🚨 Finally, we show that, not CH in general, but **positive selection of CH muts** (eg, driven by chemo or radiation), is associated with OS, PFS, and RFS. These associations persist in multivariable, adjusted models.

These findings were corroborated in 2 external cohorts - one breast cancer and one pan-cancer cohort.

Neff was associated with dose - pts with higher exposure to chemo had smaller Neff (ie, stronger positive selection).

Neff: the min. # of clonally independent alleles needed to explain the observed changes in VAF during tx in a pt.
Low Neff: strong / positive selection;
High Neff: weak / negative or no selection.
Chemo and chemo+rad exerted the strongest selective pressure on CH clones.

All post-tx CH muts were present before tx. All CHIP muts persisted at CHIP-defining VAFs; 4% of CH expanded to CHIP. Short-term, VAF changes were highest in pts tx-ed with chemo, followed by rad & chemo+rad; pts tx-ed with only hormone therapy had essentially no change in VAFs.

55% of patients had #CH before tx; 33% had #CHIP. Of those with CH, 56% had only 1 mutation (mut). We quantified clonal diversity using divergence in variant allele frequencies (VAFs) and show that as CH expands, a larger number of CH muts may arise, increasing clonal diversity.

We conducted ultra-deep error-corrected sequencing (>5000x) on serial samples from before, during, after, and long after treatment (tx) for breast cancer - with chemo, radiation, chemo+rad, or hormone therapy only - to measure CH changes over time.

#ClonalHematopoiesis / #CHIP is associated with worse survival in pts treated for solid cancers, but the evolutionary mechanisms driving differential CH dynamics are unknown. To get at this, we calculated therapeutic bottleneck sizes in pts treated for cancer.

Linking identified variants to tumor sequencing (n > 390k patients), we show that pathogenic CH (#CHIP) mutations are commonly detected across tumor types and associated with 𝘺𝘰𝘶𝘯𝘨𝘦𝘳 age in patients with solid tumors suggesting differential biology.

We identified differences in SM and CH rates and mutations across genetic ancestry groups. We provide a catalogue of identified variants.

We used allele-balance and age skewing to identify SM and CH mutations in population-level data.

So good! Me, 2014.

How times have changed - my children providing a sense of sanity. For an instant, everything else fades away.

When your lab member brings you the [handmade] joy you need to make it through the week. 🤗🧪