ArticlePDF Available

Abstract and Figures

High-grade serous ovarian cancer (HGSOC) is an archetypal cancer of genomic instability1–4 patterned by distinct mutational processes5,6, tumour heterogeneity7–9 and intraperitoneal spread7,8,10. Immunotherapies have had limited efficacy in HGSOC11–13, highlighting an unmet need to assess how mutational processes and the anatomical sites of tumour foci determine the immunological states of the tumour microenvironment. Here we carried out an integrative analysis of whole-genome sequencing, single-cell RNA sequencing, digital histopathology and multiplexed immunofluorescence of 160 tumour sites from 42 treatment-naive patients with HGSOC. Homologous recombination-deficient HRD-Dup (BRCA1 mutant-like) and HRD-Del (BRCA2 mutant-like) tumours harboured inflammatory signalling and ongoing immunoediting, reflected in loss of HLA diversity and tumour infiltration with highly differentiated dysfunctional CD8+ T cells. By contrast, foldback-inversion-bearing tumours exhibited elevated immunosuppressive TGFβ signalling and immune exclusion, with predominantly naive/stem-like and memory T cells. Phenotypic state associations were specific to anatomical sites, highlighting compositional, topological and functional differences between adnexal tumours and distal peritoneal foci. Our findings implicate anatomical sites and mutational processes as determinants of evolutionary phenotypic divergence and immune resistance mechanisms in HGSOC. Our study provides a multi-omic cellular phenotype data substrate from which to develop and interpret future personalized immunotherapeutic approaches and early detection research. Multi-modal analysis of genomically unstable ovarian tumours characterizes the contribution of anatomical sites and mutational processes to evolutionary phenotypic divergence and immune resistance mechanisms.
TME of HGSOC at single-cell resolution a, Overview of the MSK SPECTRUM cohort and specimen collection workflow. b, UMAP plot of cells profiled by scRNA-seq coloured by patient. Cell types are highlighted with grey outlines. c, Patient specificity for each cell type (Methods). Ov, ovarian. d, Number of cells identified per cell type next to a UMAP plot with cells coloured by cell type. e, Number of cells profiled per tumour site next to a UMAP plot with cells coloured by tumour site. UQ, upper quadrant. f, Site-specific enrichment of cell type composition in scRNA-seq, H&E and mpIF data fitted using a GLM. GLMs for H&E and mpIF data were separated by tumour (T) and stroma (S) regions. The colour gradient indicates the log2-transformed odds ratio (red, enrichment; blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value). g, Cell type composition based on scRNA-seq data for CD45⁻ and CD45⁺ samples. Upper panels, absolute and relative cell type numbers; lower panels, box plot distributions of sample ranks with respect to tumour site. h, Cell type composition based on H&E with lymphocyte ranks in tumour and stroma. Panels are analogous to those in g. i, Cell type composition based on mpIF with CD8⁺ T cell ranks in tumour and stroma. Panels are analogous to those in g. For c and g–i, violin plots and box plots are shown as the median, top and bottom quartiles; whiskers correspond to 1.5× interquartile range (IQR). *P < 0.05, **P < 0.01.
… 
Site specificity of immunophenotypes a, UMAP plot of T and NK cell clusters profiled by scRNA-seq. Clusters are coloured and numbered to reference cluster labels in c. b, Pairwise comparisons of kernel density estimates in UMAP space. c, Left, heatmap of average T cell state module scores (left) and signalling pathway activity scores (right) across CD4⁺ T, CD8⁺ T, innate lymphoid cell (ILC), NK and cycling cell clusters. Right, dot plot showing site-specific enrichment of T and NK cell clusters based on GLM. The colour gradient indicates the log2-transformed odds ratio (red, enrichment; blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value). d, Intra-sample diversity of T and NK cell clusters estimated by Shannon entropy with samples grouped by site (patient and sample counts shown) and intra- and inter-patient dissimilarity of T and NK cell cluster composition for pairs of samples, estimated using the Bray–Curtis distance (patient and sample pair counts shown). Pairwise dissimilarity is shown for all heterotypic pairs of sites (adnexa versus non-adnexa, adnexa versus ascites, non-adnexa versus ascites). Violin plots show the median, top and bottom quartiles; whiskers correspond to 1.5× IQR. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. e, Top, diffusion maps of the subset of CD8⁺ T cells profiled by scRNA-seq, with cells coloured by CD8⁺ T cell cluster and pseudotime. Bottom, relative expression of genes marking CD8⁺ T cell clusters in diffusion space. DC, diffusion component. f, Scaled module scores with respect to pseudotime.
… 
Malignant cell phenotypes and association with mutational signatures a, Left, UMAP plot of epithelial cells coloured by cluster. Clusters are numbered to reference cluster labels in the heatmap. Right, heatmap of scaled marker gene expression averaged per cluster, showing differentially expressed genes in rows and clusters in columns. The top two genes for each cluster are highlighted. b, Top, heatmap of average signalling pathway activity scores per site. Bottom, UMAP plots with cells coloured by signalling activity scores for pathways of interest. EGFR, epidermal growth factor receptor; MAPK, mitogen-activated protein kinase; PI3K, phosphoinositide 3-kinase; VEGF, vascular endothelial growth factor. c, Relative kernel densities showing enrichment (red) and depletion (blue) in UMAP space for pairwise comparisons of mutational signatures and sites. d, Left, estimated effects of anatomical site and mutational signature on epithelial cluster composition based on GLM. The colour gradient indicates the log2-transformed odds ratio (red, enrichment; blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value). Right, epithelial cluster compositions ranked by Cancer.cell.3 fraction. Box plot panels show distributions of scaled sample ranks by mutational signature. e,f, Distributions of signalling pathway activity scores (e) and HLA gene expression (f) in adnexal and non-adnexal samples as a function of mutational signature (patient counts shown). g, Left, intra-sample diversity of malignant cell clusters in adnexal and non-adnexal samples, with samples grouped by mutational signature and site (patient and sample counts shown). Right, intra- and inter-patient dissimilarity of malignant cluster composition for pairs of samples. Pairwise dissimilarity is shown for all pairs of sites (patient and sample pair counts shown) excluding ascites (top) and for adnexal versus non-adnexal pairs of sites (bottom). In d–g, box plots and violin plots show the median, top and bottom quartiles; whiskers correspond to 1.5× IQR. Colours in e–g are analogous to those in d. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; brackets indicate two-sided Wilcoxon pairwise comparisons in e–g.
… 
Mutational signatures as determinants of immunophenotypes a, Differences in kernel density estimates in UMAP space for pairwise comparisons of mutational signatures. b, Estimated effects of mutational signature and anatomical site on T and NK cell cluster composition based on a GLM, with models fitted excluding ascites samples. The colour gradient indicates the log2-transformed odds ratio (red, enrichment; blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value). c, Distributions of CD8⁺ T cell state module scores and JAK–STAT signalling pathway activity scores with respect to mutational signature (patient counts shown). d, Scaled module scores within the subset of CD8⁺ T cells with respect to pseudotime and mutational signature. e, Correlation of JAK–STAT signalling scores in CD8⁺ T cells in CD45⁺ samples with those in cancer cells in matched CD45⁻ samples. f, Left, intra-sample diversity of T and NK cell clusters in adnexal and non-adnexal samples estimated by Shannon entropy, with samples grouped by mutational signature (patient and sample counts shown). Right, intra- and inter-patient dissimilarity in T and NK cell cluster composition, with samples grouped by mutational signature, estimated using the Bray–Curtis distance. Pairwise dissimilarity is shown for all pairs of sites (patient and sample pair counts shown) excluding ascites (top) and for adnexal versus non-adnexal pairs of sites (bottom). g, Spatial density of CD8⁺ T cell phenotypes in adnexal and non-adnexal mpIF samples as a function of distance to the tumour–stroma interface, with samples grouped by mutational signature (Methods). In c and f, box plots and violin plots show the median, top and bottom quartiles; whiskers correspond to 1.5× IQR. Colours in f and g are analogous to those in c–e. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; brackets indicate two-sided Wilcoxon pairwise comparisons in c and f.
… 
HLA loss as a mechanism of immune escape a, Left, distribution over cells of chromosome arm 6p BAF in scRNA-seq data with ranking by median 6p BAF per cell type. Right, allelic imbalance in 6p BAF across cancer cell clusters. White vertical lines indicate the median. Chr., chromosome. b, Left, percentage of cancer cells with 6p LOH per patient. Right, site- and clone-specific percentage of cancer cells with 6p LOH. Het., heterozygous. c, Percentage of cancer cells with 6p LOH per sample as a function of mutational signature. Pie charts show the fraction of samples with heterozygous, subclonal LOH and clonal LOH 6p status. d, Percentage of patients with LOH of any HLA class I gene in the MSK-IMPACT HGSOC cohort (n = 1,298 patients) for BRCA1-, BRCA2- and CDK12-mutant and CCNE1-amplified tumours, mapping to HRD-Dup, HRD-Del, TD and FBI signatures, respectively. Error bars, 95% binomial confidence intervals. e, Percentage of cancer cells with 6p LOH per sample as a function of anatomical site. Pie charts show the fraction of samples by 6p status. f, UMAP plots of cancer cells from representative HRD-Dup and FBI cases. Density plots show site-specific 6p BAF. g, Fraction of naive and dysfunctional T cells in CD45⁺ samples as a function of the 6p LOH clonality of cancer cells in matched CD45⁻ samples. *P < 0.05; brackets indicate two-sided Wilcoxon pairwise comparisons. In b, c, e and g, 6p LOH status is defined as follows: heterozygous, percentage 6p LOH ≤ 20%; subclonal LOH, 20% < percentage 6p LOH ≤ 80%; clonal LOH, percentage 6p LOH > 80%. In c, e and g, box plots and violin plots show the median, top and bottom quartiles; whiskers correspond to 1.5× IQR. In a–e, only BAF estimates from cells with ≥10 reads aligning to 6p were considered and allelic imbalance states were assigned on the basis of the mean 6p BAF per cell as follows: balanced, BAF ≥ 0.35; imbalanced, 0.15 ≤ BAF < 0.35; LOH, BAF < 0.15 (Methods).
… 
This content is subject to copyright. Terms and conditions apply.
778 | Nature | Vol 612 | 22/29 December 2022
Article
Ovarian cancer mutational processes drive
site-specific immune evasion
Ignacio Vázquez-García1,14, Florian Uhlitz1,14, Nicholas Ceglia1, Jamie L. P. Lim1, Michelle Wu2,
Neeman Mohibullah3, Juliana Niyazov1, Arvin Eric B. Ruiz4, Kevin M. Boehm1, Viktoria Bojilova1,
Christopher J. Fong1, Tyler Funnell1, Diljot Grewal1, Eliyahu Havasov1, Samantha Leung1,
Arfath Pasha1, Druv M. Patel1, Maryam Pourmaleki1, Nicole Rusk1, Hongyu Shi1, Rami Vanguri1,
Marc J. Williams1, Allen W. Zhang1, Vance Broach2, Dennis S. Chi2, Arnaud Da Cruz Paula2,
Ginger J. Gardner2, Sarah H. Kim2, Matthew Lennon2, Kara Long Roche2, Yukio Sonoda2,
Oliver Zivanovic2, Ritika Kundra5, Agnes Viale5, Fatemeh N. Derakhshan4, Luke Geneslaw4,
Shirin Issa Bhaloo4, Ana Maroldi4, Rahelly Nunez4, Fresia Pareja4, Anthe Stylianou4,
Mahsa Vahdatinia4, Yonina Bykov6, Rachel N. Grisham6,7, Ying L. Liu6,7, Yulia Lakhman8,
Ines Nikolovski8, Daniel Kelly9, Jianjiong Gao1,5, Andrea Schietinger7,10 , Travis J. Hollmann4,13 ,
Samuel F. Bakhoum11,12 , Robert A. Soslow4, Lora H. Ellenson4, Nadeem R. Abu-Rustum2,7,
Carol Aghajanian6, Claire F. Friedman6,7, Andrew McPherson1, Britta Weigelt4,
Dmitriy Zamarin6,7 ✉ & Sohrab P. Shah1 ✉
High-grade serous ovarian cancer (HGSOC) is an archetypal cancer of genomic
instability1–4 patterned by distinct mutational processes5,6, tumour heterogeneity7–9
and intraperitoneal spread7,8,10. Immunotherapies have had limited ecacy in
HGSOC1113, highlighting an unmet need to assess how mutational processes and the
anatomical sites of tumour foci determine the immunological states of the tumour
microenvironment. Here we carried out an integrative analysis of whole-genome
sequencing, single-cell RNA sequencing, digital histopathology and multiplexed
immunouorescence of 160 tumour sites from 42 treatment-naive patients with
HGSOC. Homologous recombination-decient HRD-Dup (BRCA1 mutant-like) and
HRD-Del (BRCA2 mutant-like) tumours harboured inammatory signalling and
ongoing immunoediting, reected in loss of HLA diversity and tumour inltration
with highly dierentiated dysfunctional CD8+ Tcells. By contrast, foldback-inversion-
bearing tumours exhibited elevated immunosuppressive TGFβ signalling and immune
exclusion, with predominantly naive/stem-like and memory Tcells. Phenotypic state
associations were specic to anatomical sites, highlighting compositional, topological
and functional dierences between adnexal tumours and distal peritoneal foci. Our
ndings implicate anatomical sites and mutational processes as determinants of
evolutionary phenotypic divergence and immune resistance mechanisms in HGSOC.
Our study provides a multi-omic cellular phenotype data substrate from which to
develop and interpret future p er so na lized i mm unotherapeutic approaches and early
detection research.
The principal defining features of high-grade serous ovarian can-
cer (HGSOC) are profound structural variations in the form of copy
number alterations and genomic rearrangements, which accrue on
a genetic background of nearly ubiquitous TP53 mutation
14
. Somatic
and germline alterations in homologous recombination (HR) repair
pathway genes such as BRCA1 and BRCA2 lead to HR deficiency (HRD)
in approximately half of HGSOC cases15. Beyond gene alterations,
patients stratify by endogenous mutational processes
3,16
as inferred
from structural variation patterns in whole-genome sequencing (WGS),
including HRD subtypes (BRCA1-associated tandem duplications,
HRD-Dup; BRCA2-associated interstitial deletions, HRD-Del), CCNE1
amplification-associated foldback inversion (FBI)-bearing tumours
and CDK12-associated tandem duplicator (TD)-bearing tumours5,6.
