Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

ESMO 2017: TIL Levels Signal Pembrolizumab Response in Metastatic TNBC

TIL levels in tumour samples associate with PD-L1 expression and pembrolizumab response in metastatic triple negative breast cancer
09 Sep 2017
Breast cancer;  Cancer Immunology and Immunotherapy

Patients with metastatic triple negative breast cancer (TNBC) are more likely to show a strong response to pembrolizumab if more tumour infiltrating lymphocytes (TILs) are present in the tumour microenvironment according to findings presented at ESMO 2017, the Annual Congress of the European Society for Medical Oncology in Madrid, Spain.

Sherene Loi, Lab Head and Medical Oncologist of the Peter MacCallum Cancer Center in Melbourne, Australia, investigated TILs, which have been observed in TNBC and may represent pre-existing antitumour immunity, as a biomarker for response to immunotherapy in TNBC. Using data from the phase II KEYNOTE-086 study (NCT02447003) of pembrolizumab monotherapy for patients with previously treated metastatic TNBC and any PD-L1 expression (cohort A) and patients with previously untreated, PD-L1–positive metastatic TNBC (cohort B), the investigators determined the relationship between the quantity of stromal TILs and response to pembrolizumab.

Stromal TILs were quantified from tumour biopsies using haematoxylin and eosin (H&E)-stained slides according to a published method by one pathologist who was blinded to the clinical data, and PD-L1 expression was assessed by immunohistochemistry (IHC) and the PD-L1 IHC 22C3 pharmDx (Agilent Technologies, Carpinteria, CA, USA).

The response to pembrolizumab was assessed by central review according to RECIST v1.1 every 9 weeks for 12 months, then every 12 weeks.

Significantly higher TIL levels were observed in samples obtained from previously untreated, PD-L1–positive tumours and in lymph versus non-lymph node samples

Of the first 222 patients enrolled, 193 patients had evaluable tumour samples; 147 samples were obtained from cohort A (any level of PD-L1 expression) and 46 samples were from cohort B (PD-L1-positive only). Of these, 146 samples were newly collected and derived mostly from metastatic sites, and the 47 archival samples represented mostly primary breast tumours.

Stromal TIL levels were higher in cohort B patients who had previously untreated PD-L1–positive metastatic TNBC; the median stromal TIL level (interquartile range; IQR) was 17.5% (5 to 61.25%) in cohort B compared to 5% (1 to 10%) in cohort A (one-sided Wilcoxon rank sum p < 0.001). Stromal TIL levels were also higher in archival tumour samples versus newly collected samples at 10% (5 to 40%) versus 5% (1 to 15%; p < 0.001) and in lymph node versus non-lymph node samples at 10% (5 to 50%) versus 5% (2 to 16.25%; p = 0.01).

TIL levels emerged as a marker for higher response to pembrolizumab

The relationship between the objective response rate (ORR) and stromal TIL levels were assessed by logistic regression adjusted for cohort (A versus B) and LDH concentration (continuous).

Pembrolizumab response showed a significant relationship to stromal TIL levels; the ORR in patients with stromal TIL level greater than or equal to median as opposed to less than median was 6.4% versus 1.9% in cohort A, and 39.1% versus 8.7% in cohort B. The median (IQR) stromal TIL level in responders compared to non-responders was 10% (5 to 30%) versus 5% (1 to 10%), respectively, in cohort A and 50% (35 to 70%) versus 15% (5 to 40%), respectively, in cohort B.

In the combined cohorts, higher stromal TIL levels were associated with significantly improved ORR as a continuous variable (adjusted odds ratio 1.02; 95% confidence interval [CI] 1.00 to 1.04 [1 sided p = 0.014]). Area under the ROC curve was 0.784.

The investigators determined that PD-L1 expression assessed by the combined positive score significantly correlated with TIL levels (ρ = 0.496; one sided p < 0.001) but did not add significantly more information to a predictive model that included stromal TILs, LDH concentration, and study cohort.


The authors stated that TIL levels, as quantified on a simple H&E stained slide, can identify patients with metastatic TNBC who have a greater chance of achieving response to pembrolizumab monotherapy, particularly in the first-line setting.

Giuseppe Curigliano of the Istituto Europeo di Oncologia and University of Milan, Milan, Italy who discussed the study results summarized what did we learn: TILs assessed by H&E were significantly associated with response to pembrolizumab, particularly in the first-line setting; TILs levels, LDH concentration, and cohort were independent predictors of response to pembrolizumab monotherapy in this study; TILs and PD-L1 CPS showed significant positive correlation; and TILs can identify patients with metastatic TNBC with a greater chance of responding to pembrolizumab monotherapy. He summarised the strenghts of the study as following: TILs should be routinely assessed in primary tumour (if TN or HER2 positive), TILs should be a stratification factor for trials using immunotherapy in breast cancer, TILs reflect T cell priming, T cell attraction and potentially foreignness (mutational burden). The limits are that we are missing composition and phenotype of TILs, different immune-contexture and immune-landscape in different organs, need to assess tertiary lymphoid structures, principally located in the peri-tumoural stroma.


This trial was sponsored by Merck & Co., Inc., Kenilworth, NJ, USA.


LBA13 – Loi S, et al. Relationship between tumor infiltrating lymphocyte (TIL) levels and response to pembrolizumab (pembro) in metastatic triple-negative breast cancer (mTNBC): results from KEYNOTE-086. 

Last update: 09 Sep 2017

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.