Microvascular structural entropy: a novel approach to assess angiogenesis in pituitary tumors

Abstract

Entropy, a measure of the degree of disorder in a system, has recently been used in different morphologic studies to quantify regularity. Our aims were (a) to study the structural organization of the microvascular bed in prolactin (PRL)-producing adenomas and carcinomas, the most vascularized of pituitary tumors, by assessing microvascular structural entropy (MSE), and (b) to determine whether the degree of disorder of the capillary bed correlates with tumor cell proliferation as estimated by MIB-1 labeling, microvessel density (MVD), the most widely used method of quantifying blood vessel formation, and various clinicopathologic parameters (gender, age, tumor size and invasiveness). The morphometric study demonstrated statistically significant differences in MIB-1 labeling, MVD, and MSE between PRL-producing adenomas and carcinomas. Unlike MIB-1 labeling index (PRL-producing adenomas 1.5 +/- 0.27; carcinomas 15.0 +/- 4.04) and MVD (PRL-producing adenomas 2.7 +/- 0.34; carcinomas 4.2 +/- 0.72), the MSE values were significantly higher in adenomas (171.5 +/- 25.37) than in carcinomas (67.9 +/- 17.45). These results indicate that PRL-producing carcinomas have a less chaotic distribution of vessels than benign adenomas. In contrast to a lack of correlation between microvessel density and other morphometric parameters, a strong negative correlation was found between MSE and MIB-1 labeling index (r = 0.511, p = 0.003). It thus appears that regular, less chaotic microvascular geometry contributes to increased proliferative activity in PRL cell tumors. Analysis of MSE may provide an independent parameter of tumor behavior, and contributes to a better understanding of the role of microvasculature in pituitary tumor progression.