Human Pathology (2014) 45, 137–142
www.elsevier.com/locate/humpath
Original contribution
Histological analysis suggests an invasion-independent metastatic mechanism in alveolar soft part sarcoma☆ Nokitaka Setsu MD, PhD a,b , Akihiko Yoshida MD, PhD a,⁎, Fumiaki Takahashi ScD c , Hirokazu Chuman MD b , Ryoji Kushima MD, PhD a a
Department of Pathology and Clinical Laboratories, National Cancer Center Hospital, Tokyo, Japan Division of Musculoskeletal Oncology, National Cancer Center Hospital, Tokyo, Japan c Department of Clinical Medicine (Biostatistics), School of Pharmacy, Kitasato University, Tokyo, Japan b
Received 27 May 2013; revised 28 June 2013; accepted 3 July 2013
Keywords: Alveolar soft part sarcoma; Invasion; Metastasis; Angiogenesis
Summary Alveolar soft part sarcoma (ASPS) is a rare soft tissue tumor characterized by pseudoalveolar growths associated with abundant sinusoidal vessels. It has a high proclivity to blood-borne metastasis, but the exact mechanism of spread has not been widely discussed, and detailed histological analysis of vascular involvement is still lacking. In this study, we histologically analyzed 32 surgically resected ASPSs, with particular attention to the mode of vascular involvement. Among 188 instances of unequivocal vascular involvement, 184 (98%) were in the form of variously sized cohesive clusters that were completely enveloped by endothelial cells, confirmed by CD31 immunostaining. Discohesive intravascular tumor cells without endothelial wrapping were rare (2%). The clinical relevance of vascular involvement was supported by survival analysis where the average number of vascular involvements per slide was an independent risk factor for shorter progression-free survival. Our findings suggest that ASPSs do not actively break through the vascular walls to initiate the metastatic process. They instead suggest that ASPSs almost exclusively follow the recently postulated “invasionindependent mechanism” for entry into circulation, in which cancer cells are shed into vessels, while being completely enveloped by endothelial cells, and are subsequently entrapped at recipient organs. Because the latter mechanism is reportedly dependent on tumor angiogenesis and vascular remodeling, our data provide a morphological rationale for the use of anti-angiogenic therapy to treat ASPSs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Initially described in 1952 [1], alveolar soft part sarcoma (ASPS) is a rare neoplasm that constitutes b1% of all soft ☆
Conflicts of interest and source of funding: The authors declare no conflict of interest. This work was supported in part by the National Cancer Center Research and Development Fund (23-A-10, 23-B-12). ⁎ Corresponding author. Department of Pathology and Clinical Laboratories National Cancer Center Hospital 5-1-1 Tsukiji, Chuo-ku, Tokyo 104–0045, Japan. E-mail address: [email protected] (A. Yoshida). 0046-8177/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humpath.2013.07.045
tissue sarcomas. ASPSs mainly affect adolescents and young adults and histologically show characteristic pseudoalveolar architecture created by polygonal cells and abundant sinusoidal vessels. A specific chromosomal rearrangement, der(17)t(X;17)(p11;q25), has been reported to result in the ASPSCR1-TFE3 fusion gene in virtually all cases of ASPS [2]. Although ASPSs tend to show indolent local growth, they are notorious for frequent hematogenous metastases, especially to the lung, bone, and brain [3–7]. The metastasis may already be present at the time of initial clinical presentation. When the tumors disseminate beyond surgical control, it is difficult to eradicate the disease because ASPSs
138 are inherently resistant to conventional chemotherapy [4,5,8]. However, the patient may paradoxically survive long even after metastasis, and the exact prognosis is somewhat unpredictable for each patient. Although the high metastatic potential of ASPSs is often attributed to rich intratumoral vascularity, the exact mechanism underlying spread has not yet been described, and detailed histological analysis of vascular involvement is still lacking. In this study, we examined a series of ASPS cases with particular attention to vascular involvement with the goal of clarifying its detailed morphologic pattern. Further, we wished to determine the oncologic implication and the clinical relevance, if any, of vascular involvement in ASPSs.
N. Setsu et al. sion-free survival were defined as the time from initial surgery to death and progression, respectively. Progression was assessed according to the Response Evaluation Criteria in Solid Tumors Group (version 1.1) [9]. The survival was calculated by the Kaplan-Meier method, and the Cox regression model was applied to estimate the hazard ratio and as a multivariate model of risk factors with a backward selection approach. All P values were 2 tailed, and P b .05 was considered significant.
3. Results 3.1. Clinical findings
2. Materials and methods 2.1. Case selection and histological examination After the approval from the institutional review board, the pathology archive of the National Cancer Center Hospital, Tokyo, Japan, was searched for ASPSs. The search identified 32 ASPSs from 27 patients treated at our institution between 1986 and 2012. All the hematoxylin and eosin–stained slides available were reviewed, and in-depth analysis of morphological features related to vascular involvement was performed. In this study, vascular involvement was defined as tumor cells lying completely within vascular spaces, irrespective of the presence of attachment to the vascular wall. We limited the analysis to vascular involvement located outside the contour of the tumor mass; intratumoral events were excluded from the analysis because a high level of intratumoral vascularity hampered objective morphologic definition of vascular involvement within the tumor mass. The average number of vascular involvements per slide in which the tumor periphery was evaluable was referred to as the intravasation index.
2.2. Immunohistochemistry All the vascular involvements were immunostained with CD31 antibody (clone JC70A, 1:40; Dako, Glostrup, Denmark). D2-40 antibody (clone D2-40, 1:100, Signet Laboratories, Dedham, MA) was also used in select cases where lymphatic involvement was morphologically suspected.
