REVIEW

published: 14 May 2019

doi: 10.3389/fonc.2019.00396

Review of PD-1/PD-L1 Inhibitors in

Metastatic dMMR/MSI-H Colorectal
Cancer
André F. Oliveira*, Luís Bretes and Irene Furtado
Serviço de Oncologia Médica, Centro Hospitalar Universitário Algarve, Faro, Portugal

There are a wide range of therapies for metastatic colorectal cancer (CRC) available,
but outcomes remain suboptimal. Learning the role of the immune system in
cancer development and progression led to advances in the treatment over the
last decade. While the field is rapidly evolving, PD-1, and PD-L1 inhibitors have
a leading role amongst immunomodulatory agents. They act against pathways
involved in adaptive immune suppression resulting in immune checkpoint blockade.
Immunotherapy has been slow to impact the management of this patient population
Edited by: due to disappointing results, mainly when used broadly. Nevertheless, some patients
Giuseppe Valentino Masucci,
with microsatellite-instability-high (MSI-H) or mismatch repair-deficient (dMMR) CRC
Karolinska Institute (KI), Sweden
appear to be susceptible to checkpoint inhibitors with objective and sustained clinical
Reviewed by:
Ali H. Zaidi, responses, providing a new therapeutic option for patients with advanced disease.
Allegheny Health Network, This article provides a comprehensive review of the early and late phase trials with
United States
Lisa Villabona, the updated data of PD-1/PD-L1 inhibitors alone or in combination with other therapies
Karolinska Institute (KI), Sweden (immunotherapy, targeted therapy and chemotherapy). While data is still limited, many
*Correspondence: ongoing trials are underway, testing the efficacy of these agents in CRC. Current and
André F. Oliveira
future challenges of PD-1 and PD-L1 inhibitors are also discussed.
afoliveira@chalgarve.min-saude.pt
orcid.org/0000-0002-1118-3610 Keywords: immunotherapy, colorectal cancer, PD-1, inhibitors, PD-L1, microsatellite instability, MSI-H, dMMR

Specialty section:
This article was submitted to INTRODUCTION
Gastrointestinal Cancers,
a section of the journal
Colorectal cancer (CRC) is the third most common cancer in men and the second in women
Frontiers in Oncology
worldwide. The incidence across the globe is different in various countries, but about 55% of
Received: 22 December 2018 the cases occur in more developed regions. Despite this fact, geographic patterns are very similar in
Accepted: 26 April 2019
men and women (1). Estimates indicate that nearly half of patients with CRC are found with hepatic
Published: 14 May 2019
metastasis during the follow-up of their disease (2). Regrettably, most patients won’t undergo
Citation: surgery for metastases because of unresectability or comorbidities. Clinical outcomes have however
Oliveira AF, Bretes L and Furtado I
improved over the past two decades. Chemotherapy with biological agents remains the standard of
(2019) Review of PD-1/PD-L1
Inhibitors in Metastatic dMMR/MSI-H
care of today. Patients with metastatic colorectal cancer (mCRC) have a median overall survival
Colorectal Cancer. (OS) of 26 months for RAS-mutated and 30 months for those with RAS wild-type status (3, 4).
Front. Oncol. 9:396. Recently immunotherapy changed the treatment paradigm and its respective outcomes in many
doi: 10.3389/fonc.2019.00396 tumor types like melanoma, renal, bladder and lung cancer.

Frontiers in Oncology | www.frontiersin.org 1 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

FIGURE 1 | Deactivated T Cell (Left): when programmed-death receptor (PD-1) on the T cell binds to programmed death-ligand 1 (PD-L1) on the tumor cell, the T
cell becomes deactivated, allowing the cancer cell to evade immune attack. Inhibitors of PD-1 and PD-L1 can prevent the tumor cell from binding to PD-1, enabling
the T cell to remain active and co-ordinate an attack (on the Right). Reproduced with permission of Terese Winslow LLC. Credit: For the National Cancer Institute ©
2015 Terese Winslow LLC, U.S. Govt. has certain rights.

