Abstract Book

ESTRO 37

S567

2 actrec-Tata Memorial Centre, Surgical Oncology, Navi Mumbai, India 3 tata Memorial Hospital, Radiation Oncology, Mumbai, India 4 tata Memorial Hospital, Surgical Oncology, Mumbai, India 5 actrec-Tata Memorial Centre, Statistics, Navi Mumbai, India Purpose or Objective Comparison of subjective and objective assessment methods of cosmetic outcome following multicathetar interstitial brachytherapy implant (MIB) for early breast cancer. Material and Methods 60 patients of early breast cancer who were treated with high dose rate MIB for partial breast irradiation were followed up regularly every 6 months. In this prospective study, all consecutive patients who were 18 to 36 months post brachytherapy were screened and accrued after obtaining informed consent. One time cross-sectional assessment of the breast cosmesis was done by both physician rated subjective method using Harvard scale as well as by objective photographic method using BCCT.Core (version 3.1) software. The correlation between subjective and objective methods for the overall score as well as individual domains was done using Spearman correlation coefficient.

Conclusion In the current brachytherapy cohort, moderate correlation was observed for overall cosmesis score as well as nipple location domains while comparing physician rated subjective assessment with BCCT.core derived objective indices.

Poster: Brachytherapy: Gynaecolgy

PO-1015 Image-guided salvage radiotherapy of pelvic recurrences in post-hysterectomy endometrial cancer J. Knoth 1 , D. Berger 1 , R. Pötter 1 , M. Schmid 1 , C. Grimm 2 , V. Seebacher 2 , S. Polterauer 2 , J. Widder 1 , A. Reinthaller 2 , A. Sturdza 1 1 Medical University of Vienna, Radiation Oncology, Vienna, Austria 2 Medical University of Vienna, Gynecology, Vienna, Austria Purpose or Objective To evaluate response, local control (LC), progression free survival (PFS), overall survival (OS), and toxicity in recurrent endometrial cancer (EC) in patients undergoing image guided adaptive brachytherapy (IGABT) w/o external beam radiotherapy (EBRT). Material and Methods A retrospective analysis of consecutive patients undergoing radiotherapy for local/loco-regional recurrence of EC treated initially with surgery was performed (2001-2016). Irradiation included IGABT alone in patients with superficial tumors limited to the vaginal wall and N0 status - Group 1 (Gr1). Group 2 (Gr2) included patients with recurrences extending beyond the vaginal wall and/or to the regional lymph nodes. Treatment was standard pelvic EBRT (45 Gy) + IGABT. IGABT was performed using MRI- (70 %) or CT-guidance (30%) at least at the first fraction. The primary/residual tumor was contoured as clinical target volume; the whole vaginal wall as low risk clinical target volume. Planning aim to the whole vaginal wall was 60 Gy and to the rest tumor 85 Gy. Response rate, actuarial LC, PFS, and OS were calculated; long term toxicity was recorded using CTCAEv4 criteria. Results Median age of the cohort (n=52) was 76 years (53-91), median follow up after recurrence treatment 49 months (range 2-184). All patients had an EC type I histology initially, while 13 (25%) recurred with a more aggressive histology (Type II). Median time to recurrence after initial surgery was 26 (5-130) months. Eleven patients had previous pelvic EBRT (n=5) or BT (n=6). Response rate was 95%, actuarial LC, PFS, and OS rates at 2/5 years were 95%/95%, 74%/69%, and 85%/78%, respectively. In 4 patients (8%) ≥G3 GU toxicity (vesico-vaginal fistula, vaginal stenosis) and/or recto-vaginal fistula occurred, of whom two had previous adjuvant IGABT/ EBRT. In Gr1 (n=9), median age was 70 years (53-78), median FU 46 months (6-90). One patient had previous pelvic EBRT. All had complete remission after treatment completion. One patient developed pelvic lymph node recurrence. Actuarial LC, PFS and OS rates at 2/5 years were 100%/100%, 100%/83%, and 100%/100%, respectively. No grade ≥3 toxicity occurred. In Gr2 (n=43), median age was 77 years (53-91), median FU 49 months (2-184). Two patients had incomplete local remission and developed lung metastases within 3

Table 1: Summary of correlation statistics comparing subjective with objective domains.

Spearman correlation

p value

Overall cosmesis score Excellent-good vs. Fair- poor

0.434

0.001

Nipple

location

(NL) BRA UNR

0.285 0.349 0.449

0.027 0.006 0.000

NL NL

vs. vs.

NL vs. BCE

Breast size (BS) BS vs. BAD BS vs. BOD

-0.246 0.034

0.058 0.797

Breast

Shape

(BSh)

-0.119

0.366

BSh vs. BCD

BRA: Breast retraction score, UNR: Upward nipple retraction, BCE: Breast compliance evaluation, BAD: Breast area difference, BOD: Breast overlap difference, BCD: Breast contour difference. Results Overall, there was 76.6% agreement in the physician rated and objective methods of assessment of cosmesis. However, there was only moderate correlation between the two methods of assessments for the overall score. Among the individual domains, only nipple location correlated moderately with the software defined indices. The results of correlation statistics are summarized in Table 1. There was poor correlation between the two methods for skin colour difference and scar visibility features (results not shown), with spearman correlation coefficient ranging from -0.008 to 0.218 (p value NS) for skin colour and -0.011 to 0.160 (p value NS) for scar visibility.