HGSOC presents a distinctive clinical challenge resulting from the
widespread intraperitoneal disease at diagnosis. Long latency allows
for broad periods of clonal diversification and tumour–immune inter-
actions to unfold in the heterogeneous microenvironments of the
peritoneal cavity7,10,17. This raises key questions about how underlying
mutational processes and local tissue sites influence clonal selection,
tumour microenvironments (TMEs) and immune recognition. We car-
ried out a prospective study, capturing mutational processes from WGS,
cell phenotypes from single-cell RNA sequencing (scRNA-seq) and
https://doi.org/10.1038/s41586-022-05496-1
Received: 20 September 2021
Accepted: 28 October 2022
Published online: 14 December 2022
Open access
Check for updates
A list of afiliations appears at the end of the paper.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Nature | Vol 612 | 22/29 December 2022 | 779
spatial topology from insitu multiplexed cellular imaging in multi-site
cases of HGSOC. Our findings identify distinct immunostimulatory
and immunosuppressive mechanisms that co-segregate with sites of
disease and mutational processes, thereby defining new determinants
of immune recognition and escape in HGSOC.
Multi-site tissue biopsies (n = 160) were collected from newly diag-
nosed, treatment-naive patients (n = 42) undergoing laparoscopy or
primary debulking surgeries over a 24-month period (Fig.1a). Collec-
tion took place at anatomical sites including adnexa (that is, potential
primary lesions), omentum, peritoneum, bowel, ascites and other intra-
peritoneal sites (Extended Data Fig.1a). The clinical characteristics of all
patients are summarized in Extended Data Fig.1b and Supplementary
Table1. Patient samples were profiled using scRNA-seq on CD45+ and
CD45
flow-sorted fractions (Supplementary Table2), haematoxylin
and eosin (H&E) staining and multiplexed immunofluorescence (mpIF)
on fixed tissue sections, clinical tumour–normal sequencing of 468
cancer genes by MSK-IMPACT and whole-genome tumour–normal
sequencing (Extended Data Fig.1a,b and Methods). WGS copy num-
ber profiles were highly concordant with an external ‘meta-cohort’
derived from several HGSOC WGS studies5 (Extended Data Fig.2a).
Mutational signature inference from WGS data yielded 16 HRD-Dup,
6 HRD-Del and 14 FBI tumours (Extended Data Figs.1b and2b,c and
Supplementary Tables1 and3), with model features consistent with
previous reports5 (Extended Data Fig.2d,e), stable across multiple
computational methods18,19 and in agreement with BRCA1 and BRCA2
mutations and clinical HRD testing (Extended Data Fig.2b). Further-
more, tumours with high-level amplifications in CCNE1 (Extended Data
Fig.2f) and MYC exhibited expected distributions of gene amplifica-
tions within signature assignments and cis-acting gene expression
correlates (Extended Data Fig.2g,h).
Site-specific TMEs
We constructed a cell type map from the scRNA-seq data, quanti-
fying nine broad cellular lineages: epithelial cells, lymphoid cells
(T and natural killer (NK) cells, B cells, plasma cells), myeloid cells
(monocytes/macrophages, dendritic cells (DCs), mast cells) and stro-
mal cells (fibroblasts, endothelial cells) (Fig.1b,d, Extended Data
Fig.3a,b and Supplementary Table4). Ovarian cancer cells exhibited
high patient specificity (Fig.1c and Methods), attributed to tumour
CD45
5,000
10,000
No. cells
0
50
100
% cells
Ascites
Other
UQ
Peritoneum
Omentum
Bowel
Adnexa
CD45+
*
*
Scaled rank
(Ov cancer cell)
–1 01–1
01
–1 01–1
01
–1 01–1
01
Scaled rank
(T cell)
Tumour
600,000
1,200,000
No. cells
0
50
100
% cells
**
Other
UQ
Peritoneum
Omentum
Bowel
Adnexa
Stroma
*
Scaled rank
(lymphocyte)
Scaled rank
(lymphocyte)
Scaled rank
(CD8
+
)
Scaled rank
(CD8
+
)
Tumour
150,000
300,000
No. cells
0
50
100
% cells
Other
UQ
Peritoneum
Omentum
Bowel
Adnexa
Stroma
Cell type (H&E)
Lymphocyte
Other
Cell type (mpIF)
PanCK+
CD8+
CD68+
Other
Unknown
Patient
002
003
007
008
009
014
022
024
025
026
031
036
037
041
042
045
049
050
051
052
053
054
065
067
068
070
071
075
077
080
081
082
083
090
105
107
110
112
115
116
118
UMAP1
UMAP2
Cell type
Mast cell (1,257)
Dendritic cell (4,814)
Endothelial cell (18,531)
B cell (18,673)
Plasma cell (20,944)
Fibroblast (162,078)
Myeloid cell (201,217)
T cell (250,335)
Ov cancer cell (251,837)
Ovarian cancer cell
Fibroblast
Endothelial
cell
Plasma
cell
Mast cell
T cell
B cell
Dendritic cell
Myeloid cell
n = 929,686
scRNA
mpIF
−log10(P)log2(odds ratio)
0
125
250
≤–1.5
0
≥1.5
Region
Tumour
Stroma
a d
e
UQ
Peritoneum
Omentum
Bowel
Ascites
Adnexa
T cell
B cell
Plasma cell
Myeloid cell
Dendritic cell
Mast cell
Endothelial cell
Fibroblast
Ov cancer cell
Lymphocyte
Lymphocyte
PanCK+
CD8+
CD68+
PanCK+
CD8+
CD68+
f
g
h
−4
−2
0
2
4
6
Ov cancer
Fibroblast
Myeloid cell
T cell
Dendritic cell
Plasma cell
Endothelial cell
B cell
Mast cell
Patient
specicity
i
c
b
HGSOC cohort
No. patients = 42
No. sites = 160
scRNA-seq (156)
H&E (100)
mpIF (100)
Bulk WGS (40)
MSK-IMPACT (42)
H&E
Site
scRNAH&E mplF
TTSS
Other (29,860)
UQ (52,094)
Peritoneum (102,404)
Bowel (105,500)
Ascites (108,285)
Omentum (208,894)
Adnexa (322,649)
Fig. 1 | TME o f HGSOC at sin gle-cell r esolution . a, Overview of the M SK
SPECTRUM coh ort and spec imen collect ion workflow. b, UM AP plot of cells
profile d by scRNA-seq co loured by patien t. Cell type s are highlighte d with grey
outlines. c, Patient spe cificity for e ach cell typ e (Methods). Ov, ovarian.
d, Number of ce lls identif ied per cell ty pe next to a UMA P plot with cell s
coloured by ce ll type. e, Num ber of cells prof iled per tumo ur site next to
a UMAP plo t with cells colo ured by tumour site . UQ, upper quad rant.
f, Site-spec ific enric hment of cell ty pe composit ion in scRNA-se q, H&E and
mpIF data f itted usi ng a GLM. GLMs for H& E and mpIF data we re separated
by tumour(T) and s troma(S) regions. Th e colour gradie nt indicates t he
log2-transfor med odds rati o (red, enrichment ; blue, depletio n), and sizes
indicate the Bonferroni-corrected –log10(P value). g, Cell type compos ition
based on s cRNA-seq dat a for CD45 and CD45+ sample s. Upper pane ls, absolute
and relative ce ll type numb ers; lower panels, b ox plot distribu tions of sample
ranks wit h respect to t umour site. h, Cell t ype compo sition based o n H&E with
lymphoc yte ranks in tu mour and stroma . Panels are analogo us to those in g.
i, Cell typ e compositio n based on mpIF w ith CD8+ T cell rank s in tumour and
stroma. Pa nels are analogou s to those in g. For c and gi, vio lin plots and box
plots are show n as the median, t opand bottom quar tiles; whiskers co rrespond
to 1.5×inte rquartile rang e (IQR). *P < 0.05, **P < 0.01.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
780 | Nature | Vol 612 | 22/29 December 2022
Article
cell-specific somatic copy number alterations driving gene dosage
effects. Immune cell composition varied across anatomical sites
within patients (Fig.1e,f). Whereas CD45
fractions (ranging from
fibroblast-rich to cancer cell-rich samples) were largely conserved
between anatomical sites (Fig.1g), CD45
+
fractions (ranging from
myeloid-rich to lymphoid-rich samples) were substantially different
(Fig.1f,g and Extended Data Fig.3c). Unsurprisingly, ascites samples
were enriched for Tcells (Mann–Whitney U test, Benjamini–Hoch-
berg (BH)-corrected Q = 0.0195) and DCs (Q < 1 × 10
−4
), while adnexal
samples were comparatively depleted for Tcells (Q = 1.95 × 10
−2
), B
cells (Q = 4.1 × 10−3) and DCs (Q = 6.0 × 10−3). Among solid tumour
sites, higher lymphocyte and CD8+ Tcell fractions were found in
non-adnexal sites in scRNA-seq, whole-slide H&E and mpIF (Fig.1g–i
and Extended Data Figs.3c–e and4a–c) in both tumour and stromal
regions (Extended Data Fig.3f).
Inter-site compositional variation within patients stimulated
deeper analyses to assess immune cell phenotypic states. We identi-
fied 10 major T and NK cell clusters with 41 minor subclusters (Fig.2a,
Extended Data Fig.5a,b and Supplementary Table4), broadly defining
CD4
+
Tcells (clusters 1–10), CD8
+
Tcells (clusters 11–19), innate-like
and γδ Tcells (clusters 20–23), NK cells (clusters 24–33) and cycling cells
(clusters 34–41). Clusters were annotated on the basis of known marker
genes and cross-referenced against other published annotations20,21
(Extended Data Fig.5c). T and NK cell clusters followed a gradient
across uniform manifold approximation and projection (UMAP)
space (Fig.2b), highlighting site-specific phenotypic differencesthat
werequantified by fitting a generalized linear model (GLM) of cluster
composition (Fig.2c). In particular, naive/stem-like and central memory
CD4
+
Tcells (cluster 1) were depleted in adnexal samples and enriched
in ascites (Extended Data Fig.7a,b and Supplementary Table5). Con-
versely, dysfunctional CD4
+
and CD8
+
Tcells (clusters 3–5 and 15–17)
were depleted in ascites but enriched in adnexal and other tumour sites
(Supplementary Table5), in line with dysfunction driven by chronic
antigen exposure in solid tumours. Clusters for regulatory Tcells (7–10)
and regulatory NK cells (27–33) were also enriched in adnexal samples
(Fig.2c), potentially indicative of increased immunomodulatory feed-
back at these sites.
Comparisons of solid tumour sites within patients showed naive/
stem-like and central memory Tcell enrichment in non-adnexal sites
(22 of 31 patients) and dysfunctional Tcell enrichment in adnexal sites
(19 of 31 patients)(Extended Data Fig. 7a, vector plot). Shannon entropy
analysis indicated higher within-site variation of Tcell phenotypes in
adnexal samples, suggesting the coexistence of differentiated states
within the primary site relative to non-adnexal samples (Fig.2d).
b
T/NK cell cluster
1
2
3
45
6
7
8
9
10
11
12
13
14
15
16 17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
n = 221,315
UMAP1
UMAP2
Adnexa
Bowel
Omentum
Peritoneum
UQ
Ascites
≤–0.5 0≥0.5
–log10(P)
log2(odds ratio)
0 0.5 1.0 1.5 01 23
Mean
PROGENy
score
Cycling
02550
NK
CD8+
CD4+1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
CD4.T.reg.ISG
CD4.T.reg.3
CD4.T.reg.2
CD4.T.reg.1
CD4.T.ISG
CD4.T.dysfunc.late.2
CD4.T.dysfunc.late.1
CD4.T.dysfunc.early
CD4.T.effector.memory
CD4.T.naive.centr.mem
CD8.T.ISG.late
CD8.T.ISG.early
CD8.T.dysfunc.ISG
CD8.T.dysfunc.late
CD8.T.dysfunc.early
CD8.T.cytotoxic
CD8.T.effector.memory
CD8.T.naive.centr.mem.2
CD8.T.naive.centr.mem.1
gd.T.cell
ILC.c
ILC.b
ILC.a
NK.reg.CRTAM
NK.reg.ISG
NK.reg.KRT81.KRT86.2
NK.reg.KRT81.KRT86.1
NK.reg.CCL3
NK.reg.CD56
NK.reg.IGFBP2
NK.cytotoxic.SPON2.2
NK.cytotoxic.SPON2.1
NK.cytotoxic.GZMH
Cycling.NK
Cycling.CD8.T.6
Cycling.CD8.T.5
Cycling.CD8.T.4
Cycling.CD8.T.3
Cycling.CD8.T.2
Cycling.CD8.T.1
Cycling.CD4.T
Naive
Predysfunc.
Cytotoxic
Dysfunc.
JAK–STAT
TGFβ
TNFα
(a) Adnexa vs
(b) non-adnexa
(a) Adnexa vs
(b) ascites
(a) Non-adnexa vs
(b) ascites
−1.0 0 1.0
Enrichment
in (a) over (b)
CD8
+
T
dysfunc.
CD8
+
T
naive/mem.
CD4
+
T
naive/mem.
CD4
+
T
dysfunc.
e
1
Scaled
expression
0
DC1
DC2
Cluster
CD8.T.naive.centr.mem.1
CD8.T.naive.centr.mem.2
CD8.T.effector.memory
CD8.T.cytotoxic
CD8.T.dysfunc.early
CD8.T.dysfunc.late
CD8.T.dysfunc.ISG
CD8.T.ISG.early
CD8.T.ISG.late
ISG15CXCL13LEF1 GZMK
0
0.5
Pseudotime
Naive
Dysfunc.
ISG
f
0
1
Pseudotime
Scaled module score
Module
CD8.Naive
CD8.Predysf.
