Identification of effective therapies for colorectal cancer (CRC) remains an urgent medical need, especially for the microsatellite-stable (MSS) phenotype. In the current study, a combination of fruquintinib plus anti–PD-1 for MSS CRC therapy was investigated. First, a case of advanced MSS CRC was reported. After failure of multiline therapy, the patient finally achieved rapid response after receiving fruquintinib plus anti–PD-1 treatment. Then the effect of fruquintinib plus anti–PD-1 was verified using a murine syngeneic model of CT26 cells (MSS). The results showed that cotreatment significantly inhibited tumor growth and promote survival time for tumor-bearing mice compared with the single drug alone. In addition, fruquintinib/anti–PD-1 cotreatment decreased angiogenesis, enhanced normalization of the vascular structure, and alleviated tumor hypoxia. Moreover, the combination therapy reprogrammed the immune microenvironment by enhancing chemotactic factor release, increasing CD8+ T cell infiltration and activation, decreasing ration of regulatory T cells, and promoting M1/M2 ratio of macrophage. Finally, the enhanced antitumor effect of fruquintinib/anti–PD-1 cotreatment was significantly reversed in CD8 knockout mice compared with that in the wild-type mice. Our study indicated that combination of fruquintinib and anti–PD-1 could synergistically suppress CRC progression and altered the tumor microenvironment in favor of antitumor immune responses.

This article is featured in Top Reads, p.2555

Colorectal cancer (CRC) is the third-most common cancer worldwide and the main leading cause of death in both men and women (1). Treatments based on cytotoxic agents, including irinotecan, oxaliplatin, and fluorouracil, are still the first- or second-line treatment for metastatic CRC (mCRC) patients. In addition, biological agents targeting vascular endothelial growth factor (VEGF) or its receptor (VEGFR) and epidermal growth factor receptor (EGFR) are also available, such as bevacizumab, regorafenib, and cetuximab (2, 3). Even so, alternative therapeutic options for advanced diseases are still scarce. Therapy with immune checkpoint inhibitors, represented by anti–PD-1, has resulted in remarkable success in the treatment of various hematological and solid metastatic malignancies (47). However, in mCRC, the clinical effect of anti–PD-1 has been generally limited to tumors with the microsatellite instability–high phenotype, which accounts for 4–6% of mCRC (8). There is room for treatment development of the microsatellite stability (MSS) type, which accounts for the majority (9) and is thought to be a crucial clinical issue related to PD-1 inhibitors.

VEGF/VEGFR plays a well-characterized role in angiogenesis, and it is believed to be involved in cancer immune evasion (10). In recent years, studies have shown that antiangiogenic drugs in combination with anti–PD-1/PD-L1 therapy enhanced antitumor T cell migration and tumor growth inhibition (TGI) with a favorable toxicity profile in many cancers, such as advanced renal cell cancer (11), non–small cell lung cancer (12), and hepatocellular cancer (13). A phase 2 clinical study conducted by V. Makker showed that the proportion of objective responses in endometrial cancer patients treated with lenvatinib and pembrolizumab was higher than that in those treated with lenvatinib or pembrolizumab monotherapy, regardless of the tumor microsatellite status (14).

Fruquintinib, a new blocker of VEGFR, has been approved by the China Food and Drug Administration for the treatment of mCRC patients who have undergone at least two prior standard anticancer therapies (15). Fruquintinib, in combination with other chemotherapeutic agents, was reported to exert enhanced antitumor activity with favorable tolerance in patient-derived xenograft models (16). However, whether combination with fruquintinib is a potential therapeutic option that provides additional benefits to anti–PD-1/PD-L1 treatment for MSS CRC has not been proven in a clinical study.

In this study, we report a patient with MSS CRC who progressed after multiline therapy but achieved rapid response after receiving treatment with fruquintinib plus anti–PD-1. We then established a syngeneic model of MSS CRC and found that combination of fruquintinib and anti–PD-1 synergistically promoted the antineoplastic effect of the treatment and reprogrammed the immunosuppressive tumor microenvironment (TME).

Fruquintinib was a gift from Hutchison MediPharma Limited (Shanghai, China) and anti-mouse PD-1 Ab (CD279) (clone RMP1-14) was purchased from Bio X Cell (West Lebanon, NH). Sodium carboxymethyl cellulose (CMC.Na) used in mice was provided by Solarbio Science and Technology (Beijing, China). TUNEL assay kit was bought from Vazyme Biotech (Nanjing, China). mAbs of PCNA (2586), IFN-γ (8455), CD31 (3528), PD-L1 (13684), c-myc (9402), Caspase-3 (9662), cyclin D1 (2922), HIF-1α (36169), and α-SMA (19245) used for immunohistochemistry (IHC) and immunofluorescence analysis were purchased from Cell Signaling Technology (Danvers, MA); mAbs of CD45 (30-F11), CD4 (RM4-5), CD8 (53-6, 7), IFN-γ (XMG1.2), and Foxp3 (FJK-165) for flow cytometry were purchased from Thermo Fisher Scientific (Waltham, MA). mAbs of CD45 (30-F11), CD8a (B310527), and Arg1 (17-3697-80) for flow cytometry were purchased from eBioscience (Worcester, MA). mAbs of MHC class II (B307308), and Granzyme B (B256443) for flow cytometry were purchased from BioLegend (San Diego, CA). mAbs of CD206 (FAB741P) for flow cytometry were purchased from R&D Systems (Minneapolis, MN). mAbs of CD86 (553692), F4/80 (123108), and Foxp3 (560414) for flow cytometry were purchased from BD Biosciences (San Jose, CA).

CT26 and MC38 mouse colon carcinoma cell lines were obtained from Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Cells were cultivated in RPMI 1640 culture medium supplemented with 10% FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin and were cultured in a humidified 5% CO2 atmosphere at 37°C in incubator.

