Loading
Every tumour is a population of genetically distinct clones: in multi-region sequencing of kidney tumours, roughly two thirds of mutations were missing from at least one region. Treatment kills the dominant clones and leaves resistant minor clones to grow back, yet a single diagnostic biopsy is still treated as the whole disease.
Every cancer is a population of clones that differ by lesion, by region within a lesion, and over time. Multi-region sequencing of renal and lung tumours shows that a majority of somatic mutations are not shared by every region, so a single biopsy at diagnosis systematically under-samples the disease it is used to treat. Therapy then acts as a selection pressure: pre-existing minor subclones carrying resistance alleles expand, and new lesions can be driven by different clones from the primary. Clinical practice still assumes one genotype per patient, re-biopsies at progression are uncommon, and few trials adapt treatment to the clone that is actually growing. Longitudinal circulating tumour DNA, single-cell and spatial profiling, and evolutionary trial designs are the tools that could turn heterogeneity from an excuse into a target.
Blood tests can already detect tumour DNA. Reporting which sub-populations of the tumour are growing or shrinking, cycle by cycle, would turn the test into an evolution monitor.
Instead of hitting a tumour with the maximum dose until it stops working, adjust the dose to keep the tumour small and let drug-sensitive cells suppress resistant ones. Test this properly across several cancers.
When patients who agreed in advance die of cancer, sampling every tumour within hours reveals how the disease evolved and escaped every drug. Few hospitals can do this today.
Some tumours change fast and escape drugs quickly; others are stable. A single validated score for how evolvable a tumour is would tell doctors how aggressively to combine treatments.
When a tumour evolves resistance to one drug, it sometimes becomes weaker against another. Map these trade-offs systematically so doctors can pick the next drug to exploit them.
Map every state a cancer cell can be in, and how drugs and the surrounding tissue move it between states, into an open computational model anyone can query and improve.
Hospitals usually keep one piece of a removed tumour. Keeping three pieces from different parts would show how varied the tumour is, at almost no extra cost.
Tag every cell in a patient's lab-grown tumour with a unique DNA label, give it a drug, and read the labels to see which cells survive. This predicts which resistant clone will emerge.
When a scan shows most tumours shrinking but one growing, that odd lesion holds the escape mechanism. Sampling it, and treating it locally, should be routine.
Use tumour DNA in the blood as the signal to pause and restart a lung cancer pill, keeping the tumour in check while slowing the rise of resistant cells.
Choose two treatments so that whatever the tumour does to escape the first, it becomes easier to kill with the second. The immune system is a good candidate partner.
Some cancers escape treatment by changing into a different kind of cell that the drug no longer affects. Tumour RNA in blood could show this shift months before a biopsy would.
Build a computer model of each patient's cancer that forecasts how it will respond to each treatment option, and prove it by writing the forecast down before the real result is known.
Cancer is an evolving population, but treatment decisions are rarely made with an evolutionary biologist present. Add one to the weekly meeting and see whether decisions change.
Cells that receive only a small amount of a drug survive and adapt. Measuring where inside a tumour the drug actually reaches would show where resistance is being bred.
Flu vaccines are chosen by predicting which virus strains will dominate next season. The same forecasting maths could predict which resistance mutation a patient's tumour will develop next.
Rare cancers often share a broken cellular machine even when they arise in different organs. Grouping patients by that shared fault makes trials possible.
A drug aimed at a mutation present in every tumour cell works differently from one aimed at a mutation in only some cells. Test reports should say which is which.
Resistance mutations often exist in a tiny fraction of cells before treatment starts. Error-corrected sequencing that detects variants below 0.01 percent allele fraction could find them at diagnosis and prompt a mechanism-matched combination from day one.
A drug that shrinks a tumour by half but leaves the resistant sub-population untouched will fail. Trials should measure whether every sub-population is cleared, not just overall size.
When a targeted drug stops working, the tumour has usually changed in a way you can read. Most patients still move to the next treatment on a protocol rather than on a test of what actually happened.
New imaging shows where every cell type sits in a tumour slice. Using it to see which sub-populations are hidden from immune cells could explain why immunotherapy fails in parts of a tumour.
Blood tests tell you which tumour sub-populations are growing; scans tell you which lesions are growing. Joining the two would tell you where to biopsy or irradiate.
Resistant cancer cells can become dependent on the drug they resisted, as shown for BRAF-inhibitor-resistant melanoma in mice. Stopping the drug for a defined washout and then rechallenging, while tracking the resistance allele in blood tumour DNA, could make the tumour vulnerable to it once more.
Several big projects have sequenced the same tumours at different times and places, but their data sit apart. Bringing them together with common analysis would show general rules of how cancers evolve.
Antibody drugs need their target to still be present. After one fails, checking which surface markers remain would guide the choice of the next one instead of guessing.
Treatment is often chosen from a biopsy taken years earlier from the original tumour. The spread disease may now look different. Test it again before switching drugs.
Subsets of lung, bladder and breast cancers carry raised APOBEC enzyme activity that keeps generating new mutations, feeding resistance. Blocking APOBEC3 alongside a targeted drug aims not to kill cells but to slow the rate at which resistant variants arise; the inhibitors are still in discovery.
