Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors

Cancer Genetics and Cytogenetics 140 (2003) 1–12

Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors: dermatofibrosarcoma protuberans and giant cell fibroblastoma Avery A. Sandberga,*, Julia A. Bridgeb,* a Department of DNA Diagnostics, St. Joseph’s Hospital and Medical Center, 350 West Thomas Road, Phoenix, AZ 85013, USA Departments of Pathology/Microbiology, Pediatrics, and Orthopaedic Surgery, University of Nebraska Medical Center, 983135 Nebraska Medical Center, Omaha, NE 68198-3135, USA Received 12 September 2002; received in revised form 4 October 2002; accepted 8 October 2002

b

Dermatofibrosarcoma protuberans was first described in 1924 by Darier and Ferrand [1] in a paper entitled, “Dermatofibrosarcomas progresifs et récidivants ou fibrosarcomas de la peau” (Progressive recurrent dermatofibrosarcomas or skin fibrosarcomas). In the following year, 1925, Hoffmann further described this tumor and called it “Dermatofibrosarkoma protuberans” [2]. 1. Pathologic and clinical aspects of dermatofibrosarcoma protuberans and giant cell fibroblastoma Dermatofibrosarcoma protuberans (DFSP) is a morphologically distinct fibrohistiocytic neoplasm of the deep dermis, which generally is regarded as a low-grade malignant neoplasm of clonal origin [3]. DFSP is characterized clinically by locally aggressive growth and a high rate of local recurrence, but distant metastases (1–4%) [4] and tumorrelated deaths are very rare, unless the tumor shows fibrosarcomatous transformation [5–7]. DFSP affects predominantly young to middle-aged adults (30–50 years) and is seen mainly on the trunk and the proximal extremities. Tumors vary from plaquelike or small nodular lesions to large multinodular masses. The classic histologic features comprise a monotonous storiform or cartwheel growth pattern of uniform and cytologically bland cells with hyperchromatic and elongated nuclei and a characteristic honeycomb pattern of infiltration into subcutaneous fat (Fig. 1). Immunohistochemically, DFSP is characterized by positive staining for vimentin and CD34 [8]. The latter points to a putative derivation from CD34 dendritic cells. The P53 immunoreactivity in DFSP does not appear to be a useful predictor of tumor progression in low-grade lesions [9]. * Corresponding authors. Tel.: (602) 406-3588; fax: (602) 406-4118 (A.A. Sandberg). Tel.: (402) 559-4186; fax: (402) 559-6018 (J.A. Bridge). E-mail addresses: [email protected] (A.A. Sandberg); jbridge@unmc. edu (J.A. Bridge).

Several uncommon histologic variants of DFSP have been delineated that are important to recognize to avoid misdiagnosis with more aggressive neoplasms. Bednar tumor or pigmented DFSP is characterized by the additional presence of melanin-containing dendritic cells [10]. Myxoid DFSP is a rare lesion in which myxoid stromal changes predominate [11]. A further subset of DFSP showing focal myofibroblastic differentiation has been reported [12,13]. Occasionally, giant cell fibroblastoma-like areas are seen in DFSP [14,15], and, rarely, cases of DFSP may show granular cell changes [16,17]; however, these and other variants (e.g., atrophic [18] and palisading [19] variants) simply represent morphologic heterogeneity in DFSP and are not associated with significant differences in clinical behavior. The Bednar tumor initially was thought to be related to a storiform neurofibroma [20] of the skin. On the basis of pathologic similarities and molecular and cytogenetic features [17,21–23], however, Bednar tumor is now considered to be a pigmented variant of DFSP rather than a distinct entity [7]. Among these variants, only the fibrosarcoma (DFSP-FS) variant has a prognostic significance. The fibrosarcomatous transformation of DFSP has been considered a form of higher-grade tumor progression and appears to be associated with an adverse clinical outcome. In 1951, Penner [24] reported a case of metastasizing DFSP containing areas that were histologically indistinguishable from FS, and, during the last two decades, a number of case reports and small series of DFSP with FS-like areas have been published. Although some authors have noted an increased incidence of metastases in the FS variant of DFSP (i.e., DFSP-FS) and concluded that DFSP-FS has a less favorable prognosis than ordinary DFSP, others did not [25]. Nevertheless, the general opinion is that fibrosarcomatous change represents a form of tumor progression in DFSP and is associated with a significantly more aggressive clinical course than in ordinary DFSP, indicating a possible need for treatment intensification in such cases [6]. The

0165-4608/03/$ – see front matter © 2003 Elsevier Science Inc. All rights reserved. PII: S0165-4608(02)00 8 4 8 - 8

2

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

Fig. 1. Histologic appearance of dermatofibrosarcoma protuberans (DFSP) and its fibrosarcomatous variant. (A) Dermal location of DFSP with uninvolved zone beneath atrophic dermis. (B) Characteristic microscopic appearance of DFSP with interwoven fascicles of cells forming a storiform pattern. (C) Fascicular herringbone pattern of growth with frequent mitoses representative of the fibrosarcomatous variant of DFSP. (D) Fibrosarcomatous variant of DFSP infiltrating underlying muscle (Bowne et al., 2000) [86].

presence of fibrosarcoma as a component of DFSP, either in the original tumor or in a recurrence, has received much attention [6,26–31]. Molecular studies for the COL1A1PDGFB transcripts in FS areas of six DFSP revealed them to be present in both the DFSP and FS areas [32].

Giant cell fibroblastoma (GCF) is a clinically and morphologically distinct lesion affecting predominantly infants and children. It manifests itself as a slowly growing, painless superficial mass. Local recurrences occur, but metastases have not been reported to date [33]. There has been

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

continuous speculation on the relation between GCF and DFSP, to the extent that GCF has been called the juvenile form of DFSP [14]. Giant cell fibroblastomas are composed of loosely arranged, wavy spindle cells with a moderate degree of nuclear polymorphisms that infiltrate the deep dermis and subcutis and encircle adnexal structures in a fashion similar to DFSP [7]. The tumors vary in cellularity from those approximating that of DFSP to those that are hypocellular with a myxoid or hyaline stroma. A characteristic feature of GCF is the presence of peculiar pseudovascular spaces, which seem to reflect a loss of cellular cohesion [7]. Giant cell fibroblastomas express vimentin but lack S-100 protein markers [21]. Some GCF express CD34, a feature in common with DFSP. Although both DFSP and GCF are prone to local recurrence, only DFSP has been reported to metastasize. Despite the fact that the two lesions often bear no histologic similarities, in the past a link had been hypothesized. This theory was confirmed by the demonstration of identical chromosome changes [34–37] and molecular events [38] in these tumors. The relationship between DFSP and GCF is further illustrated by a GCF recurring as a DFSP [39], the presence of areas resembling GCF in a DFSP [15], DFSP recurring as a GCF [40], cytogenetic and immunohistochemical evidence that GCF is related to DFSP [35], similar molecular genetic findings [38,41,42], the resemblance of GCF clinicopathologically to DFSP as originally described by Shmookler et al. [14], a GCF with pigmented DFSP [37,43], and some histopathologic aspects of GCF and DFSP [7]. Although CD34 is expressed in a number of mesenchymal neoplasms [8], CD34 staining of the GCF reported by Dal Cin et al. [35] further supports the link with DFSP. 2. Dermatofibroma Both DFSP and GCF are occasionally misdiagnosed as benign lesions such as dermatofibroma and neurofibroma [4,5], leading to improper primary management. The current treatment of DFSP and GCF is entirely surgical, and recurrence rates of up to 50% following removal have been reported. Some comments are appropriate regarding dermatofibroma, with which DFSP may be confused. Dermatofibroma (fibrous histiocytoma) is perhaps the most common mesenchymal growth of the skin. This lesion is composed predominantly of fibroblastlike spindle cells, capillaries, and macrophages within a collagenous stroma, usually associated with hyperplasia and hyperpigmentation of the overlying epidermis. Some authors consider dermatofibroma (DF) to be a benign neoplasm; others contend that it is reactive. The possibility of a spectrum between DF and DFSP has been shown by the studies of Horenstein et al. [44]. Though cytogenetic studies of fibrous histiocytoma have demonstrated clonal changes compatible with a neoplastic process [45–48], as have some other studies [49] based on