Six to eight weeks old, BALB/c and C57BL/6 mice were purchased from Nanjing Tande biotechnology (Nanjing, China). CD8+ T knockout (KO) mice with C57BL/6 background were a gift from Prof. L. Ye, Third Military Medical University. They were housed with free access to pellet food and water in plastic cages at 21 ± 2°C and kept on a 12-h light/dark cycle. Animal welfare and experimental procedures were carried out in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health) and the related ethical regulations of our university. All experimental protocols were approved by the Ethics Committee of Nanjing Medical University. All efforts were made to reduce the number of animals used and to minimize animals’ suffering.

CT26 cells (6 × 105) were inoculated at the right flank of BALB/c mice. After the tumor reached 50 mm3, mice were randomized to four groups (n = 8 per group), and treatments were initiated as follows: group 1, mice were administered a daily oral gavage with 5% CMC.Na (vehicle); group 2, mice were administered a daily oral gavage with fruquintinib at 1.0 mg/kg; group 3, mice were administered with anti-mouse PD-1 at 5 mg/kg by i.p. injection every 3 d; and group 4, mice were administered with fruquintinib plus anti-mouse PD-1 Ab. MC38 cells (6 × 105) were inoculated at the flank of CD8 wild-type (WT) and CD8 KO C57BL/6 mice and were randomized to four groups separately: group 1, CD8 WT mice were administered a daily oral gavage with 5% CMC.Na (vehicle); group 2, CD8 WT mice were administered with fruquintinib (1 mg/kg, intragastric administration) plus anti–PD-1 (5 mg/kg by i.p. injection every 3 d); group 3, CD8 KO mice were administered a daily oral gavage with 5% CMC.Na (vehicle); and group 4, CD8 KO mice were administered with fruquintinib (1 mg/kg, intragastric administration) plus anti–PD-1 (5 mg/kg by i.p. injection every 3 d). Drugs were administered conforming to animal ethics. Body weight and tumor volume (TV) were measured every 2 d after treatment begin. TV was determined by measuring the largest diameter (a) and its perpendicular (b) according to the formula (a × b2)/2. The TGI (%TGI = 100 × [1 − (TV final − TV initial for drug treated group)/(TV final − TV initial for control group)]) was used for evaluation of antineoplastic effect. On the 15th day, mice were euthanized, and tumors were removed by scissors. The weight was measured by electronic balance in wet, and tumor sections were fixed in formalin. The rest of the sections were frozen in liquid nitrogen and stored at −80°C.

Tumor tissues were stained with H&E as per standard protocols and analyzed by a pathologist under light microscope (Olympus). For all staining protocols, the sections were first deparaffinized, rehydrated, and washed in 1% PBS–Tween. For IHC, the sections were treated with 2% hydrogen peroxide to block endogenous peroxidases, blocked with 3% goat serum, and incubated with specific primary Abs overnight at 4°C. The sections were then incubated with streptavidin–HRP for 40 min, stained using diaminobenzidine substrate, and counter-stained with hematoxylin. For immunofluorescence and TUNEL assay, the slides were stained with fluorescent labeled Abs or TUNEL-FITC (1:100), respectively, and then counter-stained with DAPI for 5 min. Images were acquired by confocal laser-scanning microscope (LSM 880; ZEISS) without z-stacks.

Spleen tissues from s.c. CRC model were harvested on the days indicated. Single-cell suspensions were yielded via mechanical dissociation from the collected tissues and were passed through a 70-μm cell strainer and washed twice with PBS/0.5% BSA. To study the infiltration of lymphoid and myeloid cells in the tumor tissue, mice were sacrificed 14 d after randomization/treatment, and tumors were collected. After dissection, tumors were mechanically and enzymatically dissociated using a mouse tumor dissociation kit, according to manufacturer’s recommendations (130-096-730; Miltenyi Biotec). Single-cell suspensions were stained with surface mAb: PE-Cy7/AF700-conjugated anti-CD45, BV421-conjugated anti-CD4/MHC class II (MHC-II), FITC-conjugated anti-CD8a, FITC-conjugated F4/80, and PE-conjugated CD86/CD206. For the intracellular markers, cells were incubated in 24-well, flat-bottom plates with cell stimulation mixture (eBioscience/Thermo Fisher Scientific, Waltham, MA) for 4 h under a humidified 5% CO2 atmosphere at 37°C in incubator and fixed and permeabilized with the FoxP3/Transcription Factor Staining Buffer Set (eBioscience/Thermo Fisher Scientific) according to the manufacturers’ protocols, then stained with intracellular markers: PE-conjugated Foxp3/IFN-γ, PE-Cy7–conjugated granzyme B, and APC-conjugated Arg1. For the lymphoid and myeloid cell infiltration assay in tumor tissues, viability dye eFluor 780 (65-0865-14; eBioscience) was used to identify live cells.

Tumors tissues were mechanically and enzymatically dissociated using a mouse tumor dissociation kit (130-096-730; Miltenyi Biotec) followed with Percoll isolation. The cell suspension was then filtered through a 70-mm nylon mesh, layered on a Percoll gradient (40–70%), and centrifuged at 800 × g for 20 min. The separated tumor-infiltrating lymphocyte fraction was washed twice before use. CD45+ cells represent the total leukocyte population.

The cytokines in the tumor tissues were measured by ELISA according to manufacturer’s recommendations. The CXCL9, CXCL10, IFN-γ, and TGF-β ELISA kits were obtained from MultiSciences Biotech (Hangzhou, China).

