Cytogenetic Heterogeneity in a Clear Cell Hidradenoma of the Skin Ludmila Gorunova, Fredrik Mertens, Nils Mandahl, Nils Jonsson, Bertil Persson, Sverre Helm, and Felix Mitelman
ABSTRACT: Short-term cultures from a clear cell hidradenoma, a benign skin tumor for which no chromosome data exist, were cytogenetically analyzed. A total of eight unrelated aberrant clones were identified. The karyotypic profiles of two separately processed parts of the sample - a t u m o r n o d u l e and seemingly normal adjacent dermal tissue - were different. Characteristic for the t u m o r nodule was a single abnormal clonal population consisting of three subclones: 46,XY, der(2)inv(2)(p13q23)t(2;9)(p13;q22), der(9)t(2;9)(q23;q22)•t(11;19)(q21;p13)•t(12;19)(q24;p13)/46•idem•inv(1)(p32q44)/92•idemx2. The adjacent tissue contained, in addition to the clone found in the t u m o r nodule, a spectrum of unrelated clones, the largest of which also showed clonal evolution: 45-47,XY, t(3;6)(p25;p25),t(12;17)(q15;q12), -17, + r(17)x2 [cp]/45-47,idem,inv(5)(p15q22)/90-94,idemx2. The remaining six clones found in this part were small and had simpler numerical or structural aberrations. The multiclonal pattern observed in this hidraden o m a seems to reflect both cytogenetic convergence and divergence during neoplastic progression. The presence of unrelated clones m a y be an indication that the t u m o r was of multicellular origin.
INTRODUCTION
MATERIALS AND METHODS
Cytogenetic information on benign skin n e o p l a s m s is very limited ([1], updated). Recently, clonal chromosome aberrations were described in a s p i r a d e n o m a [2]; this was the first time a skin t u m o r w i t h eccrine differentiation was demonstrated to have an abnormal karyotype. We herein describe the finding of m u l t i p l e chromosome rearrangements in a clear cell h i d r a d e n o m a (eccrine acrospiroma). This benign sweat gland t u m o r is u s u a l l y detected as a small intradermal n o d u l e covered by intact skin. Clear cell h i d r a d e n o m a s are c o m p o s e d of several cell types. In solid t u m o r parts, two cell types predominate: one large and round, with a clear glycogen-rich cytoplasm and a small dark nucleus; the second, small, polyhedral or fusiform, with slightly basophilic cytoplasm and a r o u n d e d nucleus. In addition, four types of l u m i n a l cells showing different degrees of epidermalductal, dermal-ductal, and secretory differentiation can be found [3].
Case Report
From the Departments of Clinical Genetics (L. G., Fr. M., N. M., E M.), Clinical Pathology (N. J.), and Dermatology (B. P.), University Hospital, Lund, Sweden; and the Department of Medical Genetics (S. H.), Odense University, Odense, Denmark. Address reprint requests to: Dr. Ludmila Gorunova, Department of Clinical Genetics, University Hospital, S-221 85 Lund, Sweden. Received December 1Z 1993; accepted February 24, 1994.
A n 83-year-old m a n h a d for 3 months noticed a b l u i s h subcutaneous growth in the left gluteal region. Clinical examination revealed a 3 × 3 - c m large hematoma-like and slightly fluctuating lesion. There was no history of trauma or irradiation. A p u n c h biopsy was indicative of cystic clear cell h i d r a d e n o m a and the tumor was extirpated. Except for some small cysts, the tumor was macroscopically solid. Histopathologic analysis revealed a fairly well demarcated t u m o r located in the u p p e r and m i d d l e cerium. It consisted of multiple, partly coalescent epithelial lobes that were separated by strands of collagen-rich connective tissue with some edema in the superficial parts. The epithelial lobes were composed of sheets of small p o l y h e d r a l basophilic cells mixed w i t h larger, somewhat elongated cells w i t h clear cytoplasm and a distinct cell membrane, as well as cells of intermediate appearance. The sections also revealed tubular l u m i n a and small cysts lined by cuboidal ductal cells (Fig. 1). Several loci had e p i d e r m o i d differentiation (Fig. 1); areas with oncocytic cell changes also were seen. Neither cellular pleomorphism nor necrosis was observed. The diagnosis was clear cell hidradenoma.
