Antibody-mediated transient depletion of CD4+ cells enhances the expansion of tumor-reactive CD8+ T cells and exhibits robust antitumor effects in preclinical and clinical studies. To investigate the clonal T-cell responses following transient CD4+ cell depletion in patients with cancer, we conducted a temporal analysis of the T-cell receptor (TCR) repertoire in the first-in-human clinical trial of IT1208, a defucosylated humanized monoclonal anti-CD4. Transient depletion of CD4+ cells promoted replacement of T-cell clones among CD4+ and CD8+ T cells in the blood. This replacement of the TCR repertoire was associated with the extent of CD4+ T-cell depletion and an increase in CD8+ T-cell count in the blood. Next, we focused on T-cell clones overlapping between the blood and tumor in order to track tumor-associated T-cell clones in the blood. The total frequency of blood–tumor overlapping clones tended to increase in patients receiving a depleting dose of anti-CD4, which was accompanied by the replacement of overlapping clones. The greater expansion of CD8+ overlapping clones was commonly observed in the patients who achieved tumor shrinkage. These results suggested that the clonal replacement of the TCR repertoire induced by transient CD4+ cell depletion was accompanied by the expansion of tumor-reactive T-cell clones that mediated antitumor responses. Our findings propose beneficial remodeling of the TCR repertoire following transient CD4+ cell depletion and provide novel insight into the antitumor effects of monoclonal anti-CD4 treatment in patients with cancer.

See related Spotlight on p. 601

CD4+ T cells play a bidirectional role in controlling the antitumor responses of CD8+ T cells. Although conventional CD4+ T cells promote CD8+ T-cell responses by licensing dendritic cells (DC; ref. 1), Foxp3+CD4+ regulatory T cells (Treg) inhibit them by depriving B7 costimulatory molecules from DCs (2). Contrary to the dogma that CD4+ T cells are always required for optimal CD8+ T-cell responses, several studies have demonstrated that depletion of CD4+ cells using monoclonal anti-CD4 inhibits tumor growth in mice (3–6). We previously demonstrated that enhanced oligoclonal responses of CD8+ T cells underpinned the antitumor effects of anti-CD4 administration (6, 7). Therefore, it was reasoned that simultaneous removal of the immunostimulatory and immunosuppressive effects of CD4+ T cells in tumor-bearing mice is likely to promote antitumor immunity. Based on these preclinical findings, we performed a first-in-human clinical trial of IT1208, a humanized depleting CD4 monoclonal antibody, for advanced gastrointestinal cancer in Japan (8). We found that transient depletion of CD4+ T cells was followed by an increase in CD8+ T-cell count in the blood, with tumor shrinkage in 5 of 11 patients (8). However, the nature of T-cell clonal responses following transient depletion of CD4+ cells in patients has remained elusive.

Advances in immunomonitoring technologies have provided us with multiple types of immunologic data (9). However, it remains difficult to discriminate between immune responses against the tumor and those against non-tumor antigens. Analysis of the T-cell receptor (TCR) repertoire, which represents the collection of diverse TCR alpha or beta sequences responsible for antigen recognition by T cells, has been used as a next-generation approach for monitoring T-cell responses based on antigen specificity (10). Several clinical studies have demonstrated that the enhanced clonal expansion of tumor-infiltrating lymphocytes (TIL) correlates with the clinical responses following treatment with anti–PD-1 (11–13). Unexpectedly, single-cell analysis of TILs revealed that PD-1 blockade induced an expansion of CD8+ TILs, mainly consisting of novel clones that were not observed in the tumor before treatment (14). This study suggested that, in addition to the reinvigoration of exhausted T cells (15–17), successful replacement of the tumor-reactive T-cell repertoire and expansion of novel tumor-associated clones were important for the antitumor effect of PD-1 blockade. We previously demonstrated that depletion of CD4+ cells in tumor-bearing mice increased the number of CD8+ T-cell clones that overlapped between the draining lymph node (dLN), blood, and tumor, implying that treatment with anti-CD4 could mobilize a wide variety of tumor-reactive clones into the Cancer–Immunity Cycle (7). However, whether the depletion of CD4+ cells in patients with cancer could replace the repertoire of CD8+ cells in addition to that of CD4+ cells has remained unclear.

To analyze the replacement of T-cell clones following CD4+ T-cell depletion, we performed TCR repertoire analysis in patients with gastrointestinal cancer undergoing treatment with IT1208, the monoclonal anti-CD4. By analyzing CD4+ and CD8+ T cells separately, we demonstrated that transient CD4+ cell depletion induced replacement of the CD8+ T-cell repertoire in blood, which was associated with an increase in blood CD8+ T cells. We also focused on T-cell clones overlapping between the blood and tumor in order to track tumor-associated clones in the blood. Through this blood–tumor overlapping repertoire analysis, we propose a possible association between replacement of the T-cell repertoire and antitumor T-cell responses following anti-CD4 treatment.

Study design

The blood T cells and tumor biopsy samples were collected from patients enrolled in a first-in-human phase I clinical trial of IT1208 (humanized anti-CD4; IDAC Theranostics; UMIN000026564) as described previously (8). Patients with advanced or metastatic solid tumors were administrated IT1208 at planned doses of 0.1 mg/kg (n = 4) or 1.0 mg/kg (n = 7). The first patient in each cohort was treated with one dose of IT1208 on day 1, and the other patients received two doses of IT1208 on days 1 and 8. Computed tomography was performed at baseline and every 4 weeks thereafter until disease progression or the beginning of subsequent treatment. Tumor response was evaluated as the maximum change of sum of diameter of measurable lesions per the Response Evaluation Criteria in Solid Tumors (RECIST, v1.1). The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines, following approval by the ethics board in each institution. All patients provided written informed consent for participation in the study. The patient characteristics and schedule of IT1208 administration and sample collection were summarized in Table 1.

Monitoring of immunophenotypes of peripheral blood mononuclear cells

Whole blood was collected before and after treatment with IT1208 at various time points (Pretreatment: Pre, Post1: days 22–23, and Post2: days 36–60). Peripheral blood mononuclear cells (PBMC) were separated by density-gradient centrifugation using Ficoll-Paque PLUS (GE Healthcare). In brief, the blood diluted 2-fold with RPMI-1640 (Nacalai Tesque, Inc.) were layered on Ficoll-Paque and centrifuged at 1,100 × g for 30 minutes at room temperature. PBMCs were collected, washed twice with PBS, and then stored in liquid nitrogen in CELLBANKER II (Nippon Zenyaku Kogyo Co., Ltd.).

