Sebaceous Gland and Sweat Gland Carcinomas of the Skin Clinicopathological Study and Significance of c-erbB-2 Oncoprotein Expression

585

0 1992 The Japanese Society of Pathology

Sebaceous Gland and Sweat Gland Carcinomas of the Skin Clinicopathological Study and Significance of c-erbB-2 Onco protein Expression

Takahiro Hasebe', Kiyoshi Mukai2, Kazuyuki lshihara3, Akihiro Kaneko4, and Yukio Shimosato'

Thirteen sebaceous gland carcinomas and 10 sweat gland carcinomas were examined to elucidate any important histological parameters influencing their prognosis, and the relationship between immunohistochemical expressions of c-erbB-2 oncoprotein and survival of the patients was analyzed. Sebaceous gland carcinomas with vacuolated cytoplasm in more than 50% of whole tumor area, with necrosis, and without lymphoid cell infiltration in tumor nests and stroma had a higher incidence of tumor recurrence and tumor-related death than tumors with vacuolated cytoplasm in 50% or less of whole tumor area (p<O.Ol), without necrosis, and with lymphoid cell infiltration in tumor nest and stroma ( ~ ~ 0 . 0 5 ) . Sweat gland carcinomas of a l l cases with fatal outcomes demonstrated tubular differentiation in 20% or less of whole tumor area, lymphatic permeation and desmoplastic reaction. Three sebaceous gland carcinomas and three sweat gland carcinomas were positive for c-erbB-2 oncoprotein. Two of three sebaceous gland carcinomas, and all three sweat gland carcinomas developed tumor recurrence and ended in tumor-related deaths. Sweat gland carcinomas with c- erbB-2 expression had significantly shorter survival than those with negative immunostain (p<O.Ol). Cytoplasmic appearance, tumor necrosis, and lymphoid cell infiltration in tumor nests and stroma of sebaceous gland carcinoma, and tubular differentiation, lymphatic permeation, and growth patterns of sweat gland carcinoma are considered to closely correlate to the prognosis. lmmunohisto- chemically detected c-erbB-2 oncoprotein may be an indi- cator of bad prognosis. Acta Pathol Jpn 42: 585-594, 1992.

Received Febrary 20, 1992. Accepted for publication April 27, 1992. 'Clinical Laboratory Division, National Cancer Center Hospi- tal, Tokyo, ZPathology Division, National Cancer Center Research Institute, Tokyo, 3Dermatology Division and 40ph- thalmology Division, National Cancer Center Hospital, Tokyo. Mailing address: Kiyoshi Mukai, Pathology Division, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104, Japan.

Key words: Sebaceous gland carcinoma, Sweat gland carcinoma, Histology, lmmunohistochemistry, c-erbB-2 oncoprotein

Sebaceous gland carcinoma and sweat gland carcinoma are unusual malignant adnexal tumors of the skin, which account for about 5% (1,2) and 0.05% (3- 5) of all cutaneous malignant tumors, respectively.

In spite of a number of previous publications on clinico pa t holog ica I studies of sebaceous gland ca rcino ma and sweat gland carcinoma, it has not been established which histological parameters are the best predictor of prognosis of patients with these tumors.

In this study, we attempted to answer the following questions: 1) whether there are any important histological parameters which influence the prognosis of patients with sebaceous gland carcinoma and sweat gland carcinomas, 2) if there is an expression of c- erbB-2 oncoprotein, whether or not the expression influences the prognosis of patients with these tumors.

MATERIALS AND METHODS

Subjects

Thirteen cases of sebaceous gland carcinoma and 10 cases of sweat gland carcinoma, which had been treated during the past 27 years at the National Cancer Center Hospital, were examined. All sweat gland carcinomas were classic type eccrine sweat gland carcinoma (6). Clinical information was extracted from the patient's medical record.

Paraffin blocks of each case were obtained from the files of the Pathology Laboratory, National Cancer Cen- ter Hospital. A sample of each tumor of sebaceous gland carcinoma, and sweat gland carcinoma was fixed

586 Sebaceous and Sweat Gland Carcinoma of the Skin (Hasebe et a / . )

in 20% formalin overnight at room temperature. Serial sections were cut and one section was stained by hematoxylin-eosin and examined pathologically to confirm the diagnosis and the others were used for immunohistochemical staining. Five sebaceous gland carcinomas stained for fat using frozen sections clearly revealed intracytoplasmic lipid and showed vacuolated cytoplasm in hematoxylin-eosin stained sections.