Immunotherapy With PD-1 Inhibition Mismatch Repair Deficient

Nobel Prize in Physiology or Medicine of 2018 was awarded for (dMMR)/Microsatellite Instability-High
the discovery of cancer therapy by the inhibition of negative (MSI-H) Tumors
immune regulation. Research lead by Prof. Honjo identified PD-1 Findings of previous studies with PD-1 inhibitors in mCRC were
(Programmed cell death protein-1) as a component of inhibition not notably encouraging. Responses were only rarely seen (10).
of the immune system, noting that disruption of PD-1 in pre- Further interpretation of these studies revealed a high tumor
clinical models resulted in increased immune system activity. mutational burden (TMB) as a predictive marker for response.
This key information helped initiate the development of anti-PD- The rationale was that the higher the TMB, the higher neo-
1/PD-L1 (Programmed death ligand-1) checkpoint inhibitors antigen load and therefore higher tumor immunogenicity (11).
that are now a standard of care treatments for several tumor Also of interest were findings of types of mutations that would
types (5). PD-1 is a cell surface receptor found on activated T determine neo-antigen load. It was found the greatest with by
cells, pro-B cells and macrophages. When PD-L1 is bound to PD- insertion or deletions (12). The population of patients with CRC
1, the result is a counter-inhibitory negative feedback loop. This with these characteristics are relatively small and are mainly
functions as a protective mechanism that prevents a host from an constituted by tumors with mismatch repair-deficient (dMMR)
attack by its own immune system (6, 7). Some cancers exploit this or microsatellite-instability-high (MSI-H). In this subset of
negative feedback loop by making its cells express PD-L1 in order patients it is found a large number of activated CD8-positive
to evade immunosurveillance. Binding of PD-1 to PD-L1 was cytotoxic T cells and upregulated checkpoints (e.g., CTLA 4,
shown to disable the effector function of lymphocytes, decrease PD-1, and/or PD-L1) (13, 14).
T cell receptor-mediated activation and exhaust proliferation of Besides dMMR/MSI-H patients, other subjects who respond
cytotoxic T lymphocytes in response to cancer cells. This results to PD-1/PD-L1 inhibitors have tumors with an even higher
in impaired immune activity known as T-cell exhaustion (8). T TMB known as hypermutated tumors, e.g., tumors harboring
cells can remain in a state of anergy (9) and cancer cells escape polymerase proofreading domain mutations (POLE), which are
immune surveillance. The goal of PD-1/PD-L1 inhibitors is to thought to also be susceptible to immunotherapy (15, 16).
block this inhibitory immune checkpoint molecule (Figure 1). dMMR CRCs are found in 15–20% of stage II and III
Results of this therapy are primarily unknown in CRCs. CRCs and are associated with a better prognosis than proficient

Frontiers in Oncology | www.frontiersin.org 2 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

FIGURE 2 | PRISMA flow chart of study selection.

mismatch repair (pMMR) tumors. In the metastatic setting, and English language. Since immunotherapy is a very active
dMMR CRCs represent only around 5% and are associated with research field, we also explored abstracts from known oncology
a worse prognosis (17). conferences during 2014–2018. Our systematic review focused
Researches showed that most CRC patients are not responsive on published clinical trials according to Preferred Reporting
to immunotherapy. Several factors may be involved in this lack of Items for Systematic Reviews and Meta-analyses (PRISMA)
sensitivity. Lack of T-cell infiltration, type 1 T-helper cell activity guidelines (22). The flow diagram for article selection is shown
and low immune cytotoxicity in tumor microenvironment are in Figure 2. Each identified article was analyzed and classified.
thought to be involved (18, 19). Many attempts have been Primary outcomes comprised oncologic results with progression
conducted to characterize CRC and detect subgroups susceptible free and/or overall survival and tumor response rates. We also
for individual treatments [e.g., the four consensus molecular recorded and reviewed ongoing clinical trials that are in progress
subtypes (CMS 1-4)]. CMS 1 (immune) subtype already specifies within the subject of the review. The search was done mainly on
high TMB subgroup, which is responsive to immunotherapy (20). clinicaltrials.gov with the same algorithms described above.
Although the CMS classification may help to guide clinicians and
researchers in management and treatment of CRC, there is no
clinical relevance yet (21). RESULTS
Single-Agent PD-1 Inhibitor
EVIDENCE ACQUISITION Pembrolizumab
The first study to show the clinical efficacy of PD-1 blockade
An article search was performed in PubMed and Cochrane in the microsatellite instability subset of colorectal cancer
databases to identify clinical and randomized controlled trials patients was with Pembrolizumab, an Anti-PD1 inhibitor. The
published until August 2018. Multiple algorithms that included trial examined the treatment of tumors with MMR deficiency
the following terms were used: colorectal cancer, colon following a recurrence of disease after standard treatments in a
cancer, immunotherapy, PD-1, PD-L1, MSI-H, dMMR, and wide range of malignancies. Clinical activity of pembrolizumab
microsatellite instability. Inclusion criteria used were published 10 mg/kg was evaluated in 3 patient groups: those with dMMR
full articles, clinical trials, retrospective series, meta-analyses, mCRC, pMMR and a third group that included patients with

Frontiers in Oncology | www.frontiersin.org 3 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

