Immune checkpoint blockade (ICB) therapy

TDLNs serve as a repository of anti-tumor immune cells, including CD8⁺CXCR5⁺ T cells, resident memory T cell populations,9 10 and TCF1⁺ CD8⁺ T cell populations.10–12 These cells are active in response to ICBs and can supplement the intratumoral immune-infiltrating cell repository, indicating that LNs are candidate targets for immune enhancement and reactivation.13 The efficacy of the PD-1 blockade is negated which significantly reduces exhausted CD8⁺ T cell (Tex) infiltration and eliminates increased Tex infiltration after PD-1 blockade in a mouse model when lymphocytes are locked in LNs using fingolimod (FTY720).14 This demonstrates that the PD-1 blockade can promote the infiltration of this type of Tex clone (mainly derived from TDLN), leading to anti-tumor immunity. This suggests that tumor-specific T cells in the tumor microenvironment (TME) are derived from TDLNs. TDLNs are T cell activation sites required for the response to ICB therapy15 and internally store immune checkpoint inhibitor (ICI) targets. Lymphatic endothelial cells are components of lymphatic capillaries that participate in the generation and remodeling of lymphatic vessels to accommodate the influx of immune cells from the tissues. They express high levels of programmed cell death ligand 1 (PD-L1) in vivo and in vitro, which may explain why some patients with cancer without PD-L1 expression in the TME respond to PD-1/PD-L1-targeted immunotherapy.16–18 Tumor-specific PD-1⁺ T cells colocalize with conventional dendritic cells (cDCs) expressing PD-L1,19 identifying PD-L1⁺ cDCs (and likely those of the cDC₂ lineage) as primary targets of anti-PD-L1 antibodies in TDLNs. TCF-1⁺TOX^-CD8⁺ T cells and TLDN-derived tumor-specific memory T cells are the true responders to PD-1/PD-L1 therapy in TDLNs.20 21 Anti-PD-1 therapy enhances B-cell differentiation, leading to dramatic remodeling within LNs, which is crucial for the success of PD-1 therapy in the anti-tumor immune response to tumor exposure.22 These data demonstrate that TDLNs are critical in ICB-mediated antitumor immunity, challenging the prevailing dogma that PD-1/PD-L1 checkpoints primarily function at tumor sites, and opening up additional avenues to discover biomarkers and combination immunotherapy regimens.

Local immunotherapy

ICB is a promising cancer therapy; however, it is inevitably associated with a low response rate and significant toxicity. The immune suppression status of local tumors may be higher than that of the entire body and a localized drug delivery system targeting tumors and TDLNs may achieve enhanced efficacy and reduced toxicity to some extent. The direct delivery of immunomodulators to tumors by peritumoral injection and LN perfusion may be accomplished using existing antigens, tumor-specific infiltrating lymphocytes, and APCs.23–25 This approach also confers tumor specificity and elicits a potent systemic anti-tumor immune response at lower doses with reduced toxicity compared with other approaches.26 27 In preclinical and clinical trials, a low-dose combination of anti-cytotoxic T lymphocyte-associated antigen-4 (CTLA4), anti-CD137, and anti-OX40 antibodies delivered within tumors or in a milieu of TDLNs demonstrated a potent systemic anti-tumor effect.28 Intratumoral injection of an anti-glucocorticoid-induced tumor necrosis factor (TNF) receptor agonistic antibody (Ab) induces antitumor immunity via transport to TDLNs, stimulating effector T cells, inhibiting regulatory T cell (Treg) activity in a mouse colon cancer model, and suppressing tumor growth more effectively than intraperitoneal or intravenous injections of this Ab.29

