Lung Cancer Subtypes Generate Unique Immune Responses

Abstract

Lung cancer, the leading cause of cancer-related deaths worldwide, is a heterogeneous disease comprising multiple histologic subtypes that harbor disparate mutational profiles. Immune-based therapies have shown initial promise in the treatment of lung cancer patients but are limited by low overall response rates. We sought to determine whether the host immune response to lung cancer is dictated, at least in part, by histologic and genetic differences, because such correlations would have important clinical ramifications. Using mouse models of lung cancer, we show that small cell lung cancer (SCLC) and lung adenocarcinoma (ADCA) exhibit unique immune cell composition of the tumor microenvironment. The total leukocyte content was markedly reduced in SCLC compared with lung ADCA, which was validated in human lung cancer specimens. We further identified key differences in immune cell content using three models of lung ADCA driven by mutations in Kras, p53, and Egfr Although Egfr-mutant cancers displayed robust myeloid cell recruitment, they failed to mount a CD8⁺ immune response. In contrast, Kras-mutant tumors displayed significant expansion of multiple immune cell types, including CD8⁺ cells, regulatory T cells, IL-17A-producing lymphocytes, and myeloid cells. A human tissue microarray annotated for KRAS and EGFR mutations validated the finding of reduced CD8⁺ content in human lung ADCA. Taken together, these findings establish a strong foundational knowledge of the immune cell contexture of lung ADCA and SCLC and suggest that molecular and histological traits shape the host immune response to cancer.

Figure 1

Egfr, Kras and Kp53 mice develop lung tumors and associated inflammation. (A, B) All models chronologically develop atypical alveolar hyperplasia, adenoma, and adenocarcinoma 6, 10, and 14 weeks post tumor induction. Normal lung from a non-tumor bearing wild-type mouse is depicted in the lower right corner. H&E sections, scale bars = 2 mm (A) and 500 μm (B), except lower wild-type panel = 1 mm. (C) Spectrum of disease in murine ADCA models. Data are presented as percent of mice exhibiting ≥ 1 indicated lesion at each time point post-induction (n ≥ 5 mice per group). All genotypes exhibited hyperplasia at all time points examined (not shown). Analysis of 14-week Kras ^LSL/+;Trp53^Fl/Fl mice was precluded by early mortality. (D) Percent tumor area was calculated at the indicated time points in a minimum of at least 3 representative lungs from Egfr, Kras, and Kp53 mice. (E) Body mass of tumor-bearing female mice (n ≥ 5) compared to non-tumor bearing littermate controls (n ≥ 3) at each time point post-induction.

Figure 2

Flow cytometric analysis of the inflammatory response to lung tumorigenesis. (A) Representative dot plots demonstrate the strategy used to characterize the mouse lung TME. Single cell gates (not shown) were initially applied to remove doublet cells. All subsequent gating utilized a viability marker followed by gating on the CD45⁺ population. Ly6G identified neutrophils (PMN), while remaining myeloid cells were classified as macrophage (Mac; SiglecF⁺CD11c⁺), eosinophil (Eos; SiglecF⁺CD11c⁻) or monocyte (Mono; SiglecF^loCD11b^hiLy6C⁺) from the Ly6G⁻ population. A size gate was applied for lymphocyte analysis followed by staining to identify B cells (CD3⁻CD19⁺), T cells (CD3⁺CD19⁻), and NK cells (CD3⁻CD19⁻NK1.1⁺). T cells were further classified into γδ T cells (γδTCR⁺), CD4 cells (CD4⁺CD8⁻), and CD8 cells (CD8⁺CD4⁻). T cell subtypes were identified as TH1 (CD4⁺IFNγ⁺), Treg (CD4⁺CD25⁺FoxP3⁺), and IL17A-producing T cells (CD3⁺IL17A⁺). Major lung immune cell populations in (B) Egfr mice 10 weeks post tumor induction (n ≥ 6), (C) Egfr at 14 weeks (n ≥ 4), (D) Kras at 10 weeks (n = 14), (E) Kras at 14 weeks (n = 11), (F) Kp53 at 6 weeks (n = 8), and (G) Kp53 at 10 weeks (n ≥ 8), compared to non-tumor bearing control mice (white bars, n ≥ 3). Early and late time points for each genotype are depicted with gray and black bars, respectively. Data for each cell type are displayed as the total number of live cells present within the mouse lung. (H) NKG2D median fluorescence intensity (i.e. Med.F.I.) on NK cells was examined in Egfr, Kras, and Kp53 mutant lungs compared to normal lung controls at indicated time points (n ≥ 3 per group). Asterisks indicate P < 0.05.

Figure 3

Oncogenic drivers dictate lymphocyte recruitment into the ADCA microenvironment. CD3⁺ T lymphocyte subpopulations in (A) Egfr mice 10 weeks post tumor induction (n ≥ 6), (B) Egfr at 14 weeks (n = 6), (C) Kras at 10 weeks (n ≥ 10), (D) Kras at 14 weeks (n = 11), (E) Kp53 at 6 weeks (n = 8), and (F) Kp53 at 10 weeks (n ≥ 6), compared to non-tumor bearing control mice (white bars, n ≥ 4). Data for each cell type are displayed as the total number of live cells present within the mouse lung. (G-J) CD3 immunostaining revealed that tumor-associated T cells (G, arrowheads) were commonly located at the edges of neoplastic lesions. Infiltration of CD3⁺ T cells into the tumor mass is indicated with arrows (H); note that large clusters of TA CD3⁺ cells were also present in the periphery of this lesion. CD3⁺ lymphoid aggregates formed within the vicinity of a tumor (I, dashed line, Tu) and were typically associated with airways and/or blood vessels. Although FoxP3⁺ Tregs were seldom observed infiltrating tumor masses (J), they constituted a significant portion of cells present in lymphoid aggregates. Scale bar = 250 μm. (K-L) Quantification of immune cell localization in matched tumor burden 14-week Egfr (n = 6), 10-week Kras (n = 5), and 6-week Kp53 (n = 5) mice. Results are expressed as the percent of lung lobes that contained at least one occurrence of (K, left) tumor-associated CD3⁺ cells, (center) tumor-infiltrating CD3⁺ cells, (right) CD3⁺ cells in lymphoid aggregates, or (L, left) tumor-associated FoxP3⁺ cells, (center) tumor-infiltrating FoxP3⁺ cells, or (right) FoxP3⁺ cells in lymphoid aggregates. (M) Expression of cytokine and chemokine genes in tumor-bearing lungs from 14-week Egfr and 10-week Kras mice (n = 4 per genotype). Data are presented as mean 1/ΔCt with 95% CI. Asterisks indicate P < 0.05.