CD8.Cytotoxic
CD8.Dysfunctional
JAK.STAT.pathway
a
cd
39 2533
75 2546
*
****
****
Intra-
sample Intra-
patient
Inter-
patient
3
4
5
Shannon entropy
(T/NK cells)
Site
114
23
38
14
80
21
4,978
1,392
1,786
1,086
3,136
1,266
****
****
**
****
****
****
0
0.25
0.50
0.75
1.00
Pairwise dissimilarity
(T/NK cells)
Site j
Site i
Site
Adnexa
Non-adnexa
Ascites
Mean
module
score
39
25
33
7
2
5
4
6
*
****
****
114
23
3
8
14
8
0
21
****
***
*
**
4,978
1
,
392
1,786
1
,
086
3
,136
1
,
26
6
***
*
****
****
Fig. 2 | Site specificity of immunophenotypes. a, UMAP pl ot of T and NK
cell cluste rs profiled by s cRNA-seq. Clu sters are coloure d and numbered to
reference cluster labels in c. b, Pairwis e comparisons of kernel density
estimate s in UMAP spa ce. c, Left, hea tmap of average Tcell st ate module
scores (le ft) and signallin g pathway activ ity scores (rig ht) across CD4+ T, CD 8+
T, innate lymphoid ce ll (ILC), NK and cycling cell clu sters. Rig ht, dot plot
showing sit e-specif ic enrichmen t of T and NK cell clust ers based on GL M. The
colour gradient indicates the log2-tra nsformed odds r atio (red, enrichm ent;
blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value).
d, Intra-sa mple diversity of T a nd NK cell cluste rs estimate d by Shannon
entropy wit h samples group ed by site (patie nt and sample cou nts shown) and
intra- and int er-patient dissimil arity of T and NK c ell cluster comp osition for
pairs of samp les, estimat ed using the Bray–Cu rtis dista nce (patient an d sample
pair count s shown). Pairwis e dissimilarit y is shown for all hete rotypic pair s of
sites (adnexa ver sus non-adnexa, adn exa versus ascit es, non-adnexa ver sus
ascites).Violinplot s showthe median, topand b ottomquartile s; whiskers
corresp ond to 1.5× IQ R. *P < 0.05, **P < 0.01, ** *P <0.001, ****P < 0.0 001. e, Top,
diffusi on maps of the subs et of CD8+ Tcells prof iled by scRNA-s eq, with cells
coloured by CD 8+ Tcell cluster and ps eudotime. Bo ttom, relative ex pression of
genes mark ing CD8+ Tcell cluste rs in diffusi on space. DC, d iffusion co mponent.
f, Scaled mo dule scores wit h respect to p seudotime.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Nature | Vol 612 | 22/29 December 2022 | 781
Intriguingly, analysis of site-to-site Bray–Curtis dissimilarity showed
high compositional differences between solid tumours and ascites
both within and between patients (Fig.2d and Extended Data Fig.7c).
Differentiation trajectories projected CD8
+
Tcells along two axes of (1)
terminally dysfunctional and (2) interferon-stimulated gene (ISG) Tcell
states (Fig.2e), defined by loss of naive Tcell markers and acquisition
of dysfunctional and cytotoxic traits (Fig.2f). The trajectories were
associated with loss of transcription factors expressed in naive and
central memory Tcells (TCF1 and LEF1) and acquisition of type I inter-
feron (IFN) (ISG15), cytotoxic function (GZMK) and Tcell dysfunction
(TOX, CXCL13 and PDCD1) (Extended Data Fig.7d). Notably, expression
trajectories also differed across sites, with ascites exhibiting high cyto-
toxic module scores in contrast to the high dysfunctional Tcell scores
in adnexa and omentum (Extended Data Fig.7e).
Phenotypic state composition in myeloid and DC compartments also
varied as a function of site (Extended Data Fig.6a–c). DCs clustered
1
23
4
5
6
7
89
10
1
2
3
4
5
6
9
10
7
8
Cancer.cell.1
0125 250
Cancer.cell.2
Cancer.cell.3
Cancer.cell.4
Cancer.cell.5
Cancer.cell.6
Cycling.cancer.cell.1
Cycling.cancer.cell.2
Ciliated.cell.1
Ciliated.cell.2
AdnexaNon-adnexa Ascites
Mutational
signature
HRD-Dup
HRD-Del
FBI
–log10(P)
≤–20 ≥2
log2(odds ratio)
ab
Cluster
Cancer.cell.1
Cancer.cell.2
Cancer.cell.3
Cancer.cell.4
Cancer.cell.5
Cancer.cell.6
Cycling.cancer
.cell.1
Cycling.cancer
.cell.2
Ciliated.cell.1
Ciliated.cell.2
1 2 3 4 5 6 7 8 910
CETN2
CAPS
CAPS
C20orf85
TUBA1B
HIST1H4C
PTTG1
CENPF
VEGFA
SST
S100A4
CASC1
COL3A1
COL1A1
ISG15
CXCL10
S100A9
KRT17
DAPL1
Cell type
Scaled expression
0
0.2
0.4
0.6
0.8
1
d
c
1
23
4
5
6
7
89
10
JAK–STAT NF-κBTNF TGFβHypoxia
≤−1
0
2
≥4
PROGENy
score
UMAP1
UMAP2
1
2
3
4
5
6
9
1
0
7
8
Cancer.1
Cancer.2
Cancer.3
Cancer.4
Cancer.5
Cancer.6
Cycling.cancer.1
Cycling.cancer.2
Ciliated.1
Ciliated.2
AdnexaNon-adnexa Ascites
Androgen
EGFR
Oestrogen
Hypoxia
JAK–STAT
MAPK
NF-κB
p53
PI3K
TGFβ
TNF
Trail
VEGF
WNT
Androgen
EGFR
Oestrogen
Hypoxia
JAK–STAT
MAPK
NF-κB
p53
PI3K
TGFβ
TNF
Trail
VEGF
WNT
Androgen
EGFR
Oestrogen
Hypoxia
JAK–STAT
MAPK
NF-κB
p53
PI3K
TGFβ
TNF
Trail
VEGF
WNT
−0.5 0.51.5
Mean
PROGENy
score
(a) HRD-Dup
vs (b) HRD-Del
UMAP1
UMAP2
1
23
4
5
6
7
89
10
1
23
4
5
6
7
89
10
(a) HRD-Dup
vs (b) FBI
1
23
4
5
6
7
89
10
(a) HRD-Del
vs (b) FBI
1
23
4
5
6
7
89
10
(a) Non-adnexa
vs (b) adnexa
1
23
4
5
6
7
89
10
(a) Ascites
vs (b) adnexa
1
23
4
5
6
7
89
10
(a) Ascites vs
(b) non-adnexa −1.0
−0.5
0
0.5
1.0
Enrichment
in (a) over (b)
AdnexaNon-adnexa Ascites
4,000
8,000
No. cells
0
50
100
% cells
Site
HRD-Dup
HRD-Del
FBI
Scaled rank (Cancer.cell.3)
e
f
g
14 210
24 412
16 136
30 2313
Intra-
sample
Intra-
patient
All
AllAdnexa
Non-adnexa
Inter-
patient
1.5
2.0
2.5
1.5
2.0
2.5
Shannon entropy (cancer cells)
Signature
16 126
161 8741
12 92
64 3212
*
240 13230
2,048 874184
*****
236 12618
1,040 40876
** **
0
0.25
0.50
0.75
0
0.25
0.50
0.75
Pairwise dissimilarity (cancer cells)
Signature j
Signature i
24
4
11
**
31
13
23
*
24
4
11
*
***
31 13 23
24 411
**
31 13 23
24 411
***
31 13 23
*
24 411
*
**
31 13 23
*
24 411
**
**
31 13 23
HLA-A HLA-B HLA-C B2M HLA-DRA HLA-DRB1
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
0
1
2
3
4
0
1
2
3
4
Expression
24 411
**
3113 23
24 411
**
3113 23
24 4
11
**
3113 23
24 411
3113 23
*
****
24 411
*
3113 23
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
−1
0
1
2
3
−1
0
1
2
3
PROGENy score
Adnexa
JAK–STAT NF-κBTNF TGFβHypoxia
Non-adnexa
AdnexaNon-adnexa
AdnexaNon-adnexa
UMAP1
UMAP2
n = 211,624
1
23
4
5
6
7
89
10 1
23
4
5
6
7
89
10 1
23
4
5
6
7
89
10 1
23
4
5
6
7
89
10
**
Fig. 3 | Mali gnant cell ph enotype s and assoc iation with mu tational
signatures. a, Left, UM AP plot of epithe lial cells coloure d by cluster. Clusters
are numbere d to reference clus ter labels in the h eatmap. Righ t, heatmap of
scaled mar ker gene expressio n averaged per clus ter, showing different ially
expresse d genes in rows and clu sters in column s. The top two gen es for each
cluster are highlighted. b, Top, he atmap of average sig nalling pathway ac tivity
scores pe r site. Bottom , UMAP plots w ith cells colour ed by signalling ac tivity
scoresfor pathway s of interest.EGFR , epidermal grow th factor rec eptor;
MAPK, mi togen-activa ted protein kina se; PI3K, phosp hoinositide 3 -kinase;
VEGF, vascular endoth elial growth fac tor. c, Relative kernel de nsities show ing
enrichme nt (red) and depleti on (blue) in UMAP sp ace for pairwi se compariso ns
of mutational signatures and sites. d, Left, estimat ed effect s of anatomical si te
and mutational signature on epithelial cluster composition based on GLM. The
colour gradient indicates the log2-tra nsformed odds r atio (red, enrichm ent;
blue, depletion), and sizes indicate the Bonferroni-corrected –log10(P value).
Right, epithelial cluster compositions ranked by Cancer.cell.3 fraction. Box
plot panels show distributions of scaled sample ranks by mutational signature.
e,f, Distribu tions of signall ing pathway acti vity score s (e) and HLA gene
expressio n (f) in adnexal and non-adn exal samples as a f unction of mut ational
signature (patient counts shown). g, Left, intr a-sample diversi ty of malignan t
cell cluste rs in adnexal and non -adnexal sample s, with sample s grouped by
mutational signature and site (patient and sample counts shown). Right, intra-
and inter-patient dissimilarity of malignant cluster composition for pairs of
samples . Pairwise di ssimilarit y is shown for all pairs of s ites (patien t and sample
pair count s shown) excluding asci tes (top) and for adnexal versu s non-adnexal
pairs of site s (bottom).Indg, box plots an d violin plotsshowthe me dian,
topand bottomquar tiles; whisker s correspond t o 1.5× IQR . Colours in eg
areanalogous to th osein d. *P < 0.05, **P < 0.01, ** *P < 0.001, *** *P < 0.0001;
brackets indicate two-sided Wilcoxon pairwise comparisons in eg.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
782 | Nature | Vol 612 | 22/29 December 2022
Article
into conventional DCs (cDC1s, cDC2s), mature cDCs (mDCs) and plas-
macytoid DCs (pDCs), marked by expression of CLEC9A, CLEC10A,
BIRC3 and PTGDS, respectively (Extended Data Figs.5d and6a and
Supplementary Table4). In addition, six different clusters of classical
and alternatively activatedmacrophages were identified
22,23
, as well
as cycling macrophages(Cycling.M) and phagoc yticmacrophages
(Clearing.M) (Extended Data Figs.5d,e and6b and Supplementary
Table4). Both GLMs and kernel density estimatesof cluster composi-
tion highlighted inter-site differences (Extended Data Fig.7f), including
cDC2 and M2.SELENOP depletion in ascites and enrichment in adnexa
(Extended Data Fig.6a–c and Supplementary Table5). Conversely,
M1.S100A8 macrophage fractions were decreased in adnexa and
increased in ascites (Extended Data Figs.6b,c and7f and Supplemen-
tary Table5). Similarly to Tcells, major compositional differences were
noted between solid tumour foci and ascites both within and between
patients (Extended Data Fig.6d).
Thus, phenotypic immune state differentiation for both lymphoid
and myeloid cells was strongly linked to tumour site, underlying both
within- and between-patient variation in TMEs andproviding clear
evidence that ascites immunophenotypic composition is unrepre-
sentative of solid tumours.
Tumour cell phenotypic diversification
We next defined how mutational processes in cancer cells influenced
cancer cell-intrinsic signalling and immune phenotypes. We identified
ten epithelialclusters from CD45 cells (Fig.3a, Extended Data Fig.8a
and Supplementary Table4), including cells with elevated Janus kinase
( JAK)–signal transducer and activator of transcription (STAT), nuclear
factor (NF)-κB and tumour necrosis factor (TNF) signalling (Cancer.
cell.3), transforming growth factor β (TGFβ) signalling (Cancer.cell.4)
and hypoxia (Cancer.cell.6) (Fig.3b). Mutational signature-specific
cluster enrichments included Cancer.cell.3 in HRD-Dup and Cancer.
cell.6 in FBI (Fig.3c,d, Extended Data Fig.8c,d and Supplementary
Table5). All three immune signalling pathways in Cancer.cell.3 were sub-
stantially increased in the adnexal lesions of HRD-Dup cases compared
23 512
*
33 14 20
**
23 512
****
33 14 20
*
****
CD8
Naive
CD8
Dysfunct.
JAK–STAT
pathway
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
HRD-Dup
HRD-Del
FBI
−0.3
0
0.3
0.6
−0.3
0
0.3
0.6
Module score
AdnexaNon-adnexa
AdnexaNon-adnexa
23 512
**
*
33 14 20
*
−1
0
1
2
3
−1
0
1
2
3
PROGENy score
a
c
d
CD8.Naive CD8.Dysfunctional JAK–STAT pathway
0
1
Pseudotime
Scaled module score
Signature HRD-Dup HRD-Del FBI
UMAP1
UMAP2
CD8+ T
dysfunc.
CD8+ T
naive/mem.
CD4+ T
naive/mem.
CD4+ T
dysfunc.
