Typically, ICIs are ineffective against tumors as immune cells are absent from the tumor microenvironment (TME; referred to as ‘immune-desert’ or ‘immune-excluded’) or because they cannot penetrate the tumor (15,16). Chemotherapy drugs can influence the immunogenicity of cancer cells through multiple mechanisms. First, chemotherapy drugs can alter the immune microenvironment of the tumor, thereby enhancing antitumor immune responses. For example, certain chemotherapy drugs can induce immune-mediated cell death in tumor cells, a form of cell death that activates adaptive immune responses and promotes the activation and proliferation of antitumor T cells (17). Additionally, chemotherapy drugs can improve the immune state of the TME by inhibiting the function of immunosuppressive cells and enhancing the activity of immune effector cells (18). Secondly, chemotherapeutic drugs can enhance antigen presentation by promoting the recruitment and activation of dendritic cells (DCs), thereby improving the ability of the immune system to recognize and attack tumor antigens. A study has reported that chemotherapeutic drugs can induce tumor cells to release pro-inflammatory factors, which promote the migration and maturation of DCs, thus enhancing T-cell activation (19). This mechanism not only enhances the direct cytotoxic effects of chemotherapeutic drugs but also improves their antitumor efficacy through immune modulation. Finally, chemotherapy drugs can enhance the effectiveness of immunotherapy by modulating immune suppression circuits in the TME. Chemotherapy drugs can disrupt immune suppression mechanisms in the TME, for example, by reducing the number of regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), thereby enhancing the function of effector T cells and natural killer (NK) cells (20). This immune regulatory effect provides a theoretical basis for the combined use of chemotherapy and immunotherapy, potentially improving the overall efficacy of cancer treatment.

A multicenter retrospective study showed that among patients with NSCLC who received ICIs, the prevalence of HPD was ~14%, whereas it was ~5% among those treated with chemotherapy. Within the initial 3-month period following ICI treatment, tumor progression accounted for ~15% of deaths. HPD was significantly associated with >2 metastatic sites before ICI treatment (62.5 vs. 42.6%; P=0.006) (21). Moreover, the Checkmate 227 trial reported that patients with PD-L1 expression of <1% experienced a lower progression risk with nivolumab combined with chemotherapy than nivolumab plus ipilimumab therapy at the early 3-month timepoint (22). The KEYNOTE-062 trial also reported that survival rates during the first 6–9 months were lower for patients treated with pembrolizumab monotherapy than for those treated with chemotherapy (23). These data suggest that patients at high risk of HPD may find it advantageous to receive both chemotherapy plus ICIs or to rapidly switch to chemotherapy after ICI therapy (24). In summary, the effectiveness of ICIs is enhanced by chemotherapeutic medications through their impact on the immunogenicity of cancer cells (Fig. 1).

The effectiveness of combining ICIs with chemotherapy has been previously reported. In the KEYNOTE-189 study, the effectiveness of pemetrexed in combination with platinum-based medications or a placebo, was evaluated for the treatment of non-squamous metastatic NSCLC in a double-blind phase III trial. The pembrolizumab plus pemetrexed-platinum arm achieved an objective response rate (ORR) of 48.3%, whereas the placebo plus pemetrexed-platinum arm only achieved 19.9%. In the pembrolizumab-combination group, the 5-year overall survival (OS) rate was 19.4%, whereas in the placebo-combination group, the 5-year OS rate was 11.3%. The 5-year PFS rates were 7.5 and 0.6%, respectively (25,26).

According to another study, patients in the ICIs plus chemotherapy group had significantly improved OS [not reached (NR) vs. 18.3 months; P=0.011] and PFS (14.9 vs. 6.3 months; P<0.001) times compared with those in the bevacizumab plus chemotherapy group. Moreover, as a first-line treatment, ICIs combined with chemotherapy was a more beneficial treatment for patients with adenocarcinoma who did not have driver gene alterations compared with bevacizumab combined with chemotherapy (27). Additionally, Tsai et al (28) reported that the combination of pembrolizumab with chemotherapy or radiotherapy results in a notably improved PFS for patients with treatment-naïve advanced NSCLC.