Flow cytometry analysis of patient PBMCs

Cryopreserved PBMCs were thawed and washed twice with PBS containing 10% FBS (HyClone Laboratories Inc.) and PBS, respectively. Single-cell suspensions were then processed for surface staining with an antibody cocktail (panel described in Supplementary Table S1) for 20 minutes at 4°C. Fcγ receptors were not blocked for staining. Dead cells were labeled by Fixable Viability Dye FVD V500 (eBioscience). Cells were then washed with PBS containing with 2% FBS, and fixed and permeabilized using the Foxp3 Staining Buffer Set (BD Biosciences) according to the manufacturer's protocol. Cells were subsequently stained with an intracellular stain antibody cocktail (panel described in Supplementary Table S1) for 30 minutes at 4°C. Cells were then washed with Foxp3 permeabilization buffer and resuspended in CellFix (BD Biosciences). The stained cells were detected using an LSR II Fortessa with FACSDiva software (BD Biosciences). All analyses were carried out using FlowJo software (BD Biosciences). The cell counts of each subset (/μL) were estimated the following formula: [proportion of each subset (%)] × [absolute count of lymphocytes (/μL)]. The populations assessed and the cell-surface markers used to define them are listed in Supplementary Table S2.

T-cell isolation from PBMCs

For TCR repertoire analysis, blood was taken pretreatment (Pre: day 1), posttreatment 1 (Post1: day 22), and posttreatment 2 (Post2: day 36–60), except patients who were treated with IT1208 one time, for which, blood was collected only Pre and Post1. T cells were enriched from PBMCs using the Pan T-cell Isolation Kit, human (Milteny Biotec Inc.). Cells were stained with CD8-FITC (clone BW135/80, Miltenyi Biotec), CD3-PE (clone UCHT1, TONBO Biosciences), CD4-PerCP-Cy5.5 (clone SK3, TONBO Biosciences), and Ghost Dye Red780 (TONBO Biosciences), and then CD3+ CD4+ and CD3+ CD8+ T cells were purified using a FACSAria II with Diva software (BD Biosciences). The purity of sorted cells was routinely more than 99%. After washing, the sorted cells were lysed in 1 mL of cell lysis buffer (100 mmol/L Tris-HCl pH 7.5, 1% LiDS, 500 mmol/L LiCl, 10 mmol/L EDTA, 5 mmol/L DTT) and used for TCR sequencing.

Extraction of RNA from biopsy samples

Tumor biopsies were conducted before (day −6 through day −1) and after treatment (day 29 through day 37). Samples were homogenized in TRIzol (Ambion). RNA was extracted from each sample using the RNeasy Mini Kit (QIAGEN), and amounts and purity were measured with the Agilent 2200 TapeStation (Agilent Technologies). Five micrograms of total RNA was diluted with 1 mL of cell lysis buffer and used for TCR sequencing.

TCR library construction and sequencing

TCR sequencing libraries for next-generation sequencing were prepared according to a previous report (ref. 7; GSE115425) with some modifications. Primer sequences used for library preparation are listed in Supplementary Table S3. PolyA RNAs were isolated and amplified from sorted T cells according to a previous report (GSE110711; ref. 18). To amplify the TCR cDNA containing complementarity determining region 3 (CDR3), nested PCR of the TCR locus was performed as follows. The first PCR mixture comprised 0.4 μL of 10 μmol/L primer mix (5′ WTA and Trbc_ex), 4.6 μL of template, and 5 μL of KAPA Hifi Hotstart ReadyMix (KAPA Biosystems; #KK2602). The thermal cycling conditions were programmed using SimpliAmp Thermal cycler (Applied Biosystems) as follows: denaturation at 95°C for 3 minutes, 10 cycles of denaturation for 20 seconds at 98°C, annealing for 15 seconds at 58°C, and extension for 30 seconds at 72°C, followed by a final extension at 72°C for 2 minutes. Next, 10 μL of the first PCR products were used for purification with an Agencourt AM Pure XP kit (Beckman-Coulter) at a 0.7:1 ratio (beads to sample) and eluted with 12 μL of water (Nacalai). The second PCR mixture consisted of 1.25 μL of 10 μmol/L primer mix (5′ WTA and Trbc_in-Bio), 11.25 μL of the template and 12.5 μL of KAPA Hifi Hotstart ReadyMix. The thermal cycling conditions were the same as the first PCR except for the cycle number, which was 13 cycles. Next, 25 μL of the second PCR products were purified using Agencourt AM Pure XP kit at a 0.8:1 ratio (beads to sample) and eluted in 15 μL of water. The purified PCR products were sheared randomly using NEBNext dsDNA fragmentase (New England Biolabs). The fragmentation reaction mix consisted of 6 μL of water, 2 μL of 10× Fragmentase Reaction Buffer v2, 10 μL of PCR product, and 2 μL of fragmentase. The fragmentation reaction was incubated at 37°C for 30 minutes, and then 5 μL of 0.5 M EDTA was added to stop the reaction on ice. The sheared PCR product was then purified and subjected to size selection using the Agencourt AM Pure XP kit at a 0.8:1 ratio (beads to sample) to remove large fragments, a 1.6:1 ratio (beads to sample) to remove the smaller fragments, and eluted with 20 μL of Tris-HCl (pH 8.0). To capture the TCR cDNA containing the end of constant region, the purified PCR products were incubated with 10 μL of Dynabeads M-270 Streptavidin (Thermo Fisher Scientific) for 30 minutes at room temperature, washed 3 times with B&W-T buffer [5 mmol/L Tris-HCl (pH 7.5), 1 M NaCl, 0.5 mmol/L EDTA, and 0.1% Tween-20], and once with Tris-HCl (pH 8.0), and water. The captured TCR cDNA was repaired using NEBNext Ultra II End Repair/dA-Tailing Module (New England Biolabs). The repair reaction mix contained 1.2 μL of reaction buffer, 8.3 μL of beads resuspended with water, and 0.5 μL of enzyme mix. The repair reaction was incubated at 20°C for 30 minutes, and then washed once with lysis buffer [1% LiDS, 100 mmol/L Tris-HCl (pH 7.5), 500 mmol/L LiCl, 10 mmol/L EDTA, and 5 mmol/L DTT], three times with B&W-T buffer, and once with Tris-HCl (pH 8.0). The repaired TCR cDNA was attached to the sequencing adaptor using the DNA Ligation Kit <Mighty Mix> (TaKaRa). The ligation reaction mix consisted of 1 μL of 10 μmol/L P1 Adaptor, 6.5 μL of beads resuspended in Tris-HCl (pH 8.0), and 15 μL of enzyme mix. The ligation reaction was incubated at 16°C for 60 minutes using a thermal cycler with the cover open, washed once with lysis buffer, three times with B&W-T buffer, and once with Tris-HCl (pH 8.0). The third PCR was carried out using barcoded primers to enrich the TCR cDNA flanked with sequencing adapters. The third PCR mixture consisted of 0.35 μL of 10 μmol/L trP1 primer, 1 μL of 3.5 μmol/L IonA-BC-Trbc primer, 3.65 μL of beads resuspended in water and 5 μL of KAPA Hifi Hotstart ReadyMix. The thermal cycling conditions were the same as the first PCR except for the cycle number, which was 12 cycles. The PCR product was purified and subjected to size selection using Agencourt AM Pure XP kit at a 0.75:1 ratio (beads to sample) to remove large fragments, a 0.65:1 ratio (beads to sample) to remove smaller fragments, and eluted with 20 μL of Tris-HCl (pH 8.0). Amplified TCR libraries were quantified using a KAPA Library Quantification Kit (KAPA Biosystems), and size distribution was analyzed by agarose electrophoresis and SYBR Gold staining (Thermo Fisher Scientific).