3

the expression of PDGFB, controversy continues to exist [50]. Nonetheless, there appears to be agreement that DF can readily be differentiated from DFSP on the basis of the specific cytogenetic and molecular changes in DFSP [22,51]. In the study of Wang et al. [22], fusion transcripts of COL1A1-PDGFB were detected in 10 of 12 DFSP, but in none of the 10 DF examined. The expression of tenascin, an extracellular matrix glycoprotein expressed in fibroblasts and the extracellular matrix during embryogenesis, can be used to differentiate DF from DFSP [52]. Its increased expression at the dermal–epidermal junction overlying the lesion in DF contrasts with the absence of such expression in DFSP. Expression of tenascin within the lesion does not differentiate DF from DFSP; CD34 expression, however, readily distinguishes DFSP from dermatofibroma [53]. 3. Cytogenetic aspects of DFSP and GCF Cytogenetically, DFSP is characterized by a reciprocal translocation, t(17;22)(q22;q13) (Fig. 2), and more often by a supernumerary ring chromosome derived from the translocation r(17;22) (Fig. 3) [54–58] (Table 1). Fluorescence in situ hybridization (FISH) studies have been used to demonstrate the 17;22 chromosomal composition of the ring chromosomes [55,56,59]. Other rare translocations, such as t(2; 17) or t(9;22), have also been reported [60,61]. Dal Cin et al. [35] confirmed that t(17;22)(q2122;q13) also characterizes GCF by demonstrating this translocation in a primary tumor. Moreover, these authors [35] further supported the existence of a relationship between this early childhood tumor and DFSP by finding, at recurrence of the above-mentioned GCF, a der(22)t(17;22)(q2122;q13) leading to partial trisomy 17q combined with strong CD34 expression. Simon et al. [38] have shown that both rings and translocated chromosomes present the same molecular rearrangements. Cloning of the r(17;22) or t(17;22) has revealed that the translocations result in a fusion of two genes, COL1A1 and PDGFB [38]. The COL1A1 gene is located in 17q2122 and encodes the 1(1) chain of type 1 collagen, the most abundant protein in the body, which is produced primarily by fibroblasts. The PDGFB (c-sis proto-oncogene) gene is located in 22q13 and encodes the -chain of platelet-derived growth factor (PDGF) ligand. The fusion causes deregulation of the PDGFB gene by deleting its exon 1 and placing it under the direct control of the COL1A1 gene. This rearrangement leads to an unscheduled production of PDGF, which seems to play an important role in the development of DFSP. Greco et al. [62] provided direct evidence that the rearranged PDGFB gene could transform NIH3T3 cells in a model of an autocrine mechanism. Rarely, a DFSP is seen without a ring chromosome containing material from 17q and 22q or less frequently a t(17; 22)(q22;q13), cytogenetic hallmarks of DFSP. In a recurrent DFSP in a 29-year-old woman, an additional large

4

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

Fig. 2. Karyotype of a giant cell fibroblastoma (GCF) showing a balanced translocation, 46,XY,t(17;22)(q21.33;q13.1), as the sole cytogenetic anomaly (arrows) (Craver et al., 1995) [34]. This cytogenetic change is identical to that seen in DFSP.

marker chromosome was present that included chromosome 8 or the 8q area [6]. To more definitely identify the origin of the rest of this marker chromosome, spectral karyotyping (SKY) analysis was performed. This showed that the remaining part of the marker was complex and composed of segments from chromosomes 7, 8, 17, 21, and 22 with two copies of the 17;22 fusion. Further analysis of this unusual marker by molecular means [63] showed the presence of the COL1A1-PDGFB gene fusion within this marker. Rings or marker chromosomes in DFSP containing additional sequences from chromosomes other than 17 or 22; for example, chromosomes 4 and 12, have been observed in two DFSP cases [64,65]. These results and others in the literature indicate that, in addition to the specific 17;22 fusion, amplifications of material from chromosomes 17, 22, 8, 5, 7, and 21 (in order of frequency) may play a role in DFSP development, progression, or both. The use of SKY is useful in detection of a diagnostically relevant 17;22 fusion in DFSP with unusual karyotypic features [66]. 4. Molecular aspects of DFSP and GCF Detection of COL1A1-PDGFB chimeric mRNA by a reverse transcriptase polymerase chain reaction (RT-PCR) assay has been proven to be a reliable and useful diagnostic marker for DFSP [38,41]. Wang et al. [32] showed that with some refinements, such as prolonged proteinase K treatment and selection of primers amplifying small target sequences,

use of the molecular assay of RT-PCR was also feasible for detecting the fusion transcripts in archival formalin-fixed, paraffin-embedded tissues [22]. All DFSP and GCF cases studied to date have been found to contain a fusion of COL1A1 to PDGFB (Fig. 4), as a result of rearrangements involving chromosomes 17 and 22. The transforming potential of the COL1A1-PDGFB fusion gene has been demonstrated . Shimizu et al. [67] generated a NIH3T3 cell line that expresses a tumor-derived chimeric gene. The cell line has been used to characterize the fusion protein functionally and structurally. The following discussion of the molecular events in DFSP is based on that of Simon et al. [38,68,69], who combined their experience with the cytogenetic aspects of DFSP [55–58] and results obtained through molecular studies. The discussion by Simon et al. [68,69] though speculative and heuristic in some places, presents the most comprehensive exposition of the molecular events in DFSP (Fig. 5). Type 1 collagen is a heterotrimer composed of two 1(1) chains and one 2(1) chain encoded by COL1A1 and COL1A2 genes, respectively [70]. The PDGFB gene encodes the -chain of PDGF, one of the tyrosine kinase receptors on the surface of target cells; it is the cellular homolog of the v-sis oncogene, which is known to cause simian sarcoma [71,72]. PDGFB, a homodimer composed of two -chains, is a potent mitogen for a variety of cells [73]. Although the contribution of PDGFB in tumor genesis has not been fully clarified, the coexpression of PDGFB and

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

5

Fig. 3. Karyotype of a DFSP with 50 chromosomes, including a ring chromosome (r), the latter being the most common cytogenetic change in DFSP and GCF. Trisomies 4, 5, and 8 are present. The karyotype is one of four DFSP with ring chromosomes demonstrated with FISH to contain chromosome 17 sequences [55]. In subsequent years this research group confirmed and extended the nature of these ring chromosomes (Pédeutour et al., 1994) [56–58].

its receptors has been described in many tumors, including DFSP, and the possibility of autocrine PDGFB stimulation in these tumors has been suggested [49,74,75]. To date, in all DFSP cases studied, the t(17;22) resulted in chimeric COL1A1- PDGFB RNA production, in which the PDGFB exon 1 was deleted and replaced by a variable segment of COL1A1 gene (Fig. 5) [22,38,41,62,65]. In more than 35 cases of DFSP studied to date, the reciprocal PDGFB-COL1A1 transcripts have not been described, suggesting a preponderant role for the COL1A1-PDGFB fusion gene [38]. As a consequence of the t(17;22), the sequences, located upstream of PDGFB exon 2 and known to contain repressor elements for PDGFB transcription and translation, are replaced by COL1A1 elements. The loss of PDGFB exon 1 is responsible for v-sis oncogene activation in experimental models [76,77]. One could therefore assume that the t(17; 22) rearrangement causes the production of an autocrine COL1A1-PDGFB growth factor responsible for DFSP development.

