Abstract
Sex-specific differences are increasingly recognized in various diseases but have not yet been investigated for patients with primary CNS lymphoma (PCNSL). We collected clinical information and MR-images from 130 patients with PCNSL (66 f/ 64 m). We segmented MR-images into contrast-enhanced tumor, necrosis and edema for volumetric analysis and additionally quantified anti-PCNSL immune response using anti-CD3 staining in 67 cases. Median overall survival was 10 months for females and 8 months for males (p=0.979). Sex-specific Cox-regression analyses implicated age, performance and immunodeficiency as significant prognostic markers in females, whereas only age remained as significant marker in males. We then performed cluster analyses for females (fC) and males (mC) with complete sets of clinical, imaging, and tissue phenotypes (n=55). In females, two main clusters emerged that were mainly driven by clinical performance with fC1 (n=14) featuring patients with better performance (most of whom received MTX) compared to fC2 (n=18; mOS 21 vs 3 months, p< 0.01). fC2 resolved into two subclusters with better outcome for fC2a based on larger enhancing tumor volume and high immune response (mOS 12 vs. 1 months, p< 0.01). In males, two major clusters emerged (16 vs 7 patients, mOS 5 vs 23 months, p=0.414).), which differed mainly according to treatment approach with higher prevalence of MTX-chemotherapy in mC1. Each cluster could be subdivided into 2 subclusters based on differences in clinical performance in mC1, or according to treatment strategy, i.e., combined chemo-radiotherapy vs radiotherapy-only or best supportive care (mOS 49 vs 2 months, p=0.12) in mC2. In summary, we find prognostically relevant sex-specific clusters in patients with PCNSL that implicate differential roles of tumor-related contrast enhancement and immune response in female versus treatment modality in male patients. Initial differences in cluster-defining factors need further validation in independent cohorts but might have implications for differential patient management.
Development and Evaluation of a Semi-automated Segmentation Tool and a Modified Ellipsoid Formula for Volumetric Analysis of the Kidney in Non-contrast T2-Weighted MR Images
