The researchers reported at 2017 American Society for Radiation Oncology (ASTRO) Annual Meeting the results of a prospective, randomised, multicentre trial of the International Atomic Energy Agency (IAEA) that compared loco-regional (LR) control and adverse effects (AEs) of external beam radiotherapy (EBRT) in combination with 2 different fractionation schedules of high-dose rate brachytherapy (HDRBT) with or without chemotherapy in patients with cervical cancer. They found a dose-effect relationship for tumour control.

A prospective, randomised, multicentre international trial of the IAEA tested four combinations of HDRBT and chemotherapy on cervical cancer. Eligible patients were women with stages IIB and IIIB cervical carcinoma being treated with curative intent and with no contraindications for EBRT, HDRBT and chemotherapy.

All patients were to receive EBRT, 46 Gy in 23 fractions to the pelvis. Prescribed HDRBT dose in arm A was 4 applications of 7 Gy each to point ‘A’ while in arm B it was 2 applications of 9 Gy. Arms C and D were similar to arms A and B but with cisplatin (40mg/m^2) in weeks 1 through 5.

The LR tumour control, overall survival and acute/late AEs were compared between arms. Using a/b values of 10 Gy for tumour control and 3 Gy for late-AEs, biological effective doses and equivalent doses were calculated for each arm.

Between September 2005 and May 2010, 601 patients were randomised. There were 257 patients from India, 147 cases from Peru, 76 from South Africa, 53 from Brazil, 31 from Pakistan, 19 from Morocco, and 18 from Macedonia. Average age was 48.7 year; 440 patients had stage IIB, and 161 had stage IIIB.

Overall 5-year survival was 71% for stage IIB patients and 58% for stage IIIB patients (P=0.03). The 5-year survival for all women, combined, was 67.2% (95% CI 62.7-71.2%). By treatment arm, 5-year overall survival was: 62.2% in arm A, 68.3% in arm B, 73.1% in arm C, and 65.1% in arm D. By log-rank test, stratified by centre and stage, there was no statistical difference in overall survival by study arm.

For the 440 stage IIB patients, there was no statistical difference in survival with 4 HDR versus 2 HDR, and no difference with or without chemotherapy.

Five-year tumour control was lower in arms B and D compared to arms A and C (P=0.0007).

No statistically significant difference in AEs was found. The only effect of cisplatin was an increased AEs trend in the 2x9 Gy arm B (P=0.066).

The study team concluded that a dose-effect relationship was found for tumour control. Local control was significantly superior for the arms including 4 fractions of 7 Gy HDRBT compared to 2 fractions of 9 Gy. No statistically significant differences in OS or AEs were found between arms.