Tumor-infiltrating lymphocytes in breast cancer according to tumor subtype: Current state of the art

The Breast 35 (2017) 142e150

Contents lists available at ScienceDirect

The Breast journal homepage: www.elsevier.com/brst

Review

Tumor-infiltrating lymphocytes in breast cancer according to tumor subtype: Current state of the art Cinzia Solinas a, Luisa Carbognin b, c, Pushpamali De Silva a, Carmen Criscitiello d, Matteo Lambertini e, * Molecular Immunology Unit, Institut Jules Bordet, Universit e Libre de Bruxelles (U.L.B.), Brussels, Belgium U.O.C. Oncology, University of Verona, Azienda Ospedaliera Universitaria Integrata, Verona, Italy c Breast Unit, Azienda Ospedaliera Universitaria Integrata, Verona, Italy d Division of Experimental Therapeutics, European Institute of Oncology, Milan, Italy e Department of Medicine, Breast Cancer Translational Research Laboratory, Institut Jules Bordet, Universit e Libre de Bruxelles (U.L.B.), Brussels, Belgium a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 May 2017 Received in revised form 5 July 2017 Accepted 6 July 2017

The recent success of the immune checkpoint blockade in cancer immunotherapy has modified the treatment algorithms in a variety of aggressive neoplastic diseases. Nevertheless, optimal selection of ideal candidates to these drugs remains a challenge. The presence, location and composition of a preexisting tumor immune infiltrate seem to impact on the benefit from these treatments. The association between the presence of baseline tumor-infiltrating lymphocytes (TIL) and patients' outcomes has been widely investigated in breast cancer, although immunotherapeutic strategies have historically been less successful with respect to other neoplastic diseases such as melanoma and kidney cancer. TIL extent varies and has different associations with outcomes in the various breast cancer subtypes. Furthermore, the presence of baseline high TIL has been associated with an increased benefit from some chemotherapeutic and targeted agents even though some conflicting results have been observed on this regard. This review aims to summarize the state of the art of TIL in breast cancer with a focus on their assessment, prevalence and clinical implications in the different subtypes. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Tumor-infiltrating lymphocytes Breast cancer Subtypes Prognosis

1. Introduction The recent success of the immune checkpoint blockade in cancer immunotherapy has modified the treatment algorithms in a variety of aggressive neoplastic diseases [1e4]. Nevertheless, since these new agents appear to be active only in a subset of patients, optimal selection of ideal candidates to these drugs remains a challenge. Preliminary observations revealed that the presence, location and composition of a pre-existing tumor immune infiltrate impact on the benefit from these treatments [5,6]. This has led to a rising interest in understanding the influence of the immune response on the clinical outcomes of cancer patients [7e10].

* Corresponding author. Department of Medicine, Breast Cancer Translational  Libre de Bruxelles (U.L.B.), Research Laboratory, Institut Jules Bordet, Universite Boulevard de Waterloo, 121, 1000, Brussels, Belgium. E-mail addresses: [email protected] (C. Solinas), [email protected] (L. Carbognin), [email protected] (P. De Silva), carmen.criscitiello@ ieo.it (C. Criscitiello), [email protected] (M. Lambertini). http://dx.doi.org/10.1016/j.breast.2017.07.005 0960-9776/© 2017 Elsevier Ltd. All rights reserved.

In early breast cancer, the association between the presence of a baseline immune response and patients' outcomes has been widely investigated [11]. Over the past years, a growing amount of data have shown potential clinical implications for tumorinfiltrating lymphocytes (TIL) [12e14] and immune gene signatures (which reflect the presence and the activity of the immune infiltrate) [15e18]. Specifically, the evaluation of TIL in several randomized controlled trials (RCT) revealed that TIL levels vary and have a different association with outcomes in the various breast cancer subtypes [12,19]. The association between the presence of TIL and the benefit from some chemotherapeutic and targeted agents has also been investigated [12,20e24]. However, so far, more conflicting results have been observed on this regard. This review aims to summarize the state of the art of TIL in breast cancer with a focus on their assessment, prevalence and clinical implications in the different subtypes.

C. Solinas et al. / The Breast 35 (2017) 142e150

143

2. Tumor-infiltrating lymphocytes (TIL)

3.2. TIL in luminal early-stage breast cancer

2.1. Definition and assessment

The majority of patients with luminal breast cancer included in TIL studies received adjuvant chemotherapy before starting endocrine therapy. No statistically significant correlation between TIL and survival was observed [20,21,23]. A large study including 12,439 patients found no association between cytotoxic (CD8þ) and Forkhead box protein 3 positive (Foxp3þ) T cells (quantified using IHC) and breast cancer specific survival (BCSS) [34]. A meta-analysis of RCT evaluating TIL in primary breast tumors confirmed that baseline TIL levels in luminal tumors (n ¼ 2132) did not correlate with outcomes in the adjuvant setting [35]. Taken together, these data suggest that TIL extent does not add prognostic information in this setting (Table 1). Interestingly, in primary invasive lobular breast tumors, higher levels of TIL (
20%) were associated with worse prognosis at the univariate analysis and were associated with the presence of specific somatic mutations (i.e. ARID1A, BRCA2 and TP53) [33]. These results suggest that the immune infiltrate and its clinical significance may differ by breast tumor histotype, even though further confirmation from prospective cohorts would be desirable. Several studies evaluated the association between a baseline lymphocytic infiltrate scored in pre-treatment core biopsies and the likelihood of responses after neoadjuvant therapies (Table 2). In the GeparDuo and GeparTrio RCT, patients were given neoadjuvant anthracyclines and taxanes, higher baseline TIL levels were associated with higher chance of pathologic complete response (pCR) in the overall population and in specific subgroups including the ERpositive/PR-positive (any HER2) subtype [25]. Similar findings were observed in patients with luminal tumors enrolled in the neoadjuvant GeparQuinto trial (n ¼ 209) receiving neoadjuvant anthracyclines (± bevacizumab) and taxanes (± everolimus in tumors not responding to anthracyclines ± bevacizumab) [32]. In this study, the pCR rate was 28.2% for LPBC and 8.2% for non-LPBC luminal tumors. The meta-analysis by Carbognin et al. confirmed a strong association between baseline TIL and pCR in neoadjuvant trials, regardless of the breast cancer subtype [35]. A recent pooled analysis of individual patients data (n ¼ 3771) from six neoadjuvant trials of the German Breast Group (GBG) confirmed that baseline TIL were associated with increased pCR rates in all breast cancer subtypes [36]. However, in luminal tumors (n ¼ 1366), low TIL (<10%) were associated with improved overall survival (OS). Authors speculated that high TIL infiltration in ER-positive tumors might be linked to more aggressive features and/or be associated with endocrine resistance. As of today, mechanistic data or correlative studies to support these findings are lacking. One explanation for the presence of high TIL and the link with worse prognosis may derive from the specific TIL subsets infiltrating these tumors [19]. Retrospective studies revealed that high infiltrating Foxp3þ regulatory T cells (Treg) correlated with a worse prognosis in patients with luminal early-stage breast cancer treated with adjuvant tamoxifen and/or chemotherapy [19,37,38], particularly in tumors lacking concomitant CD8þ TIL [39]. Therefore, it is possible that the presence of different leukocyte populations infiltrating the luminal breast cancer microenvironment may have different prognostic implications. A recent study evaluated the association between TIL and response after neoadjuvant therapy in luminal breast cancer [31]. Two prospective trials were included in this analysis, the LETLOB (n ¼ 73; neoadjuvant endocrine-based treatment) and the GIOB (n ¼ 38; neoadjuvant chemotherapy-based treatment) trials. Stromal TIL were centrally evaluated on samples from diagnostic corebiopsies. Tumors with stromal TIL
10% (n ¼ 28) were more often of ductal histology (p ¼ 0.02), high grade (p ¼ 0.049), and high Ki67 (p ¼ 0.02) and achieved more frequently a relative
50% Ki67

