Figure 3. Genomic evolution and... | American Society of Hematology

Figure 3.

$Genomic evolution and druggability. Mutational dynamics and evolutionary divergence in serially sampled rrDLBCL cases. (A) rrDLBCL cases that were serially sampled between treatment regimens were analyzed to compare the genomic alterations before and after treatment. Variants were classified as shared (present in both biopsies) or private (present in only 1 biopsy). For each case, the percent of shared variants between the diagnostic and relapse biopsies was calculated and plotted against time between biopsies. Biopsy site location and type of sequencing (WES or targeted) were integrated as the color and shape of each point, respectively. A negative linear correlation between genetic similarity and time between biopsies was observed (P < .00016). (B) Below the plot, a schematic example is shown: over time, a founding clone (white circle with black founding mutations) evolves into the clones present at diagnosis (green with orange private mutations) and at relapse (blue, with red private mutations). Clonal evolutionary analysis and identification of druggable alterations was performed, n = 18. (C) Schematic images of representative biopsies are shown with inferred clonal population percentages and branching evolutionary trees. Clinical information including subtype and treatment are displayed. To the right of each case, “fish plots” illustrate the clonal architecture of the 2 tumor biopsies, inferred from clonal fractions of mutation clusters and response to treatment. Mutations in genes of interest are shown in clones when they occurred, with genes involved in antigen presentation highlighted in red. (D) Druggable somatic mutations were identified using the Open Targets Platform, and oncogenic potential was estimated using CADD. Histogram of mutation counts of those variants identified as druggable and oncogenic in serially biopsied cases colored by shared vs private status, relapse vs diagnostic (n = 18, b = 37) paired Student t test (n = 16 (excluded relapse-relapse serial pairs), P = .0221). (E) Illustrative branched evolutionary trees representing founding clone mutations (gray trunk) and mutations private to diagnostic biopsies and relapsed biopsies (green and blue, respectively). Druggable mutations in genes of interest are displayed.$

Genomic evolution and druggability. Mutational dynamics and evolutionary divergence in serially sampled rrDLBCL cases. (A) rrDLBCL cases that were serially sampled between treatment regimens were analyzed to compare the genomic alterations before and after treatment. Variants were classified as shared (present in both biopsies) or private (present in only 1 biopsy). For each case, the percent of shared variants between the diagnostic and relapse biopsies was calculated and plotted against time between biopsies. Biopsy site location and type of sequencing (WES or targeted) were integrated as the color and shape of each point, respectively. A negative linear correlation between genetic similarity and time between biopsies was observed (P < .00016). (B) Below the plot, a schematic example is shown: over time, a founding clone (white circle with black founding mutations) evolves into the clones present at diagnosis (green with orange private mutations) and at relapse (blue, with red private mutations). Clonal evolutionary analysis and identification of druggable alterations was performed, n = 18. (C) Schematic images of representative biopsies are shown with inferred clonal population percentages and branching evolutionary trees. Clinical information including subtype and treatment are displayed. To the right of each case, “fish plots” illustrate the clonal architecture of the 2 tumor biopsies, inferred from clonal fractions of mutation clusters and response to treatment. Mutations in genes of interest are shown in clones when they occurred, with genes involved in antigen presentation highlighted in red. (D) Druggable somatic mutations were identified using the Open Targets Platform, and oncogenic potential was estimated using CADD. Histogram of mutation counts of those variants identified as druggable and oncogenic in serially biopsied cases colored by shared vs private status, relapse vs diagnostic (n = 18, b = 37) paired Student t test (n = 16 (excluded relapse-relapse serial pairs), P = .0221). (E) Illustrative branched evolutionary trees representing founding clone mutations (gray trunk) and mutations private to diagnostic biopsies and relapsed biopsies (green and blue, respectively). Druggable mutations in genes of interest are displayed.

This Feature Is Available To Subscribers Only