−1.0
−0.5
0
0.5
1.0
Enrichment
in (a) over (b)
(a) HRD-Dup vs
(b) HRD-Del
(a) HRD-Dup vs
(b) FBI
(a) HRD-Del vs
(b) FBI
b
≤–0.75 0
01020
≥0.75
log2(odds ratio) –log10(P)
3
4
5
2
6
1
7
8
9
10
14
15
17
16
13
18
19
11
12
23
20
21
22
24
25
26
29
28
33
27
32
30
31
34
35
36
37
38
39
40
41
CD4.T.reg.ISG
CD4.T.reg.3
CD4.T.reg.2
CD4.T.reg.1
CD4.T.ISG
CD4.T.dysfunc.late.2
CD4.T.dysfunc.late.1
CD4.T.dysfunc.early
CD4.T.effector.memory
CD4.T.naive.centr.mem
CD8.T.ISG.late
CD8.T.ISG.early
CD8.T.dysfunc.ISG
CD8.T.dysfunc.late
CD8.T.dysfunc.early
CD8.T.cytotoxic
CD8.T.effector.memory
CD8.T.naive.centr.mem.2
CD8.T.naive.centr.mem.1
gd.T.cell
ILC.c
ILC.b
ILC.a
NK.reg.CRTAM
NK.reg.ISG
NK.reg.KRT81.KRT86.2
NK.reg.KRT81.KRT86.1
NK.reg.CCL3
NK.reg.CD56
NK.reg.IGFBP2
NK.cytotoxic.SPON2.2
NK.cytotoxic.SPON2.1
NK.cytotoxic.GZMH
Cycling.NK
Cycling.CD8.T.6
Cycling.CD8.T.5
Cycling.CD8.T.4
Cycling.CD8.T.3
Cycling.CD8.T.2
Cycling.CD8.T.1
Cycling.CD4.T
Adnexa
Intra-
sample
Intra-
patient
Inter-
patient
Non-adnexa
Cycling
NK
CD8+
CD4+
Stroma Tumour
Stroma Tumour
Stroma Tumour
Stroma Tumour
Adnex
aN
on-adnexa
CD8
+
PD-1
+
TOX
CD8
+
PD-1
+
TOX
+
0
5
10
15
20
0
–100 –500+50+100–100–50 0+50 +100
5
10
15
20
Distance to tumour–stroma interface (
μ
m)
% CD8+ cells
g
Mutational
signature
HRD-Dup
HRD-Del
FBI
e
0
1
2
01
JAK−STAT score
(cancer cells)
JAK−STAT score
(CD8+ T cells)
Mutational
signature
HRD-Dup
R = 0.62, P < 2.2 × 10–16
HRD-Del
FBI
f
14 312
23 512
***
***
16 126
33 2014
****
3.5
4.0
4.5
5.0
3.5
4.0
4.5
5.0
Shannon entropy (T/NK cells)
Signature
16 136
1866451
*** *
11 63
68 2014
** *
240 15630
2,314 612238
***
23213424
1,150 268106
0
0.2
0.4
0.6
0
0.2
0.4
0.6
Pairwise dissimilarity (T/NK cells)
Signature j
Signature i
AdnexaNon-adnexa
All AllAdnexa Non-adnexa
Fig. 4 | Mutational signatures as determinants of immunophenotypes.
a, Differenc es in kernel den sity estima tes in UMAP sp ace for pairwis e
comparisons of mutational signatures. b, Estimated ef fects of mut ational
signature and anatomical site on T and NK cell cluster composition based
on a GLM, wit h models fit ted excluding as cites sample s. The colour g radient
indicate s the log2-tran sformed odds ra tio (red, enrichme nt; blue, deplet ion),
and sizes indicate the Bonferroni-corrected –log10(P value). c, Distributions
of CD8+ Tcell sta te module score s and JAK–STAT s ignalling pathw ay activit y
scores with respect to mutational signature (patient counts shown). d, S caled
module scor es within the su bset of CD8+ Tcells w ith respec t to pseudotime
and mutational signature. e, Correlation of J AK–STAT sig nalling scores in
CD8+ Tcells in CD45+ s amples with th ose in cancer c ells in matched CD 45
samples. f, Left, intr a-sample diversi ty of T and NK cell clu sters in adnexal a nd
non-adnexal s amples estim ated by Shannon en tropy, with samples g rouped by
mutational signature (patient and sample counts shown). Right, intra- and
inter-patient di ssimilarity i n T and NK cell clust er compositio n, with sample s
grouped by mu tational sig nature, estim ated using the Br ay–Curtis dist ance.
Pairwi se dissimilari ty is shown for all pair s of sites (patie nt and sample pai r
counts show n) excluding ascite s (top) and for adnexal versus no n-adnexal pairs
of sites (b ottom). g, Spatial den sity of CD8+ Tcell phe notypes i n adnexal and
non-adnexal mp IF samples as a f unction of dis tance to the tum our–stroma
interfac e, with sample s grouped by mut ational sign ature (Methods).In c and
f, box plots and v iolin plots showthe me dian, topand botto mquartiles;
whiskers cor respond to 1. 5× IQR. Colo urs in f and gareanalogous to th ose
in ce.*P < 0.05, **P < 0.01, * **P < 0.001, ** **P < 0.0001; brac kets indicate
two-sided Wilcoxon pairwise comparisons in c and f.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Nature | Vol 612 | 22/29 December 2022 | 783
with FBI cases (Fig.3e; P = 4.1 × 10−3, 5.2 × 10−3 and 5.2 × 10−3). This was
not seen in non-adnexal lesions, implying that cell-intrinsic immune
signalling in HRD-Dup cases originates in primary tumours. By contrast,
TGFβ signalling was more prominent in non-adnexal sites of FBI cases
(Fig.3e; P < 1 × 10−4), linking FBI-specific activation of TGFβ signalling to
the metastatic process. We note that within-patient differences in path-
way activity were not linked to copy number clone identity (Extended
Data Fig.9a); for example, we seedifferences in JAK–STAT pathway
activityindependently of thecopy-number profile in the same patient
(Extended Data Fig.9b,c).
Notably, cancer cell clusters differed by expression of major histo-
compatibility complex (MHC)-encoding genes (Fig.3f and Extended
Data Fig.8e–g). MHC class I genes (HLA-A, HLA-B, HLA-C and B2M) and
MHC class II genes (HLA-DRA and HLA-DRB1) were highly expressed in
Cancer.cell.3, with upregulation in HRD relative to FBI adnexal tumours
(Fig.3f), indicative of increased antigen presentation accompanied
by upregulated expression of CD274 (PD-L1) (Extended Data Fig.8h;
P = 2.8 × 10−3). While at the sample level Shannon entropy showed
similar levels of cell-intrinsic diversification across the mutational
signatures, FBI tumours exhibited statistically higher Bray–Curtis
dissimilarity in adnexa versus non-adnexa sample pairs (Fig.3g and
Extended Data Fig.8i), potentially indicating that these cancer cells
have a greater capacity for phenotypic diversity when migrating to
distal sites.
In addition, stark compositional differences of naive and dysfunc-
tional Tcells were observed as a function of mutational signature
(Fig.4a), with enrichment for naive/stem-like and central memory
Tcell clusters (1, 2, 11 and 12) in FBI tumours and dysfunctional Tcells
(3–5 and 15–17) in HRD tumours (Fig.4b, Extended Data Fig.10b and
Supplementary Table5). This was similarly reflected in higher JAK–STAT
signalling in HRD-Dup tumours (Fig.4c and Extended Data Fig.10a)
and along differentiation trajectories of Tcell phenotypes (Fig.4d).
Tcell-intrinsic and cancer cell-intrinsic JAK–STAT signalling was cor-
related in matched samples across the mutational subtypes (Fig.4e).
Higher phenotypic Tcell state diversity was found in HRD-Dup tumours,
accompanied by remarkably consistent intra-patient Bray–Curtis indi-
ces, suggesting that diversification processes were recurrent across
patients (Fig.4f).
Using mpIF, we tested whether heightened immune signalling in HRD
tumours could be attributed to reciprocal interactions between cancer
cells and immune cells in the tumour and stromal compartments in the
TME (Fig.4g and Extended Data Fig.10d). Activated CD8
+
PD-1
+
TOX
Tcells were more prevalent in non-adnexal as compared with adnexal
samples, with differences across compartments more pronounced in
HRD subtypes than in FBI cases (Fig.4g). Similarly, terminally dysfunc-
tional CD8
+
PD-1
+
TOX
+
Tcells were enriched within the peritumoural
stroma in HRD-Dup cases and in the tumour of HRD-Del cases. By con-
trast, CD8+PD-1+TOX and CD8+PD-1+TOX+ Tcells were less abundant in
FBI cases and were evenly distributed within the tumour and stroma,
implying reduced Tcell–antigen interactions (Fig.4g).
DC and macrophage phenotypic states were similarly shaped by
tumour mutational signature, with cDC2 and M2.SELENOP enrichment
Ov cancer cell
Mast cell
Monocyte
B cell
Dendritic cell
T cell
Fibroblast
Endothelial cell
Plasma cell
BAF in chr. arm 6p
0 0.25 0.5000.25 0.50
Cell type
Cycling.cancer.cell.2
Cycling.cancer.cell.1
Cancer.cell.6
Cancer.cell.5
Cancer.cell.4
Cancer.cell.3
Cancer.cell.2
Cancer.cell.1
BAF in chr. arm 6p
Cancer cell cluster
a
105
081
003
115
050
077
110
009
037
067
053
054
071
116
118
002
112
026
075
051
090
070
065
080
041
031
014
049
042
036
082
068
083
045
052
107
025
024
007
008
022
105
081
003
115
050
077
110
009
037
067
053
054
071
116
118
002
112
026
075
051
090
070
065
080
041
031
014
049
042
036
082
068
083
045
052
107
025
024
007
008
022
% cancer cells
per patient with 6p LOH
0205080100 0205080100
% cancer cells
per patient with 6p LOH
Undetermined
Mutational signature
HRD-Dup
HRD-Del
FBI
Site
Adnexa
Ascites
Bowel
Omentum
Peritoneum
UQ
Other
No. cells
500
2,500
5,000
bHet. Subclonal
LOH
Clonal
LOH
Het. Subclonal
LOH
Clonal
LOH
36/116
16/81
29/118
10/42
398/1,292
All
CDK12
CCNE1
BRCA2
BRCA1
All
FBI
HRD-Del
HRD-Dup
d
c
Het. Subclonal
LOH
Clonal
LOH
Het. Subclonal
LOH
Clonal
LOH
e
All
Ascites
UQ
Peritoneum
Omentum
Bowel
Adnexa
g
** *
CD4.T
naive.mem
CD8.T
naive.mem
CD4.T
dysfunctional
CD8.T
dysfunctional
Heterozygous
Clonal LOH
Heterozygous
Clonal LOH
Heterozygous
Clonal LOH
Heterozygous
Clonal LOH
0
10
20
30
3
6
9
10
20
30
20
40
60
% T cells
6p LOH
(cancer cells)
Heterozygous
Clonal LOH
f
Site
Bowel
Right adnexa
Left adnexa
BAF
00.250.5
0
00.250.5
0
Omentum
Ascites
Right adnexa
BAF
BAF
00.250.5
Chr. 6p BAF
HRD-Dup
022
FBI
065
6p LOH
n = 9,620
n = 7,382
6p LOH
0
010203040
20 50 80 100
% cancer cells
per patient with 6p LOH
0205080100
% cancer cells
per patient with 6p LOH
% patients with LOH
in any HLA class I gene
Fig. 5 | HLA l oss as a mech anism of immu ne escape . a, Left, distribution over
cells of chrom osome arm 6p BA F in scRNA-seq d ata with rank ing by median 6p
BAF per cell t ype. Rig ht, allelic imbala nce in 6p BAF acros s cancer cell clu sters.
White vert ical lines indi cate the media n. Chr., chromosome. b, Lef t, percent age
of cancer c ells with 6p LOH per p atient. Rig ht, site- and clon e-specif ic
percent age of cancer cel ls with 6p LOH. He t., heterozygo us. c, Percenta ge of
cancer ce lls with 6p LOH per s ample as a func tion of mutatio nal signature . Pie
charts s how the fracti on of samples wi th heterozygou s, subclonal LOH an d
clonal LOH 6p st atus. d, Percen tage of patient s with LOH of any HL A class I gene
in the MSK-IMPACT HGSO C cohort (n = 1,2 98 patients) for BRCA1-, BRCA2- and
CDK12-mutant and CCNE1-amplifie d tumours, mapp ing to HRD-D up, HRD-Del,
TD and FBI sig natures, res pectively. Error bar s, 95% binomial con fidence
intervals. e, Percentage of c ancer cells wi th 6p LOH per sampl e as a function o f
anatomic al site. Pie char ts show the fra ction of sampl es by 6p status . f, UMAP
plots of can cer cells from re presentat ive HRD-Dup an d FBI cases . Density plo ts
show site-sp ecific 6p BA F. g, Fraction of n aive and dysfunc tional Tcells in
CD45+ samp les as a funct ion of the 6p LOH clonal ity of cancer c ells in matched
CD45 samples.*P <0.05; bracket s indicate two -sided Wilcoxon pair wise
comparisons. In b,c,e and g, 6p LOH status isdef ined as follows:het erozygous,
percent age 6p LOH ≤ 20%; subclo nal LOH, 20% < perce ntage 6p LOH ≤ 80%;
clonal LOH, pe rcentage 6p LOH > 8 0%.In c, e and g, box plots and v iolin plots
showthe median, t opand bottomquart iles; whiskers cor respond to 1. 5× IQR.
In ae, only BAF es timates from ce lls with ≥10 reads al igning to 6p were
considere d and allelic imbala nce states we re assigned on t he basis of the me an
6p BAF per ce ll as follows: balance d, BAF ≥ 0.35; imba lanced, 0.15 ≤ BA F < 0.35;
LOH, BAF < 0.1 5 (Methods).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
784 | Nature | Vol 612 | 22/29 December 2022
Article
in HRD-Del cases and enrichment of M1 macrophages in FBI cases
(Extended Data Fig.11a,e). Phenotypic diversification of myeloid cells
was elevated in HRD-Dup adnexal samples with high entropy (Extended
Data Fig.11b, left); however, inter-patient Bray–Curtis dissimilarity
was statistically higher in FBI tumours, suggesting greater patient
specificity in FBI relative to HRD-Dup cases (Extended Data Fig.11b,
right). Diversification was characterized by M2.CXCL10 macrophage
enrichment in HRD-Dup and depletion in FBI (Supplementary Table5),
with FBI tumours also exhibiting fewer PD-L1 (CD274)-positive mac-
rophages (Extended Data Fig.11a,d). In line with CXCL10 being a target
of JAK–STAT signalling, macrophages in HRD-Dup, but not FBI, samples
presented higher JAK–STAT pathway activity (Extended Data Fig.11c).
Spatially, we observed elevated localization of CD68
+
macrophages in
the periphery of both HRD-Dup and FBI adnexal samples that extended
into the tumour for HRD-Dup, but not FBI, samples (Extended Data
Fig.11g).
Concordance of JAK–STAT pathway activation among all cell subtypes
implies a common upstream effector. We therefore examined type I IFN
pathway regulators in DCs, which commonly serve as a key activator of
JAK–STAT signalling. We observed a strong positive correlation between
the IFN regulator module score in DCs and JAK–STAT pathway activation
in cancer cells, Tcells and macrophages (Extended Data Fig.11h). Thus,
increased type I IFN activation in DCs in HRD-Dup tumours may serve
as an activator of JAK–STAT signalling, with downstream upregulation
of human leukocyte antigen (HLA) molecules and PD-L1 in cancer cells
and macrophages.
