Glowing Dye Showdown: 5-ALA Edges Out Fluorescein in Brain Cancer Survival

For two decades, neurosurgeons have relied on glowing dyes to help them see the invisible margins of glioblastoma, the most aggressive and lethal form of brain cancer. Now a new retrospective study from a team at the University of Turin and the University of Genova, published in the Journal of Neuro-Oncology, has delivered a finding that could reshape how these fluorescent tracers are chosen in the operating room. The researchers compared the two most widely used fluorophores, 5-aminolevulinic acid (5-ALA) and sodium fluorescein (SF), in 154 adult patients with IDH-wildtype glioblastoma, and discovered something unexpected: although the dyes performed almost identically when it came to how much tumor was removed, patients who received 5-ALA lived significantly longer.

Glioblastoma is a formidable surgical challenge precisely because it does not respect boundaries. On magnetic resonance imaging, the tumor presents as two distinct zones. The first is the contrast-enhancing core, where the blood-brain barrier has broken down and gadolinium contrast agent leaks in, highlighting the densest tumor tissue. The second is the larger, hazier region that appears bright on FLAIR sequences, a combination of infiltrating tumor cells, edema, and altered brain tissue that extends well beyond the enhancing rim. Decades of evidence, from the landmark 2006 randomized trial of Stummer and colleagues onward, have established that removing more of the enhancing tumor prolongs survival. More recently, a growing body of research has explored whether resecting into the FLAIR-hyperintense zone, sometimes called FLAIRectomy, offers additional benefit.

This is where fluorescent dyes enter the picture. 5-ALA is a prodrug that patients swallow before surgery; tumor cells metabolize it into protoporphyrin IX, a fluorescent molecule that accumulates preferentially in malignant tissue with high metabolic activity. Under blue-violet illumination, that tissue glows vivid pink-red, giving surgeons a biological signal of where tumor cells are actively proliferating. Sodium fluorescein, by contrast, is an intravenous dye that simply leaks through the disrupted blood-brain barrier, staining whatever tissue has a broken-down vascular barrier in bright green. In principle, 5-ALA marks biology while fluorescein marks barrier disruption, and the Turin-Genova team wanted to know whether that mechanistic difference translates into measurable differences in how much FLAIR-bright tissue gets removed, and whether patients survive longer as a result.

The study design was straightforward but rigorous for a retrospective analysis. Between January 2021 and January 2023, 154 adults underwent resection of supratentorial glioblastoma at the single center. Which fluorophore a patient received depended not on clinical judgment but on institutional drug availability over the study period, a quirk that actually strengthens the analysis by approximating a natural experiment. Sixty-three patients received 5-ALA, sixty-nine received sodium fluorescein, and twenty-two received both agents together. The researchers then performed volumetric analysis using semi-automated segmentation on preoperative and early postoperative MRI, calculating the extent of resection separately for the contrast-enhancing component, the FLAIR component, and the total abnormal signal. Survival outcomes were assessed with Kaplan-Meier curves and Cox proportional hazards models, the standard statistical machinery for time-to-event data in oncology.

The first surprise was what the researchers did not find. There were no statistically significant differences between the fluorophore groups in the extent of resection of the contrast-enhancing tumor (p = 0.298), the FLAIR-hyperintense volume (p = 0.060), or the total abnormal signal (p = 0.293). In other words, whichever dye lit up the operating field, surgeons ended up removing roughly the same amount of visible abnormality. The near-significant trend in FLAIR resection hinted at possible differences, but the headline volumetric numbers were essentially a tie. If the choice of dye does not change how much tissue comes out, one might reasonably expect no difference in survival either.

That expectation collapsed when the survival data came in. Patients who received 5-ALA experienced a median progression-free survival of 8.5 months compared with 6.0 months for those who received sodium fluorescein, a difference that corresponds to a hazard ratio of 1.89 (95% confidence interval 1.26 to 2.81, p = 0.002). Overall survival told an even starker story: a median of 17.0 months with 5-ALA versus 11.0 months with fluorescein, with a hazard ratio of 2.32 (95% confidence interval 1.33 to 4.04, p = 0.003). In a disease where median survival is measured in months, a six-month difference in overall survival is not a statistical curiosity; it is a clinically meaningful gap that most experimental therapies would celebrate.