The method for the assessment of TIL on hematoxylin and eosin (H&E) slides was first introduced by Denkert et al. [25]. Lately, international guidelines have been developed as a guide for pathologists to harmonize the scoring of this novel biomarker [12]. Stromal TIL are defined as the percentage of tumor stromal area containing a lymphocytic infiltrate with no direct contact with tumor cells. Intratumoral TIL are defined as intraepithelial mononuclear cells found within tumor cell nests or in direct contact with tumor cells. In breast cancer, TIL are usually more abundant in stromal compared to intratumoral areas, and the assessment of TIL in the stroma emerged as the most reproducible immune parameter scored by pathologists. Standardization of TIL scoring was the main goal of the International Breast Cancer Immuno-Oncological Biomarkers Working Group (previously named as the TIL Working Group) which includes pathologists, oncologists and immunologists [12]. A recent work from this group demonstrated that the reproducibility of TIL assessment is feasible in breast cancer, even though more efforts are needed for the improvement of the interobserver variability [26]. This study acts as an important step towards a potential use of TIL as a future biomarker in breast cancer. 2.2. Composition and organization TIL are mostly composed by T cells, followed by macrophages, B, natural killer and other immune cells [12]. Immunohistochemistry (IHC) allows the identification of the various subpopulations of TIL, by using specific leukocyte markers. A recent study demonstrated a more accurate and reproducible scoring of both stromal and intratumoral TIL by IHC with respect to H&E slides [27]. Use of specific markers for T and B cells in double stained IHC slides permits a more accurate detection of tertiary lymphoid structures (TLS) [27]. TLS are organized aggregates of B cells surrounded by T cells often localized in peritumoral areas [17,27,28]. These structures are thought to be the sites where an adaptive memory immune response can be generated [17] and their prognostic role in breast cancer remains controversial [29,30]. 3. TIL in luminal breast cancer 3.1. Incidence of TIL in luminal breast cancer Luminal tumors, defined by the expression of hormone receptors (estrogen receptor [ER] and/or progesterone receptor [PgR]), represent the most common and the least immune infiltrated breast cancer subtype [19]. Median infiltration of stromal TIL ranges between 7 and 10% [20,21,23], usually higher than intratumoral TIL (1.5e5%) [20,21,31]. In early-stage luminal breast cancer the prevalence of the lymphocyte predominant breast cancer (LPBC) phenotype (defined by the presence of
50% or
60% of either stromal or intratumoral TIL, with this cut-off differing according to each study definition) varies between 2.9 and 15% (
50% cut-off [20,21]) and 10e12% (
60% cut-off [25,32]). A well annotated retrospective study including 614 primary invasive lobular breast tumors (most of whom of luminal subtype), described median stromal TIL levels equaling to 5%, with only 9% of the tumors having
20% stromal TIL [33]. Overall, these data suggest that luminal breast tumors are characterized by low levels of stromal and intratumoral TIL with a heterogeneous prevalence of LPBC cases.

144

C. Solinas et al. / The Breast 35 (2017) 142e150

Table 1 Role of tumor-infiltrating lymphocytes (TIL) in adjuvant randomized trials [haematoxylin and eosin evaluation]. Study

Pts

Subtypes (No. pts)

TIL assay

HR DFS (p-value)

HR OS (p-value)

BIG 02e98 [20]

2009

Luminal (n ¼ 1078)

sTIL per 10% iTIL per 10% LPBC sTIL per 10% iTIL per 10% LPBC sTIL per 10% iTIL per 10% LPBC sTIL per 10% iTIL per 10% LPBC sTIL per 10% sTIL per 10% sTIL per 10% sTIL per 10% (quasi-continuous) sTIL per 10% iTIL per 10% sTIL per 10% iTIL per 10% sTIL per 10% iTIL per 10% sTIL per 10% LPBC

1.0 (0.43) 1.1 (0.33) 0.89 (0.75) 0.9 (0.071) 0.81 (0.24) 0.76 (0.37) 0.85 (0.025) 0.83 (0.1) 0.30 (0.18) 0.84 (0.005) N.S. 0.45 (0.06) 0.93 (0.55) 0.96 (0.64) 0.77 (0.02)a Arm A: 0.79 (0.0002) Arm C: 1.01 (0.85) N.R. N.R. N.R. N.R. N.R. N.R. 0.9 (0.01) 0.7 (0.13)

1.1 (0.044) 1.1 (0.22) 1.2 (0.68) 0.89 (0.12) 0.77 (0.26) 0.86 (0.68) 0.83 (0.023) 0.73 (0.035) 0.29 (0.036) 0.79 (0.003) N.S. N.S. 0.99 (0.99) 0.98 (0.84) 0.81 (0.14) Arm A: N.R. Arm C: N.R. 1.01 (0.8) 0.95 (0.6) 0.82 (0.02) 0.84 (0.055) 0.85 (0.04) 0.82 (0.04) 0.83 (<0.001) 0.48 (0.02)

HER2 pos (n ¼ 297)

TN (n ¼ 256)

ECOG 2197 & ECOG 1199 [62]

481

TN (n ¼ 481)

FinHer [23]

934

N9831 [24]

945

Luminal (n ¼ 591) HER2 pos (n ¼ 209) TN (n ¼ 134) HER2 pos (n ¼ 945)

Two French studies [21]

781

Luminal (n ¼ 463) HER2 pos (n ¼ 112) TN (n ¼ 199)

IBCSG 22-02 [64]

647

TN (n ¼ 647)

Italics signifies the p-values. Note. Results based on multivariate logistic regression are reported. Abbreviations. Pts, patients; TIL, tumor-infiltrating lymphocytes, HR, hazard ratio; DFS, disease-free survival; OS, overall survival; sTIL per 10%, stromal tumor-infiltrating lymphocytes defined as a continuous variable for each 10% increase; iTIL per 10%: intratumoral-infiltrating lymphocytes defined as a continuous variable for each 10% increase; LPBC: lymphocyte-predominant breast cancer defined as
50% of either stromal or intratumoral lymphocytic infiltration in all studies; Luminal, luminal breast cancer; HER2 pos, HER2-positive breast cancer; TN, triple negative breast cancer; N.S., not-significant; N.R., not-reported. a DFS was defined as distant disease-free survival.

Table 2 Role of tumor-infiltrating lymphocytes (TIL) in neoadjuvant randomized trials [haematoxylin and eosin evaluation]. Study

Pts

Subtypes (No. pts)

TIL assay

OR pCR (p-value)

pCR Rate LPBC vs pCR Rate non-LPBC (p-value)

GeparDuo [25]

218

All (n ¼ 218)

GeparTrio [25]

840

All (n ¼ 840)

GeparQuattro [48]

178

HER2 pos (n ¼ 178)

GeparQuinto [48]

633

HER2 neg (n ¼ 313)

N.S. 1.54 (<0.0005) 1.18 (<0.0005) 1.21 (0.001) 1.15 (¼0.03) N.R. 1.2 (0.11) 1.2 (0.01) 2.7 (0.003) N.S. N.R. 1.28 (N.S.) 4.19 (<0.001) 1.17 (N.S.) 2.17 (0.005) 0.97 (0.002) N.R. N.R. N.R.

41.7 vs. 10.8 (Focal TIL) vs. 2.8 (No TIL) (<0.0005) 40.0 vs. 15.4 (Focal TIL) vs. 7.2 (No TIL) (<0.0005) 63.8 vs. 42 (0.011) 36.6 vs. 14.3 (0.002)

NeoALTTO [45] CherLOB [46]

387 105

HER2 pos (n ¼ 387) HER2 pos (n ¼ 105)

GeparSepto [54]

1206

HER2 neg (n ¼ 810)

sTIL per 10% iTIL per 10% sTIL per 10% iTIL per 10% sTIL per 10% iTIL per 10% sTIL per 10% iTIL per 10% LPBC (
60% vs. sTIL per 10% iTIL per 10% sTIL per 10% LPBC (
60% vs. sTIL per 10% LPBC (
60% vs. sTIL per 1% sTIL per 1% iTIL per 1% LPBC (
60% vs.

HER2 pos (n ¼ 396)

LPBC (
60% vs. <60%)

N.R.

HER2 pos (n ¼ 213)

sTIL (1% log2 increase)

aOR: 1.46 (0.0091)

HER2 pos (n ¼ 320) GeparSixto [49]

580

HER2 pos (n ¼ 266) TN (n ¼ 314)

TRYPHAENA [50]

213

<60%)

<60%) <60%)

<60%)

52.6 vs. 39.5 N.S. 64.1 vs. 27.2 (<0.001) 57.3 vs. 40.0 (0.006) N.R. 64.7 vs. 25 54.3 vs. 15.6 (0.001) 61.7 vs. 57.1 (0.601) N.R.

