Limits...
Determining the origin of synchronous multifocal bladder cancer by exome sequencing.

Acar Ö, Özkurt E, Demir G, Saraç H, Alkan C, Esen T, Somel M, Lack NA - BMC Cancer (2015)

Bottom Line: Interestingly, in two patients the most common type of tumour-associated SNVs were cytosine mutations of TpC* dinucleotides (Fisher's exact test p < 10(-41)), likely caused by APOBEC-mediated deamination.Incorporating these results into our clonal model, we found that TpC* type mutations occurred 2-5× more often among SNVs on the ancestral branches than in the more recent private branches (p < 10(-4)) suggesting that TpC* mutations largely occurred early in the development of the tumour.These results demonstrate that synchronous multifocal bladder cancers frequently arise from a clonal origin.

View Article: PubMed Central - PubMed

Affiliation: Department of Urology, School of Medicine, Koc University, Istanbul, Turkey. oacar@kuh.ku.edu.tr.

ABSTRACT

Background: Synchronous multifocal tumours are commonly observed in urothelial carcinomas of the bladder. The origin of these physically independent tumours has been proposed to occur by either intraluminal migration (clonal) or spontaneous transformation of multiple cells by carcinogens (field effect). It is unclear which model is correct, with several studies supporting both hypotheses. A potential cause of this uncertainty may be the small number of genetic mutations previously used to quantify the relationship between these tumours.

Methods: To better understand the genetic lineage of these tumours we conducted exome sequencing of synchronous multifocal pTa urothelial bladder cancers at a high depth, using multiple samples from three patients.

Results: Phylogenetic analysis of high confidence single nucleotide variants (SNV) demonstrated that the sequenced multifocal bladder cancers arose from a clonal origin in all three patients (bootstrap value 100 %). Interestingly, in two patients the most common type of tumour-associated SNVs were cytosine mutations of TpC* dinucleotides (Fisher's exact test p < 10(-41)), likely caused by APOBEC-mediated deamination. Incorporating these results into our clonal model, we found that TpC* type mutations occurred 2-5× more often among SNVs on the ancestral branches than in the more recent private branches (p < 10(-4)) suggesting that TpC* mutations largely occurred early in the development of the tumour.

Conclusions: These results demonstrate that synchronous multifocal bladder cancers frequently arise from a clonal origin. Our data also suggests that APOBEC-mediated mutations occur early in the development of the tumour and may be a driver of tumourigenesis in non-muscle invasive urothelial bladder cancer.

No MeSH data available.


Related in: MedlinePlus

SNV distribution and phylogeny of tumours. a Hypothetical phylogenies with the field effect and monoclonal origin hypotheses. b Number of SNVs with respect to occurrence among samples. Tumour 1 Apex, Tumour 1 Base, Tumour 2 Base/Apex, Normal private: SNVs only in that sample. As there are only 3 normal private mutations their bar is too small to be visually noticeable. All samples shared: SNVs in all 4 samples, representing the individual genotype. Other: SNVs in the normal mucosa sample and in one or in two tumour samples. All tumours shared: SNVs in all 3 tumour samples but not in normal mucosa. Other tumour-associated: SNVs in found in one or two but not all tumours and not in normal mucosa. c Neighbour-joining tree of the 4 samples and the human reference genome based on 1628–1733 high-confidence SNVs. Bootstrap support for each internal node is indicated. d Proportion of functional SNVs among all SNVs shared among all 4 samples (n = 143–182), and SNVs shared among all 3 tumours (n = 116–152). ***: Fisher’s exact test p < 0.001
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4638097&req=5