dMMR in other malignancies (non-CRC) (13). Overall response one line of therapy with mCRC MSI-H (24). Patients received
rates were 40% (4 of 10 patients) with a 78% progression-free pembrolizumab 200 mg every 3 weeks. Endpoints included
survival rate at 20 weeks (7 of 9 patients) in the cohort of dMMR objective response rate (ORR) (primary) and duration of
mCRC. In the second group of those with pMMR, the objective response (DOR), PFS, overall survival (OS). Sixty-three patients
response rate was 0% (0 of 18 patients), with only 11% exhibiting were enrolled. The most recent results published an ORR of 32%,
Progression-Free Survival (PFS) at 20 weeks. These results two complete responses and 18 partial responses. Median PFS
highlighted the activity of Pembrolizumab in this subgroup of was 4.1 months and median OS was not reached. The 12-month
patients with dMMR mCRC, leading to the approval of the first- PFS rate was 41%, and the 12-month OS rate was 76%. This study
ever agnostic (i.e., histology and tumor-site independent) cancer revealed a durable antitumor in patients with MSI-H CRC treated
drug in May 2017 in the US. with Pembrolizumab who progressed after a first-line of therapy.
A second phase Ib trial, the KEYNOTE-028, evaluated These findings are currently being evaluated in a larger phase III
Pembrolizumab in PD-L1–positive advanced solid tumors. The trial, the KEYNOTE-177, in patients with dMMR mCRC (25).
results for the CRC cohort were published (23). Patients with
advanced and treatment-resistant PD-L1–positive were enrolled, Nivolumab
disregarding microsatellite status. Pembrolizumab in a dose Nivolumab was initially studied in phase I trial on various
of 10 mg/kg was administered every 2 weeks for up to 2 solid tumors. Only 14 patients enrolled had mCRC. A durable
years or until disease progression/unacceptable toxicity. Primary complete response was reported on a single patient. Nivolumab
endpoints were safety and overall response rate. 23 patients was further evaluated in a phase II trial on 296 patients but no
were enrolled with a median follow-up of 5.3 months. Most objective responses were achieved among those with mCRC. As
patients (n=15, 65%) experienced progressive disease. One the only responder was found to have a deficient MMR (dMMR)
partial response occurred in a patient (4%) with MSI-H CRC. tumor, this status was presumed to be a predictor of efficacy (10).
Pembrolizumab demonstrated a favorable safety profile and A phase 2 trial, CheckMate 142, provided further evidence
its antitumor activity was only observed this single patient for the use of Nivolumab in dMMR/MSI-H mCRC (26). Seventy
with MSI-H. This justified an evaluation for its use in this four patients recruited were assessed locally with dMMR/MSI-
patient population. H CRC. Seventy two percent were confirmed after central
KEYNOTE-164 was another study with Pembrolizumab. The molecular testing to have MSI-H tumors. Primary endpoint was
cohort included patients previously treated with more than an investigator-assessed overall response (iORR), achieved in 23

TABLE 1 | Results from clinical trials with PD-1/PD-L1 inhibitors.

Study/ClinicalTrials.gov Drug(s) N Patient population (iO)RR Phase Primary 12m OS

Identifier endpoint

Le et al. (13), NEJM 2015 Pembrolizumab 41 (32 CRC) dMMR:11 pMMR 21 dMMR 40% II irPFS –
NCT01876511 pMMR 0%
Lee et al. (27), JCO 2017 Pembrolizumab + azacitidine 31 30 pts with MSS mCRC 3% II ORR –
NCT02260440
Shahda et al. (28), JCO 2017 Pembrolizumab + mFOLFOX6 30 (3 MSI-H) 1st line mCRC 53% II mPFS –
NCT02375672
O’Neil et al. (23), BH 2017 Pembrolizumab 137 (23 PD-L1 positive refractory 4% Ib ORR 29,8%
NCT02054806 enrolled) mCRC
Le Dung et al. (24), Pembrolizumab 63 MSI-H mCRC treated with 32% II ORR 76%
KEYNOTE-164 ≥1 prior line
NCT02460198
NCT02788279 Atezolizumab +- Cobimetinib 363 (1.7% MSS/MSI-L mCRC 2,7% III OS –
MSI-H)
NCT01633970 Atezolizumab + FOLFOX + 23 Refractory mCRC 52% Ib Safety –
Bevacizumab
Brahmer et al. (10), NEJM 2012 Nivolumab 19 mCRC MSI unknown 0% I (multi Safety –
NCT00729664 tumors)
CheckMate142 Nivolumab 74 dMMR/MSI-H mCRC 31,1% II ORR 85%
NCT02060188
CheckMate142 Nivolumab + Ipilimumab (4 doses) 119 dMMR/MSI-H refractory 55% II ORR 85%
NCT02060188 mCRC
CheckMate142 Nivolumab + Ipilimumab (1mg/kg) Q6W 45 dMMR/MSI-H First-line 60% II ORR 83%
NCT02060188 mCRC
NCT02298946 CTX + AMP-224 + SBRT 17 mCRC 0% I Safety –

CTX, cyclophosphamide; SBRT, stereotactic body radiation therapy; mCRC, metastatic colorectal cancer; MSI, microsatellite instability; H, high; MSS, microsatellite stability; pMMR,
mismatch repair proficient; ORR, objective response rate; irORR, immune-related ORR; PFS, progression-Free Survival; OS, overall survival; RR, response rate; BRR, best RR. Details
available at: www.clinicaltrials.gov.

Frontiers in Oncology | www.frontiersin.org 4 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

TABLE 2 | Ongoing Phase II and III trials with PD-1/PD-L1 inhibitors.

