Spatiotemporal Heterogeneity Characterizes the Genetic Landscape of Pheochromocytoma and Defines Early Events in Tumorigenesis.

PURPOSE Pheochromocytoma and paraganglioma (PPGL) patients display heterogeneity in the clinical presentation and underlying genetic cause. The degree of inter- and intratumor genetic heterogeneity has not yet been defined. EXPERIMENTAL DESIGN In PPGLs from 94 patients, we analyzed LOH, copy-number variations, and mutation status of SDHA, SDHB, SDHC, SDHD, SDHAF2, VHL, EPAS1, NF1, RET, TMEM127, MAX, and HRAS using high-density SNP array and targeted deep sequencing, respectively. Genetic heterogeneity was determined through (i) bioinformatics analysis of individual samples that estimated absolute purity and ploidy from SNP array data and (ii) comparison of paired tumor samples that allowed reconstruction of phylogenetic trees. RESULTS Mutations were found in 61% of the tumors and correlated with specific patterns of somatic copy-number aberrations (SCNA) and degree of nontumoral cell admixture. Intratumor genetic heterogeneity was observed in 74 of 136 samples using absolute bioinformatics estimations and in 22 of 24 patients by comparison of paired samples. In addition, a low genetic concordance was observed between paired primary tumors and distant metastases. This allowed for reconstructing the life history of individual tumors, identifying somatic mutations as well as copy-number loss of 3p and 11p (VHL subgroup), 1p (Cluster 2), and 17q (NF1 subgroup) as early events in PPGL tumorigenesis. CONCLUSIONS Genomic landscapes of PPGL are specific to mutation subtype and characterized by genetic heterogeneity both within and between tumor lesions of the same patient.