2.3. Survival analysis All data analyses were performed within the R environment (version 2.14.0, The R Foundation for Statistical Computing, Vienna, Austria). Survival data were analyzed in 21 patients with primary tumors. Survival was correlated with age, sex, tumor size, existence of metastasis at initial presentation, and intravasation index. Overall and progres-
The study population comprised 10 men and 17 women, with an age range of 2 to 43 years (mean age, 27 years). Twenty-one tumors were primary tumors and were located in the thigh (n = 8), buttock (n = 3), periscapular region (n = 2), lower leg (n = 2), retroperitoneum (n = 2), inguinal region (n = 1), upper arm (n = 1), forearm (n = 1), and orbit (n = 1). The remaining tumors were either recurrent (n = 1) or had metastasized to the lung (n = 5), brain (n = 2), leg (n = 1), spinal canal (n = 1), or spleen (n = 1). Seven tumors were resected after embolization, and in 1 case, metastatic tissue was obtained after chemotherapy.
3.2. Histological analysis Sinusoidal vessels and a pseudoalveolar pattern were seen in all the cases, except for 3 tumors in pediatric patients (ages 2, 4, and 15 years) that predominantly exhibited solid and diffuse growth without conspicuous pseudoalveolar structures. The intratumoral sinusoidal vessels showed dense proliferation, fusion, and sometimes dilatation. The tumor periphery was evaluable in 224 slides from 32 tumors (1-15 slides per tumor). A total of 188 vascular involvements were seen in 82 slides from 19 tumors. The intravasation index was 0 to 5.7 (median, 0.3). The involved vessels were with (n = 40) or without (n = 148) an evident smooth muscle coat. In almost all (184 of 188, 98%) the vascular involvements, the intravascular tumor cells formed variously sized tight clusters (Fig. 1A-D). These clusters were completely surrounded by endothelial cells, and this envelopment was confirmed by CD31 immunostaining in all these 184 vascular involvements (Fig. 1B and 1D). Relatively large clusters often contained sinusoidal vessels (Fig. 1C and 1D). Three lymphatic involvements were identified in 1 metastatic lesion to the lung, and they were confirmed by D2-40 immunostaining. Discohesive naked intravascular tumor cells without endothelial wrapping were extremely rare (4/188, 2%). In rare sections, a small part of the peripheral tumor tissue was observed as emanating from a parent mass into an adjacent vessel (Fig. 1E). The surrounding veins were often tortuous and dilated, and
Vascular involvement of alveolar soft part sarcoma
139
Fig. 1 A and C, Intravascular ASPS cells almost always formed tight clusters that were enveloped by endothelial cells (H&E; original magnification: A, ×400; C, ×200). B and D, CD31 immunostaining highlighted the complete endothelial envelopment around the intravascular tumor clusters (CD31 immunohistochemistry; original magnification: B, ×400; D, ×200). E, In rare sections, a small part of peripheral tumor tissue was observed to emanate from a parent mass into an adjacent vessel (arrows, tumor tissue emanating from the parent mass; closed arrowheads, tumor contour with a fibrous capsule; open arrowheads, efferent vein; H&E, original magnification ×40). F, The veins adjacent to the ASPS were often tortuous and dilated, and occasionally showed mural attenuation disproportionate to the caliber of the lumen (Elastica van Gieson; original magnification ×20). H&E, hematoxylin and eosin.
occasionally showed mural attenuation somewhat disproportionate to the caliber of the lumen (Fig. 1F).
3.3. Survival analysis Survival analysis was performed for 21 patients with primary tumors. The follow-up ranged from 1 to 312 months (median, 20 months). The overall survival (OS) rates were 49% at 5 years and 27% at 10 years. The progression-free survival (PFS) rates were 23% at 5 years and 17% at 10 years. Progression occurred in 14 cases; in all cases, progression was seen in the form of the emergence of new
metastases or enlargement of existing metastatic lesions. Univariate analysis showed that a higher intravasation index was the only significant risk factor for shorter PFS (P = .00011; Table); however, none of the parameters was significant in predicting OS. Multivariate analysis showed that a higher intravasation index (P = .0034) and tumor size (P = .025) were independent risk factors for shorter PFS; nevertheless, none of the parameters remained significant for predicting OS, except for tumor size (P = .09), which may represent a potential risk factor for shorter OS. The parameter “existence of metastasis at initial presentation” was excluded from multivariate analysis of OS because no event was
140
N. Setsu et al.
Table Univariate survival analysis of 21 patients with primary alveolar soft part sarcomas Parameters
Median Progression- Overall (range) or free survival survival n (%) (P) (P)
Age (y) 29 (2-43) Sex Male 8 (38%) Female 13 (62%) Size (cm) 5 (2-12) Metastasis at presentation Yes 16 (76%) No 5 (24%) Intravasation index a 0.8 (0-3)
.28 .94
.14 .84
.083 .46
.067 .070
.00011
.37
a
Defined as the number of vascular involvements per slide that showed an evaluable tumor periphery.
observed in patients without metastasis at presentation and, as a result, the Cox regression model did not converge in the univariate analysis.