Cytogenetic Methods A 1.5 × 1.5 × 1.5-cm t u m o r sample taken for cytogenetic analysis was d i v i d e d into two parts based on gross morphology. The larger part, a t u m o r nodule, had an irregular surface and a brownish color. The second, adjacent part was comparatively thin and r e s e m b l e d dermal tissue. The two parts
26 Cancer Genet Cytogenet 77:26-32 (1994) 0165-4608/94/$07.00
© 1994 Elsevier Science Inc. 655 A v e n u e of the Americas, New York, NY 10010
Polyclonality in an Eccrine Skin T u m o r
27
Figure 1 Top) Mixture of small polyhedral cells and larger clear cells. Tubular lumina are lined by cuboidal cells (H & E, × 160). Bottom) Focus of epidermoid differentiation in solid area (H & E, x 160).
were processed separately. They were finely m i n c e d with scissors and disaggregated enzymatically in 280 U/ml collagenase II for 18 h. Whereas the cell s u s p e n s i o n obtained from the t u m o r n o d u l e was plated both on glass chamber slides and in plastic flasks, cultures from the adjacent tissue were set up only in flasks. The growth surfaces were coated with vitrogen. The cultures were m a i n t a i n e d in a
CDM4 m e d i u m [4] that was s u p p l e m e n t e d w i t h 10% fetal bovine serum. After 4-27 days, the cultures were exposed to 0.015 txg/ml of Colcemid for 12-14 hours and harvested. For glass chamber slides this p r o c e d u r e i n c l u d e d hypotonic treatment in situ with 0.25% NaC1 for 30 min and gradual fixation in methanol :acetic acid (3:1). Cells in flasks were detached from the
28
L. Gorunova et al.
Table 1
Karyotypes of clones identified in short-term cultures of clear cell h i d r a d e n o m a
Clone
Karyotype
I
46•XY•der(2)inv(2)(p13q23)t(2;9)(p13;q22)•der(9)t(2;9)(q23;q22)•t(11;19)(q21;p13)•t(12;19)(q24;p13) [165]/46,idem,inv(1) (p32q44)[32]/92,idemx2 [3] 45-47,XY,t(3;6)(p25;p25),t(12;17)(q15;q12), - 17, + r(17)x2[cp52]/45-47,idem,inv(5)(p15q22)[4]/ 90-94,idemx216] 47,XY, + 20[2] 47,XY, + Y[11] 45,X, - Y[6] 45,X, - Y,t(4;5)(q23-25;q22)[4] 46,Y,t(X;18)(q24;p11)[3] 46,XY,t(6;6)(p11;q21)[2] 46,XY[33]
II III IV V VI VII VIII Normal
surface by mild treatment with tryspin/EDTA and were, after immersion in 0.06 M KCI for 30 m i n followed by repeated fixations, dropped onto wet slides. The air-dried slides were kept at 60°C overnight, incubated in 2 × SSC at 60°C for 4 hours, and G-banded with Wright stain. Because the harvests performed w i t h i n the first week showed a very low mitotic activity in the cultures, we attempted to increase the n u m ber of mitoses by partial, repetitive harvesting whenever possible. In the cytogenetic analysis, the clonality criteria and description of karyotypes followed the recommendations of the ISCN (1991) [5]. RESULTS The cytogenetic findings are summarized in Tables 1 and 2. A total of 330 metaphases were analyzed, 90% of w h i c h were abnormal. Altogether eight aberrant clones (I-VIII) were identified (Table 1). A m i n o r fraction (2%) of the abnormal cells had n o n c l o n a l changes (Table 2).
Table 2
All the aberrant clones were in the neardiploid range. The most c o m m o n numerical change was loss or gain of the Y chromosome (clones IV-VI). Trisomy 20 was the sole anomaly i n clone III. The clonal structural rearrangements affected 13 different chromosomes (X, 1-6, 9, 11, 12, and 17-19) and comprised inversions, balanced translocations, and ring formation. The eight aberrant clones were unrelated, with the exception that the two small clones V a n d VI both had -Y. The two largest c l o n e s - I and I I - e a c h exhibited clonal evolution and consisted of three closely related subclones. Karyograms of clones I and II are presented i n Figures 2 and 3 and partial karyotypes illustrating all the other structural clonal aberrations are given in Figure 4. The cytogenetic profiles of the two separately processed parts of the s a m p l e - t h e t u m o r nodule and the macroscopically seemingly normal adjacent dermal t i s s u e - w e r e different (Table 2). A m o n g a total of 169 cells studied in the n i n e cultures from the tumor nodule, 162 cells (96%) belonged to clone I; no other aberrant clone was detected. In contrast, there was a marked heterogeneity i n the adjacent tissue. Of
Distribution of aberrant clones among short-term cultures of clear cell hidradenoma ° No. of cells in aberrant clones
Cultureb Tumor nodule C4 C10 C13 F6 F7 F8 F9 F16 F17 Total Adjacent tissue F22 F23 F27 Total
Days in culture 4 4 4 25 27 7 7 25 25 25,27 14,25 14,17
Clone I
Clone II
Clones III-VIII
No. of cells with nonclonal aberrations
No. of normal cells
1 1 9 14 19 1 2 5 117 169
1 6 14 19 1 5 116 162 27 8 3 38
° Karyotypesof the clones are given in Table 1. C, chamber slide; F, flask.
