Background: NOA-04 compared the efficacy and safety of radiotherapy versus (vs)chemotherapy in patients with newly-diagnosed, centrally assessed, supratentorial gliomas of WHO grade III and had not demonstrated a difference in efficacy parameters between arms. Here, long-term clinical follow-up for the intention-to-treat population (n = 274) and the biomarker group (n = 207) are presented. Methods: Patients with anaplastic astrocytoma (52.6%) and anaplastic oligodendroglial tumors (47.4%) were randomized 2:1:1 (A:B1:B2) to receive radiotherapy to 54-60 Gy or four 6-week cycles of procarbazine, CCNU and vincristine (PCV) or eight 4-week cycles of temozolomide. The primary endpoint was time-to-treatment-failure (TTF) defined as progression after radiotherapy and one chemotherapy in either sequence, or any time before if no further therapy could be administered. Exploratory analyses of the correlation between molecular status and TTF, progression-free (PFS) and overall survival (OS) were part of the study. Results: With a median observation period of 11.8 years (ys) median TTF [4.6 (3.4-5.1) ys vs 4.4 (3.3-5.3) ys], PFS [2.5 (1.3-3.5) ys vs 2.7 (1.9-3.2) ys], and OS [8 (5.5-10.3) ys vs 6.5 (5.4-8.3) ys] were not different between arms (A vs B1/B2). Oligodendroglial vs astrocytic histology, but more so subgroups according to CpG island methylator phenotype (CIMP) and 1p/19q co-deletion status revealed a strong prognostic value of CIMP^pos with (CIMP-codel) versus without 1p/19 co-deletion (CIMP-non-codel) vs CIMP^neg., but no differential efficacy of radio- and chemotherapy for any of the observed endpoints (PFS CIMP-codel: 8.8 (4-11.2) ys for A versus 7.6 (4.1-9.5) ys for B). In CIMP^neg. mainly astrocytic tumors hypermethylation of the O6-methyl-guanyl-DNA methyltransferasepromoter (MGMT) provided a large risk reduction for PFS in arms B1/B2, but not arm A. Conclusions: NOA-04 long-term data do not support a differential efficacy of primary temozolomide monotherapy or PCV polychemotherapy vs radiotherapy in any of the histological or molecular subgroups of anaplastic glioma. Molecular diagnosis is superior to histology. MGMT is predictive in CIMP^neg. gliomas. Clinical trial information: NCT00717210