low-grade o r b o rd e rline
lesions, and cellular m orp h olog ic alterations induced b y radia-
tio n o r chem otherapy.4,5
Concluding Remarks
Fine-needle asp iration cytology is s till w id e ly underused fo r
studying pediatric tu m o rs o f diverse locations, despite its hig h
diagnostic yie ld (overall sensib ilities and specificities between
90 and 100% ) and scarce contraind ications and com plications.
T he advantages and technical features are s im ila r to those o f
FN A in adults, b u t it occasionally requires the p a rticip ation o f
relatives o r even the use o f sedative drugs d uring the procedure
to achieve the appropriate p atient cooperation.
It is often necessary and som etim es indispensable to use spe-
cial techniques in FNA, such as im m unocytochem istry, electron
m icroscopy, cytogenetics, and m olec ular biology, to establish
the specific diagnosis a n d /o r to d eterm ine the d iffe re n tia l diag-
nosis in u nd iffe ren tia te d o r p o o rly d ifferentiated neoplasms.
A d d itio n a lly, som e m a lig n a n t ro u n d cell tu m o rs and o th e r pedi-
atric neoplasm s are fre q ue ntly lin ke d to genetic aberrations o r
fa m ilia r disorders.
948
previous page 936 ComprehensiveCytopathology 1104p 2008 read online next page 938 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off