Fig1: SNV distribution and phylogeny of tumours. a Hypothetical phylogenies with the field effect and monoclonal origin hypotheses. b Number of SNVs with respect to occurrence among samples. Tumour 1 Apex, Tumour 1 Base, Tumour 2 Base/Apex, Normal private: SNVs only in that sample. As there are only 3 normal private mutations their bar is too small to be visually noticeable. All samples shared: SNVs in all 4 samples, representing the individual genotype. Other: SNVs in the normal mucosa sample and in one or in two tumour samples. All tumours shared: SNVs in all 3 tumour samples but not in normal mucosa. Other tumour-associated: SNVs in found in one or two but not all tumours and not in normal mucosa. c Neighbour-joining tree of the 4 samples and the human reference genome based on 1628–1733 high-confidence SNVs. Bootstrap support for each internal node is indicated. d Proportion of functional SNVs among all SNVs shared among all 4 samples (n = 143–182), and SNVs shared among all 3 tumours (n = 116–152). ***: Fisher’s exact test p < 0.001

Mentions: Synchronous multifocal tumours are present in ~30 % of all non-muscle invasive urothelial carcinomas of the bladder. Two competing theories have been proposed to describe how these physically separated independent tumours arise. The clonal hypothesis suggests that multifocal tumours are formed by intraluminal or intraepithelial migration of cells that are shed from a founder tumour. In contrast, the field hypothesis proposes that a large area of cells is first partially transformed by a carcinogen and then subsequently acquires additional mutations that induce neoplastic transformation. As cancer occurs by the sequential accumulation of transformative mutations, each model would produce a different genetic signature (Fig. 1a). Various studies investigating the origin of multifocal bladder cancer have produced conflicting reports. Classic work characterizing the X-chromosomal inactivation patterns in female patients [1] and microsatellite loss of heterozygosity (LOH) [2] clearly demonstrated a clonal origin. Further, tumours analysed by array-based comparative genomic hybridization also showed monoclonality [3]. However, in a study of 21 bladder cancer patients, ~30 % of patients demonstrated significant LOH allelic differences among tumours with >40 % of patients showing some allelic variance, suggesting that these tumours arose from a field effect [4]. Additional research with microdissected bladder tumour samples also demonstrated similar results in accordance with a field effect hypothesis [5, 6]. Interestingly, in a recent study of metachronous multifocal tumours, the authors clearly demonstrated that these arose from a clonal origin [7].Fig. 1


Determining the origin of synchronous multifocal bladder cancer by exome sequencing.

Acar Ö, Özkurt E, Demir G, Saraç H, Alkan C, Esen T, Somel M, Lack NA - BMC Cancer (2015)

SNV distribution and phylogeny of tumours. a Hypothetical phylogenies with the field effect and monoclonal origin hypotheses. b Number of SNVs with respect to occurrence among samples. Tumour 1 Apex, Tumour 1 Base, Tumour 2 Base/Apex, Normal private: SNVs only in that sample. As there are only 3 normal private mutations their bar is too small to be visually noticeable. All samples shared: SNVs in all 4 samples, representing the individual genotype. Other: SNVs in the normal mucosa sample and in one or in two tumour samples. All tumours shared: SNVs in all 3 tumour samples but not in normal mucosa. Other tumour-associated: SNVs in found in one or two but not all tumours and not in normal mucosa. c Neighbour-joining tree of the 4 samples and the human reference genome based on 1628–1733 high-confidence SNVs. Bootstrap support for each internal node is indicated. d Proportion of functional SNVs among all SNVs shared among all 4 samples (n = 143–182), and SNVs shared among all 3 tumours (n = 116–152). ***: Fisher’s exact test p < 0.001
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4638097&req=5