Note. Results based on multivariate logistic regression are reported. Abbreviations. Pts, patients; TIL, tumor-infiltrating lymphocytes, HR, hazard ratio; OR, odds ratio; pCR, pathologic complete response (defined as reported by the authors); LPBC: lymphocyte-predominant breast cancer (defined as reported by the authors); sTIL per 10%, stromal tumor-infiltrating lymphocytes defined as a continuous variable for each 10% increase; iTIL per 10%: intratumoral-infiltrating lymphocytes defined as a continuous variable for each 10% increase; Luminal, luminal breast cancer; HER2 neg, HER2-negative breast cancer; HER2 pos, HER2-positive breast cancer; TN, triple negative breast cancer; N.S., not-significant; N.R.: not-reported.

suppression from baseline as compared to <10% stromal TIL tumors (55% vs 35%). In contrast, in patients treated with chemotherapy in the GIOB trial, a significant Ki67 suppression was observed only in <10% stromal TIL tumors (p ¼ 0.001) and not in the
10% stromal

TIL group (p ¼ 0.612). These findings suggest that the role of baseline TIL in luminal breast cancer might heterogeneously modulate responses based on the different types of treatments used (endocrine therapy vs chemotherapy). However, further

C. Solinas et al. / The Breast 35 (2017) 142e150

evaluations in larger studies are needed. 3.3. TIL in metastases from luminal tumors A variety of retrospective studies evaluated TIL in metastatic lesions from breast cancer, including luminal tumors. CiminoMathews et al. reported a comparison of TIL assessed in primary tumors and matched first relapses in two retrospective studies of 16 and 45 patients, respectively [40,41]. In the smaller study, luminal tumors representing 50% of the matched cases, overall metastases were characterized by lower T and B cell infiltration (p ¼ 0.0247) than their matched primary tumors [40]. Only 4 patients (3 of them with a luminal tumor) had higher TIL in metastases (vs the corresponding primary tumor), localized in lung and brain. In the overall population of the larger study from the same group, luminal tumors represented 61% of the cohort of matched cases [41]. Globally, all the lymphocyte subpopulations analyzed (including B cells, CD8þ and CD4þ T cells) were decreased in metastases as compared to the corresponding primary tumors. Similar findings were observed by Sobottka et al. who investigated TIL infiltrate at four different metastatic anatomical sites (brain, bone, liver and soft-tissues) and then compared it to corresponding primary tumors (n ¼ 87, 77% were ER-positive and 38.7% were PRpositive) [42]. In the overall population, significantly fewer TIL were observed in metastases (p < 0.001) than primary tumors. Similar results were observed in another retrospective study evaluating TIL and TLS in primary tumors and matched metastases [43]. However, in these three studies no specific information on the extent of TIL infiltrate according to breast cancer subtype was reported. Results from these retrospective cohorts, mostly composed by luminal tumors, reveal that TIL levels are usually lower in secondary lesions. Nevertheless this condition might be influenced by the organ site of relapse (i.e.: lung and brain metastases in the study by Cimino et al.) [40]. Additional data from larger and prospective cohorts are needed to better characterize TIL levels and composition in metastases from luminal tumors. 4. TIL in HER2-positive breast cancer 4.1. Incidence of TIL in HER2-positive breast cancer HER2-enriched and basal-like molecular subtypes are characterized by a higher mutational load with respect to luminal tumors [44]. Therefore, higher levels of immune infiltration are present in these subtypes. Median stromal TIL in HER2-positive tumors range between 15 and 20% [20,21,23,45,46] and median intratumoral TIL equal to 3e10% [20,21,46]. The prevalence of LPBC was approximately 10% (
60% cut-off) [24] and 11% [20] (
50% cut-off) in adjuvant trials. These values were slightly higher in neoadjuvant studies, with 16.9% (
50 cut-off%) [47] and 11.4e26.4% (
60 cut-off%) LPBC prevalence [25,46,48,49]. TIL levels are lower in the ER-positive/HER2-positive compared to the ER-negative/HER2-positive subgroup (median stromal TIL of 12.5e20% and 10e11%, respectively) [23,45]. Overall, these data suggest that in HER2-positive breast cancer TIL extent might be influenced by the ER status. Higher LPBC prevalence was observed in neoadjuvant trials, due to the possible association between TIL extent and some clinico-pathological parameters (tumor size and nodal involvement) [48]. 4.2. TIL in HER2-positive early-stage breast cancer The role of TIL in early-stage HER2-positive breast cancer has

145

been investigated in both the adjuvant (Table 1) and neoadjuvant (Table 2) settings. No anti-HER2 treatments were administered in some of the trials [20,21,25,34]. In contrast, in the most recent studies of TIL in HER2-positive disease, patients received anti-HER2 targeted treatments as mono [22e24,45e48] or dual blockade [45e47,49,50]. Two adjuvant studies evaluating TIL in HER2-positive patients not treated with trastuzumab (n ¼ 112) showed that stromal TIL were significantly associated with improved OS (HR 0.82, p ¼ 0.02, respectively) [21]. In the HER2-positive group of the BIG 02e98 trial (n ¼ 297) TIL predicted benefit from anthracyclines alone as compared to an anthracycline- and taxane-based chemotherapy in terms of both disease-free survival (DFS) and OS (interaction p ¼ 0.042 and p ¼ 0.018, respectively) [20]. Also the presence and localization of specific TIL subpopulations, specifically intratumoral CD8þ T cells was an independent predictor of improved outcome (better BCSS after adjustment for clinico-pathological parameters; HR 0.73, p ¼ 0.02) in the ER-positive/HER2-positive subgroup [34]. Taken together these findings support the prognostic effect of TIL including specific T cell subpopulations (CD8þ T lymphocytes) in HER2-positive breast cancer patients not treated with trastuzumab. Considering that pre-clinical data demonstrated the immune effects of trastuzumab capable of activating various immune effectors from both the innate and the adaptive immunity [51,52], some studies investigated the association between TIL and benefit from this treatment. Two adjuvant studies where patients were randomized to receive chemotherapy with or without trastuzumab reported conflicting results on the interaction between TIL and the benefit from trastuzumab [23,24]. In the HER2-positive cohort of the FinHER trial (n ¼ 209), increase in stromal TIL was associated with decreased distant recurrence in patients from the trastuzumab arm (n ¼ 98) [23]. On the contrary, in the TIL substudy of the N9831 trial (n ¼ 945 out of 2027 from the original study) patients benefitting more from trastuzumab had non-LPBC tumors (<60% TIL) [24]. A possible explanation for these conflicting results might be the limited number of recurrences (8 out of 94) observed in the LPBC subgroup in the N9831 trial that limited the power of the statistical analysis. Further clinical data are needed to clarify the interactions between TIL and benefit from trastuzumab. In addition, TIL assessment was performed in neoadjuvant trials with and without anti-HER2 agents. In the pooled analysis of the 6 studies from the GBG (in 4 out of 6 trials anti-HER2 agents were administered), pCR rates in HER2-positive tumors (n ¼ 1379) varied as follows: 32% in low TIL (<10%), 39% in intermediate TIL (
10e59% TIL) and 49% in high TIL tumors (
60% TIL) (p < 0.005) [36]. Ten percent increase in stromal TIL was also associated with improved OS (HR 1.12, p < 0.005) and DFS (HR 0.94, p ¼ 0.041) at the multivariate analyses. In the HER2-positive cohort from the GeparTrio study (n ¼ 254), no trastuzumab was given [25]; pCR rate was higher in the LPBC (31%) compared to the group with focal lymphocytes (19.3%) and to the one with no lymphocytes (4.3%). Five RCT evaluated TIL in neoadjuvant studies where chemotherapy was given with trastuzumab and lapatinib as mono or dual anti-HER2 blockade (n ¼ 1256) [45,46,48,49]. A recent metaanalysis estimated the association between baseline TIL levels and pCR rates including all these 5 trials [53]. High baseline TIL strongly associated with increased probability to achieve a pCR. This benefit was independent from the chemotherapy backbone (anthracyclines and taxanes vs taxanes alone) and from the antiHER2 agent(s) administered as mono or dual blockade. The association between baseline TIL and pCR rates was investigated in three neoadjuvant studies evaluating the dual anti-HER2 blockade with trastuzumab and pertuzumab: NeoSphere [47], TRYPHAENA [50] and GeparSepto [54]. Patients in the NeoSphere trial were randomized to receive neoadjuvant mono (trastuzumab