Final TCR libraries, whose lengths were 200 to 300 base pairs, were pooled and sequenced using an Ion Hi-Q Chef kit, an Ion PI Chip kit v3, and an Ion Proton Sequencer (Thermo Fisher Scientific) and an Ion Proton Sequencer or an Ion 540 Kit Chef, an Ion 540 Chip kit, and an Ion Genestudio S5 Sequencer (Thermo Fisher Scientific) according to the manufacturer's instructions, except the input library concentration (100 pmol/L) and flow number (500). The raw data have been deposited in the NCBI Gene Expression Omnibus (GEO) under the accession GSE120101.

Data processing of TCR sequencing

Adapter trimming and quality filtering of sequencing data were performed by using Cutadapt (19) and PRINSEQ-0.20.4 (20). Sequencing data were processed by MiXCR-3.0.2 (21). In MiXCR, filtered reads were aligned to reference human TCR V/D/J sequences registered in the international ImMunoGeneTics information system with the following parameters: -vParameters.geneFeatureToAlign = VTranscript -vjAlignmentOrder = JThenV, then identical sequences were assembled and grouped in clones with PCR and sequencing error correlation with the following parameters: -badQualityThreshold = 10, –separateByV = true, –only-productive = true, –region-of-interest = CDR3. The variable (V), diversity (D), and joining (J) segment of the TCRs were represented in IMGT gene nomenclature.

The list of final clones was analyzed by VDJtools-1.2.1 (22). The sequencing coverage of sample, which was defined as the ratio of total reads to the starting number of T cells, was normalized to ×5 in PBMCs and ×10 in bulk biopsy samples by “DownSample” command of VDJtools. T-cell clones were determined as TCR reads with the same TCR VJ segments and CDR3 nucleotide sequence, and the clones whose read count were less than the sequencing coverage (i.e., less than 1 read/cell) were considered as sequencing noise and discarded from the table of clones. The average number of T-cell clones in blood CD4+ and CD8+ T cells and tumor biopsies were 27,060, 10,775, and 1,704, respectively (8). The repertoires of CD4+ and CD8+ T cells shared very few clones, and the frequency of these shared clones was below 0.01%, or their frequency in CD4+ was lower than that in CD8+, and vice versa. Based on this mutually exclusive relationship between the CD4+ and CD8+ T-cell repertoires, we regarded the CD4+–CD8+ overlapping clones as contamination and categorized them into either CD4+ or CD8+ clones according to their dominance. The processed data have been deposited in the NCBI GEO under the accession GSE120101.

Calculation of clonality index of TCR repertoire

The 1 - Pielou index was used to evaluate the clonality of TCR repertoire, which was calculated using the formula |$1-\sum\limit^n_{i=1}p_i\ {\curr log}_e(p_i)/{\curr log}_e(n)$|⁠, where |$p_i$| is the frequency of clone i for a sample with n unique clones.

Identification of expanded and contracted clones

Significantly expanded/contracted clones were determined, as described in Dewitt and colleagues (23), using a Fisher exact test on an estimated count of T-cell clones, including clones detected only at one time point. The estimated count was obtained by rounding off 100,000 times the frequency of each clone, and q values corresponding to the P values of Fisher exact test was calculated using q value package in Microsoft R open 3.5.1. We adopted q < 0.0001 for threshold of expanded/contracted clones.

Statistical analysis

Statistical analyses were performed using GraphPad Prism (ver7) software (GraphPad Software). Two-tailed Wilcoxon matched-pairs signed-rank test was run on the comparison of clonality between CD4+ and CD8+ T cells in the blood within patients before IT1208 administration. Dunn multiple comparisons test was run on the comparison regarding to past anti–PD-1/PD-L1 treatment history in our cohort. All other experimental data were analyzed using the Mann–Whitney test. Asterisks to indicate significance corresponding to the following were used: n.s., not significant (P > 0.05); *, P ≤ 0.05; **, P ≤ 0.01; ***P ≤ 0.001.

Blood CD8+ T cells exhibit higher clonality at baseline

To elucidate the impact of transient depletion of CD4+ cells on the TCR repertoire in humans, we collected samples from 11 patients enrolled in the clinical trial of IT1208, including four patients with a treatment history of PD-1 blockade (Table 1; ref. 8). Blood CD4+ and CD8+ T cells were collected separately at three different time points (Pretreatment: Pre, Post1: days 22–23, and Post2: days 36–60), and tumor biopsies were collected at two different time points (Pre, day −6 to −1; and Post1, day 29–37). The CD4+ T-cell count, proportion of CD4+ T cells of blood lymphocytes, and CD4/CD8 ratio of blood T cells transiently decreased after administration of anti-CD4 (1.0 mg/kg, on days 2 and 9) and recovered to 30% to 50% of the baseline level by Post2. A similar transient decrease was observed in CD8+ T cells in the 1.0 mg/kg group. However, the count gradually increased to about 200% of the baseline level by day 29 (Supplementary Tables S4–S6; ref. 8).

First, we analyzed the baseline repertoire of CD4+ and CD8+ T cells in patients in this cohort. The clonality of the TCR repertoire, which reflects the extent of T-cell clonal expansion, is the most common index used in TCR repertoire analysis (24). Visualization of the proportion of the most abundant clones in each repertoire revealed that CD8+ T cells contained more expanded clones and exhibited higher clonality than CD4+ T cells, although clonality varied among individuals (Fig. 1A and B). For the CD8+ repertoire, the total frequency of top10 clones reached 16.7% to 68.9%, except for patient ID0006 (4.5%; Fig. 1A). This result suggested that patients in our cohort had an oligoclonal, less-diverse repertoire of blood T cells at baseline, particularly for CD8+ T cells.

