Tumor-Infiltrating Lymphocytes in Patients Receiving Trastuzumab/Pertuzumab-Based Chemotherapy


Michail Ignatiadis; Gert Van den Eynden; Salgado Roberto; Marco Fornili; Yacine Bareche; Christine Desmedt; Françoise Rothé; Marion Maetens; David Venet; Esther Holgado; Virginia McNally; Astrid Kiermaier; Heidi M. Savage; Timothy R. Wilson; Javier Cortes; Andreas Schneeweiss; Karen Willard-Gallo; Elia Biganzoli; Christos Sotiriou


J Natl Cancer Inst. 2019;111(1):69-77. 

In This Article


To our knowledge, this is the first study to evaluate associations between TIL percentage at baseline, pCR, and EFS following neoadjuvant chemotherapy and dual HER2 blockade with trastuzumab and pertuzumab in HER2-amplified early breast cancer. We provide for the first time evidence that TIL percentage at baseline and pCR provide independent prognostic information for EFS in patients treated with trastuzumab/pertuzumab-based neoadjuvant treatment. For every 10% increase in the level of baseline TILs, there was a 25% reduction in the occurrence of EFS events. These results need further validation in an independent patient cohort. Patients with a high baseline TIL percentage (≥14%) and pathological complete response had an excellent prognosis following neoadjuvant chemotherapy and dual HER2 blockade with trastuzumab and pertuzumab. Similar results to this association between baseline TIL percentage, pCR, and EFS for the trastuzumab and lapatinib combination have been reported in the NeoALTTO trial.[2]

The results of our study may have clinical implications for tailoring treatment in patients with HER2-amplified early breast cancer. For example, in 2013, pertuzumab was approved in combination with trastuzumab and docetaxel as a neoadjuvant treatment in patients with HER2-amplified breast cancer.[18] Recently, the Aphinity trial demonstrated that pertuzumab, when added to trastuzumab as an adjuvant treatment, improved invasive disease–free survival by 1%.[19] Due to the marginal benefit of the Aphinity trial, there is an urgent need to identify the patient subgroup that can derive most benefit from the addition of pertuzumab. For Aphinity, the subgroups predefined by clinic-pathological factors that derived the highest benefit were patients with node-positive disease and/or hormone receptor–negative disease;[19] the latter was also seen in the Tryphaena study. If the results of our trial on the association between TIL percentage, pCR, and EFS are further confirmed in other studies or patient cohorts that have received neoadjuvant pertuzumab-based treatment, one might consider adding pertuzumab in high-risk HER2-positive patients as part of neoadjuvant treatment, as in the Tryphaena trial. Thus, in patients with a high baseline level of TIL percentage that achieve pCR after neoadjuvant chemotherapy and dual HER2 blockade with trastuzumab and pertuzumab, trastuzumab alone as adjuvant treatment might be the appropriate treatment due to the very good prognosis for this group. In the other patient subgroups, further clinical research is needed to identify optimal approaches that might further improve outcome (eg, evaluate the role of immunotherapy in this setting).

In our study, there was no association between TIL percentage at the point of surgery where residual invasive disease was present and EFS. However, our study was largely underpowered to detect such associations. In patients with triple-negative breast cancer (TNBC) receiving neoadjuvant chemotherapy, the presence of high levels of TILs (>60%) in residual invasive disease was associated with improved EFS compared with patients with low levels of TILs.[20] More studies with adequate sample size are needed to evaluate whether such an association exists in HER2-positive breast cancer. Moreover, it is not known whether patients with TILs at residual invasive disease might benefit from immune checkpoint inhibitors. TILs have been associated with a response to the immune checkpoint inhibitor pembrolizumab in metastatic TNBC.[21] It has also been suggested that the mutational load,[22] the number of neoantigens,[23] and neoantigen heterogeneity (clonal vs subclonal neoantigens)[24] are associated with response to immune checkpoint inhibitors. Increased clonal neoantigen burden has been associated with increased benefit from immune checkpoint inhibitors in lung adenocarcinoma.[24] Thus, the information on the presence of clonal neoantigens but also a more detailed characterization of the immune landscape[25,26] of residual disease following neoadjuvant chemotherapy and dual HER2 blockade might provide more insight into the subgroup of patients that could benefit from immunotherapeutic approaches in this setting.

Additionally, we observed higher pCR rates for patients in the HER2-enriched subtype compared with other subtypes and for patients with higher ERBB2 mRNA levels, as has been observed in the CALGB 40601[13] and NeoALTTO[4] studies.

Moreover, we demonstrated that after adjustment for baseline clinicopathological characteristics and baseline TIL percentage and correction for multiple testing, increased pCR rates were associated with high expression of immune signatures related to tertiary lymphoid structures and a signature associated with benefit from adjuvant trastuzumab, as well as with high expression of genes associated with CD8+ T-cell cytolytic activity (GZMA, GZMB, PRF1) and with T-helper phenotype 1 (IFNG). Our results extend previously reported associations between pCR and IFNG in patients who received anti-HER2 treatment in the context of the Neosphere trial.[5] The fact that immune signatures and single genes were not significantly associated with EFS in our study can be related to the fact that our study is underpowered to detect such associations; thus further studies are needed in this respect.

Furthermore, in order to better understand TIL biology, we evaluated associations between TIL percentage at baseline and the expression of 800 genes related to breast cancer. As expected, increasing levels of TILs were positively correlated with immune response genes. Interestingly, TILs were inversely correlated with the expression of genes associated with EMT, angiogenesis, and T-cell inhibition, such as SNAIL1, ZEB1, NOTCH3, and B7-H3. Our results are in line with preclinical evidence suggesting that snail-induced EMT promotes immunosuppression in melanoma[27] and that mesenchymal stromal cells can suppress the immune response.[28]

Our study has some limitations. There were only 64 patients with invasive residual disease and evaluable TILs; thus the study on the association between TILs at surgery and EFS is largely underpowered. Additionally, concerning the associations between baseline TILs and either pCR or EFS, all patients received dual HER2 blockade; thus, we cannot explore single vs dual HER2 blockade biomarker questions. However, this is the first study to explore association of TILs at diagnosis with pCR and EFS for patients with HER2-positive early breast cancer who have received standard preoperative chemotherapy with trastuzumab and pertuzumab. Moreover, we acknowledge that, currently, there is not a validated cutoff for low vs high TIL percentage in this setting. Finally, our gene expression data are hypothesis-generating and need further validation.

In summary, evaluation of baseline TILs might help tailor treatment options for high-risk HER2-positive breast cancer patients who receive neoadjuvant chemotherapy and dual HER2 blockade with trastuzumab and pertuzumab. Further studies are needed to validate our results.