Furthermore, the efficacy of atezolizumab plus carboplatin plus paclitaxel (ACP group), atezolizumab plus bevacizumab plus carboplatin plus paclitaxel (ABCP group) or bevacizumab plus carboplatin plus paclitaxel (BCP group) was assessed in patients with metastatic non-squamous NSCLC by Impower150. The conclusive findings indicated that the median OS time for the ACP group was 19.0 months compared with 15.0 months for the BCP group [hazard ratio (HR), 0.86; 95% confidence interval (CI), 0.73–1.01] in patients with NSCLC lacking EGFR or ALK genomic abnormalities (29).

The KEYNOTE-407 study randomly divided patients with untreated metastatic squamous NSCLC into two groups: One group received pembrolizumab plus carboplatin and either paclitaxel or nab-paclitaxel, whilst the other group received a saline placebo plus carboplatin and the choice of paclitaxel or nab-paclitaxel. In the study, the OS and PFS rates were improved with pembrolizumab plus chemotherapy (HR, 0.71; 95% CI, 0.59–0.85) compared with placebo plus chemotherapy (HR, 0.62; 95% CI, 0.52–0.74), with 5-year OS rates of 18.4 and 9.7%, respectively (30,31). Numerous clinical studies have reported the efficacy of chemotherapy plus ICIs (12,13,32–35) (Tables I and SI).

The entry of immune cells into the tumor is facilitated by blood vessels. VEGF can induce tumor neovascularization and abnormal angiogenesis and promote the occurrence of an immunosuppressive TME (36). The presence of VEGF in circulation hinders the development and effectiveness of DCs, aiding the tumor in evading immune detection (36). Additionally, abnormal blood vessel formation leads to a reduction in the quantity and effectiveness of lymphocytes that combat tumors. When there is an increase in oxygen consumption, the neo-vasculature is unable to form, leading to direct impairment of tumor-infiltrating lymphocyte (TIL) functions due to hypoxia (37). The M2-like phenotype of tumor-associated macrophages is facilitated by the hypoxic environment within the tumor (38). Based on the immunomodulatory role of VEGF-A in different cancer types, ICIs have been combined with anti-angiogenic drugs in therapy (39). Blocking VEGF-A pharmacologically results in an augmentation of intratumoral CD8⁺ cells and a decrease in tumor growth (40). The use of anti-VEGF/VEGFR2 medications facilitates the normalization of blood vessels, leading to enhanced infiltration of cytotoxic T cells and aiding in the transportation of drugs inside tumors. Additionally, it reduces immunosuppressive cells such as MDSCs and Tregs, demonstrating a notable synergistic impact when combined with PD-1/PD-L1 inhibition (41,42) (Fig. 1).

In the CheckMate 012 phase I study, it was reported that the combination of nivolumab and bevacizumab resulted in a median PFS time of 37.1 weeks in patients with non-squamous cell carcinoma (n=12). Moreover, among patients with non-squamous cell carcinoma (n=13) and squamous cell carcinoma (n=8) treated with nivolumab monotherapy, the median PFS times were 21.4 and 16 weeks, respectively (43).

Furthermore, the impact of a combination treatment involving atezolizumab and bevacizumab was evaluated in the IMpower150 phase III trial for patients with previously untreated metastatic non-squamous NSCLC. According to the Kaplan-Meier analysis, it was reported that the ABCP group had notably extended PFS and OS times compared with the BCP group, with a median PFS time of 8.4 months for ABCP and 6.8 months for BCP (HR, 0.57; 95% CI, 0.48–0.67) and a median OS time of 19.5 months for ABCP and 14.7 months for BCP (HR, 0.8; 95% CI, 0.67–0.95) (29). Additionally, in the NCT02443324 phase I trial, the combination of ramucirumab and pembrolizumab demonstrated manageable safety and favorable antitumor efficacy in patients diagnosed with advanced gastric or gastroesophageal junction adenocarcinoma, NSCLC and urothelial cancer (44).

Another study reported that the combination of anti-PD-1/PD-L1 therapy and anti-angiogenesis therapy markedly enhanced the effectiveness of treatment in patients with advanced lung adenocarcinoma who did not respond to initial or advanced treatment. The disease control rate (DCR) was significantly greater in the group receiving both anti-PD-1/L1 and anti-angiogenesis treatment compared with the group receiving only anti-PD-1/L1 therapy (92.0 vs. 46.9%; P=0.0004). The median PFS time was 5.1 months compared with 2.0 months (HR, 0.551; 95% CI, 0.337–0.902; P=0.002), whilst the median OS time was 14.3 months compared with 8.4 months (HR, 0.549; 95% CI, 0.305–0.990; P=0.046), respectively (45).