146

C. Solinas et al. / The Breast 35 (2017) 142e150

or pertuzumab plus docetaxel) or dual-anti-HER2 blockade (trastuzumab plus pertuzumab with or without docetaxel) [47]. Differently from previous studies, in the whole study population (n ¼ 243), neither continuous stromal- nor intratumoral TIL were significantly associated with pCR in the breast. Moreover, in the arm receiving the dual anti-HER2 blockade (trastuzumab and pertuzumab) given concurrently with docetaxel (n ¼ 63), high rates of pCR in the breast were seen in non-LPBC as compared to the LPBC (
50% cut-off) subgroup. The TRYPHAENA study investigated the same anti-HER2 dual blockade in the entire study population, given concomitant to an anthracycline- and taxane- or/a carboplatintaxane neoadjuvant chemotherapy backbone (n ¼ 225 patients included) [50]. Higher baseline TIL were associated with increased probability of pCR achievement after adjusting for clinicopathological characteristics. In this study, TIL assessment in the residual disease revealed a lower TIL extent in patients who achieved a pCR. In the HER2-positive cohort of the GeparSepto (n ¼ 396) study, patients were randomized to receive nabpaclitaxel or paclitaxel followed by epirubicin-cyclophosphamide in association with trastuzumab and pertuzumab [54]. LPBC did not predict the odds for pCR (LPBC: 61.7% vs non-LPBC: 57.1%; p ¼ 0.601). Globally, the association between TIL levels and the benefit from dual anti-HER2 blockade with trastuzumab and pertuzumab in the neoadjuvant setting needs more data and further investigations are currently being conducted. In a retrospective study including 175 patients with HER2positive breast cancer treated with neoadjuvant chemotherapy ± trastuzumab, stromal TIL were assessed in matched pre-treatment biopsies and in the residual disease at the time of surgery [55]. TIL decreased during treatment in 78% of the patients, and the magnitude of this decrease was strongly associated with pCR achievement. Consistently with previous findings from NeoSphere [47], NeoALTTO [45] and GeparSepto [54] trials, baseline TIL were not associated with pCR. In the residual disease, higher TIL were observed in the presence of more aggressive patterns (i.e.: high residual burden score, higher mitotic index and >5% of tumor cellularity) and correlated with worse survival outcomes after adjusting for clinico-pathological parameters at the multivariate analysis. These findings probably indicate the presence of subpopulations with predominant immune suppressive activity (i.e. Foxp3þ vs CD8þ TIL) in the residual disease, suggesting a link with treatment resistance. Thus, this study gives insights on the possible future clinical utility of TIL assessment in the residual disease for HER2-positive breast cancer after neoadjuvant therapy. 4.3. TIL in HER2-positive metastatic disease Evidence from retrospective studies including metastases from all breast cancer subtypes [40e43] revealed that TIL extent in the secondary lesions is usually lower than in primary tumors. The ancillary study on SAFIR01 and MOSCATO trials (n ¼ 244 samples were analyzed for TIL), showed that higher levels of stromal TIL and a trend for higher programmed death-ligand 1 (PD-L1) expression was present in metastases from the HER2-positive cohort with respect to the other breast cancer subtypes [56]. Very few metastatic tumors had high (
50%) intratumoral TIL (1%) and stromal TIL (5%). Furthermore, the prevalence of programmed death-1 (PD-1) (5%) and PD-L1 (3%) expression was extremely rare. The CLEOPATRA trial investigated the addition of pertuzumab to trastuzumab and docetaxel as first-line treatment in the HER2positive advanced disease [57]. Pre-treatment stromal TIL were assessed in 678 tumors: 631 unpaired samples (93%) of primary tumors and 47 (7%) unpaired samples from metastases. The majority or relapses were cutaneous (40%), followed by hepatic (34%),

bony (15%) and pulmonary (10%). Lung metastases had higher TIL with respect to the other organ sites. For 20 cases only, matched samples were available and a trend of lower levels of stromal TIL was observed in metastases with respect to the corresponding primary tumors. TIL levels predicted better prognosis, independent from known prognostic clinico-pathological parameters. No interaction between TIL and the treatment effect of pertuzumab was observed. In the EMILIA trial, metastatic HER2-positive patients previously treated with trastuzumab and taxanes were randomized to trastuzumab emtansine (T-DM1) or lapatinib plus capecitabine [58]. TIL were evaluated in 95 cases out of 991 patients included. Around 40% of the metastatic sites were TIL negative and a worse outcome was observed in the lapatinib plus capecitabine group having high tumor infiltration. Overall, these data consistently reveal that TIL levels are low at the metastatic site(s). Their extent might be heterogeneous by organ site of relapse, with lung metastases showing a trend of higher levels of infiltration. Further data are needed to understand whether TIL might be useful tools for identifying ideal candidates to immunotherapy in the metastatic setting. 5. TIL in TN breast cancer 5.1. Incidence of TIL in TN breast cancer Triple negative disease represents the breast cancer subtype with the highest lymphocyte infiltrate, supporting the potential immunogenicity of its phenotype. In this regard, the majority of studies showed values of stromal TIL ranging from 15% to 25%, intratumoral TIL lower than 5e10% and a LPBC prevalence between 10% and 20% in primary tumors [20,21,59,60]. The clonal heterogeneity and the genomic instability, together with the higher mutational load characterizing the TN phenotype in comparison with luminal subtypes, may theoretically contribute to the stimulation of the host immune response against tumor cells [61]. 5.2. TIL in TN early-stage breast cancer The potential value of TIL in TN breast cancer has been widely investigated in patients with early-stage disease. In the adjuvant setting (Table 1), the first prospectiveretrospective analysis was conducted in a phase III trial derived from the subgroup of 256 patients with node-positive TN disease enrolled in the BIG-02-98 trial [20]. These data demonstrated a 15% (p ¼ 0.025) and 17% (p ¼ 0.023) relative reduction of risk of recurrence and a 17% (p ¼ 0.1) and 27% (p ¼ 0.035) relative reduction of risk of death for each 10% increase in stromal and intratumoral TIL. Similarly, the presence of LPBC (
50% TIL) was significantly associated with better DFS (HR 0.3, p ¼ 0.018) and OS (HR 0.29, p ¼ 0.036). The prognostic role of TIL in TN breast cancer was subsequently confirmed in the subset of 481 samples from two phase III randomized trials (Eastern Cooperative Oncology Group 2197 and 1199) [62]. At a median follow-up of over 10 years, each 10% increase in stromal TIL was significantly associated with DFS (HR 0.84, p ¼ 0.005) and OS (HR 0.79, p ¼ 0.003). Despite an observed trend confirming previous findings, the association of LPBC (
50% cut-off) and intratumoral TIL with outcome was not statistically significant. Consistent results were reported by the prospectiveretrospective analysis conducted in 134 TN samples from the FinHer trial [23]: each 10% increase in stromal TIL was associated with a 23% relative reduction of risk for distant recurrence or death (p ¼ 0.02). However, the association between stromal TIL and OS

