Correlation between magnetic resonance imaging and the level of tumor-infiltrating lymphocytes in patients with estrogen receptor-negative HER2-positive breast cancer

Abstract

Background

High levels of tumor-infiltrating lymphocytes (TILs) are associated with improved prognosis and response to therapy in patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer.

Purpose

This study investigated the associations between TIL levels and magnetic resonance imaging (MRI) findings in patients with estrogen receptor (ER)-negative HER-2 positive breast cancer.

Material and Methods

This study included 110 consecutive patients with surgically confirmed ER-negative HER2-positive breast cancers who underwent preoperative MRI from January to December 2015. Images of all lesions were reviewed in accordance with the BI-RADS lexicon by radiologists blinded to clinicopathologic findings. Tumor kinetic features were acquired by computer-aided diagnosis (CAD). Patients were divided into three TIL groups: low (<10%); intermediate (10–50%); and high (>50%). Associations between TIL levels and clinicopathologic and imaging features were evaluated; independent predictors of high and low TIL were identified by multiple logistic regression analysis.

Results

The 110 patients included 29 (26.4%) with low, 45 (40.9%) with intermediate, and 36 (32.7%) with high TIL levels. Multiple logistic regression analysis showed that older age (odds ratio [OR] = 1.08; P = 0.017), high peak enhancement (OR = 1.01; P = 0.019), positive CK5/6 (OR = 4.36; P = 0.024), and low Ki-67 (OR = 14.29; P = 0.037) were significantly associated with low TILs; low peak enhancement (OR = 1.01; P = 0.020) was significantly associated with high TILs.

Conclusion

MRI features may predict TIL levels in patients with ER-negative HER-2 positive breast cancer, enhancing the ability to diagnose and treat these patients.

Keywords

Breast cancer human epidermal growth factor receptor 2 magnetic resonance imaging tumor-infiltrating lymphocytes

Introduction

Human epidermal growth factor receptor 2 (HER2) is overexpressed in approximately 20–30% of invasive breast cancers and is associated with more aggressive disease progression, higher recurrence rate, poorer prognosis, and shorter survival (1,2). Tumor-infiltrating lymphocytes (TIL) reflect individual immunological tumor responses, with higher levels of TILs observed in highly proliferative tumors, including HER2-positive and triple-negative breast cancer (TNBC) (3,4). TIL levels show significant predictive and prognostic value (3,5–9). For example, higher TIL levels in HER2-positive breast cancer have been associated with greater response to adjuvant trastuzumab treatment (10), higher rates of pathological complete response to anthracyclin- and taxane-based chemotherapy (11), and longer recurrence-free survival (9).

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is widely used in breast cancer patients to detect lesions, assess disease extent, and monitor tumor response based on morphologic and kinetic features. Several studies have investigated the relationship between tumor subtypes and imaging findings. HER2-positive breast cancers are characterized by irregular margins, multifocality, and washout enhancing kinetic curves (12). Moreover, hormone status affects patterns of recurrence and survival in patients with HER2-positive breast cancer (13,14). An evaluation of imaging findings in patients with HER2-positive breast cancer showed that MRI phenotypes were dependent on hormone status and that independent MRI features associated with hormone negative HER2-positive breast cancers included circumscribed mass margins and non-mass enhancement (15).

Although several studies have investigated the association between MRI features and TIL levels in patients with TNBC (16,17), none to date has assessed this association in patients with HER2-positive breast cancer. Moreover, as HER2-positive breast cancers differ by hormone status, with the luminal subtype showing significantly lower TIL levels than the TNBC or HER2-subtype (5,10,18), this study investigated the associations between TIL levels and MRI findings in patients with ER-negative HER2-positive breast cancer.

Material and Methods

Study population

This retrospective study was approved by the institutional review board of our hospital, which waived the requirement for informed patient consent. The medical records of patients with surgically confirmed ER-negative HER2-positive breast cancer who underwent preoperative MR in our institution from January to December 2015 were reviewed. Of the 148 patients identified, those who received neoadjuvant chemotherapy (n = 38) were excluded. Therefore, the study population consisted of 110 women (mean age = 51.6 years, age range = 30–76 years).