As mentioned above, studies on chimeric COL1A1PDGFB in nearly 40 DFSP, GCF, and Bednar tumors have shown that fusion points are widely variable on the COL1A1 sequence, ranging from exon 7 to exon 47 [17,22,32,37,38,41,62,63] (Fig. 5). This could indicate that COL1A1 sequences play a subordinate role in the COL1A1PDGFB protein and are not essential for the potential growth factor activity. In contrast, COL1A1-PDGFB fusions always preserve PDGFB sequences, allowing for synthesis of mature PDGFB protein from the N to the C terminus cleavage sites encoded by exons 3 and 6, respectively [75]. It is possible that the contribution of COL1A1 to the chimeric protein would be only to provide the signal peptide essential for protein export. Analogous to the processing of wild-type PDGFB, the processing of the COL1A1-PDGFB chimeric proteins would further result in the production of mature PDGFB in the tumor cells. To validate this hypothesis, Simon et al. [68,69] generated fibroblastic cell lines expressing a DFSP-derived COL1A1-PDGFB gene fusion. In their report, Simon et al. [68,69] presented the structural and

6

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

Table 1 Chromosome findings in DFSP and GCF Karyotype

Reference

48,XY,8,r/47,XY,8 47,XY,r(1;?)(p36q44;?) 47,XY,r 52,XY,7,11,13,14,15,r 46,XY,der(5)t(5;13)(p11;q11),13,add(17)(p13),18,r /47,XY,idem,r 49,XX,5,8,r 47,XY,4,22,r1,r22 50,XX,4,5,8,r 7495,XXXX,2,3,4,5,7,10,10,11,11,12,12,15,19, 20,20,r,r,mar1,mar2,mar3,mar4,dmin 46,XY,t(2;17)(q33;q25) 46,XY,t(X;7)(q21.2;q11.2) 46,XY,t(17;22)(q21.33;q13.1)a 47,XX,r 47,XY,r 47,XY,22,2mar 47,XX,r 47,XY,der(3)(q21),r 49,XY,5,8,r XY, polyploid, 2 r 47,XY,8/48,XY,8,r 46,XY,t(17;22)(q2122;q13)/46,XY,der(22)t(17;22)(q2122;q13)a 48,XY,8,r/48,XY,r,mar 46,XX,t(17;22)(q22;q13),der(22)t(17;22)(q22;q13) 4751,XY,18,2der(2)t(17;22)(q22;q13),mar1,mar2 46,XY,der(22)t(17;22)(q21;q13)a 48,XY,der(?)t(?;12)(?;q15),r/50,idem,15,20/ 46,XY,ish der(22)t(17;22)t(12;17 or 22)(wcp12,HMGIC, 2196B2,wcp17,COL1A1 amp,wcp22,XXT680,PDGF3–4), r(5)(p15q35)(wcp5,D5Z1,84C11,B22A4) 47,XX,r 48,XX,5,r/47,XX,5 47,XY,r 47,XY,r/46,idem,der(5)t(5;13)(p13;q12),13,22,mar 47,XX,12,22,r 49,XX,X,5,r 49,XY,8,2r 5961,XX,X,add(5)(p13),7,i(8)(q10),9,10,13,14,15,18, del(22)(q11),2r,mar 4748,XX,8,17,der(22)t(17;22)(q22;q13) 46,XX,t(9;22)(q32;q12.2) 47,XX,r 47,XX,rb 47,XY,rc 47,XX,dic(8;?)(p11;?) or der(?;8)(?→cen→?::?→cen→:: 8q11→8qter)/47,XX,dic(8)(p23;?)c 47,XX,der(8;21;?17)(7?pter→7?p21::21q22→21q11 or 21q11→21q22::22q11→22q13::17q22→17q2425::?17? → cen→?17?::21q11→21q22:22q11→22q13::17q22→17q2425:: ?17?::21p11→cen→21q1121::8p11→cen→8qterd,e 46,XY,der(22)t(17;22)(q22;q13)a

Bridge et al., 1990 [95] Mandahl et al., 1990 [96] Örndal et al., 1992 [97] Stephenson et al., 1992 [98] Pédeutour et al., 1993 [99] Pédeutour et al., 1994,1995 [55,56] Pédeutour et al., 1994 [55] Pédeutour et al., 1994 [55] Pédeutour et al., 1994 [55] Sinovic and Bridge, 1994 [60] Craver et al., 1995 [100] Craver et al., 1995 [100] Minoletti et al., 1995 [59] Minoletti et al., 1995 [59] Minoletti et al., 1995 [59] Naeem et al., 1995 [54] Naeem et al., 1995 [54] Naeem et al., 1995 [54] Pédeutour et al., 1995 [56] Stenman et al., 1995 [101] Dal Cin et al., 1996 [35] Mandahl et al., 1996 [102] Pédeutour et al., 1996 [58] Pédeutour et al., 1996 [58] Dal Cin et al., 1997 [36]

Gisselsson et al., 1998 [64] Iwasaki et al., 1998 [103] Navarro et al., 1998 [65] O’Brien et al., 1998 [41] O’Brien et al., 1998 [41] O’Brien et al., 1998 [41] O’Brien et al., 1998 [41] O’Brien et al., 1998 [41] O’Brien et al., 1998 [41] Dobin et al., 1999 [104] Sonobe et al., 1999 [61] Vanni et al., 2000 [42] Nishio et al., 2001 [23] Nishio et al., 2001 [105] Mrózek et al., 2001 [66]

Mrózek et al., 2001 [66] Maire et al., 2002 [37]

a

Giant cell fibroblastoma. Bednar tumor. c Karyotype based on Giemsa banding. d Karyotype based on Spectral karyotyping SKY, Giemsa banding and inverted and contrast-enhanced 4,6-diamidine-2-phenylindol DAPI staining. e For another interpretation of the marker, see original paper [66]. b

functional characterization of the COL1A1-PDGFB proteins synthesized in COL1A1-PDGFB expressing cells. The results of Simon et al. [68,69] showed that transfected cell lines expressing the tumor-derived T94796 COL1A1-

PDGFB sequence became independent of growth factors, including PDGFB, and induced tumor formation in nude mice. Using specific anti–COL1A1-PDGFB and anti-PDGFB antibodies, Simon et al. [68,69] showed that cells expressing

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

7

Fig. 4. Breakpoints in COL1A1 and PDGFB in DFSP. Shown is a partial genomic structure of the COL1A1 gene from exon 1 to exon 46, with the appropriate location of the breakpoints in eight DFSP indicated by arrows pointing to the various exons affected. The breakpoint in PDGFB gene usually involves exon 2 solely. Open boxes represent untranslated regions (Wang et al., 1999) [22].

