Purpose: Since September 1980 we have been conducting a prospective randomized trial to determine the best treatment schedule for radiation therapy (XRT) on brain metastasis from lung carcinoma. The first trial (September 1980 to December 1984) was randomly allocated by two different time-dose radiotherapy schemes, i.e., 30 Gy/ten fractions/two weeks versus 50 Gy/20 fractions/four weeks. Treatment results showed no significant difference in neurological improvement and survival between the two arms and lactate dehydrogenase (LDH) as the most important prognostic factor. The present study (January 1985 to April 1992) examines two sequential trials stratified by the level of LDH enrolled 162 patients with brain metastasis from lung carcinoma.
Patients and methods: Whole brain dose was selected for 30 Gy/ten fractions/two weeks (group A, n = 46) or 50 Gy/20 fractions/four weeks (group B, n = 46) in the group with normal LDH and 30 Gy/ten fractions/two weeks (group C, n = 35) or 20 Gy/five fractions/one week (group D, n = 35) in the group with high LDH, while the treatment fields were shrunk at 30 Gy in group B if possible.
Results: The final results showed the facts that 1. the most important prognostic factor, according to Cox's multivariate analysis, was also the level of LDH in the second trial, 2. the incidence of acute side effects showed the trend toward depending upon a single dose, i.e., group A (3 Gy/fraction); 35% versus group B (2.5 Gy/fraction); 21% (p = 0.165) and group C (3 Gy/fraction); 23% versus group D (4 Gy/fraction); 45% (p = 0.044), 3. median survival time and one-year survival rates were 5.4 months and 21% in group A; 4.8 months and 17% in group B; 3.4 months and 6% in group C; and 2.4 months and 4% in group D, respectively, and survival curves showed no statistically significant difference between the two treatment groups in each LDH group, 4. improvement in neurologic function appeared to increase with total dosage escalation, i.e., 41% in group A versus 45% in group B and 35% in group C versus 21% in group D (p = 0.13).
Conclusion: A short course (30 Gy/ten fractions/two weeks) is an advantageous XRT because of the short treatment time for normal LDH and neurological improvement and minor toxicity for the high LDH group, while an optional treatment may be necessary for the selected patients.