Fig1: SNV distribution and phylogeny of tumours. a Hypothetical phylogenies with the field effect and monoclonal origin hypotheses. b Number of SNVs with respect to occurrence among samples. Tumour 1 Apex, Tumour 1 Base, Tumour 2 Base/Apex, Normal private: SNVs only in that sample. As there are only 3 normal private mutations their bar is too small to be visually noticeable. All samples shared: SNVs in all 4 samples, representing the individual genotype. Other: SNVs in the normal mucosa sample and in one or in two tumour samples. All tumours shared: SNVs in all 3 tumour samples but not in normal mucosa. Other tumour-associated: SNVs in found in one or two but not all tumours and not in normal mucosa. c Neighbour-joining tree of the 4 samples and the human reference genome based on 1628–1733 high-confidence SNVs. Bootstrap support for each internal node is indicated. d Proportion of functional SNVs among all SNVs shared among all 4 samples (n = 143–182), and SNVs shared among all 3 tumours (n = 116–152). ***: Fisher’s exact test p < 0.001
Mentions: Synchronous multifocal tumours are present in ~30 % of all non-muscle invasive urothelial carcinomas of the bladder. Two competing theories have been proposed to describe how these physically separated independent tumours arise. The clonal hypothesis suggests that multifocal tumours are formed by intraluminal or intraepithelial migration of cells that are shed from a founder tumour. In contrast, the field hypothesis proposes that a large area of cells is first partially transformed by a carcinogen and then subsequently acquires additional mutations that induce neoplastic transformation. As cancer occurs by the sequential accumulation of transformative mutations, each model would produce a different genetic signature (Fig. 1a). Various studies investigating the origin of multifocal bladder cancer have produced conflicting reports. Classic work characterizing the X-chromosomal inactivation patterns in female patients [1] and microsatellite loss of heterozygosity (LOH) [2] clearly demonstrated a clonal origin. Further, tumours analysed by array-based comparative genomic hybridization also showed monoclonality [3]. However, in a study of 21 bladder cancer patients, ~30 % of patients demonstrated significant LOH allelic differences among tumours with >40 % of patients showing some allelic variance, suggesting that these tumours arose from a field effect [4]. Additional research with microdissected bladder tumour samples also demonstrated similar results in accordance with a field effect hypothesis [5, 6]. Interestingly, in a recent study of metachronous multifocal tumours, the authors clearly demonstrated that these arose from a clonal origin [7].Fig. 1

Bottom Line: Interestingly, in two patients the most common type of tumour-associated SNVs were cytosine mutations of TpC* dinucleotides (Fisher's exact test p < 10(-41)), likely caused by APOBEC-mediated deamination.Incorporating these results into our clonal model, we found that TpC* type mutations occurred 2-5× more often among SNVs on the ancestral branches than in the more recent private branches (p < 10(-4)) suggesting that TpC* mutations largely occurred early in the development of the tumour.These results demonstrate that synchronous multifocal bladder cancers frequently arise from a clonal origin.

View Article: PubMed Central - PubMed

Affiliation: Department of Urology, School of Medicine, Koc University, Istanbul, Turkey. oacar@kuh.ku.edu.tr.

ABSTRACT

Background: Synchronous multifocal tumours are commonly observed in urothelial carcinomas of the bladder. The origin of these physically independent tumours has been proposed to occur by either intraluminal migration (clonal) or spontaneous transformation of multiple cells by carcinogens (field effect). It is unclear which model is correct, with several studies supporting both hypotheses. A potential cause of this uncertainty may be the small number of genetic mutations previously used to quantify the relationship between these tumours.

Methods: To better understand the genetic lineage of these tumours we conducted exome sequencing of synchronous multifocal pTa urothelial bladder cancers at a high depth, using multiple samples from three patients.

Results: Phylogenetic analysis of high confidence single nucleotide variants (SNV) demonstrated that the sequenced multifocal bladder cancers arose from a clonal origin in all three patients (bootstrap value 100 %). Interestingly, in two patients the most common type of tumour-associated SNVs were cytosine mutations of TpC* dinucleotides (Fisher's exact test p < 10(-41)), likely caused by APOBEC-mediated deamination. Incorporating these results into our clonal model, we found that TpC* type mutations occurred 2-5× more often among SNVs on the ancestral branches than in the more recent private branches (p < 10(-4)) suggesting that TpC* mutations largely occurred early in the development of the tumour.

Conclusions: These results demonstrate that synchronous multifocal bladder cancers frequently arise from a clonal origin. Our data also suggests that APOBEC-mediated mutations occur early in the development of the tumour and may be a driver of tumourigenesis in non-muscle invasive urothelial bladder cancer.

No MeSH data available.


Related in: MedlinePlus