Immunoradiotherapy

Enhancing tumor immunogenicity and immune cell infiltration (establishing inflammation in the TME) helps improve immune regulation in immune-cold tumors, tumors with an immune rejection phenotype, and immunodeficient tumors.30–32 Immunotherapy (iRT) significantly enhances the abscopal effects of radiation therapy. Radiation therapy may partially help overcome immunotherapy resistance by increasing the release of tumor antigens and reprogramming the TME, with TDLNs playing a crucial role in this process. The effectiveness of high-dose radiation therapy depends on the presence of CD8⁺ T cells.33 Radiation therapy can promote an immunogenic cell death thus favoring tumor antigen uptake by dendritic cells (DCs). Damage-associated molecular pattern and interferon type I, which are induced by radiation, stimulate DCs to mature and migrate to the TDLN, where they activate tumor-specific T cells.34 TDLNs are the primary site of education and expansion of tumor-specific cytotoxic immune cells. Preclinical studies support the preliminary preservation of LNs for radiotherapy or ICI treatment. Type I interferon receptor (IFNAR) signaling and cDC1s in TDLNs are crucial for activating tumor-specific cytotoxic lymphocytes and upregulating anti-tumor immune cell factors and chemokines, leading to subsequent tumor infiltration. Selective LN irradiation of TDLNs without tumor metastasis disrupts the recruitment of effector T cells driven by chemokines, weakening the effect of combination therapy between radiotherapy and immunotherapy. The preservation and activation of immunoregulatory functions in unaffected TDLNs are important. Therefore, immunological evaluation of affected and unaffected TDLNs is necessary.35–37 The activation status of TDLNs may determine the outcome of radiotherapy and the manifestation of distant effects.38 Resection of bilateral but not unilateral TDLNs significantly reduces the anti-tumor and in vitro effects of iRT stimulation in unilateral and bilateral subcutaneous xenograft models established in mice with colon adenocarcinoma cell lines; TDLNs play a key role in iRT by promoting CD8⁺ T cell infiltration and maintaining M1/M2 macrophages.39

Adoptive cell therapy (ACT)

TDLNs are a good source for ex vivo expansion of immune cells.40 41 Tumor-reactive T-cells in non-metastatic LNs can be expanded in vitro,1 indicating their potential for ACT. Lymphocytes in SLNs are the first LNs to receive tumor antigen stimulation and possess potent tumor recognition-specific immune memory and are considered an ideal starting source of cells for ACT. Phase I/II clinical studies demonstrate that adjuvant SLN-derived tumor-reactive T cell (SN-T cell) immunotherapy improves long-term survival (24-month survival) in patients with metastatic CRC (mCRC).42 These cells can replace tumor-infiltrating lymphocytes; thus, it was hypothesized that SLNs may be the ‘headquarters’ for antitumor immunity. In vitro, SLNs contain higher levels of interferon γ, TNF-α, and sFas than non-SLNs after being removed from the immunosuppressive environment, indicating that SLNs can enhance the antitumor immune function of T cells in ex vivo culture. This finding further emphasizes the specific value of SLNs in promoting the response to cancer immunotherapy and the development of SN-T-cell-based immunotherapy for CRC.43 In addition, successful therapeutic strategies require the stimulation of humoral immunity, and B cells in TDLNs may serve as effector cells in an immunotherapy model with adoptive cell transfer. Treatment with adoptively transferred B-cells and T cells results in a more effective antitumor response than treatment with B-cells or T cells alone.44 Overall, the synergistic antitumor efficacy of cotransferred activated B cells and effector T cells represents a novel approach for cancer immunotherapy.45 These findings further elucidate the biological functions of antitumor effector B cells, which were not previously studied in CRC and may constitute an alternative cellular therapy for cancer.