Finally, according to a retrospective study, the combination of ICIs and low-dose anlotinib was reported to be a viable treatment choice for advanced NSCLC in the second-line setting (46) (Tables II and SI).

Previous research has indicated that NSCLC with EGFR mutations possesses a distinct TME (47). The alteration of EGFR and EGFR-tyrosine kinase inhibitors (TKIs) could potentially impact the TME, consequently influencing the effectiveness of immunotherapy. Immunologically cold tumors, typically NSCLCs with EGFR mutations, are characterized by lower levels of CD8⁺ TILs and impaired function, leading to reduced cytotoxicity and inherent resistance to ICIs (48,49). The TME of patients with NSCLC may be modified by EGFR or ALK inhibitors, leading to an increase in CD8⁺ TILs and a decrease in immunosuppressive cells (such as Tregs). This suggests the potential for the use of PD-1/PD-L1 antibodies (50).

In addition, the use of BRAF and MEK inhibitors impacts the immune response against tumors by inducing an increase in the immune-stimulating cytokine levels, a reduction in the immunosuppressive cytokine levels and an enhancement in the infiltration and activity of T cells. BRAF and MEK inhibitors allow the immune system to identify cancerous growths and improve the response of TILs (51). KRAS-mutant patients have been reported to exhibit a higher PD-L1 tumor proportion score and tumor mutational burden, as observed in the KEYNOTE-042 study (52). Stimulating effector T-lymphocyte activation and reducing the proliferation of immunosuppressive cells are the primary mechanisms of action for other targeted therapy medications (53,54). In summary, the aforementioned mechanisms provide a theoretical basis for combining these targeted therapies with ICIs (Fig. 1).

The combination of erlotinib and atezolizumab in the treatment of advanced NSCLC has attracted significant attention. In this study, the results indicated that the combination exhibited tolerable safety and showed promising clinical activity in patients with EGFR-mutated NSCLC. The combination therapy achieved an ORR of 75%, with a median PFS of 15.4 months, while median OS was NR (55). Furthermore, the TATTON trial (NCT02143466) assessed the efficacy of osimertinib in combination with selumetinib, savolitinib or durvalumab in patients with advanced NSCLC harboring EGFR mutations. In the selumetinib, savolitinib and durvalumab arms, the ORR was 42% (95% CI, 26–59%), 44% (95% CI, 22–69%) and 43% (95% CI, 23–66%), respectively. For TKI pretreatment and TKI-naïve patients, osimertinib plus durvalumab demonstrated encouraging results. Nevertheless, the therapy led to a rise in drug-related interstitial lung disease, resulting in the discontinuation of the combined treatment regimen in this particular cohort (56).

According to a cohort study, patients in the ICIs plus anlotinib group exhibited a significantly longer median PFS compared with those in the ICIs monotherapy group (6.37 vs. 3.90 months; P<0.001) (57). Furthermore, in the study performed by He et al (58), the study group (anlotinib combined with ICIs) exhibited considerably higher DCR and ORR values compared with the control group (62.50 vs. 36.11 and 81.25 vs. 55.56, respectively; P<0.05).

In addition, in the KEYNOTE-021 clinical trial, the combination of pembrolizumab and erlotinib resulted in an ORR of 41.7%, whilst the combination of pembrolizumab and gefitinib resulted in an ORR of 14.3%. The median PFS time for pembrolizumab plus erlotinib was 19.5 months (95% CI, 3.0–19.5), whilst for pembrolizumab plus gefitinib it was 1.4 months (95% CI, 0.2–13.0). The median OS time was not achieved and 13.0 months, respectively (95% CI, 0.2-NR) (59).

There are also combination therapies for other gene mutation inhibitors under study, with the JASPER study phase II (NCT04475939) reporting that niraparib, a poly-ADP ribose polymerase-1/2 inhibitor taken orally and combined with pembrolizumab, exhibited positive clinical outcomes (60). The NCT03157089 have investigated the use of afatinib plus ICIs in the treatment of NSCLC (61) (Tables III and SI).

