ESTRO 36 Abstract Book

S198

ESTRO 36

_______________________________________________________________________________________________

Conclusion

At our knowledge this is the first report finding a

correlation between vaginal stenosis, VP volume and

vaginal dose on a relatively large serie. Further studies on

larger dataset are needed to confirm such data

OC-0368 Postoperative vaginal brachytherapy: a

quality assurance dummy-run procedure in the

PORTEC-4 trial

R. Nout

, E. Astreinidou

, M. Laman

, C. Creutzberg

Leiden University Medical Center LUMC, Department of

Radiotherapy, Leiden, The Netherlands

Purpose or Objective

As part of the quality assurance program in the ongoing

randomized multicenter PORTEC-4 trial a ‘dummy-run’

procedure for vaginal brachytherapy was mandatory

before centers could participate. The aim was to evaluate

whether the CT- or MRI based clinical target volume (CTV)

and organ at risk (OAR) delineations and the standard

treatment plans were according to the trial protocol,

especially since for many centres this involved

introduction of CT-based delineation and planning for

vaginal brachytherapy.

Material and Methods

Pelvic CT and MRI scan of a postoperative endometrial

cancer patient with a cylinder in situ were made available

to participating Dutch centers. Centers were asked to use

their own treatment planning and delineation software

and follow the study protocol in order to: 1) delineate CTV

and OAR’s: bladder, rectum, sigmoid and small bowel; 2)

reconstruct the single line source path; 3) create a

treatment plan prescribing 7 Gy at 5 mm from the surface

of the applicator (point A2, fig 1); 4) perform DVH

analysis. The CTV consisted of the proximal 4 cm of the

vagina to a depth of 3 mm from the mucosal surface. To

account for anisotropy in the longitudinal direction of the

source two points (A1 and A3) were defined at 5 mm from

the cranial applicator surface and additional points (A4-6)

were added (fig 1). The average dose between A1 and A3

should be approximately 100%, with A1 minimal 90% and

A3 maximal 110%. Central evaluation of contours and

treatment plans took place and in case of deviations from

the protocol, feedback was provided and necessary steps

in the dummy-run repeated.

Results

Fifteen centers participated, 12 centers used CT planning,

2 used MRI planning and 1 both. For 11 plans some

adjustments were required, and in 6 cases a second

revision was requested. Main reasons for adjustments

were: delineation (N=8), dose planning (N=7),

reconstruction (N=2). Three different commercially

available treatment planning systems and HDR sources

were used. Table 1 summarizes dose to points A1-6, CTV

and OAR’s of the final submissions. Consistency with the

protocol improved and interobserver differences

significantly decreased with the revisions.

Conclusion

Interobserver variation in delineation resulted in the

largest dose deviations, most pronounced for bowel on

postoperative CT. The use of a second point (A3) at the

apex was most useful for controlling the anisotropy of the

source and should be recommended for dose reporting in

routine clinical practice.