Tumor vaccines

In-situ vaccination triggers systemic anti-tumor responses through the induction of local immunity. Tumor vaccines are typically administered systemically or at non-tumor sites. This approach was employed in basic research on lymphoma, melanoma, and breast cancer to improve the efficacy of anticancer vaccines via targeted delivery to TDLNs.46–48 Improving the delivery system of tumor vaccines has partially addressed the problems of antigen presentation deficiency and low efficiency of DCs. During in situ vaccination, adjuvants are delivered to the tumor and TDLNs in the form of micelles to regulate TDLNs by activating DC migration from the tumor to the TDLNs and by directly activating resident DCs in TDLNs through lymphatic drainage, leading to anti-tumor immunity.49 The nanovaccine utilized in this study was a type of DC hybrid zinc phosphate nanoparticle designed for simultaneous delivery of antigenic peptides and photosensitive melanin. On administration to mice with MC38 tumors, the nanovaccine efficiently drained into the TDLNs and infiltrated the tumor site. Furthermore, this nanovaccine effectively delivered antigens to the DCs, thereby promoting their maturation. Additionally, it induced the proliferation and infiltration of CD8⁺ and CD4⁺ T cells, together with the secretion of immune-stimulating cytokines. Ultimately, these immunological responses exert significant anti-tumor effects.47

Optimal administration of cancer vaccines requires the induction of tumor antigen-specific T cells to ensure their effective introduction, sufficient retention in lymphoid organs, and the alleviation of immunosuppressive factors. Therefore, the combination of tumor vaccines and ICIs may have strong synergistic anti-tumor effects.50 T cell-targeted exosomes modified with costimulatory molecules, major histocompatibility complex, antigenic OVA peptides, and anti-CTLA-4 antibodies were designed (EXO-OVA-mAb) by combining the strategies of vaccines and checkpoint blockade. These exosomes exhibit enhanced binding to T cells in TDLNs, effectively induce T cell activation, improve the tumor homing of effector T cells, and increase the ratio of effector T cells/Tregs in the tumor, leading to significant tumor growth inhibition.51

Targeting TDLNs for diagnosis

Exploratory studies on non-small cell lung cancer demonstrated that TDLNs have unique characteristics that differ from those of tumor lesions and circulating immune cells. Immunophenotyping of TDLN samples is feasible. Elevated PD-1 levels in TDLNs may serve as predictive or early response biomarkers for PD-1 checkpoint blockade.54 In addition, the first studies on changes in T cell subpopulations within TDLNs after treatment with anti-PD-1 checkpoint inhibitors (NCT04082988 and NCT03446911) were recently reported. An increase in the frequency of Tregs in TDLNs may be a marker of resistance to anti-PD-1 checkpoint inhibitors. Considering this observation, the potential of biomarkers and immune spectrum of TDLNs should be conclusively explored.55

Molecular pathological diagnosis

Currently, the focus on the pathological diagnosis of CRC LNM is limited to histological approaches. The same degree of pathologically diagnosed differentiation in the primary tumor is associated with various histological types (tubular, cribriform, poorly differentiated, and mucinous) in the metastatic LNs.56 Molecular genetic changes in LNs were detected to reveal the relationship between primary lesions, metastatic LNs, and distant metastases of CRC at the genetic level. High TMB is more frequently observed in LNs than in distant organ metastases, whereas a similar pattern is observed in MSS tumors. High TMB LNMs are more commonly associated with PD-L1 overexpression when evaluating the genetic changes associated with MSS LNMs.6 However, there are numerous gaps in research on genetic information and protein expression related to LNs.52 53

Currently, the basis for immunotherapy mainly originates from PD-L1 expression in tumor cells, which is crucial to impair the cytotoxic function of T cells. However, response to ICIs is observed in some PD-L1-negative patients.57 Human and mouse studies show high levels of functional PD-L1 expression in DCs, macrophages, natural killer cells, and regulatory B cells (Bregs) in draining LNs.58–60 Simultaneously, blocking PD-L1 derived from tumor and non-tumor sources can maximize T cell anti-tumor responses and demonstrate a synergistic effect.

