Integrin-αV-mediated activation of TGF-β regulates anti-tumour CD8 T cell immunity and response to PD-1 blockade

TGF-β is secreted in the tumour microenvironment in a latent, inactive form bound to latency associated protein and activated by the integrin αV subunit. The activation of latent TGF-β by cancer-cell-expressed αV re-shapes the tumour microenvironment, and this could affect patient responses to PD-1-targeting therapy. Here we show, using multiplex immunofluorescence staining in cohorts of anti-PD-1 and anti-PD-L1-treated lung cancer patients, that decreased expression of cancer cell αV is associated with improved immunotherapy-related, progression-free survival, as well as with an increased density of CD8+CD103+ tumour-infiltrating lymphocytes. Mechanistically, tumour αV regulates CD8 T cell recruitment, induces CD103 expression on activated CD8+ T cells and promotes their differentiation to granzyme B-producing CD103+CD69+ resident memory T cells via autocrine TGF-β signalling. Thus, our work provides the underlying principle of targeting cancer cell αV for more efficient PD-1 checkpoint blockade therapy.

For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly The statistical test(s) used AND whether they are one-or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section.
A description of all covariates tested A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals) For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated Our web collection on statistics for biologists contains articles on many of the points above.

Software and code
Policy information about availability of computer code Data collection Multispectral fluorescent images were captured using the Vectra 3 microscope (PerkinElmer). Flow cytometry data were collected on BD Fortessa instruments with FACSDiva software V10 (BD). Relative luciferase activity data were collected on In Vivo Imaging System IVIS (PerkinElmer). qRT-PCR data were collected with rtqPCR step-one Plus (Applied). Proximity ligation assay (PLA) data were collected with confocal microscope SP8 (Leica).

Data analysis
Image analysis of multispectral fluorescent images was performed using InForm software (PerkinElmer). Using http://kmplot.com, analysis of gene-expression TCGA datasets derived from NSCLC patients was performed. Gene-expression data were automatically computed to generate Kaplan-Meier from TCGA datasets. Flow cytometry data were analyzed on FlowJo v10 (TreeStar) and statistical analysis performed on Prism v8 (GraphPad). Relative luciferase activity data were calculted using ROI measurements by IVIS (PerkinElmer). PLA experiments were analyzed with Icy software (Gustave Roussy). The proliferation index was calculated every other day by flow cytometric analysis using ModFit LT™ program v5.0. Regarding patients cohorts, the best cut-point for total CD8+, CD8+CD103+ and CD8+CD103neg TIL was assessed using the logrank maximization method. Survival analyses were performed using the Kaplan-Meier method and the log-rank test. All p-values inferior to 0.05 were considered statistically significant. A Cox proportional hazards regression model was used to evaluate independent prognostic factors for OS and PFS. Variables included in the final multivariate model were selected according to their clinical relevance and statistical significance in univariate analysis (p-value cut-off = 0.10). The proportional hazard hypothesis was verified with the Schoenfeld residual method. Predictive factors of disease control were tested with logistic regression in univariate and multivariate analyses. The alpha level was 5%. Statistical analyses were performed with RStudio v1.1.463 (free software environment for statistical computing and graphics).
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

April 2020
Data Policy information about availability of data All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: -Accession codes, unique identifiers, or web links for publicly available datasets -A list of figures that have associated raw data -A description of any restrictions on data availability The authors state that all data generated during this study are included in the article and its Supplementary Information file and are available from the corresponding author upon request. Source data are provided with this paper. The use of publicly available data from lung cancer cohorts were consulted on the website https://kmplot.com/analysis/index.php?p=service&cancer=lung , under the specific data product name: KM Plotter -Lung Cancer.

Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative.

Sample size
No sample size calculation was performed for in vitro studies. Every in vitro experiment was performed independently at least 3 times for subsequent statistical analysis. Animal sample size were determined using previous studies and guided by the 3R principle. Human samples sizes were driven by the availability of biopsies.
Data exclusions No data were excluded from the analysis.

Replication
We established two cohorts of NSCLC patients treated with anti-PD(L)1 immunotherapy. Regarding the number of patients, the first cohort included 106 patients, whereas the second cohort included 51 patients. The attempt to replicate the same number of patients in both cohorts was not achieved due to the long inclusion period of patients. Moreover, the results were reliably reproduced for in vitro and in vivo experiments. This included independant replication: experiments presented in this study were performed using at least 2 biological replicates, except the experiment with Nude mice or WT C57Bl/6 mice engrafted with B16F10E or B16F10E-KO and treated in parallel with anti-PD-1, a-TGFβ or a-CD103, which were performed once.
Randomization Human material was obtained from random healthy donors independently of sex, age and any other characteristics. NSCLC patients were selected based on the eligibility for the operation after informed written consent and based on the disponibility of the biological material after pathology examinations. Mice were randomized after tumour inoculation to obtain homogeneous groups.

