We thank Bouzid and colleagues for their interest in our article describing pancreatic cancer stromal types identified by image analyses of two distinct fibroblast subpopulations, ACTA2 (α smooth muscle actin)–dominant fibroblasts and FAP (fibroblast activation protein α)–dominant fibroblasts, intratumoral collagen, and CD8+ cells using whole tissue slides from 215 patients with pancreatic ductal adenocarcinomas (1).

Bouzid and colleagues have confirmed our previous findings that CD8+ cell density is generally lower in the tumor center than in the tumor margin (1), suggesting that T-cell exclusion is a hallmark of the immunosuppressive nature of pancreatic cancers. Then, their letter addresses the association between FAP expression and CD8+ cell exclusion. In our analysis on a relatively large series of cases, the proportion of FAP–dominant fibroblasts (but not for ACTA–dominant fibroblasts or collagen) showed an inverse correlation with the ratios of CD8+ cell density in the tumor center to the tumor margin (r = −0.27, P < 0.001; ref. 1). This finding was consistent with previous studies demonstrating the immunosuppressive role of FAP-expressing stromal cells in preclinical models and human specimens (2, 3). In contrast, a morphometric analysis of 31 pancreatic cancers by Bouzid and colleagues has not found any association between local FAP expression and CD8+ T-cell presence. Together, as they pointed out, one reasonable explanation is that FAP-expressing fibroblasts themselves are not a standalone driver of T-cell exclusion. Evidence indicates that a subset of cancer-associated fibroblasts (CAF) secretes many kinds of chemokines to recruit various types of immune cells, including tumor-associated macrophages, neutrophils, regulatory T cells, and MDSC (1, 2, 4), which in turn leads to local T-cell suppression in a context-dependent manner. Alternatively, there could be diversity in FAP-dominant fibroblasts (and maybe in ACTA2-dominant fibroblasts) within pancreatic cancer stroma. Indeed, a single-cell transcriptome analysis on CAFs in solid tumors has identified five or more distinct clusters in FAP-expressing CAFs, several of which likely play different roles in maintaining the immunosuppressive microenvironment (5). In addition, non–CAF-mediated mechanisms also contribute to driving T-cell exclusion from the tumor bed (3). These lines of evidence suggest heterogeneity in mechanisms underlying T-cell exclusion within the pancreatic cancer microenvironment.

Stromal therapies attempting to overcome resistance to chemotherapies and immunotherapies without detrimental effects have been, so far, largely unsuccessful. A phase IB/II study in patients with metastatic pancreatic cancers (NCT01959139) showed that the combination of PEGPH20 and modified FOLFIRINOX was poorly tolerated (3). A phase II trial (NCT01130142) designed to recapitulate the benefit of the Hedgehog inhibitor in murine pancreatic cancer models paradoxically showed worse clinical outcomes in combination with gemcitabine (3), indicating nonselective inhibition of fibroblasts by Hedgehog pathway inhibition could be harmful in the treatment of human pancreatic cancers. On the other hand, FAP-expressing fibroblasts and their secreting chemokines/cytokines have emerged as attractive therapeutic targets against several solid tumors (2, 3). For example, dual blockade of CXCR4 and PD-1 with or without chemotherapy has been safe and well tolerated, and has shown encouraging results in phase II clinical trials for patients with metastatic pancreatic cancers (4, 6). Note that the inhibition of the CXCR4–CXCL12 axis with PD-1 pathway blockade resulted in increased T cells with a decrease of MDSCs in the tumor microenvironment (4). Because adequate patient selection would enhance the potential efficacy of these targeted therapies, future biomarker studies are warranted to stratify patients based on differential pathobiological resistant pathways. Hence, a deeper understanding of intertumoral and intratumoral stromal heterogeneity in clinical samples of pancreatic cancers is clearly needed to harness stromal therapies in the treatment of this deadly disease.

See the original Letter to the Editor, p. 425

No disclosures were reported.

1.
Ogawa
Y
,
Masugi
Y
,
Abe
T
,
Yamazaki
K
,
Ueno
A
,
Fujii-Nishimura
Y
, et al
Three distinct stroma types in human pancreatic cancer identified by image analysis of fibroblast subpopulations and collagen
.
Clin Cancer Res
2021
;
27
:
107
19
.
2.
Whittle
MC
,
Hingorani
SR
. 
Fibroblasts in pancreatic ductal adenocarcinoma: biological mechanisms and therapeutic targets
.
Gastroenterology
2019
;
156
:
2085
96
.
3.
Bejarano
L
,
Jordāo
MJC
,
Joyce
JA
. 
Therapeutic targeting of the tumor microenvironment
.
Cancer Discov
2021
;
11
:
933
59
.
4.
Bockorny
B
,
Semenisty
V
,
Macarulla
T
,
Borazanci
E
,
Wolpin
BM
,
Stemmer
SM
, et al
BL-8040, a CXCR4 antagonist, in combination with pembrolizumab and chemotherapy for pancreatic cancer: the COMBAT trial
.
Nat Med
2020
;
26
:
878
85
.
5.
Kieffer
Y
,
Hocine
HR
,
Gentric
G
,
Pelon
F
,
Bernard
C
,
Bourachot
B
, et al
Single-cell analysis reveals fibroblast clusters linked to immunotherapy resistance in cancer
.
Cancer Discov
2020
;
10
:
1330
51
.
6.
Bockorny
B
,
Macarulla
T
,
Semenisty
V
,
Borazanci
E
,
Feliu
J
,
Ponz-Sarvise
M
, et al
Motixafortide and pembrolizumab combined to nanoliposomal irinotecan, fluorouracil, and folinic acid in metastatic pancreatic cancer: The COMBAT/KEYNOTE-202 trial
.
Clin Cancer Res
2021
;
27
:
5020
7
.