C. Solinas et al. / The Breast 35 (2017) 142e150

did not reach the statistical significance (HR 0.81, p ¼ 0.1), even if this could be due to the small number of events occurred. Conversely, the analysis performed in the context of two French randomized trials, suggested that for each 10% stromal and intratumoral lymphocytes increase, there was a 15% (p ¼ 0.04) and a 18% (p ¼ 0.04) relative reduction in death risk in 199 TN patients, without any differential effect of chemotherapy according to TIL levels [21]. The meta-analysis by Carbognin et al. estimated that, in the context of TN breast cancer, each 10% increment of stromal TIL was significantly associated with a chance of better OS (HR 0.84, p < 0.0001). Recently, an individual patient data pooled analysis of the same randomized trials, including 991 patients with primary TN disease, confirmed the prognostic value of stromal TIL at diagnosis in terms of both invasive DFS (HR 0.86, p < 0.0001) and OS (HR 0.83, p ¼ 0.001) [63]. In addition, similar results were subsequently reported by the analysis of 647 TN patients from the International Breast Cancer Study Group Trial 22-00 phase III trial [64]. Indeed, the magnitude of benefit of stromal TIL in terms of OS was fully comparable with previous studies (HR 0.83 for each 10% increase in stromal TIL, p < 0.001). The strength of the aforementioned data, despite differences in terms of adjuvant chemotherapeutic regimens, attrition bias rates, and inclusion criteria between trials, consists of the measurement of stromal TIL, as a continuous variable, on H&E-stained full sections, as suggested by the TIL working group's recommendations [12]. The clinical and analytical validity of these recommendations for TIL evaluation in the clinical practice was recently assessed in a retrospective series of 897 consecutive TN breast cancer [59]. This analysis suggested that each 10% increase in stromal TIL strongly predicted better distant DFS (HR 0.76, p < 0.0001) and OS (HR 0.76, p < 0.0001), strengthening the potential adoption of TIL in daily practice. Other evidences deriving from non-randomized trials showed consistent data supporting lymphocyte infiltration as a prognostic biomarker in TN breast cancer [34,38,65e70]. Likewise, this role is also confirmed in meta-analyses including these trials as well as randomized studies [71,72]. Nevertheless, in several studies lymphocytes were assessed with methods (such as IHC or tissue microarrays scoring) currently to be limited to the research setting, as recommended by the international guidelines [12,34,38,66,67]. Given that pCR is generally considered a reliable intermediate end-point for longer survival particularly for patients with TN tumors [73], the discovery and validation of predictors of response represent a milestone for both clinical research and daily practice. The evaluation of immunological infiltrate in core biopsies in the neoadjuvant setting was widely investigated in randomized and retrospective trials, particularly in TN breast cancer as well as in HER2-positive tumors [32,49,68,74]. In the exploratory analysis of the GeparQuinto trial, the impact of the LPBC phenotype (
60% TIL) was evaluated in a subset of 104 TN patients randomized to receive anthracycline and taxane-based chemotherapy without bevacizumab [32]. The pCR rate for 66 LPBC tumors reached 44.2%, compared to 31.1% for 38 non-LPBC tumors, despite a nonsignificant Fisher test (p ¼ 0.22). In the GeparSixto trial, the LPBC phenotype (
60% cut-off) was significant for pCR prediction (OR: 2.17, p ¼ 0.005) in the subset of 314 TN patients receiving anthracycline and taxane-based chemotherapy plus bevacizumab with or without carboplatin [49]. The addition of carboplatin in LPBC tumors increased the pCR rate from 43% to 74% (p ¼ 0.0005); however, the test for interaction with therapy was not significant, different from what observed in the HER2-positive subtype receiving the same chemotherapy regimen, in association with the anti-HER2 blockade with trastuzumab and lapatinib.

147

The role of TIL as predictors of pCR was confirmed in the individual patient data meta-analysis including a subgroup of 906 TN tumors from 6 German randomized trials [36]. Beyond the predictive value of TIL, this meta-analysis suggested that TN patients with high TIL at diagnosis (
60% of stromal TIL) experienced a better DFS (p ¼ 0.02). The potential prognostic role of TIL in the neoadjuvant setting was explored in a mono institution experience of 72 TN patients suggesting that TIL >10% may have an impact in terms of OS [75]. Taking into account that a subset of TN breast cancer presents a homologous recombination deficiency (HRD) and a BRCA1/2 mutation status, the association of TIL with these phenotypes was recently evaluated in a pooled analysis including 161 patients enrolled in five phase II trials and treated with a neoadjuvant platinum-based chemotherapy [76]. In this analysis, both stromal and intratumoral TIL levels were not associated with either HRD or BRCA mutation status and they were not statistically independent predictors of pCR. Recent data suggest that the quality of the response obtained with neoadjuvant treatment may potentially have a prognostic value. In this regard, a retrospective study assessed the intratumoral and stromal TIL in 278 TN cases with residual invasive disease after neoadjuvant chemotherapy [77]. At a median follow-up of 6.3 years, TIL percentage, pathological nodal status and tumor size after treatment were independent predictors of outcome. For each 10% increment in both intratumoral and stromal TIL a 14% (p ¼ 0.01) and a 15% (p ¼ 0.02) reduction in the risk of metastasis and a 14% (p ¼ 0.01 for stromal TIL; p ¼ 0.03 for intratumoral TIL) reduction of death were observed, respectively. Furthermore, the 5-year OS rate was 91% for the 27 patients with LPBC phenotype (
60% TIL) and 55% for the remaining 251 patients with non-LPBC phenotype (HR 0.19, p ¼ 0.0017). Interestingly, the majority of LPBC cases after chemotherapy displayed low TIL on the core biopsy at diagnosis, suggesting that chemotherapy could induce lymphocyte recruitment. The prognostic role of TIL in residual tumors after neoadjuvant chemotherapy was also evidenced in a retrospective analysis evaluating cytotoxic and Foxp3þ T cells, by double-staining IHC, in 131 TN patients that did not achieve pCR [78]. Furthermore, a preliminary analysis of a small series of 61 TN tumors showed that the combination of residual cancer burden and TIL emerged as a prognostic marker of recurrence (HR 0.04, p ¼ 0.018) [79]. However, in another small retrospective series [68], tumors with high TIL were associated with a not statistically significant better prognosis compared with low and intermediate TIL tumors (p ¼ 0.24). 5.3. TIL in TN metastatic breast cancer Nowadays, few data concerning the prognostic and predictive role of TIL in the metastatic setting are available. A recent retrospective study suggested that the subset of 17 advanced TN patients receiving eribulin with high TIL (
50% of stromal TIL in the primary tumor) had a significantly longer PFS compared to patients (n ¼ 5) with low TIL (p ¼ 0.033) [80]. The hypothesis that TIL may be a potential predictive marker of the therapeutic effect of eribulin needs undoubtedly further investigations. With regard to the prevalence of TIL in tumor samples of metastases from breast cancer, a retrospective analysis suggested that the percentage of stromal TIL in secondary TN lesions was lower than in the primary tumors (p ¼ 0.06) [81]. Furthermore, the overall group (including HER2-positive e TN subtypes) with low stromal TIL (<10%) had a significantly poorer OS than the intermediate TIL (10%e60%) group (HR 3.77, p ¼ 0.038). A low stromal TIL level in tumor samples from metastases and a better prognosis in TN patients with high TIL (>10%) was reported in a preliminary analysis including 52 HER2-positive and 42 TN metastatic disease

148

C. Solinas et al. / The Breast 35 (2017) 142e150

[82]. However, the role of TIL for TN as well as for other breast cancer subtypes in the advanced disease is in need of further investigation. The growing interest for immunotherapy in TN breast cancer requires new efforts to integrate the value of TIL with that of immune biomarkers, such as the PD-L1, in order to optimize the treatment strategies of this heterogeneous subtype [83].

Conflict of interests statement The authors declare no conflict of interests. Ethical approval Not applicable. References

6. Conclusions and future perspectives A growing body of evidence has demonstrated the clinical relevance of TIL in breast cancer, also supported by ongoing efforts towards a standardized assessment of this biomarker on H&E slides following international guidelines. Pre-existing immune infiltrates appeared linked with good outcomes in the most aggressive breast cancer subtypes, namely the HER2-positive and TN tumors where high TIL infiltration predicted better survival and higher likelihood of achieving pCR. Therefore, TIL may be used as one of the stratification parameters to be considered in studies investigating both de-escalating (i.e. in patients with good prognostic features), or alternatively escalating (i.e. in patients with adverse prognostic features) treatment strategies. For example, this is of great importance in HER2-positive breast cancer for better identifying who are those patients with early-stage disease that may benefit from dual anti-HER2 blockade with trastuzumab and pertuzumab [84]. Moreover, TIL might aid in the selection of candidates to immunotherapy. For instance, patients with low levels of TIL may probably require a boost of their baseline anti-tumor immune response, and thus, they might benefit from combinations of immune checkpoint blockade and chemotherapy, radiotherapy or other targeted therapies. Noteworthy, in patients not achieving a pCR, the assessment of TIL in the residual disease after neoadjuvant therapy, ideally in association with other immune biomarkers (i.e. PD-L1, CD8, Foxp3) could aid in identifying patients who may require additional treatments including immune checkpoint blockade. The role of TIL in luminal disease is more controversial. Recent evidences demonstrated a worse OS in patients with high infiltrated tumors in the neoadjuvant setting. This possibly indicates that some of these patients might benefit from strategies able to remove some mechanisms of immune suppression. In the advanced disease, metastases from breast cancer have been shown to have a low immune infiltration. However, the potential prognostic and predictive role of TIL in this setting needs further investigation. Emerging data from trials testing anti-PD-1 and anti-PD-L1 monotherapy suggested that the presence of TIL subpopulations (i.e. CD8þ T cells) and the expression of PD-L1 by immune cells correlated with increased benefit from immune checkpoint blockade in advanced breast cancer, particularly in the TN subtype [85,86]. A variety of ongoing trials are now prospectively evaluating baseline TIL and PD-L1 expression in breast cancer patients treated with immunotherapy. Results from these studies are awaited in order to clarify the clinical utility of TIL and PD-L1 in this setting.