We then analyzed the association between baseline characteristics and clonality. Patient age weakly correlated with the clonality of the CD8+ T-cell repertoire, whereas the number of previous treatments and history of PD-1 blockade showed no correlation (Supplementary Fig. S1A and S1B). Comparison of the baseline clonality between the 0.1 and 1.0 mg/kg groups revealed that the 0.1 mg/kg group had a higher clonality in both the CD4+ and CD8+ repertoire, although there was no statistically significant difference in the former (Supplementary Fig. S1C).

Changes in blood T-cell repertoire correlate with the extent of blood CD4+ T-cell depletion

In our previous study, we demonstrated that anti-CD4 treatment altered the blood T-cell repertoire of patients with cancer in a dose-dependent manner (8). To extend this finding, we assessed the correlation between the extent of repertoire alteration and changes in blood CD4+ T-cell count following anti-CD4 treatment. The decrease in CD4+ T-cell count on day 15, when the count was lowest, correlated with the alternation of both CD4+ and CD8+ T-cell repertoires from Pre to Post1, demonstrating that the change of TCR repertoire was associated with the extent of CD4+ cell depletion (Fig. 1C). Several studies have reported that an increase of clonality in the blood is associated with good prognosis following immunotherapy (11–13). Therefore, we calculated the time-dependent changes in blood T-cell repertoire clonality for each patient (Supplementary Fig. S1D and S1E). The increase in CD4+ and CD8+ T-cell clonality weakly correlated with a decrease rate of CD4+ T-cell count on day 15, although the change in CD4+ T-cell clonality was small (Fig. 1D and E). We did not observe clear tendencies in clonality changes among patients whose tumor diameter was decreased (ID0003, ID0005, ID0008, ID0009, and ID0010, hereafter “responders”) or increased (the rest of the patients; hereafter “nonresponders”) after treatment (Fig. 1F).

Anti-CD4 induces replacement of CD4+ and CD8+ T-cell clones in the blood

Despite the transient depletion and subsequent recovery of CD4+ T cells in the blood, their clonality was almost unchanged after treatment. However, it was unclear whether the repopulated repertoire retained the original clonal composition or predominant clones had changed after treatment. Moreover, it remained elusive whether the increase in CD8+ T-cell count was accounted for by the exclusive expansion of clones with higher frequency (major clones) or more diverse ones. To address these questions, we tracked the frequencies of individual clones in the blood from Pre to Post1. Significantly expanded and contracted clones were identified using Fisher exact test (Fig. 2A and B; Supplementary Fig. S2A–S2D; ref. 23). We then calculated the count and total changes in their frequency (Fig. 2C; Supplementary Fig. S2E). Expanded clones were observed not only among the major clones with a frequency over 0.1% at Pre, but also in minor clones with a frequency lower than 0.1% at Pre (Fig. 2A and B). The count of expanded clones tended to be higher in the 1.0 mg/kg group for both CD4+ and CD8+ T-cell repertoires (Fig. 2C). We also observed that contracted clones in CD8+ T cells also tended to be higher in the 1.0 mg/kg group (Fig. 2D). Total changes in the frequency of expanded and contracted clones showed a pattern similar to that observed for their count, except for total changes of CD4+ contracted clones, which seemed to reflect the higher baseline clonality of patients in the 0.1 mg/kg group (Supplementary Figs. S1E and S2F).

To extend similar analysis to the clones detected exclusively at Pre or Post1, we identified emerged or disappeared clones with significant changes in their abundance (Supplementary Fig. S2G and S2H). We found that the count and total changes of “emerged and expanded” or “disappeared and contracted” clones were higher in the 1.0 mg/kg group compared with the 0.1 mg/kg group, although the difference was not statistically significant (Supplementary Fig. S2I and S2J). Taken together, these results suggest that transient depletion of CD4+ cells induces the replacement of CD4+ and CD8+ T-cell clones and promotes the expansion of diverse clones among CD8+ T cells.

Anti-CD4–induced clonal replacement is accompanied by increased CD8+ T cells

Given the dose–response relationship between anti-CD4 treatment and clonal replacement in blood T cells, we examined whether changes in the blood T-cell count were associated with the replacement of blood T-cell clones. To test this, we calculated the Pearson correlation coefficient (R values) between TCR repertoire parameters and the CD4+ and CD8+ T-cell count relative to baseline (Fig. 3A). Most of the repertoire parameters were positively correlated with the increase of CD8+ T-cell count, except for the contraction in CD4+ T-cell clones. The count of contracted CD8+ T-cell clones (CD8.Con.count) and total changes in the frequency of expanded CD8+ T-cell clones (CD8.Exp.freq), which reflected CD8+ T-cell clone replacement, exhibited a stronger correlation than other repertoire parameters (Fig. 3B). The R values between TCR repertoire and CD8+ T-cell count gradually increased until day 29 (Fig. 3A; Supplementary Table S7). We observed similar tendencies in the correlation between the TCR repertoire and the proportion of the effector subpopulation among CD8+ T cells (Fig. 3C and D; Supplementary Fig. S3; Supplementary Table S8). These results suggested that the replacement of the blood T-cell repertoire was accompanied by the increase of total CD8+ T-cell count and the effector subpopulation among CD8+ T cells in the blood.

Based on these results, we hypothesized that the replacement of T-cell clones was caused by differences in the extent of clonal expansion between individual clones after transient CD4 depletion. Major clones before the treatment failed to repopulate, whereas some minor clones expanded to occupy the “space” emerging in the blood T-cell repertoire (Fig. 3E).

Anti-CD4 replaces the tumor-associated T-cell clones in the blood

Next, we examined whether the replacement of the TCR repertoire in the blood was associated with the antitumor effects of CD4+ cell depletion therapy. The TCR repertoire that overlapped between the blood and tumor (blood–tumor overlapping repertoire) is thought to include clones that (i) are activated in the dLN and mobilized into the tumor (25) or (ii) emigrated from the tumor and are recirculating through the blood (26). Thus, we considered that analysis of the blood–tumor overlapping T-cell repertoire could highlight tumor-reactive T-cell clones. Based on this concept, we analyzed the repertoire of tumor biopsies (Supplementary Table S9) and detected the blood–tumor overlapping clones from blood T-cell repertoire (Supplementary Tables S10 and S11). We excluded patient ID0005 from the blood–tumor overlap analysis because the quality of RNA from the tumor biopsy was insufficient to perform TCR repertoire analysis.

