New BLIP score may help predict survival in patients with NSCLC

Brain metastases remain one of the most difficult complications in advanced lung cancer, often signaling a sharp decline in survival. Now, researchers say a newly developed prognostic tool may help clinicians better estimate outcomes for patients with non-small cell lung cancer (NSCLC) and brain metastases receiving immunotherapy. The Brain-Lung Immunotherapy Prognostic (BLIP) score, described in a new study, successfully stratified patients into distinct prognostic groups using just three routine clinical factors.

The findings, published May 20, 2026, in the Nature Portfolio British Journal of Cancer, suggest the BLIP score could offer clinicians a practical tool for counseling patients, guiding treatment planning and informing decisions about intensity of monitoring and palliative care. In both the primary and validation cohorts, patients classified in the “good” prognostic group had approximately double the median overall survival compared with those in the “poor” group.

Lung cancer remains the leading cause of cancer-related death worldwide, and brain metastases are a common and particularly challenging complication in advanced NSCLC. Although immunotherapy drugs, including immune checkpoint inhibitors (ICIs) such as Keytruda (pembrolizumab) and Imfinzi (durvalumab), have improved outcomes for many patients, survival remains highly variable, in part because of differences in tumor biology, disease burden and treatment history. Existing prognostic models for brain metastases were largely developed before the immunotherapy era or rely on biomarkers such as PD-L1 expression, which may vary substantially within and between tumors over time.

To address that gap, Marcus Skribek, M.D., Ph.D., lead author and consultant clinical oncologist and postdoctoral researcher at Karolinska Institutet in Sweden, and colleagues developed and externally validated the BLIP score using real-world data from two European centers. The retrospective study included 131 patients treated at Karolinska University Hospital in Sweden and an independent validation cohort of 109 patients treated at Sotiria Thoracic Diseases Hospital of Athens in Greece. All patients had NSCLC with radiologically confirmed brain metastases and received ICIs either alone or in combination with chemotherapy.

Investigators screened 1,844 patients before applying eligibility criteria. They then analyzed clinical factors associated with survival and identified three key predictors: tumor histology, age at brain metastasis diagnosis and number of brain metastases. Patients received one point each for having non-squamous NSCLC, being younger than 65 years and having one to three brain metastases, creating a BLIP score ranging from 0 to 3.Patients were then stratified into “poor” (0–1 points) and “good” (2–3 points) prognostic groups.

In the primary cohort, median overall survival was 7 months in the poor prognosis group compared with 14.5 months in the good prognosis group. External validation produced similar findings, with median overall survival of 8 months versus 14 months, respectively.

Additional analyses suggested the BLIP score was a reliable tool for distinguishing between patients with better and worse survival outcomes. Notably, PD-L1 expression was not independently associated with overall survival and was excluded from the final model.

According to Skribek and his colleagues, the BLIP score differs from earlier prognostic tools because it was developed specifically in patients receiving immunotherapy and measures survival from the time brain metastases are diagnosed rather than treatment initiation. The researchers emphasized that the score is prognostic — not predictive — meaning it is intended to estimate survival rather than determine whether a patient will respond to immunotherapy

In their paper, the authors noted that patients categorized as having poor prognosis may benefit from earlier palliative care involvement and symptom-focused management, while those with better prognoses could be candidates for more aggressive treatment strategies or clinical trial enrollment.

“Real-world evidence can help address important evidence gaps created by restrictive clinical trial eligibility criteria,” Skribek told Managed Healthcare Executive in an email interview. “Patients with brain metastases, poor performance status, comorbidities, or other complex clinical characteristics are often underrepresented in prospective trials.”

He added that the BLIP score aims to “provide clinicians with a simple and practical prognostic tool that can improve risk stratification, patient counseling, and clinical decision-making in a population that remains underrepresented in many prospective studies.”

Although the study was retrospective and relatively small, the authors said the consistent findings across two independent cohorts support the score’s broader clinical relevance. Future prospective studies, they noted, could help refine the model further by incorporating additional biomarkers and validating its use in larger patient populations.

“Real-world studies allow us to evaluate outcomes in these populations and better understand how therapies perform in routine clinical practice,” Skribek said. “This information can complement clinical trial data and support more informed regulatory, reimbursement and clinical decision-making.”