Objective Measurement and Significance of TILs in NSCLC
Here we describe a novel method for objective, simultaneous, and compartment-specific measurement of CD3, CD8, and CD20+ TILs that is suitable for conventional FFPE tissue samples. We also show that different TIL subtypes show distinct levels in the tumor and peritumoral stroma and are associated with longer survival in NSCLC. Our findings also indicate that measurement of TILs using this method is more sensitive and provides superior prognostic information than conventional H&E-based pathologist estimation.
Determination of TILs has traditionally been performed in a semiquantitative fashion by using H&E-stained preparations or with single-color IHC. Typically, diverse areas from one slide are evaluated and a trained observer (eg, a pathologist or researcher) renders an integrated categorical estimation of the cells of interest. Also, the examiner estimates the predominance of the cells within certain tissue compartments or structures. For its simplicity and low cost, this approach has been widely used and interesting correlations regarding the biological and clinical role of TILs in diverse tumor types have been established. However, this strategy is subjective, variably reproducible, and lacks capacity for detailed simultaneous characterization of cell subtypes. Also, such determinations are limited by the absence of standardized cutoffs and operational landmarks to assess the level of TILs (eg, use of proportion of cells vs absolute number; evaluate the whole sample vs stroma vs tumor only; score the whole slide or only the invasive front, etc.). This has been recognized by an international taskforce led by Dr. Jerome Galon, which proposed the use of a standardized "Immunoscore" using chromogenic IHC for CD3 and CD8 in serial tumor sections coupled to automated quantification in different tumor areas. However, this method is largely supported by findings in colorectal cancer and not NSCLC, has limited throughput, requires relatively large amounts of tissue and lacks multiple cellular subtyping and compartment-specific capacity. Comparison of this strategy with multiplexed fluorescence will be required to determine the strengths and clinical value of each approach.
Consistent with findings in tumors from different locations and tissue types, increased total TILs has been associated with longer survival in both early-stage and advanced NSCLC. However, studies measuring single-cell subtypes using IHC have reported conflicting results with one showing association between increased CD8+ cytotoxic T cells (but not of CD4+ cells) and longer survival and others showing the opposite result. In addition, Hiraoka et al. reported an absence of survival benefit of either elevated CD8+ or CD4+ TILs alone, but a statistically significant (and independent) prognostic effect of combined high stromal CD8+ and CD4+ in 109 NSCLC samples. However, these reports include relatively small collections of cases from single institutions and without validation in an independent set. Our findings support the notion that the presence of elevated CD3- and CD8-positive T cells is consistently associated with survival, but only CD8 provides independent prognostic information in NSCLC. CD20 signal showed statistically significant association with survival only in one cohort. This is somewhat consistent with another study evaluating CD4, CD8, and CD20+ cells with chromogenic IHC in 335 NSCLC samples represented in TMAs. The authors found that a high stromal density of each marker was statistically significantly associated with longer survival. Also similar to our study, only a high proportion of CD4 and CD8+ cells in this group was independently associated with survival in multivariable analysis. Our results do not include measurements of CD4 because of limited available fluorescence channels.
The limited correlation seen between the markers and in different tumor compartments indicates that the TIL signal is heterogeneous. Nonhomogenous or clustered secretion of cytokines and chemokines known to participate in the recruitment of different TIL subtypes to the tumor microenvironment such as Interferon-γ, CXCL9, CXCL10, and CXCL13 might explain this finding. In particular, our results suggest that CD20+ cells are often located in clusters, leading to a more variable expression pattern that could explain the prognostic effect present only in one of the cohorts. The heterogeneity of TILs seen in our study raises questions regarding the minimal amount of tissue or number of fields/TMA spots required to accurately reflect the tumor-immune contexture. Future studies will be required to determine the minimal tissue sample size for assessment of the TIL markers for clinical purposes.
Our analysis has a number of limitations. One major limitation is that it includes only retrospectively collected cases lacking data on molecular tumor alterations (eg, EGFR, KRAS, ALK, or ROS1). A second issue is that the use of TMAs may underestimate or overestimate the level of TILs because of intratumoral heterogeneity of expression. Clinical assays on actual patient specimens use whole pathology sections (although samples are frequently small). Thus, examination of high numbers of fields of view seen in a histologic section may attenuate the sampling effect of TMAs. Our multiplexing approach is suitable for whole-tissue section samples; it is supported by its use in conventional biopsy slides of human tonsil used as positive control (Figure 1A). Moreover, efforts to characterize TILs in conventional whole-tissue section biopsies from tumor samples have been recently finalized and show comparable results (Brown et al., 2014 [17]). One key unaddressed issue is the potential of TILs, alone or in combination with other markers, to predict response to immunostimulatory therapies such as anti-PD-1/PD-L1 as previously suggested. Future studies measuring TIL subpopulations using QIF in whole-tissue sections specimens from patients treated with such compounds might provide a better substrate to conclusively address the predictive value of these markers.