Prognostic Para meters

The following parameters were examined to analyze the correlation with patients' prognosis : 1) invasive pattern (large solid nests, intermediate, small solid nests and strands) ; 2) the percentage area occupied by tumor cells with vacuolated cytoplasm vis-a-vis the whole tumor area of sebaceous gland carcinoma ; 3) the percentage area occupied by tubular structure in the whole tumor area of sweat gland carcinoma ; 4) foci of coagulation necrosis ; 5) lymphoid cell infiltration in the tumor nest and stroma; 6) blood vessel and lymphatic inva- sion; 7) N/C ratio; 8) nuclear findings (small and finely granular; intermediate in size; large, irregular, and coarsely granular) ; 9) nucleolar findings (small or inconspicuous ; intermediate in size ; large, irregular, and conspicuous) ; 10) number of mitotic figures per 10 high power field (HPF); 11) completeness of excision; and 12) expression of c-erbB-2 oncoprotein. On lymphoid cell infiltration in the tumor nest and stroma, cases with tumors with no or a few lymphoid cells in surrounding areas of tumor nests were considered to have no lymphoid cell infiltration, and cases with tumors with aggregations of lymphoid cells in surrounding area of tumor nests or band like infiltrations of lymphoid cells beneath tumor nests were considered as those with lymphoid cell infiltration. In the latter, various degrees of lymphoid cell infiltration in tumor nests were frequently observed. In addition, the presence of pagetoid spread in epidermis was examined in cases of sebaceous gland carcinoma.

I m m u nocy toc he mist r y

lmmunohistochemical staining of c-erbB-2 was performed by the avidin-biotin-peroxidase complex (ABC) technique (7). The primary antibody was an affinity- purified polyclonal antibody (Nichirei, Tokyo) specific for 185 kDa c-erbB-2 protein. A section of salivary gland adenocarcinoma, which was strongly positive for c- erbB-2 oncoprotein, was used each time as a positive control. As a negative control, the primary antibody was replaced with normal rabbit serum. Cell membrane staining of the tumor cells was considered as positive.

Statistical Analysis

Survival curves of patients with sebaceous gland carcinoma and sweat gland carcinoma were drawn by the Kaplan-Meier method (8) and statistical differences between the survival curves were calculated using the logrank test (9). The chi-square test was used to analyze the significance of correlation between each his- tologica I para meter and the patient's surviva I.

RESULTS

Clinical f indings

Sebaceous gland carcinoma : Sebaceous gland carcinomas accounted for 9.0% of malignant adnexal tumors of the skin in this hospital. Table 1 shows the clinical findings of cases. There were eight males and five females. The age of patients ranged from 6 to 7 9 years with an average age of 58.4 years. Ten tumors were situated on an eyelid, and one each in the buccal, nasal, and occipital skin. All patients had a solitary tumor.

A six-year-old boy (case 13) was treated with radiotherapy (60 Gy) and chemotherapy (Endoxan, Vincris- tine) for retinoblastoma at the age of three months. When he was two-year-old, the right eye was enucleated. At the age of six years, the tumor was noticed in the irradiated right lower eyelid.

Twelve patients were treated with local excision and tumors were incompletely resected in six patients (cases 1, 4, 5, 10, 12, 13). All patients received radiotherapy or chemotherapy after surgery. Four patients developed local recurrences, and one patient developed lung metastasis and died. In seven patients with completely resected tumors (cases 2, 3, 6-9, 1 l), additional radiotherapy was given to two patients for local recurrence of the tumor. The patient whose tumor was located in the nasal skin died from another cause (gastric cancer and colon cancer) without recurrence of sebaceous gland carcinoma and might belong to the Torre-Muir syndrome (1 0-1 2) associated with multiple visceral carcinomas.

Sweat gland carcinoma : Sweat gland carcinomas accounted for 6.5% of malignant adnexal tumors in this hospital. Table2 shows the clinical findings of the cases. There were six males and four females. The age ranged from 38 to 68 years with an average of 58.3 years. The tumor was situated in the skin of the chest in three patients, the thigh in two, and the lower leg, auricle, axilla, scrotum, and perineum in one each. Among 10 patients, seven had a solitary tumor and three multiple tumors.

