Table 2.

Relative risk of recurrence associated with GRP78 expression

Treatment characteristicsGRP78nEvents *Univariate analysisMultivariable analysis
HR (95% CI)PHR (95% CI)P
Overall analysis
    Total subjectsNegative42711
Positive85241.78 (0.77-4.14)0.161.76 (0.74-4.17)0.18
Subgroup analyses
        Adriamycin based §Negative34411
Positive68233.00 (1.04-8.70)0.0223.00 (1.02-8.84)0.026
        Taxanes added Negative8311
Positive1710.15 (0.016-1.46)0.0720.24 (0.020-3.00)0.24
        P for interaction (GRP78 and chemotherapy)0.0120.012
    Surgery type
        Segmental mastectomyNegative14411
Positive2150.74 (0.20-2.77)0.660.53 (0.11-2.54)0.42
Positive64193.33 (0.98-11.30)0.0272.53 (0.73-8.75)0.11
        P for interaction (GRP78 and Surgery)0.0780.28
  • * Number of recurrences.

  • Stratified analysis using propensity score (based on tumor stage, lymph node status, and grade) divided into quintiles.

  • Ps from likelihood ratio test based on Cox model.

  • § Adriamycin with one or more cyclophosphamide, 5-fluorouracil, or methotrexate.

  • Adriamycin-based chemotherapy followed by or combined with taxanes.