Brief Scientific Reports
Microstaging of Squamous Cell Carcinomas
HELMUT BREUNINGER, M.D., BARBARA BLACK, AND GERNOT RASSNER, M.D.
The clinical classification of squamous cell carcinoma, which
was established primarily by the International Union Against
Cancer (UICC), does not permit optimal estimation of expected
metastasis. The authors' results indicate that metastasis can be
more accurately estimated on the basis of invasion depth, histopathologic grading, and especially tumor thickness. One essential advantage of these criteria is that they can be established
by a histopathologist. It is interesting to note that in the authors'
collective no carcinoma less than 2 mm thick metastasized, that
is, a relatively high percentage of carcinomas (48%) can be graded
as no-risk carcinomas. The risk of metastasis for undifferentiated
carcinomas greater than 6 mm thick that have infiltrated the
musculature, the perichondrium, or the periosteum, however, is
quite high. Tumors between 2 and 6 mm thick with moderate
differentiation and a depth of invasion that does not extend beyond the subcutis can be classified as low-risk carcinomas. (Key
words: Squamous cell carcinoma of skin and lower lip; Prognostic
factor; Microstaging) Am J Clin Pathol 1990;94:624-627
THE INTERNATIONAL UNION Against Cancer
(UICC) TNM system has proved satisfactory for many
malignant tumors because staging and prognosis of the
disease can be established quickly. In most malignant skin
lesions, the diagnosis is usually established at stage I, at
which point there are no discernible metastases (T+, NO,
MO). An appropriate pT classification (local tumor
spread) is therefore extremely important for estimating
the risk of metastasis.
Squamous cell carcinoma (SCC) is classified differently
for the skin and lower lip. Both classifications are relatively
crude with respect to the clinical size of the lesion. Clinical
experience shows that most carcinomas are smaller than
2 cm and definitely metastasize. The question therefore
arises as to whether the current T classification of skin
and lower lip carcinomas is reliable for estimating the
prognosis with respect to the appearance of metastases
during the course of the disease.
The effect of invasion depth or tumor thickness on later
metastasis has been investigated by other research-
Received November 15, 1989; received revised manuscript and accepted for publication March 13, 1990.
Address reprint requests to Dr. Breuninger: Department of Dermatology of the University Tubingen, Liebermeisterstr. 25, D-7400 Tubingen
1, Federal Republic of Germany.
624
Department of Dermatology, University Hospital for Skin
Diseases and Institute for Medical Data Processing,
Tubingen, Federal Republic of Germany
ers9.io.i3.i6.22. t h e s a m p i e
investigated, however, was rela-
tively small.
Based on a large sample and a satisfactorily long followup period, we studied whether prognosis estimation can
be improved by introducing the parameters of tumor
thickness and level to the grading of histologic differentiation.
Materials and Methods
Five hundred seventy-one patients with a total of 673
SCCs were investigated. Five hundred ninety-six carcinomas were in the head region, 158 on the lower lip, and
74 on the helix. Seventy-seven carcinomas appeared on
the extremities and the trunk.
The following data were recorded for the 673 tumors:
1. Clinical size of tumor, that is the mean of the greatest
diameter (in millimeters) and the greatest diameter
perpendicular to this diameter (in millimeters);
2. Depth of invasion as determined on histologic section;
3. Tumor thickness as measured on histologic section
with an accuracy of 0.1 mm, with the use of an ocular
micrometer or a special sliding calabran for thick tumors;
4. Course of disease as observed in yearly follow-up examinations over a period of 1 to 12 years (mean, 6
years).
Results
Twenty-two patients had metastases develop during the
follow-up period (0-36 months after the operation); the
rate of metastasis was therefore 3.9%. Table 1 lists the
metastases for all SCCs: those appearing in the head region
are cited separately from those on the lower lip and helix
and on the extremities and trunk.
BRIEF SCIENTIFIC REPORTS
Vol. 94 • No. 5
625
Table 1. Rates of Metastasis
No. of
Metastases
All SCCs
Head
Lower lip
Helix
Trunk and extremities
All patients
(n
(n
(n
(n
(n
(n
=
=
=
=
=
=
22
19
7
4
3
22
673)
596)
158)
74)
77)
571)
Rate
P Values
3.3%
3.2% - , 4.4% - r
5.4% —
3.9%
3.9%
Difference P = 0.75*
• Difference P = 0.34*
• Not significant.
