Science1 publisher2 min readPublished
Twenty rapid autopsies show bladder cancer metastases mostly seeding other metastases
A University of Washington and Fred Hutch autopsy programme sequenced tumours from 20 people who died of bladder cancer to reconstruct how the disease spread. Plasmacytoid tumours turned out to limit the genomic damage chemotherapy leaves behind.
The Scientist · Science desk
What happened
- Rapid autopsies of 20 patients with carcinomas of the urothelium, performed a median of 3.7 hours after death, yielded 80 metastases, eight primary tumours and 20 benign tissue samples.
- Whole-genome sequencing of multiple lesions per patient let the team reconstruct each tumour's family tree, and metastasis-to-metastasis seeding was the dominant pattern of spread.
- Plasmacytoid and neuroendocrine variants developed their driver alterations early in the disease, and those early alterations were associated with shorter survival.
- Immune profiling separated the subtypes, with plasmacytoid tumours showing immune-inflamed profiles while squamous tumours were predominantly immunosuppressive.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint If deposits seed further deposits, clearing or irradiating one site leaves a network that can still seed, which limits what any single-lesion strategy can be expected to do in late disease.
- decision Plasmacytoid disease now has a named resistance mechanism a trial can be designed around, and that is a different kind of claim from a prognosis: it has to survive a randomised comparison before it reorders anyone's therapy.
- capability If the same subtype signal reads out of blood in living patients, identifying an aggressive variant would no longer require tissue from a metastatic site.
- constraint The programme deliberately over-sampled rare subtypes among people already dying of the disease, so the survival patterns in these 20 cases cannot be read as rates for the wider metastatic population.
Polyclonal migration is the result I would build the next study around. It means more than one clone travelling together to found a new deposit, and the authors report that more of it went with worse prognosis [7]. Set beside the seeding pattern, that describes established deposits founding further deposits, sometimes several clones at a time [6].
The cohort is small and skewed on purpose. About one in four patients with metastatic bladder cancer have tumours with squamous, plasmacytoid, neuroendocrine or sarcomatoid features [3], and this programme was built to enrich for those [4]. Eighty metastases from twenty patients is four per patient on average [15]. The tissue tally lists eight primary tumours against twenty patients [16], while the paper describes whole-genome sequencing of primary and multiple metastatic tumours from each patient [5]; the excerpt does not reconcile the two counts.
The plasmacytoid finding is the one with a target attached. Mutational signature analysis and experimental models showed those tumours using the Fanconi anaemia pathway to mitigate the genomic scarring chemotherapy causes [10]. That is a mechanism, shown in signatures and in laboratory models. Whether it should change the order in which a plasmacytoid patient receives platinum chemotherapy is a clinical question, and a trial in living patients is what settles it.
The authors set out the ambition themselves. Charting the evolutionary history of metastatic tumours, they write, "can uncover the emergence of treatment-resistant clones and identify optimal points for intervention" [18].
Histological variation was already prognostic in metastatic urothelial cancer, and its evolutionary trajectory was poorly defined [14]. The sequencing adds when things happen: the burden, heterogeneity and timing of genomic alterations differed markedly across subtypes [8]. Single-nucleus profiling found mixed cell states inside individual subtypes, and transcriptional heterogeneity was associated with survival [11]. All twenty patients had terminal disease when they entered the programme [1], so every survival comparison here is drawn among people the disease went on to kill.
What to watch
- Whether a trial opens that tests interference with the Fanconi anaemia pathway in plasmacytoid disease alongside or before platinum chemotherapy.
- Whether the subtype signal read out of post-mortem cell-free DNA reproduces in blood drawn from living patients.
- Whether the polyclonal migration and transcriptional heterogeneity survival associations hold in a cohort not selected for terminal disease.