For decades, the long-term genomic consequences of surviving cancer treatment have been poorly understood — particularly how chemotherapy, radiation, and immunotherapy reshape the DNA of healthy cells in organs nowhere near the original tumor. This question has profound implications not just for cancer survivors but for how oncologists weigh treatment intensity against future risk in patients with extended life expectancy.
Using ultra-sensitive high-depth duplex sequencing at over 30,000× coverage — a technology capable of detecting mutations present in as few as one in tens of thousands of DNA molecules — researchers analyzed 168 cancer-free tissue samples from 16 distinct organs drawn from 22 patients enrolled in the PEACE autopsy program. Every single sample carried detectable somatic mutations, ranging from roughly 300 to nearly 3,000 per sample. Sixteen distinct mutational signatures were extracted, cataloguing both endogenous and exogenous mutagenic processes. Critically, treatment-associated signatures accounted for more than 40% of all mutations in liver tissue on average, while contributing less than 10% in brain samples — a stark organ-specificity that likely reflects differential drug penetration and local DNA repair capacity. Over a quarter of driver mutations in treatment-exposed normal tissue, including alterations in the tumor suppressor TP53, were attributable to therapy itself. Immunotherapy, notably, was linked to driver mutations in PPM1D and TP53 without elevating overall mutation burden — indicating that some treatments can selectively promote oncogenic selection without being broadly mutagenic.
This work is potentially paradigm-shifting. The prevailing clinical assumption has been that systemic therapies primarily damage cancer cells, with off-target genomic effects in healthy tissue remaining biologically negligible. These findings challenge that view directly. The identification of tissue-specific positive selection — PTEN and PIK3CA variants enriched in lung, BRAF and NOTCH2 in spleen — suggests normal tissue in treated patients is actively evolving toward oncogenic states. A key limitation is the sample size: 22 patients limits statistical power and generalizability across cancer types, treatment regimens, and genetic backgrounds. Nevertheless, as an analytical framework, this study establishes that cancer therapy is itself a meaningful driver of somatic evolution in normal organs, with implications for secondary malignancy risk monitoring and the long-term health management of cancer survivors.