close

Вход

Забыли?

вход по аккаунту

?

95

код для вставкиСкачать
2413
Randomized Trials of Radiotherapy Alone
versus Combined Chemotherapy and
Radiotherapy in Stages llla and lllb Illonsmall
Cell Lung Cancer: A Meta-Analysis
M
arino et al.’ have performed a meta-analysis using data extracted
from published randomized trials comparing radiotherapy (RT) to
RT plus chemotherapy (CT) in Stage IIIA and IIIB unresectable nonsmall
cell lung cancer (NSCLC).They identified 14 trials published between 1980
and 1994. The odds ratio (OR) summarizing the results was computed for
1887 patients. The NSCLC Collaborative Group (NSCLCCG) has recently
published a meta-analysis using updated data from individual patients
on the same topic,‘ including 22 trials which accrued patients between
1965 and 1991. The analysis was performed on 3033 patients (2814
deaths). These two approaches differ not only in the trials and patients
included and the method of analysis used, but also in the results that
they produced.
The two studies had nine trials in common. Five trials included in
the Marino et al. meta-analysis were not eligible for inclusion in the
NSCLCCG meta-analysis. Two were considered confounded, one used
different RT doses per arm (Wils et al., trial 2),and in the other, concomitant cisplatin was given in both arms (Wolf et al., trial 2). A further three
trials gave cisplatin only during RT. The NSCLCCG meta-analysis excluded all trials in which CT was given only during RT, to avoid mixingup trials studying the radiosensitizing effect of CT with those studying
the cytotoxic effect. The trials included in the NSCLCCG meta-analysis,
but not in the one by Marino et al. included 9 published between 1985
and 1991 (5 full papers, 4 abstracts), 1 trial published before this date,
and 3 unpublished trials. It is not clear why the 9 trials (1106 patients)
published between 1985 and 1991 were not included in the Marino et al.
study. In the NSCLCCG meta-analysis, the proportion of Stage 111patients
was 83%.
For the 9 trials common to both studies, the number of patients
included in the analysis by Marino et al. was 1332, whereas 1439 patients
were included in the NSCLCCG meta-analysis. In the latter, the analysis
was performed on an intent-to-treat basis using data from all randomized
patients, whereas Marino et al. were restricted to the material available
in the publication. Owing to the limited amount of information in the
publications, a point estimate of survival on all eligible trials was only
possible at 2 years in the Marino et al. meta-analysis. The NSCLCCG
meta-analysis used individual patients’ survival data in a time-to-event
(log rank) test stratified by trial. This approach provides summary statistics relating to the whole survival experience rather than analyses based
at selected fixed points in time.
In the Marino et al. meta-analysis, the pooled OR for death at 2 years
was 0.70 (95% confidence interval (CI):0.5-0.9) for the 1 1 cisplatin-based
trials and 0.82 (0.5-1.3) for the 4 noncisplatin trials. No conclusion could
be drawn from the 3- and 5-year results. In the NSCLCCG meta-analysis,
0 1996 American Cancer Society
2414
CANCER June 1,1996 I Volume 77 / Number 11
the pooled hazard ratio for death was 0.87 (0.79-0.96; P
< 0.005) for the 11 cisplatin-based trials. For the 3 trials
using vinca-alkaloids or etoposide, the hazard ratio was
0.87 (0.70-1.09), and 0.96 for the 8 other noncisplatin
trials. However, the results of the three trial groups were
not significantly different. For the cisplatin-based trials,
the absolute improvement in survival was 4% (95% CI:
1-7%) at 2 year and 2% (1-4%) at 5 years. Using the same
15% survival rate at 2 years for the control group and the
published OR, it was possible to compute3 the absolute
improvement in survival for the Marino et al. meta-analysis as 5% (1-11%). Based on the results of the two metaanalyses, we don’t share the confidence of Marino et al.
on the reliability of meta-analysis based only on the data