MRI technique

Bilateral MRI was performed using a 1.5-T (Avanto; Siemens Medical Solutions, Erlangen, Germany) or 3.0-T (Skyra; Siemens Medical Solutions, Erlangen, Germany, Ingenia; Philips, Best, The Netherlands) MR scanner and a dedicated 18-channel phased-array breast coil (Siemens Medical Solutions) with the patient in a prone position. The imaging protocol included a T2-weighted sequence (1.5-T: TR/TE = 1300/131 ms, field of view [FOV] = 340 × 340 mm, matrix size = 384 × 384, and slice thickness = 1.5 mm; 3.0-T: TR/TE = 1100/131 ms, FOV = 341 × 210 mm, matrix size = 256 × 416, and section thickness = 1.5 mm) and a dynamic contrast material-enhanced fat-suppressed T1-weighted (T1W) sequence (1.5-T: TR/TE = 5.0/2.4 ms, FOV = 340 × 340 mm, matrix size = 384 × 384, section thickness = 0.9 mm; 3.0-T: TR/TE = 5.6/2.5 ms, matrix size = 360 × 360, section thickness = 0.9 mm), consisting of one unenhanced and five contrast-enhanced acquisitions, with a temporal resolution of 61 s. For all examinations, 0.2 mL/kg gadoterate meglumine (UNIRAY®, Dongkook Pharmaceutical Co., Ltd., Seoul, Republic of Korea) was power-injected (Spectris; Medrad, Pittsburgh, PA, USA) at a flow rate of 1 mL/s, followed by a 20-mL saline flush.

MRI interpretation and computer-aided diagnosis (CAD) data collection

All MR images were reviewed by two dedicated breast radiologists (Y.K. and W.J.C., with five and 10 years of experience in breast imaging, respectively) using the Breast Imaging Reporting and Data System (BI-RADs) MRI lexicon (19). Reviewers were blinded to the clinicopathologic findings. The amount of fibroglandular tissue (FGT; almost entirely fat, scattered FGT, heterogeneous FGT, or extreme FGT), background parenchymal enhancement (BPE; minimal, mild, moderate, or marked), and morphologic features of each lesion were determined. In patients with multiple lesions, only the lesion with the largest diameter was evaluated. The kinetic features of each tumor were evaluated using a commercially available CAD system (CADstream, version 6.0.1; Confirma, Kirkland, WA, USA). To measure MRI parameters, precontrast and all postcontrast T1W imaging series were transferred to a CAD system, which automatically segmented the tumors in three dimensions and calculated the tumor diameter (maximal size of the enhancing lesion), angio volume (total enhancing lesion volume), peak enhancement (highest pixel signal intensity in the first postcontrast series), and delay enhancement profiles (proportions of washout, plateau, and persistent enhancing components within a tumor). Peak enhancement was defined as the percentage of increase of the initial up-slope of the time-signal intensity curve and was calculated according to the following equation: PE = (S₁ – S₀)/S₀ × 100.

Pathologic analysis

Pathological data recorded included histologic type, invasive tumor size, nuclear grade, histologic grade, lymph node status, lymphovascular invasion, and immunohistochemical results, including expression of epidermal growth factor receptor (EGFR), cytokeratin (CK) 5/6, and Ki-67 proliferation. All tumors were histopathologically analyzed for the presence of TILs. TIL levels were defined as the proportion of each tumor and adjacent stroma infiltrated by lymphocytes and reported in 10% increments. Patients were divided into three TIL groups: low (<10%); intermediate (10–50%); and high (>50%) (20). If patients had multiple tumors, only the index tumor was included in the statistical analysis.

Statistical analysis

Clinicopathologic and MRI findings were compared in the three TIL groups, using the Chi-square test or Fisher’s exact test for categorical variables, and T-test/ANOVA or the Kruskal–Wallis test for continuous variables. Independent predictors of low (versus high or intermediate) TILs were identified by multiple logistic regression analysis. Variables with P values < 0.1 in univariate analysis were included in the multivariate analysis, using a logistic regression method with backward elimination. All statistical analyses were performed with SPSS software version 20.0 (SPSS, Chicago, IL, USA) or SAS version 9.3 (SAS Institute, Cary, NC, USA). A P value < 0.05 was considered statistically significant.