587 10 A009009T-6 920924

Table 1. Clinical Findings and Outcome of 13 Patients with Sebaceous Gland Carcinoma

Case Age 1. 55

2. 7 9 3. 61 4. 59

5. 66

6. 6 5

7. 60 8. 56 9. 7 4

10. 69

11. 45

12. 6 5

13. 6

Sex Primary site Treatment Outcome M Cheek Local excision 16 months, DOD

60 Gy, ADM+PEP Metastasis to lung F Eyelid Locai excision 19 months, AW M Nose Local excision 27 months, DNED M Eyelid Local excision 2 8 months, AD

60 Gv Local recurrence

F

M

M

Eyelid

Eyelid

Eyelid Head Eyelid Eyelid

Eyelid

Eyelid

Eyelid

-, _ _ Local excision 60 Gy, PEP Local excision 60 Gy Local excision Extensive excision Local excision Local excision 60 Gy Locai excision 60 Gy Locai excision 60 Gy Local excision 6 0 Gv

3 6 months, AD Local recurrence 36 months, AD Local recurrence 38 months, AW 38 months, AW 38 months, AW 52 months, AW

5 4 months, AD Local recurrence 60 months, AD Local recurrence 150 months, AW

AW: Alive and well with no evidence of disease, AD : Alive with disease, DOD : Dead of disease, DNED : Dead with other cause with no evidence of desease.

Table 2. Clinical Findings and Outcome of 10 Patients with Sweat Gland Carcinoma

Case Age Sex Primary site Treatment Outcome 1. 68 M Thigh Local excision 3 months, DOD

2.

3.

7.

8.

9.

10.

AD :

- 70 Gy, BLM+FT207 MMC+ NCS + Endoxan

63 M Ear Local excision 160 Gy

57 M Chest Local excision 32 Gy, MMC

68 M Scro t um Local excision 40 Gy

60 F Chest Local excision PEP + MMC

38 F Perineum Local excision

5 1 F Chest Local excision 50 Gy, ACNUfEndoxan +MMC+CQ

66 F Axilla Local excision

6 6 M Thigh Local excision

46 M Lea Local excision 60 Gy

- 230 Gy, CDDP+ADM SVDS

Metastasis to lung

5 months, DOD Metastasis to lung, thyroid, and lymph node 14 months, DOD Local recurrence, Metastasis to lymph node 14 months, DOD Metastasis to liver, kidney, and heart 16 months, DOD Metastasis t o lymph node 35 months, AD Local recurrence, Metastasis to lymph node 5 9 months, DOD Metastasis to lung

86 months, DOD Metastasis to lymph node 103 months, AD Local recurrence 108 months, DOD Metastasis to lung and liver

Alive with disease, DOD : Dead of disease.

588 Sebaceous and Sweat Gland Carcinoma of the Skin (Hasebe et a/ . )

Table 3. Relationship between Histological Findings and Prognosis of Patients with Sebaceous Gland Carcinoma

~ No. of Datients - AW AD DOD

Total Histological findinns - Tumor size (cm)

$0.5 4 >0.5 9

with vacuolated cytoplasm 5 50% 7 > 50% 6

Present 6 Absent 6

Small-intermediate 2 Large 11

Small-intermediate 4 Large and irregular 9

Small-intermediate 7 Large and irregular 6

Number of mitotic figures/lO HPF 110 3 >I0 9

Present 5 Absent 8

Present 6 Absent 6

Present 6 Absent 7

Ratio of tumor cells

Pagetoid spread

N/C ratio

Nuclear size

Nucleolar size

Foci of coagulation necrosis

Lymphoid cell infiltration

Tumor at surgical margin

0' 1 DI" * : p<O.OI

0 1 [11 unknown case: 1

0 1 [11

0 1 [11

1 PI 0

0 1 [I1 unknown case: 1

1 [l]" O** * * : p<0.05

0*' 1 [1]** unknown case: 1

AW: Alive and well with no evidence of disease, AD: Alive with disease, DOD: Dead of disease, [ ] : Number of c-erbB-2 positive case.

All tumor cells infiltrated in large solid nests. t, *I. . Significant difference between two groups in AW and AD+DOD.

In nine patients treated with local excision, six tumors (cases 1-5, 7) were incompletely resected. Eight patients received radiotherapy or chemotherapy and had tumor recurrences. Eight of 10 patients died of tumor. Lymph node metastasis developed in four patients and distant metastasis in five patients (lung, liver, kidney, thyroid). Local recurrence only was observed in one patient. One of two surviving cases had local recurrence and another lymph node metastasis.