Mutational processes drive immunoediting
We next investigated whether increased immune signalling in HRD
subtypes led to mechanisms mediating immune escape. We profiled
loss of HLA presentation machinery
24
inferred through loss of heterozy-
gosity (LOH) of chromosome arm 6p—harbouring HLA class I and class
II genes—at the single-cell level using the SIGNALS algorithm5. Predic-
tions were restricted to cancer cells (Fig.5a), with per-cell B-allele frac-
tions (BAFs) classed as balanced, imbalanced or LOH (Extended Data
Fig.12a,b) and orthogonal genomic validation from site-matched WGS
and MSK-IMPACT datasets (Extended Data Fig.12c–f). We observed
marked inter-patient heterogeneity (Fig.5b,c and Extended Data
Fig.12a,b), with clonal 6p LOH in 4 of 41 patients (10%) and subclonal 6p
LOH in 7 of 41 patients (17%; Fig.5c, left). Intriguingly, site-specific losses
were found in 4 of 41 patients (Fig.5c, right). Clonal 6p LOH was primar-
ily observed in HRD-Dup cases, whereas subclonal distributions were
more frequent in patients with FBI tumours (Fig.5c). Higher prevalence
of 6p LOH in HRD-Dup was validated in an independent cohort (n = 1,298
patients) with available MSK-IMPACT sequencing (31% in BRCA1-mutant
cases, 19% in BRCA2-mutant cases and 24% in CCNE1-amplified cases
Fig.5d). Notably, clonal 6p LOH was present in adnexal lesions in 5 of
47 samples (Fig.5e), in line with ‘early’ immune evolutionary selection
in the primary site. Patient 022 with the HRD-Dup subtype and patient
065 with the FBI subtype further showed patient-specific evolution-
ary timing of 6p LOH (Fig.5f). Functional consequences of 6p LOH in
HRD-Dup were also observed, including upregulation of JAK–STAT
signalling (Extended Data Fig.12g,h), which was most pronounced
in bowel samples (Extended Data Fig.12i), and increased presence of
dysfunctional CD4+ and CD8+ Tcells (Fig.5g). Together, association
of LOH of HLA alleles with heightened JAK–STAT signalling and Tcell
dysfunction points to ‘early’ immune-mediated evolutionary selection
of 6p loss in HRD-Dup tumours, in contrast to evolutionarily ‘late’ clonal
expansion of 6p LOH in FBI tumours.
Spatial topology of the microenvironment
The single-cell analyses above link immunophenotypic variation to
mutational signatures and tumour site. We sought to validate these
findings with tumour–immune cell interactions and spatial topolo-
gies from insitu mpIF profiling of principal immune cell types (Tcells
and macrophages) and their functional markers (PD-1, TOX, PD-L1)
(Extended Data Fig.13a). We enumerated the proximal interactions
of naive/memory (CD8+PD-1TOX), activated/predysfunctional
(CD8+PD-1+TOX) and dysfunctional (CD8+PD-1+TOX+) Tcells with
PD-L1-expressing cancer cells (pan-cytokeratin (panCK)+PD-L1+).
a
PanCK
CD8
CD68
DAPI
PD-1
PD-L1
TOX
CD8
+
PD1
+
TOX
+
PanCK+PDL1-
PanCK+PDL1+
CD8+PD1-TOX-
CD8+PD1+TOX-
CD68+PDL1-
CD68+PDL1+
Other
Unknown
007-LAdnx
[49322,9587]
082-RAdnx
[48667,10554]
HRD-Dup
HRD-Del
FBI
065-ROv
[42891,12989]
CD8+PD1+TOX-
PanCK+PDL1+
NN dist.
CD8+PD1+TOX+
PanCK+PDL1+
NN dist.
Nearest-neighbourdistance from CD8+Tcellphenotypes
to cancer cell phenotypes
200 μm200 μm200 μm
200 μm200 μm200 μm
200 μm200 μm200 μm
Cell phenotypes
Fluorescence
200 μm
200 μm
200 μm
b
58 355
164
43 105
141 222110
35 242
175
2723 114
36 4735
28 16256
31 42 119
48 56 73
0
0.01
0.02
0
0.01
0.02
0
0.01
0.02
Radial distance to nearest PanCK+PD-L1+ cell, r (μm)
Density
BowelOmentum Adnexa
Mutational
signature
HRD-Dup
HRD-Del
FBI
050100 1500 50 100150 050100 150
CD8+PD-1TOXCD8+PD-1+TOXCD8+PD-1+TOX+
169
Fig. 6 | Spati al topologi es of insitucellul ar interac tions. a, Repres entative
mpIF fi elds of view (FOVs) highlight ing common featu res of the TME and
showing one a dnexal sample p er mutational s ignature. Firs t column, raw
pseudocolour images;second column, cellular phenotypes of segmented cells;
third and fourt h columns, proximi ty of phenoty pe pairs, high lighting PD- L1–
PD-1 interactions with colour-coded phenotypes and edges depicting
nearest-neig hbour distan ces. Only edg es joining pairs o f cells within 2 50 μm
of each othe r are shown. b, Nearest-neighbour distance from CD8+ Tcell
phenoty pes to panCK+PD-L1+ canc er cells aggre gated acros s FOVs, with
samples g rouped by anatom ical site and mut ational signa ture. Vertical l ines
indicate the median nearest-neighbour distance.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Nature | Vol 612 | 22/29 December 2022 | 785
Interactions between proximal PD-L1-expressing cancer cells and
activated/predysfunctional Tcells were particularly high in bowel
samples, and dysfunctional Tcell interactions were high in both bowel
and adnexal samples (Extended Data Fig.13b,c). Omentum samples,
by contrast, exhibited relatively few proximal interactions of either
Tcell or macrophage phenotypes with PD-L1-expressing cancer cells
(Extended Data Fig.13d,e).
Mutational signatures also impacted cellular interactions, as pre-
dicted by higher receptor–ligand co-expression of PD-L1 (CD274) in
myeloid clusters and PD-1 (PDCD1) in T and NK cell clusters in HRD
subtypes derived from scRNA-seq data (Extended Data Fig.13f). In line
with these findings, the spatial organization of cellular neighbourhoods
also varied by mutational subtype, as reflected in nearest-neighbour
distances between Tcells and panCK+PD-L1+ cancer cells (interactions of
exemplar samples are shown in Fig.6a). Antigen-experienced CD8+PD-1+
Tcells within a 30-μm radius of PD-L1
+
cancer cells were common in
HRD-Dup cases but rare or absent in FBI tumours (Extended Data
Fig.14a). When combining site and signature, the shortest median
distances were observed in HRD-Dup adnexa and bowel samples, par-
ticularly in the activated/predysfunctional and dysfunctional Tcell
compartments (Fig.6b and Extended Data Fig.14b), supporting PD-L1
as a negative feedback mechanism in response to activated Tcells in
HRD tumours. Similar interactions were noted between Tcells and
CD68+PD-L1+ macrophages, which were particularly prevalent in
HRD-Del cases (Extended Data Fig.14c–f), but largely absent in FBI
tumours. Overall, mpIF analysis further highlighted site- and muta-
tional signature-dependent TMEs consistent with scRNA-seq-based
observations.
Discussion
Our results synthesize anatomical sites and mutational processes as
determinants of HGSOC TMEs and their phenotypic states. We specu-
late that, while the relative paucity of immune cells in adnexal sites is
driven by the immune privilege of the ovaries and fallopian tubes, the
predominance of dysfunctional Tcells at these sites reflects immuno-
reactivity early in cancer evolution with subsequent immune escape in
metastatic sites. In addition, contrasting cell-to-cell topological fea-
tures of omental and bowel samples indicates that specific metastatic
sites may harbour tissue-specific immunosurveillance constraints.
Moreover, high intra- and inter-TME heterogeneity highlights that
mechanisms of immune resistance are not universal in a given patient,
requiring any therapeutic approach to account for evolution of the
immune response in individual tumours.
Moreover, how mutational processes engender distinct immune
evasion mechanisms raises additional questions in view of preclinical
studies suggesting that immunogenicity in HRD tumours may lead
to improved responses to immune checkpoint blockade (ICB)2528.
Clinical evidence for this is lacking as no association between HRD
status, tumour mutational burden and response to ICB alone or in
combination with chemotherapy has been observed
11–13
. Our findings
highlight that mechanisms of immune resistance are distinct among
the mutational subtypes, including activated type I IFN signalling in
Tcells, cancer cells and myeloid cells that is particularly enriched in
HRD-Dup tumours. These data argue for multifaceted strategies for
immunotherapeutic reprogramming that consider the underlying
mutational process, in particular for FBI tumours, which are more
resistant to chemotherapy
5
and here are found to be immunologically
inert.
Altogether, our study provides an extensive multi-modal resource,
mapping the cellular constituents of HGSOC TMEs and linking them to
mutational processes and spatial context. Our findings illustrate that
even personalized approaches may be ineffective against widespread
and heterogenous disease within patients, highlighting the urgent need
for early detection before dissemination into the peritoneal cavity.
The data presented here can be leveraged broadly to contextualize
mechanistic insights into immunotherapeutic response across cancers
of genomic instability.
Online content
Any methods, additional references, Nature Portfolio reporting summa-
ries, source data, extended data, supplementary information, acknowl-
edgements, peer review information; details of author contributions
and competing interests; and statements of data and code availability
are available at https://doi.org/10.1038/s41586-022-05496-1.
1. Patch, A.-M. etal. Whole-genome characterization of chemoresistant ovarian cancer.
Nature 521, 489–494 (2015).
2. Li, Y. etal. Patterns of somatic structural variation in human cancer genomes. Nature 578,
112–121 (2020).
3. Drews, R. M. etal. A pan-cancer compendium of chromosomal instability. Nature 606,
976–983 (2022).
4. Wang, Y. K. etal. Genomic consequences of aberrant DNA repair mechanisms stratify
ovarian cancer histotypes. Nat. Genet. 49, 856–865 (2017).
5. Funnell, T. etal. Single-cell genomic variation induced by mutational processes in cancer.
Nature https://doi.org/10.1038/s41586-022-05249-0 (2022).
6. Funnell, T. etal. Integrated structural variation and point mutation signatures in
cancer genomes using correlated topic models. PLoS Comput. Biol. 15, e1006799
(2019).
7. McPherson, A. etal. Divergent modes of clonal spread and intraperitoneal mixing in
high-grade serous ovarian cancer. Nat. Genet. 48, 758–767 (2016).
8. Schwarz, R. F. etal. Spatial and temporal heterogeneity in high-grade serous ovarian
cancer: a phylogenetic analysis. PLoS Med. 12, e1001789 (2015).
9. Izar, B. etal. A single-cell landscape of high-grade serous ovarian cancer. Nat. Med. 26,
1271–1279 (2020).
10. Zhang, A. W. etal. Interfaces of malignant and immunologic clonal dynamics in ovarian
cancer. Cell 173, 1755–1769 (2018).
11. Liu, Y. L. etal. Mutations, homologous DNA repair deiciency, tumor mutational burden,
and response to immune checkpoint inhibition in recurrent ovarian cancer. JCO Precis.
Oncol. 4, PO.20.00069 (2020).
12. Ledermann, J. A. etal. Avelumab in combination with and/or following chemotherapy
vs chemotherapy alone in patients with previously untreated epithelial ovarian
cancer: results from the phase 3 Javelin Ovarian 100 trial. Gynecol. Oncol. 159, 13–14
(2020).
13. Moore, K. N. etal. Atezolizumab, bevacizumab, and chemotherapy for newly diagnosed
stage III or IV ovarian cancer: placebo-controlled randomized phase III trial (IMagyn050/
GOG 3015/ENGOT-OV39). J. Clin. Oncol. 39, 1842–1855 (2021).
14. The Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian
carcinoma. Nature 474, 609–615 (2011).
15. Konstantinopoulos, P. A., Ceccaldi, R., Shapiro, G. I. & D’Andrea, A. D. Homologous
recombination deiciency: exploiting the fundamental vulnerability of ovarian cancer.
Cancer Discov. 5, 1137–1154 (2015).
16. Macintyre, G. etal. Copy number signatures and mutational processes in ovarian
carcinoma. Nat. Genet. 50, 1262–1270 (2018).
17. Gerstung, M. etal. The evolutionary history of 2,658 cancers. Nature 578, 122–128
(2020).
18. Davies, H. etal. HRDetect is a predictor of BRCA1 and BRCA2 deiciency based on
mutational signatures. Nat. Med. 23, 517–525 (2017).
19. Nguyen, L., Martens, J. W. M., Van Hoeck, A. & Cuppen, E. Pan-cancer landscape of
homologous recombination deiciency. Nat. Commun. 11, 5584 (2020).
20. Hornburg, M. etal. Single-cell dissection of cellular components and interactions
shaping the tumor immune phenotypes in ovarian cancer. Cancer Cell 39, 928–944
(2021).
21. Zheng, L. etal. Pan-cancer single-cell landscape of tumor-iniltrating Tcells. Science
374, abe6474 (2021).
22. Gebhardt, C., Németh, J., Angel, P. & Hess, J. S100A8 and S100A9 in inlammation and
cancer. Biochem. Pharmacol. 72, 1622–1631 (2006).
23. Bohlson, S. S., O’Conner, S. D., Hulsebus, H. J., Ho, M.-M. & Fraser, D. A. Complement, C1q,
and C1q-related molecules regulate macrophage polarization. Front. Immunol. 5, 402
(2014).
24. McGranahan, N. etal. Allele-speciic HLA loss and immune escape in lung cancer
evolution. Cell 171, 1259–1271 (2017).
25. Iyer, S. etal. Genetically deined syngeneic mouse models of ovarian cancer as tools for
the discovery of combination immunotherapy. Cancer Discov. https://doi.org/10.1158/
2159-8290.CD-20-0818 (2020).
26. Zhang, S. etal. Genetically deined, syngeneic organoid platform for developing
combination therapies for ovarian cancer. Cancer Discov. https://doi.org/10.1158/2159-
8290.CD-20-0455 (2020).
27. Ding, L. etal. PARP inhibition elicits STING-dependent antitumor immunity in Brca1-deicien t
ovarian cancer. Cell Rep. 25, 2972–2980 (2018).
28. Samstein, R. M. etal. Mutations in BRCA1 and BRCA2 differentially affect the tumor
microenvironment and response to checkpoint blockade immunotherapy. Nat. Cancer 1,
1188–1203 (2021).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional afiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
786 | Nature | Vol 612 | 22/29 December 2022
Article
Open Access This article is licensed under a Creative Commons Attribution
4.0 International License, which permits use, sharing, adaptation, distribution
and reproduction in any medium or format, as long as you give appropriate
credit to the original author(s) and the source, provide a link to the Creative Commons licence,
and indicate if changes were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your
intended use is not permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a copy of this licence,
visit http://creativecommons.org/licenses/by/4.0/.
