Article Figures & Data
Figures
- Fig. 1.
Plot of natural logarithm of CYFRA before treatment on the horizontal axis versus the natural logarithm of CYFRA after the first cycle of treatment on the vertical axis. The line shows where the points should fall if the two levels are the same.
- Fig. 2.
Four Kaplan-Meier plots of survival probability versus time of follow-up in months. Top left plot, the survival stratified by clinical stage. Although the top two curves with sparse numbers of patients are for stages IIIa and IIIb, the bottom curve is for stage IV. Top right plot, the survival stratified by clinically determined response. Although the top curve is for those with PR, the bottom curve is for those without any response. Bottom left plot, the survival stratified by pretreatment level of CYFRA. Although the top curve is for those with CYFRA less than or equal to the mean (after logarithm transformation) value of 3.9 ng/ml, the bottom curve is for those with greater levels. Bottom right plot, the survival stratified by the change between pretreatment and post-treatment levels of CYFRA. Although the top curve is for those with a >27% drop in CYFRA such that the magnitude of the log(y2/y1) was less than or equal to the mean of −0.3117, the bottom curve is for those smaller relative decreases.
- Fig. 3.
Kaplan-Meier plots of estimated survival versus time in months and stratified by low versus high HS, i.e., HS < 0.284 versus HS > 0.284.
Tables
- Table 1
Significant associations between serum level of CYFRA 21-1 and stage, surgical resection, and survival of lung cancera
Author Year No. Stage Surgery Survival Pujol ’93 165 Yes Yes Stieber ’93 200 Yes Ebert ’94 177 No Yes van der Gaast ’94 212 Yes Niklinski ’94 115 Yes No Bombardieri ’94 496 Yes Molina ’94 136 Yes Takada ’95 149 Yes Wieskopf ’95 116 Yes Yes Moro ’95 105 Yes Yes Niklinski ’95 57 Yes Yes Niklinski ’95 76 Yes Yes Muraki ’96 114 Yes Lai ’96 139 Yes Ebert ’96 108 Yes Pujol ’96 314 Yes Yes Niklinski ’96 91 Yes Yes Szturmowicz ’96 78 No No Brechot ’97 116 Yes Yes Ebert ’97 129 Yes Takei ’97 70 Yes Yes Huang ’97 87 Yes Yes Nisman ’98 94 Yes Yes -
a Significant increases in stage with increasing levels of CYFRA 21-1 are denoted under the stage column by a “Yes.” Significant drops in CYFRA 21-1 after surgical resection are denoted under the surgery column by a “Yes.” Significant decreases in survival with higher levels of pretreatment CYFRA 21-1 are denoted under the survival column by a “Yes.”
-
- Table 2
Patient characteristics
No. of patients Institution Washington University 35 Syracuse Upstate Medical University/VAMC 10 Duke University 8 Durham VAMC 5 Gender Males: 37 Females: 21 Age (years) Mean: 65 Range: 38–83 Stage IIIA: 14 IIIB: 8 IV: 36 Tumor histology Adenocarcinoma: 18 Adenosquamous: 2 Large cell neuroendocrine: 2 Large cell: 19 Squamous: 17 No. of patients Chemotherapy given Carboplatin + paclitaxel 29 Carboplatin + gemcitabine 8 Carboplatin + vinorelbine 5 Cisplatin + etoposide 1 Gemcitabine + docetaxel 2 Gemcitabine + irinotecan 1 Vinorelbine + docetaxel 2 Gemcitabine 4 Vinorelbine 4 Paclitaxel 2 Distribution of responses PR: 10 Stable: 29 Progression: 19 Status at last follow-up Alive with tumor: 23 Dead of tumor: 35 Follow-up time (months) Median: 8.1 Range: 1.3–20.6 - Table 3
Cox model analysis of overall survival timea
Coefficient SE P Stage IIIa 0.14 Stage IIIb 0.16 Response (PR vs. none) 0.18 Log(y1) 0.61 0.15 0.000032 >27% Decrease in CYFRA −1.14 0.41 0.0052 -
a For this analysis, there were 38 uncensored patients. The remaining patients were either lost to follow-up or alive at the time of analysis. Stage IV was used as the default stage, and stages IIIa and IIIb were coded as dummy variables. Log(y1) is the natural logarithm of the pretreatment level of CYFRA. The coefficients and SE are provided for just a two-variable model that included just the last two variables, because the other three were not found to be significant.
-
Volume 9, Issue 5