UICC T Classification of Skin Cancer with
Corresponding Rates of Metastasis
Depth of Infiltration. Table 5 shows the different stages
of infiltration. A clearly significant increase in the rate of
metastasis was established between infiltration of the subcutis or the musculature and infiltration of the perichondrium or periosteum, or cartilage and bone.
Tumor Thickness. Grading the tumors into three classes
(<2 mm, 2-6 mm, >6 mm) revealed that no metastases
occurred in the first class, whereas there was a significant
to highly significant increase in the rate of metastasis between those that were 2-6 mm and those thicker than 6
mm (Table 6).
Five hundred fifteen SCCs were evaluated. Table 2
shows the numeric grading into T classes organized according to clinical size of tumor and the tumors in these
classes that metastasized. Infiltration into the deeper
structures was taken into consideration for T4, as the
TNM system prescribes. A significant difference was established for the rates of metastasis between Tl and T2.
No difference was found between T2, T3, and T4.
Rates of Metastasis for Histologic Grading
of Skin Cancers
Discussion
In general, a favorable prognosis was established for
the SCCs of skin and lower lip in our collective, expressed
as a low rate of metastasis—that is, 3.9% for all patients
over a mean observation period of five years. This rate
corresponds with those reported in the literature.81820
Higher rates of metastasis, however, have also been
cited.79 As the results of this investigation show, however,
precise tumor data are necessary for a reliable comparison
of the rates of metastasis because these rates can vary
considerably, depending on the tumors selected for investigation.
Data on the correlation between tumor size and rate
of metastasis are not uniform in the literature.28'IU6-20-23
The UICC TNM system postulates a dependent relationship between the clinical diameter of the tumor and metastasis. Our investigation confirms this presumption for
carcinomas of the skin and lower lip (see Table 2). The
grading, however, appears too crude. No significant dif-
Five hundred fifteen carcinomas were evaluated. The
rate of metastasis increased significantly from G2 to G4.
No difference was observed between Gl and G2 (Table 3).
UICC T Classification of Carcinoma of the Lower Lip
with Corresponding Rate of Metastasis
One hundredfifty-eightcarcinomas were evaluated. All
metastasizing carcinomas were in the Tl group. Patients
with T3 and T4 carcinomas did not come for treatment
(Table 4).
Additional Possibilities of Estimating the Prognosis
Because the rates of metastasis for the various sites of
carcinomas did not differ statistically, they are combined
in the following analyses.
Table 2. Rates of Metastasis (SCC of skin [n = 515])
TNM Classification
T,
T2
T3
T4
<2 cm
2-5 cm
>5 cm
deep infiltration
* Significant after adjusting for multiple comparisons.
(n
(n
(n
(n
=
=
=
=
421)
87)
7)
46)
No. of
Metastases
Rate
P Values
6
8
1
6
1.4%
9.2%
14.3%
13.0%
• Difference P = <0.001*
• Difference P = 1
• Difference P = 1
BREUNINGER, BLACK, AND RASSNER
626
Table 3. Rates of Metastasis According
to Grading (SCC of skin)
Gl ( n =
G2(n=
G3 (n =
G4 (n =
143)
185)
140)
47)
A.J.C.P. • November 1990
Table 6. Rates of Metastasis According to Tumor
Thickness (all SCCs)
No. of
Metastases
Rate
P Values
1
1
5
8
0.7%
0.5%
3.6%
17.0%
-Difference P= 1.0
-Difference/1 = 0.054
-Difference/1 = 0.004*
No. of
Metastases
<.! mm (n = 325)
2-6 mm (n = 288)
>6 mm (n = 60)
Rate
Level of Significance
" ^ T - Difference P < 0.001*
j 5 o % J - Difference P = 0.005*
* Significant after adjusting for multiple comparisons.
* Significant after adjusting for multiple comparisons.
associates19 demonstrated that, with regard to SCC of the
extremities and the trunk, a tumor thickness of more than
4 mm increases the risk of metastasis. We suggest grading
into three thickness classes (Table 6). The rate of metastasis increased significantly with increasing tumor thickness. Surprisingly enough, 48% of all SCCs of the skin
and lower lip occurred in the group of those less than 2
mm thick; no metastasis appeared in this group. A sharp
increase in malignancy occurred in tumors thicker than
6 mm. Our values for the metastasis of lower lip carcinomas thicker than 6 mm correspond essentially with
those reported by Fierson and associates.8
In conclusion, tumor thickness allows a differentiation
into three risk groups with a higher level of significance
than the depth of infiltration does, and a relatively large
number of carcinomas can be excluded as no-risk tumors
with high significance, which is not possible with the level
of infiltration alone.