Results

Patient and tumor characteristics

Of the 110 patients, 29 (26.4%) had low TIL, 45 (40.9%) had intermediate TIL, and 36 patients (32.7%) had high TIL (Table 1). Patients in the low TIL group were significantly older than those in the intermediate and high TIL groups (P = 0.002). Tumor sizes were in the range of 0.2–7.2 cm (mean = 2.3 cm), with no significant differences among the three groups. All tumors were invasive ductal carcinoma; 17 patients (15.5%) were positive for lymph node metastases. Both lymph node metastasis (P = 0.049) and Ki-67 proliferation (P = 0.011) correlated with TILs.

Table 1.

Patient and tumor characteristics.

Variable	All (n =110)	Low (<10%)	Intermediate (10–50%)	High (>50%)	P value	P value (Low vs. Non-low)	P value (High vs. Non-high)
Patients (n)	110 (100)	29 (26.4)	45 (40.9)	36 (32.7)
Patient age, years (mean ± SD)	51.6 ± 8.7	55.8 ± 9.4	50.0 ± 8.5	50.3 ± 7.5	0.010	0.002	0.253
Tumor size, cm (mean ± SD)	2.3 ± 1.7	2.7 ± 2.0	2.1 ± 1.5	2.2 ± 1.6	0.622	0.332	0.745
Histologic grade
Intermediate	27 (24.5)	6 (20.7)	15 (33.3)	6 (16.7)	0.191	0.574	0.180
High	83 (75.5)	23 (79.3)	30 (66.7)	30 (83.3)
Lymph node metastasis
Negative	93 (84.5)	28 (96.6)	34 (75.6)	31 (86.1)	0.049	0.039	0.751
Positive	17 (15.5)	1 (3.4)	11 (24.4)	5 (13.9)
Lymphovascular invasion
Absent	90 (81.8)	25 (86.2)	35 (77.8)	30 (83.3)	0.630	0.475	0.774
Present	20 (18.2)	4 (13.8)	10 (22.2)	6 (16.7)
EGFR
Negative	85 (77.3)	23 (79.3)	36 (80.0)	26 (72.2)	0.676	0.760	0.378
Positive	25 (22.7)	6 (20.7)	9 (20.0)	10 (27.8)
CK5/6
Negative	92 (83.6)	21 (72.4)	41 (91.1)	30 (83.3)	0.105	0.078	0.952
Positive	18 (16.4)	8 (27.6)	4 (8.9)	6 (16.7)
Ki-67
Low (<14%)	5 (4.5)	4 (13.8)	0 (0)	1 (2.8)	0.011	0.017	>0.999
High (≥14%)	105 (95.5)	25 (86.2)	45 (100)	35 (97.2)
Surgery type
Breast conservation	68 (61.8)	17 (58.6)	28 (62.2)	23 (63.9)	0.907	0.680	0.755
Mastectomy	42 (38.2)	12 (41.4)	17 (37.8)	13 (36.1)

Values are presented as n (%) unless otherwise specified. SD: standard deviation; EGFR: epidermal growth factor receptor; CK: cytokeratin.

Association of MRI features with TILs

MRI findings according to the BI-RADs lexicon are listed in Table 2. BPE differed significantly among the three different TIL groups (P = 0.008). Percent peak enhancement differed significantly, with a higher percentage in the low TIL group and a lower percentage in the high TIL group (P = 0.003). Other features did not differ significantly in the three TIL groups, although tumors with high TIL levels tended to appear as a mass, oval or round shape, with circumscribed margins. Tumors with high TIL levels also tended to appear as multifocal and/or multicentric disease, with a high proportion of washout component, and small angio volume. Tumors with low TIL levels tended to show non-mass enhancement, uncircumscribed margins, heterogeneous mass enhancement, an absence of multifocal and multicentric disease, a plateau or persistent kinetic curve, a high proportion of persistent component with low proportions of plateau and/or washout components, and a large angio volume. Examples of patients with high and low TIL levels are illustrated in Figs. 1 and 2, respectively.

Table 2.

MR features of patients with HER-2 positive breast cancer according to levels of tumor-infiltrating lymphocytes (TILs).

