The correlates of benefit from neoadjuvant chemotherapy before surgery in non-small-cell lung cancer: a metaregression analysis

Table 2 Factors associated with benefit from neoadjuvant chemotherapy in non-small-cell lung cancer

Factors	All trials		Stage 1 to 3 (mixed-stage trials)		Stage 3
	B (95% CI)^a	P value	B (95% CI)	P value	B (95% CI)	P value
Disease characteristics
Histology (squamous cell fraction)	−0.00 (−0.01 to 0.00)	0.600	0.00 (−0.01 to 0.01)	0.536	−0.00 (−0.01 to 0.00)	0.261
Stage (1 to 3 vs 3 only)	−0.06 (−0.17 to 0.05)	0.256	NR^b	NR^b	NR^b	NR^b
Treatment characteristics
Platinum type (carboplatin vs cisplatin based)	−0.07 (−0.15 to 0.04)	0.205	−0.05 (−0.18 to 0.09)	0.411	NR^c	NR^c
Generation of drugs (older vs third-generation agent protocol)	−0.03 (−0.09 to 0.14)	0.639	−0.04 (−0.19 to 0.11)	0.572	−0.09 (−0.16 to 0.34)	0.373
Number of drugs (2 vs 2 to 3 vs 3 drugs)	−0.01 (−0.09 to 0.06)	0.695	−0.03 (−0.04 to 0.11)	0.303	−0.18 (−0.27 to −0.08)	0.006
Number of chemotherapy cycles	0.00 (−0.09 to 0.10)	0.955	−0.04 (−0.19 to 0.11)	0.572	−0.01 (−0.23 to 0.21)	0.871
Trial characteristics
Trial size^d	0.00 (−0.01 to 0.01)	0.569	−0.00 (−0.02 to 0.01)	0.530	0.01 (−0.02 to 0.03)	0.514
Publication year	0.01 (−0.01 to 0.02)	0.605	−0.00 (−0.02 to 0.01)	0.535	0.03 (−0.03 to 0.09)	0.215
Origin of study (European vs Asian vs US)	−0.01 (−0.10 to 0.08)	0.885	0.03 (−0.09 to 0.15)	0.566	0.03 (−0.03 to 0.34)	0.841
Trial quality score^e	0.03 (−0.04 to 0.09)	0.373	0.04 (−0.03 to 0.10)	0.186	0.12 (−0.16 to 0.40)	0.300

^aRegression coefficient and 95% confidence interval.
^bNot relevant.
^cAnalysis not performed due to insufficient number of trials in that subgroup.
^dSquare root transformation. For analyses for trial size, analyses have not been weighed for trial size.
^eOver a scale from 0 to 3, depending on randomization method (details given vs not given), stratification criteria (reported vs not reported) and trial type (reported in full text or abstract).

ISSN: 1477-7819