Skip to main content

Table 3 Colorectal cancer patient mortality according to KRAS mutation status in 1067 BRAF -wild-type cases

From: Analyses of clinicopathological, molecular, and prognostic associations of KRAS codon 61 and codon 146 mutations in colorectal cancer: cohort study and literature review

KRAS

Total No.

Colorectal cancer-specific mortality

Overall mortality

No. of events

Univariate HR (95% CI)

Multivariate stage-stratified HR (95% CI)

No. of events

Univariate HR (95% CI)

Multivariate stage-stratified HR (95% CI)

Wild-type (codons 12, 13, 61 and 146)

582

144

1 (referent)

1 (referent)

258

1 (referent)

1 (referent)

All mutants together

485

163

1.46 (1.17-1.83)

1.19 (0.94-1.51)

256

1.32 (1.11-1.57)

1.14 (0.95-1.38)

   

P = 0.0009

  

P = 0.0018

 

Codons 12 and 13, and codons 61 and 146

       

Codons 12 and 13

434

151

1.51 (1.20-1.90)

1.27 (0.99-1.62)

235

1.36 (1.14-1.62)

1.15 (0.95-1.40)

   

P = 0.0004

  

P = 0.0007

 

Codons 61 and 146

51

12

1.02 (0.57-1.85)

0.85 (0.47-1.56)

21

1.00 (0.64-1.56)

1.07 (0.68-1.68)

Codons 12, 13, 61 and 146

       

Codon 12 mutants

328

121

1.64 (1.29-2.09)

1.45 (1.12-1.87)

183

1.45 (1.20-1.76)

1.24 (1.01-1.52)

   

P < 0.0001

P = 0.0048

 

P = 0.0001

P = 0.037

Codon 13 mutants

106

30

1.16 (0.78-1.72)

0.83 (0.55-1.25)

52

1.11 (0.82-1.49)

0.90 (0.66-1.24)

Codon 61 mutants

16

4

1.11 (0.41-3.01)

0.81 (0.29-2.26)

8

1.43 (0.71-2.90)

1.55 (0.75-3.18)

Codon 146 mutants

35

8

0.98 (0.48-2.01)

0.86 (0.42-1.78)

13

0.84 (0.48-1.48)

0.88 (0.50-1.56)

The 10 most common mutations in codons 12, 13, 61 and 146

    

c.34G>A (p.G12S)

12

6

2.44 (1.07-5.54)

0.94 (0.39-2.23)

7

1.57 (0.74-3.33)

0.77 (0.35-1.70)

   

P = 0.033

    

c.34G>C (p.G12R)

7

5

5.25 (2.13-12.9)

3.44 (1.25-9.43)

6

4.69 (2.06-10.6)

3.51 (1.42-8.70)

   

P = 0.0003

P = 0.017

 

P = 0.0002

P = 0.0067

c.34G>T (p.G12C)

42

16

1.70 (1.01-2.86)

2.33 (1.36-3.99)

25

1.56 (1.03-2.35)

1.57 (1.02-2.42)

   

P = 0.044

P = 0.0021

 

P = 0.035

P = 0.039

c.35G>A (p.G12D)

155

51

1.46 (1.06-2.01)

1.18 (0.84-1.66)

80

1.37 (1.06-1.76)

1.16 (0.89-1.51)

   

P = 0.021

  

P = 0.015

 

c.35G>C (p.G12A)

19

6

1.28 (0.56-2.90)

0.61 (0.26-1.42)

9

1.00 (0.51-1.95)

0.59 (0.30-1.17)

c.35G>T (p.G12V)

92

37

1.76 (1.22-2.52)

2.13 (1.47-3.09)

56

1.54 (1.16-2.06)

1.54 (1.14-2.08)

   

P = 0.0024

P < 0.0001

 

P = 0.0033

P = 0.0048

c.38G>A (p.G13D)

101

30

1.23 (0.83-1.82)

0.83 (0.55-1.26)

50

1.14 (0.84-1.54)

0.91 (0.66-1.25)

c.183A>C (p.Q61H)

7

2

1.06 (0.26-4.28)

0.46 (0.11-1.93)

4

1.28 (0.48-3.45)

1.16 (0.42-3.18)

c.436G>A (p.A146T)

21

3

0.55 (0.17-1.71)

0.50 (0.16-1.59)

5

0.45 (0.19-1.10)

0.51 (0.21-1.26)

c.437C>T (p.A146V)

11

4

1.94 (0.72-5.26)

1.77 (0.64-4.90)

7

2.02 (0.95-4.29)

2.10 (0.97-4.56)

  1. The multivariate, stage-stratified Cox regression model initially included sex, age, body mass index, year of diagnosis, family history of colorectal cancer, tumor location, tumor differentiation, peritumoral lymphocytic reaction, microsatellite instability, CpG island methylator phenotype, PIK3CA mutation, and LINE-1 methylation. A backward elimination with a threshold of P = 0.20 was used to select variables in the final model.
  2. For the survival analysis of mutations in the two groups of KRAS codons (codons 12 and 13, and codons 61 and 146), the P-value for significance was adjusted for multiple hypothesis testing to P = 0.05/2 = 0.025. Thus, a P-value between 0.05 and 0.025 should be regarded as of borderline significance. For the survival analysis of mutations in the four KRAS codons (12, 13, 61 and 146), the P-value for significance was adjusted for multiple hypothesis testing to P = 0.05/4 = 0.013. Thus, a P-value between 0.05 and 0.013 should be regarded as of borderline significance. For the survival analysis of the 10 most common KRAS mutations, the P-value for significance was adjusted for multiple hypothesis testing to P = 0.05/10 = 0.005. Thus, a P-value between 0.05 and 0.005 should be regarded as of borderline significance. CI, confidence interval; HR, hazard ratio.
Back to article page

Molecular Cancer

ISSN: 1476-4598