© The Author(s) 2022
1Computational Oncology, Department of Epidemiology and Biostatistics, Memorial Sloan
Kettering Cancer Center, New York, NY, USA. 2Department of Surgery, Memorial Sloan
Kettering Cancer Center, New York, NY, USA. 3Integrated Genomics Operation, Memorial
Sloan Kettering Cancer Center, New York, NY, USA. 4Department of Pathology and
Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
5Marie-Josée and Henry R. Kravis Center for Molecular Oncology, Memorial Sloan Kettering
Cancer Center, New York, NY, USA. 6Department of Medicine, Memorial Sloan Kettering
Cancer Center, New York, NY, USA. 7Weill Cornell Medical Center, New York, NY, USA.
8Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
9Department of Information Systems, Memorial Sloan Kettering Cancer Center, New York,
NY, USA . 10Immunology Program, Memorial Sloan Kettering Cancer Center, New York, NY,
USA. 11Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer
Center, New York, NY, USA . 12Department of Radiation Oncology, Memorial Sloan Kettering
Cancer Center, New York, NY, USA. 13Present address: Bristol Myers Squibb, Princeton, NJ,
USA. 14These authors contributed equally: Ignacio Vázquez-García, Florian Uhlitz. e-mail:
zamarind@mskcc.org; shahs3@mskcc.org
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Methods
Experimental methods
Sample collection. All enrolled patients were consented to an insti-
tutional biospecimen banking protocol and MSK-IMPACT
29
testing,
and all analyses were performed according to a biospecimen research
protocol. All protocols were approved by the institutional review board
(IRB) of Memorial Sloan Kettering Cancer Center. Patients were con-
sented following the IRB-approved standard operating procedures for
informed consent. Written informed consent was obtained from all
patients before conducting any study-related procedures. The study
was conducted in accordance with the Declaration of Helsinki and the
Good Clinical Practice guidelines.
We collected fresh tumour tissues from 42 patients with HGSOC at the
time of upfront diagnostic laparoscopic or debulking surgery. Ascites
and tumour tissue fromprimary and multiple metastatic sites, includ-
ing bilateral adnexa, omentum, pelvic peritoneum, bilateral upper
quadrants and bowel, were procured in a predetermined, systematic
fashion (median of four primary and metastatic tissues per patient)
and were placed in cold RPMI for immediate processing. Blood samples
were collected before surgery for the isolation of peripheral blood
mononucleated cells (PBMCs) for normal WGS. The isolated cells were
frozen and stored at −80 °C. In addition, tissue was snap frozen for bulk
DNA extraction and tumour WGS. Tissue was also subjected to formalin
fixation and paraffin embedding (FFPE) for histological, immunohisto
-
chemical and multiplex immunophenotypic characterization.
Sample processing. We profiled patient samples using five different
experimental assays:
1. CD45+ and CD45 flow-sorted cells were collected from fresh tissue
samples and processed for scRNA-seq in 156 sites from 41 patients
(~6,000 cells per site; Supplementary Table2).
2. For each specimen with scRNA-seq data, site-matchedFFPE tis-
suesections were used for whole-slide H&E staining and computa-
tional analysis (n = 100 tissue samples from 35 patients).
3. For each specimen with scRNA-seq data, site-matched FFPE tissue
sections adjacent to the H&E section were stained by mpIF for major
cell type and immunoregulatory markers (n = 1,349 quality-filtered
FOVs across 100 tissue samples from 35 patients).
4. US Food and Drug Administration-approved clinical sequencing of
468 cancer genes (MSK-IMPACT) was performed on DNA extracted
from FFPE tumour and matched normal blood specimens for each
patient (Extended Data Fig.1b).
5. Snap-frozen tissues were processed to obtain matched tumour–nor-
mal WGS data for a single representative site in n = 40 patients with
scRNA-seq, H&E and mpIF data, to derive mutational processes from
genome-wide single-nucleotide and structural variants.
scRNA-seq. Tissue dissociation. Tumour tissue was immediately pro-
cessed for tissue dissociation. Fresh tissue was cut into 1-mm pieces and
dissociated at 37 °C using the Human Tumor Dissociation kit (Miltenyi
Biotec) on a gentleMACS Octo Dissociator. After dissociation, single-cell
suspensions were filtered and washed with ammonium-chloride-
potassium (ACK) lysing buffer. Cells were stained with trypan blue, and
cell counts and viability were assessed using the Countess II Automated
Cell Counter (ThermoFisher) (for a detailed protocol, see ref. 30).
Cell sorting. Freshly dissociated cells were stained with a mixture
of GhostRed780 live/dead marker (TonBo Biosciences) and Hu-
man TruStain FcX Fc Receptor Blocking Solution (BioLegend). The
stained samples were then incubated and stained with Alexa Fluor
700 anti-human CD45 antibody (BioLegend). After staining, cells
were washed and resuspended in RPMI + 2% FCS and submitted for
cell sorting. The cells were sorted into CD45+ and CD45 fractions by
fluorescence-assisted cell sorting on a BD FACSAria III flow cytometer
(BD Biosciences). Positive and negative controls were prepared and
used to set up compensations on the flow cytometer. Cells were sorted
into tubes containing RPMI + 2% FCS for sequencing.
Library preparation. Flow-sorted tumour cells were stained with
trypan blue, and the Countess II Automated Cell Counter (Ther-
moFisher) was used to assess both cell number and viability. Following
quality control, the single-cell suspension was loaded onto a Chromium
Chip B (10x Genomics, PN 2000060). GEM generation, cDNA synthesis,
cDNA amplification and library preparation for 1,400–5,000 cells pro-
ceeded using Chromium Single-Cell 3′ Reagent kit v3 (10x Genomics,
PN 1000075) according to the manufacturer’s protocol. cDNA ampli-
fication included 12 cycles, and 0.4–419 ng of the material was used to
prepare sequencing libraries with 8–14 cycles of PCR.
Sequencing. Equimolar amounts of indexed libraries were pooled and
sequenced on a HiSeq 2500 in rapid mode or on a NovaSeq 6000 in a
28-bp/91-bp, 100-bp/100-bp or 150-bp/150-bp paired-end run using
HiSeq Rapid SBS kit v2 or NovaSeq 6000 SP, S1, S2 or S4 Reagent kit
(100, 200 or 300 cycles) (Illumina).
Bulk WGS. Bulk tumour WGS. Frozen banked tissue was cut into sec-
tions on charged microscope slides. Following histological review,
tumour tissue was microdissected if required to enrich for neoplastic
cells
31
and subjected to DNA extraction for bulk WGS. Genomic DNA was
extracted using DNeasy Blood & Tissue kits (Qiagen) and quantified on a
Qubit 3 Fluorometer using the Qubit 1× dsDNA HS Assay kit (Invitrogen).
Bulk normal WGS. PBMCs were brought up to a volume of 1 5 ml in cold
PBS, and DNA was isolated with the DNeasy Blood & Tissue kit (Qiagen,
69504) according to the manufacturer’s protocol with 1 h of incubation
at 55 °C for digestion. DNA was eluted in 0.5× buffer AE.
Sequencing. DNA quantity was measured using the Quant-iT PicoGreen
dsDNA assay (ThermoFisher, P11496), and DNA quality was assessed
with TapeStation D1000 ScreenTape (Agilent, 5067-5582). After Pico-
Green quantification and quality control with an Agilent BioAnalyzer,
500 ng of genomic DNA was sheared using an LE220-plus focused
ultrasonicator (Covaris, 500569) and sequencing libraries were pre-
pared using the KAPA Hyper Prep kit (Kapa Biosystems, KK8504) with
modifications. In brief, libraries were subjected to a 0.5× size selection
using AMPure XP beads (Beckman Coulter, A63882) after postligation
clean-up. Libraries were not amplified by PCR and were pooled in an
equal volume and quantified on the basis of their initial sequencing
performance. Samples were run on a NovaSeq 6000 in a 150-bp/150-bp
paired-end run, using the NovaSeq 6000 SBS v1 kit and an S1, S2 or S4
flow cell (Illumina).
Preparation, review and scanning of histopathology slides. Archived
FFPE tissues were used for histological review, including the assessment
of spatial topology and tumour-infiltrating lymphocytes (TILs), as well
as for immunohistochemical characterization and mpIF analysis for
mapping of the TME, in the Advanced Immunomorphology Platforms
Laboratory. Slides were originally reviewed by gynaecological patholo-
gists for diagnosis and FIGO (International Federation of Gynecology
and Obstetrics) stage assignment. Representative H&E-stained slides
from each site of interest were digitally scanned to produce virtual
slides. Two senior gynaecological pathologists then reviewed these
images for the presence and location of serous tubal intraepithelial
carcinoma (STIC), SET architecture (solid, pseudo-endometrioid and
transitional cell-like patterns), micropapillary architecture32, pres-
ence of a fimbrial ball, architectural patterns of metastatic disease33,
mitotic counts (per ten high-power fields, HPFs) and tumour cell con-
tent (viable percentage). Regions with TILs were also assessed with a
quantitative TIL score (low, <42 TILs per HPF in a hotspot; high, 42 or
more TILs per HPF in a hotspot)
32
. Histopathology slides were scanned
into whole-slide images using a Leica Aperio AT2 scanner (Leica Biosys-
tems) at ×20 magnification. The most representative tissue block was
selected for slide scanning.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Article
mpIF. Overview. We carried out multiparameter quantification of
epithelial and immune cell subsets and activation markers using the
AkoyaBio Vectra automated imaging system at the MSKCC Parker Insti-
tute for Cancer Immunotherapy. We stained whole slides of FFPE tissue
for markers of ovarian cancer cells (panCK + CK8–CK18) and of specific
leukocyte subsets, including macrophages (CD68) and cytotoxic Tcells
(CD8), known immune inhibitory proteins (PD-L1) and markers of the
activation/exhaustion status of CD8
+
Tcells (PD-1, TOX). FOVs were
chosen to include either the entire tissue with minimal field overlap if
the tissue was small or a distribution of fields with 50% stroma/tumour
at the edge plus some central areas of tumour-dense fields. Quality con-
trol was performed on marker intensities so that they fell in the range
of 5–30 arbitrary units and helped guide spectral unmixing. Lower
values might be close to background, while higher values prompted
us to check for channel spillage.
Tissue staining. Primary antibody staining conditions were optimized
using standard immunohistochemical staining on the Leica Bond RX
automated research stainer with DAB detection (Leica Bond Polymer
Refine Detection, DS9800). Using 4-μm FFPE tissue sections and serial
antibody titrations, the optimal antibody concentration was deter-
mined followed by transition to a seven-colour multiplex assay with
equivalency. Optimal primary antibody stripping conditions between
rounds in the seven-colour assay were performed following a cycle of
tyramide deposition followed by heat-induced stripping (see below)
and subsequent chromogenic development (Leica Bond Polymer
Regine Detection, DS9800) with visual inspection for chromogenic
product with a light microscope by a senior pathologist. Multiplex
assay antibodies and conditions are described in Supplementary
Table6.
Tissue sections were baked for 3 h at 62 °C in vertical slide orienta-
tion with subsequent deparaffinization performed on the Leica Bond
RX followed by 30 min of antigen retrieval with Leica Bond ER2 and
six sequential cycles of staining with each round including a 30-min
combined block and primary antibody incubation (Akoya Antibody
Diluent/Block, ARD1001).
For panCK and CK8–CK18, detection was performed using a sec-
ondary horseradish peroxidase (HRP)-conjugated polymer (Akoya
Opal Polymer HRP Ms + Rb, ARH1001; 10-min incubation). Detection
of all other primary antibodies was performed using a goat anti-mouse
Poly HRP secondary antibody or goat anti-rabbit Poly HRP secondary
antibody (Invitrogen, B40961 and B40962; 10-min incubation). The
HRP-conjugated secondary antibody polymer was detected by fluo
-
rescent tyramide signal amplification using Opal dyes 520, 540, 570,
620, 650 and 690 (Akoya, FP1487001KT, FP1494001KT, FP1488001KT,
FP1495001KT, FP1496001KT, FP1497001KT). The covalent tyramide
reaction was followed by heat-induced stripping of the primary anti-
body–secondary antibody complex using PerkinElmer AR9 buffer
(AR900250ML) and Leica Bond ER2 (90% ER2 and 10% AR9) at 100 °C
for 20 min before the next cycle (one cycle of stripping for CD68, PD-1,
PD-L1, CD8 and panCK/CK8/CK18 and two cycles of stripping for TOX).
After six sequential rounds of staining, sections were stained with Hoe-
chst (Invitrogen, 33342) to visualize nuclei and mounted with ProLong
Gold antifade reagent mounting medium (Invitrogen, P36930).
Imaging and spectral unmixing. Seven-colour multiplex-stained
slides were imaged using Vectra Multispectral Imaging System version
3 (PerkinElmer). Scanning was performed at ×20 magnification (×200
final magnification). Filter cubes used for multispectral imaging were
DAPI, FITC, Cy3, Texas Red and Cy5. A spectral library containing the
emitted spectral peaks of the fluorophores in this study was created
using Vectra image analysis software (PerkinElmer). Using multispectral
images from slides singly stained for each marker, the spectral library
was used to separate each multispectral cube into individual compo-
nents (spectral unmixing), allowing for identification of the seven
marker channels of interest, using InForm 2.4 image analysis software.
Computational methods
scRNA-seq. Overview. The pipeline was built using the 10x Genomics
Martian language and computational pipeline framework. CellRanger
software (version 3.1.0) was used to perform read alignment, barcode
filtering and unique molecular identifier (UMI) quantification using the
10x GRCh38 transcriptome (version 3.0.0) for FASTQ inputs.
Quality control. CellRanger-filtered matrices were loaded into indi-
vidual Seurat objects using the Seurat R package (version 3.0.1)
34,35
.
The resulting gene-by-cell matrix was normalized and scaled for each
sample. Cells retained for analysis had a minimum of 500 expressed
genes and 1,000 UMI counts and had less than 25% mitochondrial gene
expression. Cell cycle phase was assigned using the Seurat CellCycle-
Scoring function. Scrublet (version 0.2.1) was used to calculate and
filter cells with a doublet score greater than 0.25. Sample matrices
were merged by patient and subsequently renormalized and scaled
using default Seurat functions.
