Edited by: Benjamin Frey, University of Erlangen Nuremberg, Germany
Reviewed by: Umberto Malapelle, University of Naples Federico II, Italy; Frederique Vegran, INSERM U1231 Lipides, Nutrition, Cancer (LNC), France; Vincent Bourbonne, Centre Hospitalier Regional Universitaire (CHU) de Brest, France
This article was submitted to Cancer Immunity and Immunotherapy, a section of the journal Frontiers in Oncology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Esophageal cancer (EC) is one of the most common cancers with poor survival in the world. Nowadays, a generous number of clinical trials are underway on the use of immunotherapy in EC patients, especially the programmed death-1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors. However, only a few patients could benefit from single-agent therapy. Others need combination therapies to enhance the response rate and survival. In this review, we focus on PD-1/PD-L1 inhibitors and its combination options in EC patients. We also summarized the potential predictive biomarkers for PD-1/PD-L1 inhibitors treatment.
Esophageal cancer (EC) is the seven most common cancer and ranks second in the cause of cancer-related death worldwide (
Traditional therapies for EC patients include surgery, chemotherapy, radiotherapy (RT), and targeted therapy. However, most patients relapse quickly after the initial therapy. Meanwhile, EC patients always have a lack of oncogenic driver mutations (
In recent clinical trials of EC, interest is very high in immunotherapies, which involved immune checkpoint inhibitors (ICIs), adoptive T-cell therapy, cancer vaccines, and oncolytic viruses. Immunotherapies act on different steps of the anti-tumor immunity to enhance the host's immunity and strengthen anti-tumor responses. A series of events occurred in steps to eliminate cancer cells and were identified as the cancer-immunity cycle (
ICIs, especially programmed death-1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors, have demonstrated clinical benefit in multiple cancers and generated tremendous interest. Immune checkpoints are immunosuppressive proteins and help to maintain immunologic homeostasis in hosts (
We summarize the clinical studies of PD-1/PD-L1 inhibitors and the combination options for EC in this review.
PD-L1 overexpression has been reported in around 40% of EC patients and is related to worse OS (
Furthermore, combined positive score (CPS) is used to define the PD-L1 status instead of the tumor proportion score (TPS) for EC patients. CPS is defined as the ratio of the combining number of PD-L1 positive tumor cells and immune cells (lymphocytes, macrophages) by IHC staining to the total number of tumor cells. The maximum score is 100, and a higher score showing greater likelihood of response to PD-1/PD-L1 inhibitors (
The most studied PD-1/PD-L1 inhibitors were Nivolumab (OPDIVO, Bristol-Myers Squibb Co.), Pembrolizumab (KEYTRUDA, Merck Sharp & Dohme Co., Inc.), Avelumab (BAVENCIO, EMD Serono Inc.), Atezolizumab (TECENTRIQ, Genentech Inc.), and Durvalumab (IMFINZI, AstraZeneca Inc.). Plenty of new PD-1/PD-L1 inhibitors are being investigated now, including PD-1 inhibitors: Camrelizumab (SHR-1210, Jiangsu HengRui Medicine Co., Ltd.), Sintilimab [IBI308, Innovent Biologics (Suzhou) Co. Ltd.], Spartalizumab (PDR001, Novartis Pharmaceuticals), Tislelizumab (BGB-A317, BeiGene), Toripalimab (triprizumab, teripalimab, JS001, Shanghai Junshi Biosciences Co., Ltd), HLX-10 (Shanghai Henlius Biotech, Inc.) and PD-L1 inhibitor: SHR-1316 (Jiangsu HengRui Medicine Co., Ltd.) and CS1001 (CStone Pharmaceuticals Co., Ltd.).
ICIs have shown considerable objective response rates (ORR), durable toxicities and even prolong the OS in several cancers, including advanced EC. However, the effective rates of single-agent were 31% in melanoma and only 17% in lung cancer, respectively (
We searched the
The combination with PD-1 inhibitors in EC.
Nivolumab | NCT03044613 (CRT + Relatlimab) |
NCT02743494 |
NCT3143153 (Ipilimumab/chemo) |
NCT02476123 (Mogamulizumab) |
Pembrolizumab | NCT02844075 (CRT) |
NCT02844075 |
NCT03189719 (Chemo) |
NCT02013154 (DKN-01) |
Camrelizumab | NCT03200691 (RT) |
NCT03817658 (After CRT) |
NCT03187314 (RT) |
NCT03736863 (Apatinib) |
Sintilimab | NCT03940001 (CRT) |
NCT03748134 (Chemo) | ||
Spartalizumab | NCT01351103 (LGK974) |
|||
Tislelizumab | NCT03469557 (Chemo) |
|||
Toripalimab | NCT03985670 (Chemo) |
NCT03829969 (Chemo) |
||
HLX-10 | NCT03958890 (Chemo) |
The combination with PD-L1 inhibitors in EC.