Our study shows that the following three risk classes
can be defined:
Table 4. Rates of Metastasis (SCC
of lower lip [n = 158])
TNM
Classification
T,
T2
J]
T4
<2 cm
2-4 cm
>4 cm
deep infiltration
(n
(n
(n
(n
=
=
=
=
153)
5)
0)
0)
No. of
Metastases
Rate
7
0
4.6%
0%
—
—
ferences were established between T2 and T3 classes (Table 2).
In terms of histopathologic grading, tumors with higher
dedifferentiation metastasize more frequently.3"6'91214"18-21
Highly differentiated carcinomas, however, also metastasize. Especially clear is the rapid increase between grades
3 and 4 (Table 3).
The importance of depth of invasion for metastasis remains uncontested.'316'22 However, this has not been satisfactorily considered in the current TNM system. In this
system, the first three classes are graded according to tumor size but the T4 class according to depth of invasion.
This seems highly impractical. Our study shows a significant increase in the rate of metastasis proportional to
depth of invasion (Table 5). This then would explain the
somewhat higher but statistically insignificant rate of metastasis for lower lip and ear carcinomas because at these
sites the skin is very thin, and even small tumors will
infiltrate into deeper structures.
As with melanomas,1 the criterion "tumor thickness"
appears prognostically important. Friedmann and
1. No-risk group with carcinomas less than 2 mm thick,
2. Low-risk group with carcinomas between 2 and 6 mm
(undifferentiated tumors in this group infiltrating the
musculature or other deep structures should be graded
as malignant),
3. High-risk group with tumors more than 6 mm thick
(higher risk of undifferentiated and deep infiltrating
tumors).
Multifactorial analysis would be necessary to obtain
exact grading; our patient collective is still too small for
such analytic procedures.
Table 5. Rates of Metastasis According to Depth of Invasion (all SCCs)
No. of
Metastases
Middle of corium
Full corium
Subcutaneous fat
Muscle
Cartilage/bone
* Significant after adjusting for multiple comparisons.
(n
(n
(n
(n
(n
=
=
=
=
=
176)
181)
220)
72)
24)
0
1
9
9
3
Rate
0.6% —
4.1%
12.5%
12.5%—'
P Values
-Difference/'= 0.31
-Difference P= 0.026
-Difference P = 0.017*
-Difference P = 1*
BRIEF SCIENTIFIC REPORTS
Vol. 94 • No. 5
References
Breslow A. Thickness, cross sectional areas and depth of invasion
in the prognosis of cutaneous melanomas. Ann Surg 1970;172:
902-908.
Breuer I, Ehring F. Die Metastasierung des Plattenepithelcarcinoms
der Haut, an 11 Katamnesen und Autopsien unterucht. Arch
Derm Forsch 1972;245:277-284.
Broders AC. Squamous-cell epithelioma of the lip. A study of 537
cases. JAMA 1920;74:656-664.
Broders AC. Squamous epithelioma of the skin. Ann Surg 1921 ;73:
141-160.
Brown RG, Poole MD, Calmel PM, Behamjian VY. Advanced and
recurrent squamous carcinoma of the lower lip. Am J Surg
1976;132:492-498.
Eggert JH, Dumbach J, Steinhauser EW. Operative Therapie der
regionaren Lymphknoten bei Unterlippenkarzinomen. Der
Hautarzt 1986;37:444-449.
Epstein E, Epstein NN, Bragg K, Linden G. Metastases from squamous cell carcinoma of the skin. Arch Dermatol 1968,97:245251.
Fierson FH, Cooper PH. Prognostic factors in squamous cell carcinoma of the lower lip. Hum Pathol 1986;17:346-354.
Friedman HJ, Cooper PH, Wanebo HJ. Prognostic and therapeutic
use of microstaging of cutaneous squamous cell carcinoma of the
trunk and extremities. Cancer 1985;56:1099-1105.