ClinicalTrials.gov identifier Drug(s) Phase Patient Population Primary Endpoint Completion Date

NCT03396926 Pembrolizumab + bevacizumab + capecitabine II pMMR mCRC ORR April 2021

NCT03259867 TATE treatment + Pembrolizumab IIA Liver metastasis from CRC RR October 2021
NCT03519412 Induction (pMMR): Temozolomide Treatment: Pembrolizumab II dMMR or pMMR mCRC ORR July 2019
NCT03631407 Vicriviroc + Pembrolizumab II MSS mCRC ORR March 2025
NCT02981524 CY/GVAX with Pembrolizumab II MMR-p mCRC ORR November 2017
NCT02563002 Pembrolizumab III MSI-H/dMMR mCRC PFS, OS March 2025
NCT02437071 Pembrolizumab + RT II pMMR mCRC ORR September 2019
NCT02227667 Durvalumab II mCRC MSI-H BRR August 2021
NCT02870920 Durvalumab + Tremelimumab II Refractory mCRC OS February 2019
NCT02997228 Atezolizumab +- (Bevacizumab + mFOLFOX6) III dMMR mCRC PFS March 2022
NCT02873195 Atezolizumab + Capecitabine + Bevacizumab II Refractory mCRC PFS November 2022
NCT02291289 Atezolizumab II mCRC PFS April 2019
NCT02992912 Atezolizumab + SABR II Metastatic multi tumors PFS December 2021
NCT03050814 Avelumab + vaccine Ad-CEA II mCRC PFS November 2020
NCT03186326 Avelumab II Second line MSI-H mCRC PFS December 2018
NCT03642067 Nivolumab + Relatlimab II MSS mCRC ORR November 2021
NCT02860546 Nivolumab + TAS 102 II mCRC MSS irORR November 2017
NCT03638297 BAT1306 + Cox inhibitor II MSI-H/dMMR or High TMB RR January 2023

mCRC, metastatic colorectal cancer; MSI, microsatellite instability; MSS, microsatellite stability; pMMR, mismatch repair proficient; ORR, objective response rate; irORR, immune-related
ORR; PFS, progression free survival; OS, overall survival; RR, response rate; BRR, best RR. Details available at: www.clinicaltrials.gov.

patients (31,1%). Durable responses (≥12 months) were reported in patients with previously treated mCRC without standard
in eight (35%) of 23 responders. PFS and OS at 1 year were 50 and treatment options. The combination showed a tolerable safety
73%, respectively. An improvement in some parameters quality profile but had a little anti-tumor effect for MSS mCRC (27).
of life was also documented. Pembrolizumab was also tested with mFOLFOX6 in a phase 2
trial of untreated or unresectable CRC. 30 patients were assigned
Combinations With PD-1/PD-L1 Inhibitors to a single arm study with 3 of them being MSI-H. Treatment
The large population of mCRC patients are not dMMR/MSI-H resulted in 15 patients achieving partial responses (CR+PR =
which make up to 95% of patients. Based on previously reported 53%) and 14 stable diseases. Of note, is a single patient with
low efficacy of immunotherapy in unselected patients, several dMMR that had resection that showed complete pathologic
combination regimens with local ablation, chemotherapy or response after 2 months of therapy. Increased neutropenia was
molecularly targeted drugs have been already evaluated (Table 1) the main toxicity found. Clinical activity was identified, including
and a large number of trials in this setting are still ongoing those with pMMR (28).
(Tables 2, 3) either in selected or unselected populations. The combination of Atezolizumab an anti–PD-L1 monoclonal
The combination of immune checkpoint inhibitors with antibody with FOLFOX and Bevacizumab was studied in a
Nivolumab and Ipilimumab (anti-CTLA4) in dMMR/MSI-H multi tumor phase I trial. Preliminary data suggests that the
mCRC patients were studied in a cohort with 119 patients combination can promote immune-related activity resulting in
of the CheckMate 142. Published outcomes demonstrated a enhanced efficacy. However, more robust data is need (31).
consistent clinical effect with an ORR of 55% and a 12-weeks Atezolizumab in association with Cobimetinib a
disease control rate-rate 80% (29). Responses were durable with MEK1/MEK2 inhibitor in the MAPK pathway was also
a PFS rate of 71% and OS of 85% after 1 year. Responses studied in patients with MSS/MSI-L mCRC in the IMblaze370
were independent RAS/BRAF mutation status, PD-L1 expression trial. The results of primary analysis were presented recently. The
or Lynch syndrome history. Patients recruited were heavily study did not however meet its primary endpoint which was OS
pre-treated with majority having at least two prior lines of with a median of 8.9 month with combination vs. 8.5 month with
therapy for metastatic disease. Recently published, is another regorafenib. In this study, almost all patients included (91.7%)
cohort of the same study but in first-line chemorefractory had tumors with MSS/MSI-L status (32).
mCRC with nivolumab plus low dose ipilimumab. It resulted Lastly, PD-L2 Fc fusion protein that binds to PD-1, known
in lower toxicity and with a median of 2.6 months for patients as AMP-224, was tested in combination with stereotactic body
to respond to treatment. The ORR was 60%, the disease radiation therapy (SBRT) plus cyclophosphamide in mCRC
control rate was 84%, and 7% of patients had a complete with hepatic metastasis. This combination appeared safe and
response (30). feasible, but preliminarily data showed no objective responses.
Other studies combining Pembrolizumab with chemotherapy Expected in a near future are clinical and correlative data
were published. Pembrolizumab plus Azacytidine was evaluated including post-therapeutic radiated and non-radiated tumor
in a phase 2 trial to assess anti-tumor activity and safety biopsies (33).