presented in published papers. Although both studies
conclude in favor of cisplatin-based CT, the NSCLCCG
showed a smaller effect than the Marino et al. meta-analysis. This difference could not be explained by the inclusion of the 3 trials giving cisplatin only during RT as the
pooled OR for these 3 trials is 0.75 compared with 0.64
for the other cisplatin-based trials. Thus, if these trials
had been excluded from the Marino et al. study, the results would have shown an even larger effect of CT. To
the best of our knowledge, there are 4 other examples
where both meta-analysis based on data extracted from
publications and meta-analysis based on data from individual patients were p e r f ~ r m e d . ~In- ~three c a ~ e s , the
~-~
relative risk of death was overestimated in the literaturebased meta-analysis. In a recent example concerning
breast cancer,6 it was the meta-analysis of updated data
on individual patients with follow-up of 15 years that
showed highly significant results whereas an analysis of
the data extracted from trial publications (before 1990)
would have shown almost no benefit. The mounting evidence of different results obtained by the two approaches
illustrates the need to consider all of the randomized evidence and suggests that meta-analysis based on updated
data from individual patients may be the best method of
obtaining reliable answers to unresolved questions.
Pignon JP, Arriagada R. Role of thoracic radiotherapy in limited-stage small cell lung cancer: quantitative review based
on literature versus meta-analysis based on individual data.
J Clin Oncol 1992;10:1819-20.
5. Marino P, Pampallona S, Preatoni A, Cantoni A, invernizzi
F. Chemotherapy versus supportive care in advanced nonsmall cell lung cancer: results of a meta-analysis of the literature. Chest 1994;106:861-5.
6. Clarke M. Ovarian ablation-Why the Early Breast Cancer
Trialist’ Collaborative Group (EBCTCG) individual patient
data meta-analysis was needed [abstract].Control Clin Trials
1995;16:67S-8S.
4.
Jean-Pierre Pignon, M.D., Ph.D.
Department of Biostatistics
Institut Gustave-Roussy
94805 Villejuif Cedex, France
Lesley A. Stewart, Ph.D.
MRC Cancer Trials Ofice
5 Shaftesbuiy Road
Cambridge, UK
Author Reply
S
election criteria for our trials was reported in the
“Method” section of our paper. Probably, they were
different from Pignon et al.’s. Although we used a different method of analysis, our results were in agreement
with those of Pignon et al.; in particular both studies
conclude in favor of cisplatin-based chemotherapy.
In a previous paper,’ we discussed the potential
sources of bias affecting our meta-analysis of the literature. However, we believe that the bias was rather small
in this setting. Furthermore, the sensitivity analysis performed on the width of the confidence interval for pooled
odds ratio (OR) seems to suggest statistical significance
of the pooled OR at the 5% level. We believe that our
method, cheaper and faster than meta-analysis of the
individual data, can offer useful results when no conclusions are possible from the data of literature.
REFERENCES
I.
2.
3.
Marino P, Preatoni A, Cantoni A. Randomized trials of radiotherapy alone versus combined chemotherapy and radiotherapy in stages IIIa and IIIb nonsmall cell lung cancer: A
meta-analysis. Cancer 1995;76593401.
Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using
updated individual patients data from 52 randomised clinical trials. EM/ 1995;311:899-909.
Stewart LA, Parmar MKB. Meta-analysis of the literature or
meta-analysis of individual patient data-is there a difference? Lancet 1993;341:418-422.
REFERENCE
1.
Marino P, Pampallona S, Preatoni A, Cantoni A, invernizzi
F. Chemotherapy versus supportive care in advanced nonsmall-cell lung cancer. Results of a meta-analysis of the literature. Chest 1994;106:861-5.
Pietro Marino, M.D.
Department of Internal Medicine
University of Milano
S. Paolo Hospital, Milano, Italy
Документ
Категория
Без категории
Просмотров
2
Размер файла
177 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа