Texture analysis on MR images helps predicting non-response to NAC in breast cancer

BMC Cancer

Table 2 Association between pathologic responsiveness of breast cancer to NAC and receptor status

Pathologic response	NR	CR	PR	PR + CR	p-value^a
BI-RADS
Mass	12 (31 %)			27 (69 %)	0.51
non Mass	7 (23 %)			23 (77 %)	0.51
Biological markers
ER positivity	15 (29 %)	9 (17 %)	28 (54 %)	37 (71 %)	0.70
PgR positivity	14 (33 %)	4 (10 %)	24 (57 %)	28 (67 %)	0.19
Ki67 > 14 %	11 (21 %)	11 (21 %)	30 (58 %)	41 (79 %)	0.05
HER2 positivity	4 (15 %)	8 (31 %)	14 (54 %)	22 (85 %)	0.09
Subtypes
Luminal A	8 (62 %)	0	5 (38 %)	5 (38 %)	0.005
Luminal B/ HER2 –	4 (16 %)	5 (20 %)	16 (64 %)	21 (84 %)	0.11
Luminal B/HER2 +	3 (20 %)	4 (27 %)	8 (53 %)	12 (80 %)	0.49
Non-luminal/HER2 +	1 (10 %)	4 (40 %)	5 (50 %)	9 (90 %)	0.20
Triple-negative cancer rate	3 (50 %)	1 (17 %)	2 (33 %)	3 (50 %)	0.25
Histological grade
IDC 2	5 (20 %)	3 (12 %)	17 (68 %)	20 (80 %)	0.31
IDC 3	14 (32 %)	11 (25 %)	19 (43 %)	30 (68 %)	0.31

The number and proportions of NR, CR, PR and PR + CR patients with a given feature within all patients having this feature are given. The statistical significance of the relationship between response (NR or PR + CR) and features is then assessed (p-value^a). If a p-value < 0.05 is observed for a given feature, then we can conclude that patients’ response is associated to that feature. If a p-value > 0.05 is observed, then the null hypothesis that there is no association, cannot be rejected. Subtype Luminal A is the only feature showing a significant association with response
BI-RADS breast imaging-reporting and data system, NR non response, CR complete response, PR partial response, ER estrogen receptor, PgR progesterone receptor, Ki67 cellular marker for proliferation based on monoclonal antibody Ki-67, HER2 human epidermal growth factor receptor 2, HR hormone receptor, IDC invasive ductal carcinoma
^aSignificance of the association between response (NR or PR + CR) and features (Fisher’s exact test)

ISSN: 1471-2407