4. Discussion ASPSs are characterized by their prominent metastatic potential, exemplified by most (25/27) of our patients having experienced distant metastasis. However, the exact mechanism underlying this high proclivity is currently unclear. In this study, we showed that the intravascular ASPS cells almost exclusively formed tight clusters that were completely surrounded by endothelial cells. The clinical relevance of these intravascular ASPS clusters was supported by the survival analysis, in which the intravasation index was a significant risk factor of shorter PFS in both univariate and multivariate analyses. These data suggest that the functional unit of ASPS metastasis is likely an endothelium-wrapped cluster rather than an isolated naked cell. The foregoing observation is inconsistent with the conventional view of tumor metastasis [10–12]. According to the conventional model (Fig. 2A), (1) tumor cells actively degrade the extracellular matrix/vascular walls and migrate into the vessels; (2) the transported cells adhere to the endothelium where they exit the blood flow by penetrating the vascular wall; and (3) when a distant site constitutes an appropriate niche, the tumor cells survive and proliferate to form masses. The envelopment of the endothelium is not expected in this model where destructive penetration through vascular walls plays a critical role. Our histological observation instead fits well with the recently postulated alternative “invasion-independent pathway.” An invasion-independent mechanism of metastasis was first described by Sugino et al [13] in murine breast cancer cell lines, where cancer cells are shed into vessels as tight clusters while being completely enveloped by endothelial cells, and are subsequently entrapped at the level of
Fig. 2 Schematic diagram of tumor-cell entry into circulation according to 2 different mechanisms of metastasis. A, In the canonical invasive pathway of metastasis, tumor cells destroy the extracellular matrix and invade through vascular walls to reach the blood flow without endothelial envelopment. B, In the alternative invasionindependent pathway of metastasis, profuse sinusoidal angiogenesis transforms tumor tissue into endothelium-enveloped small clusters (“micronodules”) that shed into the blood stream.
the recipient organs. Subsequent studies using clinical samples showed that this mechanism is indeed at play in vivo by demonstrating endothelium-wrapped intravascular cancer cells in a subset of human carcinomas [14]. Other studies further demonstrated that the invasion-independent metastatic pathway primarily depends on the increase in the number of blood vessels and sinusoidal remodeling of vasculature that transforms the tumor tissue into cell clusters within the blood stream (“micronodular transformation”) (Fig. 2B), rather than cellular invasiveness associated with proteolysis and cell motility [15,16]. Notably, renal cell carcinomas, hepatocellular carcinomas, and follicular thyroid carcinomas, the 3 most common cancer types that use the invasion-independent mechanism [14], have been well known for their hypervascularity and sinusoidal vasculatures [17–19]. Therefore, given the rich vascularity and conspicuous intratumoral sinusoidal element of ASPSs, it is eminently reasonable that these sarcomas would follow this alternative pathway. The invasion-independent mechanism of metastasis may explain some of the unique clinical characteristics of ASPSs. For example, the unusually high metastatic potential may be ascribed to the maintenance of tissue organization in the
Vascular involvement of alveolar soft part sarcoma intravascular tumor cell clusters. A previous study has shown that tumor cells injected in clusters developed metastasis more effectively than singly separated cells did [20]. Similarly, the peculiar discordance between high metastatic rate and relatively indolent local growth with protracted clinical course may reflect that the tumor behavior in ASPSs, unlike many other malignancies, is determined by angiogenesis and vascular remodeling rather than by the invasive property of tumor cells. Because the invasion-independent pathway primarily depends on hypervascularity and sinusoidal remodeling [15,16], we believe that our data provide a morphologic rationale for applying anti-angiogenic therapy to ASPSs. Indeed, ASPSs have been reported to show gene upregulation and protein overexpression of angiogenic molecules such as vascular endothelial growth factor (VEGF) [21,22]. VEGF has been previously shown to induce the formation of sinusoidal structure and endothelium-enveloped intravascular tumor clusters [15]. Several studies have documented the efficacy of VEGF-targeted therapy in ASPSs [22–24] as well as in cancers with sinusoidal structure [25–28]. Future research may identify additional targetable molecules that control angiogenesis and sinusoidal remodeling in ASPSs. While the invasion-independent pathway of metastasis was originally proposed for a subset of carcinomas [14], to our knowledge, the present study is the first to document its role in sarcoma. The only other mesenchymal tumor thus far suggested to follow this mechanism is lymphangioleiomyomatosis (LAM). Kumasaka et al [29] showed that LAM cell clusters are invariably enveloped by lymphatic endothelial cells within the lymphatic spaces. The authors proposed these clusters as the basis of dissemination and disease progression of LAM. LAM belongs to the perivascular epithelioid cell tumor family, a subset of which morphologically resembles ASPSs and/or shares the TFE3 gene rearrangement with ASPSs [30]. Tortuous dilatation and irregular mural attenuation of the veins were occasionally observed adjacent to the ASPSs and might reflect altered local hemodynamics. Unlike normal capillaries, intratumoral sinusoids may permit arterial blood pressure to directly transmit to the venous system. Supporting this view, ASPSs are clinically known for their pulsation and audible bruit [31], and arteriovenous shunting can be demonstrated by angiography [32]. Alternatively, such vascular changes may represent incipient sinusoidal vascular remodeling that may be later incorporated into the tumor mass. In conclusion, this detailed histological analysis of vascular involvements in ASPSs showed virtually constant endothelial envelopment around the intravascular tumor cell clusters. These findings suggest that ASPSs almost exclusively follow the invasion-independent pathway for entry into circulation. In addition to providing insights into the metastatic mechanism, the present study also provides a morphological rationale for the use of anti-angiogenic therapy to treat ASPSs.
141
Acknowledgments The authors thank Ms. Sachiko Miura and Ms. Chizu Kina for superb technical assistance.