Indeed, the total frequency of CD4+ and CD8+ blood–tumor overlapping clones tended to increase in the blood in the 1.0 mg/kg group (8), and all of the responders showed an increase of CD8+ blood–tumor overlapping clones of over 15% (Fig. 4A). Therefore, we investigated whether this increase accompanied the replacement of overlapping clones, as seen for the whole T-cell repertoire in the blood. For this purpose, blood–tumor overlapping clones were extracted from the expanded/contracted clones in the blood (Fig. 4B and C; Supplementary Table S12). About 50% to 80% of expanded/contracted clones in the blood also increased/decreased in frequency in the tumor, although the differences between patients were large (Supplementary Fig. S4A–S4C). In the 1.0 mg/kg group, the count of expanded and contracted overlapping clones in the CD8+ T-cell repertoire tended to increase (Fig. 4D and E). The count of expanded overlapping clones in the repertoire of CD4+ T cells also increased significantly in the 1.0 mg/kg group (Fig. 4D and E). Total changes in the frequency of these expanded/contracted overlapping clones showed a similar pattern (Supplementary Fig. S5A and S5B). These results demonstrated that the increase in the blood–tumor overlap was accompanied by the replacement of overlapping clones in the blood.

Responders exhibit expansion of blood–tumor overlapping clones

To further characterize responders and nonresponders with regard to the pattern of blood–tumor overlapping T-cell repertoire replacement, we performed unsupervised hierarchical clustering using 10 parameters representing the changes in blood–tumor overlap (Fig. 5). Responders in the 1.0 mg/kg treatment group clustered together in the dendrogram. With regard to blood–tumor overlapping clones among blood CD8+ T cells, responders were characterized by an increase in total changes in the frequency of overlapping clones, count of expanded clones, and total changes in the frequency of expanded clones (Supplementary Fig. S6). There were some characteristics of nonresponders in the 1.0 mg/kg group (Fig. 5). In patient ID0006, the replacement of blood–tumor overlap was minor. In patient ID0007, the total frequency of blood–tumor overlap decreased in CD4+ and CD8+ T cells. In patient ID0011, the contraction of CD8+ blood–tumor overlapping clones was stronger than that observed in others. These results suggested that the clonal expansion of blood–tumor overlapping CD8+ T-cell clones was associated with the antitumor effect of anti-CD4 treatment.

Persistence of expanded CD8+ tumor-associated clones after anti-CD4

Finally, we investigated whether the expanded or contracted T-cell clones after anti-CD4 treatment would persist for a long period of time. To visualize the temporal changes in the frequency of clones in blood, we plotted the fold change in the frequency of each clone for Pre to Post1 (x-axis) and Post1 to Post2 (y-axis; Supplementary Fig. S7A and S7B). Both CD4+ and CD8+ T-cell clones were distributed from the top left to the bottom right, and the fold changes for Pre to Post1 and Post1 to Post2 were negatively correlated in all patients (CD4+: −0.345 to −0.588, CD8+: −0.316 to −0.530). These observations suggested that changes in the frequency of individual clones from Pre to Post1 were counteracted during Post1 to Post2.

We then focused on blood–tumor overlapping CD8+ expanded clones from Pre to Post1 (Pre–Post1 expanded/contracted clones). Most of these clones were distributed in the lower right area (Fig. 6A). However, especially within the 1.0 mg/kg group, expanded clones tended to distribute near the x-axis, suggesting that the Post1–Post2 contraction was weaker than the Pre–Post1 expansion. To confirm this, we performed temporal tracking of each expanded clone and observed that on average 57.6% of the Pre–Post1 expanded clones remained expanded at Post2 compared with Pre (Fig. 6B). These results indicated that the expansion of blood–tumor overlapping clones following anti-CD4 treatment persisted long after treatment.

To the best of our knowledge, this is the first study to characterize the clonal responses of CD4+ and CD8+ T cells after transient CD4+ cell depletion in patients with advanced cancer. We previously demonstrated that depletion of CD4+ cells increased the number and total frequency of tumor-reactive CD8+ T-cell clones in mice (7). However, the antitumor responses of T cells following transient CD4+ cell depletion in patients with cancer remained elusive. In particular, there are significant differences between the T-cell populations of mice and patients with cancer. Populations of memory or effector T cells are known to be dominant in elderly patients due to thymic involution and a long history of antigen exposure (27), whereas naïve T-cell populations have been observed as predominant in young, specific pathogen–free mice. The larger body size of patients compared with mice has enabled us to analyze temporal changes in the TCR repertoire following depletion of CD4+ cells. In this study, we demonstrated that transient depletion of CD4+ cells in patients with cancer induced replacement of CD4+ and CD8+ T-cell clones, as well as an expansion of blood–tumor overlapping CD8+ T-cell clones associated with antitumor responses.

Following anti-CD4 treatment, transient reduction and subsequent repopulation of circulating CD4+ and CD8+ T cells are observed (8). Temporal tracking of individual T-cell clones in the blood revealed that the transient depletion of CD4+ cells promoted the replacement of both CD4+ and CD8+ clones. Based on these observations, we speculated that clonal replacement was caused by differences in the extent of clonal expansion at the repopulation phase between individual clones; that is, some minor clones expanded beyond their original clone size, whereas some major clones did not recover their proportion after early reduction.

The T-cell repertoire in blood represents the circulating T-cell pool, in which T cells patrol secondary lymphoid tissues for their cognate antigens through the lymph–blood circulation. Therefore, the high clonality of the blood T-cell repertoire in our cohort at baseline implied the existence of already-expanded major clones in the circulating T-cell pool, including CD4+ effector/effector memory cells and Tregs, which suppress tumor antigen presentation by DCs through CTLA-4 (28). Thus, in the reduction phase soon after anti-CD4 administration, the transient depletion of CD4+ T cells made “space” for antigen presentation to novel tumor-reactive clones in the secondary lymphoid tissues. The transient increase of serum cytokines, such as IL6, IL8, and TNFα, immediately after administration of the antibody (8) might activate major CD8+ T-cell clones among the circulating T-cell pool and marginate them (29–31), thus also creating “space” in the repertoire of CD8+ T cells. Thereafter, during the repopulation phase, residual CD4+ and CD8+ T-cell clones might expand to occupy the “space” that emerged in the circulating T-cell pool. In this step, already-expanded major clones might exhibit compromised proliferation due to replicative senescence, in which T cells lose their replicative capacity after multiple rounds of replication (32). Therefore, these already-expanded clones might be replaced by newly expanded ones.

The important question raised in the above-described speculation was whether newly expanded clones occupying the “space” consisted of tumor-reactive T-cell clones or nontumor-specific ones. About half of the newly expanded clones in blood overlapped with those in the tumor, and some responders exhibited a greater expansion of overlapping clones among blood CD8+ T cells. Therefore, it is likely that part of the newly expanded clones were tumor-reactive and contributed to the antitumor effect. Several observations in our studies of mouse models support the notion that, following CD4+ cell depletion, the newly expanded clones are enriched for tumor-reactive T cells. First, when adoptively transferred into B16 tumor–bearing mice receiving anti-CD4 treatment, B16-reactive Pmel-1 T cells selectively proliferate, whereas polyclonal CD8+ T cells do not (6). Second, the overlapping T-cell repertoire between the dLN and tumor included T-cell clones whose reactivity against the B16 tumor was demonstrated in an independent study (7, 33). The increased expansion of blood–tumor overlapping clones in responders is also observed in patients with cancer after anti–PD-L1 treatment (34, 35). Nevertheless, we did not analyze the tumor reactivity of these overlapping clones. Thus, whether the overlapping clones were tumor-reactive or nontumor-reactive “bystanders” remained unknown (36). Cloning the TCR alpha- and beta-chain sequence pairs and validation of their reactivity through coculturing assays with tumor cells or antigen-pulsed antigen-presenting cells will provide answers to these questions.

There are several advantages to our analysis of the TCR repertoire. First, we performed TCR sequencing on purified blood CD4+ and CD8+ T cells, whereas many previous studies had not separated the two populations, which have distinct repertoire structure and immunologic roles. As CD8+ T cells are generally more oligoclonal than CD4+ T cells (37, 38), changes in the CD8:CD4 ratio can influence the results of TCR repertoire analysis of total T cells. CD8+ T cells play a central role in the antitumor immune response, whereas CD4+ T cells include immunosuppressive subpopulations such as Tregs and thus sometimes have tumor-promoting roles. Therefore, by analyzing the CD4+ and CD8+ T-cell repertoires separately, we were able to clarify TCR repertoire dynamics and provide clear immunologic insight into the antitumor responses of T cells following treatment with the anti-CD4. Another advantage of the current work was that we focused on the blood–tumor overlap in order to track clones associated with the antitumor response among blood T cells, as the TCR repertoire of tumor biopsy samples was highly variable depending on the amount of tumor tissue and intratumor heterogeneity of patients (39–42). We could characterize responders based on the changing patterns of the blood–tumor overlap, but could not do so by using the blood or TIL repertoire alone. Hence, the analysis of blood–tumor overlap has an advantage for the monitoring and evaluation of antitumor responses despite the invasiveness of tumor biopsy. A limitation of the current study is the lack of statistically significant differences in some TCR repertoire parameters between responders and nonresponders due to the small cohort sample size (n = 11). Patients in our cohort had varying cancer types, treatment history, age, and history of infections, including cytomegalovirus, all of which could influence the interpretation of changes observed in the TCR repertoire following anti-CD4 treatment (11).

In summary, our interorgan and time-course analysis of the TCR repertoire suggests that transient depletion of CD4+ cells induces the beneficial remodeling of the TCR repertoire and enhances the expansion of CD8+ tumor–reactive clones in patients with cancer. We expect that our TCR repertoire analysis of the blood–tumor overlap could be used as an early diagnostic biomarker for the evaluation of monoclonal anti-CD4 treatment efficacy.

H. Aoki reports grants from Japan Agency for Medical Research and Development and Japan Society for the Promotion of Science during the conduct of the study; other support from ImmunoGeneTeqs, Inc. outside the submitted work; and a patent for method to monitor immune response by TCR repertoire analysis pending. S. Ueha reports grants from Japan Agency for Medical Research and Development and Japan Society for Promotion of Science during the conduct of the study; other support from ImmunoGeneTeqs, Inc. and IDAC Theranostics, Inc. outside the submitted work; and a patent for method to monitor immune response by TCR repertoire analysis pending. H. Ogiwara reports a patent for method to monitor immune response by TCR repertoire analysis pending. K. Shitara reports grants and personal fees from Astellas Pharma, Eli Lilly and Company, Taiho Pharmaceutical, Merck Pharmaceutical, and Ono Pharmaceutical; personal fees from AbbVie, Yakult, GlaxoSmithKline, Bristol-Myers Squibb, Takeda Pharmaceuticals, Pfizer, and Novartis; and grants from Sumitomo Dainippon Pharma, Daiichi Sankyo, Chugai Pharma, and Medi Science outside the submitted work. T. Nakatsura reports grants from Japan Agency for Medical Research and Development and Japan Society for the Promotion of Science during the conduct of the study; grants from IDAC Theranostics, Inc. outside the submitted work; and a patent for method to monitor immune response by TCR repertoire analysis pending. S. Kitano reports grants and personal fees from Ono Pharmaceutical Co., Ltd., Boehringer Ingelheim, Daiichi Sankyo, Eisai, and Regeneron; personal fees from AstraZeneca, Chugai, Pfizer, Sanofi, Taiho, Novartis, MSD, Sumitomo Dainippon Pharma, Bristol-Myers Squibb, AYUMI Pharmaceutical Corporation, Rakuten Medical, and GlaxoSmithKline; and grants from Astellas, Gilead Sciences, PACT Pharma, and Takara Bio Inc. outside the submitted work. S. Kuroda reports personal fees from Celgene K.K. outside the submitted work. M. Wakabayashi reports personal fees from Chugai Pharmaceutical Co., Ltd. and Johnson & Johnson K.K. MEDICAL COMPANY outside the submitted work. S. Ito reports grants from IDAC Theranostics, Inc. during the conduct of the study; grants from AMED outside the submitted work; and a patent for method to monitor immune response by TCR repertoire analysis pending. T. Doi reports grants from Lilly, Merck Serono, Eisai, IQVIA, and Pfizer; grants and personal fees from Taiho, Novartis, MSD, Janssen Pharma, Boehringer Ingelheim, Sumitomo Dainippon Pharma, Daiichi Sankyo, Bristol-Myers Squibb, and AbbVie; and personal fees from Amgen, Rakuten Medical, Takeda, Otsuka, and Ono outside the submitted work. K. Matsushima reports grants from Japan Agency for Medical Research and Development and Japan Society for the Promotion of Science during the conduct of the study; other support from Kyowa-Hakko Kirin, ImmunoGeneTeqs, Inc., and IDAC Theranostics, Inc., and grants from Kyowa-Hakko Kirin and Ono outside the submitted work; and a patent for method to monitor immune response by TCR repertoire analysis pending. No disclosures were reported by the other authors.

