ABS compendium of gynecologic HDR brachytherapy fractionation
TL;DRCervix HR-CTV D90 goal 85-95 Gy EQD210; no single optimal HDR scheme for any site; doses >7 Gy/fx flagged for higher complication risk.
Portec-2EMBRACE
The EMBRACE-derived OAR limits sit below the ABS ones (rectal D2cc ≤65 vs 70-75 Gy, bladder ≤80 vs about 80-90 Gy EQD23), and the panel says lower doses are recommended in practice. For cuff brachy the panel picks no scheme. It flags that 0.5 cm-depth prescriptions need modifying on a 2.5 cm cylinder to keep surface dose acceptable.
The EMBRACE-derived OAR limits sit below the ABS ones (rectal D2cc ≤65 vs 70-75 Gy EQD23), and the panel endorses the lower values. For the cuff it names no preferred scheme and flags that depth prescriptions need adjusting on a 2.5 cm cylinder.
12 details 1 trial watching
ABS Gynecologic Cancer Task Force compendium (Brachytherapy 2019). Experts reviewed the literature, the existing ABS guidelines and their own clinical experience, then reached consensus over conference calls. Not a systematic review.
All schemes are given as EQD210 for tumor and EQD23 for OAR, so regimens can be summed and compared. Cervix image-guided target: HR-CTV D90 85-95 Gy EQD210, starting from the Point A schedules and then optimized. Interstitial needle loading should hold no more than 10-20% of total dwell time.
The cervix target range and the tighter OAR limits (rectal D2cc ≤65 Gy, bladder ≤80 Gy) come from EMBRACE, not the older ABS 2012 Point A guidance. For the cuff, Portec-2 and institutional series supply the schedules, and Sorbe is the only randomized comparison of fractionation.
Several of the source tables could not be reliably paired row-to-value here, so specific scheme-to-EQD2 mappings for Point A, interstitial and inoperable uterine schedules are not reproduced. The cervix Point A schedules derive from 2012 ABS guidance and are pre-optimization starting points.
The document is a reference list of what is acceptable, not a ranking. Its practical value is the EQD2 bookkeeping and the explicit caution on >7 Gy/fx. It leaves open which cuff scheme is best and how to load newer free-trajectory interstitial applicators.
| Setting | Target | Dose goal |
|---|---|---|
| Intact cervix, image-guided | HR-CTV D90 | 85-95 Gy (90-95 Gy for larger tumors) |
| Inoperable stage I uterine, HDR alone | Uterine CTV / GTV | 48-62.5 Gy / 80-90 Gy |
| Inoperable stage I uterine, EBRT + HDR | Uterine CTV / GTV | 65-75 Gy / 80-90 Gy |
| Post-op vaginal cuff boost after EBRT | Vaginal surface (LDR equivalent) | 65-70 Gy; 70-75 Gy for superficial positive margins |
| Vaginal recurrence of uterine cancer | Vaginal lesion | at least 75 Gy EQD2 (LDR equivalent) |
| OAR | ABS limit | EMBRACE-derived |
|---|---|---|
| Rectum/sigmoid D2cc | 70-75 Gy | ≤65 Gy (rectum) |
| Bladder D2cc | approximately 80-90 Gy | ≤80 Gy |
| Rank | Regimen | Prescription point |
|---|---|---|
| 1 | 7 Gy x 3 | 0.5 cm depth |
| 2 | 6 Gy x 5 | Vaginal surface |
| 3 | 5.5 Gy x 4 and 5 Gy x 5 | 0.5 cm depth |
| 4 | 7.5 Gy x 5 | Vaginal surface |
| EBRT | HDR boost |
|---|---|
| 45 Gy/25 fx | 5-6 Gy x 3 (to surface) |
| 50.4 Gy/28 fx | 6 Gy x 2 (to surface) |
- Optimal vaginal cuff HDR fractionation scheme
- Outcomes of 3-fraction vs 4-5 fraction cervix HDR n=41 · primary completion 2028-06 · phase 2 3-fx vs standard 4-fx HDR, definitive cervix
- Loading patterns for newer interstitial applicators
📚 Sources · 📄 1 paper
The longer read
The most useful way to read this compendium is as a statement of tolerance rather than of optimality. The panel says plainly that no single fractionation scheme is optimal for any gynecologic site. What it actually does is fence the acceptable space: a set of schedules that converge on similar EQD2 values, plus a warning that pushing dose per fraction above 7 Gy buys convenience at the cost of complication risk unless implant geometry and OAR doses are favorable. For a brachytherapist choosing between a three-fraction and a five-fraction cervix course, that framing is more honest than a single recommended regimen would be, because the comparative evidence is thin.
The interesting tension sits between the two generations of dose limits the document carries side by side. The ABS constraints of 70-75 Gy EQD23 for rectum and sigmoid D2cc and roughly 80-90 Gy for bladder are presented next to the lower EMBRACE-derived thresholds (rectal ≤65 Gy, bladder ≤80 Gy), and the authors concede that lower OAR doses are recommended in practice on the strength of the European data. Practically, a planner who treats the ABS numbers as the ceiling is working to an older standard than the panel itself endorses. The same applies to target dose. The 85-95 Gy HR-CTV D90 goal is an image-guided, EMBRACE-anchored target, while the Point A schedules are explicitly labeled pre-optimization starting points. Reaching the upper end of that range for bulky disease required hybrid intracavitary and interstitial technique in about 45% of the Vienna experience, which is a resource statement as much as a dosimetric one.
For the post-hysterectomy cuff, the evidence base is weaker than the table of schedules might suggest. The only randomized fractionation comparison cited (Sorbe, 290 pts, 2.5 Gy x 6 vs 5.0 Gy x 6 to 0.5 cm) enrolled low-risk stage IA grade 1-2 disease, a group in which most current practice would not offer brachytherapy at all, and it found equal outcomes with more vaginal shortening in the higher-dose arm. Everything else rests on institutional series and survey frequency. The most common US scheme, 7 Gy x 3 at 0.5 cm, is therefore a consensus of habit. The point about cylinder diameter is the one most likely to change a plan: a depth prescription that is reasonable on a 3 cm cylinder delivers a materially higher surface dose on a 2.5 cm cylinder, so the prescription should follow the applicator rather than the protocol name.
What would make this document wrong is mainly the passage of time. It was written before newer prospective data on reduced-fraction cervix regimens and modern interstitial applicators matured, and the authors themselves say loading patterns for free-trajectory needles are undefined. The international comparison favoring 4 x 7 Gy over 2 x 9 Gy for local control is a reminder that isoeffective EQD2 arithmetic does not guarantee isoeffective outcomes at the extremes of fraction size. Read it as a floor for safe practice and as an EQD2 lookup, and defer to image-guided dose-volume data where the two disagree.