Clinico pa tho logica I Correlation

Sebaceous gland carcinoma : Table 3 shows the relationship between histological findings and prognosis of patients with sebaceous gland carcinoma. All tumors showed an invasive pattern of large solid nests. Tumor recurrence and tumor-related death were significantly more frequent in the patients with tumors showing vacuolated cytoplasm in more than half of the tumor

area than those with tumors showing vacuolated cytoplasm in less than half of the tumor area (chi-square test, p<O.Ol) (Fig. 1). The tumors with foci of coagulation necrosis, and without lymphoid cell infiltration in tumor nests and stroma showed more frequent recurrence and tumor-related deaths than those without foci of coagulation necrosis, and with lymphoid cell infiltration in tumor nests and stroma (chi-square test, p< 0.05). Of seven patients with complete resection, two developed local recurrences and the tumors of these patients had foci of coagulation necrosis, and no lymphoid cell infiltration in tumor nests and stroma. In contrast, among the tumors of the remaining five patients without tumor recurrence, three tumors had no focus of coagulation necrosis, and lymphoid cell infiltration in tumor nests and stroma, and two tumors showed either of lymphoid cell infiltration in tumor nests and stroma or absence of coagulation necrosis. One tumor, which later developed lung metastasis, had foci of

Acta Pathologica Japonica 42 (8) : 1992

coagulation necrosis with no lymphoid cell infiltration in tumor nests and stroma. There was no significant relationship between other histological parameters and prognosis.

Three of 13 (23%) tumors showed positive stainings for c-erb6-2 oncoprotein (cases 1, 6, 13, Table 1) (Fig. 3) and developed tumor recurrence and one of the three patients died of tumor. Regarding the relationship between his t o log ica I pa ra meters and i m m uno hi st o- chemical positivity for c-erbB-2 oncoprotein, the tumor

589

Figure 1. Sebaceous gland carcinoma. The neoplastic cells have vacuolated cytoplasm, large irregular nuclei, and large conspicuous nucleoli (Hematoxylin-eosin stain).

Figure 2. Sweat gland carcinoma. The tumor cells are infiltrating in small solid nests and strands without tubular structure (Hematoxylin- eosin stain).

of case 13 which did not develop tumor recurrence showed vacuolated cytoplasm in half or less of the tumor area, no focus of coagulation necrosis, and lymphoid cell infiltration in tumor nests and stroma.

Sweat gland carcinoma : Table 4 shows the relationship between histological findings and prognosis of patients with sweat gland carcinoma. Tumors of all fatal cases showed tubular differentiation in areas of 20% or less (Fig. 2), lymphatic permeation and grew in

5 90 Sebaceous and Sweat Gland Carcinoma of the Skin (Hasebe et a / . )

small nests or strands. Tumors of the two surviving patients had tubular differentiation in more than 20% of the area with no lymphatic permeation. There was, however, no significant relationship between any of other histological parameters and prognosis of patients.

Three of 10 (30%) tumors showed positive staining for c-erbB-2 oncoprotein (cases 2-4, Table 2) (Fig. 4) and these developed nodal metastasis and/or distant metastasis to the liver, kidney, lung, and thyroid. Tumors which satisfied all the following histological

Figure 3. lmmunoh is tochemica l staining of c-erbB-2 in a case of sebaceous gland carcinoma. The neoplastic cells show marked membrane staining for c-erbB-2 oncoprotein.

Figure 4. lmmunoh is tochemica l staining of c-erbB-2 in a case of sweat gland carcinoma. The neoplastic cells show marked membrane staining for c-erb8-2 oncoprotein.

parameters were positive for c-erb6-2 oncoprotein : 1) tumor size 0.5 cm or larger, 2) tumors showing tubular differentiation in less than 20% of the tumor area, 3) tumors growing in small nests or strands, 4) tumors with lymphatic permeation.

Prognosis

Figure 5 shows the overall survival of sebaceous gland carcinoma and sweat gland carcinoma. Sebaceous

Acta Pathologica Japonica 42 (8) : 1992

c a,

591

Table 4. Relationship between Histological Findings and Prognosis of Patients with Sweat Gland Carcinoma

x 1 0 0 y 1

No. of patients AD DOD Total AW H istologica I

findings Tumor size (cm) 10.5 >0.5

S 2 0 % >20%

Large nest Intermediate Small nest/strand

N/C ratio S+l L

Small-intermediate Large and irregular

Small-intermediate Large and irregular

Tubular differentiation

Infiltrating pattern

Nuclear size

Nucleolar size

3 7

0 1 0 1

0 0 0 2

2 6 [3l

7 r31 1

0 1 0 1 0 0

3 7

0 0 0 2

0 10 - A

5 5

N m b e r of mitotic figures/lO HPF $10 3 0 1 2 111 >10 7 0 1 6 PI Present 0 0 0 0

Foci of coagulation necrosis

Absent 10 0 2 8 [31 LvmDhatic permeation

i 3 ln 5ot 5 a,

Y; a, ln m a, ln

F

.- n

p<0.05

50 100 Month ~.