10. Galinsky AW, Williams JE, Geduldig SB. Squamous cell carcinoma:
a review of the literature and case report. Journal of the American
Podiatry Association 1981;71:505-508.
Gottron HA. Zur Klinik, Pathogenese und Prognose des Hautkarzinoms. Die Medizinische 1955;4:133-140.
627
12. Grier WRN. Squamous cell carcinoma of the body and extremities.
In: Andrade R, Gumport SL, Popkin GL, Rees TD, eds. Cancer
of the skin, vol 2. Philadelphia: WB Saunders, 1976:916-939.
13. Hermanek P, Sobin LLH, eds. International Union Against Cancer
TNM classification of malignant tumours. Springer Verlag, 1987.
14. Immeraman SC, Scanlon EF, Facs MC, Knox K.L. Recurrent squamous cell carcinoma of the skin. Cancer 1983;51:1537-1540.
15. Jessen RT, Merwin CF. Identifying and treating skin malignancies.
Geriatrics 1979;34:71-78.
16. Lever W, Schaumburg-Lever G. Histopathology of the skin. Philadelphia: J. B. Lippincott, 1983.
17. Lund HZ. How often does squamous cell carcinoma of the skin
metastasize? Arch Dermatol 1965;92:635-637.
18. Mihm MC, Clark WH Jr, From L. The clinical diagnosis, classification and histogenetic concepts of the early stages of cutaneous
malignant melanomas. N Engl J Med 1971;284:1078-1082.
19. Moller R, Nielsen A, Reyman F, Hou-Jensen K. Squamous cell
carcinoma of the skin and internal malignant neoplasms. Arch
Derm 1979;115:304-305.
20. Sage HH, Casson PR. Squamous cell carcinoma of the scalp, face
and neck. In: Andrade R. Gumport SL, Popkin GL, Rees TD,
eds. Cancer of the skin, vol 2. Philadelphia: WB Saunders. 1976:
899-915.
21. Stromberg BV, Keiter JE. Wray RC, Weeks PM. Scar carcinoma:
prognosis and treatment. South Med J 1977;70:821-822.
22. Turk LL, Winder PR. Carcinomas of the skin and their treatment.
Semin Oncol 1980;7:376-384.
23. Williamson GS, Jackson R. Treatment of squamous cell carcinoma
of the skin by electrodesiccation and curettage. Can Med Assoc
J 1964;90:408-423.
Histiocytosis X
S-100 Protein, Peanut Agglutinin, and Transmission Electron Microscopy Study
FENG YE, M.D., SHU-WEI HUANG, M.D., AND HANG-JI DONG, M.D.
Twenty-seven cases of histiocytosis X (HX) for which paraffin
blocks were available were studied with the use of S-100 protein
immunohistochemistry, peanut agglutinin affinity histochemistry,
and transmission electron microscopy. The results show that
these techniques did enable identification of Langerhans'-type
histiocytes in 88.5%, 75%, and 47.4% of cases, respectively. All
techniques were proved to be of some diagnostic value for HX,
but none was able to confirm the diagnosis in all instances and
none could foretell the prognosis of the patients. This study shows
that besides the Langerhans' cells, the indeterminate cells of the
skin and the interdigitating dendritic reticulum cells of the lymph
node may also be involved in this process. Moreover, some multinucleate giant cells and foamy cells may be derived from Langerhans' and related cells. (Key words: Histiocytosis X; S-100
Received December 6, 1989; received revised manuscript and accepted
for publication April 2, 1990.
Address reprint requests to Dr. Ye: Department of Pathology, Affiliated
Hospital, Zhanjiang Medical College, Guangdong, The People's Republic
of China.
Department of Pathology, Affiliated Hospital,
Zhanjiang Medical College, Guangdong,
The People's Republic of China
protein; Immunohistochemistry; Peanut agglutinin; Affinity histochemistry; Transmission electron microscopy; Langerhans1
cells) Am J Clin Pathol 1990;94:627-631
HISTIOCYTOSIS X (HX) is a syndrome of unknown
pathogenesis consisting preponderantly of a well-differentiated histiocytic proliferation with an infiltrating pattern. Traditionally, it is subclassified into Letterer-Siwe's
disease (LSD), Hand-Schiiller-Christian's syndrome
(HSCS), and eosinophilic granuloma of bone (EGB). It
has recently been established that the demonstration of
S-100 protein (S-100) by immunohistochemistry and the