Pancreatic cancer patients are in desperate need of effective therapy virtually from the moment of their diagnosis. As we acquire more therapies, how best to deploy them, in what order and to which patients is emerging as an important clinical question. Pancreatic cancer subtypes, identifiable with common lab diagnostics in diagnostic biopsy samples, may be helpful in guiding therapy selection.

See related article by O'Kane et al., p. 4901

In this issue of Clinical Cancer Research, O'Kane and colleagues (1) present results correlating systemic response to chemotherapy to pancreatic adenocarcinoma (PDA) subtype. PDA is a deadly disease. Most patients present to medical attention with disseminated or otherwise unresectable disease, and rely on systemic, cytotoxic chemotherapy to prolong their lives and mitigate the substantial symptoms caused by the disease. Pain, fatigue, muscle wasting, weight loss, and depression are all too common manifestations of this aggressive systemic disease. As a result, patients with PDA rely on their care team to quickly and accurately select and administer effective intravenous therapy and to medically manage its side effects until it can have a beneficial effect on the disease course.

Our treatments for PDA have gradually improved over the last decade. Practitioners now choose between the doublet of gemcitabine and nano-albumin bound (Nab)-paclitaxel (GnP) and the more complicated infusional 5-Fluorouracil, leucovorin, irinotecan, and oxaliplatin (i.e., FOLFIRINOX) in the first line for fit patients. Unlike the scenarios in lung adenocarcinoma (where we treat the immune system or driver oncogene), or colon cancer (where we treat the absence of the oncogene or the anatomic location of the primary tumor), medical decision making in the treatment of pancreatic cancer is driven largely by the practitioner-assessed functional status of the patient, because the perception is that less fit patients might not do as well with one regimen or another. While rational, this approach is not empirically supported. We have not compared GnP to FOLFIRINOX head to head. We do not know whether one is better tolerated than the other in a head-to-head comparison. Perhaps more importantly, we do not have indicators of which chemotherapy regimen is more effective overall, much less which treatment most benefits an individual patient. In short, we have a long way to go personalizing treatment approaches for patients with this disease.

O'Kane and colleagues present a systematic investigation into pancreatic cancer subtypes, how to practically detect them in clinic, and the impact they may have on the treatment choices physicians and patients face in the first line. They built upon their hugely successful COMPASS trial (2), NCT02750657, a prospective trial focused on generating genomic data on patients with PDA and closely following their responses to standard therapy. COMPASS has generated RNA (3) as well as germline and somatic DNA (4) data on many patients with PDA, and here the investigators focused on 195 patients with mRNA profiling available. Here the authors had built on important findings from Moffit (5), Bailey (6), and others describing intrinsic subtypes of PDA though the use of mRNA profiling data. Using their own variant of the classification scheme pioneered by Moffit and colleagues, they correlated mRNA based subtyping (classical vs. basal) with clinical outcomes as well as nonsequencing-based laboratory classification approaches.

They were able to compare RECIST response data for 157 (29 basal and 129 Classical). They found that patients with classical tumors treated with modified FOLFIRINOX (mFFX) were more likely to survive longer and experience tumor shrinkage (Fig. 1). Lower numbers of patients treated with GnP precluded more definite conclusions as to subtype responsiveness to this popular regimen, although a trend was detected noted favoring longer survival in Basal tumor–harboring patients treated with GnP without reaching statistical significance.

Figure 1.

Pancreatic cancer subtypes, as identified by RNA in situ hybridization and IHC, may correlate with clinical responsiveness to 5-FU–based therapy.

Figure 1.

Pancreatic cancer subtypes, as identified by RNA in situ hybridization and IHC, may correlate with clinical responsiveness to 5-FU–based therapy.

Close modal

The authors additionally correlated GATA6 in situ hybridization (ISH) with the mRNA-determined subtype with encouraging results. Valiant attempts to classify tumors using GATA6 protein (for Classical subtype identification) or KRT5 (for Basal classification) are also presented and may hold promise.

These results show the importance of correlating clinical responsiveness to molecular characteristics of tumors and provide compelling evidence that we need better treatments for all patients with PDA, but particularly for the basal subtype. These conclusions build on and extend similar observations from others who have developed Radiomic (7) or NanoString-based (8) methods to identify subtypes of PDA in the clinic. O'Kane and colleagues provide compelling evidence that we may be able to improve overall survival in patients with PDA as a whole by better selecting first-line treatment, where we have the best shot at achieving a meaningful clinical response through dose-intensive systemic cytotoxic therapy. These results point us in a direction using mRNA subtypes, possibly determined by lab-friendly methods, to make decisions in the clinic. Future studies will likely consider PDA subtype as a meaningful variable in supervising responses in therapeutic trials. Further off, we may consider subtype-directed trials in which we iteratively improve outcomes for the whole by improving survival for one subtype at a time.

No potential conflicts of interest were disclosed.

E. A. Collisson is supported by the NIH (5R01CA222862).

1.
O'Kane
GM
,
Grunwald
BT
,
Jang
GH
,
Masoomian
M
,
Picardo
S
,
Grant
RC
, et al
GATA6 expression distinguishes classical and basal-like subtypes in advanced pancreatic cancer
.
Clin Cancer Res
2020
;
26
:
4901
10
.
2.
Aung
KL
,
Fischer
SE
,
Denroche
RE
,
Jang
GH
,
Dodd
A
,
Creighton
S
, et al
Genomics-driven precision medicine for advanced pancreatic cancer - early results from the compass trial
.
Clin Cancer Res
2018
;
24
:
1344
54
.
3.
Connor
AA
,
Denroche
RE
,
Jang
GH
,
Lemire
M
,
Zhang
A
,
Chan-Seng-Yue
M
, et al
Integration of genomic and transcriptional features in pancreatic cancer reveals increased cell cycle progression in metastases
.
Cancer Cell
2019
;
35
:
267
82
.
4.
Notta
F
,
Chan-Seng-Yue
M
,
Lemire
M
,
Li
Y
,
Wilson
GW
,
Connor
AA
, et al
A renewed model of pancreatic cancer evolution based on genomic rearrangement patterns
.
Nature
2016
;
538
:
378
82
.
5.
Moffitt
RA
,
Marayati
R
,
Flate
EL
,
Volmar
KE
,
Loeza
SG
,
Hoadley
KA
, et al
Virtual microdissection identifies distinct tumor- and stroma-specific subtypes of pancreatic ductal adenocarcinoma
.
Nat Genet
2015
;
47
:
1168
78
.
6.
Bailey
P
,
Chang
DK
,
Nones
K
,
Johns
AL
,
Patch
AM
,
Gingras
MC
, et al
Genomic analyses identify molecular subtypes of pancreatic cancer
.
Nature
2016
;
531
:
47
52
.
7.
Kaissis
GA
,
Ziegelmayer
S
,
Lohofer
FK
,
Harder
FN
,
Jungmann
F
,
Sasse
D
, et al
Image-based molecular phenotyping of pancreatic ductal adenocarcinoma
.
J Clin Med
2020
;
9
:
724
.
8.
Rashid
NU
,
Peng
XL
,
Jin
C
,
Moffitt
RA
,
Volmar
KE
,
Belt
BA
, et al
Purity-independent subtyping of tumors (PurIST), a clinically robust, single-sample classifier for tumor subtyping in pancreatic cancer
.
Clin Cancer Res
2020
;
26
:
82
92
.