Oncology · Tumour heterogeneity
In ovarian cancer, the clone that came back was already present before chemotherapy
Save this development or follow its topic to track what changes.
Short answer
Higher clonal expansion predicted shorter survival in a 14-patient cohort, and the clone associated with relapse was detectable in pre-treatment samples.
What happened
Schwarz and colleagues applied MEDICC to whole-genome copy-number profiles from spatially and temporally separated samples in 14 patients with high-grade serous ovarian cancer on platinum chemotherapy, with median follow-up of 31 months. Splitting clonal expansion at the median (0.74), the high group had progression-free survival of 10.1 versus 12.7 months (p=0.009) and overall survival of 23.5 versus 42.6 months (p=0.003). Digital PCR found the relapse-associated NF1 deletion already present before treatment.
Why it matters
If the resistant clone predates therapy rather than being created by it, then relapse is partly a sampling and selection problem, not only an induced-mutation problem. That changes what a single diagnostic biopsy can be expected to tell you.
Evidence
- Study type
- Phylogenetic copy-number analysis of multi-site tumour samples with survival follow-up
- Sample
- 135 samples from 14 patients (177 arrays from 18 profiled); median follow-up 31 months
- Journal
- PLoS Medicine · peer reviewed
- Replication
- A separate head-and-neck cancer analysis in this corpus reports the same direction for intra-tumour heterogeneity
- Limitations
- Fourteen patients cannot size the hazard. Clonal expansion was significant only when dichotomised at the median, not as a continuous variable, which is a warning sign in a small cohort.
What this connects to
Sources
The 2 studies this explanation is built from, by the role each plays. Every source links to PaperFren’s explanation of it and to the original paper.
Primary study
- Does ovarian-cancer clone diversity predict death?
In 14 women with high-grade serous ovarian cancer, high clonal expansion predicted worse PFS (10.1 vs 12.7 mo) and OS (23.5 vs 42.6 mo); an NF1-deleted relapse clone was already detectable before chemotherapy.
What it does not showLimitations
n=14 cannot precisely size the hazard; CE as a continuous variable was not significant in multivariable models, so the median split is a small-cohort signal, not a ready clinical test.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Supporting evidence
- Does mixed DNA inside a head-and-neck tumor predict death?
In 305 TCGA HNSCC patients, high intra-tumor MATH heterogeneity more than doubled the hazard of death (HR 2.2; 95% CI 1.4–3.3), even after HPV, TP53, grade, and N classification.
What it does not showLimitations
Retrospective multi-site TCGA data were not collected to study MATH; prospective, homogeneously treated series (especially HPV+ oropharynx) are still required before clinical use.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Before
Platinum resistance in high-grade serous ovarian cancer was commonly framed as acquired under treatment pressure, with single-site biopsies treated as representative of the tumour.
Now
Multi-site sampling shows substantial spatial heterogeneity and a pre-existing resistant clone in this cohort. With 14 patients and significance only on a median split, this is a signal worth pursuing, not a clinical test.