Biological Properties And Hormonal Therapy Of Mangioma: Study On Nude Mice Xenograft Model And Human Pecimens

Objectives:1. To establish a reliable model of human hemangioma in nude mice and to observe the biological properties of xenografted human hemangioma and the related change of glucocorticoid receptor (GR) level in xenografted hemangioma.2. To observe the effect of intralesional administration of triamcinolone acetonide on the graft hemangioma by detecting the expression of GR, vascular endothelial growth factor (VEGF), and Ki-67, and to explore the possible mechanism.3. To determine the expression of GR in hamangiomas and vascular malformations, and to investigate the correlation among expression of GR, VEGF, and Ki-67.Methods:1. The specimen of proliferating hemangioma from a male infant of 3 monthes old was obtained by surgery. The tissue was cut into small pieces 5 mm X 4mm X 3 mm in size and grafted onto nude mice subcutaneously. Gross observation and histological examination was carried out at 30th, 45th, 52th, 59th, 66th, 80th, 90th, and 120th day after graft, respectively.Immunohistochemistry and image analysis were employed to detect the expression of GR, VEGF, and Ki-67 at different time courses.2. Triamcinolone acetonide was administered intralesionally at 45th day after graft. Grafted specimens, were taken before and at 3rd day, 1st week, and 2nd week after triamcinolone acetonide administration. In mice of control group, introlesional injection of the same volume of normal saline was given. GR, VEGF and Ki-67 were detected.3. From 2001 through 2003, specimens were taken from 43 cases of infantile vascular anomalies in the West China Hospital of Sichuan University. The lesions located cutanecusly or subcutaneously. All cases denied disease history of endocrine system and had not undergone previous glucocorticoid therapy, orally or intralesionally. The lesions were classified histopathologically according to Mulliken's criteria. GR, VEGF, and Ki-67 were detected on specimens of proliferating and involuting hemangiomas, vascular malformations, and normal subcutaneous tissue respectively. Results:1. Immunohistochemistry of mouse anti-human CD31: The intergrated optical densities of positive particles of CD31 were 19656.20 + 637.55 in grafts, and 20591.44±2263.02 in human hamengioma respectively, the difference is not significant (P>0.05), suggesting that the epithelial cells in the living grafts were of human origin.2. Histopathology of grafts: Within 3-4 weeks after graft, no dramatical enlargement in size was observed. Fortieth day after graft, most grafts were observed to grow with obvious neovascular network formation surrounding the grafts. The volum of graft hemangioma achieved the maximum at about60th day after graft when the grafts appeared purple under the skin, and red under dissection. Ninetieth day after graft, some grafts turned lighter and harden, with decrease in volume. Some residuum would be recognized 120th day after transplantation. Gragfts in proliferating phase: Great amount of endothelial cells were crowded in lobes. Karyomitosis was recognizable. Vascular lumens were few, narrow, and irregular. Grafts in involuting phase: Quantity of endothelial cells decreased with enlargement of vascular lumen. Fibrous tissue increased and deposited surrounding vascular lumens. Grafts in involuted phase: Primary vascular lumens were mostly substituted by fibrous tissue. Residual lumens were narrow and lined by flat endothelial cells.3. Glucocorticoid receptor: High positive staining of GR was obtained in grafts in proliferative phase and located in the cytoplasm of plump endothelial cells, while it was hardly recognizable in grafts in involuting or involuted phases. Such positive staining was common in specimens from surgically removed human proliferating hemangiomas. While in involuting hemangiomas, vascular malformations and normal skin, little positive staining of GR was obtained.4. Vascular endothelial growth factor (VEGF): Positive staining of VEGF was recognized in grafts and located in cytoplasm of endothelial cells in proliferating phase, and was hardly recognizable in grafts in involuting and involuted phases. As to human specimens, positive staining of VEGF located in cytoplasm of proliferating endothelial cells. Little positive staining of VEGF was obtained in human specimens from involuting hemangiomas, vascular malformations and normal subcutaneous tissue.5. Ki-67: Positive level of Ki-67 in proliferating grafts was significantly higher than that of involuting or involuted ones. High level positive staining of Ki-67 was also obsen'ed in human