Total RNA was isolated using TRIzol reagent (Invitrogen), and cDNA was synthesized using the PrimeScript RT Reagent Kits (Takara) according to the manufacturers’ instructions as previous reported. The quantitative RT-PCR was performed with SYBR Premix Ex Taq (Takara) and CFX96 Real-time system (Bio-Rad Laboratories). The program for amplification was one cycle of 95°C for 2 min, followed by 40 cycles of 95°C for 10 s, 60°C for 30 s, and 95°C for 10 s (17). Primer sequences were as follows: IFN-γ, forward 5′-ACAGCAAGGCGAAAAAGGATG-3′ and reverse 5′-TGGTGGACCACTCGGATGA-3′; TGF-β, forward 5′-CCACCTGCAAGACCATCGAC-3′ and reverse 5′-CTGGCGAGCCTTAGTTTGGAC-3′; CXCL9, forward 5′-GGAGTTCGAGGAACCCTAGTG-3′ and reverse 5′-GGGATTTGTAGTGGATCGTGC-3′; CXCL10, forward 5′-CCAAGTGCTGCCGTCATTTTC-3′ and reverse 5′-TCCCTATGGCCCTCATTCTCA-3′; and β-actin, forward 5′-GTGACGTTGACATCCGTAAAGA-3′ and reverse 5′-GCCGGACTCATCGTACTCC-3′.

Data of a patient treated with fruquintinib and anti–PD-1 Ab are provided by the Department of Oncology, The First Affiliated Hospital with Nanjing Medical University. Written informed consent was obtained from the patient.

Data are expressed as mean ± SEM. Student t test and two-way ANOVA test were used for statistical analyses of the data. All statistical analyses were conducted using GraphPad Prism Software Version 7.0 (GraphPad Software, La Jolla, CA). Cases with p values <0.05 were considered statistically significant.

A 22-y-old patient with no significant past medical or family history presented with abdominal pain on April 1, 2018, and underwent an urgent left hemicolectomy. The pathology was consistent with that of invasive poorly differentiated adenocarcinoma, pT3N2bM0 (stage IIIB), with 7 of 17 lymph nodes showing positive result.

He started chemotherapy with capecitabine and oxaliplatin, but the disease progressed with a new lymph node adjacent to the abdominal aorta after three cycles. Next, the patient received a multiline line therapy including a second-line irinotecan, raltitrexed, and bevacizumab (five cycles), folinic acid, 5-fluorouracil, and oxaliplatin plus bevacizumab (one cycle), regorafenib single drug (two cycles), and regorafenib combined with tegafur (two cycles). Unfortunately, computerized tomography (CT) scan showed a continuous disease progression (PD) with sustained enlargement of the abdominal lymph nodes and increased carcinoembryonic Ag 199 (CA199), which was beyond the normal value. Thus, a new therapy was urgently needed.

Molecular profile test did not reveal the presence of KRAS and NRAS mutations, but revealed BRAF V600 mutation, and the tumor mutation burden was 6.45 mutations/megabase. Regarding PD-L1 expression on tumor cells, the tumor proportional score, and tumor immune cells proportional score were <1%. Moreover, his original tumor was confirmed by IHC to be of the MSS phenotype. On the basis of these results, he was then treated with fruquintinib plus sindilizumab (an anti-PD1 Ab) for six cycles from August 28, 2019. Surprisingly, CT scan showed that the enlarged lymph nodes in the root of the mesentery obviously became fewer and smaller after two cycles, and the diameter and number of lymph nodes continued to decrease. Moreover, CA199 decreased to 28.35 U/ml on October 22, 2019, and 13.02 U/ml on January 18, 2020, respectively (Fig. 1; detailed patient description is available in the Supplemental Fig. 1).

During the combination treatment, the patient developed moderate hypertension and slight rash, which were relieved after symptomatic treatment. To the best of our knowledge, this is the first report of successful treatment of a CRC patient using a combination of fruquintinib and sindilizumab with effective curation and favorable safety. This case report suggests that the combination of fruquintinib and anti–PD-1 therapy is an option for MSS CRC patients who fail to respond to multiline therapy.

To investigate whether fruquintinib plus anti–PD-1 exerts synergistical antineoplastic effect in vivo, CT26 mouse colon carcinoma cells (MSS) were s.c. transplanted to establish syngeneic murine models (18). There was no significant inhibition on tumor growth in mice treated with anti–PD-1 (TV: 1153.0 ± 161.2 mm3) compared with that in mice treated with vehicle (0.5% CMC.Na; TV: 1368.5 ± 131.7 mm3), with a TGI of 15.7%. Mice treated with fruquintinib alone had slower tumor growth and smaller TVs (753.8 ± 68.1 mm3), with a TGI of 45.6%. In contrast, the combination therapy group showed remarkable inhibition of tumor growth. The TV was 490.2 ± 74.6 mm3 on the day of sacrifice, and the TGI was 65.5% (Fig. 2A). There was a significant prolong of survive time in fruquintinib-treated group compared with control, which was further prolonged in the combination treatment group (Fig. 2B). The same tendency was also observed in tumor weight (Fig. 2C, 2D). Moreover, cotreatment of fruquintinib and anti–PD-1 caused no significant change in the bodyweight of mice, suggesting the safety and well tolerance of the cotherapy to some extent (Supplemental Fig. 2A).

Taken together, these data showed that a combination of fruquintinib and anti–PD-1 suppressed tumor growth in MSS CRC syngeneic without affecting the general health of the mice.

We further observed the effect of fruquintinib plus anti–PD-1 on the proliferation of tumor cells in the murine syngeneic model. Histological analyses by H&E staining showed that fruquintinib plus anti–PD-1 strongly induced massive cell damage, resulting in nuclear shrinkage, sparse arrangement and fragmentation of tumor cells (Fig. 3A). Consistent with the H&E staining results, TUNEL staining results confirmed that fruquintinib combined with anti–PD-1 triggered extensive apoptosis in tumor cells (Fig. 3B, 3C). In addition, results of IHC staining for PCNA showed sharply decreased expression of PCNA protein in the tumor tissues from the combined treatment group compared with that in the single-drug group (Fig. 3D, 3E). The same tendency was shown in proliferation and apoptosis-related proteins including c-Myc, cyclin-D1, and caspase 3 (Fig. 3F). These results showed that cotreatment with fruquintinib and anti–PD-1 led to elevated proliferation inhibition and apoptosis induction of tumor cells in vivo.