Present 6 0 0 6 [3l Absent 4 0 2 2

Present 7 0 1 6 [3l Absent 3 0 1 2

Tumor at surgical margin

Figure 7. Disease-free survival curves of patients with sebaceous gland carcinoma. Curves: A, six patients with tumors with lymphoid cell infiltration in tumor nests and stroma ; B, six patients with tumors without lymphoid cell infiltration in tumor nests and stroma.

loo? AW: Alive and well with no evidence of disease. AD : Alive with disease, DOD : Dead of disease. [ ] : Number of c-erbB-2 positive case S : Small, I : Intermediate, L : Large.

A

x

p<0.05 50 100

Month

Figure 8. Overall survival curves of patients with sweat gland carcinoma. Curves: A, three patients with tumors showing tubular differentiation in 20% or more of tumor cells; B, seven patients with tumors showing tubular differentiation in less than 20% of tumor cells.

v)

o/o p<0.05

50 100 Mon th

o/o

Figure 5. Overall survival curves of patients with sebaceous gland carcinoma and sweat gland carcinoma. Curves: A, 13 patients with sebaceous gland carcinoma ; €3, 10 patients with sweat gland carcinoma. tumor cells, nucleolar features, foci of coagulation ne-

crosis, presence of lymphoid cell infiltration in tumor nests and stroma were analyzed. Tumors with vacuolated cytoplasm in more than half of the tumor area, and with lymphoid cell infiltration in tumor nests and stroma had a significantly shorter disease-free survival period than those with vacuolated cytoplasm in less than half of

gland carcinoma showed significantly better overall survival than sweat gland carcinoma (p<0.05).

Concerning the relationship between histological parameters and prognosis of patients with sebaceous gland carcinoma, tumor size, cytoplasmic feature of

5 92 Sebaceous and Sweat Gland Carc

50

p<O.Ol

100 Mon th

Figure 9. Overall survival curves of patients with sweat gland carcinoma. Curves: A, seven patients with tumors showing negative staining for c-erbB-2 oncoprotein ; B, three patients with tumors showing positive staining for c-erbB-2 oncoprotein.

the area, and without lymphoid cell infiltration in tumor nests and stroma ( ~ ~ 0 . 0 5 ) (Figs. 6, 7). In sweat gland carcinoma, lymphatic permeation, numbers of tumors, percentage of area of tubular differentiation, pattern of tumor recurrence, and tumor size were analyzed and tumors with tubular differentiation in areas of 20% or less showed a significantly worse prognosis than tumors with tubular differentiation in more than 20% of the tumor area (p<0.05) (Fig. 8).

Tumors with c-erbB-2 o nco pro tei n demonstrated by immunohistochemistry in sweat gland carcinoma demonstrated a significantly worse prognosis than tumors without c-erbB-2 oncoprotein expression (p< 0.01) (Fig. 9). On the contrary, expression of c-erbB-2 oncoprotein did not correlate with the prognosis or disease- free survival period of patients with sebaceous gland carcinoma.

DISCUSSION

Sebaceous gland carcinoma is most frequently observed in the 7th decade (2, 13-1 7). The tumor in a six year-old boy in this study was exceptional, because it was a radiation-induced second malignant tumor.

The five-year survival rate of patients with sebaceous gland carcinoma was reported as 30 to 40%(13, 18). There is a controversy regarding the difference in the survival rate of eyelid vs. extraeyelid sebaceous gland carcinoma; one study did not find any difference(l9) and the other showed longer survival for sebaceous gland carcinoma of the eyelid (16). In this study, there was no significant difference in prognosis between eyelid and extraeyelid sebaceous gland carcinoma (Kaplan- Meier method, logrank test, data not shown).

As for the survival rate of patient with sweat gland carcinoma, Wick et a/ . (5) examined 10 cases of the classic type of eccrine gland carcinoma and reported that

:inoma of the Skin (Hasebe et a / . )

the survival rates ranged from 2 months to 10 years with an average of 34.7 months. In this study, the survival rates of classic type eccrine gland carcinoma were from 3 months to 9 years with an average of 44.3 months. Sweat gland carcinoma in this study showed significantly worse prognosis than sebaceous gland carcinoma, indicating that sweat gland carcinoma is a more malignant tumor than sebaceous gland carcinoma.

