Comparisons of tumor-infiltrating lymphocyte levels and the 21-gene recurrence score in ER-positive/HER2-negative breast cancer

Abstract

Background: Recent studies have shown that tumors with extensive tumor-infiltrating lymphocytes (TILs) have a higher probability of pathologic complete response, even in luminal/human epidermal growth factor 2 (HER2)-negative breast cancer. We compared TIL levels and the 21-gene recurrence score (RS) in estrogen receptor (ER)-positive/HER2-negative breast cancer.

Methods: We evaluated the percentage of stromal TILs in 198 ER-positive/HER2-negative patients in whom RS was obtained by examining slides of surgical specimens by standardized methodology proposed by the international TIL Working Group. TIL levels were categorized as high (≥ 60%), intermediate (11-59%), or low (≤ 10%). All tumors were treatment-naïve.

Results: Ninety-seven (49.0%), 88 (44.4%), and 13 patients (6.6%) had low, intermediate, and high TIL levels, respectively. There was a significant but weak correlation between continuous RS and continuous TIL levels (Pearson's R = 0.201, p = 0.004). The mean RS was significantly highest in high TIL tumors (17.8 ± 10.7 in low TIL tumors, 19.4 ± 8.7 in intermediate TIL tumors, and 26.2 ± 8.2 in high TIL tumors; p = 0.014). However, when we compared categorized RS and TIL levels, we found that tumors with high TIL levels tended to have higher RS (≥ 26) but it was not significant (p = 0.155). Furthermore, multivariate analysis revealed that high RS was not an independent factor associated with high TIL levels. Chemo-endocrine therapy was more frequently performed among patients with high TILs and less frequently among those with low or intermediate TILs (p < 0.001).

Conclusions: Despite of a weak correlation between continuous TIL levels and RS, we found that tumors with high TIL levels tended to have a higher RS in ER-positive/HER2-negative breast cancer. Further study is warranted considering the clinical outcomes.

Keywords: 21-gene recurrence score; Breast cancer; Tumor-infiltrating lymphocytes.

Fig. 1

Scatter plots of continuous recurrence scores and continuous tumor infiltrating lymphocytes (TILs) (Pearson’s R = 0.201; P = 0.004)

Fig. 2

Distribution of the 21-gene recurrence score (RS) groups according to tumor infiltrating lymphocyte (TIL) count. a Patients with intermediate-TIL tumors showed a distribution of categorized RS similar to that of patients with low-TIL tumors. In contrast, among the patients with high-TIL tumors, the percentage of intermediate RS was significantly higher (p = 0.007). b Tumors with higher RS (≥ 26) tended to have a higher rate of high TIL levels, but it was not significant (p = 0.155)

Fig. 3

Distributions of recurrence score (RS) according to tumor infiltrating lymphocyte (TIL) levels. a Distributions of RS differed significantly among the three groups, which were categorized by TIL levels (p = 0.001 with Kruskal-Wallis test). The median RS (interquartile) was 16 (11–23) in the low-TIL group, 18 (13.25–22.75) in the intermediate-TIL group, and 24 (22–28.5) in the high-TIL group. A post-hoc test with the Bonferroni-corrected Dunn’s procedure showed that high-TIL tumors had significantly higher median RS than low- or intermediate TIL tumors (p = 0.001 and p = 0.019, respectively), whereas intermediate-TIL tumors did not have a higher median RS than low-TIL tumors (p = 0.290). b Distributions of RS differed significantly according to the dichotomized TIL levels (p < 0.001 with Mann-Whitney U test). The median RS was 17 (12.5–23) in the low- and intermediate-TIL group and 24 (22–28.5) in the high-TIL group. Lines indicate the median values, and error bars indicate interquartile ranges

Fig. 4

Association of TIL levels with RS-guided adjuvant therapy. a The proportions of adjuvant treatments according to RS (p < 0.001). b The proportions of adjuvant treatments according to TIL levels (p < 0.001)