As VEGFR is the target of fruquintinib, we sought to confirm the on-target effect of fruquintinib. Thus, we investigated tumor neovascularization by Biocolor immunofluorescence with Ab against PECAM-1 (CD31) and α-SMA in tumor tissues (Fig. 4A). Quantification results of α-SMA+/CD31+ showed that fruquintinib single-drug treatment obviously inhibited the growth of tumor vessels and increased pericyte-covered microvessel density in tumor tissues both in the single-drug and combination groups (Fig. 4B). However, little effect was shown by anti–PD-1 single-drug therapy. We next analyzed the hypoxia status of tumor tissues through HIF1-α staining (Fig. 4C). The expression of HIF1-α was significantly decreased in fruquintinib-treated groups, indicating alleviated hypoxia level (Fig. 4D, 4E). The above results showed that inhibition of VEGFR by fruquintinib contributed to tumor vessel normalization and improvement of hypoxia in tumor tissues.

TME is intimately associated with tumor vessel normalization; hence, alleviation of hypoxia in the TME (16) and blockade of the VEGF/VEGFR pathway were reported to play important roles in inhibiting both peritumoral and tumoral regulatory T (Treg) cell infiltration, thereby suppressing anticancer immunity (10, 19). To test whether changes in Treg cells occur in the periphery, we detected immune cells in the spleen using flow cytometry. Results showed that fruquintinib monotherapy reduced the differentiation of Treg cells (percentage: 12.5%), and this effect was further enhanced by anti–PD-1 (percentage: 8.0%) (Fig. 5A). We subsequently analyzed the activation of antitumor lymphocytes. Compared with the control, fruquintinib plus anti–PD-1 increased both the proportion of CD4+ IFN-γ+ (Fig. 5B) and CD8+ IFN-γ+ T cells (Fig. 5C), although no obvious change in IFN-γ occurred following single-drug treatment. These results suggest that the combination of fruquintinib and anti–PD-1 augmented antitumor immune response by reducing Treg cells and enhancing the function of T lymphocytes.

We further analyzed changes in immune cells and their function inside the tumor tissues. An increasing tendency was shown in CD8+ T as well as CD4+ T cells (Fig. 6A–D) infiltration in tumor tissues treated with fruquintinib plus anti–PD-1, which is consistent with the result of a previous study (20). IFN-γ and granzyme B, markers of cytotoxic lymphocytes, were increased in the fruquintinib-treated groups, with the combination group showed the most abundant cells in tumor tissues (Fig. 6E, 6F). Consistent with that in the spleen, fruquintinib monotherapy reduced the differentiation of Treg cells inside the tumor tissues, and this effect was further enhanced by anti–PD-1 (Fig. 6G). Moreover, fruquintinib plus anti–PD-1 tended to alter tumor-associated macrophage (TAM) polarization. The expression of inducible NO synthase and TNF-α were increased, whereas the ratio of F4/80+CD86+MHC-II+ “M1-type–activated” TAMs was increased and the ratio of F4/80+CD206+Arg1+ “M2-type–activated” TAMs decreased, leading to the elevated M1/M2 ratio (Fig. 7). These results suggested that cotherapy with fruquintinib and anti–PD-1 enhanced T lymphocyte activation in tumors and promoted TAMs into the antitumor state, resulting in a strong antitumor immune microenvironment.

By using ELISA and quantitative PCR (qPCR), we subsequently analyzed changes in cytokine trafficking in Ag-specific antitumor T cells into colorectal tumors after treatment. We found that fruquintinib plus anti–PD-1 significantly upregulated the cytokine production and mRNA expression levels of CXCL9 and CXCL10 as well as IFN-γ in CT26 murine models (Fig. 8A, 8B). The combination therapy also decreased TGF-β secretion, which hindered T cell recruitment into tumor tissues (Fig. 8A, 8B). We further assessed tumor cell immunogenicity through Western blot analysis for PD-L1 in tumor tissues. We found that fruquintinib plus anti–PD-1 potently increased PD-L1 expression, whereas fruquintinib and anti–PD-1 alone led to no obvious difference in PD-L1 expression compared with the control (Fig. 8C). In brief, cotreatment with fruquintinib and anti–PD-1 could improve immunopermissive TME and enhance the immunogenicity of tumor cells.

To further confirm whether T cells were involved in the delayed tumor progression induced by the combination therapy, MC38 mouse colon carcinoma cells were s.c. transplanted in CD8 WT and CD8 KO C57BL/6 mice, and the effect of the combination therapy was examined. The syngeneic effect of fruquintinib plus anti–PD-1 was abrogated tremendously in CD8 KO mice (TGI: 43.5%), compared with CD8 WT mice (TGI: 69.4%) (Fig. 9A). It was the same tendency in the weight of tumor tissues obtained from the mice sacrificed (Fig. 9B, 9C). Also, no impact on the bodyweight was occurred in fruquintinib and/or anti–PD-1 therapy (Supplemental Fig. 2B). All these data indicated a pivotal role of CD8+ T cells in the syngeneic effect of cotreatment of fruquintinib and anti–PD-1.

Tumor with MSS is the most frequent colorectal neoplasm in both metastatic and nonmetastatic settings, and no single-agent anti–PD-1/PD-L1 therapy has shown any efficacy against these tumors (9, 21). Identifying the optimal combinatorial strategies to enhance the efficacy of anti–PD-1/PD-L1–based immunotherapy is an important research to combat MSS CRC (22).