Discussion
Here we describe a novel method for objective, simultaneous, and compartment-specific measurement of CD3, CD8, and CD20+ TILs that is suitable for conventional FFPE tissue samples. We also show that different TIL subtypes show distinct levels in the tumor and peritumoral stroma and are associated with longer survival in NSCLC. Our findings also indicate that measurement of TILs using this method is more sensitive and provides superior prognostic information than conventional H&E-based pathologist estimation.
Determination of TILs has traditionally been performed in a semiquantitative fashion by using H&E-stained preparations or with single-color IHC. Typically, diverse areas from one slide are evaluated and a trained observer (eg, a pathologist or researcher) renders an integrated categorical estimation of the cells of interest. Also, the examiner estimates the predominance of the cells within certain tissue compartments or structures. For its simplicity and low cost, this approach has been widely used and interesting correlations regarding the biological and clinical role of TILs in diverse tumor types have been established. However, this strategy is subjective, variably reproducible, and lacks capacity for detailed simultaneous characterization of cell subtypes. Also, such determinations are limited by the absence of standardized cutoffs and operational landmarks to assess the level of TILs (eg, use of proportion of cells vs absolute number; evaluate the whole sample vs stroma vs tumor only; score the whole slide or only the invasive front, etc.). This has been recognized by an international taskforce led by Dr. Jerome Galon, which proposed the use of a standardized "Immunoscore" using chromogenic IHC for CD3 and CD8 in serial tumor sections coupled to automated quantification in different tumor areas. However, this method is largely supported by findings in colorectal cancer and not NSCLC, has limited throughput, requires relatively large amounts of tissue and lacks multiple cellular subtyping and compartment-specific capacity. Comparison of this strategy with multiplexed fluorescence will be required to determine the strengths and clinical value of each approach.
Consistent with findings in tumors from different locations and tissue types, increased total TILs has been associated with longer survival in both early-stage and advanced NSCLC. However, studies measuring single-cell subtypes using IHC have reported conflicting results with one showing association between increased CD8+ cytotoxic T cells (but not of CD4+ cells) and longer survival and others showing the opposite result. In addition, Hiraoka et al. reported an absence of survival benefit of either elevated CD8+ or CD4+ TILs alone, but a statistically significant (and independent) prognostic effect of combined high stromal CD8+ and CD4+ in 109 NSCLC samples. However, these reports include relatively small collections of cases from single institutions and without validation in an independent set. Our findings support the notion that the presence of elevated CD3- and CD8-positive T cells is consistently associated with survival, but only CD8 provides independent prognostic information in NSCLC. CD20 signal showed statistically significant association with survival only in one cohort. This is somewhat consistent with another study evaluating CD4, CD8, and CD20+ cells with chromogenic IHC in 335 NSCLC samples represented in TMAs. The authors found that a high stromal density of each marker was statistically significantly associated with longer survival. Also similar to our study, only a high proportion of CD4 and CD8+ cells in this group was independently associated with survival in multivariable analysis. Our results do not include measurements of CD4 because of limited available fluorescence channels.
The limited correlation seen between the markers and in different tumor compartments indicates that the TIL signal is heterogeneous. Nonhomogenous or clustered secretion of cytokines and chemokines known to participate in the recruitment of different TIL subtypes to the tumor microenvironment such as Interferon-γ, CXCL9, CXCL10, and CXCL13 might explain this finding. In particular, our results suggest that CD20+ cells are often located in clusters, leading to a more variable expression pattern that could explain the prognostic effect present only in one of the cohorts. The heterogeneity of TILs seen in our study raises questions regarding the minimal amount of tissue or number of fields/TMA spots required to accurately reflect the tumor-immune contexture. Future studies will be required to determine the minimal tissue sample size for assessment of the TIL markers for clinical purposes.
Our analysis has a number of limitations. One major limitation is that it includes only retrospectively collected cases lacking data on molecular tumor alterations (eg, EGFR, KRAS, ALK, or ROS1). A second issue is that the use of TMAs may underestimate or overestimate the level of TILs because of intratumoral heterogeneity of expression. Clinical assays on actual patient specimens use whole pathology sections (although samples are frequently small). Thus, examination of high numbers of fields of view seen in a histologic section may attenuate the sampling effect of TMAs. Our multiplexing approach is suitable for whole-tissue section samples; it is supported by its use in conventional biopsy slides of human tonsil used as positive control (Figure 1A). Moreover, efforts to characterize TILs in conventional whole-tissue section biopsies from tumor samples have been recently finalized and show comparable results (Brown et al., 2014 [17]). One key unaddressed issue is the potential of TILs, alone or in combination with other markers, to predict response to immunostimulatory therapies such as anti-PD-1/PD-L1 as previously suggested. Future studies measuring TIL subpopulations using QIF in whole-tissue sections specimens from patients treated with such compounds might provide a better substrate to conclusively address the predictive value of these markers.
SHARE
![Pediatric Supportive Care (PDQ®): Supportive care - Patient Information [NCI]-Overview](https://cdnus.oss-us-west-1.aliyuncs.com/libs/images/NewsJpg/cate_6/072_215x160.jpg)