Funding source The present work did not receive any specific funding. Cinzia Solinas and Pushpamali De Silva are fellows of the Belgian Fund for le vie. Matteo Lambertini Scientific Research (FNRS)-Operation Te acknowledges the support from the European Society for Medical Oncology (ESMO) for a Translational Research Fellowship at the Institut Jules Bordet in Brussels.

[1] Franklin C, Livingstone E, Roesch A, Schilling B, Schadendorf D. Immunotherapy in melanoma: recent advances and future directions. Eur J Surg Oncol 2017 Mar;43(3):604e11. s S, Navarro A, Martinez de [2] Remon J, Pardo N, Martinez-Martí A, Cedre Castro AM, et al. Immune-checkpoint inhibition in first-line treatment of advanced non-small cell lung cancer patients: current status and future approaches. Lung Cancer 2017 Apr;106:70e5. [3] Ghatalia P, Zibelman M, Geynisman DM, Plimack ER. Checkpoint inhibitors for the treatment of renal cell carcinoma. Curr Treat Options Oncol 2017 Jan 1;18(1):7. [4] Gupta S, Gill D, Poole A, Agarwal N. Systemic immunotherapy for urothelial cancer: current trends and future directions. Cancers 2017 Jan 27;9(2):15. [5] Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJM, Robert L, et al. PD1 blockade induces responses by inhibiting adaptive immune resistance. Nature 2014 Nov 27;515(7528):568e71. [6] McGranahan N, Furness AJS, Rosenthal R, Ramskov S, Lyngaa R, Saini SK, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 2016 Mar 25;351(6280):1463e9. [7] Schatton T, Scolyer R, Thompson J, Mihm Jr M. Tumor-infiltrating lymphocytes and their significance in melanoma prognosis. In: Thurin M, Marincola FM, editors. Molecular diagnostics for melanoma. Humana Press; 2014. p. 287e324. [8] Bremnes RM, Busund L-T, Kilvær TL, Andersen S, Richardsen E, Paulsen EE, et al. The role of tumor-infiltrating lymphocytes in development, progression, and prognosis of nonesmall cell lung cancer. J Thorac Oncol 2016 Jun;11(6): 789e800. [9] Solinas C, Pusole G, Demurtas L, Puzzoni M, Mascia R, Morgan G, et al. Tumor infiltrating lymphocytes in gastrointestinal tumors: controversies and future clinical implications. Crit Rev Oncol Hematol 2017 Feb;110:106e16.  NM, Awada A, Scartozzi M. The immune infiltrate in pros[10] Solinas C, Chanza tate, bladder and testicular tumors: an old friend for new challenges. Cancer Treat Rev 2017 Feb;53:138e45. [11] Pruneri G, Vingiani A, Denkert C. Tumor infiltrating lymphocytes in early breast cancer. Breast. Available from: http://www.sciencedirect.com/science/ article/pii/S0960977617304174. [12] Salgado R, Denkert C, Demaria S, Sirtaine N, Klauschen F, Pruneri G, et al. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. Ann Oncol 2015 Feb;26(2):259e71. [13] Savas P, Salgado R, Denkert C, Sotiriou C, Darcy PK, Smyth MJ, et al. Clinical relevance of host immunity in breast cancer: from TILs to the clinic. Nat Rev Clin Oncol 2015 Dec 15. [14] Luen SJ, Savas P, Fox SB, Salgado R, Loi S. Tumour-infiltrating lymphocytes and the emerging role of immunotherapy in breast cancer. Pathol (Phila) 2017 Feb;49(2):141e55. [15] Desmedt C, Haibe-Kains B, Wirapati P, Buyse M, Larsimont D, Bontempi G, et al. Biological processes associated with breast cancer clinical outcome depend on the molecular subtypes. Clin Cancer Res 2008 Aug 15;14(16): 5158e65. [16] Teschendorff AE, Miremadi A, Pinder SE, Ellis IO, Caldas C. An immune response gene expression module identifies a good prognosis subtype in estrogen receptor negative breast cancer. Genome Biol 2007 Jan;8(8):R157. [17] Gu-Trantien C, Loi S, Garaud S, Equeter C, Libin M, de Wind A, et al. CD4þ follicular helper T cell infiltration predicts breast cancer survival. J Clin Investig 2013 Jul 1;123(7):2873e92. [18] Ignatiadis M, Singhal SK, Desmedt C, Haibe-Kains B, Criscitiello C, Andre F, et al. Gene modules and response to neoadjuvant chemotherapy in breast cancer subtypes: a pooled analysis. J Clin Oncol 2012 Jun 1;30(16): 1996e2004. [19] Dieci MV, Griguolo G, Miglietta F, Guarneri V. The immune system and hormone-receptor positive breast cancer: is it really a dead end? Cancer Treat Rev 2016 May;46:9e19. [20] Loi S, Sirtaine N, Piette F, Salgado R, Viale G, Van Eenoo F, 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 Mar 1;31(7):860e7. [21] Dieci MV, Mathieu MC, Guarneri V, Conte P, Delaloge S, Andre F, et al. Prognostic and predictive value of tumor-infiltrating lymphocytes in two phase III randomized adjuvant breast cancer trials. Ann Oncol 2015 Aug 1;26(8): 1698e704. [22] Loi S, Michiels S, Salgado R, Sirtaine N, Jose V, Fumagalli D , et al. Tumor