Avelumab | NCT03490292 (CRT) | NCT03490292 (CRT) |
||
Atezolizumab | NCT03087864 (CRT) |
UMIN000034373 (After CRT) | NCT03087864 (CRT) | NCT03170960 (Cabozantinib) |
Durvalumab | NCT02735239 (CRT/Chemo) |
NCT02639065 (After CRT + surgery) |
NCT02658214 (Chemo + Tremelimumab) |
NCT02735239(Chemo ± Tremelimumab) |
SHR-1316 | NCT03732508 (Chemo) | NCT03766178 (Nimotuzumab) |
The CROSS study showed that neoadjuvant concurrent CRT induced 23% pathologic complete response (pCR) and prolong median overall survival (mOS) (49 vs. 24 months; hazard ratio (HR) = 0.657, 95% confidence interval (CI): 0.495–0.871,
According to preclinical studies of EC, RT could induce immunogenic cell death (ICD), consequently release neoantigens, alter tumor microenvironment (TME), and finally activate the immune response (
The published data of neoadjuvant or adjuvant use of PD-1/PD-L1 inhibitors.
Nivoluzumab | NCT03044613 | Neoadjuvant | 10 | EAC | Paclitaxel + carboplatin, q1w | RT | 40.0% | 0 |
Pembrolizumab | NCT02844075 | Neoadjuvant and adjuvant | 26 | ESCC | Paclitaxel + carboplatin, q1w | 44.1 Gy/21 fr | 46.1% | – |
Avelumab | NCT03490292 | Neoadjuvant | 6 | EAC | Paclitaxel + carboplatin, q1w | 41.4 Gy/23 fr | 43.0% | – |
Atezolizumab | NCT03087864 | Neoadjuvant | 23 | EAC | Paclitaxel + carboplatin, q1w | 41.4 Gy/23 fr | 39% | – |
Atezolizumab | UMIN000034373 | Adjuvant | 56 | ESCC | Cisplatin + fluoracil, q4w, two cycles | 60 Gy/30 fr | – | – |
Durvalumab | NCT02639065 | Adjuvant | 24 | EAC or GEJ | Carboplatin + paclitaxel/cisplatin + fluoropyrimidine, surgery | RT | – | 1 (grade 3) |
Nivolumab is a humanized IgG4 monoclonal PD-1 antibody. Its neoadjuvant role was assessed in the trial NCT03044613. This study recruited 16 EC patients to receive two cycles of induction nivolumab before CRT and three additional cycles of nivolumab concurrently with CRT (
Pembrolizumab is another humanized IgG4 monoclonal PD-1 antibody. The trial NCT02844075 enrolled 28 ESCC patients to receive neoadjuvant CRT plus pembrolizumab (
As for PD-L1 inhibitors, the trial NCT03490292 tested the safety and efficacy of avelumab with CRT in esophageal or EGJ adenocarcinoma patients (
In another phase II study, PERFRCT trial (NCT03087864), resectable EAC patients were enrolled to receive five cycles of concurrent CRT and atezolizumab before surgery (
The value of postoperative chemotherapy in resectable esophageal and EGJ cancers remains uncertain in the previous trials (
For unresectable locally advanced ESCC, the definitive CRT and observation after that is the standard treatment. However, the complete response (CR) rate is only 11 to 25%, 1-year relapse-free survival (RFS) rate is 50.0% and mOS is only 9–10 months (
The adjuvant role of PD-1/PD-L1 inhibitors in EC have been reported in the 2019 ASCO meeting. A phase II trial, NCT02639065 was designed for patients with resected locally advanced esophageal or EGJ adenocarcinoma who had a viable tumor in the surgical specimen after neoadjuvant CRT and R0 resection (
In the TENERGY trial (UMIN000034373), unresectable locally advanced ESCC patients without distant metastasis were enrolled and treated with atezolizumab for up to 12 months within 4 weeks after two cycles definitive CRT (
Based on the published data, the addition of PD-1/PD-L1inhibitors as neoadjuvant or adjuvant treatment demonstrated promising efficacy with acceptable toxicity. The trials are ongoing with camrelizumab, sintilimab, and toripalimab as showed in
First-line platinum-based doublet chemotherapy provides a limited survival benefit in advanced ESCC patients. To gain better survival, an effective combination with other therapy is urgently required.