References [1] Christopherson WM, Foote Jr FW, Stewart FW. Alveolar soft-part sarcomas; structurally characteristic tumors of uncertain histogenesis. Cancer 1952;5:100-11. [2] Ladanyi M, Lui MY, Antonescu CR, et al. The der(17)t(X;17)(p11; q25) of human alveolar soft part sarcoma fuses the TFE3 transcription factor gene to ASPL, a novel gene at 17q25. Oncogene 2001;20:48-57. [3] Lieberman PH, Brennan MF, Kimmel M, Erlandson RA, Garin-Chesa P, Flehinger BY. Alveolar soft-part sarcoma. A clinico-pathologic study of half a century. Cancer 1989;63:1-13. [4] Portera Jr CA, Ho V, Patel SR, et al. Alveolar soft part sarcoma: clinical course and patterns of metastasis in 70 patients treated at a single institution. Cancer 2001;91:585-91. [5] Ogose A, Yazawa Y, Ueda T, et al, Japanese Musculoskeletal Oncology Group. Alveolar soft part sarcoma in Japan: multiinstitutional study of 57 patients from the Japanese Musculoskeletal Oncology Group. Oncology 2003;65:7-13. [6] Pennacchioli E, Fiore M, Collini P, et al. Alveolar soft part sarcoma: clinical presentation, treatment, and outcome in a series of 33 patients at a single institution. Ann Surg Oncol 2010;17:3229-33. [7] Ogura K, Beppu Y, Chuman H, et al. Alveolar soft part sarcoma: a single-center 26-patient case series and review of the literature. Sarcoma 2012;2012:907179. [8] Reichardt P, Lindner T, Pink D, Thuss-Patience PC, Kretzschmar A, Dorken B. Chemotherapy in alveolar soft part sarcomas. What do we know? Eur J Cancer 2003;39:1511-6. [9] Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47. [10] Wyckoff JB, Jones JG, Condeelis JS, Segall JE. A critical step in metastasis: in vivo analysis of intravasation at the primary tumor. Cancer Res 2000;60:2504-11. [11] Luzzi KJ, MacDonald IC, Schmidt EE, et al. Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 1998;153:865-73. [12] Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2002;2: 563-72. [13] Sugino T, Kusakabe T, Hoshi N, et al. An invasion-independent pathway of blood-borne metastasis: a new murine mammary tumor model. Am J Pathol 2002;160:1973-80. [14] Sugino T, Yamaguchi T, Ogura G, et al. Morphological evidence for an invasion-independent metastasis pathway exists in multiple human cancers. BMC Med 2004;2:9. [15] Kusters B, Kats G, Roodink I, et al. Micronodular transformation as a novel mechanism of VEGF-A-induced metastasis. Oncogene 2007;26: 5808-15. [16] Sugino T, Yamaguchi T, Hoshi N, et al. Sinusoidal tumor angiogenesis is a key component in hepatocellular carcinoma metastasis. Clin Exp Metastasis 2008;25:835-41. [17] Rosai J. Rosai and Ackerman’s Surgical Pathology. 9th ed. Maryland Heights, MO: Mosby; 2004. [18] Tanigawa N, Lu C, Mitsui T, Miura S. Quantitation of sinusoid-like vessels in hepatocellular carcinoma: its clinical and prognostic significance. Hepatology 1997;26:1216-23.
142 [19] Kendall CH, Sanderson PR, Cope J, Talbot IC. Follicular thyroid tumours: a study of laminin and type IV collagen in basement membrane and endothelium. J Clin Pathol 1985;38:1100-5. [20] Liotta LA, Saidel MG, Kleinerman J. The significance of hematogenous tumor cell clumps in the metastatic process. Cancer Res 1976;36:889-94. [21] Lazar AJ, Lahat G, Myers SE, et al. Validation of potential therapeutic targets in alveolar soft part sarcoma: an immunohistochemical study utilizing tissue microarray. Histopathology 2009;55:750-5. [22] Vistica DT, Hollingshead M, Borgel SD, et al. Therapeutic vulnerability of an in vivo model of alveolar soft part sarcoma (ASPS) to antiangiogenic therapy. J Pediatr Hematol Oncol 2009;31: 561-70. [23] Azizi AA, Haberler C, Czech T, et al. Vascular-endothelial-growthfactor (VEGF) expression and possible response to angiogenesis inhibitor bevacizumab in metastatic alveolar soft part sarcoma. Lancet Oncol 2006;7:521-3. [24] Stacchiotti S, Negri T, Zaffaroni N, et al. Sunitinib in advanced alveolar soft part sarcoma: evidence of a direct antitumor effect. Ann Oncol 2011;22:1682-90. [25] Llovet JM, Ricci S, Mazzaferro V, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008;359:378-90.
N. Setsu et al. [26] Escudier B, Eisen T, Stadler WM, et al. Sorafenib for treatment of renal cell carcinoma: final efficacy and safety results of the phase III treatment approaches in renal cancer global evaluation trial. J Clin Oncol 2009;27:3312-8. [27] Gore ME, Szczylik C, Porta C, et al. Safety and efficacy of sunitinib for metastatic renal-cell carcinoma: an expanded-access trial. Lancet Oncol 2009;10:757-63. [28] Bible KC, Suman VJ, Molina JR, et al. Efficacy of pazopanib in progressive, radioiodine-refractory, metastatic differentiated thyroid cancers: results of a phase 2 consortium study. Lancet Oncol 2010;11: 962-72. [29] Kumasaka T, Seyama K, Mitani K, et al. Lymphangiogenesismediated shedding of LAM cell clusters as a mechanism for dissemination in lymphangioleiomyomatosis. Am J Surg Pathol 2005;29:1356-66. [30] Argani P, Aulmann S, Illei PB, et al. A distinctive subset of PEComas harbors TFE3 gene fusions. Am J Surg Pathol 2010;34:1395-406. [31] Iwamoto Y, Morimoto N, Chuman H, Shinohara N, Sugioka Y. The role of MR imaging in the diagnosis of alveolar soft part sarcoma: a report of 10 cases. Skeletal Radiol 1995;24:267-70. [32] Suh JS, Cho J, Lee SH, et al. Alveolar soft part sarcoma: MR and angiographic findings. Skeletal Radiol 2000;29:680-9.