The obstacles in anti–PD-1/PD-L1 therapy for the treatment of MSS CRC may be attributed to factors such as low mutational burden, insufficient immunogenicity of tumor cells, Ag-immunosuppressive metabolic pathways, defective Ag presentation, and poorly infiltrated antitumor immune cells (5, 22). To enhance the antitumor effect, many clinical trials have sought to establish combinations of anti–PD-1 and other therapies, such as chemotherapy, targeted therapy, radiotherapy, or other immunotherapies (23). In this study, we presented the case of an MSS mCRC patient whose disease progressed after first-line chemotherapy with 5-fluorouracil, oxaliplatin, and irinotecan, as well as with multiline therapies, including chemotherapy plus bevacizumab and regorafenib monotherapy or plus chemotherapy, but achieved rapid response after receiving anti–PD-1 and fruquintinib, which is a newly marketed VEGFR inhibitor. This unexpected efficacy was consistent with previous data on lenvatinib in combination with pembrolizumab in MSS endometrial cancer (14). The mechanism of the curative effect of such combination treatment may involve the elimination of drug resistance to anti–PD-1 by fruquintinib or a synergistic improvement of the efficacy of the two drugs; in fact, it was reported that VEGF/VEGFR inhibition, which elicits antitumor immunity, can be enhanced by anti–PD-1/PD-L1 (13, 24, 25). Considering the present clinical case, combination treatment with fruquintinib plus anti–PD-1 could be a potential alternative therapy for anti–PD-1–insensitive CRC. However, it was limited to only one case, and clinical studies are needed to verify the efficacy and safety of the combined use of fruquintinib and anti–PD-1. In addition, combinations of VEGF/VEGFR and anti–PD-1/PD-L1 have exerted notable efficacy against several tumors compared with the most commonly applied therapy (26). A recent study of ramucirumab with pembrolizumab showed a manageable safety profile with favorable antitumor activity in patients with previously treated advanced gastric or gastroesophageal junction adenocarcinoma, non–small cell lung cancer, and urothelial carcinoma (26).

Fruquintinib, a VEGFR blocker widely used in cancer treatment, normalizes tumor vessels in a dose- and time-dependent manner (16, 27), which has been proven critical in the effect of fruquintinib combined with cytotoxic drugs (16). To elucidate the antitumor effect of fruquintinib plus anti–PD-1 in vivo, syngeneic murine MSS CRC models was established by s.c. implanting CT26 cells, and the synergistic antineoplastic effect of the combination treatment was observed. This synergistic effect of cotreatment with fruquintinib and anti–PD-1 may be mainly attributed to the transient normalization of the vascular structure and function as well as the pronounced vascular rarefaction. Zheng et al. (28) have shown that only efficacious anti–PD-1 can increase vessel perfusion by promoting CD8+T cell accumulation and IFN-γ production (24), indicating a potential synergistic effect between anti–PD-1 and VEGF/VEGFR. Some reports showed that the benefit of combinations of VEGF/VEGFR with anti–PD-1/PD-L1 is associated with substantial pruning of the tumor vasculature. These reports also indicated that both passive (24) (pruning of abnormal vessels) and active (25) (formation of normalized vessels through increased pericyte coverage) normalizations are critical for the efficacy of combined therapy with anti-VEGF/VEGFR and anti–PD-1/PD-L1.

Abnormal tumor vasculature fosters an immunosuppressive TME that enables tumors to evade host immunosurveillance (29, 30). Clinical studies consistently support the view that MSS-type mCRC is nonpermissive to T effector cell accumulation and is usually infiltrated with abundant immune suppressors, such as TAMs, myeloid-derived suppressor cells, and Treg cells (22). These immune evasion mechanisms could be overcome by VEGF/VEGFR blockade (10). Tumor vascular normalization through deletion of Rgs5 increases T cell infiltration into tumors and substantially improves survival after adoptive T cell transfer in mice (31). Several preclinical studies have suggested that VEGF/VEGFR blockade therapy promotes an immune-supportive microenvironment and enhances the effect of anti–PD-1/PD-L1 in several types of tumors (11, 12, 32, 33). In our study, fruquintinib played a critical role in regulating key antitumor immune responses through mechanisms such as suppression of Treg cell proliferation and enhancement of antitumor lymphocyte infiltration and function, which were both further enhanced by anti–PD-1. We also observed that in colorectal carcinoma, the expression of specific chemokines in antitumor T cells (34, 35), such as CXCL9 and CXCL10, was significantly increased in the combination treatment group. Unfortunately, in solid cancers, surface expression of chemokine receptors on activated T lymphocytes does not always match the cognate ligand expression at the tumor site (35). Moreover, in combination therapy with fruquintinib and anti–PD-1, immunosuppressive M2-like phenotype TAMs were decreased, whereas immunopermissive M1-like phenotype TAMs increased. These findings were in agreement with the results of K. Shigeta regarding a combination of DC101 (a VEGF inhibitor) and anti–PD-1.

In addition, we compared other various immune components in the TME after treatment with fruquintinib and/or anti–PD-1. The results showed that the drug combination reprogrammed the immunosuppressive TME and enhanced immunotherapy by regulating cytokine secretion and PD-L1 expression. Although the predictive value of PD-L1 expression for response to anti–PD-1/PD-L1 therapy is being debated, MSS CRC especially expressed a low level of PD-L1. Rosenbaum et al. (36) indicated that upregulation of PD-L1 in colorectal carcinoma is due to an adaptive immune response based on the association of PD-L1 expression and frequent tumor-infiltrating lymphocytes. Moreover, PD-L1 expression is correlated with significantly decreased survival in microsatellite-unstable cases. On the contrary, Li et al. (37) has shown that high PD-L1 expression is correlated with better prognosis in MSS CRC patients. Even so, multiple studies consistently reported that PD-L1 was upregulated in cancer after VEGF/VEGFR blockade (12, 13, 37, 38). Allen et al. (25) showed that PD-L1 was upregulated in tumors relapsing from antiangiogenic (sorafenib or the anti-VEGFR2 Ab DC101) therapy in pancreatic neuroendocrine tumors and polyoma middle T oncoprotein mammary carcinoma models through IFN-γ. More studies are needed to verify the direct correlation between PD-L1 upregulation and combination therapy with fruquintinib and anti–PD-1. All the above findings highlight the complexity and intricate mechanisms of interaction between these treatments in different tumors.