COL1A1-PDGFB produced 116-kD chimeric COL1A1PDGFB chains, which probably formed dimers and were processed to yield a number of cleavage products, including 30-kD PDGFB-like secreted components. These latter species could correspond to those previously characterized in studies on PDGFB processing [78]. Indeed, PDGFB is produced as 56-kD PDGFB precursor dimers first cleaved at the N-terminal level to yield cell-associated 40-kD dimers. A second cleavage on the C-terminal side releases secreted 30-kD mature dimers. The data led Simon et al. [68,69] to propose that the COL1A1-PDGFB chimeric proteins expressed in COL1A1-PDGFB transfected cells are processed to yield active PDGFB-like dimers. In addition, Simon et al. [68,69] showed that the COL1A1-PDGFB stable clones contained activated PDGF -receptors and that the conditioned media from COL1A1-PDGFB transfected cells were able to stimulate PS200 cell growth. Anti-PDGFB neutralized this effect. Simon et al. [68,69] established that COL1A1-PDGFB transfected clones produced PDGFB-like molecules that could stimulate cell growth in an autocrine or paracrine way through PDGF -receptors activation. To prove that the PDGFB-like species detected in COL1A1-PDGFB transfected cells are effectively produced by COL1A1-PDGFB protein cleavage, Simon et al. [68,69] mutated the COL1A1-PDGFB putative proteolytic cleavage sites, including the PDGFB maturation sites. The effects of the mutations were tested in transient transfection experiments using the HEK293 cell line. The results not only confirmed that COL1A1-PDGFB transfected cells produced PDGFB-like growth factors in media, but also demonstrated that these PDGFB-like molecules resulted from the proteolysis of the COL1A1-PDGFB fusion proteins. The results of Simon et al. [68,69] are somewhat at variance with those published by Greco et al. [62] and Shimizu

et al. [67]. Greco et al. [62] showed that COL1A1-PDGFBexpressing cells produced a PDGFB-like activity clearly involved in cell transformation, but they did not demonstrate the presence of either mature PDGFB or chimeric COL1A1PDGFB precursor proteins in COL1A1-PDGFB transfected cells. Shimizu et al. [67] showed the production of a chimeric COL1A1-PDGFB protein and the accumulation of cellassociated 40-kD and 24-kD PDGFB forms in COL1A1PDGFB transfected cells, suggesting that the chimeric COL1A1-PDGFB protein was processed in these cells as is also wild-type PDGFB. They did not, however, detect PDGFB secreted molecules in cell media. The results of Simon et al. [68,69] not only provide evidence that the chimeric COL1A1-PDGFB protein processed in COL1A1PDGFB transfected cells yields PDGFB-like molecules but also demonstrated the presence of a 30-kD secreted PDGFB form in cell media expressing COL1A1-PDGFB. Moreover, mutagenesis experiments suggested that unprocessed chimeric COL1A1-PDGFB species could also play a role in the cell growth stimulation. Surprisingly, when both the cleavage of the COL1A1-PDGFB fusion proteins and the release of PDGFB in cells were prevented, the mitogenic potential of the produced molecules was not modified. Although Simon et al. [68,69] could not completely exclude the presence of nonspecific cleavages of COL1A1PDGFB and the production of atypical but active PDGFB forms, this could mean that whole uncleaved chimeric COL1A1-PDGFB dimers could act as mitogenic factors. Indeed, one could expect that, as for PDGFC, a newly described protease-activated PDGF family member [79], the N-terminal collagen domain of the chimeric COL1A1-PDGFB protein inhibited its mitogenic function. PDGFC is a secreted protein and has a remarkable two-domain structure not previously observed in this family of growth factors, with an

8

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

Fig. 5. Schematic presentation of the molecular events associated with the COL1A1-PDGFB gene fusion in DFSP or GCF according to Simon et al [69]. This scheme is to serve as a guide in the discussion related to the molecular events presented by these authors and outlined in some detail in the text. In (A) is shown the chimeric COL1A1-PDGFB cDNA map. In (B) is shown the deduced chimeric protein map. In (C) is shown the deduced chimeric protein map compared with the wild-type proteins. COL1A1 sequences are shown in white and those of PDGFB as shaded areas. The black areas and white areas on the left of the chimeric protein in (C) indicate the COL1A1-PDGFB signal peptide sequences, respectively. Arrows indicate mutation points on the COL1A1-PDGFB sequence; arrowheads point to proteolytic cleavage sites. C5 and P5 refer to respective mutated pcDNA 3.1-T94796 construction. Numbers to the left of arrowheads indicate the amino acid (aa) position on the chimeric COL1A1-PDGFB protein sequence, which begins at the first methionine of the consensus COL1A1 sequence (Simon et al., 2001) [69].

N-terminal CUB domain and a C-terminal PDGF domain. Whole PDGFC is nonfunctional and the PDGF domain is activated by proteolytic cleavage of the CUB domain. The CUB domain is present in several membrane-bound and secreted proteins, and it has been hypothesized that the CUB domain of PDGFC may bind to the pericellular matrix as well as having an inhibitory function in receptor binding and activation. In contrast, in the model of Simon et al. [68,69], the N-termi-

nal collagen domain of COL1A1-PDGFB did not seem to inhibit the PDGF activity. The observation that the whole unprocessed COL1A1PDGFB chimera is able to activate PDGF -receptors and to stimulate cell proliferation suggests that the production of mature PDGFB is not necessary for its mitogenic potential [68,69]. However, except when the normal proteolytic cleavage was prevented by mutation, the full 116-kD COL1A1-

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

PDGFB chimeric chain produced in T94796-cDNA transfected cells was not detected in the extracellular cell compartment to yield secreted mature PDGFB dimers. Thus, if the whole COL1A1-PDGFB protein activates PDGF -receptors in COL1A1-PDGFB stably transfected cells, one could hypothesize that it occurs in an intracellular way as is assumed in v-sis–transformed cells [80]. The PDGF receptors pathway would be activated in a classical autocrine or paracrine way as well as by an intracellular mode. Both cleaved and uncleaved forms of the chimeric PDGFB proteins could cooperate in uncontrolled COL1A1-PDGFB cell growth. The simplest hypothesis to explain the role played by the collagen sequences is that COL1A1 sequences act as an impetus for PDGFB expression. In DFSP cells, the t(17;22) results in the juxtaposition of a COL1A1 fragment to PDGFB and in the loss of the negative regulators of PDGFB transcription and translation [68,69]. This is also the case in the simian sarcoma virus model where the v-sis activated oncogene lacks the regulatory sequences. In DFSP, an active COL1A1 promoter and COL1A1 exons replace the PDGFB negative regulatory elements and results in the production of a high level of chimeric COL1A1-PDGFB mRNAs. In this context, the t(17;22) rearrangement present in DFSP could be considered as a novel way to activate the expression of PDGFB. The COL1A1 domain of the chimeric protein also provides the signal peptide, which is a prerequisite for the proper export of the protein [68,69]. The wide size variation of the collagen domain juxtaposed to the PDGFB domain in the various proteins so far identified indicate that the collagen sequence may not play a crucial role in the COL1A1PDGFB growth factor activity. The fact that the full chimeric COL1A1-PDGFB structure has preserved mitogenic action via the PDGF -receptors suggests that the PDGFB entity of the chimera behaves as an independent growth factor domain. Studies have provided convincing evidence for direct activation of discoidin domain receptor (DDR) tyrosine kinases DDR1 and DDR2 by various collagens, including type 1 collagen [81,82]. Moreover, DDR1 and DDR2 were found to be overexpressed in a variety of human primary tumors, suggesting that these receptors might play a role in tumor formation [83]. Simon et al. [68,69] showed that the processing of the chimeric T94996 COL1A1-PDGFB protein into PDGFB dimers also resulted in the production of significant quantities of 100-kD collagenlike components in the extracellular medium. These molecules are mainly composed of the N-terminal collagen fragment of the chimeric COL1A1-PDGFB protein and retain the putative proteolytic cleavage site, which corresponds to the COL1A1 chain N-terminus propeptide removal during collagen fiber formation. In fact, the COL1A1 chains, associated as trimers with COL1A1 chains, are secreted in the extracellular medium and are cleaved at the N- and C termini before they form mature collagen fibers [84]. Thus, it is possible that the 100-

