Table 1 Univariable and multivariable analyses of time to colorectal cancer recurrence and overall survival in pooled VICTOR and QUASAR2 trial population

From: Tumour-infiltrating CD8+ lymphocytes and colorectal cancer recurrence by tumour and nodal stage

 

No.

TTR events

OS events

Univariable analysis

Multivariable analysis

Time to recurrence

Overall survival

Time to recurrence

Overall survival

HR (95% CI)

P

HR (95% CI)

P

HR (95% CI)

P

HR (95% CI)

P

Age (continuous)

1804

435

350

1.01 (1.00–1.02)

0.10

1.03 (1.02–1.04)

6.2 × 10–6

1.00 (0.99–1.01)

0.43

1.02 (1.01–1.04

1.0 × 10−4

Sex

 Male

1083

281

222

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 Female

721

154

128

0.83 (0.68–1.01)

0.063

0.89 (0.72–1.11)

0.31

0.83 (0.68–1.02)

0.075

0.89 (0.71–1.11)

0.30

Location

 Left

972

167

172

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 Right

730

167

158

1.08(0.93–1.31)

0.45

0.77 (0.62–0.96)

0.019

1.05 (0.85–1.31)

0.63

0.83 (0.65–1.06)

0.13

Stage

 II

708

115

91

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 III

1096

320

259

1.89 (1.52–2.40)

6.4 × 10−9

1.85 (1.45–2.35)

5.6 × 10−7

1.96 (1.57–2.44)

1.34 × 10−9

2.03 (1.59–2.60)

1.3 × 10−8

Primary tumour

 pT1-3

1239

248

186

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 pT4

552

191

158

1.96 (1.61–2.37)

1.0 × 10−11

2.12 (1.70–2.61)

1.1 × 10−11

2.14 (1.76–2.60)

3.0 × 10−14

2.20 (1.77–2.74)

1.3 × 10−12

BRAF mutation

 Wild-type

1412

325

250

    

1.0 (ref)

   

 Mutant

194

55

52

1.33 (1.00–1.78)

0.048

1.68 (1.25–2.27)

6.5 × 10−4

1.59 (1.18 - 2.17)

2.7 × 10−3

1.55 (1.12–2.16)

8.0 × 10−3

MMR & POLE status

 MMR-P & POLE  wild-type

1412

357

272

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 MMR-D or   POLE-mutant

230

40

42

0.68 (0.49–0.94)

0.022

1.00 (0.73–1.39)

0.98

0.72 (0.50–1.05)

0.090

0.96 (0.65–1.41)

0.85

Chromosomal instability

 CIN low

550

110

88

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 CIN high

1049

278

214

1.37 (1.10–1.71)

0.0051

1.27 (0.99–1.62)

0.063

1.17 (0.93–1.49)

0.18

1.21 (0.94–1.58)

0.14

Bevacizumab treatment

 No

1222

279

223

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

1.0 (ref)

–

 Yes

582

156

127

1.31 (1.03–1.67)

0.027

1.25 (0.96–1.63)

0.095

1.28 (1.00–1.62)

0.047

1.23 (0.95–1.60)

0.12

CD8+ cell density (continuous, log2 transformed)

1804

435

350

0.90 (0.85–0.95)

1.7 × 10−4

0.91 (0.86–0.98)

3.2 × 10−3

0.92 (0.87–0.97)

3.6 × 10−3

0.93 (0.87–0.99)

0.024

  1. Univariable hazard ratios are derived from complete case analyses. Multivariable-adjusted hazard ratios are adjusted for all other covariables listed, and represent estimates derived from ‘final’ Cox models following stepwise backward elimination of candidate variables that did not contribute to model fit using the likelihood ratio test (NB: forced entry variables and variables significantly associated with CD8+ cell density were not subjected to variable selection). Results from ‘full’ Cox models including all candidate predictors (age, sex, tumour location, disease stage, primary tumour stage, BRAF mutation, KRAS mutation, MMR-D/POLE mutation, CIN, adjuvant chemotherapy, adjuvant bevacizumab and adjuvant rofecoxib), both before and after the addition of CD8+ cell density are provided in Table S5. The addition of CD8+ cell density to the model containing all other covariables was associated with an improvement in model fit in both the ‘final’ Cox model above (Akaike Information Criterion [AIC] = 5634.0 vs. AIC 5640.6; Likelihood ratio test for comparison of nested models: P = 3.6 × 10−3), and the initial, ‘full’ Cox model (Table S3). TTR time to colorectal cancer recurrence, OS overall survival, HR hazard ratio, 95% CI 95% confidence interval, pT pathological tumour (T) stage, MMR DNA mismatch repair, MMR-P mismatch repair proficient, MMR-D mismatch repair deficient, POLE-mutant pathogenic POLE exonuclease domain mutation
Back to article page