Total no. of cells
62
28
62
28
21 5 1 27
143 14 4 161
14
20
13
19
15
8
9
16
IO
17
11
Figure 2 Representative karyogram of clone I: 46,XY,inv(l)(p32q44),der(2)inv(2)(pl3q23)t(2:9)(pl3;q22),der(9)t[2;9) (q23;q22),t(ll;l9)(q2l;pl3),t(12;19)(q24;pl3). Arrowheads indicate breakpoints.
7
6
18
12
15
14
20
13
19
9
12
16
11
17
22
10
16
21
Figure 3 Representative karyogram of clone II: 46,XY,t(3;6)(p25;p25),t(12;17)(q15;q12),r(l7). The insert shows the inv(5)(p15q22) present in a subclone. Arrowheads indicate breakpoints and the arrow a ring chromosome.
8
7
6
XY
Polyclonality in an Eccrine Skin Tumor
31
V
4
5
X
18
.i 6 Figure 4 Partial karyotypes showing the clonal structural aberrations of clones VI-VIII.Clone VI: t(4;5)(q23-25;q22) (top); clone VII: t(X;18)(q24;p11) (middle); clone VIII:t(6;6)(p11;q21) (bottom). Arrowheads indicate breakpoints. 161 cells analyzed, 38 (24%) belonged to clone I, 62 cells (38%) represented clone II, and 28 cells (17%) constituted altogether six small clones (Ill-VIII). The whole range of aberrations was found only in culture F22; two other cultures (F23 and F27) yielded results similar to those obtained in the cultures from the tumor nodule. DISCUSSION
Perhaps the most remarkable finding of the detailed cytogenetic analysis of the clear cell hidradenoma was that the
aberrant clones were so unevenly distributed between the two neighboring parts of the tumor sample. In the tumor nodule, practically all abnormal cells belonged to a single clonal population consisting of three subclones (clone I). The macroscopically normal-appearing adjacent tissue, on the other hand, contained not only clone I but also a spectrum of unrelated clones: one large with three subclones (clone II) and several small ones with simple alterations (III-VIll). The fact that only one culture (F22) from the latter sample portion showed the whole range of aberrations might lead one to ask whether the changes could not have arisen in vitro. We think this is highly unlikely. First, all the cultures were harvested within 4 weeks and the number of mitotic events was low. Second, the emergence of clone II could hardly, in view of its complex structure and the high frequency at which cells with these changes were seen, be explained as an in vitro artifact. Moreover, if culture conditions did play a significant role in our case, one would expect to see the effects in more than I of 12 analyzed separate cultures. We conclude that the cytogenetic diversity detected in the adjacent tissue already existed in vivo; the differences among cultures were probably stochastic in nature. They could be due to an uneven distribution of cell aggregates among flasks after enzymatic treatment or to a too-small number of metaphases available in cultures F23 and F27. The genetic heterogeneity demonstrated in this skin tumor seems to reflect both cytogenetic convergence (a reduction of karyotypic diversity) and divergence (an increase of genomic complexity) during neoplastic progression [6]. A scenario of cytogenetic convergence may be envisaged if one assumes that the tumor had a multicellular origin (suggested by the presence of several unrelated clones in the adjacent tissue) and that one clone gradually gained the upper hand (exemplified by the finding of a single expanded clone in the tumor nodule). On the other hand, the clonal evolution within clones I and II constitutes clear evidence of cytogenetic divergence. The relative pathogenetic importance of the different clones is difficult to assess. Features of the clone I such as 1) complex karyotypic structure; 2) widespread distribution with the highest frequency (98 %) in the tumor nodule; and 3) consistent occurrence in cultures from both parts of the sample argue that this clone is representative of the tumor parenchyma. But certain traits of clone I f - unbalanced rearrangements with unstable ring chromosomes, signs of clonal evolution, and a rather high frequency (about 40%) in the nodule-near tissue- indicate a neoplastic