The study focused on a molecularly defined subset of patients: those whose tumors carry methylation of the MGMT promoter, a chemical modification that silences a key DNA repair gene. This silencing matters enormously for treatment, because the chemotherapy drugs used against glioblastoma, temozolomide and lomustine, work by damaging tumor DNA. When the MGMT repair protein is suppressed, cancer cells cannot efficiently undo that damage, making them more vulnerable to the alkylating agents. Patients with MGMT-methylated tumors therefore respond better to chemotherapy, and the German CeTeG/NOA-09 trial demonstrated that these patients benefit from an intensified first-line regimen combining radiotherapy with both temozolomide and lomustine, also known as CCNU, rather than temozolomide alone.

Earlier work by a group led by Wiewrodt and colleagues had reported that among patients with MGMT-methylated, IDH-wildtype glioblastoma, those with blood group O lived longer than those with A, B, or AB blood types, an effect seen mainly in patients receiving standard temozolomide-based radiochemotherapy. But the association remained untested in patients receiving the intensified lomustine-temozolomide combination, and only a small fraction of the original cohort had received that regimen. The Bonn team, led by Julia Scheuble and Johannes Weller, set out to close that gap by examining an independent cohort treated uniformly under the CeTeG/NOA-09 protocol between 2008 and 2023.

The researchers assembled a cohort of 61 adults with newly diagnosed MGMT-methylated, IDH-wildtype glioblastoma, all of whom had documented ABO blood groups from routine preoperative typing and had received first-line radiochemotherapy with lomustine and temozolomide. Of these, 24 patients, or 39.3 percent, had blood group O, a proportion almost identical to the 39.2 percent prevalence of blood group O among German blood donors, which argues against any selection distortion. Baseline characteristics, including age, sex, Karnofsky Performance Status, and extent of surgical resection, were well balanced between the O and non-O groups, strengthening the credibility of the survival comparison.

The survival differences were substantial. Patients with blood group O lived a median of 42.1 months overall, compared with 23.7 months for patients with non-O blood groups, a difference that reached statistical significance with a log-rank p-value of 0.04. Progression-free survival, the time before the tumor began growing again, showed a similar trend: 21.8 months for blood group O versus 17.5 months for non-O groups, though this unadjusted comparison fell just short of significance. Median follow-up in the cohort was 56.1 months, and the majority of patients underwent surgery in the later years of the study period, reflecting the growing routine availability of molecular tumor characterization.

Crucially, the association survived statistical adjustment for the established prognostic factors. In multivariable Cox proportional hazards models accounting for age, extent of resection, and performance status, patients with non-O blood groups faced nearly double the hazard of progression, with a hazard ratio of 1.86, and more than double the hazard of death, with a hazard ratio of 2.29, compared with blood group O. Sensitivity analyses that sequentially removed individual covariates produced consistent estimates, with overall survival hazard ratios ranging from 2.02 to 2.42, indicating that the finding was not an artifact of any single adjustment variable. Tests of the proportional hazards assumption showed no violations, and salvage treatments at recurrence, including re-resection, re-irradiation, temozolomide rechallenge, bevacizumab, and regorafenib, were distributed similarly between the two blood group strata, suggesting that differences in second-line care did not drive the survival gap.

What could a blood type possibly have to do with brain cancer biology? The ABO system is defined by glycosyltransferase enzymes that decorate cell surfaces with specific sugar structures, and these glycans influence far more than transfusion compatibility. Experimental studies in non-glioma tumor models have linked fucosyltransferase activity, the enzymatic machinery that distinguishes blood group O from other types, to cellular mobility and resistance to apoptosis, the programmed cell death that chemotherapy is designed to trigger. In MGMT-methylated glioblastoma, where impaired DNA repair already heightens sensitivity to alkylating agents, the authors speculate that ABO-dependent glycosylation could theoretically modify how tumor cells respond to temozolomide or lomustine, altering the balance between treatment-induced damage and cellular survival.

The researchers are careful, however, to emphasize the limits of this mechanistic reasoning. There is currently no experimental evidence demonstrating a direct interaction between ABO glycosyltransferase activity, MGMT promoter methylation, and chemotherapy sensitivity in glioblastoma. The observed association could equally reflect differences in intrinsic tumor biology or the tumor microenvironment rather than a treatment-specific effect. The study itself carries the inherent constraints of a retrospective, single-center design with a limited sample of 61 patients, and the restriction to patients considered fit enough for intensified lomustine-temozolomide therapy produced a clinically selected population with relatively favorable performance status, limiting generalizability to frail patients. Residual confounding also cannot be excluded, and the 15-year study window may have introduced historical shifts in surgical technique, imaging, and salvage strategies, although the cohort was concentrated in the later years.

Despite these caveats, the consistency of the result across two independent cohorts, both restricted to MGMT-methylated patients and both showing the same direction of effect, gives the finding unusual weight for a biomarker study of this size. The Bonn cohort’s uniform treatment under a single protocol minimizes the treatment-related heterogeneity that has plagued earlier ABO studies, many of which predate the WHO 2021 integrated molecular classification of brain tumors and mixed together molecularly distinct entities. If validated in larger, multicenter cohorts, blood group typing, a test already performed routinely before any craniotomy, could become a virtually free addition to the prognostic toolkit, helping clinicians and patients calibrate expectations and potentially guiding the design of future trials.

The authors describe their results as hypothesis-generating, and the next step is translational: understanding whether ABO glycans genuinely modulate the response of glioma cells to alkylating chemotherapy, or whether blood group O simply tags some other, still unknown biological advantage. Either answer would deepen scientific understanding of how the most ancient of human genetic markers, the ABO system, interacts with modern cancer medicine. For now, patients with MGMT-methylated glioblastoma and blood group O can be told they carry a statistically favorable sign, while researchers pursue the biology that might one day turn that sign into a therapeutic target for everyone else.

Subject of Research: Association of ABO blood group O with survival outcomes in MGMT-methylated glioblastoma patients receiving lomustine-temozolomide radiochemotherapy

Article Title: Prognostic impact of blood group O in MGMT-methylated glioblastoma patients receiving radiochemotherapy with temozolomide and lomustine

Article References: Prognostic impact of blood group O in MGMT-methylated glioblastoma patients receiving radiochemotherapy with temozolomide and lomustine. (n.d.). https://doi.org/10.1007/s11060-026-05813-y

Image Credits: AI Generated

DOI: 10.1007/s11060-026-05813-y

Keywords: glioblastoma, ABO blood group, MGMT promoter methylation, temozolomide, lomustine, radiochemotherapy, prognostic biomarker, overall survival, progression-free survival, CeTeG/NOA-09, glycosylation, neuro-oncology