Future studies should establish the extent of the sustained beneficial and adverse effects of long-term administration of fruquintinib plus anti–PD-1. They should also validate our findings using available patient-derived tumor xenograft model. Biological markers with promising efficacy also need to be explored in the future. In conclusion, the current study is a proof of concept that fruquintinib, in combination with anti–PD-1, showed enhanced therapeutic effect in MSS CRC models by optimizing antitumor microenvironment via vascular normalization that promoted immunopermissive microenvironment. Our results also indicated that the combination of fruquintinib and anti–PD-1, which has rarely been tested in preclinical or clinical studies, may be sufficient to appropriately reprogram the immune microenvironment and enhance immunotherapy efficacy. These findings prompt future studies of this combination therapy in MSS CRC or other cancers, which might be a potential strategy to broaden the benefit of anti–PD-1/PD-L1 treatment.

We thank Prof. Lilin Ye, Third Military Medical University, for sharing CD8 KO mice with C57BL/6 background.

This work was supported by the National Natural Science Foundation of China (81871944, 81572389, and 81922067), the Jiangsu 333 Project (BRA2016517), the Six Talent Climax Foundation of Jiangsu (YY-004), Jiangsu Province Key Medical Talents (ZDRCA2016026), and Fundamental Research Funds for the Central Universities (020814380114).

The online version of this article contains supplemental material.

Abbreviations used in this article:

CMC.Na

sodium carboxymethyl cellulose

CRC

colorectal cancer

CT

computerized tomography

IHC

immunohistochemistry

KO

knockout

mCRC

metastatic CRC

MHC-II

MHC class II

MSS

microsatellite stability

qPCR

quantitative PCR

TAM

tumor-associated macrophage

TGI

tumor growth inhibition

TME

immunosuppressive tumor microenvironment

Treg

regulatory T

TV

tumor volume

VEGF

vascular endothelial growth factor

WT

wild-type.

