Table IIIa-HResponse Regression Modeling

StudyDesign/Outcome/ModelCandidate Predictors/Method for Identifying CandidatesUnivariate Results, Variable (p value)Selected Predictors/Methods for Selecting Predictors for Multivariate ModelInteractions Considered?Multivariate Model Results, Variable (p value)Discrimination/Validation Methods/ResultsCalibration/Goodness of Fit
Adjuvant Chemotherapy
Yang et al. 2003
rec. # 8840
not reported or done
Gusterson et al. 2003; rec# 43690not reported or done
Moliterni et. al. 2003; rec# 10210not reported or done
Colozza et al. 2005; rec. # 3820not reported or done
Pritchard et al. 2006; rec. # 1760not reported or done
Knoop et al. 2005; rec. # 3450not reported or done
Dressler et al. 2005, rec. # 4280not reported or done
Del Mastro et al. 2004, 2005; rec. # 48020not reported or done
Tanner et al. 2006; rec. # 1820not reported or done
Hayes et al. 2007; rec# 47610not reported or done
Martin et al. 2005 rec. # 47650not reported or done
Neoadjuvant Chemotherapy
Learn et al. 2005; rec. # 47640RCT
ORR
logistic regression
T size; nodal status; ER, PR, HER2, & p53 status; method for identifying not describedHER2- ORR:
81%, AC+D
51%, AC,
p<0.05; HER2+
ORR: 78% AC+D,
75%, AC, p=0.79
not specified or describednot mentionedORR, HER2-
adjusted odds ratio, AC+D/AC:
3.5 (95% CI: 1.2–13.0); p<0.05
not describednot mentioned
Arriola et al. 2006; rec # 950prospective series; pCR logistic regressionmethod not specified; ER, PR, Ki67 & HER2 status, tumor grade, histologyHER2 status & pCR:
p=0.03
forward stepwise logistic regressionnot applicable?not reportednot describednot reported
Park et al. 2003; rec # 9960retrospective series; response; linear regressionmethod not specified;
HER2 & topoII amplification
HER2: p=0.013

topoII: p=0.011
not specifiednot applicable?95% responders in HER2/topoII
co-amplified
(n=18),
p=0.038
not describednot reported
Zhang et al. 2003; rec # 9820not reported or done
Tulbah et al. 2002; rec# 11560not reported or done
Tinari et al. 2006; rec # 2300retrospecttive series; tumor response (cPR + MRD); univariate logistic regressionmethod not specified; HER2 & topoII amplification; age, T size, ER & PR status, grade, Ki67 statusHER2: OR = 5.28 (95% CI: 1.57–19.6); p=0.008not reportednot mentionednot reportednot describednot reported
Chemotherapy for Advanced or Metastatic Disease
390, Harris et al 2006not reported or done
5970, Di Leo et al 2004RCT;
ORR;
logistic regression
candidates: KPS, # mets, visceral involvement, HER2 status; selected as known prognostic factorsFor HER2+ pts given docetaxel, OR=3.12 (1.11–8.80), p=.03; i.e., HER2*Tx is stat. signif.Tx, HER2, Karnofsky score, # met sites; visceral involvement; Tx*HER2; Tx*visceral; stat. signif. in univariate analysisInteractions includedOutcome=response
Tx*HER2: OR Gp2/Gp1=3.64 (1.39–9.54); p=.01
(only SS factor). Remains SS after adjust for visceral and Tx*visceral
NRNR
6740, Konecny et al 2004

Grp 1: EC Grp 2: ET
Logistic regressionOR(HER2+ vs HER2-)=2.19
(95%CI:
1.31–3.65)
p=0.003

OR (Grp2 vs Grp 1 for HER2+)=3.65(95%CI: 1.51–8.86); p=0.004) OR (Grp2 vs Grp1 for HER2-)=2.07 (95%CI: 1.10–3.90, p=0.002)

Interaction Tx*HER2: P=0.308
Interactions includedAdjusted interaction Tx*HER2: p=0.256

OR (Grp2 Vs Grp 1 For HER2+)=3.64 (95% CI: 1.48–8.92; P=0.005) OR (Grp2 Vs Grp 1 For HER2-)=1.92(95% CI: 1.01–3.64; P=0.046)
not mentionedyes, but not shown

From: Appendix C. Evidence Data Abstraction Tables

Cover of HER2 Testing to Manage Patients With Breast Cancer or Other Solid Tumors
HER2 Testing to Manage Patients With Breast Cancer or Other Solid Tumors.
Evidence Reports/Technology Assessments, No. 172.
Seidenfeld J, Samson DJ, Rothenberg BM, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.