Frontiers in Oncology | www.frontiersin.org 5 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

TABLE 3 | Ongoing Phase I and II trials with PD-1/PD-L1 inhibitors.

ClinicalTrials.gov Drug(s) Phase Patient Population Primary Completion

identifier Endpoint Date

NCT02851004 BBI608 (Napabucasin) + Pembrolizumab Ib/II mCRC irORR June 2022

NCT03531632 MGD007 + MGA012 I/II mCRC Safety December
2019
NCT03274804 Maraviroc + Pembrolizumab I MSS mCRC Safety April 2022
NCT03374254 Pembrolizumab + Binimetinib (+-CT) I mCRC Safety November
2019
NCT03202758 Durvalumab + Tremelimumab + FOLFOX Ib/II Refractory mCRC – October 2022
NCT02437136 Entinostat + Pembrolizumab Ib/II pMMR mCRC – August 2020
NCT02636036 Enadenotucirev + Nivolumab I Metastatic or advanced Safety August 2019
epithelial tumors
NCT02777710 Pexidartinib + Durvalumab I Metastatic/advanced Safety March 2020
pancreatic or colorectal
cancers
NCT03206073 Durvalumab + Pexa-Vec +- Tremelimumab I/II Refractory mCRC PFS June 2019
NCT03332498 Ibrutinib + Pembrolizumab I/II Refractory mCRC Safety December
2021
NCT02886897 D-CIK and anti-PD-1 antibody I/II Multi tumors PFS February
2022
NCT02335918 Varlilumab + Nivolumab I/II Multi tumors ORR October 2019
NCT03058289 INT230-6 + Pembrolizumab I/II Multi tumors Safety May 2020
NCT02834052 Pembrolizumab + Poly-ICLC I/II pMMR CRC RR August 2020
NCT02959437 Pembrolizumab + Epacadostat + (Azacitidine I/II MSS mCRC ORR January 2021
or INCB057643)
NCT03085914 Epacadostat + Pembrolizumab + mFOLFOX6 I/II Advanced or ORR October 2020
metastatic solid tumors
NCT02903914 INCB001158 + Pembrolizumab I/II Multi tumors Safety October 2022
NCT03168139 Olaptesed pegol + Pembrolizumab I/II Refractory mCRC Safety May 2019
NCT02650713 RO6958688 + Atezolizumab Ia/Ib Refractory mCRC Safety July 2019

mCRC, metastatic colorectal cancer; MSS, microsatellite stability; pMMR, mismatch repair proficient; ORR, objective response rate; irORR, immune-related ORR; PFS, progression-free
survival. Details available at: www.clinicaltrials.gov.

DISCUSSION AND CHALLENGES Particular patients may present primary or secondary

resistance to immune checkpoint inhibitors. It will be important
Many of the published studies are early phase clinical trials, with in the future to study mechanisms of primary and acquired
limited number of mCRC patients recruited. At least six trials resistance to PD-1 blockade. Furthermore, there are many
with favorable objective response and improved progression-free rationales for the study of combination therapies with PD-1
survival in patients with MSI-H CRCs were observed. These inhibitors and compounds targeting other immune regulators
results are encouraging. However, the population of dMMR/MSI- (e.g., CTLA-4, LAG-3, OX-40, TIM-3, KIR, VISTA, GITR,
H is very low, representing only 5% of patients in the metastatic IDO-1,2, and others). They might prove together more
setting. ORR in CheckMate142 and KEYNOTE-164 was around effective or have other properties after certain resistances. The
30% with a substantially higher rate 52% reported in KEYNOTE- highlighted any many other strategies involving targeted therapy,
016; 1-year overall survival rates were very high which may chemotherapy and radiotherapy are ongoing with the aim to
indicate durable responses. (26). enhance response to immunotherapy (Tables 2, 3).
The clinical trials with PD-1/PD-L1 inhibitors also revealed Current predictive biomarkers for the efficacy of anti PD1
different results according to the patient population studied. are dMMR status and POLE mutations. POLE mutations are
Outcomes in MSS/pMMR population including a phase III associated with an ultramutated phenotype and are reported
study with Atezolizumab plus Cobimetinib were negative, not in 1–2% of colorectal cancers (35). Still a subset of patients
providing any evidence of its use in this context. with immune-sensitive tumors, can be probably identified with
In non-metastatic CRC, dMMR accounts for approximately molecular subtyping, immune-inflammation gene expression
15–20% which can be found more frequently. An adjuvant trial signature or immunoscoring (20, 35). Lastly, gut microbiome
with an anti-PD-L1 is ongoing, but it remains unknown its modulation and other therapy combinations might prove to
effectiveness in this situation. Adjuvant ipilimumab or nivolumab beneficial and enhance the effects of immunotherapy in under
have shown to extend survival in stage III or IV (resected) other conditions resistant tumors (18, 36).
but mechanisms of resistance in CRCs in this setting are not The challenges pointed out, and many others are ahead and
necessarily the overexpression of immune checkpoints (34). will include other clinical and feasibility issues. Questions like

