Tumor infiltrating lymphocytes: current pathways to a standard biomarker in breast cancer

KEYWORDS:

• Tumor-infiltrating lymphocytes
• biomarkers
• breast cancer
• tumor immune microenvironment

1. Introduction

Our understanding of breast cancer (BC) has increased significantly over recent years, with evolving knowledge on the genomic, biological, and physiological alterations associated with tumor progression. These dynamic processes are reflected in tumor biomarkers, which aid cancer diagnosis, directing therapy, and estimating prognostic endpoints in the clinic. Biomarkers obtained from biospecimens, and more recently, from quantitative imaging, have demonstrated exponential growth with advances in laboratory techniques, imaging technology, and computational capacity. Over the past decade, this has afforded new opportunities to better characterize tumor intrinsic subtypes and identify individualized factors associated with treatment refractory and recurrent breast cancer.

Among promising biomarkers, tumor-infiltrating lymphocytes (TILs) have been the focus of substantial research, with growing evidence to demonstrate their association to aggressive and highly proliferative breast cancer phenotypes. TILs are involved in multiple steps of cancer immunoediting that drives tumorigenesis and constitute cytotoxic and helper T cells, B cells, and natural killer (NK) cells [1–3]. During the early phases of carcinogenesis, recognition and clearance of malignant cells are mediated by TILs and other immune infiltrates. Subsequently, escape from immune-driven cell death ensues, as tumor cells develop an adaptive mechanism to evade immunosurveillance; this is achieved, in part, by disrupting chemokines that control ‘eat me’ signaling in the microenvironment. Immunoediting elicits increased recruitment of TILs, localized predominantly to the stroma. These dynamic changes effectively modulate the tumor immune microenvironment (TIME) to sustain the growth of the neoplasm. It has become clear from previous studies that the level of stromal TILs is correlated to breast cancer prognosis and can serve as a predictive marker of therapy response [4,5]. This has been extensively studied in triple-negative (TNBC) and HER2-positive breast cancer subtypes.

2. TIL biomarkers in breast cancer

Measuring TILs and characterizing their temporal changes to monitor or predict therapy response is an area of clinical importance; this has been well studied in neoadjuvant chemotherapy (NAC) for high-risk breast cancer. A recent pooled analysis by Denkert et al. demonstrated that TILs assessed prior to NAC are predictive of drug response. Tumors with high TILs had the highest rate of pathologic complete response to systemic therapy irrespective of the molecular subtype [5]. Interestingly, univariate analysis showed that only patients with triple-negative tumors and HER2-positive breast cancer achieved longer disease-free survival for every 10% increase in TILs [5]. In another study, Hamy et al. measured the changes in TILs by comparing paired samples from the pretreatment biopsies to the post-treatment excisional specimens. Larger decreases in TILs were associated with a pathologic complete response in HER2+ breast cancer. Tumors that demonstrated >25% TILs after NAC carried a greater risk of exhibiting a higher Residual Cancer Burden (RCB) score (HR = 3.23 (p = 0.03)) in univariate analysis [6]. In the TNBC-setting, higher TIL levels post-NAC were associated with prolonged outcome (i.e. recurrence-free survival and overall survival) and added further prognostic information to RCB, particularly in RCB class II [7]. Moreover, high on-treatment TILs are associated with a clinical benefit in immunotherapy-treated TNBC patients and in patients with HER2+ breast cancer-treated anti-HER2 targeted therapy [8,9]. In the adjuvant setting, TILs have also been studied as a prognostic marker. Loi and colleagues conducted a pooled analysis of 2148 patients with early-stage TNBC. Stromal TILs (sTILs) were investigated as a biomarker for prognostic endpoints including invasive disease-free survival (iDFS), distant disease-free survival (D-DFS), and overall survival (OS) [10]. The quantity of sTILs was a significant parameter in multivariable analysis; an increase of 10% sTILs corresponded to a hazard ratio (HR) of 0.87 for iDFS, HR = 0.83 for D-DFS, and HR = 0.84 for OS (HR = 0.84). There was also a strong association between higher tumor grade and increased sTIL values (p = 0.001) [10]. This study demonstrates that the quantity of sTILs measured at diagnosis is a prognostic indicator for survival in triple-negative disease and is equivalent to biomarker level of evidence IB. Prospective clinical trials are being developed that will investigate whether TILs can be used as a binary variable to de-escalate chemotherapy options and thus obtain levels of evidence IA. Until that time, TILs should be used for prognostic purposes as a complement to other prognostic variables in TNBC such as tumor size and lymph node status. For predictive purposes, TIL assessment can be used as a complement to PDL1-assays, as all these assays in breast cancer rely on the TILs [11].