Figure 4

CD8⁺ cell content and function correlates with lung ADCA subtype. (A) The number of CD8⁺ cells/mm² was tabulated for each core section present on a TMA of lung ADCA cases annotated for EGFR and KRAS mutational status. Cases were ranked from lowest to highest CD8 content prior to unblinding for genotype. Shown are representative images of EGFR- (left) and KRAS-mutant (right) ADCA. Scale bar = 100 μm. 65.4% (17 of 26) of EGFR-mutant cases were scored as CD8-low versus 41.5% (22 of 53) of KRAS-mutant cases (Fisher's exact test, P = 0.0392). (B) PDL1 Med.F.I. was assessed on pulmonary EpCAM⁺ epithelial cells and macrophages from 10-week Egfr (n = 5) and 14-week Kras mice (n = 4). Expression compared to normal lung (n ≥ 4) is shown in bottom panels. (C) Splenocytes from non-tumor bearing wild-type mice were labeled with CFSE and incubated with protein homogenate generated from wild-type normal lung (NL) or tumor-bearing lung from 10 week Kras or Egfr mice, or with media alone. The cells were subsequently stained with anti-CD8 and a viability marker and analyzed for CFSE intensity; representative plots are shown. For each genotype (n ≥ 3) the percentage of proliferating CD8⁺ T cells was determined after normalization to the media control. Statistical differences were assessed by one-way ANOVA with Tukey's post-test. (D) Flow cytometric analysis of T cell function in Egfr mice compared to wild-type control, gated from single, live, CD45⁺CD3⁺ parent population. Lymphocytes were gated as CD62L⁺CD44⁻ (i.e. Naïve), CD62L⁺CD44⁺ (T_CM, central memory), and CD62L⁻CD44⁺ (T_EM/Eff, effector memory/effector). PD1 expression was assessed on CD8⁺ T cells only. CD8⁺ and CD4⁺ T cell populations were examined at 10 (E, F) and 14 weeks (G, H) post-induction of mutant Egfr (n = 6 tumor-bearing lungs and n ≥ 3 controls per group), respectively. (I) Percent lung tumor area of Kras mice treated with anti-CTLA4 (n = 7) or isotype (n = 6) with representative H&E sections. Scale bar = 500 μm. (J, K) Flow cytometric analysis of CD8⁺ and CD4⁺ T cell populations in anti-CTLA4 and isotype-treated Kras mice (n = 5 per group). Asterisks indicate P < 0.05.

Figure 5

IL17A cytokine production and impact in Kras mutant lung ADCA. (A) Representative dot plot demonstrating the relative production of IL17A by γδ T and non-γδ T cells; gated from single, live, CD45⁺CD3⁺ parent population. (B) Quantification of IL17A cytokine's cellular source in 14 week Kras tumor-bearing lungs (n = 5). (C-F) Spontaneously arising lymphomas occurred at high frequency in K.RORγt animals, commonly impacting (C) thymus, (D) spleen, (E) liver, and (F) lung tissues. H&E sections, scale bars = 250 μm, except lung panel (F) = 500 μm. (G) Representative H&E images from 14-week Kras and Kras.Tgd mice. Scale bar = 500 μm. (H) Flow cytometric analysis of lungs from tumor-bearing Kras and Kras.Tgd mice (n = 6 each) at 14 weeks post-induction. Data for each cell type are displayed as the total number of live cells present within the mouse lung. Asterisks indicate P < 0.05.

Figure 6

Tumor-associated inflammation in SCLC. (A) Rbp53 mice develop SCLC tumors within 1 year of AdCre exposure. H&E section, scale bar = 1 mm. (B) Flow cytometric analysis of lungs from tumor-bearing Rbp53 mice (n = 4) and non-Cre exposed control mice (n = 3), approximately 10 months after tumor induction. IHC staining reveals that CD45⁺ immune cells are generally located in the SCLC tumor periphery (arrowheads, C), with some clustering of cells into organized lymphoid structures (arrows, D). Some large CD45⁺ cells, likely macrophages, were observed in alveolar spaces (arrowheads, D) proximal to the tumor. Little to no leukocyte tumor infiltration was observed (E). Scale bars = 1 mm (C) and 100 μm (D, E). (F) The ratio of CD3⁺ T cells to sum myeloid population in SCLC mice was significantly increased compared to three ADCA models, as assessed by one-way ANOVA with Tukey's post-test. (G) The percentage of CD45⁺ live cells present in two resected specimens of human SCLC was greatly reduced compared to five human ADCA specimens. (H) Leukocyte population summary of the flow cytometric analyses, shown as percent of live cells, for a representative normal mouse lung, 10-week Egfr, Kras, and Kp53 ADCA and mouse SCLC are displayed. Abbreviations: NL, non-adjacent normal lung; Tu, tumor. Asterisks indicate P < 0.05.