Contributions
Abstract: S1573
Type: Oral Presentation
Presentation during EHA23: On Sunday, June 17, 2018 from 08:45 - 09:00
Location: Room A6
Background
Exome-wide characterization of somatic mutations in large cohorts of chronic lymphocytic leukemia (CLL) patients provided a comprehensive catalogue of alterations that putatively drive the disease. But how the genomic features of a patient’s CLL cells relate to individual disease kinetics remains poorly understood.
Aims
We leveraged the indolent growth dynamics of CLL to perform a time-resolved analysis of growth rates and genomic patterns of leukemia clones from 113 patients, spanning decades-long disease courses.
Methods
For high-resolution characterization we selected 21 CLL patients who progressed to need for treatment, through the CLL Research Consortium, housed at UCSD. We performed whole exome sequencing on 2-6 serial samples spanning the period from diagnosis to start of chemoimmunotherapy. Using pre-treatment measurements of white blood cell counts (WBC), we determined growth kinetics over time. We validated findings in a cohort of pre-treatment samples from 92 CLL patients enrolled at the Dana-Farber Cancer Institute (DFCI, Boston, MA).
For whole-exome sequencing (WES), we used Broad Institute sequencing and analysis protocols. Cancer cell fractions of mutations were estimated using the ABSOLUTE algorithm to calculate the purity, ploidy, and absolute DNA copy-numbers of each sample. To infer the clonal structure in serial samples, we employed a multi-sample phylogeny and clustering analysis tool PhylogicNDT. Growth rates per subclone were estimated based on a Monte Carlo Markov Chain algorithm integrating the uncertainty of the cluster assignment, given the proposed phylogeny and WBC measurements. Reference lists for SNVs and InDels in known putative CLL driver genes as well as for recurrent CNAs were concatenated based on previous sequencing studies of large CLL cohorts.
Results
We found that CLL commonly demonstrates not only exponential expansion but also logistic growth, which is sigmoidal and includes stabilization at a certain steady state level. In patients, whose serial samples underwent whole-exome sequencing, we found that dynamic changes in the CLL disease course were shaped by the activity of genetic events that were already present in earliest indolent stages. Each growth pattern was associated with marked differences in genetic composition, pace of disease progression and extent of clonal evolution. Finally, using an integrative analysis of somatic mutations and growth rates, we quantified the growth advantage putative CLL drivers confer in vivo. Accelerated growth of subclones was strongly enriched for presence of well-established statistically identified CLL drivers, for which we could also directly quantify the impact of specific mutations, such as second hits in the major tumor suppressor genes TP53 and ATM.
Conclusion
By integrating information on tumor growth and genetic evolution information, we observed complex patterns even during the pre-treatment period, with various growth behavior and expansion rates amongst subclones over time. Our computational framework enabled us to assess the degree of growth acceleration of genetically defined subclones over their parental clones. This analysis provides clear evidence for the growth-accelerating role and potential synergies of subclonal driver mutations, while demonstrating the frequent existence of growth-neutral subclones that did not contain drivers and likely represent genetic drift. Our real-life in vivo data thus provide a conceptual framework for understanding the growth trajectories of individual populations of CLL cells.
Session topic: 5. Chronic lymphocytic leukemia and related disorders – Biology & Translational Research
Keyword(s): Chronic Lymphocytic Leukemia, Clonal expansion, Progression, Somatic mutation
Abstract: S1573
Type: Oral Presentation
Presentation during EHA23: On Sunday, June 17, 2018 from 08:45 - 09:00
Location: Room A6
Background
Exome-wide characterization of somatic mutations in large cohorts of chronic lymphocytic leukemia (CLL) patients provided a comprehensive catalogue of alterations that putatively drive the disease. But how the genomic features of a patient’s CLL cells relate to individual disease kinetics remains poorly understood.
Aims
We leveraged the indolent growth dynamics of CLL to perform a time-resolved analysis of growth rates and genomic patterns of leukemia clones from 113 patients, spanning decades-long disease courses.
Methods
For high-resolution characterization we selected 21 CLL patients who progressed to need for treatment, through the CLL Research Consortium, housed at UCSD. We performed whole exome sequencing on 2-6 serial samples spanning the period from diagnosis to start of chemoimmunotherapy. Using pre-treatment measurements of white blood cell counts (WBC), we determined growth kinetics over time. We validated findings in a cohort of pre-treatment samples from 92 CLL patients enrolled at the Dana-Farber Cancer Institute (DFCI, Boston, MA).
For whole-exome sequencing (WES), we used Broad Institute sequencing and analysis protocols. Cancer cell fractions of mutations were estimated using the ABSOLUTE algorithm to calculate the purity, ploidy, and absolute DNA copy-numbers of each sample. To infer the clonal structure in serial samples, we employed a multi-sample phylogeny and clustering analysis tool PhylogicNDT. Growth rates per subclone were estimated based on a Monte Carlo Markov Chain algorithm integrating the uncertainty of the cluster assignment, given the proposed phylogeny and WBC measurements. Reference lists for SNVs and InDels in known putative CLL driver genes as well as for recurrent CNAs were concatenated based on previous sequencing studies of large CLL cohorts.
Results
We found that CLL commonly demonstrates not only exponential expansion but also logistic growth, which is sigmoidal and includes stabilization at a certain steady state level. In patients, whose serial samples underwent whole-exome sequencing, we found that dynamic changes in the CLL disease course were shaped by the activity of genetic events that were already present in earliest indolent stages. Each growth pattern was associated with marked differences in genetic composition, pace of disease progression and extent of clonal evolution. Finally, using an integrative analysis of somatic mutations and growth rates, we quantified the growth advantage putative CLL drivers confer in vivo. Accelerated growth of subclones was strongly enriched for presence of well-established statistically identified CLL drivers, for which we could also directly quantify the impact of specific mutations, such as second hits in the major tumor suppressor genes TP53 and ATM.
Conclusion
By integrating information on tumor growth and genetic evolution information, we observed complex patterns even during the pre-treatment period, with various growth behavior and expansion rates amongst subclones over time. Our computational framework enabled us to assess the degree of growth acceleration of genetically defined subclones over their parental clones. This analysis provides clear evidence for the growth-accelerating role and potential synergies of subclonal driver mutations, while demonstrating the frequent existence of growth-neutral subclones that did not contain drivers and likely represent genetic drift. Our real-life in vivo data thus provide a conceptual framework for understanding the growth trajectories of individual populations of CLL cells.
Session topic: 5. Chronic lymphocytic leukemia and related disorders – Biology & Translational Research
Keyword(s): Chronic Lymphocytic Leukemia, Clonal expansion, Progression, Somatic mutation