It is reported that the differentiation of sebaceous gland carcinoma influences the prognosis, and poorly differentiated tumors with finely vacuolated cytoplasm, large irregular nuclei, and foci of coagulation necrosis show worse survival rates than differentiated tumors with abundantly vacuolated cytoplasm, small nuclei, and no focus of coagulation necrosis (1 3-1 5). Among tumors already examined in previous reports, several tumors infiltrated in small nests or strands. Those tumors showed histological feature resembling that of sweat gland carcinoma and had more lymphatic permeation than tumors that grew in large solid nests. In this study, there were no tumor infiltrating in small solid nests or strands, so we could not investigate whether the growth pattern of tumor influenced tumor survival. Among tumors growing in large solid nests, however, there were tumor cells with finely vacuolated cytoplasm and histological features similar to sweat gland carcinoma, which indicates that the growth pattern of tumor may possibly be the most important factor influencing survival. In addition, five of six tumors with vacuolated cytoplasm in more than half of the tumor area had large, irregular, and conspicuous nucleoli, while six of seven tumors with vacuolated cytoplasm in less than half of the tumor area had small to intermediate sized nucleoli, which reveals that the former has significantly more irregularly shaped large nucleoli than the latter (p<O.Ol, chi-square test). Since the features of nucleoli reflects the cell proliferative activity and the nucleoli of rapidly proliferating tumor cells are large, irregular, and pleomorphic (20, 21), it is considered that tumors with vacuolated cytoplasm in more than half of the tumor area proliferate more actively than tumors with vacuolated cytoplasm in less than half of the tumor area. Therefore, if tumors grow in large solid nests, the cytoplasmic characters and nucleolar features of tumor cells would be important parameters to on which to base estimations of prognosis.

The presence of lymphoid cell infiltration in tumor nests and stroma seems to indicate immunopotential to prevent proliferation of tumor cells (22), and tumors with foci of coagulation necrosis seem to proliferate more aggressively than those without foci of coagulation necrosis (23, 24).

Amplification of the c-erbB-2 oncogene was detected

Acta Pathologica Japonica 4 2 (8) : 1992 593

mainly in adenocarcinomas of various organs (25-27), and correlated with the aggressive clinical behavior of human breast cancer (28, 29). Correspondence between gene amplification, and immunohistochemical staining of c-erb6-2 has been reported, and it has been proved that the immunohistochemically detected expression of c-erb6-2 oncoprotein in tumors is useful for the prediction of the prognosis of patients with tumors (30- 34). In this study, expression of c-erb6-2 oncoprotein in sweat gland carcinoma influenced the prognosis of patients, and, in sebaceous gland carcinoma, considering the fact that tumors positive for c-erb6-2 oncoprotein developed either local recurrence or lung metastasis, the expression of c-erb6-2 oncoprotein appears to be an important prognostic factor.

Although we attempted to investigate which is the most important factor influencing prognosis among various histological parameters and c-erb6-2 oncoprotein expression by multivariate analysis, the numbers of patients with sebaceous gland carcinoma and sweat gland carcinoma were too small for the analysis.

With regard to the treatment of sebaceous gland carcinoma and sweat gland carcinoma, the best treatment is complete tumor resection (35-38), combined in the latter with regional lymph node dissection because of the high frequency of nodal metastasis. It was reported that radiotherapy and chemotherapy for sebaceous gland carcinoma and sweat gland carcinoma were of no benefit in many cases (39-44).

In this study, lymph node metastasis and distant organ metastasis were very rare in cases with sebaceous gland carcinoma, and local recurrence tended to occur in patient's with incompletely resected tumors. Therefore, the appropriate primary treatment for sebaceous gland carcinoma would be aggressive and extensive resection of tumor. The patients with tumors having the following histological parameters should receive additional extended skin excision after initial excisional biopsy : 1) tumors with vacuolated cytoplasm in more than half of the tumor area, 2) tumors with foci of coagulation necrosis, 3) tumors without lymphoid cell infiltration in tumor nests and stroma, and 4) tumors with expression of c-erbB-2 oncoprotein.

As for sweat gland carcinoma, some patients had multiple tumors and most of the cases developed lymph node and/or distant organ metastases. Therefore, initial complete extensive local excision with regional lymph node dissection and examination of internal organs for metastases before and after excision must be performed in all patients. In addition, since the histology of sweat gland carcinoma resembles that of breast carcinoma and since Swanson et a/. (45) reported that 7 of 2 3 eccrine ductal carcinomas had reactivity for estrogen receptor

protein, the patients with sweat gland carcinoma should be treated in the same manner as breast carcinoma, including hormone therapy.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

1 4.

15.

16.