Several studies evaluated the efficacy of PD-1/PD-L1 inhibitors plus RT as first-line treatment in advanced EC patients. As mentioned before, RT could enhance the anti-tumor immunity and induce a synergistic effect with PD-1/PD-L1 inhibitors.
The combination of RT with camrelizumab has been tested in a phase II, single-arm study for patients with locally advanced EC intolerant to or refused CRT (NCT03187314) (
Another phase Ib trial NCT03222440 evaluated camrelizumab with RT as first-line therapy in 20 ESCC patients and observed two (11.1%) patients had CR and 13 (72.2%) patients with PR (
Chemotherapy could also facilitate the anti-tumor response. When combined with immunotherapy, chemotherapy could promote the presentation of tumor antigen, enhance the filtration of CTL and improve the efficacy of checkpoint inhibition (
The phase II trial, NCT03469557 evaluated the tolerability of tislelizumab combined with chemotherapy (cisplatin and fluorouracil) as first-line treatment in Chinese patients with inoperable, locally advanced ESCC (
The phase III study KEYNOTE-590 (NCT03189719, MK-3475-590) of pembrolizumab plus chemotherapy (cisplatin and fluorouracil) as first-line therapy for advanced or metastatic EC is ongoing. Its China Extension study, NCT03881111 is also underway now.
Since the different mechanisms of ICIs, the combination of PD-1/PD-L1 inhibitors with other ICIs may be significant. Cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) on T cells interact with peripheral membrane B7 on APCs and impede the activation of T cells (
The phase I trial of NCT02658214 evaluated the safety of durvalumab and tremelimumab in combination with chemotherapy (cisplatin and fluorouracil) as first-line treatment for advanced ESCC patients (
The trial CheckMate 648 (NCT03143153) was a randomized, phase III study to compare nivolumab plus ipilimumab or nivolumab combining with chemotherapy (cisplatin and fluorouracil) vs. chemotherapy alone as first-line treatment in unresectable, recurrent, or metastatic ESCC patients. Future results will provide more safety and efficacy data on the combination of PD-1/PD-L1 inhibitors and CTLA-4 inhibitors.
NCT03603756 was a phase II study of camrelizumab combined with apatinib [a vascular endothelial growth factor receptor 2 (VEGFR2) inhibitor] and chemotherapy (liposomal paclitaxel and nedaplatin) as the first-line treatment for advanced ESCC (
For patients who failed in the treatment of standard platinum-based doublet chemotherapy, second-line, and subsequent treatment options are limited. The 5-year survival rate was ~20% in all stages and only 5% in advanced patients (
The published data of PD-1/PD-L1 inhibitors in advanced EC.
Nivolumab | ATTRACTION-1 | II, single-arm, second-line | 65 | Japanese ESCC | 17.2% | 11.17 | 1.5 | 10.8 |
ATTRACTION-3 | III, randomized controlled, second-line | 210 | ESCC | 19.0% | 6.9 | 1.7 | 10.9 | |
Pembrolizumab | KEYNOTE-028 | II, single-arm, second-line or subsequent | 23 | EC | 30% | 15.0 | 1.8 | 7.0 |
18:5 | ESCC vs. EAC | 28%vs. 40% | – | – | – | |||
KEYNOTE-180 | II, single-arm, second-line or subsequent | 121 | EC | 9.9% | Not reached | 2.0 | 5.8 | |
63:58 | ESCC vs. EAC | 14.3 vs. 5.2% | Not reached | – | – | |||
58:63 | PD-L1 (+) vs. (–) | 13.8 vs. 6.3% | Not reached | – | – | |||
KEYNOTE-181 | III, randomized controlled, second-line | – | EC | – | – | – | 7.1 | |
– | ESCC | 16.7% | 9.3 | – | 8.2 | |||
107 | PD-L1 (+) | 21.5% | 9.3 | 2.6 | 9.3 | |||
85:22 | PD-L1 (+) ESCC vs. EAC | 22.0 vs. 18.0% | 9.3 vs. Not reached | 3.2 vs. 2.1 | 10.3 vs. 6.3 | |||
62 | Chinese EC | 16.1% | Not reached | – | 8.4 | |||
Camrelizumab | NCT02742935 | I, single-arm, second-line or subsequent | 30 | Chinese ESCC | 33.3% | – | 3.6 | – |
NCT03099382 | III, randomized controlled, second-line | 228 | Chinese ESCC | 20.2% | – | – | 8.3 | |
