Fig. 3
From: Clonal replacement and heterogeneity in breast tumors treated with neoadjuvant HER2-targeted therapy
Intra-tumor heterogeneity and clonal replacement in treated primary breast tumors. a Heterogeneity of protein-altering mutations in driver and targetable genes pre- and post-treatment, with multi-region sampling. b Cellular prevalence of inferred mutational clusters across treatment in P5 (PyClone). T1 is pre-treatment and T2 is post-treatment. The purple cluster reflects truncal mutations and the pink cluster reflects a new clone arising post-treatment. c Schematic describing geographic killing and clonal replacement. Untreated primaries must have regions of high homogeneity for geographic killing to mimic clonal replacement. d Values of tHFR in patient-treated tumors (dashed lines) compared to distributions of tHFR in virtual untreated tumors. For each case, tHFR is computed using 1, 2, and, when possible, 3 post-treatment samples (averaged over all possible combinations of subsets of post-treatment samples) to compare to the distribution derived from the virtual tumors for 1, 2, and 3 post-treatment samples. tHFR corresponds to the percentage of clonal mutations across all post-treatment samples that are absent or rare in a single pre-treatment sample. P1 and P5 exhibited tHFR values consistent with clonal evolution (not geographic killing). In certain cases, at least two post-treatment samples were necessary to discriminate between pre-treatment heterogeneity (P3) and clonal evolution (P5), as demonstrated within the simulation framework. Source data for panels a, b, and d are provided as a Source Data file