Variable	All (n = 110)	Low (<10%) (n = 29)	Intermediate (10–50%) (n = 45)	High (>50%) (n = 36)	P value	P value (Low vs. Non-low)	P value (High vs. Non-high)
Patients (n)	110 (100)	29 (26.4)	45 (40.9)	36 (32.7)
Amount of fibroglandular tissue
Almost entirely fat	4 (3.6)	1 (3.4)	2 (4.4)	1 (2.8)	0.880	0.966	0.521
Scattered	49 (44.5)	12 (41.4)	18 (40.0)	19 (52.8)
Heterogeneously dense	34 (30.9)	10 (34.5)	16 (35.6)	8 (22.2)
Extremely dense	23 (20.9)	6 (20.7)	9 (20.0)	8 (22.2)
Background parenchymal enhancement
Minimal	74 (67.3)	22 (75.9)	22 (48.9)	30 (83.3)	0.008	0.231	0.091
Mild	24 (21.8)	7 (24.1)	13 (28.9)	4 (11.1)
Moderate	8 (7.3)	0 (0)	7 (15.6)	1 (2.8)
Marked	4 (3.6)	0 (0)	3 (6.7)	1 (2.8)
Lesion type
Mass	91 (82.7)	21 (72.4)	37 (82.2)	33 (91.7)	0.124	0.087	0.084
Non-mass enhancement	19 (17.3)	8 (27.6)	8 (17.8)	3 (8.3)
Mass shape
Oval or round	20 (22.0)	5 (23.8)	6 (16.2)	9 (27.3)	0.523	0.817	0.358
Irregular	71 (78.0)	16 (76.2)	31 (83.8)	24 (72.7)
Mass margin
Circumscribed	8 (8.8)	1 (4.8)	3 (8.1)	4 (12.1)	0.637	0.457	0.397
Non-circumscribed	83 (91.2)	20 (95.2)	34 (91.9)	29 (87.9)
Mass internal enhancement
Homogenous	10 (11.0)	1 (4.8)	6 (16.2)	3 (9.1)	0.729	0.568	0.908
Heterogeneous	70 (76.9)	17 (81.0)	27 (73.0)	26 (78.8)
Rim	11 (12.1)	3 (14.2)	4 (10.8)	4 (12.1)
Multifocal or multicentric disease
Absent	38 (34.5)	7 (24.1)	16 (35.6)	15 (41.7)	0.330	0.170	0.273
Present	72 (65.5)	22 (75.9)	29 (64.4)	21 (58.3)
Kinetics
Washout	94 (85.5)	23 (79.3)	41 (91.1)	30 (83.3)	0.338	0.357	0.660
Plateau or persistent	16 (14.5)	6 (20.7)	4 (8.9)	6 (16.7)
Persistent component, % (mean ± SD)	49.7 ± 27.8	56.6 ± 26.3	46.4 ± 26.5	48.4 ± 30.5	0.291	0.123	0.723
Plateau component, % (mean ± SD)	30.1 ± 15.3	27.2 ± 13.9	32.9 ± 16.6	29.0 ± 14.6	0.266	0.243	0.590
Washout component, % (mean ± SD)	20.0 ± 20.5	16.0 ± 17.8	20.6 ± 19.4	22.6 ± 23.6	0.711	0.443	0.967
Peak enhancement, % (mean ± SD)	253.9 ± 119.4	306.6 ± 158.1	255.4 ± 99.4	209.7 ± 87.7	0.003	0.017	0.001
Angio volume, cm³ (mean ± SD)	5.1 ± 6.6	6.8 ± 8.9	5.2 ± 6.5	3.8 ± 3.9	0.324	0.299	0.156

Values are presented as (%) unless otherwise specified. SD: standard deviation.

Logistic regression analysis

Univariate analysis showed that variables significantly associated with low TILs included older age (odds ratio [OR] = 1.09, 95% confidence interval [CI] = 1.03–1.15, P = 0.004), non-mass lesion enhancement (OR = 2.42, 95% CI = 0.86–6.81, P = 0.093), high peak enhancement (OR = 1.01, 95% CI = 1.00–1.01, P = 0.009), absence of lymph node metastasis (OR = 6.25, 95% CI = 0.80–50.00, P = 0.080), positive CK 5/6 (OR = 2.71, 95% CI = 0.95–7.73, P = 0.063), and low Ki-67 (OR = 12.50, 95% CI = 1.37–100.00, P = 0.025) (Table 3). High TILs were significantly associated with mass lesions (OR = 3.03, 95% CI = 0.82–11.11, P = 0.096) and low peak enhancement (OR = 1.01, 95% CI = 1.00–1.01, P = 0.009).

Table 3.

