Customize your JAMA Network experience by selecting one or more topics from the list below.
Copyright 2015 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Quantifying tumor-infiltrating lymphocytes (TILs) in breast cancer is a frequently used approach to gain insight into the tumor’s immune activity. In this issue of JAMA Oncology, Salgado and colleagues1 evaluated TIL levels as a predictive and prognostic factor in patients participating in the NeoALTTO trial. This is a well-conducted biomarker study measuring 1 prospectively defined marker in a prospectively collected clinical trial cohort by 2 pathologists, renowned for TIL assessment, who analyzed the complete trial cohort, independently adhering to recently published guidelines.2 The results were combined thereafter. The correlation between the assessments of the 2 pathologists was 74%. Salgado and colleagues1 demonstrate that the presence of stromal TILs assessed in pretherapeutic core biopsies prior to the start of neoadjuvant treatment indicate a higher probability of achieving a pathological complete response (pCR), as well as a higher probability of event-free survival.2 In general, factors indicating a higher probability of achieving a pCR, for example, negative hormone receptor status, are associated with a higher chance of relapse but not with better survival. As a matter of fact, in this analysis the correlation between TIL level and survival is linear whereas the correlation between TIL level and pCR is nonlinear. However, this is not further discussed and it can be argued that it is a chance finding because the steep increase in pCR at 5% TILs seems to be confined to the 2 groups with single anti–human epidermal growth factor receptor 2 (anti-HER2) therapy. A couple of other factors might have influenced these results. First, there seems to be no difference in pCR in the group with 5% or less TILs between trastuzumab and lapatinib treatment. Second, lapatinib-treated patients were excluded from the survival analyses, whereas these patients had been included in the pCR analyses. Third, an important part of the therapy was given after surgery and did not influence the pCR rate but only the survival. Last, the follow-up time for patients with a hormone receptor positive–tumor is rather short, at a median of 3.77 years.
It would have been interesting to see the change in TIL levels within the first 6 weeks when patients only received the anti-HER2 treatment without chemotherapy, especially because after these first 6 weeks patients receiving lapatinib alone had a higher clinical response than those receiving trastuzumab,3 but at the end of neoadjuvant therapy, the lapatinib group had a lower pCR rate and in some studies the lapatinib cohort even had a significantly lower pCR rate.4 This suggests a rapid development of lapatinib resistance, which could be influenced by the local immune system.
Whereas many studies have shown that TILs are a good prognostic factor, TILs seem not yet established as a predictive marker for anti-HER2 treatment. Recent data from the adjuvant Alliance N9831 trial, in which patients received standard anthracycline-taxane chemotherapy with or without trastuzumab, demonstrated a better survival for the group with a tumor of lymphocyte-predominant breast cancer (LPBC) phenotype receiving chemotherapy alone; however, the LPBC phenotype was not predictive for trastuzumab therapy benefit but rather the opposite.5 In fact, patients with an LPBC phenotype tumor did not at all benefit from adding trastuzumab therapy (interaction term for LPBC phenotype, 0.042). There might be a difference between assessing the presence of TILs in a surgical specimen, where it is more likely to assess more B lymphocytes because they are at the invasion front, and in the core biopsies, where it is more likely to assess T cells in the center of the tumor.
The ultimate question remains whether TIL assessment should be integrated into routine histology practice and, if so, what is the clinical value and consequence. Do patients with a high level of TILs (cutoff at 5%, 12.5%, or 40%) or an LPBC phenotype (simply more TILs than tumor cells) need less aggressive treatment or vice versa? Probably neither is the case because the pCR rate is markedly higher in all TIL groups using the double blockade consisting of trastuzumab plus lapatinib in addition to paclitaxel.1 Pathological complete response is an excellent surrogate marker for long-term outcome and remained, together with level of TILs, the only independent prognostic factor in this analysis. It may be that patients achieving a pCR with initially high level of TILs could have been spared the anthracycline-containing postneoadjuvant therapy. But this cannot be answered using these data and remains speculation. Therefore, although TIL assessment is simple, cheap, and can be easily integrated into routine histopathology reporting to assess the prognosis as well as the treatment response, it remains only a semiquantitative marker. An immune gene panel—although single immune markers are highly correlated with TILs6—could outperform TIL assessment.
Corresponding Author: Sibylle Loibl, MD, PhD, German Breast Group, c/o GBG-Forschungs GmbH, Martin-Behaim-Strasse 12, 63263 Neu-Isenburg, Germany (firstname.lastname@example.org).
Published Online: April 30, 2015. doi:10.1001/jamaoncol.2015.0851.
Loibl S. The Dual Role of Tumor-Infiltrating Lymphocytes in Human Epidermal Growth Factor Receptor 2–Positive Primary Breast Cancer: Two Sides of the Same Coin? JAMA Oncol. 2015;1(4):455–456. doi:10.1001/jamaoncol.2015.0851
Create a personal account or sign in to: