Rau, Sabrina

Intratumoral heterogeneity and clonal evolution are pivotal in the progression and metastasis of melanoma. However, when combined with variable tumor cellularity, intratumoral heterogeneity limits the sensitivity and accuracy of uncovering a cancer’s clonal evolution. In this study, we combined fluorescence-activated cell sorting (FACS) with whole-exome sequencing (WES) to investigate the clonal composition and evolutionary patterns in seven melanoma biopsies obtained from three patients, each having both primary site and metastatic samples. We employed a multiparameter ploidy sorting approach to isolate tumor populations based on DNA ploidy and melanoma biomarkers (SOX10 or S100), enabling us to investigate clonal evolution with high resolution. Our approach increased the mean tumor purity from 70% (range 19–88%) in unsorted material to 91% (range 87–96%) post-sorting. Our findings revealed significant inter- and intratumor heterogeneity, with one patient exhibiting two genomically distinct clonal tumor populations within a single primary site biopsy, each giving rise to different metastases. Our findings highlight the critical role of intratumoral heterogeneity and clonal evolution in melanoma, especially when analyzing tumor trajectories. The unique combination of multiparameter FACS and WES provides a powerful method for identifying clonal populations and reconstructing clonal evolution. This study provides valuable insights into the clonal architecture of melanoma and lays the groundwork for future research with larger patient groups.

Abstract Background Genomic intratumor heterogeneity is an increasingly recognized phenomenon in various solid tumors and is now extensively being investigated by next-generation sequencing (NGS) approaches. However, most methods are not designed to resolve the complexity of mixed cell populations. Therefore important information may be lost, eventually leading to assumptions that may not reflect the true nature of the disease. We have therefore developed the method of Multiparameter Ploidy Profiling (MPP). We applied this technology on tumor specimens from malignant melanoma and from non-small cell lung cancer (NSCLC). Methods MPP involves isolation of nuclei from tissues, multiparameter flow-sorting of tumor populations by DNA content and specific tumor markers, and profiling genomes using high resolution array comparative genomic hybridization (aCGH) and NGS. As some tumors are diploid by flow cytometry, specific lineage markers, e.g. SOX10 (melanoma) and TTF1 (NSCLC) were used to separate these from normal cells. Fluorescence-in-situ hybridization (FISH) and immunohistochemistry (IHC) were performed on sorted nuclei and on whole tissue sections to validate results. Results MPP allowed us to sort different tumor populations within a single tumor specimen, while excluding non-neoplastic contamination in downstream analysis with aCGH and NGS. We separated three different tumor populations from a single primary NSCLC with a TTF1 multiparameter sort. We detected significant differences in copy numbers, e.g. a CDKN2A homozygous deletion only present in the diploid tumor population as well as differences in allelic frequencies of gene mutations (e.g. EGFR exon 20). Further sequencing of the diploid TTF1 negative population proved to be concordant to genomic DNA obtained from a tumor free lymph node. Finally, we used the population-specific genomic data in order to infer the clonal relationship of all tumor populations and to postulate the evolution of these tumor populations within single tumor specimens. Conclusions The power of MPP consists of sorting pure tumor populations and of deciphering the level of intratumor heterogeneity based on ploidy. Most importantly MPP enables the separation of diploid tumor populations from diploid, benign stromal cells that can serve as reference for genomic profiling studies, where germline DNA is otherwise not available. To our knowledge, diploid tumor populations were for first time purifyed from normal cells in a multiparameter sort and genomically characterized. We also demonstrated that the analysis of diploid tumor populations within a single biopsy is crucial for the understanding of the clonal composition and evolution of a given tumor. MPP on multiple biopsies from individual patients is necessary to deeply investigate the clonal evolution and the behavior of a diploid tumor population and their potential impact on metastasis and therapy response. Citation Format: Thomas Lorber, Sabrina Rau, Valeria Perrina, Michael Barrett, Christian Ruiz, Lukas Bubendorf. Multiparameter ploidy profiling: a powerful tool to investigate the genomics of diploid tumor populations. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4924. doi:10.1158/1538-7445.AM2015-4924