Clinicopathologic and MRI features associated with low and high tumor-infiltrating lymphocytes (TILs) by logistic regression analysis.

	Low TIL				High TIL
	Univariate analysis		Multivariate analysis		Univariate analysis		Multivariate analysis
Characteristics	Unadjusted OR (95% CI)	P value	Adjusted OR (95% CI)	P value	Unadjusted OR (95% CI)	P value	Adjusted OR (95% CI)	P value
Age
Old vs. Young	1.09 (1.03–1.15)	0.004	1.08 (1.01–1.15)	0.017	0.97 (0.93–1.02)	0.252
MR lesion type
Non-mass enhancement vs. Mass	2.42 (0.86–6.81)	0.093			0.33 (0.09–1.22)	0.096
Kinetic curve
Washout vs. Persistent or plateau	2.35 (0.49–11.19)	0.284			0.31 (0.10–0.97)	0.045
Peak enhancement
High vs. Low	1.01 (1.00–1.01)	0.011	1.01 (1.00–1.01)	0.019	0.99 (0.99–1.00)	0.009	0.99 (0.99–1.00)	0.020
Lymph node metastasis
Positive vs. Negative	0.16 (0.02–1.25)	0.080			0.92 (0.30–2.89)	0.892
CK5/6
Positive vs. Negative	2.71 (0.95–7.73)	0.063	4.36 (1.21–15.70)	0.024	1.03 (0.35–3.02)	0.952
Ki-67
High vs. Low	0.08 (0.01–0.73)	0.025	0.07 (0.01–0.85)	0.037	2.00 (0.22–18.57)	0.542

OR: odds ratio; CI: confidence interval; MR: magnetic resonance; CK: cytokeratin.

Multivariate analysis showed that low TILs were significantly and independently associated with older age (OR = 1.08, 95% CI = 1.01–1.15, P = 0.017), high peak enhancement (OR = 1.01, 95% CI = 1.00–1.01, P = 0.019), positive CK 5/6 (OR = 4.36, 95% CI = 1.21–15.70, P = 0.024), and low Ki-67 (OR = 14.29, 95% CI = 1.18–100.00, P = 0.037). Low peak enhancement (OR = 1.01, 95% CI = 1.00–1.01, P = 0.020) was a significant independent predictor of high TIL.

Discussion

The results of the present study suggest that the clinicopathologic and MRI features of ER-negative HER2-positive breast cancers differ according to TIL levels. Peak enhancement was independently associated with both low and high TIL levels, with low peak enhancement being significantly associated with high TILs and high peak enhancement with low TILs. The different patterns of CAD-generated kinetic features in tumors with high and low TIL levels may be associated with tumor angiogenesis, as shown by contrast enhancement patterns on DCE-MRI, and may predict survival outcomes.

TIL levels have been found to be a positive prognostic biomarker in patients with HER2-positive breast cancer. TILs have been used as a continuous parameter; Dieci et al. reported strong prognostic impact for overall survivals with a 78% 10-year overall survival (OS) rate in the high-TIL group and 57% in the low-TIL group (21). Another study by Ignatiadis et al. demonstrated 25% reduction in the hazard for an event-free survival event in every increase of 10% TIL levels (22).

In view of the enhancement kinetics, low TILs significantly correlated with high peak enhancement and our result were similar with previous studies that high peak enhancement being significantly associated with poor disease-free survival (DFS) and OS (23 –25). An investigation of the correlations between CAD-generated kinetic features of breast cancer and survival in patients with primary operable breast cancer showed that higher peak enhancement and washout component were significantly associated with poorer DFS (24). The present study showed a significant association between high TIL and low peak enhancement; however, the proportion of washout component tended to be higher in patients with high TIL. Computer-aided analysis tools make it possible to depict kinetic parameters pixel-by-pixel as a parametric image, describing the variation in blood flow within a lesion and throughout the entire breast. A study investigating the correlations between the kinetic features of breast cancer and histologic prognostic factors found that higher lesion volume and stronger initial enhancement were independent covariates predicting greater tumor aggressiveness (26). Total tumor volume, initial peak enhancement, and time to peak enhancement showed significant negative correlations with survival, whereas persistent curve type showed a positive correlation with survival (23). Our study found that angio volume and the proportion of persistent component tended to be higher in patients with low TIL, although these differences were not statistically significant.