www.elsevier.com/locate/humpath
Original contribution
Histological analysis suggests an invasion-independent metastatic mechanism in alveolar soft part sarcoma☆ Nokitaka Setsu MD, PhD a,b , Akihiko Yoshida MD, PhD a,⁎, Fumiaki Takahashi ScD c , Hirokazu Chuman MD b , Ryoji Kushima MD, PhD a a
Department of Pathology and Clinical Laboratories, National Cancer Center Hospital, Tokyo, Japan Division of Musculoskeletal Oncology, National Cancer Center Hospital, Tokyo, Japan c Department of Clinical Medicine (Biostatistics), School of Pharmacy, Kitasato University, Tokyo, Japan b
Received 27 May 2013; revised 28 June 2013; accepted 3 July 2013
Keywords: Alveolar soft part sarcoma; Invasion; Metastasis; Angiogenesis
Summary Alveolar soft part sarcoma (ASPS) is a rare soft tissue tumor characterized by pseudoalveolar growths associated with abundant sinusoidal vessels. It has a high proclivity to blood-borne metastasis, but the exact mechanism of spread has not been widely discussed, and detailed histological analysis of vascular involvement is still lacking. In this study, we histologically analyzed 32 surgically resected ASPSs, with particular attention to the mode of vascular involvement. Among 188 instances of unequivocal vascular involvement, 184 (98%) were in the form of variously sized cohesive clusters that were completely enveloped by endothelial cells, confirmed by CD31 immunostaining. Discohesive intravascular tumor cells without endothelial wrapping were rare (2%). The clinical relevance of vascular involvement was supported by survival analysis where the average number of vascular involvements per slide was an independent risk factor for shorter progression-free survival. Our findings suggest that ASPSs do not actively break through the vascular walls to initiate the metastatic process. They instead suggest that ASPSs almost exclusively follow the recently postulated “invasionindependent mechanism” for entry into circulation, in which cancer cells are shed into vessels, while being completely enveloped by endothelial cells, and are subsequently entrapped at recipient organs. Because the latter mechanism is reportedly dependent on tumor angiogenesis and vascular remodeling, our data provide a morphological rationale for the use of anti-angiogenic therapy to treat ASPSs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Initially described in 1952 [1], alveolar soft part sarcoma (ASPS) is a rare neoplasm that constitutes b1% of all soft ☆
Conflicts of interest and source of funding: The authors declare no conflict of interest. This work was supported in part by the National Cancer Center Research and Development Fund (23-A-10, 23-B-12). ⁎ Corresponding author. Department of Pathology and Clinical Laboratories National Cancer Center Hospital 5-1-1 Tsukiji, Chuo-ku, Tokyo 104–0045, Japan. E-mail address: [email protected] (A. Yoshida). 0046-8177/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humpath.2013.07.045
tissue sarcomas. ASPSs mainly affect adolescents and young adults and histologically show characteristic pseudoalveolar architecture created by polygonal cells and abundant sinusoidal vessels. A specific chromosomal rearrangement, der(17)t(X;17)(p11;q25), has been reported to result in the ASPSCR1-TFE3 fusion gene in virtually all cases of ASPS [2]. Although ASPSs tend to show indolent local growth, they are notorious for frequent hematogenous metastases, especially to the lung, bone, and brain [3–7]. The metastasis may already be present at the time of initial clinical presentation. When the tumors disseminate beyond surgical control, it is difficult to eradicate the disease because ASPSs
138 are inherently resistant to conventional chemotherapy [4,5,8]. However, the patient may paradoxically survive long even after metastasis, and the exact prognosis is somewhat unpredictable for each patient. Although the high metastatic potential of ASPSs is often attributed to rich intratumoral vascularity, the exact mechanism underlying spread has not yet been described, and detailed histological analysis of vascular involvement is still lacking. In this study, we examined a series of ASPS cases with particular attention to vascular involvement with the goal of clarifying its detailed morphologic pattern. Further, we wished to determine the oncologic implication and the clinical relevance, if any, of vascular involvement in ASPSs.
N. Setsu et al. sion-free survival were defined as the time from initial surgery to death and progression, respectively. Progression was assessed according to the Response Evaluation Criteria in Solid Tumors Group (version 1.1) [9]. The survival was calculated by the Kaplan-Meier method, and the Cox regression model was applied to estimate the hazard ratio and as a multivariate model of risk factors with a backward selection approach. All P values were 2 tailed, and P b .05 was considered significant.
3. Results 3.1. Clinical findings
2. Materials and methods 2.1. Case selection and histological examination After the approval from the institutional review board, the pathology archive of the National Cancer Center Hospital, Tokyo, Japan, was searched for ASPSs. The search identified 32 ASPSs from 27 patients treated at our institution between 1986 and 2012. All the hematoxylin and eosin–stained slides available were reviewed, and in-depth analysis of morphological features related to vascular involvement was performed. In this study, vascular involvement was defined as tumor cells lying completely within vascular spaces, irrespective of the presence of attachment to the vascular wall. We limited the analysis to vascular involvement located outside the contour of the tumor mass; intratumoral events were excluded from the analysis because a high level of intratumoral vascularity hampered objective morphologic definition of vascular involvement within the tumor mass. The average number of vascular involvements per slide in which the tumor periphery was evaluable was referred to as the intravasation index.
2.2. Immunohistochemistry All the vascular involvements were immunostained with CD31 antibody (clone JC70A, 1:40; Dako, Glostrup, Denmark). D2-40 antibody (clone D2-40, 1:100, Signet Laboratories, Dedham, MA) was also used in select cases where lymphatic involvement was morphologically suspected.