Regulatory T cells expressing high levels of CTLA-4 and PD-1 accumulate abundantly in TDLNs and metastatic LNs, reflecting the tumor-driven immunosuppressive environment of TDLNs.61 Benign LNs can be used to predict the response to immunotherapy, with benign LNs in non-responders showing significantly higher expression of cytotoxic T-cell markers, a higher proliferation index (Ki67), and macrophages expressing higher levels of PD-L1.62 PD-L1 and PD-1 are abnormally expressed in the metastatic LNs of T4 stage rectal cancer, and PD-L1 and PD-1 protein expression is more common in patients with T4 stage rectal cancer with LNM than in those without LNM.63 This pattern demonstrates the potential contribution of PD-L1 derived from non-tumor cells. Therefore, a comprehensive evaluation of PD-L1 expression may be a more accurate approach to predict responses to PD-1/PD-L1 blockade therapy than monitoring PD-L1 expression in tumor cells alone.64–66 This provides a new direction for optimizing ICI efficacy prediction and developing clinical biomarker strategies.

Currently, biomarkers derived from tumors or serum samples are widely recognized; however, the immune information carried by TDLNs before or during treatment has been ignored. Measuring the expression levels of biomarkers such as PD-1 and PD-L1 and determining the TMB in TDLNs may help guide the selection of treatments for pMMR patients with high expression levels of ICI response targets.67 Therefore, future clinical trials of ICIs should include the molecular pathological diagnosis of metastatic LNs and monitoring changes in cellular biomarker levels before and after treatment to understand the predictive value of these findings.

Immune spectrum

The immune score reflects the outcome of the battle between local host immune cells and cancer cells, and the potential for immunotherapy may depend on the host immune status. Thus, consideration of host immune status can allow for more accurate targeting or expansion of the population benefitting from ICIs to a certain extent. The TNM immune staging system is the first standardized immune-based cancer classification method that originated from CRC research and incorporates consensus immune scoring into the TNM classification system.68 The combination of CD20⁺ B and CD8⁺ T cells constituted the TB score. Combining the immune score with tumor growth patterns allow the prediction of patient survival independent of MSS-related and MSI-related molecular features, and the patient population with a high density of memory and cytotoxic T and B cells experienced prolonged disease-specific survival independent of MSI status.69

However, MSI status does not consistently correlate with a high immune score, with a significant percentage of MSI tumors possessing a low T cell density, similar to that of MSS tumors (26%–35%).70 This is a positive and useful indication that starting an evaluation with the knowledge of the patient’s immune status may allow for a broader application of immunotherapy. Tumor progression in the context of host immunity does not unilaterally depend on the tumor but on the patient’s immune function, which determines the patient’s ability to tolerate tumor invasion and respond to treatment.71 TDLNs can capture different features of tumor neoantigens and T cell activation and even reflect the status of immune-inhibitory cells, which may provide information for the precise stratification of patients for immunotherapy and predict efficacy.72 73 CellTypist is a machine-learning tool for the rapid and accurate annotation of cell types that was used to construct a human cross-tissue immune cell atlas that uncovered tissue-specific characteristics of immune cells.74 This type of approach may be used to develop a TDLN-specific immune scoring system based on tumor immune scoring to encompass the immune spectrum of TDLNs, and more clinical studies should be conducted to evaluate the practicality and promotion of the TDLN immune spectrum.

TDLNs as therapeutic targets

Clinical studies on immunotherapy that do not restrict enrolment based on MMR status are a good first step in overcoming this bottleneck in immunotherapy. The difference in responses between dMMR and pMMR tumors is mainly attributed to variations in TMB, tumor neoantigens, and T-cell infiltration.75 76 Notably, CD8⁺PD-1⁺ T cell infiltration can predict the response of pMMR tumors, suggesting that some pMMR tumors are immunoreactive despite not being dMMR at the molecular level.77 78 This indicates that the MMR status is not the sole factor determining immune status and immunotherapy response. Analysis of the immune characteristics of patients with MSI-H/dMMR CRC revealed that they primarily exhibit a higher TMB and tumor neoantigen burden, more robust lymphocyte infiltration of tumor tissue, and higher levels of markers for cytotoxic T lymphocytes, helper T cells, T follicular helper cells, and tissue-resident memory T cells.