C. Solinas et al. / The Breast 35 (2017) 142e150

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

infiltrating lymphocytes (TILs) indicate trastuzumab benefit in early-stage HER2-positive breast cancer (HER2 BC). Cancer Res 2013;73(24 Suppl). abstract S1-05. Loi S, Michiels S, Salgado R, Sirtaine N, Jose V, Fumagalli D, 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 Aug 1;25(8):1544e50. Perez EA, Ballman KV, Tenner KS, Thompson EA, Badve SS, Bailey H, et al. Association of stromal tumor-infiltrating lymphocytes with recurrence-free survival in the N9831 adjuvant trial in patients with early-stage HER2positive breast cancer. JAMA Oncol 2016 Jan;2(1):56e64. Denkert C, Loibl S, Noske A, Roller M, Müller BM, Komor M, et al. Tumorassociated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer. J Clin Oncol 2010 Jan 1;28(1): 105e13. Denkert C, Wienert S, Poterie A, Loibl S, Budczies J, Badve S, et al. Standardized evaluation of tumor-infiltrating lymphocytes in breast cancer: results of the ring studies of the international immuno-oncology biomarker working group. Mod Pathol 2016. Buisseret L, Desmedt C, Garaud S, Fornili M, Wang X, Van den Eynden G, et al. Reliability of tumor-infiltrating lymphocyte and tertiary lymphoid structure assessment in human breast cancer. Mod Pathol 2017 Jun 16 [Epub ahead of print]. Buisseret L, Garaud S, Wind A de, Eynden GV den, Boisson A, Solinas C, et al. Tumor-infiltrating lymphocyte composition, organization and PD-1/PD-L1 expression are linked in breast cancer. OncoImmunology 2017 Jan 2;6(1): e1257452. Lee HJ, Kim JY, Park IA, Song IH, Yu JH, Ahn J-H, et al. Prognostic significance of tumor-infiltrating lymphocytes and the tertiary lymphoid structures in HER2positive breast cancer treated with adjuvant trastuzumab. Am J Clin Pathol 2015 Aug;144(2):278e88. Lee HJ, Park IA, Song IH, Shin S-J, Kim JY, Yu JH, et al. Tertiary lymphoid structures: prognostic significance and relationship with tumour-infiltrating lymphocytes in triple-negative breast cancer. J Clin Pathol 2016 May 1;69(5):422e30. Dieci MV, Frassoldati A, Generali D, Bisagni G, Piacentini F, Cavanna L, et al. Tumor-infiltrating lymphocytes and molecular response after neoadjuvant therapy for HRþ/HER2 breast cancer: results from two prospective trials. Breast Cancer Res Treat 2017 Mar 13:1e8. Issa-Nummer Y, Darb-Esfahani S, Loibl S, Kunz G, Nekljudova V, Schrader I, et al. Prospective validation of immunological infiltrate for prediction of response to neoadjuvant chemotherapy in HER2-negative breast cancer e a substudy of the neoadjuvant GeparQuinto trial. PLoS One 2013 Dec 2;8(12): e79775. Desmedt C, Salgado R, Buisseret L, Zoppoli G, Fornili M, Van den Eynden G, et al. Lymphocytic infiltration in invasive lobular breast cancer. Cancer Res 2016;76(4 Suppl 1):2016. Ali HR, Provenzano E, Dawson S-J, Blows FM, Liu B, Shah M, et al. Association between CD8þ T-cell infiltration and breast cancer survival in 12 439 patients. Ann Oncol 2014 Aug 1;25(8):1536e43. Carbognin L, Pilotto S, Nortilli R, Brunelli M, Nottegar A, Sperduti I, et al. Predictive and prognostic role of tumor-infiltrating lymphocytes for early breast cancer according to disease subtypes: sensitivity analysis of randomized trials in adjuvant and neoadjuvant setting. Oncol 2016 Mar;21(3): 283e91. Denkert C, von Minckwitz G, Darb-Esfahani S, Ingold Heppner B, Klauschen F, Furlanetto J, et al. Evaluation of tumor-infiltrating lymphocytes (TILs) as predictive and prognostic biomarker in different subtypes of breast cancer treated with neoadjuvant therapy - a meta-analysis of 3771 patients. Cancer Res 2017;77(4 Suppl). abstract S1-09. Bates GJ, Fox SB, Han C, Leek RD, Garcia JF, Harris AL, et al. Quantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapse. J Clin Oncol 2006 Dec 1;24(34):5373e80. Liu S, Lachapelle J, Leung S, Gao D, Foulkes WD, Nielsen TO. CD8þ lymphocyte infiltration is an independent favorable prognostic indicator in basal-like breast cancer. Breast Cancer Res 2012;14:R48. Liu S, Foulkes WD, Leung S, Gao D, Lau S, Kos Z, et al. Prognostic significance of FOXP3þ tumor-infiltrating lymphocytes in breast cancer depends on estrogen receptor and human epidermal growth factor receptor-2 expression status and concurrent cytotoxic T-cell infiltration. Breast Cancer Res 2014;16:432. Cimino-Mathews A, Ye X, Meeker A, Argani P, Emens LA. Metastatic triplenegative breast cancers at first relapse have fewer tumor-infiltrating lymphocytes than their matched primary breast tumors: a pilot study. Hum Pathol 2013 Oct;44(10):2055e63. Cimino-Mathews A, Thompson E, Taube JM, Ye X, Lu Y, Meeker A, et al. PD-L1 (B7-H1) expression and the immune tumor microenvironment in primary and metastatic breast carcinomas. Hum Pathol 2016 Jan;47(1):52e63. Sobottka B, Pestalozzi B, Fink D, Moch H, Varga Z. Similar lymphocytic infiltration pattern in primary breast cancer and their corresponding distant metastases. Oncoimmunology 2016 Jun;5(6):e1153208. Solinas C, Boisson A, Brown D, de Wind A, Van Den Eynden G, Garaud S, et al. Tumor infiltrating lymphocytes and tertiary lymphoid structures in paired primary tumors and metastases from breast cancer patients. Ann Oncol 2016;27(suppl_6). Luen S, Virassamy B, Savas P, Salgado R, Loi S. The genomic landscape of breast

149

cancer and its interaction with host immunity. Breast 2016 Oct;29:241e50. [45] Salgado R, Denkert C, Campbell C, Savas P, Nuciforo P, Aura C, et al. Tumorinfiltrating 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 Jul 1;1(4):448.  L, Ficarra G, Bisagni G, et al. Integrated [46] Dieci MV, Prat A, Tagliafico E, Pare evaluation of PAM50 subtypes and immune modulation of pCR in HER2positive breast cancer patients treated with chemotherapy and HER2targeted agents in the CherLOB trial. Ann Oncol 2016 Aug 2. mdw262. [47] Bianchini G, Pusztai L, Pienkowski T, Im Y-H, Bianchi GV, Tseng L-M, et al. Immune modulation of pathologic complete response after neoadjuvant HER2-directed therapies in the NeoSphere trial. Ann Oncol 2015 Dec 1;26(12):2429e36. [48] Heppner BI, Untch M, Denkert C, Pfitzner BM, Lederer B, Schmitt WD, et al. Tumor-infiltrating lymphocytes: a predictive and prognostic biomarker in neoadjuvant treated HER2-positive breast cancer. Clin Cancer Res 2016 Jan 1. clincanres.2338.2015. [49] Denkert C, Minckwitz G von, Brase JC, Sinn BV, Gade S, Kronenwett R, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without carboplatin in human epidermal growth factor receptor 2epositive and triple-negative primary breast cancers. J Clin Oncol 2015 Mar 20;33(9):983e91. [50] Ignatiadis M, Van den Eynden GG, Salgado R, Fornili M, Desmedt C, Holgado E, et al. Tumor infiltrating lymphocytes before and after dual HER2 blockade in HER2-amplified early breast cancer: a TRYPHAENA substudy. ASCO Meet Abstr 2016 May 20;34(15_suppl):11507. [51] Stagg J, Loi S, Divisekera U, Ngiow SF, Duret H, Yagita H, et al. AntieErbB-2 mAb therapy requires type I and II interferons and synergizes with antiePD-1 or anti-CD137 mAb therapy. Proc Natl Acad Sci 2011 Apr 26;108(17):7142e7. [52] Charlebois R, Allard B, Allard D, Buisseret L, Turcotte M, Pommey S, et al. PolyI: C and CpG synergize with anti-ErbB2 mAb for treatment of breast tumors resistant to immune checkpoint inhibitors. Cancer Res 2017 Jan 15;77(2): 312e9. [53] Solinas C, Ceppi M, Lambertini M, Scartozzi M, Buisseret L, Garaud S, et al. Tumor-infiltrating lymphocytes in HER2-positive breast cancer patients treated with neoadjuvant chemotherapy plus trastuzumab, lapatinib or their combination: a meta-analysis of randomized controlled trials. Cancer Treat Rev 2017 Jun;57:8e15. [54] Loibl S, Jackisch C, Schneeweiss A, Schmatloch S, Aktas B, Denkert C, et al. Dual HER2-blockade with pertuzumab and trastuzumab in HER2-positive early breast cancer: a subanalysis of data from the randomized phase III GeparSepto trial. Ann Oncol 2017 Mar 1;28(3):497e504. [55] Hamy A-S, Pierga J-Y, Sabaila A, Laas E, Bonsang-Kitsis, Laurent C, et al. Stromal lymphocyte infiltration after neoadjuvant chemotherapy is associated with aggressive residual disease and lower disease-free survival in HER2positive breast cancer. Ann Onc 2017. [56] Arnedos M, Filleron T, Dieci MV, Adam J, Robbins PB, Loi S, et al. Genomic and immune characterization of metastatic breast cancer (mBC): An ancillary study of the SAFIR01 & MOSCATO trials. Ann Onc 2014;25(suppl_4). iv116eiv136. [57] Luen SJ, Salgado R, Fox S, Savas P, Eng-Wong J, Clark E, 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 Jan;18(1): 52e62. [58] Badve SS, Gokmen-Polar Y, Hoersch S, Xu J, Ruschoff J, de Haas S, et al. Role of tumor infiltrating lymphocytes (TILs) in HER2þ metastatic breast cancers (MBC) treated with trastuzumab emtansine (T-DM1) or lapatinib plus capecitabine (LþP) (EMILIA trial). J Clin Oncol 2016;34(Suppl 15). [59] Pruneri G, Vingiani A, Bagnardi V, Rotmensz N, De Rose A, Palazzo A, et al. Clinical validity of tumor-infiltrating lymphocytes analysis in patients with triple-negative breast cancer. Ann Oncol 2016 Feb 1;27(2):249e56. [60] Stanton SE, Adams S, Disis ML. Variation in the incidence and magnitude of tumor-infiltrating lymphocytes in breast cancer subtypes: a systematic review. JAMA Oncol 2016 Oct 1;2(10):1354e60. [61] Shah SP, Roth A, Goya R, Oloumi A, Ha G, Zhao Y, et al. The clonal and mutational evolution spectrum of primary triple-negative breast cancers. Nature 2012 Jun 21;486(7403):395e9. [62] Adams S, Gray RJ, Demaria S, Goldstein L, Perez EA, Shulman LN, et al. Prognostic value of tumor-infiltrating lymphocytes in triple-negative breast cancers from two phase III randomized adjuvant breast cancer trials: ECOG 2197 and ECOG 1199. J Clin Oncol 2014 Jul 28:1e9. [63] Loi S, Drubay D, Adams S, Francis PA, Joensuu H, Dieci MV, et al. Pooled individual patient data analysis of stromal tumor infiltrating lymphocytes in primary triple negative breast cancer treated with anthracycline-based chemotherapy. Cancer Res 2016;76(4 Suppl). abstract S01-03. [64] Pruneri G, Gray KP, Vingiani A, Viale G, Curigliano G, Criscitiello C, et al. Tumor-infiltrating lymphocytes (TILs) are a powerful prognostic marker in patients with triple-negative breast cancer enrolled in the IBCSG phase III randomized clinical trial 22-00. Breast Cancer Res Treat 2016 Jul;158(2): 323e31. [65] Krishnamurti U, Wetherilt CS, Yang J, Peng L, Li X. Tumor-infiltrating lymphocytes are significantly associated with better overall survival and diseasefree survival in triple negative but not estrogen receptor positive breast