2.3. Survival analysis All data analyses were performed within the R environment (version 2.14.0, The R Foundation for Statistical Computing, Vienna, Austria). Survival data were analyzed in 21 patients with primary tumors. Survival was correlated with age, sex, tumor size, existence of metastasis at initial presentation, and intravasation index. Overall and progres-
The study population comprised 10 men and 17 women, with an age range of 2 to 43 years (mean age, 27 years). Twenty-one tumors were primary tumors and were located in the thigh (n = 8), buttock (n = 3), periscapular region (n = 2), lower leg (n = 2), retroperitoneum (n = 2), inguinal region (n = 1), upper arm (n = 1), forearm (n = 1), and orbit (n = 1). The remaining tumors were either recurrent (n = 1) or had metastasized to the lung (n = 5), brain (n = 2), leg (n = 1), spinal canal (n = 1), or spleen (n = 1). Seven tumors were resected after embolization, and in 1 case, metastatic tissue was obtained after chemotherapy.
3.2. Histological analysis Sinusoidal vessels and a pseudoalveolar pattern were seen in all the cases, except for 3 tumors in pediatric patients (ages 2, 4, and 15 years) that predominantly exhibited solid and diffuse growth without conspicuous pseudoalveolar structures. The intratumoral sinusoidal vessels showed dense proliferation, fusion, and sometimes dilatation. The tumor periphery was evaluable in 224 slides from 32 tumors (1-15 slides per tumor). A total of 188 vascular involvements were seen in 82 slides from 19 tumors. The intravasation index was 0 to 5.7 (median, 0.3). The involved vessels were with (n = 40) or without (n = 148) an evident smooth muscle coat. In almost all (184 of 188, 98%) the vascular involvements, the intravascular tumor cells formed variously sized tight clusters (Fig. 1A-D). These clusters were completely surrounded by endothelial cells, and this envelopment was confirmed by CD31 immunostaining in all these 184 vascular involvements (Fig. 1B and 1D). Relatively large clusters often contained sinusoidal vessels (Fig. 1C and 1D). Three lymphatic involvements were identified in 1 metastatic lesion to the lung, and they were confirmed by D2-40 immunostaining. Discohesive naked intravascular tumor cells without endothelial wrapping were extremely rare (4/188, 2%). In rare sections, a small part of the peripheral tumor tissue was observed as emanating from a parent mass into an adjacent vessel (Fig. 1E). The surrounding veins were often tortuous and dilated, and
Vascular involvement of alveolar soft part sarcoma
139
Fig. 1 A and C, Intravascular ASPS cells almost always formed tight clusters that were enveloped by endothelial cells (H&E; original magnification: A, ×400; C, ×200). B and D, CD31 immunostaining highlighted the complete endothelial envelopment around the intravascular tumor clusters (CD31 immunohistochemistry; original magnification: B, ×400; D, ×200). E, In rare sections, a small part of peripheral tumor tissue was observed to emanate from a parent mass into an adjacent vessel (arrows, tumor tissue emanating from the parent mass; closed arrowheads, tumor contour with a fibrous capsule; open arrowheads, efferent vein; H&E, original magnification ×40). F, The veins adjacent to the ASPS were often tortuous and dilated, and occasionally showed mural attenuation disproportionate to the caliber of the lumen (Elastica van Gieson; original magnification ×20). H&E, hematoxylin and eosin.
occasionally showed mural attenuation somewhat disproportionate to the caliber of the lumen (Fig. 1F).
3.3. Survival analysis Survival analysis was performed for 21 patients with primary tumors. The follow-up ranged from 1 to 312 months (median, 20 months). The overall survival (OS) rates were 49% at 5 years and 27% at 10 years. The progression-free survival (PFS) rates were 23% at 5 years and 17% at 10 years. Progression occurred in 14 cases; in all cases, progression was seen in the form of the emergence of new
metastases or enlargement of existing metastatic lesions. Univariate analysis showed that a higher intravasation index was the only significant risk factor for shorter PFS (P = .00011; Table); however, none of the parameters was significant in predicting OS. Multivariate analysis showed that a higher intravasation index (P = .0034) and tumor size (P = .025) were independent risk factors for shorter PFS; nevertheless, none of the parameters remained significant for predicting OS, except for tumor size (P = .09), which may represent a potential risk factor for shorter OS. The parameter “existence of metastasis at initial presentation” was excluded from multivariate analysis of OS because no event was
140
N. Setsu et al.
Table Univariate survival analysis of 21 patients with primary alveolar soft part sarcomas Parameters
Median Progression- Overall (range) or free survival survival n (%) (P) (P)
Age (y) 29 (2-43) Sex Male 8 (38%) Female 13 (62%) Size (cm) 5 (2-12) Metastasis at presentation Yes 16 (76%) No 5 (24%) Intravasation index a 0.8 (0-3)
.28 .94
.14 .84
.083 .46
.067 .070
.00011
.37
a
Defined as the number of vascular involvements per slide that showed an evaluable tumor periphery.
observed in patients without metastasis at presentation and, as a result, the Cox regression model did not converge in the univariate analysis.
