Background

Human normal tissues serve as a stage for evolution, where countless somatic mutation clones become intermingled like a mosaic as we age. In esophageal epithelial tissue, even in healthy individuals, cells harboring mutations in NOTCH1 or the tumor suppressor gene TP53 form clusters and compete for space due to aging, alcohol consumption, and smoking. The problem is that it remains unclear how neoadjuvant chemoradiotherapy, used to reduce tumor size before surgery in esophageal cancer patients, affects surrounding normal epithelial cells.

Existing research has primarily focused on drug resistance in cancer cells themselves or genetic mutations in blood stem cells. While the surge of specific clones surviving treatment stress in blood has been identified, it remains unclear whether similar genetic restructuring occurs in solid organ epithelial tissues. As many patients undergo toxic treatments, tracking the evolutionary pathways of normal mucosal cells is considered a key to evaluating tissue regeneration and the potential risk of secondary cancers.

Key Findings

A research team led by Professor Philip Jones at the University of Cambridge and the Wellcome Sanger Institute provided an answer by precisely analyzing normal esophageal tissue from 70 patients who underwent esophageal cancer surgery. The researchers used 21 untreated patients as a control group and closely examined 7 patients in the 5-fluorouracil (5-FU)-based FLOT treatment group, 31 patients in the platinum-based combination therapy group (ECX and EOX), and 11 patients in the chemoradiotherapy group (CROSS). The method precisely tracked 23,692 somatic mutations by performing targeted deep sequencing of 324 cancer-related genes at an average depth of over 500x across 1,995 grid tissues, each measuring 2 square millimeters (mm²).

The analysis revealed that chemotherapy did not generate a large number of new mutations. Mutation density and single nucleotide substitution (SBS) signatures were similar regardless of treatment status. This indicates that the treatment strongly induced Darwinian selection among mutation clones already present in normal tissue, rather than creating new variants.

The patterns of clonal selection also showed distinct differences depending on the treatment method. In patients receiving CROSS therapy combined with radiation, the area of TP53 and PPM1D mutation clones was approximately twofold larger than in the untreated group. More than 90% of PPM1D mutations are concentrated in exon 6, a pattern that suppresses p53 signaling and supports cell survival amidst radiation damage. In mouse esophageal organoid experiments, cells with this mutation also demonstrated a clear survival advantage in a radiation-exposed environment.

In contrast, in the FLOT chemotherapy group, mutations in the antioxidant regulator NFE2L2, the stem cell factor RAC1, and the cell growth driver MTOR were selected. Protein structure analysis revealed that NFE2L2 mutations induce alterations in the KEAP1 binding site, thereby neutralizing 5-FU toxicity, while RAC1 and MTOR mutations are interpreted as mechanisms promoting cell proliferation. In essence, only normal cells capable of withstanding drug stress survived selectively.

Significance and Outlook

This study is the first to identify that anticancer treatment reshapes the clonal composition of normal epithelium, allowing drug-resistant cells to flourish. The surviving mutant clones act as a shield to immediately repair mucosal damage and prevent tissue collapse. However, there is also a concern that as cells accumulating cancer driver mutations widely occupy the normal mucosa, the probability of secondary cancer development may increase in the long term.

Clues for drug development and biomarker discovery were also revealed. Structural modeling confirmed that the MTOR variant, which emerged following FLOT therapy, maintained its binding affinity for rapamycin without developing target resistance, thereby preserving drug responsiveness. The analysis suggests that by exploiting the defense strategies of clones surviving in normal tissue, it is possible to identify combination targets to overcome treatment resistance in cancer cells.

However, a limitation is that the study sample size was 70, with a limited number of patients in specific treatment groups. The researchers explained that large-scale long-term follow-up studies are needed to determine whether these mutation clones actually trigger long-term esophageal dysfunction or malignant tumors.