Frontiers in Oncology | www.frontiersin.org 6 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

how to increase or reinforce the efficacy of immunotherapy, how to find predictive biomarkers of efficacy. dMMR/MSI-
the optimal duration and combinations of treatments (37) H tumors only account for a small 5% of mCRC. There is
and indications of surgical interventions after immunotherapy no doubt that extending the benefit of immunotherapy to a
(38) will be slowly unraveled. Also, feasibility will undoubtedly broader, microsatellite stable population would be the next but
be discussed since the higher costs of this drugs can add a nevertheless difficult step.
substantial burden to healthcare systems, as the example of Many trials are already in progress exploring combinations
Canada (39). strategies. Other important topics that in the future will be
important to address are the role of immunotherapy and anti-
CONCLUSIONS PD1 therapy in the prevention of CRC, conversion therapy
of potentially resectable liver metastases, and adjuvant or
The outcome of CRC patients has improved considerably neoadjuvant treatment. Trials already published have been
over the past two decades. The efficacy of systemic therapies received by gastrointestinal oncology community with great
and biomarker-based treatments has been predominant in this enthusiasm, and future studies with PD-1 inhibitors mCRC will
positive change. A better understanding of the interaction further help us to decipher the many pieces of a big puzzle
between a tumor and the immune system in the last decade led with immunotherapy.
to the development of new agents, in particular, the PD-1/PD-
L1 inhibitors. AUTHOR CONTRIBUTIONS
Even though PD-1 inhibitors have shown efficacy in dMMR
CRCs, there are still many questions which need to be answered, All authors listed have made a substantial, direct and intellectual
e.g. how could we conceive a response on pMMR CRCs and contribution to the work, and approved it for publication.

REFERENCES the immunogenic phenotype: a pan-cancer analysis. Lancet Oncol. (2017)

