IROCK RCC SABR contouring guidelines
ForLocalized RCC treated with SABR, incl. IVC thrombus, recurrence, post-RFA
TL;DRFirst international consensus on target delineation for localized RCC SABR: 16 experts, median DSC 0.85 vs consensus, >0.70 in every case.
The two contested target calls are settled: include the entire tumor thrombus and the entire RFA cavity, and these were also the cases with most variation (DSC 0.85 and 0.75; HD 64.60 mm in Case 1). The 5 mm PTV assumes 4DCT, vacuum cushion, IV contrast and daily CBCT, so a center without those shouldn't copy the margin.
In a medically inoperable localized RCC with IVC tumor thrombus or recurrence after RFA, this supports covering the full thrombus or full ablation cavity in the iGTV; it does not settle the dose or fraction number, which the panel left institution-specific.
The two contested target calls are settled: include the entire tumor thrombus and the entire RFA cavity, the cases with most variation (DSC 0.85 and 0.75). The 5 mm PTV assumes 4DCT, vacuum cushion, IV contrast and daily CBCT, so it does not transfer unchanged to other workflows.
10 details
Contouring consensus: 16 IROCK radiation oncologists contoured 4 CT-only cases; a STAPLE algorithm produced the 95% consensus contour, refined over 2 online meetings (May and June 2024) until 100% agreement on statements.
Panel eligibility: ≥10 RCC SABR cases treated; 14 of 16 had treated ≥10 pts in the prior 12 months. Cases: >10 cm tumor with IVC thrombus, central tumor at the hilum, post-nephrectomy recurrence, post-RFA residual.
iGTV equals ITV since no microscopic-spread margin is added. Most common PTV: uniform 5 mm on iGTV with supine vacuum cushion, 4DCT, IV contrast and daily CBCT. Participants used 1, 3, or 5 fractions; OAR constraints drawn from FASTRACK-II and AQuOS-II protocols.
Overall median DSC 0.85 (range 0.40-0.95), MDA 2.17 mm, HD 9.00 mm; median DSC >0.70 in every case. Greatest variation in Case 1 (superior thrombus extent) and Case 4 (cavity inclusion).
| Case | Scenario | iGTV recommendation |
|---|---|---|
| 1 | Tumor >10 cm with IVC tumor thrombus | Primary tumor + entire tumor thrombus; exclude bland thrombus |
| 2 | Central tumor abutting renal hilum | All visible tumor, account for internal motion; may extend into hilum |
| 3 | Local recurrence post-nephrectomy | All visible gross tumor; surgical clips need not be included |
| 4 | Residual tumor post-RFA | Visible gross recurrence + entire ablation cavity |
| Case | DSC | MDA (mm) | HD (mm) |
|---|---|---|---|
| 1 | 0.85, 0.79-0.85 | 6.69, 5.88-8.79 | 64.60, 64.44-86.40 |
| 2 | 0.90, 0.84-0.93 | 1.55, 1.00-2.09 | 7.92, 5.35-9.98 |
| 3 | 0.91, 0.88-0.93 | 1.42, 1.18-1.89 | 6.18, 6.00-7.12 |
| 4 | 0.75, 0.60-0.79 | 2.50, 2.12-2.99 | 9.00, 9.00-12.39 |
FASTRACK-II established prospective phase II support for RCC SABR; this document standardizes the target and OAR definitions that trial-era protocols left variable. AQuOS-II is prospectively testing whether ipsilateral parenchyma and hilum contours matter for toxicity.
CT-only images; MRI might have reduced variation. Participants did not recontour after consensus, so the post-consensus agreement is assumed rather than measured.
The guideline chooses OAR priority over coverage and accepts heterogeneous target dose, which trades some target coverage for bowel safety. Whether hilum or cortex sparing reduces renal toxicity remains unanswered and is deliberately left optional.
Expert contouring consensus; agreement metrics measure contour overlap, not outcome. Standardizes target definition for an emerging indication; not validated prospectively.
- Does renal hilum or cortex sparing reduce renal toxicity?
- Patterns of failure with the 5 mm PTV margin
- Does MRI-based contouring reduce target variation?
📚 Sources · 📄 1 paper
The longer read
The value of this paper sits less in its agreement statistics than in the decisions it makes on the scenarios where RCC SABR practice has been most improvised. Tumor thrombus, ablation cavities, post-nephrectomy recurrences near bowel, and central tumors at the hilum are exactly the cases a center starting a kidney SABR program is likely to see, because they are the patients surgeons and interventional radiologists hand off. Before this, a radiation oncologist faced with an IVC thrombus or a failed RFA had protocol documents from trials like FASTRACK-II and AQuOS-II for dose and constraints, but no shared rule for where the target ends.
The agreement metrics should be read carefully. A DSC is overlap with a consensus that was itself built from the same participants' contours through STAPLE weighting, so high agreement is partly circular. The more informative numbers are the outliers: Case 1 carried an HD of 64.60 mm and Case 4 a DSC of 0.75, and these are precisely the two cases where the panel had to make a policy decision (how far up the thrombus, how much cavity). In other words, experienced contourers agree on tumor edges; they disagree on intent. The guideline resolves intent by choosing inclusion, the whole thrombus and the whole cavity, on the rationale that an in-cavity recurrence outside the field would be hard to retreat. That is a reasonable judgment, but it is a judgment, and it will enlarge targets in exactly the cases where OARs are hardest to spare.
The panel's answer to that tension is explicit: OAR constraints outrank target coverage, with heterogeneous dose and the hotspot inside the iGTV. This mirrors how SABR is delivered for pancreas and other upper abdominal targets and is probably the most transferable statement in the paper. It also means that reported coverage objectives are aspirational in the difficult cases, which matters for any future trial using these contours as a quality benchmark.
The renal substructure question is handled with appropriate restraint. The panel declined to make the hilum a dose-limiting structure, noting no evidence that sparing it reduces renal artery stenosis or ureteric stricture, and warned that pushing dose away from it may raise dose to functioning parenchyma. That is the right default until AQuOS-II reports.
What would make this wrong? Prospective patterns-of-failure data showing marginal recurrences with the 5 mm PTV, or toxicity signals from the larger thrombus and cavity targets. The margin also rests on a specific workflow (vacuum immobilization, 4DCT, contrast, daily CBCT); centers without that chain should not import the number unchanged. The guideline defines geometry, not dose, and the fractionation question remains institution-specific.