9

kD collagenlike molecules could be processed in the extracellular medium and could participate in anomalous collagen fiber formation. Several mechanisms could cooperate in the establishment of a transformed phenotype in DFSP cells expressing COL1A1-PDGFB: one, the more likely, is through the activation of the PDGF -receptor pathway by overexpressing either mature PDGFB or whole chimeric COL1A1-PDGFB chains; the second, more hypothetical mechanism is through DDR activation by producing high quantities of collagenlike structures [68,69]. 5. Microarray studies in DFSP A study [85] of 9 DFSP using 42,000 spotted cDNA microarrays showed high expression of a large and distinct group of genes that were expressed at much lower levels in a group of 29 other spindle-cell tumors (monophasic synovial sarcomas, gastrointestinal stromal tumors, leiomyosarcomas, and a fibromatosis). DFSP showed high expression of a large distinct group of genes that were expressed at much lower levels in the other tumors. This group included PDGFB and one of its modulators, osteonectin. Other genes, such as AKT1 and STAT3, that are normally activated by PDGF were found to be downregulated compared with the levels in the other soft-tissue tumors. Linn et al. [85] stated that microarrays are a useful tool in the characterization of molecular changes in DFSP and can highlight unexpected features in molecular pathways that may be targeted for rationally designed therapy. 6. Some clinical aspects of DFSP The prognosis after surgical resection with negative and sometimes positive microscopic margins for patients with DFSP is very good [86]; however, increased age, high mitotic index, and increased cellularity are predictors of poor outcome. The DFSP-FS variant represents a much more aggressive tumor, with metastatic potential [87,88]. Patients who are treated with curative intent for DFSP-FS should undergo aggressive attempts at complete surgical resection. Patients with recurrent classic DFSP without evidence of adverse prognostic features may benefit from conservative management, especially in the setting of potentially unresectable disease. Differentiation of dermatofibroma from DFSP, though it can readily be established through genetic analyses, continues to be the subject of other approaches, such as the use of stromelysin 3 expression [89]. Cytogenetic, fluorescence in situ hybridization (FISH), and molecular analyses for the establishment or confirmation of the diagnosis of DFSP are likely to find crucial clinical application in the future, particularly in problem cases [90,91]. The possible effectiveness of STI571 (imatinib), a PDGF tyrosine kinase inhibitor, in the treatment of DFSP has been described [92]. The translocation involving 22q13.3 (the region for the PDGFB gene) was not with chromosome 17

10

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12

(FISH studies). The administration of STI571 to patients with metastatic sarcoma arising from DFSP has been recently reported [93]. One patient failed to respond and the other, at the time of the report, showed a partial but ongoing response with resolution of superior vena cava syndrome and shrinking of metastatic lung lesions. Another report [94] indicated that STI571 may be useful for patients with locally advanced DFSP when other options for local therapy are limited. Acknowledgments All figures have been reproduced with appropriate permissions. The authors wish to thank Mrs. Molly Harrington and Mrs. Irma Contreras of the St. Joseph’s Hospital and Medical Center for assistance in literature search and Kimberly Christian of the University of Nebraska Medical Center. This work was supported in part by the John A. Wiebe, Jr. Children’s Health Care Fund, Nebraska Department of Health LB595 and NIH/NCI P30CA 36727. Jan Vaughan, as usual, deserves our thanks for the preparation of the manuscript. References [1] Darier J, Ferrand M. Dermatofibrosarcomes progressifs et récidivants ou fibrosarcomes de la peau. Ann Dermatol Syphiligr 1924;5: 545–8. [2] Hoffmann E. Über das knollentreibende Fibrosarkom der Haut (Dermatofibrosarkoma protuberans). Dermatol Z 1925;43:1. [3] Allan AE, Tsou HC, Harrington A, Stasko T, Lee X, Si SP, Grande DJ, Peacocke M. Clonal origin of dermatofibrosarcoma protuberans. J Invest Dermatol 1993;100:99–102. [4] Rutgers EJ, Kroon BB, Albus LC, Gortzak E. Dermatofibrosarcoma protuberans: treatment and prognosis. Eur J Surg Oncol 1992;18: 241–8. [5] Fish FS. Soft tissue sarcomas in dermatology. Dermatol Surg 1996; 22:268–73. [6] Mentzel T, Beham A, Katenkamp D, Dei Tos AP, Fletcher CDM. Fibrosarcomatous (“high-grade”) dermatofibrosarcoma protuberans: clinicopathologic and immunohistochemical study of a series of 41 cases with emphasis on prognostic significance. Am J Surg Pathol 1998;22:576–87. [7] Weiss SW, Goldblum JR. Enzinger and Weiss’s soft tissue tumors, 4th edition. St. Louis: Mosby, 2001, pp 491–516. [8] Van de Rijn M, Rouse RV. CD34: a review. Appl Immunohistochem 1994;2:71–80. [9] Goldblum JR, Frank TS, Poy EL, Weiss SW. p53 Mutations and tumor progression in well-differentiated liposarcoma and dermatofibrosarcoma protuberans. Int J Surg Pathol 1995;3:35–42. [10] Dupree WB, Langloss JM, Weiss SW. Pigmented dermatofibrosarcoma protuberans (Bednar tumor): a pathologic, ultrastructural, and immunohistochemical study. Am J Surg Pathol 1985;9:630–9. [11] Frierson HF, Cooper PH. Myxoid variant of dermatofibrosarcoma protuberans. Am J Surg Pathol 1983;7:445–50. [12] Calonje E, Fletcher CDM. Myxoid differentiation in dermatofibrosarcoma protuberans and its fibrosarcomatous variant: clinicopathologic analysis of 5 cases. J Cutan Pathol 1996;23:30–6. [13] O’Connell JX, Trotter MJ. Fibrosarcomatous dermatofibrosarcoma protuberans: a variant. Mod Pathol 1996;9:273–8.