Species go extinct when a second disaster hits a population already shrunk by a first one. Apply the same logic: hit the tumour with a different kind of drug when it is smallest, rather than waiting for it to grow back.
Tumour DNA in blood can be told apart by chemical marks as well as mutations. Marks are more numerous and cheaper to read, so they could track more sub-populations for less money.
Chromosomally unstable, often whole-genome-doubled tumours survive constant chromosome mistakes by depending on the motor protein KIF18A, which diploid cells do not need. Blocking it kills unstable cancer cells while sparing normal ones; inhibitors are in early trials in ovarian and other cancers.
If a drug relies on one marker, the tumour can survive by dropping it. A drug that recognises two markers at once makes that escape harder.
Brain tumours release little DNA into blood because of the blood-brain barrier. Sonobiopsy briefly opens the barrier with focused ultrasound and microbubbles, raising circulating tumour DNA severalfold so a blood sample can replace repeat surgical biopsy for diagnosis and resistance monitoring in glioma and brain metastases.
Personal cancer vaccines target a list of mutations, some present in only part of the tumour, so the tumour can escape by losing them. Restricting vaccines and T-cell products to clonal mutations shared by every tumour cell, identified by multi-region sequencing, should close that escape route.
In a minority of men, prostate cancer escapes hormone drugs by becoming a different kind of cell that no longer needs the androgen receptor. By the time a biopsy shows it, the treatment options are almost gone. The genetic changes that allow the switch are detectable years earlier, and nobody is looking for them.
The latest in a line of negative targeted-therapy trials in triple-negative disease (EGFR, VEGF, iniparib, now AKT): a modest delay in progression that did not translate into survival, in the same year that an antibody-drug conjugate did. It is why the roadmap treats pathway-targeted small molecules as the road not taken.
Relapse after surgery is driven by particular subclones that can be identified in the primary tumour and tracked in blood, which argues for evolution-aware adjuvant strategies. The pollution finding reframes carcinogenesis: some agents promote already-mutant cells rather than causing mutations.
Lung cancer in never-smokers is not smokers' lung cancer with the smoking removed; it is a different set of diseases with a different clock. The slow-growing piano subtype in particular is the argument that a screening test aimed at never-smokers would need to look for something other than what low-dose computed tomography was built to find.
The genetic nosology that precision-medicine trials in diffuse large B-cell lymphoma now use to pick patients. It gives a mechanism, not just a label: two of the four subtypes point at a drug class that already exists.
One of the two foundational genetic classifications of diffuse large B-cell lymphoma. Neither has yet changed what a patient receives outside a trial, but together they are the reason precision-medicine trials in this disease now select by genetics rather than by cell of origin.
Cancers are defined as much by the tissue they come from as by the mutations they carry, which is why the same drug can work in one organ and fail in another with the same mutation. TCGA is the shared public dataset behind most modern biomarkers and target discovery.
One bottleneck page on OnCo cites this paper by its DOI; this record gives the citation a page of its own so a reader can follow it without leaving OnCo. Read the abstract above alongside the citing page listed under Related; the record was created automatically from the Europe PMC entry and its figures have not been checked by hand.
The foundation of minimal residual disease testing in lung cancer: a blood test that says a patient will relapse months before a scan does, and says which part of the tumour is doing it. Whether acting on that signal changes outcome is what the ctDNA-guided trials are for.
Shares Switch drugs at maximum response, not at relapse, An open atlas of collateral sensitivity for every approved targeted drug, Find the parts of a tumour the drug never reaches, Two-target antibody drugs to close the antigen escape route.
Shares Intra-tumour heterogeneity, Gerlinger: a single biopsy misses most of the mutations in a kidney tumour, Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution, TRACERx 421: the full-cohort picture of how lung cancer evolves and which subclones drive relapse.
Shares Label every targetable mutation as truncal or branch on the report, Punctuated evolution of prostate cancer genomes, Chromoplexy, Make resistance a diagnosis: sequence at every progression and choose the next line from what the tumour became.
Shares Cancer Grand Challenges, Gerlinger: a single biopsy misses most of the mutations in a kidney tumour, Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution, TRACERx 421: the full-cohort picture of how lung cancer evolves and which subclones drive relapse.
Shares TRACERx 421: the full-cohort picture of how lung cancer evolves and which subclones drive relapse, Integrative genomic profiling of human prostate cancer, Genomic and evolutionary classification of lung cancer in never smokers, TRACERx first 100: tracking how lung cancers evolve, and how chromosomal chaos predicts relapse.
Shares Match each blood-detected clone to the lesion it comes from on the scan, A national rapid research autopsy network for end-stage cancer, Oligoprogression, The evolutionary history of lethal metastatic prostate cancer.
Shares Track clones in blood with methylation patterns instead of mutations, Make resistance a diagnosis: sequence at every progression and choose the next line from what the tumour became, Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution, Natera.
Shares A standard evolvability score for every tumour, Label every targetable mutation as truncal or branch on the report, The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma, Gerlinger: a single biopsy misses most of the mutations in a kidney tumour.