Skeptics of retrospective studies will immediately ask whether the 5-ALA group was simply healthier, and the authors anticipated that objection. The survival advantage remained statistically significant after adjustment for the classic prognostic factors in glioblastoma: age, preoperative Karnofsky Performance Status, MGMT promoter methylation status, and adjuvant treatment. More strikingly, the association persisted even in multivariable models that accounted for residual tumor volumes, the extent of resection treated as a continuous variable, and the RANO resect category, a validated classification system for resection outcomes in glioblastoma. If the amount of tissue removed was the same, and the amount of tissue left behind was the same, and the patients were otherwise comparable, what could explain the survival difference?

The authors offer a provocative hypothesis: a more biologically selective resection within the FLAIR-hyperintense region. Because 5-ALA fluorescence depends on metabolic activity within tumor cells rather than on passive leakage through a leaky barrier, it may guide surgeons toward removing the portions of the FLAIR zone that actually harbor viable, proliferating tumor, while sparing tissue that is merely edematous. Sodium fluorescein, staining everything behind a disrupted blood-brain barrier, may be less discriminating, leaving behind metabolically active tumor cells scattered through non-fluorescent FLAIR tissue even when the total resected volume looks identical on postoperative imaging. Two surgeries can remove the same number of cubic centimeters and yet differ profoundly in how many tumor cells remain. This hypothesis, the researchers are careful to note, remains to be tested prospectively, and the retrospective, single-center design means residual confounding cannot be excluded.

The findings land in a field that is actively debating how far surgery should go. The RANO resect group has worked to standardize the radiological definition of supramaximal resection, and studies combining histopathology with advanced imaging, including work showing that [18F]DOPA PET identifies metabolically active tumor beyond contrast-enhancing margins, have reinforced the idea that the FLAIR zone is not a monolith. Some regions contain dense infiltrative disease; others contain little more than vasogenic edema. A dye that distinguishes between them at the bedside would be a genuine advance, and the new data suggest 5-ALA may already be doing something along those lines, even if the mechanism has not been definitively proven. Meanwhile, combined approaches using both dyes are being explored in lower-grade gliomas and in technical series of over one hundred glioblastoma cases, raising the possibility that dual fluorescence could one day capture the strengths of both agents.

Cautious interpretation is still essential. The study is retrospective, from a single institution, and the choice of dye was driven by drug availability rather than randomization, so unmeasured differences between the groups, including temporal changes in care over the two-year window, could contribute to the survival gap. The combined 5-ALA plus fluorescein group was small, with only twenty-two patients, limiting the power of subgroup comparisons. Yet the internal consistency of the result, its persistence across multiple adjustment models, and its alignment with a plausible biological mechanism make it difficult to dismiss. For neurosurgeons choosing a fluorophore tomorrow morning, the message is that equal resection volumes may not mean equal oncological outcomes, and that the biology of the dye matters as much as the brightness of the glow. For patients with glioblastoma, a disease that has stubbornly resisted every therapeutic advance of the past two decades, the suggestion that a simple preoperative choice could add months of survival is a rare and welcome piece of good news, and a compelling argument for the prospective trials that must now follow.

Subject of Research: Fluorescence-guided resection of IDH-wildtype glioblastoma using 5-aminolevulinic acid versus sodium fluorescein

Article Title: Impact of 5-ALA and sodium fluorescein on FLAIR volume extent of resection and survival in IDH-wildtype glioblastoma

Article References: Margherita, C., Pietro, Z., Andrea, B., Leonardo, B., Lorenzo, D., Francesco, B., Melcarne, A., Giovanni, M., Roberta, R., Diego, G., & Fabio, C. (2026). Impact of 5-ALA and sodium fluorescein on FLAIR volume extent of resection and survival in IDH-wildtype glioblastoma. Journal of Neuro-Oncology, 179(3), Article 86. https://doi.org/10.1007/s11060-026-05802-1

Image Credits: AI Generated

DOI: 10.1007/s11060-026-05802-1

Keywords: glioblastoma, 5-aminolevulinic acid, sodium fluorescein, fluorescence-guided surgery, extent of resection, FLAIR, IDH-wildtype, progression-free survival, overall survival, neurosurgery, neuro-oncology, MRI volumetry