18:1009–21. doi: 10.1016/S1470-2045(17)30516-8
1. Richman DM, Tirumani SH, Hornick JL, Fuchs CS, Howard S, Krajewski 13. Le DT, Uram JN, Wang H, Bartlett BR, Kemberling H, Eyring AD, et al. PD-
K, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: 1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. (2015)
IARC CancerBase No. 11. Lyon: International Agency for Research 372:2509–20. doi: 10.1056/NEJMoa1500596
on Cancer (2013). Available online at: http://globocan.iarc.fr, accessed 14. Llosa NJ, Cruise M, Tam A, Wicks EC, Hechenbleikner EM, Taube JM, et al.
on day/month/year The vigorous immune microenvironment of microsatellite instable colon
2. Manfredi S, Lepage C, Hatem C, Coatmeur O, Faivre J, Bouvier AM. cancer is balanced by multiple counter-inhibitory checkpoints. Cancer Discov.
Epidemiology and management of liver metastases from colorectal cancer. (2015) 5:43–51. doi: 10.1158/2159-8290.CD-14-0863
Ann Surg. (2006) 244:254–9. doi: 10.1097/01.sla.0000217629.94941.cf.A 15. Bourdais R, Rousseau B, Pujals A, Boussion H, Joly C, Guillemin
3. Van Cutsem E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, A, et al. Polymerase proofreading domain mutations: new
Aderka D, et al. ESMO consensus guidelines for the management of opportunities for immunotherapy in hypermutated colorectal
patients with metastatic colorectal cancer. Ann. Oncol. (2016) 27:1386– cancer beyond MMR deficiency. Crit Rev Oncol Hematol. (2017)
422. doi: 10.1093/annonc/mdw235 113:242–8. doi: 10.1016/j.critrevonc.2017.03.027
4. Venook AP, Niedzwiecki D, Lenz H-J, nnocenti F, Fruth B, Meyerhardt 16. Gong J, Wang C, Lee PP, Chu P, Fakih M. Response to PD-1 blockade in
JA, et al. Effect of first-line chemotherapy combined with cetuximab or microsatellite stable metastatic colorectal cancer harboring a pole mutation.
bevacizumab on overall survival in patients with KRAS wild-type advanced J Natl Compr Canc Netw. (2017) 15:142–7. doi: 10.6004/jnccn.2017.0016
or metastatic colorectal cancer: a randomized clinical trial. JAMA. (2017) 17. Colle R, Cohen R, Cochereau D, Duval A, Lascols O, Lopez-Trabada D, et al.
317:2392–401. doi: 10.1001/jama.2017.7105 Immunotherapy and patients treated for cancer with microsatellite instability.
5. Nobel Assembly at Karolinska Institutet. Available online at: https:// Bull Cancer. (2017) 104:42–51. doi: 10.1016/j.bulcan.2016.11.006
www.nobelprize.org/uploads/2018/10/press-medicine2018.pdf (accessed 18. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-
December, 2018). Pagès C, et al. Type, density, and location of immune cells within
6. Oiseth S, Aziz M. Cancer immunotherapy: a brief review of the history, human colorectal tumors predict clinical outcome. Science. (2006) 313:1960–
possibilities, and challenges ahead. J Cancer Metastasis Treatment. (2017) 4. doi: 10.1126/science.1129139
3:250–61. doi: 10.20517/2394-4722.2017.41 19. Calon A, Lonardo E, Berenguer-Llergo A, Espinet E, Hernando-Momblona X,
7. Chen L, Han X. Anti-PD-1/PD-L1 therapy of human cancer: past, present, Iglesias M, et al. Stromal gene expression defines poor-prognosis subtypes in
and future. J Clin Invest. (2015) 125:3384–91. doi: 10.1172/JCI80011 colorectal cancer. Nat Genet. (2015) 47:320–9. doi: 10.1038/ng.3225
8. Zinselmeyer BH, Heydari S, Sacristán C, Nayak D, Cammer M, Herz J, et al. 20. Guinney J, Dienstmann R, Wang X, de Reyniès A, Schlicker A, Soneson
PD-1 promotes immune exhaustion by inducing antiviral T cell motility C, et al. The consensus molecular subtypes of colorectal cancer. Nat Med.
paralysis. J Exp Med. (2013) 210:757–74. doi: 10.1084/jem.20121416 (2015) 21:1350–56. doi: 10.1038/nm.3967
9. de Guillebon E, Roussille P, Frouin E, Tougeron D. Anti program death-1/anti 21. Becht E, de Reyniès A, Giraldo NA, Pilati C, Buttard B, Lacroix L, et al.
program deathligand 1 in digestive cancers. World J Gastrointest Oncol. (2015) Immune and stromal classification of colorectal cancer is associated with
7:95–101. doi: 10.4251/wjgo.v7.i8.95 molecular subtypes and relevant for precision immunotherapy. Clin Cancer
10. Brahmer JR, Tykodi SS, Chow LQ, Hwu WJ, Topalian SL, Hwu P, et al. Safety Res. (2016) 22:4057–66. doi: 10.1158/1078-0432.ccr-15-2879
and activity of anti-PD-1 antibody in patients with advanced cancer. N Engl J 22. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred
Med. (2012) 366:2455–65. doi: 10.1056/NEJMoa1200694 reporting items for systematic reviews and meta-analyses: the PRISMA
11. Giannakis M, Mu XJ, Shukla SA, Qian ZR, Cohen O, Nishihara R, et al. statement. PLoS Med. (2009) 6:e1000097. doi: 10.1371/journal.pmed.1000097
Genomic correlates of immune-cell infiltrates in colorectal carcinoma. Cell 23. O’Neil BH, Wallmark JM, Lorente D, Elez E, Raimbourg J, Gomez-
Rep. (2016) 15:857–65. doi: 10.1016/j.celrep.2016.03.075 Roca C, et al. Safety and antitumor activity of the anti-PD-1 antibody
12. Turajlic S, Litchfield K, Xu H, Rosenthal R, McGranahan N, Reading pembrolizumab in patients with advanced colorectal carcinoma. PLoS ONE.
JL, et al. Insertion-and-deletion-derived tumour specific neoantigens and (2017) 12:e0189848. doi: 10.1371/journal.pone.0189848

