A randomized trial of prophylactic cranial irradiation (PCI) versus no PCI in extensive disease small cell lung cancer after a response to chemotherapy (EORTC 08993-22993)

Reviewer: Christopher Dolinsky, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 6, 2007

Presenter: Bernard Slotman
Presenter's Affiliation: EORTC Radiation Oncology and Lung Cancer Groups
Type of Session: Plenary


  • At diagnosis, patients with small cell lung cancer have a high rate of brain metastasis (around 20%) and this number rises to >50% at 2 years.
  • Brain metastases have a major impact on physical and psychological functioning.
  • After they have developed, brain metastases generally have a poor response to systemic therapy and brain radiotherapy.
  • Prophylactic cranial irradiation (PCI) reduces the risk of brain metastases in limited stage (LD) SCLC and does not lead to an increase in the risk of late neuropsychological dysfunction.
  • PCI improves survival in LD SCLC patients in complete remission following local therapy (Auperin et al. 1999).
  • This study was performed to determine if PCI impacts outcomes in patients with extensive stage small cell and have responded to radiation therapy

Materials and Methods

  • Patients with extensive stage SCLC (ED) received 4 to 6 cycles of chemotherapy and if they had any response to chemotherapy, were randomized to either PCI or no PCI.
  • Patients were stratified by institute and performance status.
  • PCI was given to doses of 20 to 30 Gy in 5 to 12 daily fractions.
  • Eligibility criteria included: WHO performance status of 0-2, documented ED-SCLC before start of chemotherapy, any response after 4 to 6 cycles of chemotherapy, maximum of 5 weeks between chemotherapy and randomization, no evidence of brain or leptomeningeal metastases, no use of corticosteroids, no prior radiotherapy to the head and neck and no prior and/or current other malignancy.
  • CT or MRI was not required prior to randomization unless there was at least 1 key symptom suggesting brain metastasis such as: headache, nausea/vomiting, signs of increased intracranial pressure, cognitive and/or affective disturbances, seizures or focal neurologic symptoms.
  • 286 patients were accrued between February 2001 and March 2006.
  • There were no significant differences between the two arms in terms of patient related or disease specific prognosticators.
  • At the time of randomization, both arms had similar percentages of patients who had persistent primary disease (76% vs. 77%, p=NS) or metastatic disease (70% vs. 73%, p=NS).
  • 94% of patients who were supposed to get PCI received full treatment, and 1 patient in the control arm received PCI after demanding it.
  • About 2/3 of patients in the PCI arm received 20 Gy in 5 fractions, while the remainder of patients were split between 30 Gy in 10 fractions and 24-30 Gy in 8-12 fractions.


  • Acute toxicity was generally mild in the PCI arm with headache, nausea/vomiting, fatigue and skin reactions being the most common.
  • Headache was the only grade 3 acute toxicity seen, and it was reported in 3% of cases.
  • Late radiation toxicities were seen in 35% of patients, with only 2.2% having grade 3 toxicity (severe headache, severe CNS dysfunction).
  • 89% of patients were followed until either progression or death.
  • Symptomatic brain metastases were the first type of recurrent event in 9% of the PCI arm and 35% of the control arm.
  • PCI significantly reduced the rate of symptomatic brain metastasis (1 year rate 15% vs. 40%, HR 0.27, p<0.001).
  • There was no significant difference seen in the rate of extracranial progression between the two arms.
  • PCI significantly improved failure free survival (6 months 23.4% vs. 15.5%, HR 0.76, p=0.02).
  • PCI significantly improved overall survival (1 year 27.1% vs. 13.3%, HR 0.68, p=0.003).

Author's Conclusions

  • PCI significantly reduces the risk of symptomatic brain metastases.
  • No difference was seen for the time to extra-cranial progression.
  • PCI significantly prolongs failure free survival and overall survival.
  • PCI is well tolerated.
  • Patients with extensive stage small cell lung cancer who respond to chemotherapy should routinely be offered PCI.

Clinical/Scientific Implications

The authors presented a well designed and executed phase III randomized trial. These results have really dropped a bombshell. Previous data demonstrated a clear survival benefit with PCI in LD SCLC by about 5%. Outcomes for therapy in this disease have been fairly dismal, and ED SCLC has an awful prognosis. This trial showed a doubling of overall survival with a therapy that (at least in the field of oncology) is cheap, safe, and easy to deliver. In fact, toxicities may even be reduced beyond the results described in this study by using more protracted radiation schedules. 20 Gy in 5 fractions is a little more aggressive than most American clinicians would employ because of the risk of late effects. This trial should change the standard of care in this disease, and the authors are correct when they conclude that patients with extensive stage small cell lung cancer should routinely receive PCI.

Frequently Asked Questions