Prabhu 2014

Methods	Design: RCT (with observational arm – non‐randomised) Country: USA. Multicentred Accrual dates: 31 October 1998 to 27 June 2002, with long‐term follow‐up (results up to 5 years reported in the published paper). Trial reg: NCT00003375 Funding: it was reported that there was no commercial sponsorship, but in 'Conflicts of interest' it appeared that several investigators had received compensation from commercial organisations: Pharmacyclics, Merck Serono, Genentech, Bristol‐Myers Squibb, Merck, Novartis, Elekta, GlaxoSmithKline. It was not clear whether the compensation was outside this study. NIH funding.
Participants	No. randomised: (254 originally randomised) 251 eligible for evaluation. No. analysed: 230 included in analyses of cognitive function. Inclusion/exclusion criteria: WHO grade 2 glioma age 40 or more with any extent of resection or less than 40 with subtotal resection/biopsy. (Histologically confirmed grade 2 astrocytoma, oligodendroglioma or mixed oligoastrocytoma). Karnofsky performance status 60% or greater, neurological functioning score 3 or less and supratentorial location. Age: median age in RT arm 40 and 41 in RT + chemotherapy. Range overall 18 to 82 (ages 18 to 39 with subtotal resection or 40 or more with total resection and KPS > 60 and neuro ≤ 3) Gender: male 77/126, female 49/126 in RT arm, male 65/125 and female 60/125 in RT + Chemo + RT (lower proportion of females (39% ) in RT arm versus 48% in RT + chemo arm (NS)) Glioma type: confirmed grade 2 astrocytoma (65) , oligodendroglioma (107) or mixed oligoastrocytoma (79) Glioma grade: grade 2 (low grade) Resection/biopsy: biopsy 119/251, partial resection 107/251, total resection 25/251 Anti‐epileptics/SSRIs: not stated. Duration of follow‐up: results up to year 5 reported for cognitive outcomes (survival follow‐up ongoing). At 4, 8,12, 18 and 24 months and annually thereafter. MMSE evaluated at each follow‐up point but discontinued with tumour progression.
Interventions	Arm 1: (n = 128 randomised, 122 analysed) radiotherapy alone (54 Gy in 30 fractions of 1.8 Gy) over 6 weeks Arm 2: (n = 125 randomised, 116 analysed) radiotherapy plus chemotherapy. Following radiotherapy (as arm 1) patients received 6 cycles of procarbazine (60 mg/m² orally per day on days 8 and 21 of each cycle), lomustine (110 mg/m² on day 1 of each cycle) and vincristine (1.4 mg/m² (maximum 2 g) IV on days 8 and 29 of each cycle. The cycle length was 8 weeks.
Outcomes	Reported review outcomes: cognitive function assessed by MMSE at 1, 2, 3 and 5 years from randomisation. Significant MMSE decline was defined as a decrease of more than 3 points and gain as an increase in score of more than 3 points compared with baseline. Other reported study outcomes: survival reported in main trial report
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Published papers and trial registration do not state how randomisation was carried out
Allocation concealment (selection bias)	Unclear risk	Described as randomised trial with parallel assignment and randomisation stratified by tumour type but methods of allocation concealment were not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	It was described as an open label trial with no masking in the trial registration
Blinding of outcome assessment (detection bias) All outcomes	High risk	For some outcomes (survival), lack of masking may not have been important, but for assessment of cognitive function lack of blinding may have introduced bias
Incomplete outcome data (attrition bias) All outcomes	High risk	There was considerable loss to follow‐up due to death and a large proportion of patients (approximately 1/3) had no MMSE assessment at 1 year. There seems to have been fewer responses at all time points in the radiotherapy plus chemotherapy arm.
Selective reporting (reporting bias)	Unclear risk	The trial was registered but there was very little information about methods
Other bias	Unclear risk	It was not explained why large numbers of patients were not assessed using the MMSE