4. Impact of adjustment for other prognostic factors.
| Study ID | Technique | Sample type | CpGs analysed (PCR‐based tests) | Threshold for methylated | Unadjusted HR (95% CI) | Adjusted HR (95% CI) | Factors adjusted for (if an adjusted HR reported) | Source of results | Risk of bias (Domain 6) |
| Bady 2012 (M‐GBM) | Bead array | Frozen | 31 and 83 | > 0.358 | 6.46 (2.41 to 17.35) | 6.51 (2.42 to 17.54) | Age | IPD | Low |
| MSP | NR | 76–80 and 84–86 | NR | 7.21 (2.37 to 21.99) | 7.38 (2.41 to 22.60) | Age | IPD | Low | |
| Dahlrot 2018 (NS cohort) | DIF | FFPE | N/A | < 0.2 | 1.60 (0.95 to 2.71) | 1.50 (0.93 to 2.43) | Age, ECOG performance status and gender | Directly reported | Low |
| Hsu 2015 | IHC | FFPE | N/A | < 10% | 2.12 (1.32 to 3.42) | 1.80 (1.01 to 3.21) | Age, sex, KPS, extent of resection, bevacizumab treatment and IDH1 status | Directly reported | Low |
| MSP | FFPE | 76–80 and 84–87 | NR | 2.39 (1.42 to 4.02) | 2.62 (1.50 to 4.55) | Age, sex, KPS, extent of resection, bevacizumab treatment and IDH1 status | Directly reported | Low | |
| PSQ | FFPE | 76–79 | > 5% | 2.66 (1.49 to 4.76) | 2.51 (1.46 to 4.33) | Age, sex, KPS, extent of resection, bevacizumab treatment and IDH1 status | Directly reported | Low | |
| qMSP | FFPE | 77–80 and 84–87 | > 0.04% | 2.75 (1.51 to 5.04) | 2.65 (1.47 to 4.76) | Age, sex, KPS, extent of resection, bevacizumab treatment and IDH1 status | Directly reported | Low | |
| McDonald 2013 | MSP | FFPE | 76–80 | NR | 1.64 (0.95 to 2.83) | 1.63 (0.95 to 2.81) | Age | IPD | Low |
| PSQ | FFPE | 74–78 | > 8% | 1.96 (1.16 to 3.33) | 1.68 (0.99 to 2.84) | Age | Directly reported | Low | |
| Thon 2017 | MSP | Frozen | 76–80 and 84–87 | NR | 3.33 (1.82 to 6.25) | 3.23 (1.72 to 6.25) | Age, RTOG and KPS | Directly reported | Low |
| Sequencing | Frozen | 75–99 (unclear) | > 50% | 3.33 (1.82 to 6.25) | 3.23 (1.72 to 6.25) | Age, RTOG and KPS | Directly reported | Low | |
| Yamashita 2018 | MSP | Frozen | 76–80 and 84–87 | NR | 2.36 (1.62 to 5.05) | 1.63 (0.86 to 3.09) | Surgery (gross‐total resection vs other) and MGMT status by PCR‐HRM | Directly reported | High (model includes other MGMT status using alternative method) |
| PCR‐HRM | Frozen | 72–89 | > 10% | 2.51 (1.63 to 4.83) | 2.36 (1.20 to 4.65) | Surgery (gross‐total resection vs other) and MGMT status by MSP | Directly reported | High (model includes other MGMT status using alternative method) | |
| Yang 2012 | IHC | FFPE | N/A | < 10% | 1.07 (0.35 to 3.31) | 1.50 (0.37 to 6.04) | Age and extent of resection | IPD | Low |
| MSP | FFPE | 76–80 and 84–87 | NR | 1.35 (0.44 to 4.16) | 0.99 (0.29 to 3.45) | Age and extent of resection | IPD | Low |
CpG: 5'‐cytosine‐phosphate‐guanine‐3'; CI: confidence interval; DIF: double immunofluorescence; ECOG: Eastern Cooperative Oncology Group; FFPE: formalin‐fixed paraffin embedded; HR: hazard ratios; IDH: isocitrate dehydrogenase; IHC: immunohistochemistry; IPD: individual participant data; KPS: Karnofsky performance status; MGMT: O6‐methylguanine–DNA methyltransferase; MSP: methylation‐specific polymerase chain reaction; N/A: not applicable; NR: not reported; PCR: polymerase chain reaction; PCR‐HRM: polymerase chain reaction with high‐resolution melting; PSQ: pyrosequencing; qMSP: quantitative methylation‐specific polymerase chain reaction; RTOG: Radiation Therapy Oncology Group prognostic factor class.