EORTC 08993

Background

Phase 3 RCT (EORTC 08993). 286 patients (143 per arm) with extensive-stage SCLC who had any response to chemotherapy. Enrolled across EORTC institutions. Unlike the Auperin meta-analysis (limited-stage CR patients), this trial specifically addressed PCI in extensive SCLC responders. Brain imaging was not required before enrollment — a key limitation.

Interventions and follow up

Arm A: PCI 20 Gy in 8 fractions or 24–30 Gy (physician discretion) + follow-u
Arm B: No PCI + follow-u
Primary endpoint: Time to symptomatic brain metastase
mFollow up: 12+ month

Results

1-yr symptomatic brain met risk: 14.6% (PCI) vs 40.4% (no PCI), HR 0.27 (95% CI 0.16–0.44), P<.001
Median OS: 6.7 months (PCI) vs 5.4 months (no PCI)
1-yr OS: 27.1% vs 13.3%
Median disease-free survival: 14.7 vs 12.0 weeks

Adverse events

Main adverse events: PCI had side effects including fatigue, headache, and nausea but did not have a clinically significant effect on global health status at 3 months. No systematic neurocognitive assessment was performed. Some patients reported hair loss and memory concerns not formally assessed.

Conclusions

PCI significantly reduced symptomatic brain metastases (HR 0.27) and improved both OS and disease-free survival in extensive SCLC responders. This was the first phase 3 evidence supporting PCI specifically in extensive SCLC, leading to its widespread adoption.

Key Limitations

Key Limitations: Brain imaging (CT or MRI) was NOT required before enrollment — patients with undetected brain metastases at entry may have been included, inflating the apparent PCI "benefit" and biasing toward the treatment arm. The Takahashi 2017 Japanese phase 3 trial (which required MRI before enrollment) found no OS benefit for PCI in ES-SCLC when patients had confirmed absence of brain metastases on MRI and underwent regular MRI surveillance. The PCI doses used were lower than those in the Auperin meta-analysis. OS benefit was modest (6.7 vs 5.4 months) and not the primary endpoint.

Clinical Context

EORTC 08993 was pivotal in establishing PCI for extensive SCLC and drove guideline adoption worldwide. However, the Takahashi 2017 trial fundamentally challenged this practice: in Japanese ES-SCLC patients with confirmed MRI-negative brains who received regular MRI surveillance, PCI showed NO OS benefit (HR 1.27, p=0.094, actually trending toward harm). Current guidelines now allow either PCI or MRI surveillance every 2–3 months for extensive SCLC patients; patient preference and neurocognitive concerns are factored into the shared decision-making process.

References

References: Slotman B et al, N Engl J Med 2007 (EORTC 08993)