Elements of the immune system, such as immune checkpoints, and the diverse and intricate alterations in cytokine levels, serve a role in regulating the immune response of the body against tumors in the TME (62). The impact of a PD-1/PD-L1 blockade is also affected by these variables (63). Compared with single use, the combined use of ICIs has a greater blocking effect and lower rate of drug resistance. Numerous research studies are currently assessing several combinations of cancer immunotherapy medications (64). Several checkpoint inhibitors possess distinct modes of operation. CTLA-4 inhibitors mainly act in the lymph nodes by restoring the induction and proliferation of activated T cells, and inhibitors of PD-1 function by acting on the outer region of the tumor location to hinder the neutralization of cytotoxic T cells (65). Moreover, research on cytokine therapy has reported that the administration of cytokine agonists results in the proliferation and activation of CD4⁺ and CD8⁺ T cells and NK cells (66). In addition to immunotherapy, alternative treatment options such as chimeric antigen receptor (CAR)-T cell therapy, tumor vaccines and other emerging therapies are also necessary. The immunosuppressive effect within the tumor may be decreased, and the simultaneous use of ICIs may impact the growth and stimulation of T cells, resulting in a synergistic effect (67,68). Inhibitors of co-stimulatory or co-inhibitory receptors function as co-stimulatory molecules, promoting CD4⁺ and CD8⁺ T-cell activation, clonal expansion, survival, cytokine synthesis and the formation of T-cell memory (69).

The growth of cancer cells relies heavily on nutrients, leading to depletion in the TME. To adapt to the conditions of limited nutrients, immune cells are prompted to modify their glucose, amino acid and lipid metabolism (70). Cancer is characterized by the emerging phenomenon of metabolic reprogramming (71). Throughout the process of metabolic reprogramming, the effectiveness of the antitumor response is undermined by immune cells as they undergo differentiation or polarization into phenotypes that suppress the immune system (72). Therefore, the malfunctioning metabolisms may impact the immune response against tumors and trigger an immunosuppressive TME to evade immune detection (Fig. 1).

T-cell immunoglobulin and mucin domain-containing 3, T-cell immunoglobulin and ITIM domain (TIGIT), lymphocyte activation gene 3, CD47 and V-domain Ig suppressor of T-cell activation are part of the upcoming wave of immune checkpoints (73–77). Furthermore, the combination of additional types of ICIs have demonstrated a synergistic impact in clinical investigations. The results of the CheckMate 227 trial indicated that the combination of nivolumab and ipilimumab exhibited lasting effectiveness in individuals with advanced NSCLC. The median OS time was 17.1 months (95% CI, 15.0–20.1) and 15.7 months (95% CI, 13.3–18.7) when receiving nivolumab plus ipilimumab and nivolumab, respectively, in patients with a tumor PD-L1 expression of ≥1%. In patients with a PD-L1 of <1%, the OS time was 17.2 months (95% CI, 12.8–22.0) for nivolumab plus ipilimumab compared with 12.2 months (95% CI, 9.2–14.3) for chemotherapy (78). A phase Ia/Ib non-randomized controlled trial evaluated the safety and antitumor activity of the anti-TIGIT antibody, tiragolumab, and its combination with atezolizumab in patients with advanced solid tumors. In the metastatic NSCLC cohort, the ORR was 46% and the DCR was 77%. The median response duration was 24.2 months (95% CI, 9.7-NR) (79). Additionally, in the CITYSCAPE clinical trial, the median PFS time was 5.4 months (95% CI, 4.2-not estimable) in the tiragolumab plus atezolizumab group compared with 3.6 months (95% CI, 2.7–4.4) in the placebo plus atezolizumab group (HR, 0.57; 95% CI, 0.37–0.90; P=0.015) (80). Numerous clinical trials have validated the effectiveness and safety characteristics of several combinations of ICIs (81–83).

Within the TME, high levels of adenosine (ADO), an immunosuppressive metabolite, have been reported to regulate tumor immunity. According to the existing evidence, ADO inhibits the production of cytokines and the growth of CD8⁺ and CD4⁺ T cells, as well as the cytotoxic activity of CD8⁺ cells (84). CD39 and CD73 are the principal enzymes involved in ADO generation and are expressed on cells within the TME. CD39 and CD73 impact the immune response by inhibiting the function of effector T cells and maintaining the stability of immunosuppressive regulatory cells (85). Moreover, the NCT02503774 study is testing a combination of oleclumab (anti-CD73) with durvalumab in NSCLC. However, the inhibitors of CD39 have yet to progress into clinical development. An ongoing study, NCT04148937, is evaluating the effectiveness of LY3475070 (a CD-73 inhibitor) in combination with pembrolizumab for clinical purposes. There are currently four main ADO receptors available: A1R, A2AR, A2BR and A3R (86). Several ongoing clinical trials are assessing the efficacy of ADO receptor inhibitors plus ICIs. OX40, also known as CD134 or TNF receptor (TNFR)SF4, belongs to the super-family of TNFR. OX40 is expressed by activated T cells, including CD4⁺ and CD8⁺ T cells (87,88). In several preclinical models, tumor regression was observed when using anti-OX40 monoclonal antibodies and OX40L-Fc fusion proteins (89,90). Ongoing clinical investigations are assessing the therapeutic potential of humanized anti-OX40 agonists combined with ICIs against diverse advanced solid tumors, including NSCLC.

