Testing tumors from different anatomic sites for clonal relatedness using somatic mutation data

Evaluating statistical approaches to define clonal origin of tumours using bulk DNA sequencing: context is everything

Clonal analysis of tumour sequencing data enables the evaluation of the relationship of histologically distinct synchronous lesions, such as co-existing benign areas, and temporally distinct tumours, such as primary-recurrence comparisons. In this review, we summarise statistical approaches that are commonly employed to define tumour clonal relatedness using data from bulk DNA technologies. We discuss approaches using total copy number, allele-specific copy number and mutation data, and the relative genomic resolution required for analysis and summarise some of the current tools for inferring clonal relationships. We argue that the impact of the biological context is critical in selecting any particular approach, such as the relative genomic complexity of the lesions being compared, and we recommend considering this context before employing any method to a new dataset.

Molecular evidence for a clonal relationship between synchronous uterine endometrioid carcinoma and ovarian clear cell carcinoma: a new example of "precursor escape"?

Synchronous endometrial and ovarian carcinomas (SEOCs) that share the same endometrioid histology are generally considered as the result of metastatic spread from one organ to another. However, SEOCs with different histologies are regarded as distinct primary lesions that arise independently from each other. This study was undertaken to compare the mutational landscape of SEOCs with different histologies to confirm or refute the hypothesis of an independent origin. Four patients with synchronous uterine endometrioid carcinoma (UEMC) and ovarian clear cell carcinoma (OCCC) were examined. UEMCs were accompanied by endometrial hyperplasia/endometrioid intraepithelial neoplasia, whereas endometriosis was evident in two cases. Paired UEMC and OCCC specimens were subjected to mutation analysis with massively parallel sequencing. Surprisingly, we found that 50% (2/4) of paired SEOCs with different histologies shared the same somatic mutations, some of which localized in cancer driver genes. Clonality analyses indicated that these tumors were clonally related to each other. Notably, 75% (3/4) of the study patients had Lynch syndrome. The cancer-specific survival figures of patients with synchronous UEMCs and OCCCs were more favorable than those observed in a historical cohort of patients with isolated stage 2/3 OCCCs. Taken together, we set forth a potential explanation that considers clonally related SEOCs as a result of "precursor escape" - whereby precursor cells of endometrial cancer spread beyond the uterus to reach the pelvis and eventually evolve into an OCCC under an increasing mutational burden. KEY MESSAGES: • SEOCs characterized by different histologies are rare. • All cases of SEOCs were accompanied by endometrial hyperplasia. • Fifty percent of SEOCs were clonally related to each other. • Shared mutations in cancer driver genes were evident among SEOCs. • Clonally related SEOCs may be a result of "precursor escape." • Lynch syndrome is highly prevalent in patients with UEMC and synchronous OCCC. • The prognosis of synchronous UEMC and OCCC was favorable.