H. Aoki: Conceptualization, software, formal analysis, methodology, writing–original draft, writing–review and editing. S. Ueha: Conceptualization, formal analysis, supervision, writing–original draft, project administration, writing–review and editing. S. Shichino: Formal analysis, methodology, writing–original draft, writing–review and editing. H. Ogiwara: Investigation, writing–review and editing. K. Shitara: Conceptualization, resources, writing–review and editing. M. Shimomura: Investigation, writing–review and editing. T. Suzuki: Formal analysis, writing–original draft, writing–review and editing. T. Nakatsura: Conceptualization, supervision, funding acquisition, project administration, writing–review and editing. M. Yamashita: Investigation, writing–review and editing. S. Kitano: Conceptualization, formal analysis, supervision, project administration, writing–review and editing. S. Kuroda: Conceptualization, data curation, writing–review and editing. M. Wakabayashi: Conceptualization, data curation, formal analysis, writing–review and editing. M. Kurachi: Writing–original draft, writing–review and editing. S. Ito: Conceptualization, funding acquisition, writing–original draft, writing–review and editing. T. Doi: Conceptualization, resources, supervision, project administration, writing–review and editing. K. Matsushima: Conceptualization, supervision, funding acquisition, writing–original draft, project administration, writing–review and editing.

This work was supported by the Japan Agency for Medical Research and Development (AMED) under grant numbers 16768526 and 19187773, and the Japan Society for the Promotion of Science under grant numbers 20281832 and 17929397. The authors thank D. Komura and S. Ishikawa for advice on data analysis, C.Y. Chen for useful comments for writing the manuscript, and A. Yamashita for expert technical assistance. They also thank Editage (www.editage.com) for English language editing.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Borst
J
,
Ahrends
T
,
Bąbała
N
,
Melief
CJM
,
Kastenmüller
W
. 
CD4+ T cell help in cancer immunology and immunotherapy
.
Nat Rev Immunol
2018
;
18
:
635
47
.
2.
Togashi
Y
,
Shitara
K
,
Nishikawa
H
. 
Regulatory T cells in cancer immunosuppression — implications for anticancer therapy
.
Nat Rev Clin Oncol
2019
;
16
:
356
71
.
3.
Yu
P
,
Lee
Y
,
Liu
W
,
Krausz
T
,
Chong
A
,
Schreiber
H
, et al
Intratumor depletion of CD4 + cells unmasks tumor immunogenicity leading to the rejection of late-stage tumors
.
J Exp Med
2005
;
201
:
779
91
.
4.
Den Boer
AT
,
Van Mierlo
GJD
,
Fransen
MF
,
Melief
CJM
,
Offringa
R
,
Toes
REM
. 
CD4+ T cells are able to promote tumor growth through inhibition of tumor-specific CD8+ T-cell responses in tumor-bearing hosts
.
Cancer Res
2005
;
65
:
6984
9
.
5.
Choi
BK
,
Kim
YH
,
Kang
WJ
,
Lee
SK
,
Kim
KH
,
Shin
SM
, et al
Mechanisms involved in synergistic anticancer immunity of anti-4-1BB and anti-CD4 therapy
.
Cancer Res
2007
;
67
:
8891
9
.
6.
Ueha
S
,
Yokochi
S
,
Ishiwata
Y
,
Ogiwara
H
,
Chand
K
,
Nakajima
T
, et al
Robust antitumor effects of combined anti-CD4-depleting antibody and anti-PD-1/PD-L1 immune checkpoint antibody treatment in mice
.
Cancer Immunol Res
2015
;
3
:
631
40
.
7.
Aoki
H
,
Ueha
S
,
Shichino
S
,
Ogiwara
H
,
Hashimoto
SI
,
Kakimi
K
, et al
TCR repertoire analysis reveals mobilization of novel CD8 + T cell clones into the cancer-immunity cycle following anti-CD4 antibody administration
.
Front Immunol
2019
;
10
:
1
13
.
8.
Shitara
K
,
Ueha
S
,
Shichino
S
,
Aoki
H
,
Ogiwara
H
,
Nakatsura
T
, et al
First-in-human phase 1 study of IT1208, a defucosylated humanized anti-CD4 depleting antibody, in patients with advanced solid tumors
.
J Immunother Cancer
2019
;
7
:
195
.
9.
Duffy
D
. 
Standardized Immunomonitoring: separating the signals from the noise
.
Trends Biotechnol
2018
;
36
:
1107
15
.
10.
Schrama
D
,
Ritter
C
,
Becker
JC
. 
T cell receptor repertoire usage in cancer as a surrogate marker for immune responses
.
Semin Immunopathol
2017
;
39
:
255
68
.
11.
Riaz
N
,
Havel
JJ
,
Makarov
V
,
Desrichard
A
,
Urba
WJ
,
Sims
JS
, et al
Tumor and microenvironment evolution during immunotherapy with nivolumab
.
Cell
2017
;
171
:
934
49
.
12.
Roh
W
,
Chen
P
,
Reuben
A
,
Spencer
CN
,
Peter
A
,
Miller
JP
, et al
Integrated molecular analysis of tumor biopsies on sequential CTLA-4 and PD-1 blockade reveals markers of response and resistance
.
Sci Transl Med
2017
;
9
:
1
11
.
13.
Yusko
E
,
Vignali
M
,
Wilson
RK
,
Mardis
ER
,
Hodi
FS
,
Horak
C
, et al
Association of tumor microenvironment T-cell repertoire and mutational load with clinical outcome after sequential checkpoint blockade in melanoma
.
Cancer Immunol Res
2019
;
7
:
458
65
.
14.
Yost
KE
,
Satpathy
AT
,
Wells
DK
,
Qi
Y
,
Wang
C
,
Kageyama
R
, et al
Clonal replacement of tumor-specific T cells following PD-1 blockade
.
Nat Med
2019
;
25
:
1251
9
.
15.
Huang
AC
,
Postow
MA
,
Orlowski
RJ
,
Mick
R
,
Bengsch
B
,
Manne
S
, et al
T-cell invigoration to tumour burden ratio associated with anti-PD-1 response
.
Nature
2017
;
545
:
60
5
.
16.
Kamphorst
AO
,
Pillai
RN
,
Yang
S
,
Nasti
TH
,
Akondy
RS
,
Wieland
A
, et al
Proliferation of PD-1+ CD8 T cells in peripheral blood after PD-1–targeted therapy in lung cancer patients
.