4. Discussion ASPSs are characterized by their prominent metastatic potential, exemplified by most (25/27) of our patients having experienced distant metastasis. However, the exact mechanism underlying this high proclivity is currently unclear. In this study, we showed that the intravascular ASPS cells almost exclusively formed tight clusters that were completely surrounded by endothelial cells. The clinical relevance of these intravascular ASPS clusters was supported by the survival analysis, in which the intravasation index was a significant risk factor of shorter PFS in both univariate and multivariate analyses. These data suggest that the functional unit of ASPS metastasis is likely an endothelium-wrapped cluster rather than an isolated naked cell. The foregoing observation is inconsistent with the conventional view of tumor metastasis [10–12]. According to the conventional model (Fig. 2A), (1) tumor cells actively degrade the extracellular matrix/vascular walls and migrate into the vessels; (2) the transported cells adhere to the endothelium where they exit the blood flow by penetrating the vascular wall; and (3) when a distant site constitutes an appropriate niche, the tumor cells survive and proliferate to form masses. The envelopment of the endothelium is not expected in this model where destructive penetration through vascular walls plays a critical role. Our histological observation instead fits well with the recently postulated alternative “invasion-independent pathway.” An invasion-independent mechanism of metastasis was first described by Sugino et al [13] in murine breast cancer cell lines, where cancer cells are shed into vessels as tight clusters while being completely enveloped by endothelial cells, and are subsequently entrapped at the level of
Fig. 2 Schematic diagram of tumor-cell entry into circulation according to 2 different mechanisms of metastasis. A, In the canonical invasive pathway of metastasis, tumor cells destroy the extracellular matrix and invade through vascular walls to reach the blood flow without endothelial envelopment. B, In the alternative invasionindependent pathway of metastasis, profuse sinusoidal angiogenesis transforms tumor tissue into endothelium-enveloped small clusters (“micronodules”) that shed into the blood stream.
the recipient organs. Subsequent studies using clinical samples showed that this mechanism is indeed at play in vivo by demonstrating endothelium-wrapped intravascular cancer cells in a subset of human carcinomas [14]. Other studies further demonstrated that the invasion-independent metastatic pathway primarily depends on the increase in the number of blood vessels and sinusoidal remodeling of vasculature that transforms the tumor tissue into cell clusters within the blood stream (“micronodular transformation”) (Fig. 2B), rather than cellular invasiveness associated with proteolysis and cell motility [15,16]. Notably, renal cell carcinomas, hepatocellular carcinomas, and follicular thyroid carcinomas, the 3 most common cancer types that use the invasion-independent mechanism [14], have been well known for their hypervascularity and sinusoidal vasculatures [17–19]. Therefore, given the rich vascularity and conspicuous intratumoral sinusoidal element of ASPSs, it is eminently reasonable that these sarcomas would follow this alternative pathway. The invasion-independent mechanism of metastasis may explain some of the unique clinical characteristics of ASPSs. For example, the unusually high metastatic potential may be ascribed to the maintenance of tissue organization in the
Vascular involvement of alveolar soft part sarcoma intravascular tumor cell clusters. A previous study has shown that tumor cells injected in clusters developed metastasis more effectively than singly separated cells did [20]. Similarly, the peculiar discordance between high metastatic rate and relatively indolent local growth with protracted clinical course may reflect that the tumor behavior in ASPSs, unlike many other malignancies, is determined by angiogenesis and vascular remodeling rather than by the invasive property of tumor cells. Because the invasion-independent pathway primarily depends on hypervascularity and sinusoidal remodeling [15,16], we believe that our data provide a morphologic rationale for applying anti-angiogenic therapy to ASPSs. Indeed, ASPSs have been reported to show gene upregulation and protein overexpression of angiogenic molecules such as vascular endothelial growth factor (VEGF) [21,22]. VEGF has been previously shown to induce the formation of sinusoidal structure and endothelium-enveloped intravascular tumor clusters [15]. Several studies have documented the efficacy of VEGF-targeted therapy in ASPSs [22–24] as well as in cancers with sinusoidal structure [25–28]. Future research may identify additional targetable molecules that control angiogenesis and sinusoidal remodeling in ASPSs. While the invasion-independent pathway of metastasis was originally proposed for a subset of carcinomas [14], to our knowledge, the present study is the first to document its role in sarcoma. The only other mesenchymal tumor thus far suggested to follow this mechanism is lymphangioleiomyomatosis (LAM). Kumasaka et al [29] showed that LAM cell clusters are invariably enveloped by lymphatic endothelial cells within the lymphatic spaces. The authors proposed these clusters as the basis of dissemination and disease progression of LAM. LAM belongs to the perivascular epithelioid cell tumor family, a subset of which morphologically resembles ASPSs and/or shares the TFE3 gene rearrangement with ASPSs [30]. Tortuous dilatation and irregular mural attenuation of the veins were occasionally observed adjacent to the ASPSs and might reflect altered local hemodynamics. Unlike normal capillaries, intratumoral sinusoids may permit arterial blood pressure to directly transmit to the venous system. Supporting this view, ASPSs are clinically known for their pulsation and audible bruit [31], and arteriovenous shunting can be demonstrated by angiography [32]. Alternatively, such vascular changes may represent incipient sinusoidal vascular remodeling that may be later incorporated into the tumor mass. In conclusion, this detailed histological analysis of vascular involvements in ASPSs showed virtually constant endothelial envelopment around the intravascular tumor cell clusters. These findings suggest that ASPSs almost exclusively follow the invasion-independent pathway for entry into circulation. In addition to providing insights into the metastatic mechanism, the present study also provides a morphological rationale for the use of anti-angiogenic therapy to treat ASPSs.
141
Acknowledgments The authors thank Ms. Sachiko Miura and Ms. Chizu Kina for superb technical assistance.
