Investigation of the optimal combination of external beam radiotherapy and high-dose-rate intracavitary brachytherapy in definitive radiotherapy for uterine cervical cancer patients

Abstract

Purpose: Intracavitary brachytherapy (ICBT) and external beam radiotherapy (EBRT) are both essential components of definitive radiotherapy for patients with uterine cervical cancer. From previous phase II trials, reduced cumulative central dose using midline block (MLB) did not compromise the treatment outcome while reducing late toxicity rate compared to other studies. However, no randomized evidence is available and long term results are needed to confirm the efficacy and safety of this treatment approach. We aimed to assess the efficacy and toxicity of low cumulative central dose using MLB during EBRT. Patients and Methods: Between January 1988 and December 2010, a total of 1559 patients with uterine cervical cancer (FIGO stage IB 410, stage IIA 133, stage IIB 1016) who received definitive radiotherapy (n = 1054, 67.6%) or platinum-based chemoradiotherapy (n = 504, 32.4%) consisting of EBRT and high-dose-rate intracavitary brachytherapy (HDR-ICBT) were retrospectively analyzed. The median EBRT dose was 45.0 Gy (range, 30.6？60.0 Gy) in 1.8 Gy per fraction and median HDR-ICBT dose prescribed at point A was 30 Gy (range, 12？63 Gy) in median 5 Gy (3.0？6.0 Gy) per fraction. During EBRT, tumor response was checked every week and when sufficient response was achieved to place the ICBT applicator, MLB of 4 cm width and 8？10 cm in height was placed (n = 1195, MLB group). For patients with slow tumor response during EBRT, full dose was applied without MLB (n = 364, non-MLB group). MLB was performed after ≤ 27 Gy (n = 229), > 27 Gy and ≤ 36 Gy (n = 847), or > 36 Gy (n = 119) of EBRT. The rectal and bladder doses were estimated using doses at the International Commission on Radiation Units and Measurements points. To calculate the cumulative dose from EBRT and ICBT, the biologically equivalent dose in 2-Gy fractions (EQD2) using the linear quadratic model was used (α/β value of 3 for normal tissue and 10 for tumor). Propensity score matching was also performed to balance the characteristics between MLB and non-MLB group. Results: Median follow-up period was 89.0 months (range, 2.4-320.2 months). The 10-year overall survival (OS), progression-free survival (PFS), regional recurrence (RR), and local recurrence (LR) rates were 82.3%, 74.7%, 2.6%, and 9.5%, respectively. The 10-year OS, PFS, RR, and LR (all Ps <0.05) were significantly superior in the MLB group compared to the non-MLB group. The MLB group was older in age, had smaller tumor size, lower FIGO stage, higher pelvic and para-aortic lymph node metastases rate than the non-MLB group. EQD2point A (72.9 Gy vs. 86.4 Gy), EQD2rectal (64.5 Gy vs. 74.8 Gy), and EQD2bladder (67.4 Gy vs. 75.9 Gy) were all significantly lower in the MLB group (all Ps <0.05). Grade ≥2 late rectal toxicity was significantly lower in MLB group (8.1% vs. 11.5%, P = 0.045). There was no significant difference in late genitourinary and small bowel toxicity. After all patient and tumor characteristics were well balanced using propensity score matching, the 10-year OS, PFS, RR, and LR were similar between the MLB and non-MLB group (all Ps >0.05) despite the lower EQD2point A (72.9 Gy vs. 86.4 Gy; P <0.001). Conclusion: Lowering the cumulative central dose using EBRT with MLB according to tumor response may reduce rectal toxicity without compromising treatment outcome. The efficacy of MLB should further be evaluated in prospective trials.