Kiebert 1998

Methods	Design: sub‐study of a RCT Country: multinational sites ‒ 14 out of 27 sites contributed to the QoL substudy Accrual dates: April 1985 to September 1991 Trial reg: EORTC 22844 (Karim 1996) Funding: NR
Participants	No. randomised: 379 No. analysed: 109 of 113 that complete a baseline QoL questionnaire Inclusion/exclusion criteria: all adult patients (age 16 to 65 years) having a definite histopathologic diagnosis of low‐grade astrocytomas (G1 and G2), oligodendroglioma, and mixed oligoastrocytomas of the supratentorial areas. Grade 1 (pilocytic) astrocytoma, if totally excised, was excluded, while grade 2 astrocytoma, even if totally excised, was included. Oligodendrogliomas and mixed oligoastrocytomas were included. The patients had to have been in reasonable good general condition as indicated by performance score after surgery: Karnofsky index ≥ 60 and WHO score ≤ 2. Neurologic deficit status was also recorded and defined: 1 = no deficit; 2 = some deficit but with adequate functioning for useful work; 3 = moderate functional impairment with movement difficulties, moderate dysparesis, paresis, and visual or memory impairment; 4 = major functional impairment; and 5 = lack of conscious response. The patients in categories 4 and 5 were excluded from this trial. Patients with pregnancy or gross hepatic, renal, or cardiovascular diseases of malignancy other than curable skin cancers were excluded. However, patients thought to be cured of cancer for at least 5 years before inclusion in the protocol were eligible. Age: < 35 years (40), 35 to 44 years (35), ≥ 45 years (38) Gender: low dose: 24 female, 33 male; high dose: 28 female, 28 male. Glioma type: low dose: astrocytoma (37), oligodendoglioma (15), mixed (5); high‐dose: astrocytoma (35), oligodendoglioma (17), mixed (4) Glioma grade: low dose: 0 or 1 (6), > 1 (51); high dose: 0 or 1 (6), > 1 (50) Resection/biopsy: low dose: < 50% tumour excised (23), ≥ 50% tumour excised (34); high dose: < 50% tumour excised (30), ≥ 50% tumour excised (26) Anti‐epileptics/SSRIs: NR Duration of FU: 2+ years
Interventions	Arm 1: (n = 57) low dose RT (45 Gy in 5 weeks) Arm 2: (n = 56) high dose RT (59.4 Gy in 6 weeks)
Outcomes	Reported review outcomes: QoL (self‐reported scale) including physical, social, psychological, and symptom domains. Signs and symptoms were rated using a Likert scale 1 to 4 (4 = severe); Rand HIS‐Physical capacities scale. Other reported outcomes: Survival (OS, PFS) Timing of follow up: 3, 6, 12, 18, 24 months and then annually. Data were analysed in 2 time points ‒ immediately post RT and at 7 to 15 months.
Notes	Longer term (2+ year) follow‐up data have not been reported because "compliance with further follow‐up was so poor that analysis of these latter data were considered inappropriate". At the 7 to 15 month follow‐up, there was no significant difference in neurological impairment observed, and no significant difference was found in the proportion of patients with the worst neurological scores (data were not shown). Emotional functioning and leisure time activities were significantly worse with high dose (P = 0.009 and P = 0.017, respectively). Authors found no major difference in QoL overall but some individual QoL items were worse with high‐dose RT. In the primary study, high‐dose RT did not lead to better survival than low‐dose RT.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Central randomization with stratification according to institution and grade Baseline characteristics of patients that completed the QoL questionnaires were not significantly different to those of the whole sample
Allocation concealment (selection bias)	Low risk	"Central randomization"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported
Incomplete outcome data (attrition bias) All outcomes	High risk	Long‐term (2+ year) follow‐up data have not been reported because "compliance with further follow‐up was so poor that analysis of these latter data were considered inappropriate".
Selective reporting (reporting bias)	Unclear risk	Not possible to make a judgement
Other bias	Unclear risk	None noted