Largest LITT Registry Shows Complete Ablation and Smaller Tumors Drive Survival Gains

Why This Study Matters

Laser interstitial thermal therapy (LITT) is a minimally invasive, MRI-guided neurosurgical technique that ablates intracranial tumors, radiation necrosis, and epileptic foci using hyperthermia. It has been adopted across a range of oncologic diagnoses — newly diagnosed and recurrent gliomas, brain metastases, and radiation necrosis — particularly for lesions in deep or high-surgical-risk locations where open craniotomy carries substantial morbidity.

Despite growing use since FDA clearance of the NeuroBlate System, prospective, multicenter, real-world outcomes data at scale have been limited. The LAANTERN registry was established in 2015 to close that gap by systematically capturing procedural, survival, and quality-of-life data from LITT cases across community and academic centers nationwide. This analysis, encompassing 787 tumor patients followed through the registry's 2023 close, is described by its authors as the largest prospective evaluation of LITT for intracranial tumors to date.

Study Design

LAANTERN is a prospective, multicenter, single-arm registry — not a randomized controlled trial — capturing real-world outcomes for patients treated with LITT using the FDA-cleared NeuroBlate System at 25 U.S. institutions between 2015 and 2023. Because it is a postmarket registry without a comparator arm, it was designed to accommodate real-world evidence collection rather than to test a single pre-specified hypothesis; the authors note that some analyses were exploratory and developed after the Data Analysis Plan was finalized.

Each participating site obtained Institutional Review Board approval, and patients provided written consent. Extent of ablation (EOA) was determined by the treating neurosurgeon based on postprocedural imaging and thermal-dose threshold coverage, categorized in bands from less than 10% to 100%. Overall survival and progression-free survival were defined relative to either initial diagnosis (for newly diagnosed primary tumors) or the LITT procedure itself (for recurrent and metastatic disease). Survival was analyzed using Kaplan-Meier methods and Cox proportional hazards modeling; adverse events were independently adjudicated by an external physician safety committee.

Patient Population

Of the 787 patients analyzed, 445 (56.5%) had primary brain tumors (291 recurrent, 120 newly diagnosed, 34 radiation necrosis) and 342 (43.5%) had metastatic lesions (164 recurrent, 163 radiation necrosis, 15 newly diagnosed). Glioblastoma was the most common primary tumor pathology (190 of 445), followed by grade-4 astrocytoma (79 of 445). Biopsy-proven radiation necrosis (163 of 342) was the most common metastatic-cohort pathology, followed by non-small cell lung cancer (57 of 342) and breast cancer (35 of 342).

The median patient age was 59.6 years; 18 patients were pediatric (under 22 years at consent), with a median pediatric age of 15 years. Just over half of primary tumors (51.7%) were classified as deep-seated — locations such as the basal ganglia, thalamus, insula, and corpus callosum that typically carry elevated surgical risk with open approaches.

Comparing 2014–2019 with 2020–2023, procedural efficiency improved over time: mean procedure time fell from 191.8 to 164.2 minutes (P<.0001), ICU utilization dropped from 46.7% to 27.9% (P<.0001), and median length of stay decreased from 33.4 to 31.6 hours (P<.0001).

Primary Outcome Results: Extent of Ablation and Survival

Because LAANTERN is a registry rather than a randomized trial, there is no single primary endpoint in the traditional sense. Instead, the analysis centered on how extent of ablation (EOA) and lesion size relate to overall survival (OS) and progression-free survival (PFS) across disease subtypes. Near-total ablation (91% or greater) was achieved in 75.5% of primary tumors and 85.9% of metastatic lesions.

In conclusion, this largest-to-date prospectively collected cohort of 787 patients supports the consideration of LITT as a minimally invasive and safe cytoreductive tool for select patients with primary and metastatic brain tumors, including radiation necrosis. Associated outcomes are optimized by ablating when lesions are smaller and by maximizing the extent of ablation.
— Study authors, Discussion section

Subgroup Analyses

Lesion size compounded the effect of ablation extent in several disease subtypes. In recurrent metastatic tumors, lesions smaller than 7.9cc (roughly 2.47cm in diameter) had substantially longer median PFS than larger lesions (1.2 vs. 0.5 years, P=.001); no comparable lesion-size effect on PFS was observed for metastatic radiation necrosis (P=.22).