proliferating hemangiomas while it was very low in human specimens of involuting hemangiomas, vascular malformations, and normal subcutaneous tissue.6. Effect of triamcinolone acetonide injection on grafts: Significant decrease of graft volume was observed in the experimental group compared to that in control group, and the grafts in experimental group turned harden and whitish. Under microscopy, grafts of experimental group were mainly composed of lipofibrous tissue. Collapse and blockage of vascular lumens were frequent. While control grafts involved large amount of proliferative capillaries and signs of destruction was not observed. With administration of triamcinolone acetonide injectionon, the stainings of GR, VEGF, and Ki-67 turned weaker significantly. Significant differences of GR, VEGF, and Ki-67 staining were obtained between experimental group and control group, and between experimental group and normal subcutaneous tissue group.7. Coefficient correlation of GR to Ki-67 and VEGF: Coefficient correlations of GR to Ki-67 and VEGF were 0.766(P<0.01) and 0.643(P<0.01), respectively. Coefficient correlation of VEGF to Ki-67 was 0.763(P<0.01). After intralesional admin stration of triamcinolone acetonide, the coefficient correlations of GR to VEGF and to Ki-67 were 0.766(P<0.01) and 0.643(P<0.01), respectively. The coeffecient correlation of VEGF to Ki-67 was 0.567(P<0.05). In human hemangioma specimens, the coeffecient correlations of GR to Ki-67 and to VEGF were 0.945(P<0.01) and 0.962(P<0.01), respectively. The coeffecient correlation of VEGF to Ki-67was0.963(P<0.01). Conclusions:1. Planting of human subcutaneous hemangioms onto nude mice subcutaneously is a feasible way to establish in vivo model of human hemangioma. In this model, the histological and biological properties of human hemangoma were retained with stable growth which resembles that of natural pattern. Living grafts held 84.4% of all grafts. Being convenient to operate and duplicable are also priorities worthy noting. When compared to other contemporary models such as in vitro cell culture, cockscomb, spleen, and polyoma-induced vascular tumors in animal, this model is more reliable and valid in that it in a way resembles the natural development of human hemangioma.2. Strong positive staining of GR was observed in grafts in proliferating phase and human specimens of proliferating hemangiomas, while in grafts in involuting and involuted phase and specimens of human involuting hemangiomas and vascular malformations, little, if any, positive staining of GR could be observed. GR level was positively related to those of VEGF and Ki-67 in grafts and hurran hemangiomas. After intralesional administration of triamcinolone acetonide, the grafts involuted within 2 weeks with VEGF and Ki-67 positive staining decreased. At the same time, the grafts volume enlarged and positive staining levels of VEGF and Ki-67 elevated after intralesional injection of normal saline in control group. It suggests that intralesional administration of triamcinolone acetonide may lead to decrease of VEGF secretion and then inhibit proliferative activity of endothelial cells, and promote involution of hemangioma. After intralesional administration oftriamcinolone acetonide positive staining level of GR was decreased as well as those of VEGF and Ki-67. Expression of GR was positively related to those of VEGF and Ki-67. It suggests that the decrease of GR might related to biding of GR by glucocorticoid. GR level might play an important role in development of hemangiomas. The effect of external glucocorticoid on hemangioma might be mediated by GR, i.e., glucocorticoid binds to it's specific receptor and forms a glucocorticoid-GR complex, and the complex interacts with glucocorticoid response element in target gene, and then inhibit secretion of VEGF. With decreased VEGF secretion, proliferation of endothelial cells is hampered and the hemangioma involutes. In vasclular malformations and involuting hemangiomas, weak or no expression of GR, VEGF, and Ki-67 could be observed. With low quantity of GRs in vascular malformation, glucocortscoid may be difficult to play its biological role, so glucocorticoids are kss likely effective in treatment of vascular malformations. Meanwhile, this also suggests that the biological properties of vaslular malformations and hemangioma are different indeed. This investigation presents further evidence for the validity and clinical implication of Mulliken's classification of vascular anomalies, and the conventional classification of "angioma" cann' t show the difference of the basic biological properties among vascular anomalies.