[14] Shmookler BM, Enzinger FM, Weiss SW. Giant cell fibroblastoma: a juvenile form of dermatofibrosarcoma protuberans. Cancer 1989; 64:2154–61. [15] Beham A, Fletcher CDM. Dermatofibrosarcoma protuberans with areas resembling giant cell fibroblastoma: report of two cases. Histopathology 1990;17:165–82. [16] Banerjee SS, Harris M, Eyden BP, Hamid BNA. Granular cell variant of dermatofibrosarcoma protuberans. Histopathology 1990;17: 375–8. [17] Maire G, Pédeutour F, Coindre J-M. COL1A1-PDGFB gene fusion demonstrates a common histogenetic origin for dermatofibrosarcoma protuberans and its granular cell variant. Am J Surg Pathol 2002;26:932–7. [18] Zelger BW, Öfner D, Zelger BG. Atrophic variants of dermatofibroma and dermatofibrosarcoma protuberans. Histopathology 1995; 26:519–27. [19] Llatjós R, Fernández-Figueras MT, Díaz-Cascajo C, Ribera M, Ariza A. Palisading and Verocay body-prominent dermatofibrosarcoma protuberans: a report of three cases. Histopathology 2000;37:452–5. [20] Bednar B. Storiform neurofibromas of the skin, pigmented and nonpigmented. Cancer 1957;10:368. [21] Fletcher CD. Giant cell fibroblastoma of soft tissue: a clinico-pathologic and immunohistochemical study. Histopathology 1988;13: 499–508. [22] Wang J, Hisaoka M, Shimajiri S, Morimitsu Y, Hashimoto H. Detection of COL1A1-PDGFB fusion transcripts in dermatofibrosarcoma protuberans by reverse transcription-polymerase chain reaction using archival formalin-fixed, paraffin-embedded tissues. Diagn Mol Pathol 1999;8:113–9. [23] Nishio J, Iwasaki H, Ishiguro M, Ohjimi Y, Yo S, Isayama T, Naito M, Kikuchi M. Supernumerary ring chromosome in Bednar tumor (pigmented dermatofibrosarcoma protuberans) is composed of interspersed sequences from chromosomes 17 and 22: a fluorescence in situ hybridization and comparative genomic hybridization analysis. Genes Chromosomes Cancer 2001;30:305–9. [24] Penner DW. Metastasizing dermatofibrosarcoma protuberans: a case report. Cancer 1951;4:1083–6. [25] Wrotnowski U, Cooper PH, Shmookler BJ. Fibrosarcomatous change in dermatofibrosarcoma protuberans. Am J Surg Pathol 1988;12:287–93. [26] Goldblum JR. CD34 positivity in fibrosarcomas which arise in dermatofibrosarcoma protuberans. Arch Pathol Lab Med 1995;119: 238–41. [27] Sato N, Kimura K, Tomita Y. Recurrent dermatofibrosarcoma protuberans with myxoid and fibrosarcomatous changes paralleled by loss of CD34 expression. J Dermatol 1995;22:665–72. [28] Hisaoka M, Okamoto S, Morimitsu Y, Tsuji S, Hashimoto H. Dermatofibrosarcoma protuberans with fibrosarcomatous area: molecular abnormalities of the p53 pathway in fibrosarcomatous transformation of dermatofibrosarcoma protuberans. Virchows Arch 1998; 433:323–9. [29] Morimitsu Y, Hisaoka M, Okamoto S, Hashimoto H, Ushijima M. Dermatofibrosarcoma protuberans and its fibrosarcomatous variant with areas of myxoid differentiation: a report of three cases. Histopathology 1998;32:547–51. [30] Hamada M, Hirakawa N, Fukuda T, Furue M, Hori Y, Tsuneyoshi M. A progression to dermatofibrosarcoma protuberans with a fibrosarcomatous component: A special reference to the chromosomal aberrations. Pathol Res Pract 1999;195:451–60. [31] Goldblum JR, Reith JD, Weiss SW. Sarcomas arising in dermatofibrosarcoma protuberans: a reappraisal of biologic behavior in eighteen cases treated by wide local excision with extended clinical follow up. Am J Surg Pathol 2000;24:1125–30. [32] Wang J, Morimitsu Y, Okamoto S, Hisaoka M, Ishida T, Sheng W, Hashimoto H. COL1A1-PDGFB fusion transcripts in fibrosarcomatous areas of six dermatofibrosarcoma protuberans. J Mol Diagn 2000;2:47–52.

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12 [33] Enzinger FM, Weiss SW. Soft tissue tumors, 3rd edition. St. Louis: Mosby–Year Book, 1995. [34] Craver RD, Correa H, Kao YS, Van Brunt T, Golladay ES. Aggressive giant cell fibroblastoma with a balanced 17;22 translocation. Cancer Genet Cytogenet 1995;80:20–2. [35] Dal Cin P, de Wever I, Brock P, Casteels-Van Daele M, Van Damme B, Van den Berghe H. Cytogenetic and immunohistochemical evidence that giant cell fibroblastoma is related to dermatofibrosarcoma protuberans. Genes Chromosomes Cancer 1996;15:73–5. [36] Dal Cin P, Polito P, Van Eyken P, Sciot R, Hernandez JM, Garcia JL, Van den Berghe H. Abnormalities of chromosomes 17 and 22 in giant cell fibroblastoma. Cancer Genet Cytogenet 1997;97:165–6. [37] Maire G, Martin L, Michalak-Provost S, Gattas GJF, Turc-Carel C, Lorette G, Pédeutour F. Fusion of COL1A1 exon 29 with PDGFB exon 2 in a der(22)t(17;22) in a pediatric giant cell fibroblastoma with a pigmented Bednar tumor component: evidence for age-related chromosomal pattern in dermatofibrosarcoma protuberans and related tumors. Cancer Genet Cytogenet 2002;134:156–61. [38] Simon M-P, Pédeutour F, Sirvent N, Grosgeorge J, Minoletti F, Coindre J-M, Terrier-Lacombe M-J, Mandahl N, Craver RD, Blin N, Sozzi G, Turc-Carel C, O’Brien KP, Kedra D, Fransson I, Guilbaud C, Dumanski JP. Deregulation of the platelet-derived growth factor @@@B-chain gene via fusion with collagen gene COL1A1 in dermatofibrosarcoma protuberans and giant-cell fibroblastoma. Nat Genet 1997;15:95–8. [39] Alguacil-Garcia A. Giant cell fibroblastoma recurring as dermatofibrosarcoma protuberans. Am J Surg Pathol 1991;15:798–801. [40] Coyne J, Kaftan SM, Craig RDP. Dermatofibrosarcoma protuberans recurring as a giant cell fibroblastoma. Histopathology 1992;21: 184–7. [41] O’Brien KP, Seroussi E, Dal Cin P, Sciot R, Mandahl N, Fletcher JA, Turc-Carel C, Dumanski JP. Various regions within the alphahelical domain of the COL1A1 gene are fused to the second exon of the PDGFB gene in dermatofibrosarcomas and giant-cell fibroblastomas. Genes Chromosomes Cancer 1998;23:187–93. [42] Vanni R, Faa G, Dettori T, Melis GB, Dumanski JP, O’Brien KP. A case of dermatofibrosarcoma protuberans of the vulva with a COL1A1/PDGFB fusion identical to a case of giant cell fibroblastoma. Virchows Arch 2000;437:95–100. [43] Zámecník M, Michal M. Giant-cell fibroblastoma with pigmented dermatofibrosarcoma protuberans component. Am J Surg Pathol 1994;18:736–40. [44] Horenstein MG, Prieto VG, Nuckols JD, Burchette JL, Shea CR. Indeterminate fibrohistiocytic lesions of the skin: is there a spectrum between dermatofibroma and dermatofibrosarcoma protuberans? Am J Surg Pathol 2000;24:996–1003. [45] Seruca R, Carneiro F, Soares P, Fonseca E, Castedo S. Clonal cytogenetic abnormalities and telomeric associations in a fibroxanthoma of the stomach. Genes Chromosomes Cancer 1992;5:407–9. [46] Vanni R, Marras S, Faa G, Licheri S, Daniele GM, Fletcher CDM. Cellular fibrous histiocytoma of the skin: evidence of a clonal process with different karyotype from dermatofibrosarcoma. Genes Chromosomes Cancer 1997;18:314–7. [47] Vanni R, Fletcher CDM, Sciot R, Dal Cin P, de Wever I, Mandahl N, Mertens F, Mitelman F, Rosai J, Rydholm A, Tallini G, Van den Berghe H, Willén H. Cytogenetic evidence of clonality in cutaneous benign fibrous histiocytomas: a report of the CHAMP Study Group. Histopathology 2000;37:212–7. [48] Calonje E. Is cutaneous benign fibrous histiocytoma (dermatofibrosarcoma) a reactive inflammatory process or a neoplasm? Histopathology 2000;37:278–80. [49] Smits A, Funa K, Vassbotn FS, Beausang-Linder M, af Ekenstam F, Heldin C-H, Westermark B, Nistér M. Expression of platelet-derived growth factor and its receptors in proliferative disorders of fibroblastic origin. Am J Pathol 1992;140:639–48. [50] Zelger BG, Zelger B. Dermatofibroma (fibrous histiocytoma): an inflammatory or neoplastic disorder? Histopathology 2001;38:378–81.