Blinding
Blinding was used during measurement of tumours in different treatment groups, and also during the analysis where each mouse was sorted by their numbers and not by the experimental group. The assessment of αV staining was performed by an expert thoracic pathologist blinded from clinical data. For assessment of αV staining experiments, the investigators were blinded to group allocation during data collection. For in vitro experiments, the blinding was not relevant because the analysis of flow cytometry data was performed with fixed gates for each group.
Reporting for specific materials, systems and methods We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material, system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.

Validation
All commercial antibodies were used (assay and species) according to the manufacturer's instructions. Validation of flow cytometry antibodies was performed by the titration to determine the optimal concentration through the series of the dilutions: 1/400, 1/200, 1/100, 1/50, 1/25, 1/11. Optimal concentration was defined by calculating the stain index.

Authentication
We regularly authenticate our NSCLC cell lines by testing recognition by autologous CTL clones and HLA-A2 expression when applicable. Purchased B16F10 cell line was visualy authenticated. The reporter mink lung epithelial cell line Mu.1LV was not authenticated.

Mycoplasma contamination
All the cell lines are mycoplasma-free and were regularly tested for mycoplasma contamination.
Commonly misidentified lines (See ICLAC register) No commonly misidentified cell lines were used in the study.

nature research | reporting summary
April 2020 Animals and other organisms Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research

Laboratory animals
Female C57BL/6J mice (7-9 weeks old) were purchased from Envigo. The female athymic nude Crl:NU(NCr)-Foxn1nu mice (6 weeks) were inbred in-house. Mice were housed in Super Mouse 750TM ventilated cages. Cages contained 1/8" corn cob bedding. Lights were on a 12h ON/OFF cycle, room temperature was set to 22°C with the variance of +/-2°C and ambient humidity conditions. A standard food diet (Envigo) and water (purified by reverse osmosis) were provided ad libitum.

Wild animals
Study did not involve wild animals.
Field-collected samples Study did not involve samples collected in the field.

Ethics oversight
All animals were housed at Gustave Roussy's animal facility and treated in accordance with guidelines established by the institutional animal committee (CEEA n°026: 2018-056-16280), after receiving the legal approval from French Minister of Higher Education, Research and Innovation under the procedure number APAFIS#16281-2018072515064652v2.
Note that full information on the approval of the study protocol must also be provided in the manuscript.

Human research participants
Policy information about studies involving human research participants

Population characteristics
Healthy donor blood samples were collected from the French blood bank (Etablissement français du Sang (EFS); agreement number N°12/EFS/079). Fresh NSCLC tumours were obtained from the Centre chirurgical Marie Lannelongue. FFPE tumour samples from NSCLC patients were obtained from patients treated with anti-PD(L)1 immunotherapy in a variety of settings in Gustave Roussy between 2012 and 2020. In the first imunotherapy-treated cohort, 55% of patients were over 65y old, 73% were male, 32% had KRAS mutated NSCLC, 46% received platinum-based chemotherapy, 15% chemoradiation, 30% other therapies and 8% had no prior therapy before received anti-PD-(L)1 immunotherapy, and 59% had adenocarcinoma. In the second imunotherapy-treated cohort, 47% of patients were over 65y old, 55% were male, 38% had KRAS mutated NSCLC, 57% received platinum-based chemotherapy, 6% chemoradiation, 22% other therapies and 8% had no prior therapy before received anti-PD-(L)1 immunotherapy, and 76% had adenocarcinoma. The treatment-naïve cohort of NSCLC patients has been previously described ( Note that full information on the approval of the study protocol must also be provided in the manuscript.

Flow Cytometry
Plots Confirm that: The axis labels state the marker and fluorochrome used (e.g. CD4-FITC).
The axis scales are clearly visible. Include numbers along axes only for bottom left plot of group (a 'group' is an analysis of identical markers).
All plots are contour plots with outliers or pseudocolor plots.
A numerical value for number of cells or percentage (with statistics) is provided.

Methodology Sample preparation
For human experiments: For freshly isolated TIL and tumour cells, human lung tumours were dissociated mechanically and enzymatically using a Tumor Dissociation Kit (Miltenyi,. Mononuclear cells were then isolated by a Ficoll-Hypaque gradient. Tumour cells were isolated by magnetic separation using Tumor Cell Isolation Kit (Miltenyi,. For PBL, mononuclear cells were then isolated by a Ficoll-Hypaque gradient. Cell surface and intracellular staining of human cells was performed on single-cell suspensions. Dead cells were excluded using the Live/Dead Fixable Blue Dead Cell Stain Kit (Invitrogen 186684). For intracellular staining, cells were fixed, permeabilized with the FoxP3 staining buffer set according to the manufacturer's instructions (eBioscience 00-5523-00).