MRI diffusion and perfusion markers link MYD88/CD79B co-mutation to poorer outcomes in primary CNS lymphoma

Quantitative diffusion and perfusion MRI may offer a non-invasive way to identify an aggressive molecular subtype of primary central nervous system lymphoma (PCNSL), according to a retrospective study published in Cancer Imaging. The research found that patients whose tumours carried mutations in both the MYD88 and CD79B genes showed distinct imaging signatures, and that an imaging-predicted co-mutation status was associated with shorter survival and lower rates of complete response.

MYD88 and CD79B alterations are key molecular drivers in PCNSL, with prognostic and therapeutic implications.

The researchers analysed data from 92 patients with PCNSL. Of these, 60 carried an MYD88 mutation, 52 carried a CD79B mutation, and 40 had mutations in both genes.

Patients with MYD88/CD79B co-mutation showed significantly lower ADCmin and rADCmin values, together with higher CBVmax and rCBVmax values, compared with cases without the co-mutation. The adjusted P values were 0.020, 0.001, 0.001 and less than 0.001, respectively.

To test how well these parameters could discriminate between genomic subtypes, the researchers used receiver operating characteristic analysis with stratified fivefold cross-validation. The cross-validated discriminatory performance was moderate to good, with areas under the curve ranging from 0.652 to 0.861.

The prognostic findings were particularly notable. Out-of-fold, imaging-predicted MYD88/CD79B co-mutation status was significantly associated with shorter progression-free survival and shorter overall survival, as well as a lower complete response rate, with all reported P values below 0.05.

Survival analysis was conducted using Kaplan-Meier estimates with log-rank tests, and Cox proportional hazards regression was used to assess prognostic factors for progression-free survival and overall survival. Through this analysis, diffusion and perfusion metrics were found to be independently associated with survival outcomes, indicating that the imaging parameters carried prognostic information beyond a simple binary mutation classification.

The authors concluded that quantitative DWI and DSC-PWI parameters may aid non-invasive characterisation of MYD88/CD79B mutation status and preliminary prognostic stratification in PCNSL. The cautious framing of the conclusion is worth emphasising: the study describes associations observed in a single retrospective cohort of 92 patients, and the authors describe the prognostic stratification as preliminary rather than established.

Several limitations follow from the study design. The findings arise from a retrospective analysis, and the reported discriminatory performance, while cross-validated, was moderate to good rather than definitive, with some area under the curve values as low as 0.652. The imaging-predicted co-mutation status is a statistical association with outcomes, not proof that the imaging parameters cause or directly determine prognosis.

Subject of Research: Quantitative diffusion and perfusion MRI for characterising MYD88/CD79B mutation status and prognosis in primary CNS lymphoma

Article Title: Quantitative diffusion and perfusion MRI for non-invasive MYD88/CD79B genotype characterization and prognostic stratification in primary CNS lymphoma

Article References: Zhou, X., Yu, S., Yu, F., Zeng, Z., Xing, Z., & Cao, D. (2026). Quantitative diffusion and perfusion MRI for non-invasive MYD88/CD79B genotype characterization and prognostic stratification in primary CNS lymphoma. Cancer Imaging. https://doi.org/10.1186/s40644-026-01145-2

Image Credits: AI Generated

DOI: 10.1186/s40644-026-01145-2

Keywords: primary CNS lymphoma, MYD88, CD79B, diffusion MRI, perfusion MRI, ADC, cerebral blood volume, genotype, prognosis, cross-validation, biomarkers, Cancer Imaging