Collected resected primary tumor blocks are categorized as exhibiting ‘high’ or ‘low’ immune responses according to the type of infiltrating lymphocytes based on the hypothesis that the efficacy of ICIs in MSS CRC can be enhanced by increasing immune cell infiltration and T cell activation in the TME.79 The efficacy of chemotherapy and anti-angiogenic Ab therapy in combination with pembrolizumab was compared in patients with pMMR mCRC exhibiting high immune scores and/or high mutation genotypes (NCT04262687, Recruiting). Clinical studies on immunotherapy without enrolment limitations based on MMR status demonstrate the pre-existing role of immunotherapy in patients with operable stage I-III CRC. Downstaging of large-volume tumors in the setting of planned resection may be the most likely benefit of neoadjuvant immunotherapy and may even allow organ preservation75 80 (table 1). We anticipate that more prospective studies on the use of immunotherapy for MSS/pMMR CRC will establish a new paradigm for CRC immunotherapy (table 2).

A recent review proposed an optimal approach for the local delivery of immune modulators to TDLNs to overcome the blockade of early and late-stage DC activation in cancers. This strategy leads to effective systemic anti-tumor immunity by promoting T cell recruitment to the tumor, even in cases where the primary tumor is removed (but TDLNs are present). The direct immune modulation of TDLNs may induce effective systemic anti-tumor immunity.81 The presence of CD169⁺ macrophages and an abundance of CD8⁺ tumor-infiltrating lymphocytes in the regional LNs of patients with CRC indicate a better prognosis independent of the dMMR/MSI-H status, suggesting that the immune cell infiltration status of LNs can be used to classify an anti-tumor group that differs from dMMR CRC.82

TDLNs can accumulate in high concentrations and exhibit high drug bioavailability after local delivery of ICIs, which leads to the activation of effector and memory T cells, thereby offering durable protection against disease recurrence while reducing toxicity.83 84 Recently, numerous clinical studies have explored the efficacy of topical delivery of ICIs (NCT03889912, NCT03316274, NCT04090775, NCT03755739, and NCT04274816). However, further clinical research is required to explore the value of TDLNs for immunotherapy stratification and efficacy prediction based on molecular pathological diagnosis and the immune spectrum of TLDNs.

Conclusions and perspectives

TDLNs are key determinants of immune activation and tolerance and can reflect the characteristics involved in the battle between immune and tumor cells. This information has helped us further understand the role of LNs in the progression of CRC and will ultimately help us decide whether to preserve or remove LNs. In addition, this information will aid in the clinical consideration of the timing of immunotherapy, since the removal or retention of TDLNs may alter the response to immunotherapy. Ongoing clinical trials on neoadjuvant immunotherapy should also investigate changes in TDLNs and metastatic LNs (including LN size and structure, immune cell type, subtype, activation status, and function) after treatment with ICIs since this knowledge may help determine the most effective treatment timing. After clarifying whether to remove or preserve TDLNs, we need to consider how to target immune therapy for TDLNs. This requires improvement in the diagnostic techniques for TDLNs. Molecular pathology diagnosis is important to identifying targets for immune therapy and the immune spectrum can be used to help evaluate patient responses to immune therapy. The combination of these two factors can help predict the effectiveness of immunotherapy. It will be important to focus on the integration of disciplines, such as oncology, surgery, pathology, and radiology to explore more effective future treatment strategies.

The contribution of TDLNs to the response to immunotherapy raises questions regarding the decision to dissect LNs in CRC. Do we need different dissection strategies based on whether the LNs are infiltrated by tumor cells? Additionally, the efficacy of immunotherapy is related to different dissection strategies. There is an urgent need for non-invasive detection techniques for LNs, together with more clinical trials to find answers. Currently, the pathological detection of tumor cell infiltration limits the acquisition of immune information related to resected LNs. Performing molecular pathology and immune spectrum diagnoses of TDLNs will facilitate significant advances in selecting immunotherapy options and predicting treatment efficacy.