Prabal Deb1, Dibyajyoti Boruah, Vibha Dutta. 1. Department of Pathology, Armed Forces Medical College, Pune-411040, Maharashtra, India. debprabal@gmail.com
Abstract
INTRODUCTION: Alterations of microvasculature are integral to CNS neoplasia, and a diagnostic feature of high-grade gliomas. The objectives of this study were two fold: First, to correlate morphometrically measured microvessel density (MVD), microvessel caliber (VC), and percentage of total microvessel area (%TVA) with WHO histologic grade in various types of primary CNS tumors. Second, to evaluate if such a correlation could be further refined by using mathematical derivatives of measured parameters namely coefficient of variation of VC (COofVC), microvessel cross-sectional area (VCSA), and percentage of total VCSA (%TVCSA). MATERIALS AND METHODS: Various microvessel parameters were assessed in a variety of 30 primary CNS tumors as consecutively encountered in routine surgical pathology practice including gliomas, meningiomas and others by image morphometry using CD34-immunostained sections. We introduced a novel method of effectively determining VC. Results were correlated with tumor type and grade. Appropriate statistical analysis was performed. RESULTS: Microvessel characteristics, especially VC (p<0.0022), VCSA (p<0.0164), CVofVC (p<0.0001), %TVCSA (p<0.0002) and %TVA (p<0.0003) of tumors were significantly greater than normal tissue. MVD increased in all tumors, excepting meningiomas, and was significantly higher in gliomas (p<0.0062). MVD showed negative correlation with VC (r=-0.808) and VCSA (r=-0.848) in the normal brain but was less significant in tumors. Unlike tumors, caliber distribution of microvessels in normal brain was noted to follow a Gaussian pattern. Histological grades of tumors showed positive correlation with MVD (r=0.547), VC (r=0.606), CVofVC (r=0.623), VCSA (r=0.485), %TVCSA (r=0.783) and %TVA (r=0.603). Calculated scores, estimated from multiple regressions of vessel parameters, correlated well with histological grade, with S2 (calculated using all measured as well as mathematically derived microvessel parameters) being better than S1 (calculated using measured parameters: MVD and VC). CONCLUSION: Tumor grades positively correlated with all microvessel parameters, with %TVCSA displaying the best. The correlation of %TVA with tumor grade was weaker than %TVCSA mainly due to the impact of MVD. These findings emphasize the value of VC as effectively measured using our novel method and best illustrated by its derivative %TVCSA (an indicator of blood flow), in addition to the well-recognized value of MVD in tumor prognostication. Multiple regressions of microvessel parameters provided the best correlation with grade. Morphometric analysis of microvessels in CNS tumor facilitates a better understanding of the tumor grade, tumor progression and overall prognosis.
INTRODUCTION: Alterations of microvasculature are integral to CNS neoplasia, and a diagnostic feature of high-grade gliomas. The objectives of this study were two fold: First, to correlate morphometrically measured microvessel density (MVD), microvessel caliber (VC), and percentage of total microvessel area (%TVA) with WHO histologic grade in various types of primary CNS tumors. Second, to evaluate if such a correlation could be further refined by using mathematical derivatives of measured parameters namely coefficient of variation of VC (COofVC), microvessel cross-sectional area (VCSA), and percentage of total VCSA (%TVCSA). MATERIALS AND METHODS: Various microvessel parameters were assessed in a variety of 30 primary CNS tumors as consecutively encountered in routine surgical pathology practice including gliomas, meningiomas and others by image morphometry using CD34-immunostained sections. We introduced a novel method of effectively determining VC. Results were correlated with tumor type and grade. Appropriate statistical analysis was performed. RESULTS: Microvessel characteristics, especially VC (p<0.0022), VCSA (p<0.0164), CVofVC (p<0.0001), %TVCSA (p<0.0002) and %TVA (p<0.0003) of tumors were significantly greater than normal tissue. MVD increased in all tumors, excepting meningiomas, and was significantly higher in gliomas (p<0.0062). MVD showed negative correlation with VC (r=-0.808) and VCSA (r=-0.848) in the normal brain but was less significant in tumors. Unlike tumors, caliber distribution of microvessels in normal brain was noted to follow a Gaussian pattern. Histological grades of tumors showed positive correlation with MVD (r=0.547), VC (r=0.606), CVofVC (r=0.623), VCSA (r=0.485), %TVCSA (r=0.783) and %TVA (r=0.603). Calculated scores, estimated from multiple regressions of vessel parameters, correlated well with histological grade, with S2 (calculated using all measured as well as mathematically derived microvessel parameters) being better than S1 (calculated using measured parameters: MVD and VC). CONCLUSION:Tumor grades positively correlated with all microvessel parameters, with %TVCSA displaying the best. The correlation of %TVA with tumor grade was weaker than %TVCSA mainly due to the impact of MVD. These findings emphasize the value of VC as effectively measured using our novel method and best illustrated by its derivative %TVCSA (an indicator of blood flow), in addition to the well-recognized value of MVD in tumor prognostication. Multiple regressions of microvessel parameters provided the best correlation with grade. Morphometric analysis of microvessels in CNS tumor facilitates a better understanding of the tumor grade, tumor progression and overall prognosis.