Toripalimab | NCT02915432 | I/II, single-arm, second-line or subsequent | 59 | Chinese ESCC | 18.6% | 11.2 | – | – |
ATTRACTION-1 (ONO-4538-07/JapicCTI-No.142422) study was a single-arm, phase II study enrolled 65 Japanese ESCC patients with unresectable or recurrent EC and were refractory to or intolerant to standard chemotherapy (fluoropyrimidine and platinum) (
Furthermore, the randomized, phase III trial, ATTRACTION-3 (NCT02569242, ONO-4538-24/CA209-473) is performing to compare nivolumab with chemotherapy (docetaxel or paclitaxel) as second-line therapy in unresectable or recurrent ESCC patients with PD-L1-unselected (
KEYNOTE-028 (NCT02054806) was a single-arm, phase Ib trial that enrolled PD-L1 positive (CPS > 1) locally advanced or metastatic EC patients to receive pembrolizumab (
KEYNOTE-180 (NCT02559687) was a single-arm, phase II study which enrolled patients with locally advanced or metastatic ESCC and EAC (including Siewert type I EGJ adenocarcinoma) who refractory to at least two prior systemic treatments (
Since pembrolizumab was certified as an effective and safe third-line treatment in the trial KEYNOTE-028 and KEYNOTE-180, the trial KEYNOTE-181 (NCT02564263) evaluated its upfront use as second-line treatment. The trial enrolled 628 patients with locally advanced or metastatic EC who progressed on or after the standard chemotherapy (
Camrelizumab is a novel humanized high-affinity IgG4-kappa PD-1 monoclonal antibody that independently developed by Chinese biopharma. In the phase I study, NCT02742935, 30 ESCC patients failed to at least one systemic treatment were enrolled (
The phase III trial, ESCORT study (NCT03099382), compared camrelizumab and chemotherapy (docetaxel or irinotecan) as the second-line treatment for advanced ESCC. The latest results were reported at the 15th OESO World Conference as an oral presentation (
Toripalimab is another humanized PD-1 monoclonal antibody developed by Chinese biopharma. The latest data of a phase Ib/II trial, NCT02915432, was presented on the 2019 ASCO meeting (
Based on these studies, pembrolizumab has been approved by the Food and Drug Administration (FDA) as the second-line treatment for recurrent, locally advanced or metastatic ESCC with PD-L1 positive (CPS ≥ 10) (
To boost the efficacy of PD-1/PD-L1 inhibitors in EC patients, we concentrate on its combination with different immunoregulatory factors to activate anti-tumor immunity (
The PD-1/PD-L1 inhibitors combined with immunoregulatory factors.
CTLA-4 inhibitor | Ipilimumab | NCT03416244 |
LAG-3 inhibitor | Relatlimab, LAG525 | NCT03044613, NCT02460224 |
OX40 agonist | INCAGN01949 | NCT03241173 |
GITR agonist | INCAGN01876 | NCT03277352 |
ICOS inhibitor | KY1044 | NCT03829501 |
IDO1 inhibitor | Epacadostat | NCT03592407, NCT03277352 |
CCR-4 inhibitor | Mogamulizumab | NCT02476123, NCT02946671 |
M-CSF inhibitor | MCS110 | NCT03785496 |
WNT inhibitor | LGK974 | NCT01351103 |
DKK-1 inhibitor | DKN-01 | NCT02013154, NCT03818997 |
PDE-5 inhibitor | Tadalafil | NCT03993353 |
SHP2 inhibitor | TNO155 | NCT04000529 |
Lymphocyte activation gene-3 (LAG-3, CD223) is regularly expressed on activated T cells and natural killer (NK) cells. The expression and upregulation of LAG-3 and PD-1 on tumor-infiltrating lymphocytes (TILs) leading to the inactivate of effector T cells and cause tumor growth (
OX40 (CD134) is a T cell co-stimulatory receptor and mainly expressed on activated T cells, and regulatory T (Treg) cells (
Indoleamine 2.3-dioxygenase-1 (IDO1) is an intracellular enzyme and overexpressed in various cancers. As a metabolic mediator, IDO1 enzyme transfer tryptophan to tryptophan catabolites (
There are also novel multi-target antibodies, including SL-279252 (PD-1 and OX40L inhibitor) and INBRX-105 [PD-L1 and 41BB (CD137) inhibitor]. The safety and efficacy of these agents may be reported in forthcoming studies.
