TOPLINE

A cerebrospinal fluid (CSF) biomarker panel, combining MYD88 mutation testing with interleukin 10 (IL-10) and C-X-C motif chemokine ligand 13 (CXCL13) measurements, showed high specificity for CNS lymphoma and correctly identified more than half of the CNS lymphoma cases. With further validation, this panel could help reduce reliance on invasive brain biopsy in selected patients, the authors suggested.

METHODOLOGY

• Brain biopsy remains the reference standard for diagnosing primary CNS lymphoma but can be risky and contribute to diagnostic delays. CSF biomarkers, such as the MYD88 mutation, IL-10, and CXCL13, have shown promise in identifying CNS lymphoma, but most evidence has come from retrospective studies. This study tested their diagnostic value in routine practice.
• Researchers prospectively enrolled 255 consecutive patients (median age, 61 years; 67% men) with suspected primary CNS lymphoma from five hospitals in Netherlands between March 2021 and May 2025.
• Participants underwent CSF analysis for the MYD88 L265P mutation via droplet digital polymerase chain reaction and measurement of IL-10, CXCL13, and five additional biomarkers. Researchers used decision tree analysis to develop and evaluate a diagnostic panel that incorporated the three key biomarkers.
• Final diagnoses were established using brain biopsy and clinical evaluation, with 50 patients ultimately diagnosed with CNS lymphoma (44 primary and 6 secondary).

TAKEAWAY

• MYD88 mutations were detected in 28 of 50 patients with CNS lymphoma. MYD88 mutation testing alone demonstrated a sensitivity of 56.0%, specificity of 98.5%, positive predictive value of 90.3%, and negative predictive value of 90.0%.
• Combining MYD88 positivity with elevated IL-10 and/or CXCL13 yielded a sensitivity of 54% and a specificity of 100%, correctly classifying 27 of 50 CNS lymphoma cases. A negative result for all three biomarkers provided a negative predictive value of 99.4% (162 of 163).
• Among biopsy-confirmed primary CNS lymphoma cases, 13 (45%) met the biomarker criteria that could, with further validation, support a noninvasive diagnosis in selected clinical scenarios.
• In univariable analyses, six of seven chemokines demonstrated significant diagnostic value with the area under the curve (AUC); IL-10 achieved the highest AUC of 0.96, followed by CXCL13 at 0.95. In multivariable analysis, MYD88 (odds ratio [OR], 71.90) and IL-10 (OR, 2.82; P < .0001) were independently associated with CNS lymphoma, while CXCL13 was retained in the panel on the basis of model selection despite not reaching statistical significance (OR, 1.39; P = .081).

IN PRACTICE

“We validated the diagnostic value of cerebrospinal fluid biomarkers [MYD88 L265P mutation, CXCL13, and IL-10] for primary CNS lymphoma,” the authors of the study wrote, adding that “in the appropriate clinical context, this approach potentially increases minimally invasive diagnoses” and could reduce the need for brain biopsy in certain patients with suspected primary CNS lymphoma.

SOURCE

The study was led by Josephus L.M. van Rooij, MD, St. Antonius Hospital, Utrecht, Netherlands. It was published online in The Lancet Haematology.

LIMITATIONS

The diagnostic algorithm and decision tree were developed and validated within the same cohort, introducing the risk of overfitting. The relatively small number of CNS lymphoma events may have limited the precision and stability of multivariable effect estimates. Clinical suspicion was determined by treating physicians without central review, potentially introducing heterogeneity and affecting estimates of specificity and negative predictive value.

DISCLOSURES

The study received financial support from Stichting Annie van Koeverden. The authors reported no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.