Skip to main content
Insights into Imaging

Table 8 Change in ADC between baseline and during NST; (i) percentage change, (ii) absolute change, (iii) ADC ratios baseline

From: Factors affecting the value of diffusion-weighted imaging for identifying breast cancer patients with pathological complete response on neoadjuvant systemic therapy: a systematic review

First author

After N cycles

pCR versus non-pCR (mean ± SD%)

ADC percentage change cutoff (%)

AUC

Sens (%)

Spec (%)

PPV (%)

NPV (%)

(i) ΔADC %

Li [44]

1

 

6.5

0.63

50

78

55

 

Pereira [18]

1

Overall: 44.36 ± 6.7 versus 7.54 ± 2.3 p =  < 0.001

      
  

TN: 53 versus 7 p = 0.002

      
  

Luminal B: 42 versus 16 p = 0.009

      
  

HER2-ov.exp: 43 versus 7 p = 0.055

      

Zhang [24]

2

68.2 ± 49.6 versus 10.4 ± 26.3

 

0.877

    

Partridge [23]

3 (= 3 weeks)

Overall: 18 ± 20 versus 16 ± 21; p = 0.48

 

0.53

    
  

HR−/HER2−: 14 ± 15 versus 15 ± 18; p = 0.94

 

0.51

    
  

HR+/HER2−: 22 ± 18 versus 15 ± 22; p = 0.18

 

0.61

    
  

HR−/HER2+: 25 ± 26 versus 32 ± 28; p = 0.52

 

0.61

    
  

HR+/HER2+: 14 ± 23 versus 18 ± 23; p = 0.43

 

0.58

    

Bedair [20]

3

49 versus 21 p = 0.03 → ADC

      
  

Other model based metrics:

      
  

45 versus 32 p = 0.04 → DDC

      
  

36 versus 23 p = 0.14 → D

      
  

− 29 versus 5 p = 0.05 → f

      
  

7 versus 5 p = 0.68 → α

      

Fangberget [65]*

4

54.7 versus 18.5 p = 0.111

      

Partridge [23]

12 (= 12 weeks)

Overall: 50 ± 49 versus 36 ± 44; p = 0.017

 

0.60

    
  

HR−/HER2−: 33 ± 36 versus 26 ± 40; p = 0.33

 

0.57

    
  

HR+/HER2−: 75 ± 43 versus 35 ± 40; p < 0.001

 

0.76

    
  

HR−/HER2+: 63 ± 65 versus 35 ± 57; p = 0.40

 

0.67

    
  

HR+/HER2+: 40 ± 43 versus 56 ± 56; p = 0.53

 

0.56

    

First author

After N cycles

pCR versus non-pCR (× 103 mm2/s)1

ADC change cutoff (× 103 mm2/s)

AUC

Sens (%)

Spec (%)

PPV (%)

NPV (%)

(ii) ΔADC

Yuan [22]**

1

 

Luminal A2: 0.5589

0.845

87.3

73.4

  
   

Luminal B3: 0.5746

0.865

89.4

83.4

  
   

Basal-like4: 0.5854

0.879

89.9

82.6

  
   

HER2 enr.4: n.r.

0.783

n.r.

n.r.

  

Che [19]5

2

− 0.45 (− 0.67, − 0.29) versus 0.07 (− 0.16, − 0.01) p < 0.001

− 0.163

0.924

100

73.7

64.3

100

First author

Time points

pCR versus non-pCR6

ADC ratio cutoff (A.U.)

AUC

Sens (%)

Spec (%)

PPV (%)

NPV (%)

(iii) ADC-ratio of two time points

Ramirez-Galván [25]

1 cycle/pre

1.08 ± 0.4 versus 1.12 ± 0.09

≤ 1.09

0.641

85.9

58.6

  
 

2 cycles/pre

1.30 ± 0.28 versus 1.10 ± 0.10

> 1.14

0.807

79.2

79.7

  
 

3 cycles/pre

1.35 ± 0.28 versus 1.10 ± 0.15

> 1.08

0.826

100

66.7

  
 

Post/pre

1.49 ± 0.20 versus 1.13 ± 0.01

> 1.25

0.938

100

83.8

  
  1. ADC apparent diffusion coefficient, AUC area under the curve, HER2-enr. human epidermal growth factor receptor 2 enriched, HR hormone receptor, pCR pathologic complete response, Sens sensitivity, Spec specificity, PPV positive predictive value, NPV negative predictive value, α intravoxel heterogeneity index, Δ representing change
  2. *31 MRI at pre NAC and after 4 cycles 27 MRI’s
  3. **Data in full-text was reported based on different NST (started with taxanes or anthracyclines, or taxanes and anthracyclines) and the molecular subtypes
  4. 1There has been chosen to use the exact numbers (positive and negative) in order to avoid misinterpretation, when definitions are not mentioned in the full-text
  5. Median and interquartile range in change in ADC for Che et al. [19]
  6. 2Compared after 1 cycle with anthracyclines
  7. 3Compared after 1 cycle of taxanes
  8. 4Compared after 1 cycle of anthracyclines and taxanes
  9. 5(Parameter-baseline)–(parameter after two cycles), change in true diffusion (D)
  10. 6Ratio: ADC time point after baseline/ADC baseline
Back to article page