References [1] Christopherson WM, Foote Jr FW, Stewart FW. Alveolar soft-part sarcomas; structurally characteristic tumors of uncertain histogenesis. Cancer 1952;5:100-11. [2] Ladanyi M, Lui MY, Antonescu CR, et al. The der(17)t(X;17)(p11; q25) of human alveolar soft part sarcoma fuses the TFE3 transcription factor gene to ASPL, a novel gene at 17q25. Oncogene 2001;20:48-57. [3] Lieberman PH, Brennan MF, Kimmel M, Erlandson RA, Garin-Chesa P, Flehinger BY. Alveolar soft-part sarcoma. A clinico-pathologic study of half a century. Cancer 1989;63:1-13. [4] Portera Jr CA, Ho V, Patel SR, et al. Alveolar soft part sarcoma: clinical course and patterns of metastasis in 70 patients treated at a single institution. Cancer 2001;91:585-91. [5] Ogose A, Yazawa Y, Ueda T, et al, Japanese Musculoskeletal Oncology Group. Alveolar soft part sarcoma in Japan: multiinstitutional study of 57 patients from the Japanese Musculoskeletal Oncology Group. Oncology 2003;65:7-13. [6] Pennacchioli E, Fiore M, Collini P, et al. Alveolar soft part sarcoma: clinical presentation, treatment, and outcome in a series of 33 patients at a single institution. Ann Surg Oncol 2010;17:3229-33. [7] Ogura K, Beppu Y, Chuman H, et al. Alveolar soft part sarcoma: a single-center 26-patient case series and review of the literature. Sarcoma 2012;2012:907179. [8] Reichardt P, Lindner T, Pink D, Thuss-Patience PC, Kretzschmar A, Dorken B. Chemotherapy in alveolar soft part sarcomas. What do we know? Eur J Cancer 2003;39:1511-6. [9] Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47. [10] Wyckoff JB, Jones JG, Condeelis JS, Segall JE. A critical step in metastasis: in vivo analysis of intravasation at the primary tumor. Cancer Res 2000;60:2504-11. [11] Luzzi KJ, MacDonald IC, Schmidt EE, et al. Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 1998;153:865-73. [12] Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2002;2: 563-72. [13] Sugino T, Kusakabe T, Hoshi N, et al. An invasion-independent pathway of blood-borne metastasis: a new murine mammary tumor model. Am J Pathol 2002;160:1973-80. [14] Sugino T, Yamaguchi T, Ogura G, et al. Morphological evidence for an invasion-independent metastasis pathway exists in multiple human cancers. BMC Med 2004;2:9. [15] Kusters B, Kats G, Roodink I, et al. Micronodular transformation as a novel mechanism of VEGF-A-induced metastasis. Oncogene 2007;26: 5808-15. [16] Sugino T, Yamaguchi T, Hoshi N, et al. Sinusoidal tumor angiogenesis is a key component in hepatocellular carcinoma metastasis. Clin Exp Metastasis 2008;25:835-41. [17] Rosai J. Rosai and Ackerman’s Surgical Pathology. 9th ed. Maryland Heights, MO: Mosby; 2004. [18] Tanigawa N, Lu C, Mitsui T, Miura S. Quantitation of sinusoid-like vessels in hepatocellular carcinoma: its clinical and prognostic significance. Hepatology 1997;26:1216-23.
142 [19] Kendall CH, Sanderson PR, Cope J, Talbot IC. Follicular thyroid tumours: a study of laminin and type IV collagen in basement membrane and endothelium. J Clin Pathol 1985;38:1100-5. [20] Liotta LA, Saidel MG, Kleinerman J. The significance of hematogenous tumor cell clumps in the metastatic process. Cancer Res 1976;36:889-94. [21] Lazar AJ, Lahat G, Myers SE, et al. Validation of potential therapeutic targets in alveolar soft part sarcoma: an immunohistochemical study utilizing tissue microarray. Histopathology 2009;55:750-5. [22] Vistica DT, Hollingshead M, Borgel SD, et al. Therapeutic vulnerability of an in vivo model of alveolar soft part sarcoma (ASPS) to antiangiogenic therapy. J Pediatr Hematol Oncol 2009;31: 561-70. [23] Azizi AA, Haberler C, Czech T, et al. Vascular-endothelial-growthfactor (VEGF) expression and possible response to angiogenesis inhibitor bevacizumab in metastatic alveolar soft part sarcoma. Lancet Oncol 2006;7:521-3. [24] Stacchiotti S, Negri T, Zaffaroni N, et al. Sunitinib in advanced alveolar soft part sarcoma: evidence of a direct antitumor effect. Ann Oncol 2011;22:1682-90. [25] Llovet JM, Ricci S, Mazzaferro V, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008;359:378-90.
N. Setsu et al. [26] Escudier B, Eisen T, Stadler WM, et al. Sorafenib for treatment of renal cell carcinoma: final efficacy and safety results of the phase III treatment approaches in renal cancer global evaluation trial. J Clin Oncol 2009;27:3312-8. [27] Gore ME, Szczylik C, Porta C, et al. Safety and efficacy of sunitinib for metastatic renal-cell carcinoma: an expanded-access trial. Lancet Oncol 2009;10:757-63. [28] Bible KC, Suman VJ, Molina JR, et al. Efficacy of pazopanib in progressive, radioiodine-refractory, metastatic differentiated thyroid cancers: results of a phase 2 consortium study. Lancet Oncol 2010;11: 962-72. [29] Kumasaka T, Seyama K, Mitani K, et al. Lymphangiogenesismediated shedding of LAM cell clusters as a mechanism for dissemination in lymphangioleiomyomatosis. Am J Surg Pathol 2005;29:1356-66. [30] Argani P, Aulmann S, Illei PB, et al. A distinctive subset of PEComas harbors TFE3 gene fusions. Am J Surg Pathol 2010;34:1395-406. [31] Iwamoto Y, Morimoto N, Chuman H, Shinohara N, Sugioka Y. The role of MR imaging in the diagnosis of alveolar soft part sarcoma: a report of 10 cases. Skeletal Radiol 1995;24:267-70. [32] Suh JS, Cho J, Lee SH, et al. Alveolar soft part sarcoma: MR and angiographic findings. Skeletal Radiol 2000;29:680-9.