REFERENCES

Doxanas MT and Green WR. Sebaceous gland carcinoma : Review of 40 cases. Arch Ophthalmol 102:

Pardo FS, Wang CC, Albert D, and Stracher MA. Sebaceous carcinoma of the ocular adnexa : Radiother- apeutic management. Int J Radiat Oncol Biol Phys

Hirsh LF, Enterline HT, Rosato EF, and Rosato FE. Sweat gland carcinoma. Ann Surg 174: 283-286, 1971. Cabbabe EB. Eccrine gland adenocarcinoma of the chin. Plast Reconstr Surg 69: 521-523, 1982. Wick MR, Goellner JR, Wolfe JT, and Daniel WP. Adnexal carcinomas of the skin: I. Eccrine car cinomas. Cancer 56 : 1147-1 162, 1985. Lever WF and Schaumberg-Lever G. Histopathology of the Skin. J.B. Lippincott, Philadelphia, 1990 : 635- 636. Hsu SM, Raine L, and Fanger H. The use of avidin biotin-peroxidase complex (ABC) in rmmunoperoxidase technique : A comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem

Kaplan EL and Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:

Pet0 R, Pike MC, Armitage P, et a/. Design and analysis of randomized clinical trials requsring pro- longed observation of each patient I I : Analysis and examples. Br J Cancer 35: 1-39, 1977. Rulon DB and Helwig EB. Multipie sebaceous neoplasms of the skin: An association with multiple visceral carcinomas, especially of the colon. Am J Clin Pathol 60: 745-752, 1973. Finan MC and Connoly SM. Sebaceous gland tumors and systemic disease : A cltnicopathologrc analysis. Medicine 63 : 232-242, 1984. Graham R, McKee P, McGibbon D, and Heyderman E. Torre Muir syndrome: An association with isolated sebaceous carcinoma. Cancer 55 : 2868-2873, 1985. Chou NI and Ping-Kuan KUO. Meibomian gland carcinoma : A cltnicopathological study of 156 cases with long-period follow-up of 100 cases. Jpn J Ophthal- mol 23: 388-401, 1979. Russell WG, Page DL, Hough AJ, and Rogers LW Sebaceous carcinoma of meibomian gland origin : The diagnostic importance of pagetoid spread of neoplastic cells. Am J Clin Pathol 70: 504-51 1, 1980. Rao NA, Hidayat AA, McLean IW, and Zimmerman LE Sebaceous carcinomas of the ocular adnexa : A cltnicopathoiogic study of 104 cases, with five-year follow-up data. Hum Pathol 13: 113-122, 1982. Wick MR, Goellner JR, Woife JT, and Daniel WP. Adnexai carcinomas of the skin: I I Extraocular sebaceous carcinomas. Cancer 56 : 11 63-1 172, 1985.

245-249, 1984.

17 : 643-647, 1989

29 : 577-580, 1981

457-481, 1958.

594 Sebaceous and Sweat Gland Carcinoma of the Skin (Hasebe et a / . )

17.

18.

19.

20.

21.

2 2.

2 3.

24.

25.

2 6.

2 7.

28.

2 9.

30.

Lemos LB, Santa Cuz DJ, and Baba N. Sebaceous carcinoma of the eyelid following radiation therapy. Am J Surg Pathol 2 : 305-311, 1978. Boniuk M and Zimmerman LF. Sebaceous carcinoma of the eyelid, eyebrow, caruncle, and orbit. Trans Am Acad Ophthalmol Otolaryngol 72 : 619-642, 1968. Mellette JR Jr, Amonette RA, Gardner JH, et al. Car- cinoma of sebaceous glands on the head and neck: A report of 4 cases. J Dermatol Surg Oncol 7 : 404- 407, 1981. Verheijen R, Kuijpers HJH, Schlingemann RO, et a/. Ki-67 detects a nuclear matrix-associated proliferation-related antigen : I. lntracellular localization during interphase. J Cell Sci 92 : 123-130, 1989. Jhiang SM, Yaneva M, and Busch H. Expression of human proliferation-associated nucleolar antigen p 120. Cell Growth Differ 1 : 319-324, 1990. Tefany FJ, Barnetson RStC, Halliday GM, McCarthy SW, and McCarthy WH. lmmunocytochemical analysis of the cellular infiltrate in primary regressing and non-regressing malignant melanoma. J Invest Der- matol 97 : 197-202, 1991. Meyer JS, Prey MU, Babcock DS, and McDivitt RW. Breast carcinoma cell kinetics, morphology, stage, and host characteristics: A thymidine labeling study. Lab Invest 54 : 41-51, 1986. Barnard NJ, Hall PA, Lemoine NR, and Kadar N. Proliferative index in breast carcinoma determined in situ by Ki67 immunostaining and its relationship to clinical and pathological variables. J Pathol 152 :