Major cell type identification. Major cell type assignments were com
-
puted for each patient with CellAssign (version 0.99.2)36 using a set of
curated marker genes. Marker genes were compiled for nine major cell
types related to HGSOC (Supplementary Table4). These major cell
types were defined as Tcells, B cells, plasma cells, myeloid cells, DCs,
mast cells, endothelial cells, fibroblasts and ovarian cancer cells. Before
running CellAssign, cells with zero expression for all marker genes were
removed from the count matrix. Cell-specific size factors were com-
puted using scran (version 3.11). Default CellAssign parameters were
used with a design matrix of patient batch labels. CellAssign returned
a probability distribution over the major cell types, and individual cells
were labelled by the resulting most probable cell type.
Dimensionality reduction. Principal-component analysis (PCA) was
performed on the filtered feature-by-barcode matrix. UMAP embed-
dings including cohort-level and patient-level embeddings for all major
cell types were based on the first 50 principal components. UMAP
embeddings of major cell type supersets (see below) were based on
the 50 batch-corrected harmony components. Diffusion map embed-
dings and pseudotime estimates were computed using the R package
destiny (v3.0.1) for the subset of CD8+ Tcells37.
Batch correction and integration. Major cell types identified across
samples were split into six supersets: (1) Tcells; (2) B cells and plasma
cells; (3) myeloid cells, DCs and mast cells; (4) fibroblasts; (5) endothe-
lial cells; and (6) ovarian cancer cells. For each superset, the R package
harmony (version 0.1) was used for batch correction to account for
patient-specific effects38.
Clustering and cell subtype identification. Graph-based cluster-
ing was performed for each superset using the Louvain algorithm
implemented in Seurat (version 3.0.1) at three different resolutions
(0.1, 0.2 and 0.3). Differential expression between identified clusters
was computed using a two-sided Wilcoxon rank-sum test as imple-
mented in Seurat FindMarkers. Final results were filtered on log(fold
change) > 0.25 and Benjamini–Hochberg-adjusted P < 0.05. Clusters
were annotated on the basis of marker genes identified in differential
gene expression analysis. Patient-specific clusters not represented
across the full cohort were identified using relative entropy. Relative
entropy per cluster was defined as the maximum entropy per cluster
divided by the empirical entropy of patient compositions. Clusters with
a relative entropy of <0.8 were considered patient-specific clusters and
disregarded for downstream analyses.
For Tcell clusters, Tcells and NK cells were clustered in two steps.
Initial coarse-grained clustering resulted in ten different T and NK cell
clusters, including four CD4
+
Tcell clusters, three CD8
+
Tcell clusters,
two NK cell clusters and one cycling T/NK cell cluster (Extended Data
Fig.5a). Subclustering identified a total of 41 distinct fine-grained clus-
ters, broadly defining major Tcell and NK cell subtypes (Fig.2a and
Extended Data Fig.5b). These included populations of CD4
+
naive and
central memory cells (expressing IL7R and TCF7), CD4+ effector memor y
Content courtesy of Springer Nature, terms of use apply. Rights reserved
cells (IL7R, CCL5 and KLRB1), early and late dysfunctional CD4+ Tcells
(expressing dysfunctional Tcell markers CXCL13, TOX2 and PDCD1), reg-
ulatory Tcells (FOXP3 and IL2RA) and type 17 helper Tcells (KLRB1, RORA
and RORC). In the CD8+ compartment, we also identified populations
of naive/central memory (expressing KLF2, KLF3 and TCF7), activated/
cytotoxic (GZMH, GZMK and HLA-DR) and early and late dysfunctional
(CXCL13, TOX2, LAG3, HAVCR2, TIGIT and PDCD1) Tcells. Notably, the
early dysfunctional cluster, in addition to exhaustion-associated genes,
was characterized by expression of CXCR6 and ITGAE, commonly used
to define tissue-resident memory Tcells. In the innate compartment,
we similarly identified several clusters, including a γδ Tcell cluster
and several NK cell clusters. Finally, in all compartments, we identi-
fied populations of cells marked by expression of type I IFN response
genes such as ISG15 and IFIT3, herein named CD4-ISG, CD8-ISG and
NK-ISG, with strong upregulation of the JAK–STAT signalling pathway
as the dominant feature of these cells (Fig.2b). The remaining clusters
consisted of cycling T and NK cells expressing S phase markers such
as MKI67 and G2M markers such as TOP2A (Supplementary Table4).
For myeloid cell clusters, cDCs of the myeloid lineage were separated
into cDC1s, cDC2s and mDCs, marked by expression of CLEC9A, S100B
and BIRC3, respectively (Extended Data Figs.5d and6a). pDCs were
marked by expression of PTGDS. Macrophage clusters were described
with respect to their classical (M1-like) or alternative (M2-like) polari-
zation. Six different clusters encompassing both classical and alter-
natively activated macrophages were identified, as well as a cluster
of cycling macrophages (Cycling.M) and a cluster of actively phago-
cytic macrophages (Clearing.M). The M1-like and M2-like clusters were
labelled according to the top genes defining the clusters (M1.S100A8,
M2.CXCL10, M2.SELENOP, M2.MARCO, M2.COL1A1, M2.MMP9)
(Extended Data Figs.5d and6b). Among these, the M1.S100A8 cluster
was the only unambiguous M1-type macrophage cluster, marked by
expression of pro-inflammatory calcium-binding protein genes S100A8
and S100A9
22
. The M2.CXCL10 cluster was characterized by expres-
sion of both M1 (for example, CXCL10) and M2 (for example, PDL1 and
C1QC) markers. CXCL10 is an established downstream target of type
I and type II IFN signalling and was found to be expressed along with
other CXC-motif chemokines (CXCL9 and CXCL11). The remaining M2
clusters were all marked by high expression of complement component
C1QC, which is known to promote M2 polarization23.
InferCNV copy number clonal decomposition. InferCNV (version
1.3.5)39,40 was used to identify large-scale copy number alterations
in ovarian cancer cells classified by CellAssign. To do this, 3,200
non-cancer cells were randomly sampled from the cohort and used
as the set of reference ‘normal’ cells. After subtracting out reference
expression in non-cancer cells, chromosome-level smoothing and
denoising with InferCNV, we derived a processed expression matrix
that represented copy number signals. Cancer cell subclusters were
identified by ward.D2 hierarchical clustering and the ‘random_trees’
partition method using P < 0.05.
Gene signature scores. Cell state scores were calculated for the
exhausted phenotype within the set of Tcells using a manually curated
list of genes as input to the Seurat AddModuleScore method
40
. The
curated list of genes was derived from a review of single-cell analyses
of CD8+ Tcell states in human cancers41 (Supplementary Table4).
Patient specificity. Patient specificity scores were computed by using
a shared nearest-neighbour graph. For a given cell, patient specificity
was defined as the observed fraction of nearest neighbours divided by
the expected fraction of nearest neighbours in the patient subgraph.
Here the expected fraction of neighbours from the same patient was
defined as the global fraction of cells for each patient. Scores were log
2
transformed. Hence, a positive patient specificity score indicates an
over-representation of cells derived from the same patient among its
nearest neighbours, a negative score indicates an under-representation
of cells from the same patient and a score of 0 reflects a perfectly mixed
neighbourhood of patient labels.
Intra- and inter-patient variation in cluster composition. To calcu-
late intra-sample diversity of cluster composition, we used the Shannon
entropy H:
Hpp=− log
c
C
cc
=1
where pc is the proportional abundance of cluster cand C is the total
number of clusters.
To estimate the similarity or dissimilarity between samples, we used
the Bray–Curtis dissimilarity index D for samples i and j, defined as
DNN
NN
=12∑ min( ,)
∑+
c
C
c
ic
j
c
C
c
i
c
C
c
j
=1
=1 =1
where
Nc
i
and
Nc
j
are the counts for cluster c in samples i and j, respec-
tively, and C is the total number of clusters. This measure D takes val-
ues between 0 (identical samples:
NN
=
c
ic
j
for all j) and 1 (disjoint
samples:
N>0
c
i
implies
N=0
c
j
). We only considered the triangular
distance matrix D such that i < j. The pairwise distance matrix was
estimated by randomly subsampling the dataset with a minimum num-
ber of cells per sample and averaging over the subsampled datasets
after 100 iterations. We then evaluated intra- and inter-patient dis-
similarity on the basis of the distributions of the off-diagonal elements
in the averaged distance matrix (for example, all pairs of adnexal sam-
ples or all pairs of HRD-Dup samples).
These definitions were used to estimate the intra-sample diversity,
intra-patient dissimilarity and inter-patient dissimilarity of cluster
composition of cell states within each major cell type superset (cancer
cells, Fig.3g; T and NK cells, Figs.2d and 4f; myeloid cells, Extended
Data Figs.6d and 11b). Rarefaction of samples was applied in estimation
of the Bray–Curtis dissimilarity matrix on the basis of the number of
cells for each subset (n = 400 cells per sample).
Finally, we also used non-metric multidimensional scaling (NMDS)
to visualize the pairwise distances of cell type abundances in
low-dimensional space. We used the pairwise dissimilarity matrix D
to calculate the rank order of the Bray–Curtis distance and project
differences in cluster composition in two dimensions using NMDS (can-
cer cells, Extended Data Fig.8i; T and NK cells, Extended Data Figs.7c
and10c; myeloid cells, Extended Data Figs.7h and11f ).
GLMs of cluster composition. To estimate the effect of mutational
signatures and tumour site specificity on the composition of cell clus-
ters, we considered a GLM where we included interactions between
signature, site and cluster identity for each major cell type defined in
the scRNA-seq, H&E and mpIF data. The data matrix included the counts
of every cluster c, sampled from site s in a patient with mutational signa-
ture subtype m. Using a binomial linear model, one can analyse counts
of repeated observations of cell types or cell states as binary choices:
NpN~Bin(,)
c
c
where N
c
is the cellcount for cluster c in a sample, N is the total number
of cells in the sample and the probability to detect the cluster can be
described by the logit function βX
l
og
=.
p
p1−
c
c
To account for the effect of mutational signature and anatomical
tumour site on the cluster abundance observed in scRNA-seq data, we
formulated a GLM of the observed cell counts Nc for a cell type or cell
state described by the logit function, which is distributed as
p
pββxx βxlog1− Normal(+ ++++
,)
c
c
ccmmsscm cm cs cs
0ε
2
where β
0
is a shared constant baseline per cluster that must be inferred;
βc, βm and βs are individual fixed-effect terms to be inferred; βcm and
βcs are cluster–signature and cluster–site interaction effects to be
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Article
inferred; x
c
, x
m
and x
s
are elements of the model design matrix X; and
σ
ε
represents measurement noise. We note that for each cluster c we
had multiple measurement replicates of N
c
across signatures and sites.
This formulation was used to fit a GLM of major cell types (Fig.1f). We
also used this formulation to separately fit GLMs of cluster composition
for each superset of coarse-grained immune cell types (T and NK cells,
Extended Data Fig.7b; myeloid cells, Extended Data Figs.7g and 11e)
and GLMs of cluster composition for fine-grained immune cell states
(T and NK cells, Fig.2c; DCs, Extended Data Fig.11a; macrophages,
Extended Data Fig.11a).
To model the abundance of major cell types in the scRNA-seq data
from CD45+ and CD45 samples, the GLM included a covariate for
CD45
+/−
flow sorting with additional fixed-effect sorting coefficients β
f
and additional cluster sorting interactions β
cf
to be inferred, plus an
additional element x
f
in the model design matrix (Fig.1f). Similarly,
GLMs for H&E and mpIF data accounted for differences in cell type
abundance observed in the tumour and stroma regions, incorporat-
ing a covariate for the tumour or stroma region counts with additional
fixed-effect region coefficients β
r
and additional cluster–region coef-
ficients βcr to be inferred, plus an additional element xr in the model
design matrix (Fig.1f).
To quantify interactions between mutational signature and anatomi-
cal tumour site, we also fitted GLMs with an additional interaction term:
p
pββxx
βxxx
log1− Normal(+ +++
++ ,)
c
cccmmsscm cm
cs
cs
csm
csm
0
ε
2
where βcsm terms were cluster-specific signature–site interaction
effects to be inferred. This formulation was used to fit GLMs of cluster
composition of cell states within each major cell type superset, both
for fine-grained clusters (cancer cells, Fig.3d; T and NK cells, Fig.4b;
DCs, Extended Data Fig.11a; macrophages, Extended Data Fig.11a)
and coarse-grained clusters (T and NK cells, Extended Data Fig.10e;
myeloid cells, Extended Data Fig.11e).
PD-1 and PD-L1/PD-L2 co-expression analysis. To determine poten-
tially interacting cell type subclusters for the receptor-ligand pair PD-1–
PD-L1/PD-L2, we first computed the fraction of sender cells (cancer cell
or myeloid cell clusters) expressing the PD-L1 and PD-L2 ligands (CD274
or PDCD1LG2 read counts >0 in >10% of cells) and the fraction of receiver
cells (Tcell clusters) expressing the PD-1 receptor (PDCD1 read counts
>0 in >10% of cells) for every patient. Co-expression networks were
constructed as follows: for a given group of patients of the same muta-
tional subtype (Extended Data Fig.13f), an edge was drawn between
sender cell clusters and receiver cell clusters if the ligands (CD274 or
PDCD1LG2) and receptor (PDCD1) were co-expressed in the sender and
receiver subclusters for at least 50% of patients in that group.
Bulk WGS. Alignment. Sequencing reads were aligned to human ge-
nome reference GRCh37 (hg19) using the Burrows–Wheeler aligner
(BWA-MEM) v0.7.17-r1188 (https://sourceforge.net/projects/bio-bwa/).
Single-nucleotide variants and indels. Single-nucleotide variants
(SNVs) and indels were called using mutationSeq (version 4.3.8; model
v4.1.2.npz) available at https://github.com/shahcompbio/mutationseq.