11

[51] Nishio J, Iwasaki H, Ohjimi Y, Ishiguro M, Isayama T, Naito M, Iwashita A, Kikuchi M. Overrepresentation of 17q22qter and 22q13 in dermatofibrosarcoma protuberans but not in a dermatofibroma: a comparative genomic hybridization study. Cancer Genet Cytogenet 2002;132:102–8. [52] Kahn HJ, Fekete E, From L. Tenascin differentiates dermatofibroma from dermatofibrosarcoma protuberans: comparison with CD34 and factor XIIIa. Hum Pathol 2001;32:50–6. [53] Kutzner H. Expression of the human progenitor cell antigen CD34 (HPCA-1) distinguishes dermatofibrosarcoma protuberans from fibrous histiocytoma in formalin-fixed, paraffin-embedded tissue. J Am Acad Dermatol 1993;28:613–7. [54] Naeem R, Lux ML, Huang S-F, Naber SP, Corson JM, Fletcher JA. Ring chromosomes in dermatofibrosarcoma protuberans are composed of interspersed sequences from chromosomes 17 and 22. Am J Pathol 1995;147:1553–8. [55] Pédeutour F, Coindre J-M, Sozzi G, Nicolo G, Leroux A, Toma S, Miozzo M, Bouchot C, Hecht F, Ayraud N, Turc-Carel C. Supernumerary ring chromosomes containing chromosome 17 sequences: a specific feature of dermatofibrosarcoma protuberans? Cancer Genet Cytogenet 1994;76:1–9. [56] Pédeutour F, Simon M-P, Minoletti F, Sozzi G, Pierotti MA, Hecht F, Turc-Carel C. Ring 22 chromosomes in dermatofibrosarcoma protuberans are low-level amplifiers of chromosome 17 and 22 sequences. Cancer Res 1995;55:2400–3. [57] Pédeutour F, Lacour JP, Perrin C, Huffermann K, Simon MP, Ayraud N, Turc-Carel C. Another case of t(17;22)(q22;q13) in an infantile dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1996;89:175–6. [58] Pédeutour F, Simon MP, Minoletti F, Barcelo G, Terrier-Lacombe MJ, Combemale P, Sozzi G, Ayraud N, Turc-Carel C. Translocation, t(17;22)(q22;q13), in dermatofibrosarcoma protuberans: a new tumor-associated chromosome rearrangement. Cytogenet Cell Genet 1996;72:171–4. [59] Minoletti F, Miozzo M, Pédeutour F, Sard L, Pilotti S, Azzarelli A, Turc-Carel C, Pierotti MA, Sozzi G. Involvement of chromosomes 17 and 22 in dermatofibrosarcoma protuberans. Genes Chromosomes Cancer 1995;13:62–5. [60] Sinovic J, Bridge JA. Translocation (2;17) in recurrent dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1994;75:156–7. [61] Sonobe H, Furihata M, Iwata J, Ohtsuki Y, Chikazawa M, Taguchi T, Shimizu K. Dermatofibrosarcoma protuberans harboring t(9; 22)(q32;q12.2). Cancer Genet Cytogenet 1999;110:14–8. [62] Greco A, Fusetti L, Villa R, Sozzi G, Minoletti F, Mauri P, Pierotti MA. Transforming activity of the chimeric sequence formed by the fusion of collagen gene COL1A1 and the platelet derived growth factor b-chain gene in dermatofibrosarcoma protuberans. Oncogene 1998;17:1313–9. [63] Maire G, Pédeutour F, Mrózek K, Rys J, Iliszko M, Limon J. COL1A1-PDGFB gene fusion in dermatofibrosarcoma protuberans: molecular analysis of a case with an unusual large marker containing sequences from chromosomes 7, 8, 17, 21, and 22. Cancer Genet Cytogenet 2002;135:197–9. [64] Gisselsson D, Höglund M, O’Brien KP, Dumanski JP, Mertens F, Mandahl N. A case of dermatofibrosarcoma protuberans with a ring chromosome 5 and a rearranged chromosome 22 containing amplified COL1A1 and PDGFB sequences. Cancer Lett 1998;133:129–34. [65] Navarro M, Simon M-P, Migeon C, Turc-Carel C, Pédeutour F. COL1A1-PDGFB fusion in a ring chromosome 4 found in a dermatofibrosarcoma protuberans. Genes Chromosomes Cancer 1998; 23:263–6. [66] Mrózek K, Iliszko M, Rys J, Babinska M, Niezabitowski A, Bloomfield DE, Limon J. Spectral karyotyping reveals 17;22 fusions in a cytogenetically atypical dermatofibrosarcoma protuberans with a large marker chromosome as a sole abnormality. Genes Chromosomes Cancer 2001;31:182–6. [67] Shimizu A, O’Brien KP, Sjöblom T, Pietras K, Buchdunger E, Col-

12

[68]

[69]

[70] [71]

[72]

[73]

[74]

[75] [76]

[77]

[78]

[79]

[80]

[81]

[82] [83]

[84] [85]

