nt12dreuar3esd

278 | Nature | Vol 579 | 12 March 2020

Article

strength of immune response, or the ‘set point’^23 of the immune system
against tumour or foreign antigens. Our study presents an alternative
to the widely presumed mechanism that immune checkpoint blockade
acts on chronically stimulated T cells to reverse a terminally differenti-
ated or ‘exhausted’ state observed originally during chronic virus infec-
tion^24 ,^25. Recent observations have also questioned this model, showing
that the phenotype of terminal exhaustion is epigenetically locked
and difficult to alter^26 ,^27. In addition, recent studies have linked clini-
cal response to stem-like memory CD8+ T cells rather than exhausted
ones^28. Overall, our study suggests that non-exhausted T cells and T cell
clones supplied from the periphery may be key factors in explaining
patient variability and clinical benefit from cancer immunotherapy.
Clinical benefit could arise from a direct effect of PD1 or PDL1 block-
ade on effector T cells^29 or other non-exhausted T cells, or because
blockade can increase the production or effectiveness of commit-
ted anti-tumoural T cells only in those patients with an ongoing T cell
response and continued replenishment of tumour-infiltrating lympho-
cytes. As a practical consequence of this study, the observed correla-
tions between TCR repertoires of dual-expanded clones in tumours
and those of peripherally expanded clones suggest that sampling and
identifying expanded clones in blood may characterize the TCR com-
position of clinically relevant intratumoural T cells, expanding the
possibilities for ‘liquid biopsies’^30 ,^31.

Online content

Any methods, additional references, Nature Research reporting sum-
maries, source data, extended data, supplementary information,
acknowledgements, peer review information; details of author con-
tributions and competing interests; and statements of data and code
availability are available at https://doi.org/10.1038/s41586-020-2056-8.

1. Mellman, I., Coukos, G. & Dranoff, G. Cancer immunotherapy comes of age. Nature 480 ,
   480–489 (2011).


2. Shulman, Z. et al. Transendothelial migration of lymphocytes mediated by
   intraendothelial vesicle stores rather than by extracellular chemokine depots. Nat.
   Immunol. 13 , 67–76 (2012).


3. Guo, X. et al. Global characterization of T cells in non-small-cell lung cancer by single-
   cell sequencing. Nat. Med. 24 , 978–985 (2018).


4. Zhang, L. et al. Lineage tracking reveals dynamic relationships of T cells in colorectal
   cancer. Nature 564 , 268–272 (2018).


5. Yost, K. E. et al. Clonal replacement of tumor-specific T cells following PD-1 blockade.
   Nat. Med. 25 , 1251–1259 (2019).