Proc Natl Acad Sci U S A
2017
;
114
:
4993
8
.
17.
Huang
AC
,
Orlowski
RJ
,
Xu
X
,
Mick
R
,
George
SM
,
Yan
PK
, et al
A single dose of neoadjuvant PD-1 blockade predicts clinical outcomes in resectable melanoma
.
Nat Med
2019
;
25
:
454
61
.
18.
Shichino
S
,
Ueha
S
,
Hashimoto
S
,
Otsuji
M
,
Abe
J
,
Tsukui
T
, et al
Transcriptome network analysis identifies protective role of the LXR/SREBP-1c axis in murine pulmonary fibrosis
.
JCI Insight
2019
;
4
:
e122163
.
19.
Martin
M
. 
Cutadapt removes adapter sequences from high-throughput sequencing reads
.
EMBnet.journal
2013
;
17
:
10
2
.
20.
Schmieder
R
,
Edwards
R
. 
Quality control and preprocessing of metagenomic datasets
.
Bioinformatics
2011
;
27
:
863
4
.
21.
Bolotin
DA
,
Poslavsky
S
,
Mitrophanov
I
,
Shugay
M
,
Mamedov
IZ
,
Putintseva
EV
, et al
MiXCR: software for comprehensive adaptive immunity profiling
.
Nat Methods
2015
;
12
:
380
1
.
22.
Shugay
M
,
Bagaev
DV
,
Turchaninova
MA
,
Bolotin
DA
,
Britanova
OV
,
Putintseva
EV
, et al
VDJtools: unifying post-analysis of T cell receptor repertoires
.
PLoS Comput Biol
2015
;
11
:
1
16
.
23.
DeWitt
WS
,
Emerson
RO
,
Lindau
P
,
Vignali
M
,
Snyder
TM
,
Desmarais
C
, et al
Dynamics of the cytotoxic T cell response to a model of acute viral infection
.
J Virol
2015
;
89
:
4517
26
.
24.
Ye
B
,
Smerin
D
,
Gao
Q
,
Kang
C
,
Xiong
X
. 
High-throughput sequencing of the immune repertoire in oncology: applications for clinical diagnosis, monitoring, and immunotherapies
.
Cancer Lett
2018
;
416
:
42
56
.
25.
Chen
DS
,
Mellman
I
. 
Oncology meets immunology: the cancer-immunity cycle
.
Immunity
2013
;
39
:
1
10
.
26.
Gros
A
,
Parkhurst
MR
,
Tran
E
,
Pasetto
A
,
Robbins
PF
,
Ilyas
S
, et al
Prospective identification of neoantigen-specific lymphocytes in the peripheral blood of melanoma patients
.
Nat Med
2016
;
22
:
433
8
.
27.
Thome
JJC
,
Grinshpun
B
,
Kumar
B V
,
Kubota
M
,
Lerner
H
,
Sempowski
GD
, et al
Long-term maintenance of human naïve T cells through in situ homeostasis in lymphoid tissue sites
.
Sci Immunol
2016
;
1
:
eaah6506
.
28.
Qureshi
OS
,
Zheng
Y
,
Nakamura
K
,
Attridge
K
,
Manzotti
C
,
Schmidt
EM
, et al
Trans-endocytosis of CD80 and CD86: a molecular basis for the cell-extrinsic function of CTLA-4
.
Science
2011
;
332
:
600
3
.
29.
Aulitzky
WE
,
Tilg
H
,
Vogel
W
,
Aulitzky
W
,
Berger
M
,
Gastl
G
, et al
Acute hematologic effects of interferon alpha, interferon gamma, tumor necrosis factor alpha and interleukin 2
.
Ann Hematol
1991
;
62
:
25
31
.
30.
Hendriks
HR
,
Korn
C
,
Mebius
RE
,
Kraal
G
. 
Interferon-gamma-increased adherence of lymphocytes to high endothelial venules
.
Immunology
1989
;
68
:
221
6
.
31.
Cavender
D
,
Saegusa
Y
,
Ziff
M
. 
Stimulation of endothelial cell binding of lymphocytes by tumor necrosis factor
.
J Immunol
1987
;
139
:
1855
60
.
32.
Kurachi
M
,
Kakimi
K
,
Ueha
S
,
Matsushima
K
. 
Maintenance of memory CD8+ T cell diversity and proliferative potential by a primary response upon re-challenge
.
Int Immunol
2007
;
19
:
105
15
.
33.
Shitaoka
K
,
Hamana
H
,
Kishi
H
,
Hayakawa
Y
,
Kobayashi
E
,
Sukegawa
K
, et al
Identification of tumoricidal TCRs from tumor-infiltrating lymphocytes by single-cell analysis
.
Cancer Immunol Res
2018
;
6
:
378
88
.
34.
Snyder
A
,
Nathanson
T
,
Funt
SA
,
Ahuja
A
,
Buros Novik
J
,
Hellmann
MD
, et al
Contribution of systemic and somatic factors to clinical response and resistance to PD-L1 blockade in urothelial cancer: an exploratory multi-omic analysis
.
PLoS Med
2017
;
14
:
1
24
.
35.
Rytlewski
J
,
Deng
S
,
Xie
T
,
Davis
C
,
Robins
H
,
Yusko
E
, et al
Model to improve specificity for identification of clinically-relevant expanded T cells in peripheral blood
.
PLoS One
2019
;
14
:
1
10
.
36.
Duhen
T
,
Duhen
R
,
Montler
R
,
Moses
J
,
Moudgil
T
,
De Miranda
NF
, et al
Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors
.
Nat Commun
2018
;
9
:
2724
.
37.
Li
HM
,
Hiroi
T
,
Zhang
Y
,
Shi
A
,
Chen
G
,
De
S
, et al
TCR repertoire of CD4+ and CD8+ T cells is distinct in richness, distribution, and CDR3 amino acid composition
.
J Leukoc Biol
2016
;
99
:
505
13
.
38.
Yoshida
K
,
Cologne
JB
,
Cordova
K
,
Misumi
M
,
Yamaoka
M
,
Kyoizumi
S
, et al
Aging-related changes in human T-cell repertoire over 20 years delineated by deep sequencing of peripheral T-cell receptors
.
Exp Gerontol
2017
;
96
:
29
37
.
39.
Reuben
A
,
Gittelman
R
,
Gao
J
,
Zhang
J
,
Yusko
EC
,
Wu
CJ
, et al
TCR repertoire intratumor heterogeneity in localized lung adenocarcinomas: an association with predicted neoantigen heterogeneity and postsurgical recurrence
.
Cancer Discov
2017
;
7
:
1088
97
.
40.
Chen
Z
,
Zhang
C
,
Pan
Y
,
Xu
R
,
Xu
C
,
Chen
Z
, et al
T cell receptor β-chain repertoire analysis reveals intratumour heterogeneity of tumour-infiltrating lymphocytes in oesophageal squamous cell carcinoma
.
J Pathol
2016
;
239
:
450
8
.
41.
Shi
L
,
Zhang
Y
,
Feng
L
,
Wang
L
,
Rong
W
,
Wu
F
, et al
Multi-omics study revealing the complexity and spatial heterogeneity of tumor-infiltrating lymphocytes in primary liver carcinoma
.
Oncotarget
2017
;
8
:
34844
57
.
42.
Joshi
K
,
Robert de Massy
M
,
Ismail
M
,
Reading
JL
,
Uddin
I
,
Woolston
A
, et al
Spatial heterogeneity of the T cell receptor repertoire reflects the mutational landscape in lung cancer
.
Nat Med
2019
;
25
:
1549
59
.