Cancer cells have the ability to secrete diverse immune-modulating cytokines to inhibit T cells and create an environment of immune suppression (91). The interaction between TGF-β, the most crucial cytokine for suppressing the immune system, and epithelial-mesenchymal transition (EMT), along with its impact on the TME, could potentially serve as a resistance mechanism to ICIs (92). TGF-β mediates EMT by activating both the Smad signaling pathway and non-Smad signaling pathways. The Smad signaling pathway is a classic pathway of TGF-β signaling, where TGF-β activates Smad2/3 through its receptor, which then forms a complex with Smad4 and enters the nucleus to regulate gene expression (93). Additionally, TGF-β further promotes EMT through non-Smad signaling pathways, such as the PI3K/AKT and MAPK pathways (94). Several types of inhibitors exist for the TGF-β pathway, such as antibodies that hinder the binding of TGF-β to receptors, molecules that impede TGF-β receptor kinases and TGF-β ligand traps. Both galunisertib and vactosertinib are inhibitors of small-molecular receptor kinases. These inhibitors hinder ATP binding to TGF-β receptors, effectively obstructing the activation of Smad2 and Smad3 in response to TGF-β (95,96). Ongoing clinical trials (NCT02423343 and NCT04515979) are being performed to evaluate the efficacy of galunisertib combined with nivolumab and vactosertinib combined with pembrolizumab in NSCLC.

BEMPEG, also known as NKTR-214, is an IL-2 pathway agonist that specifically targets CD122 and can be attached to several detachable polyethylene glycol chains. NKTR-214 has the potential to enhance the growth and stimulation of CD8⁺ T cells and NK cells. Multiple clinical trials have investigated the utilization of NKTR-214 in conjunction with ICIs for several advanced solid malignancies (66,97).

The use of CAR-T cell treatment has been increasingly utilized in different types of solid tumor (98). The NCT03060343 trial is involving the administration of autologous CAR-T cells that targeted PD-L1 and CD80/CD86 to assess the safety, tolerability and engraftment potential for the treatment of recurrent or refractory NSCLC during its initial stage. During a phase I clinical trial (NCT03330834), the safety and effectiveness of anti-PD-L1 CAR-T cell therapy is being evaluated in patients with advanced NSCLC who had a positive PD-L1 status.

Cancer vaccines represent an exciting approach for cancer immunotherapy. In the NCT02897765 phase Ib trial, the combination of a personalized neoantigen-based vaccine termed NEO-PV-01 and a PD-1 inhibitor was reported to be effective in treating patients with advanced melanoma, NSCLC or bladder cancer. No negative responses associated with the therapy were identified in the study, and the combined application was reported to be safe. Neoantigen-specific CD4⁺ and CD8⁺ T-cell responses were triggered by the combination (99). Furthermore, in the NCT02823990 phase II study, the effectiveness of TG4010 (a vaccine consisting of modified vaccinia virus Ankara, human mucin 1 and IL-2) combined with nivolumab is being assessed in individuals with NSCLC who had received previous treatment.

Furthermore, oncolytic virus is a promising tumor immunotherapy agent. The oncolytic virus has the potential to selectively kill tumors and activate innate and adaptive immune responses in the host (100). The use of oncolytic virotherapy has been reported to enhance the infiltration of T cells within the tumor and result in an improvement in anti-PD-1 immunotherapy (101). Currently, ongoing clinical trials aim to investigate the combination of oncolytic virus and ICIs in solid tumors, such as NSCLC.

Finally, through the activation of T cells mediated by the T-cell receptor, zoledronic acid (ZA) has been reported to enhance antitumor immunity. In a mouse model of LL2 lung cancer, the combination of ZA and ICIs resulted in an increase in CD8⁺ IFN-γ⁺ T cells and γδ T cells, as well as a decrease in CD11b cells in both the circulation and TILs. The combination group exhibited markedly inhibited tumor growth (102) (Tables IV and SI).