6. Schenkel, J. M. & Masopust, D. Tissue-resident memory T cells. Immunity 41 , 886–897
   (2014).


7. Mackay, L. K. et al. Hobit and Blimp1 instruct a universal transcriptional program of tissue
   residency in lymphocytes. Science 352 , 459–463 (2016).
   8. Kumar, B. V. et al. Human tissue-resident memory T cells are defined by core
   transcriptional and functional signatures in lymphoid and mucosal sites. Cell Rep. 20 ,
   2921–2934 (2017).
   9. Im, S. J. et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy.
   Nature 537 , 417–421 (2016).
   10. Miller, B. C. et al. Subsets of exhausted CD8+ T cells differentially mediate tumor control
   and respond to checkpoint blockade. Nat. Immunol. 20 , 326–336 (2019).
   11. Balin, S. J. et al. Human antimicrobial cytotoxic T lymphocytes, defined by NK receptors
   and antimicrobial proteins, kill intracellular bacteria. Sci. Immunol. 3 , eaat7668 (2018).
   12. Thommen, D. S. et al. A transcriptionally and functionally distinct PD- 1 + CD8+ T cell pool
   with predictive potential in non-small-cell lung cancer treated with PD-1 blockade. Nat.
   Med. 24 , 994–1004 (2018).
   13. Sun, Q., Hao, Q. & Prasanth, K. V. Nuclear long noncoding RNAs: key regulators of gene
   expression. Trends Genet. 34 , 142–157 (2018).
   14. Delmas, V., Stokes, D. G. & Perry, R. P. A mammalian DNA-binding protein that contains a
   chromodomain and an SNF2/SWI2-like helicase domain. Proc. Natl Acad. Sci. USA 90 ,
   2414–2418 (1993).
   15. Gaide, O. et al. Common clonal origin of central and resident memory T cells following
   skin immunization. Nat. Med. 21 , 647–653 (2015).
   16. Simoni, Y. et al. Bystander CD8+ T cells are abundant and phenotypically distinct in
   human tumour infiltrates. Nature 557 , 575–579 (2018).
   17. Scheper, W. et al. Low and variable tumor reactivity of the intratumoral TCR repertoire in
   human cancers. Nat. Med. 25 , 89–94 (2019).
   18. Mariathasan, S. et al. TGFβ attenuates tumour response to PD-L1 blockade by contributing
   to exclusion of T cells. Nature 554 , 544–548 (2018).
   19. Fehrenbacher, L. et al. Atezolizumab versus docetaxel for patients with previously treated
   non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised
   controlled trial. Lancet 387 , 1837–1846 (2016).
   20. McDermott, D. F. et al. Clinical activity and molecular correlates of response to
   atezolizumab alone or in combination with bevacizumab versus sunitinib in renal cell
   carcinoma. Nat. Med. 24 , 749–757 (2018).
   21. Tumeh, P. C. et al. PD-1 blockade induces responses by inhibiting adaptive immune
   resistance. Nature 515 , 568–571 (2014).
   22. Araujo, J. M. et al. Effect of CCL5 expression in the recruitment of immune cells in triple
   negative breast cancer. Sci. Rep. 8 , 4899 (2018).
   23. Chen, D. S. & Mellman, I. Elements of cancer immunity and the cancer-immune set point.
   Nature 541 , 321–330 (2017).
   24. Wherry, E. J. & Kurachi, M. Molecular and cellular insights into T cell exhaustion. Nat. Rev.
   Immunol. 15 , 486–499 (2015).
   25. Topalian, S. L., Drake, C. G. & Pardoll, D. M. Immune checkpoint blockade: a common
   denominator approach to cancer therapy. Cancer Cell 27 , 450–461 (2015).
   26. Khan, O. et al. TOX transcriptionally and epigenetically programs CD8+ T cell exhaustion.
   Nature 571 , 211–218 (2019).
   27. Scott, A. C. et al. TOX is a critical regulator of tumour-specific T cell differentiation. Nature
   571 , 270–274 (2019).
   28. Sade-Feldman, M. et al. Defining T cell states associated with response to checkpoint
   immunotherapy in melanoma. Cell 175 , 998–1013 (2018).
   29. Yan, Y. et al. CX3CR1 identifies PD-1 therapy-responsive CD8+ T cells that withstand
   chemotherapy during cancer chemoimmunotherapy. JCI Insight 3 , e97828 (2018).
   30. Schumacher, T. N. & Scheper, W. A liquid biopsy for cancer immunotherapy. Nat. Med. 22 ,
   340–341 (2016).
   31. Hogan, S. A. et al. Peripheral blood TCR repertoire profiling may facilitate patient
   stratification for immunotherapy against melanoma. Cancer Immunol. Res. 7 , 77–85
   (2019).
   Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
   published maps and institutional affiliations.

© The Author(s), under exclusive licence to Springer Nature Limited 2020

Tumour

singletons (t)

n = 27

0.5 2.0

IMmotion150Sunitinib

IMmotion150Atezo+Bev

AtezolizumabIMmotion150

DocetaxPOPLARel

AtezolizumabPOPLAR

AtezolizumabIMvigor210

0.950.21058

0.941.00000

0.990.44972

0.940.18418

0.810.00028

1.130.99999

Tumour

multiplets (T)

n = 19

0.5 2.0

1.098.7 × 10–1

0.623.4 × 10–6

0.764.1 × 10–5

0.851.7 × 10–2

0.684.2 × 10–8

0.847.9 × 10–9

Dual

expanded (D)

n = 24

0.5 2.0

1.036.4 × 10–1

0.651.2 × 10–10

0.733.7 × 10–7

0.832.9 × 10–3

0.721.8 × 10–7

0.812.3 × 10–14

CD8A ± T

0.21.05.0

n = 21

9

4

3

2

5

156

38

45

38

43

40

CD8A ± D

0.21.05.0

n = 16

5

5

3

3

9

161

42

44

38

42

36

CD8A ± CCL5

0.21.05.0

n = 25

8

5

5

4

8

152

39

44

36

41

37

Adj P < 0.05 P ≥ 0.05

P < 0.05 P ≥ 0.05

CD8Ahigh, signaturelow

PFS hazard ratio PFS hazard ratio CD8Ahigh, signaturehigh

ab

Fig. 4 | Survival analysis of tissue expansion gene signatures. a, Gene hazard
ratios. Each gene signature for a tissue expansion pattern involving tumour
tissue (columns) is represented by hazard ratios in treatment arms (rows)
after dichotomization within the clinical trial. Vertical jitter distinguishes
overlapping dots. Hazard ratio < 1 (left of vertical lines) indicates greater PFS.
Mean hazard ratio and one-sided P values are shown from a one-sample z-test
on hazard ratios for genes in the signature, highlighted with associated data in
red when Bonferroni-adjusted P < 0.05. b, Survival analysis of signatures with

high CD8A expression. Plots summarize results from a dual-variate survival

analysis of dichotomized values of CD8A gene expression and tissue expansion

signature scores. Points and segments show PFS hazard ratio and 95%

confidence intervals from two groups: ‘CD8Ahigh, signaturelow’ or ‘CD8Ahigh,

signaturehigh’. Pairs of groups correspond to each treatment arm in a. Vertical

lines mark hazard ratio = 1. The number of patients in each group is shown,

highlighted with the associated hazard ratio and confidence interval in red

when a two-sided P < 0.05 from a Cox proportional-hazards model.