Frontiers in Oncology | www.frontiersin.org 7 May 2019 | Volume 9 | Article 396

Oliveira et al. PD-1/PD-L1 Inhibitors in Colorectal Cancer

24. Le Dung T, Kavan P, Kim TW, Burge ME, Cutsem EV, Hara H, cancer. Ann Oncol. (2018) 29:mdy208-003. doi: 10.1093/annonc/mdy
et al. KEYNOTE-164: pembrolizumab for patients with advanced 208.003
microsatellite instability high (MSI-H) colorectal cancer. J Clin Oncol. 33. Duffy AG, Makarova-Rusher OV, Fioravanti S, Walker M, Venkatesan A,
(2018) 36:3514. doi: 10.1200/jco.2018.36.15_suppl.3514 Abi-Jaoudeh N, et al. A pilot study of AMP-224, a PD-L2 Fc fusion
25. Koustas E, Papavassiliou AG, Karamouzis MV. KEYNOTE-177: protein, in combination with stereotactic body radiation therapy (SBRT) in
Phase 3, open-label, randomized study of first-line pembrolizumab patients with metastatic colorectal cancer. J Clin Oncol. (2016) 34 (Suppl.
(Pembro) versus investigator-choice chemotherapy for mismatch S4):560. doi: 10.1200/jco.2016.34.4_suppl.560
repair-deficient (dMMR) or microsatellite instability-high (MSI- 34. Weber J, Mandala M, Del Vecchio M, Gogas HJ, Arance AM, Cowey CL, et al.
H) metastatic colorectal carcinoma (mCRC). J Clin Oncol. (2018) Adjuvant nivolumab versus ipilimumab in resected stage III or IV melanoma.
36:TPS877. doi: 10.1200/jco.2018.36.4_suppl.tps877 N Engl J Med. (2017) 377:1824–35. doi: 10.1056/NEJMoa1709030
26. Overman MJ, McDermott R, Leach JL, Lonardi S, Lenz HJ, Morse MA, 35. Domingo E, Freeman-Mills L, Rayner E, Glaire M, Briggs S, Vermeulen L,
et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient et al. Somatic POLE proofreading domain mutation, immune response, and
or microsatellite instability-high colorectal cancer (CheckMate 142): an prognosis in colorectal cancer: a retrospective, pooled biomarker study. Lancet
open-label, multicentre, phase 2 study. Lancet Oncol. (2017) 18:1182– Gastroenterol Hepatol. (2016) 1:207–16. doi: 10.1016/S2468-1253(16)30014-0
91. doi: 10.1016/s1470-2045(17)30422-9 36. Sivan A, Corrales L, Hubert N, Williams JB, Aquino-Michaels K,
27. Lee JM, Sun W, Bahary N, Ohr J, Rhee JC, Stoller RG, et al. Earley ZM, et al. Commensal Bifidobacterium promotes antitumor
Phase 2 study of pembrolizumab in combination with azacitidine immunity and facilitates anti-PD-L1 efficacy. Science. (2015)
in subjects with metastatic colorectal cancer. J Clin Oncol. (2017) 350:1084–89. doi: 10.1126/science.aac4255
35:3054. doi: 10.1200/jco.2017.35.15_suppl.3054 37. Emens LA, Ascierto PA, Darcy PK, Demaria S, Eggermont AMM,
28. Shahda S, Noonan AM, Bekaii-Saab TS, O’Neil BH, Sehdev A, Shaib Redmond WL, et al. Cancer immunotherapy: opportunities and challenges
WL, et al. A phase II study of pembrolizumab in combination with in the rapidly evolving clinical landscape. Eur J Cancer. (2017) 81:116–
mFOLFOX6 for patients with advanced colorectal cancer. J Clin Oncol. (2017) 29. doi: 10.1016/j.ejca.2017.01.035
35:3541. doi: 10.1200/jco.2017.35.15_suppl.3541 38. Bello DM, Panageas KS, Hollmann TJ, Shoushtari AN, Chapman P, Postow
29. Overman MJ, Lonardi S, Wong KYM, Lenz HJ, Gelsomino F, Aglietta MA, et al. Outcomes of patients with metastatic melanoma selected for surgery
M, et al. Durable clinical benefit with nivolumab plus ipilimumab in after immunotherapy. In: 2018 Society of Surgical Oncology Annual Cancer
DNA mismatch repair-deficient/microsatellite instability-high metastatic Symposium, Vol. 25. New York, NY: Springer (2018). p. S7.
colorectal cancer. J Clin Oncol. (2018) 36:773–9. doi: 10.1200/jco.2017. 39. Virik K, Wilson RB. 1223P Potential drug cost impact of dual agent
76.9901 immunotherapy (DAIO) with nivolumab (N) plus ipilimumab (I) in
30. Lenz HJ, van Cutsem E, Limon ML, Wong KY, Hendlisz A, Aglietta patients with DNA mismatch repair deficient (dMMR) metastatic colorectal
M, et al. LBA18_PR Durable clinical benefit with nivolumab cancer (mCRC) in Canada. Ann Oncol. (2018) 29 (Suppl. 8):mdy288-
(NIVO) plus low-dose ipilimumab (IPI) as first-line therapy in 094. doi: 10.1093/annonc/mdy288.094
microsatellite instability-high/mismatch repair deficient (MSI-H/dMMR)
metastatic colorectal cancer (mCRC). Ann Oncol. (2018) 29 (Suppl. Conflict of Interest Statement: The authors declare that the research was
8):mdy424-019. doi: 10.1093/annonc/mdy424.019 conducted in the absence of any commercial or financial relationships that could
31. Bendell JC, Powderly JD, Lieu CH, Eckhardt SG, Hurwitz H, be construed as a potential conflict of interest.
Hochster HS, et al. Safety and efficacy of MPDL3280A (anti-PDL1)
in combination with bevacizumab (bev) and/or FOLFOX in patients Copyright © 2019 Oliveira, Bretes and Furtado. This is an open-access article
(pts) with metastatic colorectal cancer (mCRC). J Clin Oncol. (2015) distributed under the terms of the Creative Commons Attribution License (CC BY).
33:704. doi: 10.1200/jco.2015.33.3_suppl.704 The use, distribution or reproduction in other forums is permitted, provided the
32. Bendell J, Ciardiello F, Tabernero J, Tebbutt N, Eng C, Di Bartolomeo M, original author(s) and the copyright owner(s) are credited and that the original
et al. LBA-004 Efficacy and safety results from IMblaze370, a randomised publication in this journal is cited, in accordance with accepted academic practice.
Phase III study comparing atezolizumab+ cobimetinib and atezolizumab No use, distribution or reproduction is permitted which does not comply with these
monotherapy vs regorafenib in chemotherapy-refractory metastatic colorectal terms.

Frontiers in Oncology | www.frontiersin.org 8 May 2019 | Volume 9 | Article 396