Clinical variables may reflect tumor biology and influence patient outcomes. The present study found that older age, positive CK5/6, and low KI-67 were significant independent predictors of low TIL, with positive CK5/6 having the highest OR (4.36) for low TILs on multivariate analysis. Positive CK 5/6 expression was a significant independent predictive factor for worse DFS (27). CK5/6 is a basal marker associated with negative hormone status and shorter DFS, and basal phenotype have a more aggressive course than non-basal phenotypes (28).

The present analysis showed that HER2-positive breast cancer with high levels of TILs tended to have mass lesions with multifocal multicentricity and a high proportion of washout pattern, a pattern generally considered to be indicative of malignancy and imaging features of HER2-positive breast cancer (12,15). The results of the present study also demonstrated that HER2-positive breast cancer with low TIL levels tended to be non-mass enhanced lesions, without multifocal multicentricity, and was likely to have a persistent pattern with a persistent or plateau kinetic curve. Although our results are defined in clinically used MR features according to the BI-RADs 5th edition, heterogeneity of breast cancer is well-known and may be in consideration (29). Wu et al. extracted 17 computational DCE-MRI features and showed significant association with TIL levels in breast cancer (30). In this study, four imaging features were significantly associated with TILS, including tumor volume, cluster shade of signal enhancement ratio (SER), mean SER of tumor-surrounding BPE, and proportion of BPE.

This study had several limitations. First, this study was retrospective in design and included patients from a single tertiary referral center, which may limit the generalizability of the study results. Second, the sample size was relatively small, which may also limit the generalizability of these findings. Third, most features that differed among patient groups were not statistically significant. Thus, generalization of our results requires in the future with validation study.

In conclusion, the MRI phenotypes and clinicopathologic features of ER-negative HER2-positive breast cancers differ according to TIL levels. Low TILs were significantly associated with older age, high peak enhancement, positive CK 5/6, and low Ki-67, whereas high TILs were significantly associated with low peak enhancement. MRI features may have the potential to predict TIL levels in patients with HER-2 positive breast cancer, thereby enhancing the ability to diagnose and treat these patients.

Footnotes

Declaration of conflicting interests

The author(s) declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Slamon

Clark

Wong

, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987; 235:177–182.

Hudis

CA.

Trastuzumab–mechanism of action and use in clinical practice.

N Engl J Med 2007; 357:39–51.

Loi

Sirtaine

Piette

, et al. Prognostic and predictive value of tumor-infiltrating lymphocytes in a phase III randomized adjuvant breast cancer trial in node-positive breast cancer comparing the addition of docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG 02-98. J Clin Oncol 2013; 31:860–867.

Denkert

Loibl

Noske

, et al. Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer. J Clin Oncol 2010; 28:105–113.

Salgado

Denkert

Campbell

, et al. Tumor-infiltrating lymphocytes and associations with pathological complete response and event-free survival in HER2-positive early-stage breast cancer treated with lapatinib and trastuzumab: a secondary analysis of the NeoALTTO trial. JAMA Oncol 2015; 1:448–454.

Lee

Kim

Park

, et al. Prognostic significance of tumor-infiltrating lymphocytes and the tertiary lymphoid structures in HER2-positive breast cancer treated with adjuvant trastuzumab. Am J Clin Pathol 2015; 144:278–288.

Luen

Salgado

Fox

, et al. Tumour-infiltrating lymphocytes in advanced HER2-positive breast cancer treated with pertuzumab or placebo in addition to trastuzumab and docetaxel: a retrospective analysis of the CLEOPATRA study. Lancet Oncol 2017; 18:52–62.

Liu

, et al. The predictive and prognostic role of stromal tumor-infiltrating lymphocytes in HER2-positive breast cancer with trastuzumab-based treatment: a meta-analysis and systematic review. J Cancer 2017; 8:3838–3848.

Perez

Ballman

Tenner

, et al. Association of stromal tumor-infiltrating lymphocytes with recurrence-free survival in the N9831 adjuvant trial in patients with early-stage her2-positive breast cancer. JAMA Oncol 2016; 2:56–64.

10.

Loi

Michiels

Salgado

, et al. Tumor infiltrating lymphocytes are prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial. Ann Oncol 2014; 25:1544–1550.

11.

Denkert

von Minckwitz

Brase

, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without carboplatin in human epidermal growth factor receptor 2-positive and triple-negative primary breast cancers. J Clin Oncol 2015; 33:983–991.

