Chemotherapy Can Leave a DNA Fingerprint in Childhood Tumours Within Months
An international study of more than 600 childhood tumours identified mutation patterns linked to specific cancer treatments, sometimes within 91 days. The findings may support earlier monitoring of resistant disease, but they do not show that chemotherapy causes every relapse or provide a clinical test yet.
Chemotherapy is designed to damage cancer cells, but surviving cells can carry a record of that encounter in their DNA. An international study has now mapped those records across hundreds of childhood tumours, revealing treatment-linked changes months after therapy began.
The 30-second summary
- What happened? Researchers combined whole-genome sequencing with treatment histories from more than 600 tumours in 544 children and identified mutation patterns associated with several therapies.
- Why does it matter? Those patterns could eventually help doctors detect when surviving cancer is changing under treatment and adjust care earlier.
- What is the catch? The study largely examined tumours that returned, spread or survived treatment. It does not prove that chemotherapy caused every relapse, and there is no routine blood test based on the findings yet.
KEY NUMBER
Some treatment-linked DNA signatures were detectable just 91 days after therapy began.
A tumour can keep a record of treatment
Every cancer cell accumulates mutations. Some appear through ordinary cell division, some come from the biology that created the cancer, and others bear recognisable patterns left by an outside force. Researchers call these patterns mutational signatures.
The new study, published in Nature on July 22, 2026, asked whether therapies leave signatures in childhood tumours and how quickly those marks appear. The team analysed more than 600 samples from 544 patients in Canada, Australia and the United States, linking whole-genome data to drug type, dose and timing.
Tumours sampled after therapy carried nearly three times as many mutations unique to that tumour, and roughly twice as many total acquired mutations, as untreated samples. The researchers connected distinct signatures to four commonly used chemotherapy agents.
Why platinum treatments stood out
Platinum drugs damage DNA so severely that cancer cells struggle to copy it. They are important treatments for several childhood cancers, but the same mechanism can leave a recognisable mutation pattern in cells that survive.
According to SickKids, 48% of evaluable tumours exposed to platinum chemotherapy showed a detectable platinum-related signature within 18 months. The research paper reports the results through several analytical thresholds, so that figure should not be interpreted as the share of all children who will relapse.
The study also found that platinum therapies accounted for a large portion of treatment-related mutations. This does not mean the drugs failed. Most children with cancer now survive, and chemotherapy remains lifesaving. The finding instead reveals how treatment can shape the small population of tumour cells that remains.
From a genomic record to an early warning
Today, clinicians mainly discover treatment resistance through scans, symptoms, biomarkers or visible tumour growth. A genomic signature might offer an earlier clue that surviving cells are adapting under pressure.
Researchers envision a future blood test that searches fragments of tumour DNA for these patterns. If validated, it might help doctors identify a concerning change sooner, intensify a different therapy or reduce exposure when treatment is producing harm without sufficient benefit.
That possibility remains prospective. This study identified signatures in stored tumour genomes; it did not run a clinical trial in which treatment decisions were changed on the basis of a real-time blood result.
What the study does not say
The work may sound alarming because it connects treatment with additional mutations. Yet cancer therapy is always a balance between immediate benefit and later risk. A child with an aggressive tumour may need a DNA-damaging drug because failing to control the cancer is the more urgent danger.
The relevant question is therefore not whether chemotherapy alters DNA, which is central to how many drugs work. It is whether clinicians can use those alterations as information, choosing the right dose, combination and monitoring plan for each patient.
Before we overstate the result
- The dataset was enriched for tumours that survived, relapsed or metastasised, so it cannot estimate risk for every treated child.
- A treatment-associated signature shows exposure and mutation, not proof that the drug alone caused a later cancer event.
- The reported percentages depend on whether a tumour contained enough mutations for a signature to be detected.
- No prospective trial has yet shown that changing therapy in response to these signatures improves survival or reduces long-term harm.
What happens next
The next step is to follow patients over time and test whether signatures in blood reliably predict resistance, relapse or late effects before standard methods do. Any test would also need to show that acting on its result helps patients, not merely that it detects more abnormalities.
The study does not weaken the case for effective chemotherapy. It adds resolution to a difficult picture. By reading the molecular record left in surviving cells, doctors may eventually learn not only whether a treatment attacked the cancer, but also how the cancer changed in response.
Sources and citations4 sources
External references used to support the reporting in this article.
Published by
NewTqnia Editorial
Technology & innovation desk