Tal M, Wetzler M, Josefberg Z, et a/. Sporadic amplification of the HER2/neu protooncogene in adenocarcinomas of various tissue. Cancer Res 48 :

King RC, Kraus MH, and Aaronson SA. Amplification of a novel v-erbB-related gene in human mammary carcinoma. Science 229: 974-976, 1985. Semba K, Kamata N, Toyoshima K, and Yamamoto T. A v-erbB-related protooncogene, c-erbB-2, in distinct from the c-erbB-l /epidermal growth factor-receptor gene and is amplified in a human salivary gland adenocarcinoma. Proc Natl Acad Sci USA 8 2 : 6497- 6501, 1985. Zhou D, Battifora H, Yokota J, Yamamoto T, and Cline MJ. Association of multiple copies of the c-erbB-2 oncogene with spread of breast cancer. Cancer Res

Slamon DJ, Clark GM, Wong SG, et a/. Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science

Venter DJ, Tuzi NL, Kukar S, and Gullick W. Over- expression of the c-erbB-2 oncoprotein in human breast carcinomas : lmmunohistochemical assessment

287-295, 1987.

151 7-1 520, 1988.

47 : 61 23-61 25, 1987.

235: 177-182, 1987.

31.

32.

3 3.

34.

35.

3 6.

37.

3 8.

39.

40.

41.

42.

43.

44.

45.

correlates with gene amplification. Lancet 2 : 69-72, 1987. Berger MS, Locher GW, Saurer S, et a/ . Correlation of c-erbB-2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading. McCann A, Dervan PA, Johnston PA, Gullick WJ, and Carney DN. c-erbB-2 oncoprotein expression in primary human tumors. Voravud N, Foster S, Gilbertson JA, Sikora K, and Waxman J. Oncogene expression in cholangiocar- cinoma and in normal hepatic development. Hum Pathol 20 : 1163-1168, 1989. Falck VG and Gullick WJ. c-erbB-2 oncogene product staining in gastric adenocarcinoma : An immunohistochemical study. J Pathol 159: 107-111, 1989. Folberg R, Whitaker DC, Tse DT, and Nerad JA. Recurrent and residual sebaceous carcinoma after Moh’s excision of the primary lesion. Am J Ophthal- mol 103: 817-823, 1987. Futrell JW, Krueger GR, Morton DL, and Ketcham AS. Carcinoma of sweat gland in adolescents. Am J Surg Pathol 123: 594-597, 1972. Wertkin MG and Bauer JJ. Sweat gland carcinoma. Current concepts of surgical management. Arch Surg

Okada N, Ota J, Sat0 K, and Kitano Y. Metastasizing eccrine sweat gland carcinoma. Report of a case. Arch Dermatol 120 : 768-769, 1984. Morris DM, Sanusi ID, and Lanehart WH. Carcinoma of eccrine sweat gland : Experience with chemotherapy, autopsy findings in a patient with metastatic eccrine carcinoma, and review of the literature. J Surg Oncol 31 : 26-30, 1986. Nunery WR and Welsh MG. Recurrence of sebaceous carcinoma of the eyelid after radiation therapy. Am J Ophthalmol 96 : 10-1 5, 1983. Harari PM, Shimm DS, Bangert JL, and Cassady JR. The role of radiotherapy in the treatment of malignant sweat gland neoplasms. Cancer 65 : 1737-1740, 1990. Swanson JD, Pazdur R, and Sykes E. Metastatic sweat gland carcinoma : Response to 5-fluorouracil infusion. J Surg Oncol 42 : 69-72, 1989. Sridhar KS, Benedetto P, Otrakji CL, and Charyulu KK. Response of eccrine adenocarcinoma to tamoxifen. Cancer 64 : 366-370, 1989. Chow CW, Campbell PE, and Burry AF. Sweat gland carcinomas in children. Cancer 53 : 1222-1 227, 1984. Swanson PE, Mazoujian G, Mills SE, Campbell RJ, and Wick MR. lmmunoreactivity for estrogen receptor protein in sweat gland tumors. Am J Surg Pathol 15 :

Cancer Res 48: 1238-1243, 1988.

Cancer 65 : 88-92, 1990.

11 1 884-885, 1976.

835-841. 1991.

Documents

Sebaceous Gland and Sweat Gland Carcinomas of the Skin Clinicopathological Study and Significance of c-erbB-2 Oncoprotein Expression