1734MO - Hypoxic bladder cancers have a poorer outcome following hypofractionated vs conventionally fractionated radiotherapy in the BC2001 and BCON randomised trials

Background

Radical treatment for muscle-invasive bladder cancer (MIBC) involves radiotherapy with radiosensitiser or cystectomy currently selected subjectively (patient/clinician preference). The West 24-gene bladder cancer hypoxia signature predicted benefit from hypoxia-modifying radiosensitisation (carbogen+nicotinamide, CON) in BCON trial. The BC2001 trial showed combining chemotherapy (5-FU, mitomycin-C) with radiotherapy (RT) improves loco-regional progression free survival. Both BCON and BC2001 allowed conventional (62Gy/32 fractions) or hypofractionated (55Gy/20 fractions) RT. We tested two hypotheses: (1) our hypoxia signature does not predict benefit from concurrent chemotherapy; and (2) accelerated treatment reduces reoxygenation and is less favourable for patients with hypoxic tumours.

Methods

Pre-treatment biopsy samples were obtained from 312 patients from BC2001. RNA was extracted and full transcriptomic data generated using Affymetrix Clariom S arrays. Patients were stratified using the cohort median of the median expression of 24-signature genes hypoxia score into hypoxia-high and -low groups. Data for BCON (n=151) were available from a published study.

Results

Hypoxia score was prognostic for overall survival in the combined BC2001 cohort in univariable and multivariable (HR=1.29; 95% CI 0.97-1.71; p=0.075) analyses with no interaction between hypoxia score and treatment arm (p=0.92) Patients with hypoxic tumours had a poorer prognosis following hypofractionated compared with conventional fractionation in both BC2001 (n=298, HR 1.80 95% CI 1.08-2.91; p=0.023) and the BCON RT only arm (n=75; HR 14.2; 95% CI 1.7-119; p=0.015). CON abrogated the detrimental effect of hypoxia with hypofractionated RT.

Conclusions

Hypoxic MIBC has a poor prognosis and exhibits fraction sensitivity, which is mitigated with hypoxia modification. Use of hypoxia score to personalise treatment needs testing in a biomarker-stratified trial.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

NIHR CRUK.

Disclosure

All authors have declared no conflicts of interest.