We also used Strelka (version 2.8.2) with default parameter settings
to identify somatic SNVs and indels42. Both SNVs and indels were then
annotated for variant effects and gene-coding status using SnpEff4
(version 5.0e). We identified a set of high-confidence SNVs by taking the
intersection of the high-probability calls predicted from mutationSeq
(with probability ≥0.9) and the somatic SNVs predicted by Strelka. The
high-confidence set of SNVs was further filtered by removing positions
that fell within either of the following regions: (1) the UCSC Genome
Browser blacklists (Duke and DAC) and (2) regions defined in the ‘CRG
Alignability 36mer track’ with more than two nucleotide mismatches,
requiring a 36-nucleotide fragment to be unique in the genome even
after allowing for two differing nucleotides. Postprocessing on this
set of high-confidence SNVs and somatic indels from Strelka involved
removing known variants (both SNVs and indels), which were obtained
from the 1000 Genomes Project (release 20130502) and dbSNP (version
dbsnp 142.human 9606). The set of high-confidence somatic SNVs and
indels passing the above filters were then used in feature computation
for mutational signature analysis and in neoantigen prediction.
Rearrangements. Rearrangement breakpoints were predicted using
lumpy (version 0.2.12)43 executed by SpeedSeq (version 0.1.08)44 and
destruct (version 0.4.18) derived from nFuse
45
, available at https://
github.com/amcpherson/destruct. In brief, destruct extracted discord-
ant and non-mapping reads from BAM files and realigned the reads
using a seed-and-extend strategy. Split alignment across a putative
breakpoint was attempted for reads that did not fully align to a single
locus. Discordant alignments were clustered according to the likeli-
hood that they were produced from the same breakpoint. Multiply
mapped reads were assigned to a single mapping location using previ-
ously described methods46. Finally, heuristic filters removed predicted
breakpoints with poor discordant read coverage of sequence flanking
predicted breakpoints.
We applied stringent three-step filtering criteria to identify
high-confidence breakpoint calls for downstream analysis as follows:
Step 1: Breakpoints that were predicted by both algorithms, lumpy
and destruct, were taken forward.
Step 2: We removed (1) breakpoints from regions with poor map-
pability, (2) events with a break distance of ≤30 bp and (3) breakpoints
annotated as a deletion with a breakpoint size of <1,000 bp. Fur ther-
more, only high-confidence breakpoints that had at least five support-
ing reads in the tumour sample and no read support in the matched
normal sample were used in the analysis. The breakpoints were fur-
ther filtered by removing positions that fell in either of the following
regions: (1) UCSC Genome Browser blacklists (Duke and DAC) and (2)
regions defined in the ‘CRG Alignability 36mer track’ with more than
two nucleotide mismatches, requiring a 36-nucleotide fragment to be
unique in the genome even after allowing for two differing nucleotides.
Step 3: Predictions with a small break distance and a low number of
supporting reads in tumour samples were excluded.
Copy number. Genome-wide allele-specific copy number was called in
matched tumour–normal WGS samples using ReMixT
47
and TitanCNA
48
with default parameters. A parameter grid search for multiple purity
and ploidy solutions was carried out, and the top solution was selected
after manual assessment of the copy number segmentations. All tumour
samples were run with ploidy = 2 and ploidy = 4 initializations.
Myriad HRD test. We used a commercial assay (Myriad Genetics ‘my-
Choice CDx’) to test for genome-wide LOH, the number of chromo-
somal breakpoints in large-scale state transitions and telomeric allelic
imbalance. If the resulting HRD score was greater than 42, the sample
was deemed to be HRD.
Targeted sequencing (MSK-IMPACT). Genomic DNA isolated from
FFPE tumour tissue and matched normal blood was subjected to hy-
bridization capture and sequenced with deep coverage (700×)49. Variant
calling for the MSK-IMPACT gene panel and copy number analysis were
performed using the MSK-IMPACT clinical pipeline (https://github.
com/mskcc/Innovation-IMPACT-Pipeline).
Mutational signatures. We analysed mutational signatures by integrat-
ing SNVs and structural variations detected by bulk WGS in a unified
probabilistic approach called multimodal correlated topic models
(MMCTM)
6
. MMCTM analysis enables robust determination of mu-
tational signatures and their correlation structure and delineation of
subgroupings based on point mutation signatures
50
and structural
variations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
We estimated signature probabilities for bulk WGS samples in the
MSK SPECTRUM cohort (n = 40) using MMCTM, on the basis of SNV
and structural variation signatures inferred from HGSOC (n = 170)
and triple-negative breast cancer (n = 139) bulk whole genomes (total
n = 309) (Extended Data Fig.2b). By clustering the meta-cohort of
309 HGSOC and triple-negative breast cancer samples using UMAP
and HDBSCAN51, we used the meta-cohort as a training dataset to fit
a k-nearest-neighbour (kNN) classifier and applied the kNN classifier
to the SPECTRUM samples (n = 40), assigning them into one of four
strata defined solely by SNV and structural variation signature prob-
abilities. A nearest-neighbour graph was built using a Euclidean dis-
tance metric, and classification into strata was computed by a majority
vote of the k nearest neighbours of the unknown test sample (k = 30),
requiring m votes for an assignment (m = 25). The four strata included
those with samples enriched for (1) BRCA1-associated HRD point muta-
tion signatures accompanied by tandem duplications (HRD-Dup),
(2) BRCA2-associated HRD point mutation signatures accompanied
by interstitial deletions (HRD-Del), (3) foldback inversions mediated
by breakage–fusion bridge cycles (FBI) and (4) a group of ambigu-
ous samples near the classifier decision boundaries (‘Undetermined’)
(Extended Data Fig.2c).
To validate the MMCTM mutational signatures, we used two
independent computational methods (Extended Data Fig.2b).
We applied HRDetect18 to validate HRD status on the basis of SNV
signatures previously associated with HRD (SBS3, SBS8), short
microhomology-mediated indels (ID8) and rearrangement signatures
(RS3, RS5). Samples with an HRDetect score of >0.1 were defined as HRD.
We also applied CHORD
19
to validate HRD status and stratification of
HRD-Dup from HRD-Del cases. CHORD incorporates SNVs, indels and
structural variations and relies on duplications (1–100 kb) to distinguish
BRCA1-like from BRCA2-like HRD.
WGS-derived HRD signatures were in agreement for seven of seven
cases with BRCA1 or BRCA2 loss (Extended Data Fig.2b). WGS-based and
standard-of-care HRD status were concordant in five of six cases. The
discordant case (024) was deemed HRD by all three independent meth-
ods for WGS signature inference (MMCTM, HRDetect and CHORD).
Focal ampliications and deletions. We used WGS copy number
inferred by ReMixT47 to classify copy number changes as focal am-
plifications and deletions in the MSK SPECTRUM cohort. For focal
amplifications, we calculated the percentile of each gene with respect
to the cumulative distribution of total copy number changes across the
genome. On the basis of the mean copy number across each gene, we
classified high-level amplifications as those in the top 2% of bins with a
log2-transformed change over ploidy greater than 1. For homozygous
deletions, we considered gene copy number in overlapping segments
and we classified segments that were 10 kb or greater in size with a mean
copy number less than 0.5 as homozygous deletions.
Similarly, we used IMPACT copy number inferred by FACETS52 to
delineate focal amplifications and homozygous deletions in the MSK
IMPACT HG SOC cohort. Focal amplifications and deletions were identi-
fied on the basis of the median copy number log ratio persegment, only
considering segments shorter than 10 Mb with ten or fewer genes to
suppress arm-level events. Segments with a total copy number greater
than 8 were considered as high-level amplifications. Homozygous dele-
tions were called for segments with a total copy number of 0.
HLA LOH. To detect allele-specif ic copy number LOH of the HLA locus in
single cells profiled by scRNA-seq, we inferred allele-specific alterations
on chromosome arm 6p, which harbours HLA class I and II genes, using
SIGNALS
5
. We first called germline heterozygous single-nucleotide
polymorphisms (SNPs) in the scRNA-seq tumour data using cellSNP
53
.
As input, we used the set of heterozygous SNPs identified in the cor-
responding normal WGS dataset for each sample. The liftover script
provided in cellSNP was used to lift over SNP coordinates from the
GRCh37 (hg19) reference genome to the GRCh38 reference genome.
Following genotyping, we aggregated SNP counts across all cells and
defined the B allele as the allele with the lowest allele frequency for
each SNP. As SNP counts are very sparse in scRNA-seq data, we then
aggregated cell-level counts of the B allele across chromosome arms
to compute the BAF for each arm in each cell. We then generated a
cell-by-chromosome arm BAF matrix and incorporated this into the
Seurat gene expression objects. To assign allelic imbalance states (bal-
anced, imbalanced, LOH) to chromosome arms in each cell, we used
the mean BAF of each arm per cell as follows: balanced, BAF ≥ 0.35;
imbalanced, 0.15 ≤ BAF < 0.35; LOH, BAF < 0.15. Documentation and
code are available at https://shahcompbio.github.io/signals/.
To validate our observations of allele-specific alterations on chro-
mosome arm 6p in relation to the HLA locus, we detected gene-level
HLA class I LOH from tumour and matched normal WGS data, as well as
from tumour and matched normal MSK-IMPACT data, using LOHHLA
24
.
To validate HLA LOH status by WGS, we used 40 tumour–normal
pairs from 40 patients. Tumour purity and ploidy were estimated using
ReMiXT47 and used for subsequent HLA LOH analysis. To validate HLA
LOH status by MSK-IMPACT, we selected 1,298 tumour–normal pairs
from 1,298 patients in the MSK-IMPACT cohort with HGSOC histology
based on an HGSOC or HGSFT OncoTree classification54. This cohort
did not include MSK-IMPACT samples from patients who were part of
the MSK SPECTRUM cohort.
Patient HLA references were built from tumour and normal reads
using Polysolver (v4)55, for both WGS and MSK-IMPACT data. Tumour
purity and ploidy from the WGS datasets were estimated using ReMixT47
and used for subsequent HLA LOH analysis. Similarly, tumour purity and
ploidy for the MSK-IMPACT datasets were estimated using FACETS
52
.
HLA LOH was called for an allele in the tumour sample using LOHHLA.
LOH was observed for each HLA gene if the estimated copy number
was <0.2 and the statisticalsignificance of the allelic imbalance was
P < 0.01, testing for pairwise differences in log(R) values between the
two HLA homologues (paired t test).
Digital histopathology. We built a training dataset of cellular annota
-
tions for scanned H&E images. Expert delineation and quantification
of cell and tissue types present in the H&E slides was carried out on
MSK Slide Viewer, a computational pathology interface for review
and annotation of histopathology images. Nuclear segmentation was
carried out using StarDist, a method for nuclear detection based on the
U-Net neural network architecture56,57. Membrane segmentation was ap-
proximated using a cell expansion of 3 μm of the nuclear boundary. The
training dataset encompasses a set of 61 slides from a representative set
of patients and sites. To classify regions of tumour, stroma, vasculature
and necrosis, we trained an artificial neural network (ANN)-based pixel
classifier using QuPath (v0.2.3)56, which operates on higher-order pixel
features over multiple channels and scales within an image. In addition,
lymphocytes and ‘other’ cells were annotated in 19 of these slides by a
researcher using MSK Slide Viewer. After importing these annotations
into QuPath, along with cellular segmentations and feature vectors
generated from StarDist, we trained an ANN-based cellular classifier
that operates over cellular measurements to identify lymphocytes. We
then applied these models for inference across100 whole-slide H&E im-
agesfrom 35 patients. Segmentation yielded a total of 24,628,462 cells
across samples,and weused the model outputs to compute statistics
on lymphocyte densities and other spatially derived measurements.
mpIF. We carried out nuclear segmentation based on DAPI intensity
using the watershed algorithm in QuPath (v0.2.3)
56
, setting a mini-
mum DAPI threshold of 1 arbitrary unit with an expected nucleus area
ranging between 5 μm
2
and 100 μm
2
. Membrane segmentation was
approximated using a cell expansion of 3 μm of the nuclear boundary.
Starting from 1,349 quality-filtered FOVs across 100 tissue samples
from 35 patients, segmentation yielded a total of 10,892,612 cells. To
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Article
annotate regions of tumour and stroma, we trained a pixel classifier with
examples of panCK
+
(tumour) and panCK
(stroma) regions. Following
nuclear segmentation, we extracted the pixel intensities per cell for
functional markers expressed in the cytoplasm (panCK, CD68, CD8,
PD-1, PD-L1) and in the nucleus (TOX) to define cell types and cell states.
All channels were manually thresholded in at least one FOV per slide, and
marker positivity was determined by setting these thresholds on the
mean pixel intensity. Segmented objects that were double or triple posi-
tive for multiple cell type markers (panCK, CD68, CD8) were counted as
separate cells, yielding a total of 12,359,463 single cells. Marker assign-
ments were used to define cell states of epithelial cells (panCK
+
PD-L1
,
panCK
+
PD-L1
+
), macrophages (CD68
+
PD-L1
, CD68
+
PD-L1
+
) and CD8
+
Tcells (CD8+PD-1TOX, CD8+PD-1+TOX, CD8+PD-1+TOX+).
Analysis of spatial topology comprised estimation of spatial densi-
ties and intercellular nearest-neighbour distances. Spatial density
estimates as a function of distance to the tumour–stroma boundary
were obtained by aggregating cell counts within 10μm distance bands
from the boundary in each FOV, grouped across FOVs and normal-
ized by the total number of cells for a given phenotype of interest.
Error bars were calculated as the standard error of the probability p
of observing a given phenotype as , where N was the total number of
cells in the distance band. Intercellular distances between nearest
neighbours were calculated using the distance matrix rij for cells i
and j, where the value of the (i, j) element in the matrix was the radial
distance from cell i to cell j. After computing per-cell nearest neigh-
bours, the summary statistics over nearest-neighbour distances were
estimated for each phenotype. Proximity counts for phenotypes
within a fixed radius R were also determined on the basis of per-cell
nearest neighbours.
Reporting summary
Further information on research design is available in theNature Port-
folio Reporting Summary linked to this article.
Data availability
Datasets generated and analysed in this study are available for gen-
eral research use and are documented in Synapse (https://www.syn-
apse.org/msk_spectrum). Open-tier datasets not requiring access
approval are available for download via Synapse (accession number
syn25569736: https://www.synapse.org/msk_spectrum). Controlled-
tier datasets requiring access approval are available by requesting
authorisation to the Data Access Committee via dbGaP (accession
number phs002857.v1.p1: http://www.ncbi.nlm.nih.gov/projects/gap/
cgi-bin/study.cgi?study_id=phs002857.v1.p1). An interactive visualiza-
tion interface for the scRNA-seq data from this study is available via
CELLxGENE (https://cellxgene.cziscience.com/collections/4796c91c-
9d8f-4692-be43-347b1727f9d8). The WGS and MSK-IMPACT datasets
are available for browsing via cBioPortal (https://www.cbioportal.org/
study/summary?id=msk_spectrum_tme_2022).
Code availability