150

C. Solinas et al. / The Breast 35 (2017) 142e150

cancers. Hum Pathol 2017 Jan 30. [66] Mahmoud SMA, Paish EC, Powe DG, Macmillan RD, Grainge MJ, Lee AHS, et al. Tumor-infiltrating CD8þ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol 2011 May 20;29(15):1949e55. [67] West NR, Kost SE, Martin SD, Milne K, deLeeuw RJ, Nelson BH, et al. Tumourinfiltrating FOXP3þ lymphocytes are associated with cytotoxic immune responses and good clinical outcome in oestrogen receptor-negative breast cancer. Br J Cancer 2013 Jan 15;108(1):155e62. [68] Hida AI, Sagara Y, Yotsumoto D, Kanemitsu S, Kawano J, Baba S, et al. Prognostic and predictive impacts of tumor-infiltrating lymphocytes differ between Triple-negative and HER2-positive breast cancers treated with standard systemic therapies. Breast Cancer Res Treat 2016 Jul 1;158(1):1e9. [69] Mori H, Kubo M, Yamaguchi R, Nishimura R, Osako T, Arima N, et al. The combination of PD-L1 expression and decreased tumor-infiltrating lymphocytes is associated with a poor prognosis in triple-negative breast cancer. Oncotarget 2017 Jan 17;8(9):15584e92. [70] Tian T, Ruan M, Yang W, Shui R, Tian T, Ruan M, et al. Evaluation of the prognostic value of tumor-infiltrating lymphocytes in triple-negative breast cancers. Oncotarget 2016 Jun 15;7(28):44395e405. [71] Ibrahim EM, Al-Foheidi ME, Al-Mansour MM, Kazkaz GA. The prognostic value of tumor-infiltrating lymphocytes in triple-negative breast cancer: a metaanalysis. Breast Cancer Res Treat 2014 Dec 1;148(3):467e76. [72] Mao Y, Qu Q, Chen X, Huang O, Wu J, Shen K. The prognostic value of tumorinfiltrating lymphocytes in breast cancer: a systematic review and metaanalysis. Plos One 2016 Apr 13;11(4):e0152500. [73] Cortazar P, Zhang L, Untch M, Mehta K, Costantino JP, Wolmark N, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet 2014 Jul 18;384(9938):164e72. [74] Ono M, Tsuda H, Shimizu C, Yamamoto S, Shibata T, Yamamoto H, et al. Tumor-infiltrating lymphocytes are correlated with response to neoadjuvant chemotherapy in triple-negative breast cancer. Breast Cancer Res Treat 2012 Apr;132(3):793e805. [75] Sampaio C, de Lima V, de Andrade V, Neotti T, Tavares M, Sessa V, et al. Tumor-infiltrating lymphocytes (TILs) is associated with improved overall survival in triple-negative breast cancer (TNBC) patients treated with neoadjuvant chemotherapy (NAC). Cancer Res 2017;77. abstract P06-09-44. [76] Telli M, Badve S, Vinayak S, Silver D, Isakoff S, Kaklamani V, et al. Evaluation of tumor infiltrating lymphocytes (TILs) and their association with homologous recombination deficiency and BRCA1/2 mutation status in triple-negative

[77]

[78]

[79]

[80]

[81]

[82]

[83]

[84]

[85]

[86]

breast cancer (TNBC): a pooled analysis. Cancer Res 2017;77. abstract P0609-09. Dieci MV, Criscitiello C, Goubar A, Viale G, Conte P, Guarneri V, et al. Prognostic value of tumor-infiltrating lymphocytes on residual disease after primary chemotherapy for triple-negative breast cancer: a retrospective multicenter study. Ann Oncol 2014 Mar 1;25(3):611e8. Miyashita M, Sasano H, Tamaki K, Hirakawa H, Takahashi Y, Nakagawa S, et al. Prognostic significance of tumor-infiltrating CD8þ and FOXP3þ lymphocytes in residual tumors and alterations in these parameters after neoadjuvant chemotherapy in triple-negative breast cancer: a retrospective multicenter study. Breast Cancer Res 2015;17:124. Asano Y, Kashiwagi S, Goto W, Takada K, Takashima T, Morisaki T, et al. Prediction of survival after neoadjuvant chemotherapy for breast cancer by evaluation of tumor-infiltrating lymphocytes (TILs) and residual cancer burden (RCB). Cancer Res 2017;77. abstract P02-05-05. Kashiwagi S, Asano Y, Goto W, Takada K, Takahashi K, Noda S, et al. Use of Tumor-infiltrating lymphocytes (TILs) to predict the treatment response to eribulin chemotherapy in breast cancer. Plos One 2017 Feb 6;12(2):e0170634. Ogiya R, Niikura N, Kumaki N, Bianchini G, Kitano S, Iwamoto T, et al. Comparison of tumor-infiltrating lymphocytes between primary and metastatic tumors in breast cancer patients. Cancer Sci 2016 Dec;107(12):1730e5. Dieci MV, Giaratano T, Miglietta F, Griguolo G, Orvieto E, Falci C, et al. Tumor infiltrating lymphocytes in recurrent HER2þ and triple negative breast cancer: prognostic value according to tumor phenotype. Cancer Res 2017;77. abstract P02-05-20. Denkert C, Liedtke C, Tutt A, von Minckwitz G. Molecular alterations in triplenegative breast cancerdthe road to new treatment strategies. Lancet 2017 Jun 17;389(10087):2430e42.  NF, Solinas C, de Azambuja E. Adjuvant trastuzumab: a Lambertini M, Ponde 10-year overview of its benefit. Expert Rev Anticancer Ther 2017 Jan;17(1): 61e74. Schmid P, Cruz C, Braiteh FS, Eder JP, Tolaney S, Kuter I, et al. Atezolizumab in metastatic TNBC (mTNBC): long-term clinical outcomes and biomarker analyses. Abstract #2986. AACR; 2017. Dirix LYY, Takacs I, Nikolinakos P, Jerusalem G, Arkenau H-T, Hamilton EPP, et al. Avelumab (MSB0010718C), an anti-PD-L1 antibody, in patients with locally advanced or metastatic breast cancer: a phase Ib JAVELIN solid tumor trial. Cancer Res 2016;76(4 Suppl). abstract S1-04.