1
Dekker
,
E.
,
P. J.
Tanis
,
J. L. A.
Vleugels
,
P. M.
Kasi
,
M. B.
Wallace
.
2019
.
Colorectal cancer.
Lancet
394
:
1467
1480
.
2
Stintzing
,
S.
,
D. P.
Modest
,
L.
Rossius
,
M. M.
Lerch
,
L. F.
von Weikersthal
,
T.
Decker
,
A.
Kiani
,
U.
Vehling-Kaiser
,
S. E.
Al-Batran
,
T.
Heintges
, et al
FIRE-3 Investigators
.
2016
.
FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab for metastatic colorectal cancer (FIRE-3): a post-hoc analysis of tumour dynamics in the final RAS wild-type subgroup of this randomised open-label phase 3 trial.
Lancet Oncol.
17
:
1426
1434
.
3
Li
,
J.
,
S.
Qin
,
R.
Xu
,
T. C. C.
Yau
,
B.
Ma
,
H.
Pan
,
J.
Xu
,
Y.
Bai
,
Y.
Chi
,
L.
Wang
, et al
CONCUR Investigators
.
2015
.
Regorafenib plus best supportive care versus placebo plus best supportive care in Asian patients with previously treated metastatic colorectal cancer (CONCUR): a randomised, double-blind, placebo-controlled, phase 3 trial.
Lancet Oncol.
16
:
619
629
.
4
Hirsch
,
L.
,
L.
Zitvogel
,
A.
Eggermont
,
A.
Marabelle
.
2019
.
PD-Loma: a cancer entity with a shared sensitivity to the PD-1/PD-L1 pathway blockade.
Br. J. Cancer
120
:
3
5
.
5
Jacquelot
,
N.
,
T.
Yamazaki
,
M. P.
Roberti
,
C. P. M.
Duong
,
M. C.
Andrews
,
L.
Verlingue
,
G.
Ferrere
,
S.
Becharef
,
M.
Vétizou
,
R.
Daillère
, et al
.
2019
.
Sustained type I interferon signaling as a mechanism of resistance to PD-1 blockade.
Cell Res.
29
:
846
861
.
6
Powles
,
T.
,
I.
Durán
,
M. S.
van der Heijden
,
Y.
Loriot
,
N. J.
Vogelzang
,
U.
De Giorgi
,
S.
Oudard
,
M. M.
Retz
,
D.
Castellano
,
A.
Bamias
, et al
.
2018
.
Atezolizumab versus chemotherapy in patients with platinum-treated locally advanced or metastatic urothelial carcinoma (IMvigor211): a multicentre, open-label, phase 3 randomised controlled trial. [Published erratum appears in 2018 Lancet 392: 1402.]
Lancet
391
:
748
757
.
7
Robert
,
C.
,
L.
Thomas
,
I.
Bondarenko
,
S.
O’Day
,
J.
Weber
,
C.
Garbe
,
C.
Lebbe
,
J. F.
Baurain
,
A.
Testori
,
J. J.
Grob
, et al
.
2011
.
Ipilimumab plus dacarbazine for previously untreated metastatic melanoma.
N. Engl. J. Med.
364
:
2517
2526
.
8
Le
,
D. T.
,
J. N.
Uram
,
H.
Wang
,
B. R.
Bartlett
,
H.
Kemberling
,
A. D.
Eyring
,
A. D.
Skora
,
B. S.
Luber
,
N. S.
Azad
,
D.
Laheru
, et al
.
2015
.
PD-1 blockade in tumors with mismatch-repair deficiency.
N. Engl. J. Med.
372
:
2509
2520
.
9
Hermel
,
D. J.
,
D.
Sigal
.
2019
.
The emerging role of checkpoint inhibition in microsatellite stable colorectal cancer.
J. Pers. Med.
9
:
5
.
10
Georganaki
,
M.
,
L.
van Hooren
,
A.
Dimberg
.
2018
.
Vascular targeting to increase the efficiency of immune checkpoint blockade in cancer.
Front. Immunol.
9
:
3081
.
11
Wallin
,
J. J.
,
J. C.
Bendell
,
R.
Funke
,
M.
Sznol
,
K.
Korski
,
S.
Jones
,
G.
Hernandez
,
J.
Mier
,
X.
He
,
F. S.
Hodi
, et al
.
2016
.
Atezolizumab in combination with bevacizumab enhances antigen-specific T-cell migration in metastatic renal cell carcinoma.
Nat. Commun.
7
:
12624
.
12
Zhao
,
S.
,
S.
Ren
,
T.
Jiang
,
B.
Zhu
,
X.
Li
,
C.
Zhao
,
Y.
Jia
,
J.
Shi
,
L.
Zhang
,
X.
Liu
, et al
.
2019
.
Low-dose Apatinib optimizes tumor microenvironment and potentiates antitumor effect of PD-1/PD-L1 blockade in lung cancer.
Cancer Immunol. Res.
7
:
630
643
.
13
Shigeta
,
K.
,
M.
Datta
,
T.
Hato
,
S.
Kitahara
,
I. X.
Chen
,
A.
Matsui
,
H.
Kikuchi
,
E.
Mamessier
,
S.
Aoki
,
R. R.
Ramjiawan
, et al
.
2020
.
Dual programmed death receptor-1 and vascular endothelial growth factor receptor-2 blockade promotes vascular normalization and enhances antitumor immune responses in hepatocellular carcinoma.
Hepatology
71
:
1247
1261
.
14
Konecny
,
G. E.
2019
.
Inhibition of PD-1 and VEGF in microsatellite-stable endometrial cancer.
Lancet Oncol.
20
:
612
614
.
15
Shirley
,
M.
2018
.
Fruquintinib: first global approval.
Drugs
78
:
1757
1761
.
16
Sun
,
Q.
,
J.
Zhou
,
Z.
Zhang
,
M.
Guo
,
J.
Liang
,
F.
Zhou
,
J.
Long
,
W.
Zhang
,
F.
Yin
,
H.
Cai
, et al
.
2014
.
Discovery of fruquintinib, a potent and highly selective small molecule inhibitor of VEGFR 1, 2, 3 tyrosine kinases for cancer therapy.
Cancer Biol. Ther.
15
:
1635
1645
.
17
Guo
,
W.
,
Y.
Sun
,
W.
Liu
,
X.
Wu
,
L.
Guo
,
P.
Cai
,
X.
Wu
,
X.
Wu
,
Y.
Shen
,
Y.
Shu
, et al
.
2014
.
Small molecule-driven mitophagy-mediated NLRP3 inflammasome inhibition is responsible for the prevention of colitis-associated cancer.
Autophagy
10
:
972
985
.
18
Limagne
,
E.
,
M.
Thibaudin
,
L.
Nuttin
,
A.
Spill
,
V.
Derangère
,
J. D.
Fumet
,
N.
Amellal
,
E.
Peranzoni
,
V.
Cattan
,
F.
Ghiringhelli
.
2019
.
Trifluridine/tipiracil plus oxaliplatin improves PD-1 blockade in colorectal cancer by inducing immunogenic cell death and depleting macrophages.
Cancer Immunol. Res.
7
:
1958
1969
.
19
Terme
,
M.
,
S.
Pernot
,
E.
Marcheteau
,
F.
Sandoval
,
N.
Benhamouda
,
O.
Colussi
,
O.
Dubreuil
,
A. F.
Carpentier
,
E.
Tartour
,
J.
Taieb
.
2013
.
VEGFA-VEGFR pathway blockade inhibits tumor-induced regulatory T-cell proliferation in colorectal cancer.
Cancer Res.
73
:
539
549
.
20
Huang
,
H.
,
E.
Langenkamp
,
M.
Georganaki
,
A.
Loskog
,
P. F.
Fuchs
,
L. C.
Dieterich
,
J.
Kreuger
,
A.
Dimberg
.
2015
.