12.

Elias

Adams

Wisner

, et al. Imaging features of HER2 overexpression in breast cancer: a systematic review and meta-analysis. Cancer Epidemiol Biomarkers Prev 2014; 23:1464–1483.

13.

Vaz-Luis

Ottesen

Hughes

, et al. Impact of hormone receptor status on patterns of recurrence and clinical outcomes among patients with human epidermal growth factor-2-positive breast cancer in the National Comprehensive Cancer Network: a prospective cohort study. Breast Cancer Res 2012; 14:R129.

14.

Lee

Park

, et al. Two histopathologically different diseases: hormone receptor-positive and hormone receptor-negative tumors in HER2-positive breast cancer. Breast Cancer Res Treat 2014; 145:615–623.

15.

Song

Bae

Chang

, et al. MR and mammographic imaging features of HER2-positive breast cancers according to hormone receptor status: a retrospective comparative study. Acta Radiol 2017; 58:792–799.

16.

Ku YJ, Kim HH, Cha JH, et al. Correlation between MRI and the level of tumor-infiltrating lymphocytes in patients with triple-negative breast cancer. AJR Am J Roentgenol 2016;207:1146–1151.

17.

Ku YJ, Kim HH, Cha JH, et al. Predicting the level of tumor-infiltrating lymphocytes in patients with triple-negative breast cancer: Usefulness of breast MRI computer-aided detection and diagnosis. J Magn Reson Imaging 2018;47:760–766.

18.

Loi

Tumor-infiltrating lymphocytes, breast cancer subtypes and therapeutic efficacy.

Oncoimmunology 2013; 2:e24720.

19.

Morris

Comstock

Lee CH . ACR BI-RADS magnetic resonance imaging. In: CJ

D’Orsi

Sickles

Mendelson

, et al., editors. ACR BI-RADS atlas breast imaging reporting and data system. 5th ed. Reston, VA: American College of Radiology, 2013:23–71.

20.

Hida

Ohi

Evaluation of tumor-infiltrating lymphocytes in breast cancer; proposal of a simpler method.

Ann Oncol 2015; 26:2351.

21.

Dieci

Mathieu

Guarneri

, et al. Prognostic and predictive value of tumor-infiltrating lymphocytes in two phase III randomized adjuvant breast cancer trials. Ann Oncol 2015; 26:1698–1704.

22.

Ignatiadis

Van den Eynden

Roberto

, et al. Tumor-infiltrating lymphocytes in patients receiving trastuzumab/pertuzumab-based chemotherapy: a TRYPHAENA substudy. J Natl Cancer Inst 2019; 111:69–77.

23.

Dietzel

Zoubi

Vag

, et al. Association between survival in patients with primary invasive breast cancer and computer aided MRI. J Magn Reson Imaging 2013; 37:146–155.

24.

Kim

Kang

, et al. Computer-aided diagnosis-generated kinetic features of breast cancer at preoperative MR imaging: association with disease-free survival of patients with primary operable invasive breast cancer. Radiology 2017; 284:45–54.

25.

Tuncbilek

Tokatli

Altaner

, et al. Prognostic value DCE-MRI parameters in predicting factor disease free survival and overall survival for breast cancer patients. Eur J Radiol 2012; 81:863–867.

26.

Baltzer

Vag

Dietzel

, et al. Computer-aided interpretation of dynamic magnetic resonance imaging reflects histopathology of invasive breast cancer. Eur Radiol 2010; 20:1563–1571.

27.

Chung SR, Choi WJ, Cha JH, et al. Prognostic factors predicting recurrence in invasive breast cancer: An analysis of radiological and clinicopathological factors. Asian J Surg 2019;42:613–620.

28.

Abd El-Rehim

Pinder

Paish

, et al. Expression of luminal and basal cytokeratins in human breast carcinoma. J Pathol 2004; 203:661–671.

29.

Sun

, et al. Heterogeneous enhancement patterns of tumor-adjacent parenchyma at MR imaging are associated with dysregulated signaling pathways and poor survival in breast cancer. Radiology 2017; 285:401–413.

30.

Teng

, et al. Magnetic resonance imaging and molecular features associated with tumor-infiltrating lymphocytes in breast cancer. Breast Cancer Res 2018; 20:101.