In newly diagnosed grade-4 astrocytoma, patients with ≥91% EOA (n=8) had significantly longer PFS than those with ≤90% EOA (n=4): 0.84 vs. 0.25 years (P=.0065). The corresponding OS comparison was also statistically significant (P=.0004); the source article reports this OS finding only as "0.31 years v NA" without clarifying which ablation group each value belongs to, so it is reproduced here as stated rather than resolved by inference. Given the very small subgroup sizes (n=4 and n=8), this finding should be interpreted with particular caution.

Hazard ratios above 1.0 indicate increased risk (worse survival) relative to the comparator; ratios below 1.0 indicate reduced risk. Several confidence intervals cross 1.0, meaning those individual comparisons did not reach statistical significance even where a trend is apparent.

Safety Profile

The overall per-person adverse event (AE) rate was 12.8% (110 AEs across 101 of 787 patients). Of the 110 AEs recorded, 90 (82%) were neurologic in nature, and 72 (65.5%) resolved completely; 17 (15.5%) were classified as serious by the external safety committee. The per-person risk of severe or irreversible AEs was 2.3% (18 of 787 patients).

Surgical mortality was 0.25% (2 of 787 patients: one death related to edema, one related to hemorrhage). Surgical infection rate was 0.5% (cellulitis, n=1; pneumonia, n=1; wound dehiscence, n=2). Thirty-day readmission rates were 2.7% for primary tumor patients and 3.5% for metastatic patients. No adverse events were reported in the pediatric subgroup (n=18). Longer procedures (greater than 190 minutes) trended toward more complications (15% vs. 10%–11% for shorter procedures, P=.091), and larger lesion volume was associated with more complications in the primary-tumor cohort (16.4% vs. 30% AE rate, P=.0237) but not in the metastatic cohort.

Seizure control also improved markedly: among patients with primary tumors, seizure prevalence fell from 54.1% at baseline to 11.0% at 6 months (P<.0001) and 19.9% at 12 months (P<.0001). Steroid cessation (28 or more consecutive days off steroids) was achieved in 82%–98% of patients depending on cohort, and bevacizumab initiation within 30 days of LITT was minimal across all groups (0%–4.1%).

Interpretation and Broader Context

The authors frame these findings within existing benchmarks: reported median OS for glioblastoma with current standard-of-care therapy is typically 15–18 months, or roughly 2.3 years when complete surgical resection is achieved. In this registry, newly diagnosed GBM patients who achieved near-total ablation reached a comparable median OS of 2.1 years — while those with incomplete ablation fared markedly worse (0.36 years), a gap the authors attribute in part to underlying tumor biology and lesion characteristics rather than treatment failure alone.

With these results, the authors position LITT as a minimally invasive option with a low surgical mortality rate (0.25%) and infection rate (0.5%), particularly suited to deep-seated lesions that carry elevated risk under open craniotomy. They call for further study directly comparing LITT with craniotomy, especially in anatomically challenging regions such as the insula, where clinical equipoise between the two approaches may exist.

Limitations

As a prospective registry without a control arm, the study cannot establish causation the way a randomized trial would, and the authors are explicit about this limitation. Selection bias is a central concern: larger tumors may have been directed toward open surgery, while patients considered poor surgical candidates may have been preferentially directed toward LITT. The ≤90% EOA group is also not monolithic — for grade-4 GBM, 89% of patients in that group actually achieved 51%–90% ablation rather than representing outright ablation failures, and median lesion volumes differed substantially by EOA category (21.5cc for 51%–90% EOA vs. 8.2cc for 100% EOA), meaning baseline tumor size itself may partly explain the survival differences observed.

Other limitations include declining study completion over time (fewer than half of patients completed 1-year follow-up, and only 1%–2% reached 5-year follow-up), incomplete molecular data (MGMT status and some IDH/TERT/EGFR data were not available on all patients), and a Data Analysis Plan that the authors acknowledge was designed for a real-world registry rather than a hypothesis-driven trial, with some analyses developed as ad hoc exploration after data lock.