Considering the low effective rate of ICI in EC, predictive biomarkers are needed to decide patients more likely to react to ICI and identify patients resistant to ICI and need a combination or alter treatment. Predictive biomarkers identified in NSCLC including PD-L1 status, tumor mutation burden (TMB), mismatch repair deficiency (MMR), and microsatellite instability (MSI). Whether they make a similar role in EC is uncertain.
As mentioned before, ESCC is more widespread in Asia, and clinical trials reveal higher ORRs in ESCC patients with PD-1/PD-L1 inhibitors compared with EAC. ATTRACTION-1 reported an ORR of 17% (95% CI: 10–28%) in 65 advanced Japanese ESCC patients treated with nivolumab (
Regarding the better response to PD-1/PD-L1 in ESCC, Asian patients may be the more beneficial population. The latest results of Chinese patients in KEYNOTE-181 were presented on the 2019 ESMO meeting (
Thus, Asian and non-Asian patients may have different efficacy even within ESCC. Further analysis is needed to help draw an accurate conclusion.
The predictive role of PD-L1 is still controversial in EC. Some studies believe that PD-L1 expression in tumor and immune cells is associated with the efficacy of PD-1/PD-L1 inhibitors (
In some PD-L1 positive patients, the efficacy of PD-1/PD-L1 inhibitors is still low. Besides the heterogeneity of PD-L1 expression, TILs may be the possible reason. TILs consist of a group of heterogenous lymphocytes that infiltrate the tumor and participated in anti-tumor response. The high level of TILs in TME is correlated to better survival in patients with EC (
The relationship of PD-L1 status and the efficacy of PD-1/PD-L1 inhibitors remains uncertain, and the function of other factors in TME still needed to be considered and provide more evidence in the future.
TMB is the number of non-synonymous somatic gene mutations (Mb) of sequenced DNA and higher TMB tumors are likely to produce more neoantigens, induce a specific T cell response, and further enhance the anti-tumor immunity. High TMB correlates with clinical benefit from ICIs in patients with melanoma and NSCLC (
The number of TMB varies in different cancers. TMB is low in EC according to the reports of patients from western countries. (
Mismatch repair genes are genes that replace nucleotides incorrectly incorporated during DNA replication. Deficient DNA mismatch repair (dMMR) means a lack of these genes and produce a lot of short repeated sequences in the DNA (microsatellite) and more tumor-specific mutation (higher TMB) (
Despite the potential biomarkers mentioned above, many different predictive biomarkers are also studied now. The trial NCT02915432 analyzed the amplification of the chromosome 11q13 region in ESSC patients received toripalimab. Forty-eight percentage (24/50) patients had 11q13 amplification, which resulted in elevated mRNA expression of corresponding genes, including Cyclin D1 (CCND1) and fibroblast growth factor family members (FGF3/4/19) (
In another trial about camrelizumab, ESCC patients with an increased baseline lactate dehydrogenase (LDH) had lower ORR (
Currently, the predictive role of a single biomarker is limited, combined prediction models of multiple biomarkers may be available in the future.
Based on the previous results, PD-1/PD-L1 inhibitors were durable and effective in EC, though many questions remain unanswered. Firstly, most trials are single-arm designed. More randomized controlled trials are demanded to compare the efficacy of PD-1/PD-L1 inhibitors and control treatment. Secondly, the control treatment in the current studies was chemotherapy alone rather than other more effective therapies, such as CRT. Thirdly, the response rates of PD-1/PD-L1 inhibitors alone were limited. Its combination with chemotherapy, RT, targeted drugs or other immune modulates may improve the anti-tumor activity. It is extremely important to identify patients who most likely gain clinical benefit from PD-1/PD-L1 inhibitors. More predictive biomarkers are investigated to refine the optimal patient for single-agent treatment and those require combination therapies. We also include the AEs of PD-1/PD-L1 inhibitor alone or combined with others, especially the incidence of pneumonitis.
HY and LY designed the study. HY collected data of clinical trials and drafted the manuscript together with KW. TW, ML, BL, and SL coordinated, edited, and completed the drafting of the manuscript. LY revised and edited the final version of the manuscript for important intellectual content. All authors read and approved the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
1
2
3
4