onc brain

The longer read

The number that matters here is not 0.81 but 0.088, and the honest reading is that PEACE-2's target-volume randomization returned a direction without a conclusion. A hazard ratio whose interval runs from 0.63 to 1.03 is compatible with a moderate benefit of elective pelvic coverage and equally compatible with none, and a 4-point separation in 7-year cPFS is exactly what a trial powered for something larger looks like when the underlying event rate comes in below plan. That is the first thing a reader should take from the plenary slide, and it is different from concluding that pelvic RT does nothing.

What makes the trial interesting is the population, not the intervention. The eligibility criteria select for what the field has called very high risk, at least two of Gleason ≥8, T3-T4, and PSA ≥20, and the investigators' own closing slide reports that fewer than one in ten of these men died of prostate cancer in the first decade. A cohort with that little cancer-specific mortality cannot generate the event volume a target-volume question needs, and any incremental benefit from covering the pelvis has correspondingly little room to appear in a clinical-progression endpoint. The label promised a high-event population and the trial did not get one. That is a statement about how the field defines risk from clinicopathologic factors alone, and it explains a null result at least as well as a claim that nodal irradiation is inert.

Against POP-RT, the comparison is not clean enough to call a contradiction. That trial was single-center with roughly 224 men and reported a whole-pelvis benefit that pushed elective nodal coverage into routine practice for high-risk disease. PEACE-2 is larger and multicenter, which usually means the effect regresses toward the null, but the two trials also differ in the population they assembled and in what fraction of enrolled men actually harbored nodal disease. A reader who took POP-RT as settling the question should downgrade that confidence; a reader who concludes PEACE-2 refutes it is overreading a p-value of 0.088.

The staging issue cuts both ways and deserves more weight than it usually gets. Nodal status was established by conventional imaging or choline PET/CT, both of which miss pelvic nodal disease that PSMA PET would find. Undetected node-positive men in the prostate-only arm are precisely the patients elective coverage is meant to help, so their presence should have biased toward a positive result, not away from it. That the trial still fell short is the more informative version of the finding. The mirror-image problem is what happens going forward: a modern PSMA-staged N0 population is cleaner and therefore has even less occult nodal disease to treat, so the case for routine pelvic coverage in that population is weaker still, which is close to what the investigators said themselves.

What would have to be true for this to be wrong is a real subgroup, defined by something other than the three clinicopathologic factors used here, in which nodal coverage carries a meaningful effect that the whole-cohort analysis dilutes. The investigators pointed at biomarkers as the way to find it. Until then, the practical read for a radiation oncologist is that elective pelvic coverage in conventionally-staged, very-high-risk N0M0 disease is a defensible option rather than an evidence-mandated one, and the toxicity and quality-of-life secondaries, absent from these slides, become the deciding input rather than the efficacy signal.