3. Translational pathways

Despite multiple clinical studies demonstrating the predictive and prognostic values of TILs in breast cancer, routine implementation in standard practice is a slow process. In some ways, this is beneficial to identify the strengths and weaknesses and to establish cogent methods for biomarker analysis. Indeed, only 3–5% of candidate markers make it to the clinic [12]. Establishing standard guidelines, coordinating studies, building scientific collaborations, developing accessible (online) clinical tools, and including patients, industry, and the regulatory authorities are among several driving mandates of the not-for-profit International Immuno-Oncology Biomarker Working Group (www.tilsinbreastcancer.org). Evidentiary standards need to be met for clinical adoption [13,14]; some of these criteria have been addressed in previous TIL analyses, including (1) establishing standardized evaluation practices that are repeatable and reproducible [15], (2) conducting large-scale, well-powered studies, and (3) defining biomarker indications and identifying the target population for the intended use. Other considerations for widespread application include analysis to ascertain an improvement in care approaches (i.e. clinical utility), such as cost-effectiveness and a demonstration that the investigational biomarker does not yield a lower performance than other predicate markers.

The success of biomarkers in the clinic may also be attributed to readily accessible techniques that can be carried out with existing infrastructure in the pathology department. Pathology laboratories worldwide use hematoxylin–eosin (HE) slides for each diagnosis. Due to the general availability of this technique, it has also permitted access to analyzing TILs in low-to-middle-income countries as a screening tool before activating costly PDL1-assays. Wein and colleagues recently outlined three major priorities for TILs as a standard biomarker [16]. First, TILs may be valuable predictive biomarkers for immunotherapies. There is growing evidence to show an association between response to immune checkpoint inhibitors and TIL quantity; this interaction has been studied in early clinical trials as well as in late-stage phase 3 TNBC-trials [17–19]. To obtain definitive evidence on the clinical utility of TILs as a biomarker for selection of patients for immunotherapies in daily practice, clinical trials prespecifying TILs in the design as an integral biomarker are therefore imperative. Second, TILs as a surrogate marker for prognosis can aid in risk stratification of triple-negative and HER2+ breast patients; this can potentially help guide de-escalation treatment strategies for low-risk patients. Ongoing studies are investigating TILs as a variable to down- or upgrade stage in TNBC. There is preliminary data (unpublished observations) that stage I patients with low TILs have a worse outcome than stage II patients with a high number of TILs. If validated, this may be the practice changing information needed as clinicians rely on stage to determine their treatments. Lastly, TILs can be integrated into other prognostic models, such as the residual cancer burden index (RCBI). This can potentially refine prognostic signatures and aid drug therapy strategies in the neoadjuvant and adjuvant settings [16].

4. Regulatory pathways

Looking beyond well-designed studies and the clinical evidence to date, acceptance of TIL biomarkers in oncologic practice requires peer consensus between all stakeholders, namely, pathologists, oncologists, patients, scientists, industry, and the regulatory authorities. Subsequently, assays may be incorporated into standard guidelines from professional organizations, such as the American Society of Clinical Oncology (ASCO) and the European Medical Oncology Association (ESMO). Depending on the indications and intended use of TILs, attaining regulatory acceptance is needed according to the regional policies, for example, through health agencies such as the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA). It remains unknown, however, regarding the regulatory framework needed for TILs, and this varies regionally. Within the context of drug development studies, health agencies play an important role in overseeing processes related to drug development, including the use of biomarkers to evaluate drug efficacy. Applications to the biomarker qualification program through the FDA would be required to define the context of use (COU) and to meet the rigorous evidentiary standards for review prior to approval [12,13]. Other regulatory pathways include 510(k) programs administered by the FDA to evaluate the intended use of predictive, prognostic, and theragnostic biomarker assays. An integral step in this process includes risk analysis for patients concerning false negatives or false positives; in essence, applications must demonstrate substantial evidence for good biomarker performance for clinical endpoints [20]. Taken together, achieving these regulatory approvals may boost confidence in TILs as a standard biomarker and advance routine incorporation into prospective drug-driven clinical trials by the pharmaceutical industry and everyday practice.

5. Conclusion and future perspectives

In conclusion, there has been substantial work in the assessment of TILs, which has been coupled with a concerted international effort. Previous studies have shown the validity of TILs to characterize breast cancer immune phenotypes, as well as their importance for more in-depth multiplex analysis of the immune infiltrate, as well as their utility as a predictive and prognostic biomarker. Interestingly, HE-slides remain as the starting point for assessing immune infiltrates in breast tumors. With the overarching aim of improving oncologic management and ultimately patient outcomes, the direction moving forward will be to drive widespread access to assessment tools, incorporate TIL evaluation into emerging clinical trials, and work with regional health agencies to carry out regulatory approval (where applicable). Although TILs are still not yet fully realized into the diagnostic and treatment pathway, there are indeed indicators that point to a promising assay to enhance precision oncology strategies. Only unbiased quality of data over speed can help in the correct implementation and use of TILs in our current and future clinical trials and daily practices.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

A reviewer on this manuscript has received travel support from Pfizer, Novartis, Amgen and Daiichi Sankyo. The remaining reviewers have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

WT Tran and FI Lu have received grant funding from the Tri-Council (CIHR) Government of Canada’s New Frontiers in Research Fund (NFRF, Grant # NFRFE-2019-00193). WT Tran has also received funding from the Terry Fox Research Institute (TFRI, Grant #1083), the Women’s Health Golf Classic Foundation Fund, and the CAMRT Research Grant (Grant #2021-01).