Advertisement

Correlation of CAD Peak Lesion Enhancement with Quantitative Tumor Angiogenesis to Non-Invasively Assess Furhman Grades I-IV in Patients with Clear Cell Renal Cell Carcinoma

Login to Access Video or Poster Abstract: MP22-07
Sources of Funding: None

Introduction

To assess if Computer Aided Detection (CAD) of peak lesion attenuation discriminates among Fuhrman Grades I-IV and correlates with an increase in tumor angiogenesis in clear cell RCC (ccRCC) on four-phase MDCT.

Methods

We reviewed a cohort of patients with ccRCC and preoperative multiphasic multidetector CT imaged with a 4-phase renal mass protocol (unenhanced, corticomedullary (C), nephrographic (N), and excretory (E)). A whole lesion 3D contour was obtained in all phases with proprietary software. The CAD algorithm determined a 0.5cm diameter region of peak enhancement ≤300HU within the 3D lesion contour. For assessment of quantitative angiogenesis, immunohistochemical staining for CD34 to determine microvessel density (MVD) was performed. T-tests were used to compare peak multiphasic enhancement and microvessel density among Fuhrman grades I-IV. P values less than 0.05 were considered to be significant.

Results

107 patients (71(64%) men and 40(35%) women) with 111 unique ccRCC lesions (16 (14%) Fuhrman grade I, 64 (58%) Fuhrman grade II, 23 (21%) Fuhrman Grade III, 8 (7%) Fuhrman grade IV) were analyzed. In the C phase, CAD peak lesion enhancement discriminated grade I from II (150 HU vs. 185 HU, p=0.006), I from III (150 HU vs. 178 HU, p=0.054), I from IV (150 HU vs 229HU, p<0.001). This directly correlated with an increase in quantitative angiogenesis (MVD): I from II (2134 mm2 vs. 4710 mm2, p=0.004), I from III (2134 mm2 vs. 5162 mm, p=0.001), I from IV (2134 mm2 vs 6076 mm2, p=0.057).

Conclusions

CAD peak lesion enhancement discriminates Fuhrman grades 1-IV on multiphasic CT and correlates with an increase in tumor angiogenesis. This may be helpful to triage patients to active surveillance, interventional therapy or, if validated, may be useful to monitor results with anti-angiogenic therapy.

Funding

None

Authors
Heidi Coy
Jonathan Young
Michael Douek
Matthew Brown
Anthony Sisk
James Sayre
Steven Raman
back to top