onc brain

International expert consensus exercise under IROCK. 16 radiation oncologists independently contoured 4 clinical scenarios plus 2 investigational renal substructures, with contours analyzed by STAPLE (95% CI) and statements refined over 2 consensus meetings.

Target scenarios: large (>10 cm) tumor with IVC tumor thrombus, central tumor abutting the renal hilum, post-nephrectomy local recurrence, and ablation cavity recurrence after RFA. Consensus: any tumor thrombus or ablation cavity is included in the target volume, bland thrombus is excluded, surgical clips can be excluded when uninvolved, and the iGTV must account for internal motion.

Contour agreement metrics rather than clinical outcomes: dice similarity coefficient (DSC), Mean Distance to Agreement (MDA), and Hausdorff Distance (HD) per case.

Median DSC 0.85, median MDA 2.17 mm, median HD 9.00 mm. Median DSC was >0.70 for every case, the paper's bar for good agreement; 2-way ANOVA showed DSC and HD differed by case but not by participant.

Agreement was weakest exactly where the guidance is most needed: Case 1 (superior extent of IVC thrombus) and Case 4 (ablation cavity) carried the largest MDA/HD spread. The renal cortex and hilum are labeled investigational, since neither substructure has been separately correlated with post-SABR renal function, so contouring them does not yet come with a constraint to meet.

The instruction to let OAR constraints outrank target coverage is the planning-philosophy statement here, and it differs from the coverage-first habit of most SABR sites. Combined with expert radiology review to separate tumor from bland thrombus, the guideline is mostly about avoiding the two errors that inflate a renal target: over-including thrombus and over-including cavity.

Expert consensus contouring guideline, no outcome data. Standardizes an emerging practice rather than testing it; agreement metrics are process surrogates.

Prospective observational cohort inside the Göteborg-1 PSA screening trial. Of 1052 screen-detected cancers diagnosed 1995 to 2014, 494 (47%) had AS as primary strategy; 488 low- or intermediate-risk men analysed. Kaplan-Meier estimates for treatment-free, failure-free, PC-specific, and overall survival.

Very low risk (T1c, Gleason 6, PSA density <0.15, <3 positive cores) 251, low risk 129, intermediate risk (T1-2, Gleason ≤7, PSA <20 ng/ml) 108. Median age 66 yr, median PSA 4.1 ng/ml. High-risk (Gleason ≥8) excluded.

PSA every 6-12 mo, repeat biopsy only on signs of clinical progression (PSA and/or stage); early rebiopsy if <2 mm cancer on the diagnostic core. Treatment recommended on PSA, grade, or stage progression, or by patient preference.

RT was the discontinuation route for 44 of 232 men leaving AS, versus 141 prostatectomy and 47 hormonal. Post-treatment PSA relapse was defined as ≥0.2 ng/ml after RP or nadir +2 ng/ml after radiation.

Treatment-free, failure-free, PC-specific, and overall survival from diagnosis. Failure counted PC death, metastasis, hormonal treatment, or PSA relapse after curative treatment. Curves truncated at 22 yr, and 24 yr / 19 yr in risk-stratified analyses.

PC-specific survival 94% at 25 yr; 14 PC deaths against 264 all-cause deaths. Treatment-free survival 38% and failure-free survival 68% at 22 yr; overall survival 32% at 25 yr.

No predefined surveillance protocol, so biopsy timing varied with individual urologist practice and failure detection is not standardised. Risk assignment used sextant biopsies before 2009 with prediagnostic MRI in only 21 men, so some intermediate-risk failures may be understaged high-risk disease enrolled at baseline.

Failure accrued without a plateau: no safe window for stopping surveillance emerges across 22 yr. With OS 32% at 25 yr against 94% PC-specific survival, competing mortality, not prostate cancer, decides most of these men's outcome.

Prospective single-strategy cohort, no randomised comparator to immediate treatment; risk stratification predates MRI and 12-core biopsy, limiting transfer to current AS entry criteria.