onc brain

The longer read

The shape of this evidence base matters more than any single trial in it. Almost everything supporting metastasis-directed therapy in prostate cancer was generated in the metachronous oligorecurrent setting, in trials the speaker herself characterised as small and heterogeneous in inclusion criteria and endpoints. That is why MDT has not converted into guideline language despite years of positive-sounding readouts: the trials were sized to detect progression-free survival differences in populations selected on imaging, and progression-free survival in a PSA-monitored, scan-monitored disease responds to ablation almost mechanically. Removing the lesions you can see delays the next scan-detected event. Whether it changes the disease is a separate question, and it is the one the field has not answered.

Against that background, ARTO is the outlier worth arguing about. It is presented here as the first randomised trial suggesting an overall survival and prostate-cancer-specific survival benefit from adding MDT to systemic therapy, and if it holds it is the only result in the session that would move MDT from a reasonable option to an expected one. Two things temper it. It was run in castration-resistant oligometastatic disease, which is not where most MDT is offered, so the population that gained is not the population most readers are treating. And WOLVERINE, testing the same idea in hormone-sensitive disease, gained progression-free and radiographic progression-free survival and delayed castration resistance without a significant overall survival difference. The two results are not incompatible, but they diverge on the only endpoint that settles the argument, and the conservative read is that one survival signal in one setting is a hypothesis rather than a standard.

The de-escalation strand is the more interesting one for a radiation oncologist and the least mature. SOLAR enrolled 21 patients to ask whether testosterone can be allowed to recover after MDT while PSA stays suppressed, which reframes ablation as a way of buying a systemic-therapy holiday rather than as an intensification on top of it. The comparison drawn against SATURN, with outcomes appearing more favourable in synchronous than metachronous disease, is a cross-trial comparison of separately selected populations and cannot support the inference that earlier MDT works better. It is a reasonable hypothesis, and it is the kind of hypothesis that has previously survived several small series and then failed a randomised test.

Two negatives are worth carrying out of the session. RAVENS found no progression-free or metastasis-free survival gain from adding Ra-223 to MDT, which argues against assuming radioligand therapy and ablation are additive; LUNAR will test the same premise with Lu-177 and should not be expected to succeed simply because the isotope is newer. And the selection criterion itself, a lesion count of up to five, is arbitrary in the way anatomic thresholds usually are. The PSMA PET burden work raised at the end is the more defensible direction: burden measured on the imaging that now defines the disease should stratify better than counting spots on it. Until trials randomise on biology rather than lesion count, MDT will keep generating progression-free survival wins in populations chosen to produce them, and the de novo synchronous question that STAMPEDE2, PLATON, TERPS and OLIGOPRESTO are built to answer will stay the one that actually decides practice.