VEGF suppresses T-lymphocyte infiltration in the tumor microenvironment through inhibition of NF-κB-induced endothelial activation.
FASEB J.
29
:
227
238
.
21
Le
,
D. T.
,
J. N.
Durham
,
K. N.
Smith
,
H.
Wang
,
B. R.
Bartlett
,
L. K.
Aulakh
,
S.
Lu
,
H.
Kemberling
,
C.
Wilt
,
B. S.
Luber
, et al
.
2017
.
Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade.
Science
357
:
409
413
.
22
Galon
,
J.
,
D.
Bruni
.
2019
.
Approaches to treat immune hot, altered and cold tumours with combination immunotherapies.
Nat. Rev. Drug Discov.
18
:
197
218
.
23
Chinai
,
J. M.
,
M.
Janakiram
,
F.
Chen
,
W.
Chen
,
M.
Kaplan
,
X.
Zang
.
2015
.
New immunotherapies targeting the PD-1 pathway.
Trends Pharmacol. Sci.
36
:
587
595
.
24
Schmittnaegel
,
M.
,
N.
Rigamonti
,
E.
Kadioglu
,
A.
Cassará
,
C.
Wyser Rmili
,
A.
Kiialainen
,
Y.
Kienast
,
H. J.
Mueller
,
C. H.
Ooi
,
D.
Laoui
,
M.
De Palma
.
2017
.
Dual angiopoietin-2 and VEGFA inhibition elicits antitumor immunity that is enhanced by PD-1 checkpoint blockade.
Sci. Transl. Med.
9
: eaak9670.
25
Allen
,
E.
,
A.
Jabouille
,
L. B.
Rivera
,
I.
Lodewijckx
,
R.
Missiaen
,
V.
Steri
,
K.
Feyen
,
J.
Tawney
,
D.
Hanahan
,
I. P.
Michael
,
G.
Bergers
.
2017
.
Combined antiangiogenic and anti-PD-L1 therapy stimulates tumor immunity through HEV formation.
Sci. Transl. Med.
9
: eaak9679.
26
Herbst
,
R. S.
,
H. T.
Arkenau
,
R.
Santana-Davila
,
E.
Calvo
,
L.
Paz-Ares
,
P. A.
Cassier
,
J.
Bendell
,
N.
Penel
,
M. G.
Krebs
,
J.
Martin-Liberal
, et al
.
2019
.
Ramucirumab plus pembrolizumab in patients with previously treated advanced non-small-cell lung cancer, gastro-oesophageal cancer, or urothelial carcinomas (JVDF): a multicohort, non-randomised, open-label, phase 1a/b trial.
Lancet Oncol.
20
:
1109
1123
.
27
Zhang
,
Y.
,
J. Y.
Zou
,
Z.
Wang
,
Y.
Wang
.
2019
.
Fruquintinib: a novel antivascular endothelial growth factor receptor tyrosine kinase inhibitor for the treatment of metastatic colorectal cancer.
Cancer Manag. Res.
11
:
7787
7803
.
28
Zheng
,
X.
,
Z.
Fang
,
X.
Liu
,
S.
Deng
,
P.
Zhou
,
X.
Wang
,
C.
Zhang
,
R.
Yin
,
H.
Hu
,
X.
Chen
, et al
.
2018
.
Increased vessel perfusion predicts the efficacy of immune checkpoint blockade.
J. Clin. Invest.
128
:
2104
2115
.
29
Huang
,
Y.
,
J.
Yuan
,
E.
Righi
,
W. S.
Kamoun
,
M.
Ancukiewicz
,
J.
Nezivar
,
M.
Santosuosso
,
J. D.
Martin
,
M. R.
Martin
,
F.
Vianello
, et al
.
2012
.
Vascular normalizing doses of antiangiogenic treatment reprogram the immunosuppressive tumor microenvironment and enhance immunotherapy.
Proc. Natl. Acad. Sci. USA
109
:
17561
17566
.
30
Voron
,
T.
,
O.
Colussi
,
E.
Marcheteau
,
S.
Pernot
,
M.
Nizard
,
A. L.
Pointet
,
S.
Latreche
,
S.
Bergaya
,
N.
Benhamouda
,
C.
Tanchot
, et al
.
2015
.
VEGF-A modulates expression of inhibitory checkpoints on CD8+ T cells in tumors.
J. Exp. Med.
212
:
139
148
.
31
Shrimali
,
R. K.
,
Z.
Yu
,
M. R.
Theoret
,
D.
Chinnasamy
,
N. P.
Restifo
,
S. A.
Rosenberg
.
2010
.
Antiangiogenic agents can increase lymphocyte infiltration into tumor and enhance the effectiveness of adoptive immunotherapy of cancer.
Cancer Res.
70
:
6171
6180
.
32
McDermott
,
D. F.
,
M. A.
Huseni
,
M. B.
Atkins
,
R. J.
Motzer
,
B. I.
Rini
,
B.
Escudier
,
L.
Fong
,
R. W.
Joseph
,
S. K.
Pal
,
J. A.
Reeves
, et al
.
2018
.
Clinical activity and molecular correlates of response to atezolizumab alone or in combination with bevacizumab versus sunitinib in renal cell carcinoma. [Published erratum appears in 2018 Nat. Med. 24: 1941.]
Nat. Med.
24
:
749
757
.
33
Makker
,
V.
,
D.
Rasco
,
N. J.
Vogelzang
,
M. S.
Brose
,
A. L.
Cohn
,
J.
Mier
,
C.
Di Simone
,
D. M.
Hyman
,
D. E.
Stepan
,
C. E.
Dutcus
, et al
.
2019
.
Lenvatinib plus pembrolizumab in patients with advanced endometrial cancer: an interim analysis of a multicentre, open-label, single-arm, phase 2 trial.
Lancet Oncol.
20
:
711
718
.
34
Vignali
,
D.
,
M.
Kallikourdis
.
2017
.
Improving homing in T cell therapy.
Cytokine Growth Factor Rev.
36
:
107
116
.
35
Sharma
,
R. K.
,
Z. S.
Chheda
,
V. R.
Jala
,
B.
Haribabu
.
2015
.
Regulation of cytotoxic T-Lymphocyte trafficking to tumors by chemoattractants: implications for immunotherapy.
Expert Rev. Vaccines
14
:
537
549
.
36
Rosenbaum
,
M. W.
,
J. R.
Bledsoe
,
V.
Morales-Oyarvide
,
T. G.
Huynh
,
M.
Mino-Kenudson
.
2016
.
PD-L1 expression in colorectal cancer is associated with microsatellite instability, BRAF mutation, medullary morphology and cytotoxic tumor-infiltrating lymphocytes.
Mod. Pathol.
29
:
1104
1112
.
37
Li
,
Y.
,
L.
Liang
,
W.
Dai
,
G.
Cai
,
Y.
Xu
,
X.
Li
,
Q.
Li
,
S.
Cai
.
2016
.
Prognostic impact of programed cell death-1 (PD-1) and PD-ligand 1 (PD-L1) expression in cancer cells and tumor infiltrating lymphocytes in colorectal cancer.
Mol. Cancer
15
:
55
.
38
Chen
,
Y.
,
R. R.
Ramjiawan
,
T.
Reiberger
,
M. R.
Ng
,
T.
Hato
,
Y.
Huang
,
H.
Ochiai
,
S.
Kitahara
,
E. C.
Unan
,
T. P.
Reddy
, et al
.
2015
.
CXCR4 inhibition in tumor microenvironment facilitates anti-programmed death receptor-1 immunotherapy in sorafenib-treated hepatocellular carcinoma in mice.
Hepatology
61
:
1591
1602
.

The authors have no financial conflicts of interest.

Supplementary data