nature of this clone as well. The significance of the smaller clones (III-VIII)with simple aberrations is more problematic. It has recently been disputed whether the small clones with simple numerical changes or seemingly balanced structural rearrangements frequently reported in tumors of the skin and head and neck belong to the tumor parenchyma; also fibroblasts and other stromal cells may carry chromosome alterations that presumably are carcinogenetically irrelevant [7-9]. It is in this context worthy of note that the small unrelated clones were observed only in a culture derived from tissue which was located between the tumor nodule and completely normal dermis. The cultures from this adjacent tissue contained 5 times more normal metaphases than did the tumor nodule (Table 2). In these cultures, furthermore, fibroblastoid cells
32
were frequent, i n contrast to the p r e d o m i n a n t l y epithelial growth pattern seen in the cultures from the tumor nodule. In view of the complex cytogenetic picture of the clear cell hidradenoma and the lack of previous data o n this tumor type, speculations about tumor-specific aberrations seem premature. In the only previous report [2] on eccrine skin tumors (a morphologically b e n i g n spiradenoma, but with malignant behavior}, both the primary tumor and its metastases showed the changes -5, del(16}(q22), a n d a marker chromosome. That karyotype thus has no c o m m o n denominator with our cytogenetic findings; the only cytogenetic similarity between the two sweat gland tumors is the neardiploid chromosome number. It may be noteworthy that one of the structural rearrangements in our case-t(11;19) (q21;p13)-looks identical to an aberration reported in two Warthin's tumors [10, 11], which are b e n i g n salivary gland neoplasms with ductal differentiation. Whether the t(11;19) really signifies an early cytogenetic alteration c o m m o n to these tumor types remains to be seen. This study was supported by grants from the Swedish Cancer Society and the Medical Faculty of Lund University. Dr. Gorunova is on leave from the Institute of Cytology of the Russian Academy of Sciences, St. Petersburg, Russia.
REFERENCES 1. Mitelman F (1991): Catalog of Chromosome Aberrations in Cancer, 4th Ed. Wiley-Liss, New York.
L. Gorunova et al.
2. Dijkhuizen T, Van Den Berg E, Nikkels PGJ, Hoekstra HJ, De Jong B (1992): Cytogenetics of a case of eccrine spimdenoma. Hum Pathol 23:1085-1087. 3. Lever WF, Schaumberg-Lever G (1990): Histopathology of the Skin. Lippincott, Philadelphia. 4. Petersen OW, van Deurs B (1988): Growth factor control of myoepithelial-cell differentiation in cultures of human mammary gland. Differentiation 39:197-215. 5. ISCN (1991): Guidelines for Cancer Cytogenetics, Supplement to An International System for Human Cytogenetic Nomenclature. Mitelman F (ed.), Karger, Basel. 6. Heim S, Mandahl N, Mitelman F (1988): Genetic convergence and divergence in tumor progression. Cancer Res 48:5911-5916. 7. Atkin NB, Baker MC (1990): Cytogenetic studies of basal cell carcinomas. Cancer Genet Cytogenet 49:281-282. 8. Mertens F, Jin Y, Heim S, Mandahl N, Jonsson N, Mertens O, Persson B, Salemark L, Wennerberg J, Mitelman F (1992): Clonal structural chromosome aberrations in nonneoplastic cells of the skin and upper aerodigestive tract. Genes Chrom Cancer 4: 235-240. 9. Jin Y, Mertens F, Mandahl N, Helm S, Oleg~rd C, Wennerberg J, Bi6rklund A, Mitelman F (1993): Chromosome abnormalities in eighty-three head and neck squamous cell carcinomas: influence of culture conditions on karyotypic pattern. Cancer Res 53:2140-2146. 10. Bullerdiek J, Haubrich J, Meyer K, Bartnitzke S (1988): Translocation t(ll;19)(q21;p13.1)as a sole chromosome abnormality in a cystadenoma (Warthin's tumor) of the parotid gland. Cancer Genet Cytogenet 35:129-132. 11. Mark J, Dahlenfors R, Stenman G, Nordquist A (1989): A human adenolymphoma showing the chromosomal aberrations del(7)(p12p14-15) and t(11;19)(q21;p12-13). Anticancer Res 9: 1565-1566.