A.A. Sandberg, J.A. Bridge / Cancer Genetics and Cytogenetics 140 (2003) 1–12 lins VP, Heldin C-H, Dumanski JP, Östman A. The dermatofibrosarcoma protuberans-associated collagen type I1/platelet-derived growth factor (PDGF) B-chain fusion gene generates a transforming protein that is processed to functional PDGF-BB. Cancer Res 1999;59:3719–23. Simon M-P, Navarro M, Roux D, Terrier-Lacombe M-J, Coindre J-M, Pouysségur J. Transforming properties of chimerical protein COL1A1PDGFB generated by dermatofibrosarcoma protuberans-associated translocation t(17;22)(q22;q13.1). Cancer Genet Cytogenet 2001;128:82. Simon M-P, Navarro M, Roux D, Pouysségur J. Structural and functional analysis of a chimeric protein COL1A1-PDGFB generated by the translocation t(17;22)(q22;q13.1) in dermatofibrosarcoma protuberans (DP). Oncogene 2001;20:2965–75. Dalgleish R. The human type I collagen mutation database. Nucleic Acids Res 1997;25:181–7. Doolittle RF, Hunkapiller MW, Hood LE, Devare SG, Robbins KC, Aaronson SA, Antoniades HN. Simian sarcoma virus onc gene, v-sis, derived from the gene (or genes) encoding a platelet-derived growth factor. Science 1983;221:275–7. Waterfield MD, Scrace GT, Whittle N, Stroobant P, Johnsson A, Wasteson Å, Westermark B, Heldin C-H, Huang JS, Deuel TF. Platelet-derived growth factor is structurally related to the putative transforming protein p28sis of simian sarcoma virus. Nature 1983;304:35–9. Heldin CH, Ostman A, Ronnstrand L. Signal transduction via platelet-derived growth factor receptors. Biochim Biophys Acta 1998; 1378:F79–113. Kikuchi K, Soma Y, Fujimoto M, Kadono T, Sato S, Abe M, Ohhara K, Takehara K. Dermatofibrosarcoma protuberans: increased growth response to platelet-derived growth factor BB in cell culture. Biochem Biophys Res Commun 1993;196:409–15. Westermark B, Sorg C. Cytokines, Vol. 5. Basel: Karger S, 1993, pp 1–167. Rao CD, Pech M, Robbins KC, Aaronson SA. The 5 untranslated sequence of the c-sis/platelet-derived growth factor 2 transcript is a potent translational inhibitor. Mol Cell Biol 1988;8:24–92. Dirks RP, Jansen HJ, Onnekink C, De Jonge RJ, Bloemers HP. DNase-I-hypersensitive sites located far upstream of the human c-sis/ PDGF-B gene comap with transcriptional enhancers and a silencer and are preceded by (part of) a new transcription unit. Eur J Biochem 1993;216:487–95. Östman A, Thyberg J, Westermark B, Heldin C-H. PDGF-AA and PDGF-BB biosynthesis: proprotein processing in the Golgi complex and lysosomal degradation of PDGF-BB retained intracellularly. J Cell Biol 1992;118:509–19. Li X, Ponten A, Aase K, Karlsson L, Abramsson A, Uutela M, Backstrom G, Hellstrom M, Bostrom H, Li H, Soriano P, Betsholtz C, Heldin CH, Alitalo K, Ostman A, Eriksson U. PDGF-C is a new protease-activated ligand for the PDGF alpha-receptor. Nat Cell Biol 2000;2:302–9. Fleming TP, Matsui T, Molloy CJ, Robbins KC, Aaronson SA. Autocrine mechanism for v-sis transformation requires cell surface localization of internally activated growth factor receptors. Proc Natl Acad Sci USA 1989;86:8063–7. Shrivastava A, Radziejewski C, Campbell E, Kovac L, McGlynn M, Ryan TE, Davis S, Goldfarb MP, Glass DJ, Lemke G, Yancopoulos GD. An orphan receptor tyrosine kinase family whose members serve as nonintegrin collagen receptors. Mol Cell 1997;1:25–34. Vogel W, Gish GD, Alves F, Pawson T. The discoidin domain receptor tyrosine kinases are activated by collagen. Mol Cell 1997;1:13–23. Alves F, Vogel W, Mossie K, Millauer B, Höfler H, Ullrich A. Distinct structural characteristics of discoidin I subfamily receptor tyrosine kinases and complementary expression in human cancer. Oncogene 1995;10:609–18. Royce PM, Steinmann BU. Connective tissue and its heritable disorders: molecular, genetic, and medical aspects. New York: Wiley-Liss, 1993. Linn SC, West RB, Zhu S, Nielsen T, Goldblum JR, Patel R, Rubin BP, Alter O, Brown PO, Botstein D, van de Rijn M. Genome-wide mRNA expression profiling of dermatofibrosarcoma protuberans using cDNA microarrays. Mod Pathol 2002;15:18A.

[86] Bowne WB, Antonescu CR, Leung DHY, Katz SC, Hawkins WG, Woodruff JM, Brennan MF, Lewis JJ. Dermatofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution. Cancer 2000;88:2711–20. [87] Zelger B, Zelger B. Sarcomas arising in dermatofibrosarcoma protuberans: collision or illusion? [letter]. Am J Surg Pathol 2001;25:1106–7. [88] Goldblum JR, Reith JD, Weiss SW. Letter-to-the-editor. Am J Surg Pathol 2001;25:1107–8. [89] Cribier B, Noacco G, Peltre B, Grosshans E. Stromelysin 3 expression: a useful marker for the differential diagnosis dermatofibroma versus dermatofibrosarcoma protuberans. J Am Acad Dermatol 2002;46:408–13. [90] Sozzi G, Minoletti F, Miozzo M, Sard L, Musso K, Azzarelli A, Pierotti MA, Pilotti S. Relevance of cytogenetic and fluorescent in situ hybridization analyses in the clinical assessment of soft tissue sarcoma. Hum Pathol 1997;28:134–42. [91] Sandberg AA, Anderson WD, Fredenberg C, Hashimoto H. Dermatofibrosarcoma protuberans of breast. Cancer Genet Cytogenet 2003, in press. [92] Sjöblom T, Shimizu A, O’Brien KP, Pietras K, Dal Cin P, Buchdunger E, Dumanski JP, Östman A, Heldin C-H. Growth inhibition of dermatofibrosarcoma protuberans tumors by the platelet-derived growth factor receptor antagonist STI571 through induction of apoptosis. Cancer Res 2001;61:5778–83. [93] Rubin BP, Schuetze SM, Eary JF, Norwood TH, Mirza S, Conrad EU, Bruckner JD. Molecular targeting of platelet-derived growth factor B by imatinib mesylate in a patient with metastatic dermatofibrosarcoma protuberans. J Clin Oncol 2002;20:3586–91. [94] Maki RG, Awan RA, Dixon RH, Jhanwar S, Antonescu CR. Differential sensitivity to imatinib of 2 patients with metastatic sarcoma arising from dermatofibrosarcoma protuberans. Int J Cancer 2002;100:623–6. [95] Bridge JA, Neff JR, Sandberg AA. Cytogenetic analysis of dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1990;49:199–202. [96] Mandahl N, Heim S, Willén H, Rydholm A, Mitelman F. Supernumerary ring chromosome as the sole cytogenetic abnormality in a dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1990;49:273–5. [97] Örndal C, Mandahl N, Rydholm A, Willén H, Brosjö O, Heim S. Mitelman F. Supernumerary ring chromosomes in five bone and soft tissue tumors of low or borderline malignancy. Cancer Genet Cytogenet 1992;60:170–5. [98] Stephenson CF, Berger CS, Leong SPL, Davis JR, Sandberg AA. Ring chromosome in a dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1992;58:52–4. [99] Pédeutour F, Coindre J-M, Nicolo G, Bouchot C, Ayraud N, TurcCarel C. Ring chromosomes in dermatofibrosarcoma protuberans contain chromosome 17 sequences: fluorescence in situ hybridization. Cancer Genet Cytogenet 1993;67:149. [100] Craver RD, Correa H, Kao Y, Van Brunt T. Dermatofibrosarcoma protuberans with 46,XY,t(X;7) abnormality in a child. Cancer Genet Cytogenet 1995;80:75–7. [101] Stenman G, Andersson H, Meis-Kindblom JM, Röijer E, Kindblom L-G. FISH analysis of supernumerary ring chromosome in dermatofibrosarcoma protuberans. Int J Oncol 1995;6:81–6. [102] Mandahl N, Limon J, Mertens F, Arheden K, Mitelman F. Ring marker containing 17q and chromosome 22 in a case of dermatofibrosarcoma protuberans. Cancer Genet Cytogenet 1996;89:88–91. [103] Iwasaki H, Ohjimi Y, Ishiguro M, Isayama T, Fujita C, Kaneko Y, Kikuchi M, Shinohara N. Supernumerary ring chromosomes and nuclear blebs in some low-grade malignant soft tissue tumours: atypical lipomatous tumours and dermatofibrosarcoma protuberans. Virchows Arch 1998;432:521–8. [104] Dobin SM, Diaz JA, Silva MT, Truss LM, Donner LR. Translocation (17;22)(q22;q13) in a case of subcutaneous dermatofibrosarcoma protuberans in an adult. Cancer Genet Cytogenet 1999;109:86–7. [105] Nishio J, Iwasaki H, Ohjimi Y, Ishiguro M, Isayama T, Naito M, Kaneko Y, Kikuchi M. Supernumerary ring chromosomes in dermatofibrosarcoma protuberans may contain sequences from 8q11.2ter and 17q21qter: a combined cytogenetic and comparative genomic hybridization study. Cancer Genet Cytogenet 2001;129:102–6.