Table 2 Multivariable analysis of pCR in relation to clinicopathological characteristics in the cohort

From: Cross-modal deep learning model for predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer

Characteristics

OR

95% CI

P-value

Phenotypes

   

 HR+/HER2−

-

-

-

 Triple-negative

7.127

(1.735–29.272)

0.006**

 HR−/HER2+

7.139

(3.792–13.440)

<0.001***

 HR+/HER2+

1

  

ER

   

 Negative

4.171

(2.571–6.769)

<0.001***

 Low positive

8.141

(2.875–23.053)

<0.001***

 Positive

1

  

Ki-67(%)

1.021

(1.010–1.033)

<0.001***

Chemotherapy regimen

   

 Use of anti-HER2 agents in HER2-positive cancer

2.892

(1.356–6.169)

0.006**

 Taxane- and platinum-based

-

-

-

 Taxane-based

-

-

 

 Anthracycline with taxane

1

  
  1. P < 0.05 indicates that the predictive variables are independently associated with pCR.
  2. OR odds ratio, CI confidence interval, ER estrogen receptor.
  3. **P < 0.01; ***P < 0.001.
Back to article page