Article Figures & Data
Figures
- Figure 1.
Workflow for development of a prognostic gene expression signature. From the learning series of 95 stage II and III colorectal cancer samples, a filtered gene expression data set was used as input for survival modeling. From 1,000 iterations of lasso-penalized multivariate modeling, 7 models were reported as optimal for survival prediction more than 50 times (middle bar plot). For each of the gene expression signatures, patients were dichotomized to good and poor prognosis groups according to all the possible stepwise increases in amounts of genes being expressed at levels associated with poor prognosis (bottom). All 28 possible stratifications (pale and dark blue boxes in heatmap) were tested for univariate associations with patient survival, yielding significant associations for 22 (dark blue).
- Figure 2.
Statistical characteristics of genes in the identified prognostic expression signature. The 7 genes in the expression signature (blue) had (A) low P values from univariate Cox proportional hazards analyses (median P = 0.02, Wald test for predictive potential) and (B) high variances in gene expression signals (median 2.4, log2 scale), compared with the remaining 3,091 genes included for survival modeling (gray; median P = 0.3; median variance 0.3). C, the Pearson correlations of expression signals between the 7 genes in the signature were generally weak (absolute values, 0.006–0.55).
- Figure 3.
Survival curves for patients with stage II and III colorectal cancer in the 3 independent series stratified by the 7-gene expression signature. A, patients in the learning series assigned to the poor prognosis group had a 10-year relapse-free survival rate of 9%, significantly poorer than the 62% survival rate for patients in the good prognosis group. The corresponding 5-year survival rates for patients in (B) the test series and (C) external validation series, were 49% compared with 78% and 45% compared with 81%, respectively. All survival curves were calculated by Kaplan–Meier statistics and compared by log-rank tests. Relapse or death from colorectal cancer within the 10 or 5 years of follow-up was regarded events, and patients with no events within the indicated period of follow-up were censored.
- Figure 4.
Survival curves in each stage for patients in the 3 independent series. Patients with stage II (left) and III (right) colorectal cancer were individually stratified according to the 7-gene expression signature. In the learning series, (A) 19% of stage II patients [HR, 6.6 (2.7–16.1)] and (B) 28% of stage III patients [HR, 2.6 (1.2–5.7)] were assigned to the poor prognosis group. In the test series, (C) 16% of stage II patients [HR, 3.3 (0.8–13.3)] and (D) 18% of stage III patients [HR, 2.3 (0.6–8.8)] were assigned to the poor prognosis group. In the external validation series, only (E) 10% of stage II patients were assigned to the poor prognosis group [HR, 1.6 (0.3–6.9)], compared with (F) 21% of stage III patients [HR, 4.1 (2.0–8.2)]. All survival curves were calculated by Kaplan–Meier statistics and compared by log-rank tests. Relapse or death from colorectal cancer within the 10 or 5 years of follow-up was regarded events, and patients with no events within the indicated period of follow-up were censored.
Tables
- Table 1.
Characteristics of the three colorectal cancer sample series
Characteristic Learning series (n = 95) Test series (n = 77) Validation seriesa (n = 215) Age at diagnosis (mean ± SD) 66 ± 11.7 73 ± 13.5 66 ± 13.3 Gender Male 46 33 115 Female 49 44 100 Stage II 52 44 108 III 43 33 107 Location Right 27 46 85 Left 31 20 77 Rectum 37 11 22 Unknown 31 Mean follow-up, y (min–max) All patients 5.9 (0.3–10) b 3.3 (0.2–5) c 3.1 (0.04–5.0) c Patients with event 1.8 (0.3–7.7)b 2.1 (0.7–3.8)c 1.5 (0.1–3.4)c Patients with no event 10 (10–10)b 3.7 (0.2–5.0)c 3.5 (0.04–5.0)c No. of events, stage IId 21 9 13 No. of events, stage IIId 27 10 32 MSI-high 7 14 NA Adjuvant chemotherapy, stage III Yes 0 17 63 No 43 16 28 Unknown 0 0 16 Year of surgery/biobanking 1987–1989 2005–2007 NA ColoGuidePro Positive predictions (true; false) 20; 2 6; 7 16; 18 Negative predictions (true; false) 45; 28 51; 13 152; 29 - Table 2.
Prognostic stratification of stage II and III colorectal cancer patients by clinical parameters and the 7-gene expression classifier ColoGuidePro
Test series Validation seriesa Univariate Multivariate Univariate Multivariate Parameter HRb (95% CI) Pc HRb (95% CI) Pc HRb (95% CI) Pc HRb (95% CI) Pc Gender (male vs. female) 1.3 (0.5–3.3) 0.5 1.3 (0.5–3.5) 0.6 1.1 (0.6–2.0) 0.8 0.9 (0.5–1.7) 0.8 Age at diagnosis (≥70 vs. <70 y) 0.8 (0.3–1.9) 0.6 0.6 (0.2–2.1) 0.5 0.8 (0.4–1.4) 0.4 1.1 (0.6–2.1) 0.8 Tumor stage (III vs. II) 1.8 (0.7–4.3) 0.2 1.3 (0.4–4.1) 0.7 2.7 (1.4–5.2) 0.002 2.6 (1.3–5.4) 0.009 Tumor location (right vs. left and rectum) 0.3 (0.1–0.9) 0.03 0.4 (0.2–1.2) 0.1 0.8 (0.4–1.4) 0.4 0.8 (0.4–1.4) 0.4 MSI (MSI-high vs. MSI-low and MSS) 0.2 (0.03–1.4) 0.1 0.4 (0.04–3.1) 0.4 NA NA NA NA Adjuvant chemotherapy (yes vs. no) 1.1 (0.4–3.0) 0.9 0.6 (0.1–2.6) 0.5 1.5 (0.8–2.7) 0.2 0.9 (0.5–1.8) 0.8 ColoGuidePro 2.9 (1.1–7.5) 0.03 3.2 (1.1–9.3) 0.03 3.7 (2.0–6.8) <0.001 3.1 (1.6–5.8) 0.001 Abbreviations: MSI, microsatellite instability; MSS, microsatellite-stable; NA, not available.
-
↵aGEO accession numbers GSE14333 and GSE17538 (n = 215).
-
↵bHRs and corresponding 95% CIs from univariate or multivariate Cox proportional hazards analysis as indicated. Event is relapse or death from colorectal cancer within 5 years. Censoring is no event or lost to follow-up within 5 years.
-
↵cP values from Wald test of predictive potential.
-
Supplementary Data
Files in this Data Supplement:
- Supplementary Methods, Tables 1-2, Figures 1-4 - PDF file - 292K, The Supplementary Material contains Supplementary Methods, Supplementary Tables S1 and S2, and Supplementary Figures S1-S4 Table S1. Genes in the prognostic expression signature Table S2. Cox proportional hazards analyses for patients in the learning series Figure S1. Cross-validated partial likelihood and number of active predictors in the learning series as a function of the penalty parameter Figure S2. Expression in the learning series of the genes in the 7-gene prognostic signature. Figure S3. Survival curves for stage III patients in the validation series. Figure S4. Survival curves for stage III patients in the validation series aged 75 years or older
Volume 18, Issue 21
