Reviewers: John Han-Chih Chang, MD and Kenneth Blank, MD
Source: Journal of Clinical Oncology 1997, Volume 15 (9): pages 3030 - 37.
There are a handful of prospective studies looking at the timing of thoracic or chest irradiation (CI) in the treatment of limited stage small cell lung cancer (LS-SCLCa). Add this one to that list, but be careful about the conclusions you draw from it. To recap, the Cancer and Leukemia Group B (CALGB), National Cancer Institute - Canada (NCI - Canada), Komaki et al, and Jeremic et al all ran trials looking at the timing of CI. In 1987, Perry et al from the CALGB published the results from a 3 arm trial looking at chemotherapy (CTX) with cytoxan, vincristine and VP16/Adria CI (either early with the first cycle of CTX or late beginning with the fourth cycle). There was no difference in the local control (LC) and 3 year overall survival (OS) in giving the CI early or late. Murray et al from the NCI - Canada published their data in 1993. Patients were randomized either to receive CI with the 2nd cycle of CTX (consisting of Cytoxan/ Ariamycin/Vincristine) or later with the 6th cycle of CTX. Median progression free survival (PFS) and OS were statistically significant along with the incidence of brain metastases. All the data was in favor of CI earlier rather than later. Komaki et al performed two consecutive prospective trials of interdigitating and then concurrent CTX/radiation therapy (XRT) and published in 1995. There was overwhelming evidence of concurrent cisplatinum and VP16 CTX/XRT was much more effective in LC, DM, brain metastases and OS than the series of patients that got cytoxan, vincristine, cisplatinum, and VP16 CTX/XRT alternating weekly. Jeremic et al published his data on a prospective randomized trial comparing CI early (with 1st cycle of carboplatinum and VP16 CTX) versus late (with the third cycle of CTX). There was a significant LC benefit and a trend towards a significant improvement in OS with CI early.
The data from the Aarhus Lung Cancer Group from Denmark performed a randomized trial again comparing the sequencing of CTX/XRT for the treatment of LS-SCLCa were presented in this article.
Nearly 200 patients were entered into the study from 1981 to 1989 and randomized to receive initial CI (ICI) or CI later (LCI). The actual sequence of the treatments was shown in figure 1 in the article. It revealed that the CI was given in a split course fashion and never concurrent with the CTX. A dose of 4000 or 4500 cGy was delivered total with half given over a two week period. These periods of CI were separated by 21 days in which a cycle of CTX was delivered. Thus, it usually required approximately 50 days to complete the CI. ICI patients had their 1st 2000 or 2250cGy of CI prior to any CTX. LCI patients had their 1st period of CI after the 6th cycle of CTX. The dose of CI was changed from 4000 to 4500cGy after Oct 1984. Prophalactic cranial irradiation (PCI) was given to all patients in the ICI group and most of the LCI group (after Oct 1984) at the beginning of treatment. The CTX consisted of 3 cycles of VP16/cisplatinum (EP) and 6 cycles of cytoxan/adriamycin/vincristine (CAV). EP was given as the 1st two cycles and then as the 7th cycle. The CAV cycles occupy the rest of the every 21 day schedule. EP is what is given between the two CI periods.
Patients scheduled to get CI, only 75 -78% received the prescribed dose. Median survival was 10.5 months for ICI and 12 months for LCI (p = NS). No significant difference in incidence of distant metastases or in field recurrence between the two arms. The increase in dose of 4000 to 4500cGy also yielded no evidence of benefit. The toxicity profile included hematological, neuro-toxicity and cardio-toxicicty. No grade 5 (fatal) complications were demonstrated.
This trial does little to answer the question of early versus late CI in combination with chemotherapy. This was a trial of early versus late sequential CI combined with CTX. We already have knowledge that this is inferior to concurrent CTX/XRT in Komaki et al and the fact that it appears that CALGB and NCI - Canada (concurrent regimines) had higher median survivals than both arms of this Danish trial. The dose of CI was substandard, and treatment time was protracted. One quarter of the patients did not achieve the total dose of prescribed CI, but most were due to intercurrent disease or death due to disease. Treatment time for 4500cGy to chest would normally require 5 weeks of 5 days a week XRT. The Danish group describe a protracted delayed course of XRT which may be affecting the outcome. Therefore, we concluded that from this Danish trial that there is no difference between early and late sequential CTX/XRT in terms of OS or LC, but both methods appear less effective as concurrent CTX/XRT.